Thursday, February 18, 2010

Another Simple Argument Against Any General Theory of Consciousness

... related to my first Simple Argument, but spun out a bit differently.

(1.) The history of philosophy shows that no theory of consciousness can avoid having some highly unintuitive consequences. (Or more cautiously, the history suggests that. The strength of the conclusion turns in part on the strength of this premise.)

For example, if functionalism is true, some very weird assemblages will be conscious. If consciousness depends upon material constitution, then beings behaviorally indistinguishable from us but materially different might entirely lack consciousness. And: Intuitive notions of consciousness seem to involve sharp boundaries not present in the evolution or development of conscious systems. And so on.

(2.) Therefore, something apparently preposterous must be true of consciousness.

(3.) Therefore, reflection on what is intuitively true -- and metaphysical speculations that depend on such intuitions -- cannot be a reliable guide to consciousness. (What such speculations yield, as is evident from the literature, is a variety of idiosyncratic hunches.)

(4.) Empirical observation of physical structure and behavior also cannot settle the question of which preposterous things are true, because their interpretation depends on prior assumptions about consciousness. (For example: Does observing such-and-such a functional structure establish that consciousness is present? Only given such-and-such functionalist assumptions.)

(5.) So we're stuck.

If we are stuck, the live options seem to be mysterianism (we will never know the truth about consciousness) or eliminativism (the concept of "consciousness" is broken to begin with, so good riddance).

27 comments:

metabenny said...

Hello! I am just curious as to what specifically you were thinking of regarding your first premise and the history of philosophy....

Adam Arico said...

If this argument holds for consciousness, then it would also seem to hold for all of the traditional philosophical questions. History of philosophy also shows (or at least suggests) that no analysis of knowledge, no moral system, no theory of causation... can avoid some very counter-intuitive consequences. Do you think we are in the same position in these other philosophical domains? Or do you think one of your premises holds uniquely for consciousness (like, maybe, premise 4)?

Peter said...

Isn't the aim to provide, not just the right answer, but the right kind of explanation, so that the things which initially seemed counter-intuitive suddenly (aha!)turn out to make perfect sense. At the moment all the answers look counter-intuitive, but when we've got the correct perspective, we'll be saying 'How stupid not to have thought of that!'

Eric Schwitzgebel said...

Thanks for the comments, folks!

Metabenny: I'm not sure I understand your question. I presented some examples immediately after the first paragraph; but maybe it wasn't clear to you that I was thinking of those sorts of discussions (Putnam, Block, etc.) as "history"? Going farther back, we see struggles with the mind-body problem back at least to Descartes.

Adam: I think philosophical topics vary with respect to this question. For example, I don't think it's clear that a theory of knowledge or an ethical theory has to have flat-out counterintuitive consequences, even if there are disagreements at the margins and all theories will have some consequences that some (perhaps minority) of people will find unintuitive. But to the extent it becomes clear that all theories do have robustly counterintuitive consequences about non-marginal cases, then I think that the armchair method collapses, as it does in the metaphysics of consciousness.

On the empirical end, too, I think there can be disanalogies between consciousness and other philosophical topics. I don't think the empirical study of knowledge, for example, requires the assumption of highly contentious theories about knowledge in order to get started.

Peter: Yes, that's the optimistic scenario, and things often work like that in other domains. I see no grounds for that optimism, however. Suppose panpsychism, for example, is a live option. I just don't see how the empirical study of physical structures and patterns of behavior could prove or disprove it.

Gilbert Wesley Purdy said...

"I was looking for consciousness in all the wrong places / Looking for consciousness in too many faces / Searching your eyes, looking for traces / Of what.. I'm dreaming of..."

I followed the links to your various posts relating to consciousness and am aware of a bunch of intelligent folks looking for something they cannot define or describe. You "know" that it is related to this or that -- or mostly to the fact that you feel conscious and you feel that your feeling of consciousness has to do with your senses and/or an ability for complex thought. Please forgive me for stating that it should be clear that your senses have less to do with consciousness than unconsciousness (more generally referred to, in human beings, as "subconsciousness"). Moreover, a definition of consciousness- per-se (remember, I said "per se") has nothing at all to do with the senses. Period.

As for complex thought processes, wouldn't we agree that the subconscious seems to do some pretty complex thinking (better said "processing," perhaps) without the conscious mind being deployed? In fact, if all complex processing were conscious we'd die within a few days. That's the whole point.

To look for the boundaries of a persistent feeling inside yourself (or that feels like it is inside of yourself (a self, I might add, that you feel you are)), shared to some degree by other human beings, as indicated by anecdotal evidence, to begin with, is a scientifically and philosophically untenable approach.

Graham said...

Hi Eric,

Could you give an example of the kind of "weird assemblage" that you refer to at the beginning of your post?

Also, when you talk about "material constitution", are you just referring to, say, the chemicals that the parts of a mind are made of, or something else?

Cheers,
Graham

Michael Metzler said...

Given the recent rapid development of the mind sciences, it does not seem as though we should be stuck here. The science makes suggestions about how to even use the introspective side. Take for example how human understanding appears tied to embodied simulation. The constant work of our imagination - whether an evanescent occurrence of episodic recollection or the more opaque tools that allow us to follow a narrative or empathize with someone's pain - is to some degree associated with what we would call consciousness, and yet it is not prototypically what philosophers refer to as "consciousness" as they go about there folksy armchair instantiation of their sinecures. This 'in-between' aspect of our experience as largely 'non-conscious' creatures would seem to give us much to work from (and what is scientific exploration but the use of the human imagination?). It should be expected, also, that our experience is latent with plenty of vestiges, permitting analogies to other species and subspecies - the White Whale, among others.

Autumnal Harvest said...

Is asking for a "General Theory of Consciousness" asking a meaningful question? I'm not well-read on theories of consciousness, but it seems to me that the theories that you describe differ in how to define consciousness (i.e. which states to choose to put in the "bin" of "things that are conscious"), rather than differing in empirical predictions. Given different definitions, all of which agree with our intuitions about how to define things in some ways, but not in others, I'm not sure what it means to ask which one is correct. If we both have precise definitions for chair, but mine includes footstools, while yours excludes them, it doesn't seem meaningful to debate whether a footstool is "really" a chair. I think there's nothing to do with that question other than say "it depends on your definition of 'footstool.'" Similarly, if someone asks whether Searle's Chinese room really "understands Chinese," I think there's not much to say other than "it depends on your definition of 'understands.'" Definitions can be more useful, or less useful, but I don't see how you can about their correctness - you're just arguing about what things you choose to put in a human-created bin.

The only case where I can see it's important to decide what things go in which human-created bins is when you have a pre-existing body of knowledge about how objects in each of those bins will behave, so that a statement about which bin is appropriate for an object is tantamount to a statement about how that object will behave. For example, we've already accumulated detailed and useful rules about how things that are classified as "particles" behave, and how things that are classified as "waves" behave, so arguing over whether something in a certain situation belongs in the "particle" bin or the "wave" bin is a useful and meaningful argument about what set of rules will apply. But if we find something in a certain situation matches each bin partially, then arguing over whether that something is "really a particle or wave" becomes meaningless. It seems to me that arguing about what things go in the bin of consciousness is like that. Although, like I said, I'm not too well-read on the philosophy of consciousness, so I may be missing something basic.

Adam Pautz said...

Hi Eric, your simple argument and your other simple argument are interesting. (And, cool blog btw - first time poster, long time reader.) A few small thoughts.

(1) At the end you mention 'mysterianism' and 'eliminativism'. Another possibility is 'the indeterminacy view'. We cannot know because there is no fact of the matter. Indeed, I am somewhat inclined to think that, if a certain kind of reductive physicalism is true, then terms for consciousness might highly indeterminate in reference (bewteen functional properties, neural properties, environment involving properties, etc). In an extreme form, the idea is that no one theory of consciousness is determinately true; and many issues (about externalism/internalism, consciousness w/o attention, BIVs, multiple realizability, the nation of china, etc) are simply indeterminate. Sider takes a similar view about personal identity in his 'Criteria of Personal Identity and the Limits of Conceptual Analysis'. Alston (in his last book) took a similar view about 'justification'. I am surprised this view is not more discussed in the case of consciousness. I have another argument (different from 'there is no way to decide') for thinking that, if a certain kind of reductive physicalism is true, then this sort of indeterminacy view is right (but I will not try to explain here).

(Btw, I don't mean to suggest that this is my own view. I reject the relevant kind of reductive physicalism. I think people have properties that are super natural even though their supervenience bases might be really unnatural. So these properties are 'reference magnets' for terms of consciousness. I think that this represents an interesting way of avoiding the indeterminacy view - although it may not avoid the skepticism you voice about our *knowledge* of the extension of conscious properties in weird cases.)

(2) The skeptical arguments you give are very similar to those that David Papineau gives in *Thinking about Consciousness* and 'Can There be a Science of Consciousness?'. He also defends the indeterminacy view (the only defender I know of). You might want to check it out - I think you'd like what he says.

(3) A very vague point: I think I agree with you that the extension of conscious properties is extremely hard to know - e. g. whether the nation of china can have them, or a BIV, or whatever. But it still might be that we can have justified beliefs about the 'structure' of these properties - e. g. that they have an intentional structure, as on intentionalism. In that case, skepticism about whether we can know the extension of conscious properties in certain cases might be compatible with interesting theories of them.

Btw, do you develop, or plan to develop, these points anywhere (maybe in the book you're working on)? Bc I'm very interested in this general issue - which, as I said, I think should be discussed more.

Allen said...

So we have our orderly conscious experiences and we want to explain them. To do this, we need some context to place these experiences in. So we postulate the existence of an orderly external universe that “causes” our experiences. But then we want to explain what caused this orderly external universe…why the particular initial conditions and causal laws that result in what we observe?

Since nothing can explain itself, the only option seems to be to postulate the existence of a much larger multiverse whose own conditions and laws provide an explanatory background for ours. But then what explains that multiverse? A multi-multiverse?

That line of thought seems to lead to the need for an infinite series of ever larger contexts against which to explain the previous context that we used to explain the previous context that we used to explain the fact of our initial experiences.

Surely that can't end well...

So this is basically Kant's first antinomy of pure reason. Either there is a first cause, which itself is uncaused, OR there is an infinite chain of prior causes stretching infinitely far into the past. But why this particular infinite chain as opposed to some other? In fact, why our particular "infinite chain of prior causes" or "first cause" instead of Nothing existing at all?

It seems that either way (infinite chain or first cause), at the end you are left with only one reasonable conclusion: There is no reason that things are this way. They just are.

BUT...we could have said that about our experiences to start with and saved ourselves the trouble of postulating a whole multiverse of multiverses! I'm kind of thinking of Philip Goff's Ghosts here.

Why do our conscious experiences of an orderly reality exist? That's just the way things are, there is no reason for it.

Do you think?

Ken Marable said...

Personally, I'm of a mind that a general theory of consciousness isn't really possible either... at this point. To speak in analogies, it's like Maxwell trying to come up with the Grand Unified Theory of physics before bothering to develop the theory of electromagnetism.

Our understanding of consciousness is still rough enough that discussions often conflate terms or, as is most often the case discuss "conciousness" as some unified phenomena to be explained. My opinion is that "consciousness" is best understood as a mixture of a wide variety of psychological processes from perception/qualia to beliefs & desires to language to self-awareness to other-awareness to memory to pretence, etc. All of these and more are conscious phenomena and all play a role in "consciousness", but I don't see us having enough of an understanding of each individual portion (let alone even a clearly defined list of all the individual portions) to even begin trying to develop a theory that underlies all of these.

We need our electromagnetism and relativity and quantum mechanics fully understood before we can try for a grand unified theory. And with consciousness, we need a list of what even falls under that heading, and then what theories best account for portions of that list. And THEN we can look at those theories to try and find an underlying theory or theories to account for those.

Maybe it's just science envy on my part, but after returning to philosophy of mind after a 14 year break, so amny arguments appear circular and boil down to "You theory handles X great, but can't account for Y." "Yes, but even if you explain Y well, my account of X is better than yours." and so on. Looking at it from a meta-level, it really appears many philosophers and cognitive scientists are addressing different questions but not acknowledging that fact. It all gets lumped together in trying to develop A theory of mind, or a single general theory of consciousness. This, to me, seems to be leading to less progress rather than more.

Lastly, I have seen in some of your older posts allude to problems with comprehending lesser consciousness and the transition from non-conscious to possessing consciousness. If we understand consciosuness as a blend of many processes rather than a bizarre on-off switch of there or not-there, this doesn't appear to be a problem to me. Going back the evolutionary (or developmental) timeline, indivdual processes drop away so that a being has most all the same concsiousness as us except self-awareness, for example. And each of these processes drops away or simplifies until we get to the most basic level of very simple perception with the barest of information. I have to brush up on my biology, but I imagine there are extremely primitive organisms with perceptions that are little more than binary light/no light or food/not food. The transition of an organism not having any perceptions to having that drastically limited of perception, to me, seems easy to comprehend. Of course, if I try to imagine myself as that base organism, it seems dramatic, but that because I'm carrying all of my consciousness baggage with me. but that's a limit on my intuition, not on my theory.

Eric Schwitzgebel said...

Thanks for the comments, folks!

Gilbert, I'm not sure I understand your comment.

Graham: Weird assemblages would include arrangements of beer cans and wire in outer space (Searle), an empty-headed robot controlled by a nation agreeing to instantiate a computer program (Block), or naturally occurring cities or countries, if they show suitable functional organization (Paul Churchland).

Adam: Very interesting thoughts, thanks! On (1): Many philosophers do think that "consciousness" is an ambiguous term (e.g., Block 1995; Lycan 1996), but then they go on to focus on one (or more) of the ways of disambiguating. WRT personal identity or justification, once you've disambiguated you're pretty close to done with the philosophical work. WRT to consciousness, maybe, you've just begun. On (2): Thanks for the tip! I've read other parts of that book, but not the chapter you mention. On (3): Maybe, but if panpsychism is in play, I wonder how much you can get this way. On your final point: In my forthcoming book I briefly present these ideas (in the Preface at the end of Chapter 6), but I haven't developed them in detail.

Eric Schwitzgebel said...

Michael: I agree that the mind sciences are rapidly developing on some fronts. On other fronts -- maybe the more important fronts for this question -- it's less clear, I think.

Autumnal: I'm sympathetic with that perspective on many philosophical questions, but not this one. There is a fact of the matter whether the Chinese room is conscious in the sense of "conscious" I am now using (phenomenally conscious). That fact cannot be settled by stipulative definition, as far as I can see. Nor do I see how empirical evidence of the sort available to science could settle it non-question-beggingly. At least, that's my thought.

Allen: That seems a bit off topic to me.

Ken: The optimists may be vindicated in the end, but I'm not sure optimism is justified. What is the basis of that optimism? On your last point, I'm of two minds, as you probably know!

Adam Pautz said...

Thanks, Eric. I think we should distinguish between ambiguity and indeterminacy. ('There is a bank over there' is ambiguous but can have a determinate truth-value in a context.) The view I'm describing (which is basically Papineau's, at least in his book) is that most issues about consciousness are indeterminate - this is quite different from Ned's (etc.) ambiguity view, and more radical.

Could you please say a bit more about "WRT personal identity or justification, once you've disambiguated you're pretty close to done with the philosophical work. WRT to consciousness, maybe, you've just begun."? Sounds interesting, but I didn't quite follow (my fault).

Gilbert Wesley Purdy said...

I also suggested that looking for a Theory of Consciousness in the details of how the conscious mind registers the input of the senses (by all indications, the single most popular line of proceeding) does not seem promising. First of all, the consciousness-per- se (that is to say, in and of and only of itself) should be understood to be an entity apart from all extrinsic relationships that exist also for the subconsciousness. If a being that is subconscious can process sensory data then the processing of sensory data cannot define consciousness.

There is a further problem with defining/deriving consciousness from sensory data. I would suggest that it is much like trying to define a high-wire act safety net by attaching multiple sensors to every strand where it is attached to the frame of the net. While legitimate data will be collected, and it will indeed describe properties of the net, the method of proceeding is never likely to arrive at the answer that the net is composed of fiber ropes, in a two dimensional matrix format, and designed to prevent injury to human beings due to falling from generally injurious heights. Instead vast numbers of definitions like "an entity which registers X foot lbs along plane Y when a guy named Bob falls into it" will gratifyingly accrue -- if not forever, at least for a very very long time until the mass of data arrives at a correct answer which inspection of the net itself, and its context, would probably have arrived at in half an hour.

A genuine theory is not derived from data but rather gives the researcher some hope of knowing what data might be useful so that she or he isn't left with the option of filling super computer after super computer with the data from whatever the most recent best guess was as to what data was important. I don't mean to suggest that you don't know this every bit as well as I, but, no present model having succeded in distancing the field, the research seems to have fallen back more or less upon collecting data in the hope that something will eventually emerge from it.

As for my final point, in the initial post, I stand by my original hint: "As for complex thought processes, wouldn't we agree that the subconscious seems to do some pretty complex thinking (better said "processing," perhaps) without the conscious mind being deployed? In fact, if all complex processing were conscious we'd die within a few days. That's the whole point." The answer isn't that consciousness give us the capability of complex thought/processing but that it gives us some "further adaptation".

Well, one thing is for sure: philosophy is long and comments are short. Sigh...

Gilbert Wesley Purdy said...

"Gilbert, I'm not sure I understand your comment."

I'm a little bummed that the point of my comment didn't come through at all. But then I was trying to suggest a confusion that seems to pretty much infuse present approaches to "Theory of Mind," and, perhaps I tried to portray the confusion a bit too confusedly.

At least I'm provided an excuse to comment once again. First, the Waylan Jennings "remix" was hopefully a way to express the overall theme of my initial comment. I might have hoped that it suggested that your #2 above ("Therefore, something apparently preposterous must be true of consciousness.") might be true because consciousness needs to be looked for in a radically different fashion than philosophers seem presently to be looking for it.

While we may feel that the experience of qualia is central to conciousness, maybe our feeling, profoundly shared though it may be, is wrong. Maybe consciousness itself is nothing like what it feels like. More to the point, regardless of the legitimacy (or lack there of) of our feelings, in this matter, they have long and wisely been considered the least effective data for scientific or philosophical research -- so much so that the rules of research prohibit their use. They lead to the confusion I tried to portray in my initial comment and which you yourself seem to say you feel.

Gilbert Wesley Purdy said...

Aaaargh! Somehow I managed to post the two parts of the previous comment in reverse order. Please be advised.

Autumnal Harvest said...

"There is a fact of the matter whether the Chinese room is conscious in the sense of "conscious" I am now using (phenomenally conscious)."

Can you clarify what it means for the fact of the matter to be one way or the other? Correct me if I'm wrong, but seems to me that the background assumption behind your sentence must be either (1) there is a well-agreed upon definition of the word "conscious" among English speakers general enough to cover non-humans, or (2) there is a naturally occurring category of "things that are conscious," pre-existing before human attempts to determine what is in this naturally occurring category. But (1) does not seem to be true, and while (2) is possible, I'm not sure I see the evidence for it.

Eric Schwitzgebel said...

Adam: Yes, indeterminacy and ambiguity are different. Sometimes the one follows from the other, though: The ambiguity of the meaning of "person" can make it indeterminate what exactly one means by it and thus indeterminate whether some entity is a person in the relevant sense. I had thought you were referring to that sort of argument.

On your second question: Regarding personal identity, my own view is that there are different legitimate ways to think about what a "person" is. Disambiguate the ways and the puzzlement disappears, or rather is replaced by the project of pointing out the practical advantages of one way of thinking about personhood (which is a legitimate philosophical project, but not the way the personal identity debate is usually conceived).

Regarding consciousness: The term "conscious" has a variety of uses (including not knocked-out and including knowing about something, as well as the phenomenal sense). The phenomenal sense may or may not be unambiguous and unproblematic, but let's suppose it is. Once you've clarified that that's the sense you mean, you have clarified your target, but the hard work (e.g., defending materialism) is still before you.

Eric Schwitzgebel said...

Gilbert: That helps, but I'm not sure you aren't using terms differently from me in a way I don't understand. How could consciousness be different from what it feels like? Is this just a way of saying that consciousness is different from how we judge it to be (a statement with which I would agree)?

I agree in general that theory and data always come together. This will result in different theorists seeing the data differently. Usually in science the gap is bridgeable over time, but I don't see much hope for that regarding theories of consciousness.

I also agree that very complex thinking can transpire without consciousness -- at least if we reject certain absurd-seeming views of consciousness. What this implies, exactly, about the functional role of consciousness is unclear, though.

Eric Schwitzgebel said...

Autumnal: You are right that I am probably assuming one or both of those things. I am inclined to regard both as true (though 1 requires qualification to rule out some of the ambiguities in the term). But among the potential responses to the puzzlement I feel about consciousness would be to reject those assumptions.

Would you say that there is no natural kind or property in the universe that corresponds to the philosophers' concept of "phenomenal consciousness"?

Autumnal Harvest said...

I do tend to think that, although that's perhaps more gut instinct than the result of particularly careful thought. I think that categories that exist in nature, like "X is an electron" are discrete and can be well-defined in principle (even if we might not yet have the precise defintion). Conversely, if it seems that if a category could never be defined in a way that had sharp, well-defined boundaries, then my intuition is that the boundaries are fuzzy because they're manmade (e.g "is X red"). I have trouble believing that phenomenal consciousness can ever be defined in a way that gives a sharp, well-defined boundary between a zygote and an adult, because the physical progress from one to the other is so gradual and continuous (I assume that under any reasonable definition, the latter has it, and the former doesn't). I'm not claiming that this is a proof, it just makes it seem implausible to me that "things that possess phenomenal consciousness" can be a well-defined natural category. Where do your plausibility intuitions differ from mine?

Gilbert Wesley Purdy said...

"That helps, but I'm not sure you aren't using terms differently from me in a way I don't understand."

Both in ways I realize and in ways I don't, I'm sure. Some of the difference is surely due to the fact that you have these conversations daily with a group that has developed a more or less specialized vocabulary. Some, on the other hand, is due to my looking at consciousness from quite a different perspective.

"Is this just a way of saying that consciousness is different from how we judge it to be (a statement with which I would agree)?"

While I am saying that, I am not just saying that.

"What this implies, exactly, about the functional role of consciousness is unclear, though."

I look forward to running an alternative theory by you (a theory which I've adhered to for years, I would claim, with excellent results) but I must first take care of some reviewing commitments. To get the theory down on paper will, of course, take a bit of time after that.

It's been fun.

Eric Schwitzgebel said...

Autumnal: I'd say my intuitions are the opposite of yours. Most of the world is fuzzy, except where we work to create sharp borders (and except maybe at the quantum level). My philosophy of science is straight-up Cartwright and Dupre.

Gilbert: I look forward to seeing that paper!

Autumnal Harvest said...

I also think "most of the world is fuzzy." But it my case, it's because I think almost all categories are man-made, rather than natural.

Eric Schwitzgebel said...

How about biological species?

marty monteiro said...

I wonder whether we cannot make transparent the mind in the 'science of consciousness". My book is just released titled Model Of Man: Mind & Matter - Mind & Morality. Look for comments and TOC
http://martymonteiro.wordpress.com or
http://marty-monteiro.blogspot.com or
http://www.strategicpublishinggroup.com/title/ModelOfMan.html