Botched Zombies?
Topic: Mind

Last night I read through Searle's critique of Chalmers in "...Mystery..".  Searle either misrepresented Chalmers pretty bad or just refuses to go into detail so some of his refutations weren't very convincing. Searle is the master of common sense examples and that makes him fun to read. But sometimes you get the feeling he's being sloppy. That could be because he really is being sloppy or he just thinks the subject matter doesn't deserve serious attention.

Searle rejects Chalmers' Zombie argument by basically saying he can imagine a world where the laws of nature are different, yet the microstructure the same, and so pigs fly. That being the case, we can conclude flight isn't physical. This seems obviously wrong. Chalmers' response printed in the book is essentially that in Searle's world, we'd notice a difference, a lot of matter in the air that previously wasn't e.g., pigs flying. What he means by this is that Searle didn't take the thought experiment far enough. The imaginable world for the zombie argument isn't just an alternate reality of this one where the furniture has merely been rearranged. Chalmers means a molecule by molecule exact replica of this world. Me sitting at this very computer typing this very sentence and so on. Searle's response to that was more or less a repetition of his initial rebuttal, but emphasizing everything could be the same physically but the governing laws different. Searle's angle doesn't work even though there's a hint of something I do agree with that I'll explain later.

The exchange was interesting to me because it demonstrates there's even confusion among masters when they're arguing something as subtle as supervenience physicalism. Chalmers' description, which paints two pictures, one like ours exactly and one like ours but with excess matter in the sky brings to mind David Lewis' articulation of physicalism via dot matrix printers. The two worlds are nothing more than the arrangement of the dots. From SEP, the resulting picture created by the dots supervenes exactly on the placement of the dots and likewise everything in the world supervenes on the physical. But pressing this analogy might buy Chalmers more than he can really afford. True, Chalmers didn't invoke Lewis' argument specifically, I am, but his response was basically a three-dimensional version of it. Searle was wrong, according to him, because in a 3-d mapping of the world, there would be more matter in Searle's counterfactual sky than in our sky and the thesis requires one-to-one physical identity. Now this creates for me an unknown in Chalmers' thinking and I'll certainly need to follow up on it. It very well could be that this response to Searle is all that he felt was warranted to expose the most obvious of Searle's mistakes. Maybe Chalmers would have more to say given more space. But this mistake on Searle's part wasn't his most important mistake, and failure to take that up makes it confusing for us lay readers.

Supervenience physicalism could be articulated in dot-matrix style by saying every piece of matter the same accross worlds, that would probably be right. But it doesn't make explicit the fact that <i>all physical laws</i> would also be the same. The analogy is just that, the dots on the paper don't just represent dots of matter, but the laws of physics as well. So Searle imagining a world where the laws of physics violate the laws of physics is a contradiction.

Chalmers for sure has a tactical advantage in going for zombies rather than pigs flying because it's easy to see a pig fly, but ever since Leibniz first posed it, seeing consciousness has been THE problem for philsophers of mind. Alan Turing gives us the best case scenario for intuitively grasping something like consciousness by bringing the problem exactly to our personal everyday level of interaction. Leibniz crushes this intuition by blowing up the brain so big you could walk through it, and then demanding, "Let's see you find the thought now?" Chalmers goes in the other direction to promote the same thesis, take a birds eye view of the whole universe and the personalities just don't shine through. I'm not saying Chalmers is sloppy, he's one of the clearest I've read, I'm just saying the presentations in these thought experiments sometimes cloud our judgement of the actual content.

And this is where I come back to my hint of sympathy with Searle. If you're selling molecule by molecule duplicates of the entire universe, then our focus is taken off the "laws" of those molecules as a matter of presentation even though the content doesn't hinge explicitly on that. Our imaginations are primed for easier agreement. But in the end, Chalmers no doubt believes his world of zombies is the exact physical description, including both molecules and laws, as our own. positing pigs flying just isn't the same.

Searle on Dennett's Qualia
Now Playing: TBE
Topic: Mind

I'm reading Searle's Problems of Consciousness right now. I skipped ahead last night to his fight with Dennett since I've discussed Dennett on a number of occasions. It's Dennett's view on qualia that is of particular interest to me because it's so unusual. But I just can't take his advice all the way. My rejection of Dennett is more or less the same as Searle's, based on Cartesion certainty. The problem is, I'm not sure I buy into Cartesian certainty.

Descartes' "I think therefore I am" sets out to establish the thinking self. Searle (like I'm tempted to) has sort of reworked Descartes' epistemological statement of certainty into a phenomenal one, "I feel therefore I am." It doesn't matter how mistaken those feelings are as the mistake itself captures the phenomenal.

Comparing Pinker, Dennett, Fodor, and Churchland
Topic: Mind

I want to compare some of the ideas of four naturalists/athiests within the philosophy of mind or cognitive science who have also written for the lay audience, Daniel Dennett, Stephen Pinker, Jerry Fodor, and Paul Churchland. A lot of what I said about LOTH and connectionism in the last two posts lay the foundation for this one. Granted, there are no stereotypes and certainly there are subtle distinctions here I'm going to miss, this is going to be pretty darn broad sweeping. Let's start with breaking down the mind into it's two commonly attested components of intentions and qualia.

Dennett is an eliminativist on qualia, as I discussed a few posts ago. Pinker and Fodor believe qualia exist but don't have much to say about it other than it's a mystery. Churchland is a little more specific in that he argues against thought experiments which move from qualia to dualism. Fodor, however, has developed qualia arguments against functionalism (not physicalism).

Dennett and Pinker are both disciples of Fodor's early work on LOTH and the computational model of mind. Pinker and Dennett could probably be regarded as functionalists of some sort while Fodor sees the computational model as breaking down for higher-level cognitive processes. Churchland is an explicit eliminativist towards intentions. Dennett's position is nuanced because his "multiple drafts" theory renders intentions and hence, any computations which compose intentions, merely as an "intentional stance." There isn't an explicit "I" who has intentions, but if you take the sum total of all the various brain processes running together, you get a "center of gravity" that can be studied on an intentional level to make useful descriptions, descriptions that are more meaningful than purely physical ones would be. So Dennett is teased into Churchlands direction but stops short.

On a hardware level, Pinker and Dennett both appear to accept connectionism, or implementationalism since they run the computational version of mind - a serial processor - on top of that hardware. Churchland is a radical connectionist, we can scrap folk psychology altogether, symbol processors and all that, and work on a completely new model of thinking. Fodor is strongly opposed to neural net models though I can't figure out if he'd reject the possibility on a physical level. I think he might because of the intrinsically holistic nature of PDP's but that's a discussion I'm not ready for as there seems to be a lot of uncertainty on what different contributors to the field view as holism, and why.

Dennett and Pinker are pioneers in darwinian psychology which builds on Fodor's work on modularity. Fodor follows Chomsky's nativism thesis which argues that some abilities, such as learning a langauge, are innate. These innate abilities, along with peripheral functions like sight and hearing are inborn. While Fodor doesn't attempt a neurological account of what makes for a module, we have a pretty good analogy I think with computer circuitry. Specifically, with specialized circuitry that we'd now call ASICs. These modules are more or less hardcoded to perform specific functions. Think of the computer chip on your video card doing the work as a function of circuit design as opposed to the CPU on your computer doing the same manipulations virtually, following a software program. It's easy to imagine this kind of specialized function when it comes to sight or hearing but becomes more abstract when positing language, and things drastically beyond langauge. Fodor never believed modularity (and computation) can explain too much beyond peripheral functions and language. Pinker and Dennett believe modules explain just about everything. They opt for "massive modularity", that just about all we do in life arises from some kind of modular process that won out in a Darwinian battle. Fodor has been very critical of Darwinian psychology and this apparently has made Dennett to view him somewhat as an enemy. The lines of battle in my view are largely drawn by the criteria for what can be considered modular. The brain is very complex and perhaps Fodor's criteria are too narrow. But relax those assumptions too much and everything imaginable might be included by virtue of having nothing more than a broad definition. Dedicated circuitry for langauge? Maybe. But for stuff like reading fiction novels? There are just too many easy darwinian stories we can make up and too few computer models which would exceed the domain specific, encapsulated ASIC-like structure of Fodor to secure them.

Finally, it might be interesting to note that Fodor, Pinker, and Dennett are all rationalists as opposed to empericists. I like to expose little facts like this to a skeptic oriented audience because the perception of many skeptics is that empericism rules the day in every way. In fact, the further one goes in the direction of Pinker and Dennett, the deeper the committment to rationalism. Of course, ontologically, these three are  physicalists. Churchland would lean, as connectionists do, away from nativism (and rationalism). He's a functionalist, but of course his model of causal relations is connectionist rather than traditional computational.

 

Connectionism
Topic: Mind
In Cognitive Science, the main competitor to theories based on LOTH is connectionism. Connectionism tries to come up with a more biological-like hardware model for how the brain functions. The brain is just a mass of interconnected neurons, so maybe there is a way to model this in a more life-like way on a physical level. The computer science that later has become the study of "neural nets" is called parallel distributed processing. For a great introduction, see here. Essentially, you have we'll call, a lower level layer of neurons that bring in physical data, or inputs. Then a series of internal neurons that decide what to do, e.g., "fire", depending on whether a certain predefined criteria is met. Rather than being programmed by a formal language, nets must be "trained" and therefore any rules they follow are implicit in that they are distributed accross the entire network. Neural nets are therefore innately holistic. Training a net would consist of something like, feeding a net a picture of a face and now try to get it to pick out similar pictures. This kind of pattern recognition is something neural nets do very well, better than conventional computers. It would seem, that such recognition is more lifelike - how the eyes and brain would actually function. Another advantage of nets that make them seem more like a real brain is the way they naturally degrade as nodes are removed wheras, cut into the silicon on a conventional computer, and the whole thing falls apart.

But nets don't do everyting better. In fact, the key failure of nets is the attraction of LOTH and the classical model, nets don't model higher cognitive functions very well. This is what Fodor calls the systemics problem. A real mind it appears learns the formulas for things like sentences, it gets a system down, so to speak. Recall from my last post I said that LOTH is brilliant because of the insight that representation would have to go beyond just pictures, pictures alone aren't complex enough to model thinking. A net then, would naturally be very good at coming up with the picture part of a representation but not good at coming up with the sentential component. And at this point is where the controversy comes in.

A neural net can actually be trained to do sentential representation LOTH style. Maybe it's harder than in conventional computing, but hey, maybe in the future we'll figure out ways to do it better. In this case, however, a net isn't much of a departure from the classic model, rather it's a way to implement the classical model on a different hardware scheme, one that seems more brainlike on a physical level. Those who take this road are called implementationalists. The interesting school of connectionism though would be radical connectionism. They would say forget about LOTH. But under that option, what then, could be the alternative for explaining intentions, which seem naturally, very sentence like? The alternative is eliminativism. Just do away with intentions. Of course, that's a pretty radical position to hold as folk psychology can no longer be said to be useful. In such a view talking about "John wanting ice cream" has as much to do with the mind as humors have to do with health.

LOTH & Computation
Topic: Mind

Jerry Fodor is one of the most important figures in the philosophy of mind and cogvitive science. In the 70's he came up with this idea to put a "language of thought" or "mentalese" in between spoken language and things in the world. This is both a realist and atomistic theory. Normal words reference thought language, and the thought language references the actual things in reality, representation. At rock bottom then there is a collection of atomistic thoughts that discretely link to actual things in reality.  Atomic thoughts would somehow be combined to form molecular thoughts and ultimately, intentionality. Key to mentalese, is that thought works on the same principles as sentential logic and computation. As I've said before, phil mind usually breaks down to two things, the study of qualia, and of intentions. Fodor's theories laid the groundwork for what's now called cognitive science. The task at hand here ignores qualia and focuses on the problems related to intentions and how thinking actually happens. It's trying to explain intentionality.

LOT gives a good starting point for taking a crack at the mind for a number of reasons. Since I'm not an encylopedia, I'll just list a couple things that impress me as highly important. For one, computation gives us at least a stab at a reasonable idea of what thinking is, and how a mechanical device, such as a brain can think. We can simply view thinking as computation, or recursively manipulating symbols which are representations of reality. Second, Fodor justifies the importance of manipulation by the rules of logic within thought langauge brilliantly. Take a basic first stab at mental representation to be some kind of a vague picture. The problem then, is that mental pictures could never be articulate enough to express complex intentions. Only something that's like a formal language could do that. Further, recursion is the only known process by which we could indefinitely make sentences out of words that are meaningful but that we've never heard before. Rules built in to the mind could be worth millions of pictures.

Fodor has always been cautious in the extent to which he believes his theories can explain the mind. Consciousness is no less mysterious and the kind of thinking that is most typical of humans, abduction, is difficult to explain. He's become enemies with some of the more outspoken and AI hopeful cognitive scientists like Pinker and Dennett.

I think there almost has to be at least at a low level, some kind of a connection between the brain/nerveous system and a computer. When catching a baseball there just has to be some kind of computation involved.  But at higher levels, like consciously working through a math problem, how does that happen? There even seems to me, to be a tension between straightforward computation and what we call "thinking" or maybe "consciousness" in ordinary language. For instance, I seem to be thinking very little when I'm driving home. But if I'm working out a hard math problem, I'm trying to remember rules, figure out different ways I can tackle the issue, and by the time I'm done I might be very hungry - I'm really thinking hard. Where as my friend Tarski from the boards would probably look at the same problem and the answer just might automatically pop into his head, any working out of steps being in retrospect as a means of explanation to someone like me.

 

 

23 Minutes in Hell
Mood:  on fire
Topic: Lectures On Doubt
I was stuck in Walmart last night with the fam. I'm not a big shopper, so I tried to find something in their book section and the most interesting publication I found was called "23 minutes in Hell" by some cross-eyed praiser of the Lord. But OMG, there are 40+ reviews of it on Amazon. 23 Minutes

It's one of these books where you can tell at the outset that the subject matter can be condensed into two pages but somehow they've got to stretch it out long enough to make it a book. So you can skip the lay preaching and his call to the ministry and all that and read the actual account of hell which takes five or ten minutes. First thing that's funny about this account is it's heavily annotated. Virtually every sentence he utters references some Bible scripture or another moron's NDE. Ron Livingston did the same thing with his "Sealed Portion." Some advice for up and coming prophets: follow Joseph Smith's example. Make up your story, publish it, and let others be the scholars and unravel the intricities. Don't show us all the stuff you read prior to your bad dream or episode of sleep paralysis. So what happens. He is taken from his bed, the knowledge that he's a Christian is stripped from him, and he is put in a 10x15x15 brick dungeon with two or three grotesque demons pacing about angry, smelly, and ready to hurt him. They beat him up, tear apart his flesh, throw him against the wall. Then from this enclosed space, he somehow can see outward a mile wide pit of fire. Inside the pit there are stone segregations so each person has their own individual fireplace. It's very important not to let the immates have any communication with each other - they have to be completely alone. I'll admit I was reading fast so I missed whether he was actually thrown in or just about to be thrown in and then Jesus came, gave him his memory back, and sent him away with the ole, "you had to experience this to warn the world" message. Some xtians are luckier than others I suppose. Some get to experience a lot of money, power, and sex and then repent, but had to go through all that stuff to warn the world about it.

The author goes on and on about how deformed the demons were, but I wonder what his standard of comparison was. The demons were faster, stronger, and endlessly superior to humans particularily in physical feats so it sounds to me like their design was rather sophisticated. The author also complains about how thirsty he was and there not being a drop of water anywhere in hell. Of course, human bodies are, as the sand entities in Star Trek put it, "bags of mostly water". And the demons were obviously constituted of water as well. If for no other reason, we know this because their flesh was hanging down, rotting, and smelling. Things don't rot without moisture, sorry. In heaven I suppose, he'd be drinking glacier water all the time and enjoying it with his wife - as he did on earth. Would he also be eating and having sex? Finally, he knows all this stuff about hell "intuitively" as he says, he doesn't know how he knows it, he just does. Like the demons are 1,000 times stronger than humans. That there is no water. That there are endlessly more pits like that one he described was a mile wide, and so on. Let me help you Mr. Wiess. You know these things because you're making up the story. The author of the story knows the answers about the story the characters inside don't.

Dennett's Qualia Arguments
Now Playing: Edited
Topic: Mind

See here for some of Dennett's arguments against qualia. He holds a pretty unique position in that he claims there are no "raw feels" as such, no basic, ineffible apprehensions. He argues by manipulating some of the well-known thought experiments to demonstrate that they don't really work. He then introduces some of his own, to convince us that talk about qualia is inconsistent and confused. Thought experiments according to him leave something to be desired because they don't open the door to the obvious conclusion we're suppose to draw. For instance, in the neurosurgical prank, a scientist re-wires your brain to invert your color spectrum. So you wake up, and apples look blue (or something).¹ According to Dennett, the qualia shift would justify us to conclude qualia neurons have been tampered with, hence supporting their existence. But he argues that the optic nerve (qualia) could be to blame or the memory centers (not qualia) for the 'perceived' mixup, since there's no way to decisively tell, such a scenerio loses its force because our "immediat apprehension" doesn't decide. He also gives an example of two coffee testers who grow tired of coffee, one whose "tastes have changed" (matured preferences) and another whose "sense of taste" has changed (qualia input centers). How to decide which? Again, the difference can either be with the input, "raw feel" or qualia, or within the memory centers. His scenarios all turn on this same possible undecidability. Another example inverts taste so sweet is salty and salty sweet. If the subject compensates, have the qualia themselves changed or have the memories? It seems like a stretch to buy into the plausibility of such a scenario but there might be something to it.

 I remember when I lost my old glasses, and got new ones - with the little lenses as opposed to the big ones - the whole world seemed crazy. I almost felt like I was going to pass out, the optometrist looked like a flat cartoon, the world seemed like it was very, very small. In fact, for weeks afterwords, I miscalculated food portions. I kept thinking I was getting less than I really was. But now, everything seems exactly the same with or without my glasses save the clarity. So if I'm understanding Dennett I have to ask, did my eyes in their ability to directly instill a "raw feel" adjust or did some other facet of my brain kick in to "figure it out" (as my optometrist said would happen) and make me think that the new world I was seeing is the same as the old one? It doesn't matter what the answer is, the point is that my apprehension of the supposed "qualia" is useless in deciding. And that goes against the thought experiments which attempt to show the immediate apprehension making all the difference - telling us something we couldn't have known otherwise. It seems to me, the structure of the argument is along the same lines of other philosophical investigations which cast doubt on the primacy of "facts" as independent of "interpretation" or "evidence" having a primacy above "theory." I'm not saying I completely buy into Dennett here, but I have to say given my leanings towards confirmation holism and the like that it's difficult to dismiss.

¹ I was thrown off here because I think the Wikipedia entry on Qualia is wrong in their explanation. They say regarding the 'prank', "it follows that we are imagining a change in a property which determines the way things look to us, but which has no physical basis." Yet Dennett says regarding the same, "and we later discover, if you like, just how the evil neurophysiologists tampered with your neurons to accomplish this." So how can the prank have no physical basis if neurons are being tampered with - why in fact call it the "neurosurgical prank?" I think the SEP entry on qualia answers this. Color inversion arguments have been apparently offered with varying strength, and lining up Dennett's quote, "It seems to us that the standard verificationist.." with the SEP quotation of the same it would follow from SEP that Block and Fodor were arguing aginst functionalism, not physicalism with the "prank" even though a color inversion argument could (impractically) be made against physicalism. Maybe I'll try to edit it and see what happens. heh.

Bad Theology or Just Bad Manners?
Topic: Lectures On Doubt

Adam Corolla inverviewed Shirley Phelps today on his radio show. She's the daughter, I think, of Fred.  What she wanted to tell the world in this instance was, that the recent Amish girls who were murdered were perfectly deserving of the 'crime' and that there is no difference between the Amish and the Taliban. As outrageous as her position sounds, the difference between Shirley and most Christians (and otherwise religious folks) who understand their own theology is mostly a matter of social grace.  Virtually every incarnation of God damns the greater portion of humanity with a lot of misery and pain and retribution for, ultimately, incorrectly placed beliefs. If religious belief were like the contents of an irritable bowel, then the governing criteria for social acceptability isn't so much the precise constitution of that content, but whether those contents are released publically or privately.

That I believe, is a very real, serious distinction that needs to go into the evaluation of religious rationality. Civilized professionals with a degree of self-reflection typically try to nuance the manifistation of God's hand in the real world and blurr the lines on who might be taking it hard in the end. Even if they have secret hopes or beliefs, or are constrained when pressed to logically commit to a God who's damnation is way overdone and mostly arbitrary, our evaluation of that theology should ultimately take such public resistance into consideration.

 I remember pushing a wealthy and very nice Calvinist once on my mission into admitting that God is going to damn babies, and that the attitude one takes toward that situation is simply, "Well, that's tough for them, isn't it?" If reduced to it's ultimate logical implications alone, this man's doctrine was most likely as despicable as Shirley's. But, "this man's doctrine" was clearly superior in its community expression. When that's taken into account, the victory I felt during the encounter in retrospect should have been deflated a great deal. On the one hand, it might be said that I provoked his thought, yet on the other, perhaps all I did was akin to badgering a well-adjusted non-theist into admitting he has (inappropriate) sexual fantasies.

 

Three Degrees of Physicalism
Now Playing: Edited
Topic: Mind

Why the word "physicalism" anyway? Apprently the move to talk about "physicalism" rather than "materialism" (SEP) is rooted in the need to distinguish between the basic position of monism, in this case what the kind science today or in the future should be able to explain by reference to "things", from other kinds of materialism. Of course, what constitutes physicalism or even the various suggested kinds of physicalism is debatable but for those of us just getting our feet wet, there are three main distinctions.

The strongest thesis of physicalism is Type Physicalism, usually associated with identity theory. For every type of mental event, there must be a type of physical event. The familiar pain = 'C fibre' event holds here. I noted a few days ago that this thesis is probably too strong because of the fact that it's difficult to talk about an animal being in pain that has different neurological structures than a human (multiple realizibility).

Supervenience Physicalism is in the middle - actually, it's a little bit of a different way of talking about physicalism. SP is usually considered "minimal" physicalism, or the very weakest thesis that can qualify as physicalism.  SP asserts that all mental properties necessarily change with an underlying change in the physical, and vice versa. Hence, as discussed in an earlier post, we can't define a zombie with 1-1 physical correspondence that has any kind of a different mental life. Mental properties must pace physical properties. Type physicalism guarantees supervenience.

Finally, there is token physicalism which says that for every actual instance of something, there must be a physical something.  Property dualists are token physicalists. The mind can't exist without the brain, but mental properties can change independent of it - that is, the mental doesn't supervene on the physical. Because dualism is permitted, TP can't be considered minimal physicalism. So for instance, David Chalmers who is a property dualist argues that physicalism is false, and that's perfectly consistent with token physicalism. Functionalists are also token physicalists. Whether or not functionalism meets the criteria for minimal physicalism as defined by supervenience is debatable.

Type/token relate to each other also in the following way. A type is like a dollar. A token is like a particular dollar bill, or a quarter. A token physicalist can get a dollar by a couple of quarters and fifty pennies. Maybe a gold standard analogy applies here. A type physicalist will define a dollar as x amount of gold. A paper dollar or four quarters are worthless, or at least they ultimately aren't actually a dollar.

 

 

quick note

Just got back from a short vacation. Stayed in the Marriott. Man, I've never been pushed so hard in my life to drink alcohol. Everytime I called up room service there was a plea to try some special wine or other kind of alcohol.  The main floor was beautiful, all centered around a bar - I stopped and had a couple shots of Jager. It was the only place I could find where I could get a bottle of water. The top floor had a cool rotating bar also. If you partake of the Marriott ammendities as the Hotel guides you, by the time you get to the little blue book in the top drawer you'll be too drunk to turn the pages.

Another note, since I sometimes plug continental phil here. On the plane back, the woman in the seat in front of me looked like she was correcting college papers. My eyes, bored, found themselves scanning down the pages as she turned them (she held them up high, I didn't have to work for it) and I saw references to foucault and Lyotard. So, ok, I had to read a section here and there. The theme of the three or four papers I got glimpses of seemed to be globalization. Good Lord, it was like "paradigm this" and "paradigm that". There's a very big downside to that tradition.