Thursday, October 10, 2013

Oerter and the indeterminacy of the physical


Many readers will recall some worthwhile exchanges on causality and motion that I had some time back with physicist Robert Oerter.  (You’ll find my contributions to our discussion here, here, and here.  Oerter exhibited a lapse in judgment more recently, but we should forgive that.)  In a recent post, Oerter comments on James Ross’s argument for the immateriality of the intellect -- an argument Ross put forward in his Journal of Philosophy article “Immaterial Aspects of Thought” and his book Thought and World, and which I have developed and defended at length in my ACPQ article “Kripke, Ross, and the Immaterial Aspects of Thought.”  What follows are some remarks on Oerter’s remarks.

Ross’s argument, to simplify greatly, is this: (1) All formal thinking is determinate, but (2) No physical process is determinate, so (3) No formal thinking is a physical process.  By “formal thinking” what is meant is thinking that conforms to patterns of the sort familiar from mathematics and logic, such as adding, subtracting, squaring, reasoning via modus ponens or modus tollens, and so forth.  Ross offers several considerations in support of premise (1), including arguments to the effect that the premise cannot coherently be denied.  For one thing, to defend a rejection of premise (1) will require making use of the very patterns of reasoning the rejecter denies we ever really apply.  (For instance, you will have to apply forms of reasoning like modus ponens in an argument to the effect that we never determinately reason according to modus ponens.)  For another thing, even to deny premise (1) requires that one determinately grasp precisely the patterns one denies we have a determinate grasp of.  (For instance, you have determinately to grasp what modus ponens is in the first place and how it differs from other patterns of reasoning in order to go on to deny that we ever determinately grasp what modus ponens is.)

In defense of premise (2), Ross draws on a number of thought experiments from contemporary analytic philosophy, including Kripke’s “quus” paradox and Quine’s “gavagai” example.  I’ve discussed Kripke’s paradox here at the blog, though for the most detailed exposition and defense of this and the other elements of Ross’s argument, you need to read the ACPQ article.  But if you haven’t done so, you might find it useful to read at least the earlier post on Kripke before continuing, since I will make reference below to points made therein.

“Determinacy” has nothing to do with “determinism”

Oerter makes two main points, which I will consider in reverse order.  Alluding to our earlier exchange about causality, he writes:

I have to say it is exceedingly odd to see Feser defending physical indeterminism here.  In our discussions of quantum mechanics and causation he argued strenuously that there is no such thing as physical indeterminism - not even in the case of quantum mechanics (where nearly all physicists accept fundamental indeterminism).  So I'm wondering how Feser can square real, physical and logical indeterminism with the principle of causality.

But Oerter is confused here on two counts.  First -- and not really relevant to the current subject matter -- in the post Oerter is citing I did not “argue strenuously that there is no such thing as physical indeterminism.”  On the contrary, I explicitly denied that defending the principle of causality (which is what I was doing) had anything to do with “determinism” in the sense of the view that all events are necessitated by antecedent events and laws of nature.  Oerter is making the mistake, common among physicists who write about philosophy, of conflating “causality” and “determinism.”  They are simply not the same thing, certainly not as philosophers, especially Aristotelian philosophers, use the term “causality.”  Deterministic causality is at most one notion of causality among others.

But as I say, that isn’t really relevant to the subject at hand, which brings us to Oerter’s second confusion.  When contemporary philosophers like Ross, Quine, et al. talk about “determinacy” and “indeterminacy” vis-à-vis the content of thought processes and language, what they are saying simply has nothing to do with the debate over determinism and indeterminism vis-à-vis physical causality.  Suppose Mitt Romney said last year “I am determined to win this election!” and someone replied “That’s exceedingly odd -- I thought that Romney, as a Mormon, believed in free will, and now he’s saying that his actions are determined!  Maybe he’s a compatibilist?”  Obviously such a response would fail to see that “determined” doesn’t mean the same thing in every context.  Oerter’s making a similar mistake here.

In fact, Ross, Quine, Kripke, et al. would say that the physical facts are “indeterminate,” in the specific sense they have in mind, even if it turned out that “determinism” of the causal sort were true.  So what sense do they have in mind?  Here’s a very simple example to illustrate it.  Consider the symbol: Δ  It has a number of physical features, such as being black, having three straight sides, having a certain size, etc.  Now, what exactly is it that Δ is a symbol of?  Does it symbolize triangles in general?  Black triangles in particular?  A slice of pizza?  A triangular UFO?  A pyramid?  A dunce cap?  Your favorite Kate Bush video?

There’s nothing in the physical properties of Δ that entails any of these interpretations, or any other for that matter.  The physical properties are “indeterminate” in the sense that they don’t fix one particular meaning rather than another.  The same is true of any further symbol we might add to this one.  For example, suppose the sequence T-R-I-A-N-G-L-E appeared under Δ.  There is nothing in the physical properties of this sequence, any more than in Δ, that entails or fixes one particular meaning rather than another.  Its physical properties are perfectly compatible with its signifying triangles themselves, or the word “triangle,” or some weird guy who calls himself “Triangle,” or your favorite trip hop acid jazz, or any number of other things.

What Ross, Kripke, et al. are saying when they say that the physical is indeterminate is that no collection of physical facts, and indeed not even the entirety of physical facts, entails any particular meaning rather than another.  That would include all the facts about deterministic causation, if causal determinism turned out to be true.  For example, even if it so happened that every single time anyone saw Δ or T-R-I-A-N-G-L-E he were rigidly causally determined to utter “That definitely represents triangles, and not a slice of pizza, or a UFO, or some oddball acid jazz music!” there would be nothing about the physical properties of that sequence of sounds, or of its causal relations to any other collection of sounds, brain states, bodily motions, etc., that would by itself entail that the meaning it has is the one we would naturally tend to associate it with.

It is important to emphasize that there is nothing essentially anti-materialist or anti-physicalist about this claim.  Ross is a critic of materialism, but many prominent philosophers who are materialists or physicalists -- Quine, Daniel Dennett, Bernard Williams, Alex Rosenberg, and others -- have taken precisely the same view.  They hold that for any collection of physical facts, no matter how large, there is nothing about them that entails one specific meaning rather than another.  Not only could a materialist agree with Ross’s premise (2), many materialists do agree with it.

Metaphysical not epistemological

Oerter’s second point is the suggestion that for all Ross has shown, the indeterminacy to which he calls our attention is really only epistemological rather than metaphysical.  Take Oerter’s example of a mechanical computing device and the question what function it is computing.  While there may not be (Oerter allows) any way to know from the physical facts alone which function it computes, that does not entail that there isn’t a fact of the matter, metaphysically, about which function it is computing.  But physicalism is challenged only if we have a metaphysical rather than merely epistemological indeterminacy.

It seems that Oerter has not read my ACPQ paper, wherein I address this sort of objection.  In fairness to Oerter, Ross does say things in “Immaterial Aspects of Thought” that give the impression that some of the considerations he adduces have more epistemological than metaphysical significance.  (For example, Oerter quotes a passage that makes reference to incompatible predicates that are each “empirically adequate.”)  That implication is gone in Ross’s restatement of his argument in Thought and World, and Kripke makes it quite clear that the issue is not an epistemological one.   

Oerter supposes that when Ross says that the behavior of a machine is indeterminate vis-à-vis the function it is computing, it is the past behavior of the machine that is in question.  And he seems to think that this leaves it open that some future behavior, or even just possible behavior, would in principle determine what function it is computing.  The problem is just that our information is limited to the past behavior. 

But that is not the problem.  For not only is any set of past behaviors consistent with incompatible functions, but any set of future behaviors is consistent with them, and indeed any set of possible behaviors is consistent with them.   As Kripke points out, you might think that melting wires or slipping gears count as malfunctions, but relative to an eccentric program they might count as the machine doing exactly what it is supposed to be doing, whereas if the wires failed to melt or the gears failed to slip, that would be (relative to such a program) a malfunction.  Either way, the physical properties of the machine won’t tell you, no matter how long its operations continue or could continue.  Take the complete list of physical behaviors a given machine does exhibit or could exhibit -- a calculator’s outputs, the words and images on a computer screen, the noises a robot makes, or even a machine sputtering, melting, or emitting smoke and sparks.  There are always going to be alternative incompatible programs (even if eccentric ones like a program for computing Kripke’s “quus” function) that the machine’s behavior is consistent with.  You could take such-and-such behavior as a malfunction in a machine that was running program A, but it could also -- for all any collection of physical facts could in principle entail -- be a machine that is functioning properly as it runs program B. 

Of course, you could ask the programmer himself what the machine is supposed to be doing, but that just reinforces the point that the physical facts about the machine itself can’t tell you.  And of course, if the materialist claims that the programmer’s mind is itself just a program running in his brain, then appealing to his claims about what the machine is doing really just kicks the problem back a stage.  For now the problem is that we need to know what program his brain is really running, and any possible set of physical behaviors he exhibits -- including his speech behavior, considered just in terms of its physical properties, and including anything going on inside his brain -- is compatible with alternative incompatible programs.  Nor will appealing to natural selection help, since for any program we conjecture natural selection has put into us, there is going to be an alternative program with equal survival value, and the biological facts will be indeterminate between them.  There will be no reason in principle to hold that it is the one program that natural selection put into us rather than the other.  (Again, see the post about Kripke and computationalism cited above.)

Quine, Dennett, Rosenberg, et al. -- who, remember, are not attacking physicalism but rather drawing out the implications of the physicalism they accept -- have made points just like these ones themselves, and thus embrace indeterminacy.  The dispute between these physicalists on the one hand, and Ross and myself on the other, is whether they can do this coherently.  Ross and I argue that they cannot.  But at least they see that indeterminacy is a bullet a consistent physicalist has to bite. 

68 comments:

Untenured said...

This is almost too nice of you, Ed. Equating "indeterminacy" with causal "indeterminism"? That is like assuming "intentionality" means "intending" to do something. These kinds of rudimentary mistakes are infallible signs of dealing with a rank amateur.

But, then again, Keith Parsons could successfully pose as a heavyweight in the philosophy of religion without understanding what a brute fact is, so I guess its par for the course these days.

Vincent Torley said...

Hi Ed,

Thanks for this very useful summary of your argument against materialism. Two quick questions:

1. Granting that no collection of physical facts could ever entail a determinate meaning, does it necessarily follow that no set of physical facts could ever entail a determinate function, as you seem to be claiming? The latter strikes me as a very strong claim: I think if we found a knife on Mars, we'd all say it was for cutting, and if we found a calculator with an add button and circuitry that matched ours, we'd say it was for "plussing," not "quussing." What's more, if we found a Martian with a heart and circulatory system like ours, we'd surely conclude it was for pumping blood. In any case, I think that only your first claim - that physical facts cannot entail a determinate propositional meaning - is required, in order to successfully defeat materialism.

2. If a physical process has no determinate propositional meaning, doesn't that imply that thinking (which has a propositional meaning) must be a non-physical process? And if thinking is, then choosing is, too. And if choosing is a non-physical process, then we have to say that this non-physical process somehow causes processes in the brain, which means that "acts of will" sit atop a causal chain.

But if you say that there are no immaterial acts of will sitting at the top of the chain, but brain processes with an immaterial aspect, then surely the same could be said about thoughts, too: maybe the meaning is just the immaterial aspect of a process whose physical aspect is a brain process. But that sounds like neutral monism.

Comments?

Witten said...

I still don't understand what the metaphysical problem with the machine is. Even if even in principle we can't know what a machine is designed to do*, that doesn't mean that it wasn't designed to do something. I'm sure I'm missing something, because to me this just seems to be restating the problem of induction, we can always be wrong.

*of course we totally can.

George LeSauvage said...

@Witten:

Reread the 2nd part of the article. Your comment seems to be identical to that which Feser adresses, that the matter is simply empirical; that we cannot know what the machine is meant for. That isn't his point, which is that whatever it does is itself compatible with any number of incompatible descriptions.

I also wonder if determining the function of a machine is exactly induction. But that is more a puzzlement than an argument. I can see some ways in which induction is relevant ("OK, we've tested the saw for driving nails, and that's not the answer."), but whether induction is the whole story is another matter.

@Vincent Torley:

I think it does sound a bit like neutral monism here, as at other times classical Aristotelianism can sound a bit like Cartesianism, or like functionalism. Most modern theories look as if they tore a chunk out of classical philosophy, and decided to set it up as the whole answer.

Witten said...

I don't understand how one of those things is a metaphysical problem.

"whatever it does is itself compatible with any number of incompatible descriptions."

What does that mean other than: "we can never be certain what a machine was designed to do." It seems obvious to me that we could never be logically certain of something like that.

Tony said...

and if we found a calculator with an add button and circuitry that matched ours, we'd say it was for "plussing," not "quussing."

Vincent, all the calculators I have ever had were limited in the number of digits you could use, most of them would just pop up with an "error" sign. There is no way of distinguishing whether the calculator is carrying out function f or g, where they are defined by

f(x+y)= x + y, or

g(x+y) = x + y for some domains over x and y, and = "error" for other domains over x and y.

It is undeniable that f and g are two different functions, and the machine doesn't tell you which is "intended". Nor can the machine tell you that its most successful use was to sit in a pocket and stop a bullet intended for the president, and its second most happy use was as a paperweight on Al Franken(stein)'s desk helping to make him pretend to be intelligent.

To be a little less silly, suppose a person nefariously designs a calculator to do everything right EXCEPT to give the result 4+5=8, the machine does everything it was designed to do. But the "meaning" of the circuitry cannot be known to be "plus" from its actual state, nor cannot it be determined from the state of the circuitry whether the designer carried out his design correctly and achieved "plus except 4+5" or failed what to create was supposed to be "plus" because it isn't really. What is the meaning of a mistake?

Tony said...

Not to mention that

circuitry that matched ours,

already borrows from the meaning we have given the circuitry.

Jinzang said...

Here's a story about the meaning of a computer program, from the old days.

A junior programmer was asked to document the purpose of the computer program in a lab at Johns Hopkins University. He came across one that he couldn't figure out, it was doing a series of calculations that made no sense. So he asked an older programmer what it was for.

Back in the old days computers had front panels with lights that indicated the contents of the main registers. This was so that if the computer crashed, you could read the state of the computer when it crashed off the front panel and diagnose the problem.

This lab did military research and whenever the brass came to visit, they would run the program mentioned. It would cycle the lights on the front panel in an impressive way, making the military think they were getting their money's worth.

So, yes, you can't tell the meaning of a program from the physical operations performed, as this story shows.

Crude said...

This lab did military research and whenever the brass came to visit, they would run the program mentioned. It would cycle the lights on the front panel in an impressive way, making the military think they were getting their money's worth.

And it wouldn't tell us where to find the last Golden Ticket either. :\

(Explanation for those who don't get it.)

יאיר רזק said...

" “Determinacy” has nothing to do with “determinism” "

I note that, nevertheless, me and Oerter agreed (I think), on the comments to his post, that indeterminism does imply indeterminancy. Given physical indeterminism, no physical system can (perfectly) implement a (non-stochastic) "pure function". So rather than saying there might be some other functions, we can make the stronger claim that it can't be a specific function (such as modus ponens, or addition, or so on).

" You could take such-and-such behavior as a malfunction in a machine that was running program A, but it could also -- for all any collection of physical facts could in principle entail -- be a machine that is functioning properly as it runs program B."

There is a subtle distinction here - you can ask what program the machine is actually running, and that will be B. Or you can ask what program the machine is "supposed" to be running, and that can be A. But these are not the same question.


I am not really familiar with most of the thinkers you mention, but I believe Dennett is an elimintavist or weak emergenist in relation to such topics. I have to say I fail to see how the main argument would apply to such views. Both the premises are wrong as neither thinking nor physical processes are determinate under eliminativism, while under weak emergence premise (2), at least, would fail as the mental supervenes on the physical (uniquely - regardless of how many ways we can assign meaning, the meaning intended by the programmer's brain, as a mental entity, is unique). But of course I haven't delved into the paper or surrounding discussions, so perhaps these already address such concerns.

Cheers,

Yair

Crude said...

Yair,

I am not really familiar with most of the thinkers you mention, but I believe Dennett is an elimintavist or weak emergenist in relation to such topics. I have to say I fail to see how the main argument would apply to such views.

I think the typical reply here is, if someone responds to the argument Ed has laid out with an eliminativist position, then they're off in the land of incoherency anyway.

(uniquely - regardless of how many ways we can assign meaning, the meaning intended by the programmer's brain, as a mental entity, is unique)

'Unique intended meaning' would be determinate meaning, wouldn't it? And Dennett, if I recall, also explicitly rules out any and all intrinsic meaning. All meaning must be derived. And derivations are never unique.

Crude said...

Also,

I note that, nevertheless, me and Oerter agreed (I think), on the comments to his post, that indeterminism does imply indeterminancy. Given physical indeterminism, no physical system can (perfectly) implement a (non-stochastic) "pure function". So rather than saying there might be some other functions, we can make the stronger claim that it can't be a specific function (such as modus ponens, or addition, or so on).

And here, if it's being denied that we're 'really' doing modus ponens... well, I think Ross goes into this.

Debilis said...

I love a lot of the comments here, but I did want to add something:

There is a lot of talk about what one can or can't tell from looking at the physical components of a machine/program. Even if unintentional, this seems to be getting back into epistemology, rather than metaphysics.

I think Feser's real point wasn't that one can't tell what a program is for, but that it literally isn't for anything at all so long as one assumes physicalism.

The Deuce said...

I have to say it is exceedingly odd to see Feser defending physical indeterminism here.

Just want to register a big ole' facepalm at this before reading further.

George LeSauvage said...

@Debilis:

"I think Feser's real point wasn't that one can't tell what a program is for, but that it literally isn't for anything at all so long as one assumes physicalism."

Yes. You said it much better than I. The machine (from a physicalist perspective) isn't doing anything like performing a task. There is nothing but some gears turning, or electricity running through certain circuits. Calling it a "program" is imposing a non-physical description.

Vincent Torley said...

Regarding the possibility of our being able to tell what a machine's function is, it seems we need to distinguish three possible meanings of function: (i) the useful task that it performs, by virtue of its design; (ii) the mathematical function that it picks out, by virtue of the results it gives when used for calculations; and (iii) the use that was privately intended by its maker. Now I agree that (iii) by itself is unknowable, ad that no finite set of calculations can ever narrow down the mathematical function of an adding machine to "plus" (deviant alternatives such as "quus" are always conceivable), but I would argue that (i) is knowable from inspecting the machine's circuitry. If the machine had been specifically intended for "quussing" then it would have been built differently. The proper function of the machine is simply addition within a restricted range of integer inputs - which is why you get an error message when the number you input is too big.

donjindra said...

Ed,

An interesting turn of events.

Personally, I disagree that "physical properties are 'indeterminate' in the sense that they don’t fix one particular meaning rather than another." But if you believe it, it fundamentally undermines your school of philosophy.

When "no collection of physical facts ... entails any particular meaning rather than another," then we can no longer fix a final cause.

If melting wires or slipping gears don't necessarily count as malfunctions in our eccentric program, then bad valves and clogged blood vessels don't necessarily count as malfunctions in our eccentric hearts. The heart's function is indeterminate.

Conjecturing about what "natural selection has put into us" is of no help because biological facts are just more data and just as indeterminate.

In short, we cannot derive final cause from nature no matter how good the data is.

Natural Law, being dependent on these functions, is also indeterminate.

And, unfortunately, any proof of God that relies on any form of causation -- especially final cause -- is rendered indeterminate as well.

Anonymous said...

Physical properties are indeterminate. But then again, AT philosophers don't state that matter only has physical properties. See hylemorphism.

Anonymous said...

But Enlightenment is the Principle of existence, and when there is Enlightenment, then the purpose of existence is obvious. The purpose of existence in manifest form is to Transform manifestation, to Transfigure human existence or the human body-mind-complex (whatever it is altogether), to glorify Being in form.

This world IS itself Unqualified Existence, Being, Consciousness, Intelligence, Love-Bliss, Power, Form, and Beauty.

The ultimate nature of the world and how it is arising is inherently and tacitly obvious, if you remain in a state of total psycho-physical oneness with whatever and all that arises.

To remain in a state of total psycho-physical oneness with whatever and all that presently arises, you must necessarily, and always presently, Realize inherently Love-Blissful Unity with whatever and all that presently arises.

Separation, or total psycho-physical contraction from the world, or whatever and all that is presently arising, is, unfortunately, precisely the first and constant, and inherently problematic thing done by ALL those who make efforts to find out, or to account for, how the world is arising, and What Is its Ultimate Nature.

Separation or total psycho-physical self-contraction (better known as sin) is the first and foundation gesture made by anyone who has a problem, or who is seeking, or who is making an effort to account for anything whatsoever.

The entire edifice, or tower of babel/babble, of Aristotlean-Thomistic mind-games is a classic example/product of the sinners mind.

The same is the case with all Christian philosophy, so called "theology", and all secular/academic philosophy too.

יאיר רזק said...

"I think the typical reply here is, if someone responds to the argument Ed has laid out with an eliminativist position, then they're off in the land of incoherency anyway."

Fair enough; but you've just eliminated a lot of philosophers as incoherent...

(As well as myself, I suspect. I'm not clear on what's supposed to be "determinate" and in what sense, so I'm not sure. I'm definitely eliminitavist as to the existence of a metaphysical meaning to the string of letter-forms "triangle".)

"'Unique intended meaning' would be determinate meaning, wouldn't it? And Dennett, if I recall, also explicitly rules out any and all intrinsic meaning. All meaning must be derived. And derivations are never unique."

I think that's different meanings of "meaning". Assuming I recall correctly that he maintains that conscious experience is emergent, he'd have to concede a singular subjective intended meaning, regardless of how meaning is constructed in language or so on.

Every specific thinker will have his own views, including very subtle turns and twists. I just don't see how the premises would be generally accepted under eliminativism or emergence.

"And here, if it's being denied that we're 'really' doing modus ponens... well, I think Ross goes into this."

Good. Unfortunately, I haven't read him.

"I think Feser's real point wasn't that one can't tell what a program is for, but that it literally isn't for anything at all so long as one assumes physicalism."

Quite so. I have to say I fail to see how he argues for it, however. Oerter is right that Feser/Ross only argues within epistemology. He even phrases his conclusion as "the physical facts about the machine itself can’t tell you". (That these facts are not restricted to the past is immaterial.)

Of course, I could be missing all the arguments elsewhere that he keeps referring to. And I haven't read.

I don't think "pure functions" - such as logic's rules - are determinate in a mechanistic world, such as that of physicalism. Not in the strong sense of the word that Feser is using. So I agree with his main thesis. I just don't see him arguing for it. Not in this post. I see only the Problem of Undertermination.

Timotheos said...

"This world IS itself Unqualified Existence, Being, Consciousness, Intelligence, Love-Bliss, Power, Form, and Beauty.

The ultimate nature of the world and how it is arising is inherently and tacitly obvious, if you remain in a state of total psycho-physical oneness with whatever and all that arises."

@ anon

There is a contraction between these two propositions. If the world was truly unqualified being, it could not have any becoming, and therefore, any arising.

Unqualified being is that which is just purely actual, and has no tendency to non-being or has any potency. But if there is no potency, there is no becoming.

But if the world is how you claim, then it would both be becoming, since it is arising, and it would be pure being, with no mixture of becoming, a simple contraction.

So, your pantheism or hegelian absolute evolutionism (You seem to be expressing both) will not work, since the world would both be separate and not be separate from God, who is the real unqualified being.

Crude said...

Fair enough; but you've just eliminated a lot of philosophers as incoherent...

Actually, when last I read up on it, eliminative materialists were quite a minority even among materialists. Everything I've read has led me to believe it's a rather fringe position.

Besides - what's wrong with concluding that a lot of philosophers have a view which is ultimately incoherent?

I think that's different meanings of "meaning". Assuming I recall correctly that he maintains that conscious experience is emergent, he'd have to concede a singular subjective intended meaning, regardless of how meaning is constructed in language or so on.

Well, Dennett explicitly denies intrinsic meaning, and if all meaning is derived then the idea of 'singular subjective meaning' seems to go out the window too. If you do think there is intrinsic meaning, that's great. But I also think taking such a position leaves materialism/naturalism behind altogether. (I also recognize that both positions have become so beaten that pretty much anyone can still call themselves such if they really want to.)

Every specific thinker will have his own views, including very subtle turns and twists. I just don't see how the premises would be generally accepted under eliminativism or emergence.

What I've read seems to back up Ed's statement here: "It is important to emphasize that there is nothing essentially anti-materialist or anti-physicalist about this claim. Ross is a critic of materialism, but many prominent philosophers who are materialists or physicalists -- Quine, Daniel Dennett, Bernard Williams, Alex Rosenberg, and others -- have taken precisely the same view. They hold that for any collection of physical facts, no matter how large, there is nothing about them that entails one specific meaning rather than another. Not only could a materialist agree with Ross’s premise (2), many materialists do agree with it."

That's just one man's testimony, but hey, testimony's all that's being offered so far in reply.

Good. Unfortunately, I haven't read him.

He's linked right in the OP, if you wish to.

Quite so. I have to say I fail to see how he argues for it, however. Oerter is right that Feser/Ross only argues within epistemology. He even phrases his conclusion as "the physical facts about the machine itself can’t tell you". (That these facts are not restricted to the past is immaterial.)

I think it's a pretty straightforward metaphysical claim to make, and I don't think your classification is correct. But more than that - who cares about the classification? If it's being denied, then deny it: Explain/argue how the physical facts alone can give us purpose in the relevant sense. (Or so I'd say to someone who believes as much. You don't seem to, so...)

Actually - can you explain how you see the problem of underdetermination being distinct from what you see Feser as arguing for?

Crude said...

Personally, I disagree that "physical properties are 'indeterminate' in the sense that they don’t fix one particular meaning rather than another." But if you believe it, it fundamentally undermines your school of philosophy.

When "no collection of physical facts ... entails any particular meaning rather than another," then we can no longer fix a final cause.


Talking about 'physical facts not fixing any particular meaning' is an analysis of the materialist / naturalist conception of matter. Obviously that's not going to apply to the non-materialist, particularly A-T philosophers who have a pretty different view of matter, etc.

So nope, no problem here.

Step2 said...

And Dennett, if I recall, also explicitly rules out any and all intrinsic meaning. All meaning must be derived. And derivations are never unique.

All meaning is context sensitive. So unless you make a requirement for a unique objective context – which is what formal logic does within the subjective imagination, there is always a different possible meaning.

This subject always makes me think of this song.

Will Dunkirk said...

My difficulty with indeterminism is as follows.

Presumably we know the difference between what is a rock and what is a cat. If we did not know the difference, then we would not be able to succesfully distinguish between rocks and cats. But we can succesfully distinguish between them. This implies that there is something unique and definite about the two beings, such that we can thereby recognize and distinguish betwen them. Now, for a thing to be cat and a thing to be rock, implies that is has certain ascertainable characteristics that make that thing what it is or satisfies us that it is such a thing and not some other. This implies parameters, so to speak, of some sort: i.e., it implies that the thing falls within a certain range - like an equation - that, when within that range, qualifies it succesfully as being this specific kind of thing. Of course, the 'this specific kind of thing' is just a Form. Now these parameters must be limited: there cannot be an infinite number of necessary characterists that must be satisfied because we could never process an infinite number. But, again, we can tell the difference between rocks and cats. Therefore their necessary characteristics are limited.
Again, if being a house required satisfying an infinite number of characteristics, we could never of course finish actually building a house. However, we can and do build and complete houses. Similarly with health or healing: if we could not determine health and healthiness, a doctor could never hope to succesfully produce this in a patient. But doctors do produce this in patients. Therefore things are determinate, and it doesn't matter whether the thing in question is a product of man (e.g. a house) or a product of nature (e.g. rocks and cats): if, in order for a thing to come to exist, it would have to acquire an infinite number of characteristics, it could never come into existence, for the process of producing that thing would go on forever and never be complete. But definite and determinate things are observed to come into (and go out of) existence.

Now doesn't this 'limitness' imply a necessary determinancy in things? And doesn't that imply final causality? For if a thing never achieved an end, wouldn't this imply the impossibility or realizing ends? E.g., wouldn't it necessarily mean neither man nor nature could ever possibly succesfully produce something? As, e.g., a house is in some sense actually contained in the builder? Or healthiness in a doctor? Or (more) cats in cats?

Am I way off track here?

Crude said...

All meaning is context sensitive.

Insofar as you're talking about the letters c-a-t and how it 'means' cat, sure. With regards to intentionality? That's what's under dispute. And Dennett, etc, say it's all derived, with nothing intrinsic to speak of.

Step2 said...

If your only experience of cats is as predators in the wild you do have a different intention from urban dwellers when referencing cats. Even among urban dwellers there are going to be differences based on cultural history and expectation. A person in ancient Egypt viewed cats with reverence and elsewhere they treat cats as food instead of pets. So there are major contextual differences in how cats function and their proper relation to humans. The intrinsic differences are found in their physical characteristics compared to other animals, which is what is claimed to be meaningless.

Crude said...

If your only experience of cats is as predators

You're dealing with a different issue than what the talk about material-under-materialism and intentionality is taking aim at.

יאיר רזק said...

@Crude: I think we’ll be best served by answering your final question:

“Actually - can you explain how you see the problem of underdetermination being distinct from what you see Feser as arguing for?”

This is how I see the discussion: Ross/Feser raise a series of arguments. Oerter replies, his chief claim being that these arguments show a Problem of Underdetermination - meaning that the data isn’t enough to establish which theory is correct (A, B, or so on) - whereas what Ross/Feser need to show is that no theory is correct. It’s “There is no way to know if the computer is running A or B” versus “The computer isn’t running neither A or B”. Feser then replies that there is actually more data, and that even with all physical data the problem of underdetermination remains. This reply notes a minor problem of with Oerter's argument (he's only considering the past), but does not address Oerter's main argument, which still applies.

"If it's being denied, then deny it: Explain/argue how the physical facts alone can give us purpose in the relevant sense. (Or so I'd say to someone who believes as much. You don't seem to, so...)"

Well, the question is to whom the argument is addressed. I don't think it successfully targets physicalists. Here's how I see it:

In the beginning of Oerter’s post, he defined physicalism as the position that everything is reduced to the physical. I later suggested to label this position as ‘materialism’, and I’ll stick to this label. Under materialism, I can see how Feser’s result will lead to his desired conclusion. If there are no facts except physical facts, and that the computer is running A or B isn’t a physical fact, then the computer isn’t running A nor B under materialism.

But ‘physicalism’ is typically defined (as far as I know) as also including what supervenes on the physical. So it also includes facts about what supervenes and such. It’s thus open to the physicalist to claim the physical facts do determine some non-physical facts. For example, a brain in certain states will produce a certain mental feeling, which can serve to anchor a determinate sense of the "meaning" the programmer has in mind.

Of course, the physicalist is also free to deny determination and adopt an eliminative position. I don’t see why, for example, he can’t say that there is no metaphysical truth about whether the computer is “really” doing A or B; both are valid ways of looking at what it’s doing. To my limited knowledge, eliminative approaches are still influential if not dominant in the field, due to the two Churchland's and Dennett's influence (although Dennett himself doesn't commit to any -ism).

Maybe Ross/Feser address these issues in the paper(s). And maybe such solutions are not available to the physicalist when it comes to whether "rational thought" is determinate. But even if that is so, I don't think the outlines argument by itself addresses physicalism; you need to augment it to reach premises the physicalist would accept.

Then again - Feser is versed in the literature so he may be right that the professional physicalists don't reject the premises. I just don't see why they won't reject them; the standard physicalist approaches seem to lead to the rejection of the premises.

As to my own beliefs - I frankly prefer not to argue over my opinions, but they can best be labeled as "panpsychism".

Yair

Crude said...

Yair,

I'm not trying to draw you into an argument over your own views - I just made that comment based on what you said here, and my own recollection. Anyway...

It’s “There is no way to know if the computer is running A or B” versus “The computer isn’t running neither A or B”. Feser then replies that there is actually more data, and that even with all physical data the problem of underdetermination remains.

I'm just not seeing the problem here. If all the physical data does not 'fix' the meaning/intentionality/whatever, and if this is conceded, then it looks like a claim that's going to take things exactly where Feser wants to go.

Maybe the problem shows up in the following portion.

Under materialism, I can see how Feser’s result will lead to his desired conclusion. If there are no facts except physical facts, and that the computer is running A or B isn’t a physical fact, then the computer isn’t running A nor B under materialism.

Alright. Agreed here.

But ‘physicalism’ is typically defined (as far as I know) as also including what supervenes on the physical. So it also includes facts about what supervenes and such. It’s thus open to the physicalist to claim the physical facts do determine some non-physical facts. For example, a brain in certain states will produce a certain mental feeling, which can serve to anchor a determinate sense of the "meaning" the programmer has in mind.

I think part of the problem here are the labels 'materialism' and 'physicalism' - the latter of which is absurdly malleable. The short answer here, as far as I understand, is that if you're saying that this specific brain state/brain process X 'means' (Y), and if you're decisively pointing out the beginning and end of this process or state, then you're just back to good ol' final causes and the like anyway. The description you give is problematic in part because it talks about 'feelings', but my knowledge of A-T is that 'feelings/phantasms' are the sort of thing brains can do (if we reject the materialist concept of brains, etc), but they argue why this isn't going to apply to grasping and dealing with universals.

Tony said...

it seems we need to distinguish three possible meanings of function: (i) the useful task that it performs, by virtue of its design; (ii) the mathematical function that it picks out, by virtue of the results it gives when used for calculations; and (iii) the use that was privately intended by its maker.

Vincent, "useful" for number (i) depends on intention: what sort of thing do we (or might we want?

(ii) assumes that there is a distinct body of "functions" that can be identified by a distinct series of "calculations". How do you know that you have included ALL of the calculation algorithms within the set? What about the function "smash the = sign with a hammer 13 times with 8 pounds of force?" There isn't any limit to these, so you again have indeterminacy. Indeed, what defines a "calculation" and why do you think this object is "for" calculation only and not other meanings?

As far as I can see, you are just borrowing meaning from outside the object and applying it.

Austin said...

You have an interesting taste in music, Prof. Feser.

Greg Johnson said...

off topic post:

http://www.youtube.com/channel/UCxg5qHXuLGMFyjJc7VWICjQ/discussion

could someone help me out with this guy? I go under ClassicPhilosophy being a huge fan of Feser's but I cant seem to make any points without him attacking them.He purports to have found a flaw in Aquinas' First Way; im trying to show his errors, and its proving exceedingly difficult (probably because my explanatory power is exactly zero. Any advice is much appreciated.

Scott W. said...

could someone help me out with this guy?

Here is the best help I can offer: http://imgs.xkcd.com/comics/duty_calls.png

יאיר רזק said...

"I'm just not seeing the problem here. If all the physical data does not 'fix' the meaning/intentionality/whatever, and if this is conceded, then it looks like a claim that's going to take things exactly where Feser wants to go"

What Feser needs to show is "(2) No physical process is determinate". Showing that all physical data dos not fix a meaning simply doesn't achieve that. He needs more.

"The short answer here, as far as I understand, is that if you're saying that this specific brain state/brain process X 'means' (Y), and if you're decisively pointing out the beginning and end of this process or state, then you're just back to good ol' final causes and the like anyway."

I'm sorry, but I just don't see it. If brain state X 'means' Y, the... I don't really that anything follows. I certainly don't see why final causes follow. Even if they do follow, however, I fail to see how this harms the physicalist's case, specifically against Feser's argument: he's still successfully denying (2).

Yair

Crude said...

Yair,

What Feser needs to show is "(2) No physical process is determinate". Showing that all physical data dos not fix a meaning simply doesn't achieve that. He needs more.

Well, you agree that he doesn't need more with regards to the materialist. It's "the physicalist" you think he has a problem with. More below.

I'm sorry, but I just don't see it. If brain state X 'means' Y, the... I don't really that anything follows. I certainly don't see why final causes follow.

What do you take final causes to be?

Even if they do follow, however, I fail to see how this harms the physicalist's case, specifically against Feser's argument: he's still successfully denying (2).

Some of the more thorough and well-read A-T philosophers may want to correct me here, but again: once you're admitting that such and such brain state X 'means'/'points towards' determinate thought Y, then you're right back to formal and final causes anyway. At that point you're 'denying 2' by taking on a conception of matter and mind that the A-T proponent is trying to advance, and which historically was denied by the moderns.

Which is why I said that the labels seemed to be part of the problem here. I think it's entirely possible for 'physicalists' to redefine 'matter' and 'physical' to the point where it's almost indistinguishable from (among other things) an A-T view, just with a somewhat different vocabulary. If so, at that point, successfully denying 2 just doesn't mean all that much.

George LeSauvage said...

If I may commit a faux pas here, I must point out, that to me, "supervenience" seems to come to little more than "Well, 'epiphenomenalism didn't go over very well. What other word can we come up with?'

Scott said...

@George LeSauvage:

"If I may commit a faux pas here, I must point out, that to me, "supervenience" seems to come to little more than "Well, 'epiphenomenalism didn't go over very well. What other word can we come up with?'"

Heh. Well, as regards mental events/phenomena, I see your point, and practically it's not easy to tell the difference.

Strictly speaking, though, someone who says mental properties supervene on physical properties isn't also committed, as the epiphenomenalist is, to denying that mental events/processes have any causal effects on physical events/processes. That is, it could in principle be the case that mental properties supervene on physical properties and also have physical effects. (I don't know what Yair's own view is, of course.)

donjindra said...

"Talking about 'physical facts not fixing any particular meaning' is an analysis of the materialist / naturalist conception of matter. Obviously that's not going to apply to the non-materialist, particularly A-T philosophers who have a pretty different view of matter, etc."


Obviously it *is* going to apply to the A-T philosophers since they get to their philosophy by assuming those same physical facts. They assume their ability to fix function and final cause from those facts alone, and they *must* do this prior to deducing such things as hylemorphism. Otherwise A-T philosophy is an exercise in circularity.

Scott said...

@Yair:

"What Feser needs to show is '(2) No physical process is determinate'. Showing that all physical data dos not fix a meaning simply doesn't achieve that. He needs more."

I'm afraid I don't see why. His argument (like Ross's) is that formal thinking is not a (purely) physical process, and the precise sort of indeterminacy that therefore concerns him is specifically indeterminacy with respect to meaning. Why doesn't "[s]howing that all physical data [do] not fix a meaning" suffice to show that "[n]o physical process is determinate" in that sense?

Crude said...

Obviously it *is* going to apply to the A-T philosophers since they get to their philosophy by assuming those same physical facts.

Not as far as I've read, no - they're operating with a different metaphysical view, a different view of what counts as 'physical' to begin with, etc. There's some overlap in terms of what counts as physical ('weight' would be physical under both a materialist and an A-T view), but it's not total. Which, from what I understand, renders everything else you have to say here entirely off-base.

יאיר רזק said...

@ Scott: "... the precise sort of indeterminacy that therefore concerns him is specifically indeterminacy with respect to meaning. Why doesn't "[s]howing that all physical data [do] not fix a meaning" suffice to show that "[n]o physical process is determinate" in that sense?"

What Feser shows is something like "Given all physical data, it is impossible to infer whether brain state Y conforms to meaning A, or meaning B, or..."

What he needs for (2) is something like "Given all physical and non-physical data, brain state Y does not actually conform to meaning A, or B, or..."

These are quite different. For the materialist, there is no data but physical data so the first point of difference disappears. He then cannot maintain determinancy while at the same time conceding the problem of undertermination.

The physicalist, however, has recourse to another set of facts - facts about supervinience. He can therefore argue that given all physical and non-physical data brain state Y conforms to meaning A, while at the same time conceding that given only all physical data one cannot infer what meaning brain state Y conforms to.

Yair

Robert Oerter said...

Thanks for your comments, Prof. Feser. I have a new reply up on my blog if you care to discuss the matter further.

יאיר רזק said...

@ Crude:

"What do you take final causes to be?"

Explanations in terms of what something is directed to.

"I think it's entirely possible for 'physicalists' to redefine 'matter' and 'physical' to the point where it's almost indistinguishable from (among other things) an A-T view, just with a somewhat different vocabulary. If so, at that point, successfully denying 2 just doesn't mean all that much."

We agree then the physicalist can deny 2, the question is what then remains of "physicalism".

The contrast with AT-metaphysics is clear, for example, in Feser's latest post, where he notes that a soul can persist in mental activity without a physical body, given divine "sustenance" from god. This is in sharp contrast to the core physicalist thesis, that every mental fact supervenes on physical facts.

So I disagree - physicalism remains quite distinct from AT.

Specifically on final causes, note that physicalists general don't deny them. The argument was rather about whether final causes are rare and not fundamental (physicalism), or endemic and metaphysical(AT theism). But even this is a minor issue, not the core of physicalist worldviews. I can certainly see people arguing for, say, a singularity in the future as the "final cause" of our universe, who nevertheless are very much physicalists. (Frank Tipler does something very similar with his Omega Point, only he then adds further assumptions to make his view a (rather unique) theological one. This shows well it's those extra assumptions that make the difference between physicalism and theism.)

Yair

Scott said...

@Yair:

"What Feser shows is something like 'Given all physical data, it is impossible to infer whether brain state Y conforms to meaning A, or meaning B, or...'

What he needs for (2) is something like 'Given all physical and non-physical data, brain state Y does not actually conform to meaning A, or B, or...'

These are quite different."

Yes, they are, but I think they're also each different from what Ed is trying to show.

His argument is not about inference; that's an epistemological issue, and he's at some pains to make clear that he's arguing a metaphysical point. That point is, to phrase it along your lines, something like "Given all physical data, brain state Y is indeterminate as to meaning A, or B, or . . . " And that, so far as I can see, is exactly what he needs for (2).

As I understand your reply (and do please correct me if I'm wrong), you're saying that physicalism in principle allows that supervenient non-physical properties, processes, and/or phenomena could make up the insufficiency here and serve to fix a meaning even though the physical properties, processes, and/or phenomena alone didn't.

If that's what you mean, then I don't think it's right. Supervenience here just means that the physical stuff is sufficient to determine the non-physical stuff; if the physical-plus-non-physical were sufficient for determinacy of meaning, then so should the physical alone be.

Crude said...

Yair,

This shows well it's those extra assumptions that make the difference between physicalism and theism.

It's not 'physicalism' and 'theism' which is relevant here, but 'physicalism' and non-naturalism/non-materialism. And physicalism is entirely compatible with various forms of theism, which the Frank Tipler example oddly enough shows: Tipler flat out regards his view as theistic. He's just one example of that, if a little esoteric.

I also don't think Frank Tipler's Omega Point is a good example of 'final cause.'

The contrast with AT-metaphysics is clear, for example, in Feser's latest post, where he notes that a soul can persist in mental activity without a physical body, given divine "sustenance" from god. This is in sharp contrast to the core physicalist thesis, that every mental fact supervenes on physical facts.

Again, 'God' just isn't relevant to this particular discussion - you don't have to be a theist to abandon physicalism and naturalism. And if someone is arguing that intentionality is built into the physical, if meaning is 'intrinsic' and such-and-such brain pattern intrinsically means X, then materialist and physicalist views fall by the wayside in favor of something else. It certainly sounds like a form of A-T to me.

Sure, someone can just keep redefining 'physical' if they like. I also have no doubt someone can redefine it to the point where the God of classical theism is just 'the ultimate physical cause, the physical first mover'. But at that point, what's it all matter anyway? And I think similar can be said of this issue.

Specifically on final causes, note that physicalists general don't deny them.

Physicalists in general do seem to, at least in the relevant sense. You say physicalists regard final causes as 'rare and not fundamental'. First, if they're not 'fundamental', I'm not sure you're talking about final causes at all anymore. And rare? If brain states are intrinsically directed towards X - and, unless I misunderstand you, this is a view you are defending physicalists on behalf of - then final causes are extraordinarily common and fundamental after all. But if that 'direction' is not fundamental - if it's just derived, an interpretation - then it looks like the physicalist view is going to collapse into the materialist view, which you don't question the relevance of Feser's criticism regarding.

Scott said...

@Yair:

Perhaps more fundamentally, the very sort of supervenience at issue here is exactly what Ross and Feser are arguing against. They're saying that the physical alone is indeterminate with regard to some of the very "mental states" that are supposed by some to supervene on them.

"But physicalists can take supervenient non-physical data into account" isn't a cogent reply to their argument. That argument is about the strictly physical, and its burden is precisely that the "non-physical data" are not simply supervenient on the physical.

Scott said...

(My first paragraph should end "on it" rather than "on them.")

donjindra said...

"Not as far as I've read, no - they're operating with a different metaphysical view, a different view of what counts as 'physical' to begin with, etc."

Quoting from The Last Superstition:

Page 62, "There are objective essences, natures, or forms of things, just as Plato says; but our knowledge of them derives from the senses, and is grounded in ordinary objects of our experience, just as common sense holds."

Page 70, "The final cause of a thing is also the central aspect of its formal cause; indeed, it determines its formal cause. For it is only because a thing has a certain end or final cause that it has the form it has -- hence hearts have ventricles, atrias, and the like precisely because they have the function of pumping blood."

So we can assume that final cause is crucial and is required in understanding anything, even formal cause. Yet while formal cause is grounded in the senses, and everyone can grasp that, where is final cause grounded? What "sense" gives us the ability to "discover" the final cause, purpose or function of an object? How do we apply this ability? How do we know whether or not we're applying the Aristotelian doctrine correctly?

The word, "doctrine," is appropriate because no proof of the four causes is offered. There's a complaint that the modern "rejection of the four causes was a sheer stipulation, an act of pure intellectual willfulness." (page 72) Fine. But since no proof was offered in the first place, no disproof was required.

Yet now that he's thrown down the gauntlet, Feser must show that the four causes were more than sheer stipulation on Aristotle's part. He must show that his ability to discover the function(s) of a calculator are more than an act of pure intellectual willfulness. He must show that his metaphysical position gives him that necessary advantage in finding the true final cause(s), or any final cause, of any object. We'll need clear directions on how to build his compass.

For proof we'll need more than an act of intuition because a physicalist has no need to deny his own intuition. Feser must do this prior to invoking his final cause. Because if he invokes that, it's circular. And since he's already said nothing can be understood without invoking final cause, he's in a bit of a predicament. Nevertheless, the problem must be solved. Until he does that, I don't see how he can claim he's on a better foundation than the physicalist.


In short, the metaphysical claim that an object has one true function is not the same as proving you've found it.

Crude said...

Quoting from The Last Superstition:

I'll quote as well.

Page 112: "On this 'mechanical' picture of the universe as a kind of clockwork, everything that exists in the physical world is made up of (or "is reducible to") purely material parts which by themselves have no goal, purpose, or meaning, and these parts interact with other bits of material stuff according to a stripped-down version of Aristotle's "efficient cause"."

So, like I said - they operate with a different metaphysical view, a different view of what counts as 'physical' to begin with. What's "physical" for the Aristotilean is not necessarily so for the materialist or physicalist. On the flipside, "physicalist" is so loose and broad at this point that it's hardly of much use to refer to anyway.

Yet while formal cause is grounded in the senses, and everyone can grasp that, where is final cause grounded? What "sense" gives us the ability to "discover" the final cause, purpose or function of an object?

In part, the same "sense" that gives us the ability to "discover" the final cause: observation coupled with reflection, logical argument, etc. Doesn't seem all that controversial.

The word, "doctrine," is appropriate because no proof of the four causes is offered. There's a complaint that the modern "rejection of the four causes was a sheer stipulation, an act of pure intellectual willfulness." (page 72) Fine. But since no proof was offered in the first place, no disproof was required.

Actually, the complaint goes far beyond that: it's argued that the modern materialist view is ultimately incoherent, and that attempts by to save it in a physicalist way tend either to collapse into incoherency, or collapse into a broadly A-T view anyway.

And what do you mean by 'required'? As in 'you're allowed to do so, and no one will show up at your house and have you committed'? Sure - but who cares about that? Embrace whatever metaphysical view you want. Just don't ask as if the one you've rejected is disproven unless it actually was - and you're conceding that it wasn't.

In short, the metaphysical claim that an object has one true function is not the same as proving you've found it.

"So what?" is a valid reply here. You're apparently taking the position that a final cause has to be demonstrated with utter certainty on the lines of "2 + 2 = 4", and anything short of that means someone can ignore the view. That seems like a silly standard - we don't demand it in most areas of areas, including science.

If you want to see how he's on a better foundation than the physicalist - watch them try to account for (among other things) the mind, and trying to do so without slipping into either incoherency or a broad A-T view anyway. And personally? I think when the defense amounts to the physicalist trying to prove that, if you turn your head and squint your eyes and ignore the physicalist problems, maybe-possibly the two views break even, Ed's made one hell of an advance anyway.

Scott said...

@donjindra:

"Yet while formal cause is grounded in the senses, and everyone can grasp that, where is final cause grounded?"

In the fact that final causes are necessary if causation is to make sense at all. If causes have effects, as they must if they're to be regarded as "causes" at all, then there must be final causes.

"What 'sense' gives us the ability to 'discover' the final cause, purpose or function of an object?"

The intellect—which is a faculty rather than a sense, but it doesn't have anything to work on without input from sensory perception.

donjindra said...

it's argued that the modern materialist view is ultimately incoherent, and that attempts by to save it in a physicalist way tend either to collapse into incoherency, or collapse into a broadly A-T view anyway.

The materialist view cannot collapse into the dualist A-T view -- not according to what Feser writes above. That's partly what I've been saying.

Either we start with the A-T dualist view, or we arrive there. Those are the only two possibilities.

If we start there, all proofs of that A-T dualism are question-begging. That's all Ross's paper does.

If we arrive there, we had to start from somewhere else. But that "somewhere else" is going to be something like materialism and therefore -- according to Feser -- it will be impossible to advance beyond that stage because we will not be able to make sense out of the only stuff we count as real -- matter.

This is the dilemma.

Scott said...

@donjindra:

"If we arrive there, we had to start from somewhere else. But that 'somewhere else' is going to be something like materialism[.]"

Why?

Scott said...

@donjindra:

"Feser must show that the four causes were more than sheer stipulation on Aristotle's part."

I'm pretty sure Aristotle himself already did that in e.g. Book II of his Physics. (See especially Part 8, where he defends final causes; they're the most controversial item on his list, and they were so in his own day as well.)

Scott said...

@donjindra:

"[T]he metaphysical claim that an object has one true function is not the same as proving you've found it."

Well, I think we've found part of the problem.

First of all, and less importantly, there's no metaphysical claim here that any object has just "one true function." At bottom, there's just the fairly uncontroversial observation that causes have effects and the unaccountably controversial conclusion that they must therefore have them by nature—be in some way "directed" to them. There's no reason an object can't have more than one function or be "directed' to more than one effect.

Second, and more importantly, finding the function of a specific object, say a heart, isn't a matter of metaphysics at all. I hardly think you're likely to deny that our understanding of the heart's function (pumping blood) isn't well-grounded empirically.

(That's why your question—"How do we know whether or not we're applying the Aristotelian doctrine correctly?"—is off-base. Finding the function of an object doesn't involve "applying" a "doctrine" any more than recognizing that a red object isn't also blue involves "applying" the Principle of Non-Contradiction. We're exercising our natural reason in accordance with a principle, not drawing conclusions from that principle or trying to make something conform to it.)

So your statement is technically true, but its relevance to whatever point you're trying to make is far from clear.

Scott said...

Er, "likely to deny that it is."

יאיר רזק said...

@Crude: "It's not 'physicalism' and 'theism' which is relevant here, but 'physicalism' and non-naturalism/non-materialism. And physicalism is entirely compatible with various forms of theism, which the Frank Tipler example oddly enough shows"

Alright, accepted.

We have some minor disagreements, but mostly -

"You say physicalists regard final causes as 'rare and not fundamental'. First, if they're not 'fundamental', I'm not sure you're talking about final causes at all anymore. And rare? If brain states are intrinsically directed towards X - and, unless I misunderstand you, this is a view you are defending physicalists on behalf of - then final causes are extraordinarily common and fundamental after all. But if that 'direction' is not fundamental - if it's just derived, an interpretation - then it looks like the physicalist view is going to collapse into the materialist view, which you don't question the relevance of Feser's criticism regarding."

Even if all human actions are "Intentional", intentionality is still extremely rare in the large picture.

The physicalist need not commit to fundamental intentionality or meaning (the two are not the same). He can maintain a weak emergence thesis where one description is still determinate, while not being fundamental. He can maintain a strong emergence thesis where one "description" is indeed correct metaphysically. Or he can maintain panpsychist position where "meaning" of brain states is borne out of composition rules rather than being fundamental or endemic. (I personally consider the first two appraoches a failure and the third close to my own, but this is besides the point.)

In all these cases, the physicalist can affirm (1) while denying (2).

And I must emphasize again that the physicalist is free to deny determination altogether. Certainly in the strong and unique metaphysical sense Feser seems committed to. Remember many physicalists even deny the self in this sense - their own self isn't a unique things that exists at the metaphyiscal level, but rather a rough and vague collection of particles and void.

"Sure, someone can just keep redefining 'physical' if they like. "

Physicalism maintains that everything is physical or supervenient on the physical. We seem to agree that this thesis is broad enough to avoid Feser's objections. I think we should leave it at that.

יאיר רזק said...

"That argument is about the strictly physical, and its burden is precisely that the "non-physical data" are not simply supervenient on the physical."

Sorry, but I don't see their argument as succeeding to meet this burden. All they show is that we cannot deduce which is the determinate function from the physical alone. This is not the same as showing that the non-physical data - the psychophysical rules - are not of the supervenience kind.

Scott said...

@Yair:

"All they show is that we cannot deduce which is the determinate function from the physical alone."

No, that isn't all they show; the argument isn't about epistemology at all, as you can learn easily enough by rereading the passage headed Metaphysical not epistemological.

יאיר רזק said...

"No, that isn't all they show; the argument isn't about epistemology at all, as you can learn easily enough by rereading the passage headed Metaphysical not epistemological."

Out of respect for you, I reread the section. All I can see is the "there is more data" argument, and an argument from authority. I hope I don't have to waste words on why more data doesn't move us away from epistemology, and why an argument from authority isn't. An argument.

Maybe in the vast literature Feser refers to in his argument from authority there are actual arguments for why the issue is more than epistemological. Maybe there are such arguments in his own paper, too. All I'm saying is that in the arguments presented so far - I don't see any.

Yair

Scott said...

@Yair:

Take the complete list of physical behaviors a given machine does exhibit or could exhibit -- a calculator's outputs, the words and images on a computer screen, the noises a robot makes, or even a machine sputtering, melting, or emitting smoke and sparks. There are always going to be alternative incompatible programs (even if eccentric ones like a program for computing Kripke's "quus" function) that the machine's behavior is consistent with. You could take such-and-such behavior as a malfunction in a machine that was running program A, but it could also -- for all any collection of physical facts could in principle entail -- be a machine that is functioning properly as it runs program B.

That's a (short but, I think, sufficient summary of a) metaphysical argument, not an epistemological one. It's not an argument that there are "more data," it's not an appeal to authority, and it's not an attempt to show merely that we can't deduce what program is being run from the physical facts alone; it's an argument that the physical facts alone are not sufficient to determine what program is being run.

Now, you may think that argument doesn't succeed. But you won't show that merely by saying that there might be supervening non-physical facts that, together with the physical ones, do suffice to determine what program being run. The whole point of the argument itself is that this isn't the case—that physical facts don't determine non-physical facts in this way. You don't show that an argument is bad by saying that it hasn't taken into account the possibility that its conclusion is false.

יאיר רזק said...

"It's not an argument that there are "more data,""

Yes it is. Feser is saying that given more data - about the future and counterfactuals - we are faced with the same problem; the problem of undertermination Oerter already acknowledged we face on less data (the past alone).

"it's not an attempt to show merely that we can't deduce what program is being run from the physical facts alone"

Yet that is what it achieves.

"it's an argument that the physical facts alone are not sufficient to determine what program is being run."

How does it show this ? All that it shows is that there are multiple functions consistent with the data. How do you move from that to "there are no functions that metaphysically determine the data"? It's like arguing that because there are infinitely many ways to reach the digit series 3.14, I didn't actually determine which numbers to just write by the first digits of pi. These are two distinct levels, the epistemic and the actual. The argument never moves above the epistemic.

"The whole point of the argument itself is that this isn't the case—that physical facts don't determine non-physical facts in this way. "

The argument just doesn't reach that far. It isn't really discussing what is "determinate" metaphysically, only the epistemic boundaries on that.

"You don't show that an argument is bad by saying that it hasn't taken into account the possibility that its conclusion is false."

You certainly do show that an argument is bad by saying that it hasn't taken the conclusion's negation seriously. In this case - you can't attack physicalism using an argument whose premises fail if common physicalist theses (emergence) are true. You can certainly attack these thesis; but that's a different matter.

Note that this is just a "proof of principle", showing that physicalism, specifically, has the resources to deny (2). Oerter's main point, that this argument is epistemic rather than metaphysical, is more general.

Yair

Anonymous said...

Do any of those multiple functions come just from the physical facts?

Scott said...

@Yair:

"How do you move from that to 'there are no functions that metaphysically determine the data'?"

You don't.

You've got the wrong end of the stick here. As Anon implies, the argument is that the data don't determine the function, not the other way around.

And that's the metaphysical argument. Since all the physical data together still don't suffice to determine a specific function (like addition), and yet we know a specific function is being implemented (we know we're adding rather than "quadding" or anything else), that function must be determined at least in part by something non-physical.

That's why it's wrong-headed to reply But the non-physical data might supervene on the physical data. In admitting that more than one function is compatible with the physical data, you've already acknowledged that this isn't the case—the physical data aren't sufficient to determine the non-physical. That's the whole point of the blinkin' argument.

donjindra said...

"Second, and more importantly, finding the function of a specific object, say a heart, isn't a matter of metaphysics at all. I hardly think you're likely to deny that our understanding of the heart's function (pumping blood) isn't well-grounded empirically."

I'm tempted to agree. But Feser states "no collection of physical facts ... entails any particular meaning rather than another." This seems to contradict your sentences. I think the theory is that the physicalist is a de facto dualist. He is so because the mere fact that he can detect final cause commits him to hylemorphism whether he admits it or not. But I don't see how this claim can be accepted unless it's shown how hylemorphism gives one that ability. That's why I asked for a proof and why one is necessary before continuing. Until then, your position is indistinguishable from radical relativism.

יאיר רזק said...

@Scott:

I think we're mostly in agreement, actually - we both agree that the "function must be determined at least in part by something non-physical", and the only real argument that I can see between us is whether this is in contrast to the physicalist position or an integral part of it. I still maintain the physicalist can deny (2) precisely by invoking psycho-physical correspondence rules as his "non-physical something".

Incidentally, I've just been exposed to the most impressive such attempt that I've seen yet - the work of Giulio Tononi. He suggests that consciousness is to be understood as integrated information. I note that this implies panpsychism, which he readily admits.

I would very much appreciate any commentary on Tononi's position from an Aristotelian-Thomist perspective - how does his theory relate to the hylomoprhic theory of the soul AT advocates ?

http://vimeo.com/53787308

Cheers,

Yair

Scott said...

@Yair:

"Incidentally, I've just been exposed to the most impressive such attempt that I've seen yet - the work of Giulio Tononi."

Same here; someone else recently posted a link to one of his papers in a thread on this site. I've only had time to glance over it but it does look pretty interesting.

Scott said...

I see comment moderation has been enabled on this thread, so I won't reply further here in order not to make extra work for Ed.