Sunday, December 12, 2010

Quote of the Day

Chiefly, however, determinism is self-stultifying. If my mental processes are totally determined, I am totally determined either to accept or to reject determinism. But if the sole reason for my believing or not believing X is that I am causally determined to believe it I have no ground for holding that my judgment is true or false. J. R. Lucas has put the point cogently with reference to Marxist and Freudian forms of determinism thus. 'The Marxist who says that all ideologies have no independent validity and merely reflect the class interests of those who hold them can be told that in that case his Marxist views merely express the economic interests of his class, and have no more claim to be judged true or valid than any toher view. So too the Freudian, if he makes out that everybody else's philosophy is merely the consequence of childhood experiences, is, by parity of reasoning, revealing merely his delayed response to what happened to him when he was a child.' Lucas then makes the same point with regard to a person who maintains, more generally, that our behaviour is totally determined by heredity and environment. 'If what he says is true, he says it merely as the result of his heredity and environment, and of nothing else. He does not hold his determinist views because they are true, but because he has such-and-such a genetic make-up, and has received such-and-such stimuli; that is, not because the structure of the universe is such-and-such but only because the configuration of only one part of the universe, together with the structure of the determinist's brain, is such as to produce that result.'

The exact force of this criticism is sometimes missed. Certainly on deterministic premisses determinism may be true. But we should not have any grounds for affirming that it is true or therefore for knowing that it is so. In order to obtain these grounds we must be free from all determining factors in order to assess the evidence according to its own worth. This principle applies to the assessment of all truth-claims (including those of Christianity). Freedom from determining factors is therefore required in the cognitive as much as in the moral sphere.

Huw Parri Owen
Christian Theism: A Study in Its Basic Principles
(quoting J. R. Lucas, The Freedom of the Will)

17 comments:

Hiero5ant said...

But if the sole reason for my believing or not believing X is that I am causally determined to believe it I have no ground for holding that my judgment is true or false.

I'm not sure if this is supposed to be offered as satire, but on the off chance it's not, it's one of the more obviously false things I've heard all month.

Consider that the antonym of "determined with respect to inputs" is "random with respect to inputs".

I would much rather trust a thermometer which was determined with respect to temperature than one which was random with respect to temperature. Wouldn't anyone?

In order to obtain these grounds we must be free from all determining factors in order to assess the evidence according to its own worth.

I for one prefer persons who make assertions which are nonrandom with respect to their data. As far as "the evidence's own worth", I'm not sure that means anything at all. If the evidence of my engine misfiring came toe-tagged with a bright yellow sticker saying "ignition cables", I wouldn't've dropped all that cash on my mechanic. Who (hopefully) does not issue pronouncements randomly with respect to prior misfiring engines.

Jim S. said...

You're missing the point of the argument. A thermometer which was determined with respect to temperature would certainly be reliable, if its reliability was independently ascertained (otherwise, how would we know it's reliable?). If our only method of achieving such ascertation was itself determined, we would require independent ascertation of this method -- and so on. So we're faced with an infinite regress where nothing is ultimately ascertained. The only way to stop this is to affirm that some step along the way is not determined.

Your claim that our only two options are determinism or randomness is false. The claim being made here is not that our thought processes be random, but that they be self-determined, not determined by factors entirely external to us. If they were, we would not be able to independently ascertain their reliability.

Finally, this argument has a very illustrious tradition behind it. Recent proponents include Karl Popper, William Hasker, Norman Malcolm, and many others. Before you smear their claims as "obviously false", you should probably see what they say, since they deal extensively with precisely the type of objections you've raised.

Ignorance said...

I really like this quote, thanks for sharing.

I think the argument can not be discarded as quickly as Hiero5ant did, because the mutual exclusivity of various human beliefs would be quite a problem for that rejoinder. If thermometers would vary as much as human beliefs, I think one would be well entitled to send them back to the sales(wo)man or to the producer!

Timothy Mills said...

I really don't see how determinism comes into it at all. The reliability of our beliefs is based on how they were caused, not whether they were caused. The real question is this: does the causal link between the phenomenon and our belief about the phenomenon preserve truth?

My approach to truth is an admittedly vague notion of correspondence. I have not done any philosophical investigation of this approach's philosophical merit, so feel free to point out any problems with it.

To respond directly to the quote you give:

If what he says is true, he says it merely as the result of his heredity and environment, and of nothing else.

In one (rather trivial) sense, this is true: if one's beliefs are determined, then the beliefs could not have been otherwise given the same causal antecedents. But, as I said above, the point about justified belief is not about the presence or absence of causality, but about its structure. Is the structure of your brain such that it tends to respond to stimuli by constructing true beliefs (as the thermometer)? Or is it such that its response to stimuli is not systematically correlated with constructing true beliefs?

How do I know that my mind tends to respond to stimuli by constructing true beliefs? Well, the beliefs I form seem to work when I act on them, so it seems reasonable to go with a working hypothesis that they are true beliefs. Often enough, they do not work - thus I have good evidence that my mind is not a perfect truth-preserving thing. Both the general success and the occasional failure are explained, in a more distal causal way, by the mechanistic processes of evolution.

If this seems unsatisfactory to you, that's fine. Perhaps you could explain how denying determinism buys you a better explanation?

(I should probably reiterate my occasional caveat: I am not actually convinced that our existence is entirely deterministic, though I do suspect it is completely material. I simply think that questions of determinism are given far too much weight by people raised in a tradition that elevates libertarian free will to a central role in the idea of personal identity.)

Hiero5ant said...

You're missing the point of the argument. A thermometer which was determined with respect to temperature would certainly be reliable, if its reliability was independently ascertained (otherwise, how would we know it's reliable?).

Surely things can be reliable independently of our ability to ascertain whether they are reliable; do you see how the above slips between the two notions?

And conceding that determinism is not in contradiction to reliable beliefs pretty much gives away the store. All you're left with is a kind of vanilla skepticism which "self-determination" as such is powerless against. Does anyone suggest that there is no such thing as bias, or that any given person is immune from bias? Do you really have to be a card-carrying Marxist Materialist to posit that the opposition by millionaires to a tax on millionaires might be driven in part by economic self-interest?

Everyone I know operates as though they are the unbiased ones, and people who disagree with them are “in thrall” to this or that social or economic or sexual pressure. But most people in their thoughtful moments will affirm that they probably are biased, and seek to minimize it.

If our only method of achieving such ascertation was itself determined, we would require independent ascertation of this method -- and so on. So we're faced with an infinite regress where nothing is ultimately ascertained. The only way to stop this is to affirm that some step along the way is not determined.

Of Course! Flipping a coin somewhere along the line is much more truth preserving!

Your claim that our only two options are determinism or randomness is false. The claim being made here is not that our thought processes be random, but that they be self-determined, not determined by factors entirely external to us. If they were, we would not be able to independently ascertain their reliability.

I used to think I knew what “self-determined” meant, and I think you think you know what it means, but I don’t think so. “Non-random with respect to itself”? Huh?

Once I tried to express exactly what I thought I meant, I found myself confounded. Do you know anyone who you believe is “self-determined”? Here is a thought experiment then. Imagine you show them five red apples and ask them how many apples there are. Can you predict their response with any degree of confidence? What pattern, specifically, would I expect to see from a “self-determined” system that would allow me to differentiate it from a system that was either random with respect to its inputs, or (at least weakly) predictable with respect to its inputs?

Timothy’s “The reliability of our beliefs is based on how they were caused, not whether they were caused” is a good sloganized summary of this view. One starts feeling the intrinsic obviousness of this when one tries to operationalize “self-determined”. Try it and see!

Ilíon said...

"Your claim that our only two options are determinism or randomness is false. The claim being made here is not that our thought processes be random, but that they be self-determined, not determined by factors entirely external to us. If they were, we would not be able to independently ascertain their reliability."

"Randomness" as the cause of beliefs or thoughts -- which would be as much slavery as deterministic cause-and-effect is -- is the only way determinists can "account" for freedom.

Jim S. said...

Hiero5ant, you're still missing the point of the argument. The claim being made -- at least with this particular quote -- is not that determinism precludes a belief being true. Acknowledging this is not giving away the store, since virtually all epistemologists recognize that "true belief" is not sufficient for something to qualify as knowledge. The claim being made here is that determinism could only allow for true belief; as such, it could never account for knowledge. In which case, we could never know anything, including whether determinism is true or not.

Let me put it this way: if you're programmed to believe determinism, that takes away your epistemic right to believe it. You'll continue believing it, because you can't help it; and it may very well be true. But you could never know it's true. You've been programmed. In order to really know it, you would have to be able to judge it independently of your programming; in other words, you would have to not be determined.

As for self-determination, I think this is a fairly intuitive concept. Determinism holds that we are determined. Self-determination holds that we are determiners. Of course this doesn't mean we're free in all possible respects, just that we're not determined in all possible respects.

Ilíon said...

Determinism is 'Bulverism' Through A Looking Glass (and also on steriods). I say "through a looking glass" because it applies to all the determinist sees, himself included.

The "regular" 'Bulverist' asserts that "you're only saying that because you're [X]." That is, while he denies that his opponent has rational reasons for saying or believing the claim the 'Bulverist' denies, but rather that there are only a causes which cause the opponent to say or believe it; and he thereby dismisses the view he wishes to deny.

Nevertheless, the "regular" 'Bulverist' continues to hold that *he* is able (as are those who agree with him) to believe and say what he believes and says for reasons rather than merely due to causes. The "regular" 'Bulverist' continues to hold that *he* is able to rationally comprehend truth; he continues to hold that *he* is able to to possess knowledge.

Or, to put it another way, determinism is 'Bulverism' with that gaping logical inconsistency worked out.

Ilíon said...

... in my previous comment, I misformed the link to 'Bulverism'

Timothy Mills said...

The claim being made here is that determinism could only allow for true belief; as such, it could never account for knowledge.

Rather than trying to refute this rather bold assertion, I am inclined to simply counter-assert: yes, a deterministic framework can account for knowledge.

What are the elements required for a claim to knowledge (over, for example, mere belief, or even true belief)? Here are some of the things that philosophers traditionally suggest must be the case in order for an agent S to "know" a proposition P:

(1) Statement P is true.
(2) S believes P to be true.
(3) S is justified in believing P to be true.

Items (1) and (2) are obviously compatible with determinism. Item (3) hangs on the meaning of "justified", but I would say that reliabilism provides a framwork for justification that is compatible with determinism.

These cover the conventional "justified true belief" model of knowledge. It has come under criticism, notably by Gettier. But the revisions proposed to address the shortcomings don't seem to make any trouble for determinism. For example,

(4) If P were true, S would believe P to be true.
(5) If P were not true, S would not believe P to be true.

All of these speak to structural relationships between the belief and the fact, rather than to any non-determinism on the part of the believer.

So again, I reject your assertion that some non-deterministic "self-determination" is necessary for a claim to "knowledge". (Note, though, that the approach to knowledge I'm describing does not presuppose or require determinism to be true.)

Timothy Mills said...

"Randomness" as the cause of beliefs or thoughts -- which would be as much slavery as deterministic cause-and-effect is -- is the only way determinists can "account" for freedom.

Ilíon, I agree that randomness buys us nothing in the way of meaningful freedom. But you are really failing to see things from a compatibilist standpoint. Compatibilists - who may or may not be determinists - find the important elements of free will in things which are compatible with determinism: things like the ability to act on our own desires (whatever the "ultimate" causal source of those desires). You may disagree that this is an important or central element of "free will", but you cannot deny that this move expresses a definition of free will that is compatible with determinism.

I find it a perfectly satisfying and intuitively compelling approach to free will. Although I do not commit myself to a deterministic worldview, I happily call myself a compatibilist.

On the other side of the coin, I would like to reiterate Hiero5ant's challenge to operationalize "self-determined". Can someone address this, either in this discussion or through links? So far, it looks like mere hand-waving that doesn't actually achieve any explanation.

Ilíon said...

"Ilíon, I agree that randomness buys us nothing in the way of meaningful freedom. But you are really failing to see things from a compatibilist standpoint."

Indeed! I do my best to never see things from an inherently incoherent or self-constadictory point of view.

Hiero5ant said...

Let me put it this way: if you're programmed to believe determinism, that takes away your epistemic right to believe it.

I know that’s the claim, but I don’t see a good argument in support of it beyond a sort of emotive distaste of the notion. (“You mean she was a robot all along? How horrid!”) Epistemic “rights”, like any other normative claim, are expressive rather than descriptive, so saying that someone lacks a “right” to a knowledge claim reduces to your own subjective inclination (not) to lend it credence. But why would you do that if you had good reasons to believe that the pressure waves coming out of an upright mammal’s mouth bore a nonrandom relationship to the truth? That’s like cutting off your epistemic nose to spite your metaphysical face!

Recall my assertion that no one knows what they mean when they say “self-determined”. Have you tried to operationalize your concept? What were the results? Isn’t it the case that when you try to spell it out, knowledge is a subset of processes which form beliefs nonrandomly with respect to truth? (Specifically, you have non-reasons-responsive nonrandom information systems like pregnancy tests being nonrandom with respect to pregnancy, and reasons-responsive nonrandom systems like people.)

You'll continue believing it, because you can't help it; and it may very well be true. But you could never know it's true. You've been programmed. In order to really know it, you would have to be able to judge it independently of your programming; in other words, you would have to not be determined.

In other words, you would have to make a decision whose output is random with respect to whether it’s true or not! I’m in the habit of not referring to the outcome of those kinds of processes as reliable knowledge.

But you’d have to get clear on exactly what you’re doing when you say the meteorology major who spends each morning reviewing the satellite data and has a remarkable track record of making true predictions nonetheless “has no right” to claim to know it will rain. Does it mean you don’t believe him? If not, why not? If you do, then what does “you have no right” amount to in practice? I’m pretty sure this is a purely noncognitive designation. Yay human! Boo, robots!

As for self-determination, I think this is a fairly intuitive concept. Determinism holds that we are determined. Self-determination holds that we are determiners.

The one hardly contradicts the other. It’s like saying you can’t be a father if you are also a son. Saying “he was The Determinator” is just giving a rhetorical pat on the back to a certain sort of nonrandom information transmission process from data to linguistic output. I remain unconvinced that anyone can describe a set of inputs and outputs which is neither random nor nonrandom.

Also, get well soon. Holidays where you're too sick to drink too much and wake up sick the next day are never fun :(

Timothy Mills said...

Indeed! I do my best to never see things from an inherently incoherent or self-constadictory point of view.

Ilíon, as I am a compatibilist, I obviously do not (currently) find the position incoherent or self-contradictory. Perhaps you could expand on your claim.

Ilíon said...

"Ilíon, as I am a compatibilist, I obviously do not (currently) find the position incoherent or self-contradictory. Perhaps you could expand on your claim."

No. For there is no way that anyone can explain that "A != !A" to anyone determined to not see it.

Timothy Mills said...

Ilíon (and everyone), I would like to probe this issue further, if you're willing. After all, it is a rather important metaphysical issue.

It is certainly possible that you have some insight that I lack. In the spirit of cooperation, I offer an explicit formulation of my view of free will. This way, you can more easily point out any contradictions.

I see compatibilism as asserting the following:

1. Free will refers to one's ability to act as one intends to act.

2. If a person intends to perform an action, is free from constraints that would prevent them from performing that action, and so actually does perform that action, then the action was an exercise of free will.

3. One's intention to act in a certain way may or may not be completely predetermined by other factors.

4. The causal antecedents of an intention to act are immaterial to whether the act was free; all that matters in the exercise of free will is the liberty to translate an intention into action.

5. Specifically, the metaphysical possibility that, given the exact same physical situation, the person might have acted otherwise, is unnecessary to the exercise of free will.

Points 1 and 2 offer a definition; points 3, 4, and 5 outline how this definition is compatible with determinism.

Timothy Mills said...

To elaborate on point 5 from my perspective as a metaphysical naturalist, suppose that intentions are physically rooted in the brain. Then the exact same physical situation would include the exact same intentions. It would be odd (perhaps contradictory) to say that free will depended on one's ability to act against one's intentions.

In my experience, the above definition aligns fairly well with common uses of "free will" as a concept. Generally, what seems to be important to individuals when they talk about free will is that they are able to act in accordance with their intentions. Counterfactuals tend to be used to discuss different scenarios (different facts, beliefs, or intentions), rather than the crucial "if everything were exactly the same I might have done otherwise" (the metaphysical counterfactual that libertarian definitions of free will depend on).

I find this definition more appealing than the libertarian definition of free will for a couple of reasons:

A. It does not depend on dubious metaphysical presuppositions. Libertarian free will absolutely requires that the world be non-mechanistic. (That is, it requires that there are causal factors besides deterministic laws and random chance.)

B. Moral responsibility seems to be a more tractable concept within a compatibilist framework than within an incompatibilist framework. To quote the Wikipedia article on compatibilism: "The Compatibilist might argue that determinism is not just compatible with any good definition of free will, but actually necessary. If one's actions are not determined by one's beliefs, desires, and character, then how could someone possibly be held morally responsible for that action?"

I acknowledge that the compatibilist definition of free will is not satisfying to everyone. Many people seem to have a reflexive insistence that the libertarian definition is the only appropriate one. That doesn't bother me. I'm glad that there are multiple hypotheses competing in the marketplace of ideas.

But your claim that compatibilism is inherently self-contradictory seems unsupported to me. I hope that, if it is true, you would point to the contradiction. It may be that I am more open to changing my mind than you think. But even if I'm not, it can't hurt to share your reasoning, to help anyone out there who has not yet irrevocably committed themselves to an untenable position.