Relativity and Responsibility

I'm reading Galen Strawson's recent article, "Free Agents", which appears in the same Philosophical Topics issue where Kadri Vihvelin's paper that we discussed a couple of weeks ago is found.  In it he makes this suggestive claim (where 'U-freedom' refers to the sort of freedom that we need to be ultimately morally responsible for our actions, in Strawson's sense):

"The basic argument that U-freedom is impossible is entirely a priori, but there are also extremely strong a posteriori reasons for thinking it impossible.  It seems unavoidable if Einstein's theory of special relativity is anything like correct, for example -- a point that has received surprisingly little discussion in recent debate about free will." (p. 392)

Two questions about this: (1) How might such an a posteriori argument against U-freedom from special relativity go?  And (2) How could such an argument show that U-freedom is impossible, given that the special theory of relativity is, at best, contingently true?

Comments

[Corey Tomsons]

"... a point that has received surprisingly little discussion in recent debate about free will"?

Goodness. There's scads of material on this topic, though he will need to read up on some philosophy of science (mostly philosophy of space-time). Lawrence Sklar, Nancy Cartwright, and Tim Maudlin would be a start.

As for how recent this scholarship is, perhaps he needs to consider that at least since the 1980s philosophers were aware that Special Relativity isn't 'true enough' for it to anchor determinism in the 'real world'.

John Earman in chapter 4 of his 1986 book, A Primer on Determinism (Dordrecht: Reidel), talks about this.

["Q" the Enchanter]

On the second question, I presume Strawson means physical (rather than logical or metaphysical) impossibility. (Whereas the a priori argument would encompass "impossibility" in all the relevant senses.)

In re Corey's comment, I'm sure Strawson is well aware that SR and determinism are incompatible. All he's saying, though, is that SR is also incompatible with U-freedom. I imagine the argument amounts to something like that quantum randomness is (for reasons unstated) in just as much tension with the relevant sense of "freedom" as determinism is.

[Randy Clarke]

A wild guess: perhaps the idea is that in order for me to complete an infinite squence of choices within a finite time, something would have to accelerate to superluminal speed.

[Blake]

My guess is that he means that GTR entails a block universe in which there is no difference between past and future -- both are fixed.

[Corey Tomsons]

Excellent point about the distinction, and I should have taken note of it.

[Eddy Nahmias]

Correct me if I am wrong, but I don't think determinism and special relativity are incompatible, are they? The prevalent interpretation of quantum theory says that quantum mechanics is incompatible with determinism (i.e., it is irreducibly indeterministic), but SR is thought to be incompatible with quantum theory, no?

I suspect Strawson thinks SR suggests a block universe such that we cannot do otherwise in a way required for UR (though Randy's idea is interesting too). As such, he can say UR is impossible in this sense: Necessarily, in any universe in which SR holds, there is no free will. Replace SR with "determinism" and you have the incompatibilist thesis. So, he needs to be more precise--UR is not impossible simpliciter for a posteriori reasons--but (on this hypothesis) UR is impossible in any universe where SR holds.

I would say the same thing about this thesis as I do about determinism. It seems mistaken to think that a fundamental theory of physics (e.g., determinism or SR) is relevant to human freedom. The sciences relevant to human freedom seem to be the sciences that deal with human behavior (e.g., psychology, neuroscience, etc.).

[Neal]

Hmm...I guess I was thinking Strawson must have had something more interesting in mind than what has been suggested here.

As far as Randy's suggestion is concerned, I thought that Strawson thought completing an infinite sequence of actions was impossible in any case, even apart from issues about cosmic speed limits. Am I wrong about that?

And as far as the block universe suggestion, though some have worried about that, I think it's pretty clear that it is not threatening to free will. Eternalism is just the view that past, present, and future are all equally real -- not that they couldn't have been different. I guess I was hoping there was some more interesting problem than this fairly typical fatalist worry.

Thanks for your feedback.

[Kip Werking]

Neal:

That quote has always puzzled me too.

Eddy:

I, too, strongly doubt that SR and determinism are incompatible. It would be very surprising, if they were, considering that Einstein was opposed to viewing the world as indeterministic ("God does not play dice").

I honestly don't know what Strawson meant by his comment. I would be interested to learn what he did mean.

[Joe]

In his article "Determinism: What We Have Learned and What We Still Don't Know" (Freedom and Determinism, Campbell, et. al.), John Earman argues that "classical spacetimes provide unfriendly and even hostile environments for determinism" (23). Thus, according to Earman, determinism is in conflict with Newtonian Physics but not Special Relativity, which adopts Minkowski spacetime.

But I doubt that Strawson's argument is intended to go this way: If Special Relativity, then Determinism; if Determinism, then No-U-Freedom.

I think that Eddy's guess is a better one: If Special Relativity, then Block Universe; if Block Universe, then No-U-Freedom.

For one thing, this is "a point that has received surprisingly little discussion in recent debate about free will." I agree with Neal that this line of reasoning is flawed but that can't be a reason for thinking that Strawson doesn't endorse it. I also think that Strawson's Basic Argument is flawed! In any event, I'd like to know more about the details of this line of reasoning -- whether or not it is the one that Strawson has in mind.

As for Neal's second question, I think that Strawson must have meant something different than what he said. Perhaps it goes something like the suggestion above, where the conclusion is that no one has U-freedom (given Special Relativity), not that U-freedom is impossible.

[Neal]

Joe,

If you are interested to read more about whether eternalism is compatible with freedom, I recommend L. Nathan Oaklander's piece, "Freedom and the New Theory of Time", originally printed in Robin Le Poidevin, Questions of Time and Tense, Oxford: Clarendon, 1998. Oaklander's collection of essays called The Ontology of Time (Prometheus, 2004) also includes the essay, along with a large number of other great papers.

[Joe]

Thanks, Neal!

[Cihan Baran]

Eddy,

You write that "it seems mistaken to think that a fundamental theory of physics (e.g., determinism or SR) is relevant to human freedom. The sciences relevant to human freedom seem to be the sciences that deal with human behavior (e.g., psychology, neuroscience, etc.)."

I am curious as to why you think that way. I mean psychological studies reveal only that "the subject was inclined" or "tempted" to do such and such a thing. Neuroscientific studies show that people with this brain pattern "are more likely" to exhibit this kind of behavior and so on. These studies (can?) never make deterministic predictions about human behavior. Even some classical studies such as Milgram's experiment or Asch's has a huge majority acting irrationally, it is still a majority (not everyone). (Just as an aside, I think Philip Zimbardo wants to write a book about "heros" -people who do the right thing and don't conform in these experiments- now that he has completely retired from Stanford.)

So doesn't it make more sense to think that physics with its exceptionless causal laws is more apt to rob us of free will? Or do you think that probabilistic laws -as those that might be obtained from neuroscience or psychology- can be just as threatening? Or do you think that there will be some sort of a breakthrough, say, in neuroscience that will allow the formulation of strictly causal laws in these sciences? (Or perhaps you think that physical determinism allows for the right causal chain -whatever that might be- in the production of a free action, whereas the sciences you mention challenge this right causal chain? For instance, even though physical determinism is through, our actions may very well still be the result of our deliberation. However, if something like situationism - the idea that what we do has more to do with the situation we are in than our dispositions- is true, then the cause of our actions is the situation we are in.)

[Cihan Baran]

Oh by the way, Scientific American has a new neuroscience/neurobiology blog. In fact, I thought about Eddy's comment while I was reading it. I highly recommend it:

http://blog.sciam.com/index.php?cat=33

[Kip Werking]

Joe wrote:

"I think that Eddy's guess is a better one: If Special Relativity, then Block Universe; if Block Universe, then No-U-Freedom."

I'm confused: what does "block universe" mean, above and beyond simple determinism? I thought of the two as roughly synonymous?

[Joe]

Kip,

The 'block universe' is the four-dimensional space-time manifold. The assumption of the block universe, that time is like space and that past and future events are just as real as present events, makes no claim about whether or how the events of the world are causally connected to each other.

[Randy Clarke]

Neal:

You might be right that Strawson thinks it's strictly impossible for any being to complete an infinite sequence of choices. I had thought that he somewhere qualified the impossibility with "at least for creatures like us," but I can't find any such qualification.

And it does seem likely that what he's got in mind, as far as SR goes, is the reality of the future. I agree with you that that's a poor reason to deny free will.

Joe:

William James describes the universe as a block when he's characterizing determinism. Has the expression come to mean something less specific?

[Eddy Nahmias]

Joe, in trying to guess what I might mean (and come on, you know me well enough to know what I mean, don't you?), you write:
"perhaps you think that physical determinism allows for the right causal chain -whatever that might be- in the production of a free action, whereas the sciences you mention challenge this right causal chain? For instance, even though physical determinism is thorough, our actions may very well still be the result of our deliberation. However, if something like situationism - the idea that what we do has more to do with the situation we are in than our dispositions- is true, then the cause of our actions is the situation we are in."

You got it. I've written about the situationism threat. The worry, as I put it, is not so much that our actions are causally influenced more by situational factors than by dispositional factors (including character traits), but that the situational influences are often ones we (a) do not recognize as influencing us and (b) do not accept as legitimate influences on our actions, such that (c) it looks like we are being influenced (significantly in some cases) to act against the reasons we have accepted or would accept as *our* reasons for acting. (I am agnostic about how much the psychological studies have demonstrated these potential threats--which is why I'm a "neurotic compatibilist" who thinks we have free will but perhaps less than we think.)

This is one of several "bypassing" threats I think are much more important than any potential threat posed by determinism. One reason I think it is more important is because I think ordinary people recognize it as a more significant threat (and I have what I take to be some experimental phil evidence for this claim).

So (as I put it with students), if the physics geeks at MIT tell us that the fundamental laws of physics are determistic, that won't make people reconsider our fundamental human relationships and practices (e.g., in the way Smilansky suggests it would). But if the psychology geeks at Harvard or the neuroscientists at UCSD tell us that the way we make decisions is primarily caused by situational influences that bypass our conscious reasoning or by neural processes that bypass our conscious intentions, well, then I think people would think our beliefs and practices are mistaken in significant ways and may need to be revised. And if they tell us that the way these processes work is ultimately affected by indeterministic quantum processes so the processes are irreducibly probabilistic, I don't think people will feel any more at ease. (I should point out that I then make sure to refer to us as the philosophy geeks, but that's just to sound fair--I still think we're cooler than the physicists and psychologists.)

[Joe]

Randy,

In the Internet Encyclopedia of Philosophy listing on Time, Bradley Dowden uses the term 'block universe' as synonomous with 'eternalism.' That is how I was using it. (I was really just following Eddy's use.) Among philosophers, the term 'eternalism' is probably more common but 'block universe' is used fairly regularly.

I forgot about the William James use of the term. Thanks for the reminder! Note that James thinks that eternalism is just as much of a threat to free will and alternative possibilities of action as determinism. He writes:

"This of course leaves the creative mind subject to the law of time. And to anyone who insists on the timelessness of that mind I have no reply to make. A mind to whom all time is simultaneously present must see all things under the form of actuality, or under some form to us unknown. If he thinks certain moments as ambiguous in their content while future, he must simultaneously know how the ambiguity will have been decided when they are past. So that none of his mental judgements can possibly be called hypothetical, and his world is one from which chance is excluded. Is not, however, the timeless mind rather a gratuitous fiction? And is not the notion of eternity being given at a stroke to omniscience just another way of whacking upon us the block-universe, and of denying that possibilities exist? - just the point to be proved. To say that time is an illusory appearance is only a roundabout manner of saying there is no real plurality, and that the frame of things is an absolute unit. Admit plurality, and time may be its form." (James "The Dilemma of Determinism" (1948), p. 63, fn. 10)

[Cihan Baran]

Hey Eddy,

What you write would sound all fine and good to me if it weren't for the Zygote argument. In my opinion, what the Zygote argument does is that it establishes a link between simple causal determinism and forms of determinism that robs us of control -i.e. manipulation, behaving without being conscious of what the effective influences are...etc.

(I have a feeling you deny the second premise of the Zygote Argument.)

So I agree that threat of situationism or our consciousness being bypassed or similar worries are the ones we should care about BUT I also believe that physical determinism has an identical effect on us.

PS: You confused me with Joe by the way.

[Eddy Nahmias]

Cihan (sorry to you and Joe for misreading who had posted the query), I actually think both premises of the zygote argument can be plausibly challenged, and the best way to do so (as is often the case in responding to arguments) is to argue that, to the extent the first premise is developed to become plausible, the second premise becomes less plausible, and to the extent the second premise is developed to become plausible, the first premise becomes less plausible.

I won't develop the response here, but basically I think that the intuition that Ernie is not free and responsible is driven by the manipulation being presented like paradigmatic manipulation cases, in which the agent is "messed with" in such a way that he does something that he wouldn't have done without being messed with. But I don't see why determinism suggests a similar "messing with" agents to make them act in ways they wouldn't otherwise. And to the extent that the manipulation is presented in a highly non-paradigmatic way, such that the agent is "pushed" to do what he would have done anyway, the intuition that he is not free and responsible is lessened and the difference between determinism and this sort of "manipulation" is minimized. (The counterfactuals in these cases are interestingly like the ones in the situationist cases I discussed above: the agent is intuitively less free to the extent he acts on motives he does not know about and that he would not accept as legitimate motives were he to know about them--at least not unless he was also manipulated to accept them...)

I don't see why "physical determinism has an identical effect on us." Why does determinism suggest that our conscious deliberations, etc., are bypassed or that we generally act on motives we have not accepted or would not accept?

[Cihan Baran]

Hey Eddy,

I can sort of guess what your response would look like.
I am imagining something like:

Let us assume that if Ernie's zygote normally ("freely") developed (without Diane's intervention) it would A at some time t after his birth. (This is the counterfactual you have in mind?)

Now to the extent that Diane's manipulation of his atoms leads Ernie to do A at t, Diane serves as a Frankfurtian intervener. (Deny premise 1.) To the extent that Diane's intervention leads Ernie to do A' (anything different from A), Diane and natural determinism are different. (Deny premise 2, i.e. your 2 premise approach...)

This definitely seems like the best reponse I have heard so far. However, note that most incompatibilists wouldn't grant you the assumption (counterfactual): that Ernie would have freely done A.

I guess I need to think more. This certainly shows that the Zygote Argument is perhaps not as strong as I had thought it was.

Thanks!

P.S. I am deluged with work to the point of sleep deprivation. I'll probably not be able to reply. (Next week, I won't have internet access.) If I do so, that's becuase I am procrastinating.