Nelkin on Experimental Philosophy

I was happy to see the recent post about the superb article by Nahmias, Coates, and Kvaran in Midwest Studies.  I hope this doesn't end up stealing any attention from that earlier post, but I thought it might be helpful to draw your attention to another article in the same special issue. 

The article I have in mind is Dana Nelkin's 'Do We Have a Coherent Set of Intuitions about Moral Responsibility?' There, Nelkin goes through almost all of the recent work on moral responsibility within experimental philosophy and argues that it is actually possible to capture all of the intuitions uncovered by this work in a single coherent theory. 

One thing I found especially striking in her paper was the explanation she offers for cases in which people seem to think that moral responsibility is not compatible with determinism.  Her suggestion is that, when people hear that an agent's behavior is entirely determined, they conclude that the agent is not truly making decisions at all.  In arguing for this claim, she writes:

Additional anecdotal support for the idea... is provided by my experience presenting some of the relevant experimental results to an interdisciplinary academic audience that contained a number of psychologists.  Several objected to the set-up of the experiments in which subjects are supposed to respond to agents performing deliberate actions in a deterministic scenario. In particular, they worried that the scenarios might already beg a key question in describing actions in such terms at all in a deterministic world, and that the scenarios were actually incoherent because determinism precludes deliberate actions done for reasons [!].

I feel certain that Nelkin is getting at something important here -- that people ordinarily do show a tendency to think that determinism precludes even the possibility of taking into account certain reasons and then making a decision.  Yet this phenomenon strikes me as a highly puzzling one.  Just on the face of it, the concept of determinism doesn't seem to have much to do with the question as to whether people can genuinely make decisions.  Why then would people be drawn to the idea that determinism and decision are incompatible?

 

Comments

[Randy Clarke]

"Why then would people be drawn to the idea that determinism and decision are incompatible?"

An interesting question. One hypothesis: Perhaps many people conflate determinism with a view on which behavior is produced by mechanisms that run their courses entirely independently of thought (or anything mental).

My experience in the classroom, though, is that even when one has cleared up that confusion, many students (at least initially) find compatibilism crazy.

[Eddy Nahmias]

The "determinism entails we are robots" picture is rampant. I will try to find examples from intro texts, scientists' discussions of free will, etc. for my post on determinism.

In response to Randy, research on how hard (or easy) it is for people to give up beliefs would be relevant. I think there is such research suggesting that it is very difficult to give up beliefs even in the face of countervailing evidence. So, it may be that students (scientists and folk) are told or imagine that determinism entails (or suggests) we are mindless robots, then adopt what *would be* the correct view in light of that (false) belief--namely, that determinism is incompatible with determinism--and then have a hard time seeing how compatibilism makes any sense once we try to disabuse them of that implication of determinism.

Consider how hard it is (was) for many people to become convinced that invading Iraq was a bad idea...

(Oops, I didn't mean that to sound like a bad analogy/ad hominem against incompatibilists!)

[Mike in MI]

"Just on the face of it, the concept of determinism doesn't seem to have much to do with the question as to whether people can genuinely make decisions."

If you think decision is simply a matter of coming to be in one psychological state rather than another, then you can separate decision from determinism. But if you think that decision involves choosing between outcomes that are all possible prior to the decision, and that decision is the immediate cause that makes one of them actual and not the others, then determinism renders decision illusory--i.e., people don't GENUINELY make decisions in a deterministic universe (which is not necessarily the same as being mindless--showing we are not mindless in a determinsitic universe does not show that we make genuine decisions in a deterministic universe).

[Kip Werking]

"My experience in the classroom, though, is that even when one has cleared up that confusion, many students (at least initially) find compatibilism crazy."

I can't speak for others, but I would say that this describes my own thought experience. There is a two step journey, and each step involves overcoming a bias.

Step 1: recognizing that determinism doesn't conflict with computation, thinking, feeling, decision making, calculating, control (not unlike a thermostat), etc.

Step 2: recognizing that, despite our strong gut feeling that at least part of our personalities/decisions are self created, in an ultimate way, they, in fact, are not---that we are just complex robots, who make decisions and perform calculations in a way compatible with determinism.

What I find fascinating---and what I am just starting to understand---are compatibilists who only focus on Step 1. Then they quit, as if the journey is over, and the conclusion final. This strikes me as premature. Non-realist types (however few we are), completely agree with Step 1. But we are unimpressed, because we recognize that the journey is not over. There is still Step 2 with which to grapple.

This is why I pose the following challenge to compatibilists:

1. Are you an orthodox compatibilist that simply denies the existence of the bias in Step 2? Do you simply deny that people ever think of themselves in this self-creating way? I call this view SCOFF.

2. Do you agree that the bias in Step 2 is very real, but insist that it does not infect beliefs about fw and/or moral responsibility? On this view, it might be true that we think of ourselves in this extravagant, overblown way, but, thankfully, we don't need these extravagant powers to have fw and/or moral responsibility. Actually, fw and/or moral responsibility *only* requires something much weaker than most people think they have---something consistent with determinism. I call this view Phew!, as in "Phew! For a minute there I thought we were going to lose fw and mr, because our views of our own powers and abilities are so extravagant and overblown, but fortunately, it turns out that we still have fw and mr even with our weaker, real powers/abilities.

3. Are you a revisionist type compatibilist, who acknowledges the bias is Step 2, and acknowledges that we use fw and/or mr to refer to these extravagant and overblown powers, but insists that we change the meaning of these terms, and should revise their definitions, so that they only refer to things compatible with determinism?

So that is my challenge: SCOFF, PHEW, revisionist, something else, and/or some combination. I'd be very interested to hear where any compatibilist stands on this issue.

[Mike T.]

A potential third option is that when people imagine a totally determinate world they imagine that determinism “atomizes” agents into the causal stream. They become indistinguishable from the scenery so that there is simply nothing on which to pin a “decision,” since that seems to be the kind of thing that requires an agent.

[Paul Torek]

The Phew!, the proud, the Real Compatibilists*. Sign me up for that.
*as Kadri Vihvelin would say.

Not that there's a hard and fast distinction between vanilla compatibilism and revisionism. In the web of belief, all is connected. Cut one strand, and everything shifts at least a bit - but sometimes, only a tiny bit.

[quidnunc]

"if you think that decision involves choosing between outcomes that are all possible prior to the decision, and that decision is the immediate cause that makes one of them actual and not the others, then determinism renders decision illusory"

Evaluating possibilities seems compatible with a deterministic universe to me; if that's what making a decision is you can be deflationary about our degree of control but it's not like making a decision requires realizing alternatives as if they actually happen. It's an act of picking out relevant features and making a choice.

I can see how it would appear illusory within an inflated view of the will but it seems to me to presuppose metaphysics easily rendered silly. When people say "anything is possible" they're typically being optimistic about their desires and the value of hard work, not imagining that the universe and the people in it will bend to imagination.

[Mike in MI]

If the metaphysics of the "inflationary" concept of FW be "rendered silly", that still doesn't get you compatibilism. It just gets you determinism. And the metaphysics that supposes that genuine MR is possible in a deterministic universe is at least as easy to "render silly" as that of the "inflationary" concept of FW. The real puzzle about intuitions is why anybody who understands the implications of the principle that "ought" implies "can" would consider MR possible in a deterministic universe.

Here's another take on the stages of compatibilism:

1. Aporia, in which the subject accepts determinism but refuses to deny MR, despite the apparently obvious reasons for thinking they can't both be true. Rather than accepting the dilemma, the subject chooses to view the problem as a puzzle to be solved.

2. Equivocation, in which the subject tries to make the problem go away by redefining key terms.

3. Frustration, when the subject sees that incompatibilists remain unmoved by her verbal dodges, and realizes that she and they are rehashing the same arguments that have been stated and restated for centuries.

4. Complacency, in which the subject convinces herself that the problem has been solved. The subject absolves herself of having to take incompatibilism seriously, and, rather than trying to find new arguments, dismisses it with question-begging epithets like "inflationary".

[quidnunc]

(i) I would separate the problem of MR from the problem of free will. I wasn't speaking to the former - and definitely wasn't trying to be adversarial. Sorry if it appeared that way. It was stream of thought. My writing clearly needs a lot of work.

(ii) If anyone's redefining terms it's because the previous models and controversies were grounded in concepts that have changed. I came to philosophy via cognitive science and inherited my understanding through it. I'm not redefining terms, only giving my opinion about what's most plausible from the reading I've done. Dennett wrote Elbow Room long before I started reading.

(iii) I believe there's a relevant distinction between fake and feeble. Eliminativism typically trades on a contrast to demarcate and declares some thing x fake or non existent but that can be problematic if there are plausible alternative models. I do personally think there's a Cartesian intuition behind an incompatibilism that presupposes the ability to literally do otherwise and would point to alternatives that do take into account all the relevant psychological findings while maintaining philosophical sophistication (e.g. see recent work by Jenann Ismael)

[quidnunc]

Regarding the point you make about MR, "the metaphysics that supposes that genuine MR is possible in a deterministic universe"

What is the problem exactly? If "morality" evolved to regulate and deter behavior blame is a tool to identify whom it is directed at (see for example Christopher Boehm, Hierarchy in the forest for examples on a small scale). The morality system itself has causal power. History unfolds and it's part of that unfolding. Do people really care about the ability to do otherwise or is it more about the "disposition" in the flow of the person to do wrong according to the standards of society which are expected to be internalized by all actors? Reason does mitigate blame but it's limited because its primary purpose in my view is to regulate behavior not trace history and imagine a change that would alter behavior before the act. I'm not at all familiar with the literature in metaethics so let me know if I'm making a fundamental mistake.

[Mike in MI]

There's the problem right there. Incompatibilists are typically interested in morality, not "morality". Take anything that outrages you--9/11, Darfur, whatever. Introspect on that outrage for a moment. Doesn't it take a certain kind of doublethink to feel that sense of outrage and then turn around and say, "Well, that's horrible, but Bin Laden can't really help being the way he is, and the events of 9/11 couldn't have turned out any other way, and my own emotional reaction is nothing but an evolved response, like the feeling of disgust I experience when I smell rotten meat."?

Taking MR seriously means affirming that "ought" implies "can", and correlatively, "ought not to have" means "need not have", etc. for other tenses. It's not just about the past. Many of us perceive a discrepancy between the world as it is and the world as we think it should be, or between the persons we are and the persons we think we should be. One can try to do something about that discrepancy, or one can try to rationalize away that perception. Doing the former (without doublethink) requires one to think that it MATTERS how things turn out, and that our choices make a difference, and that therefore our deliberations are not analogous to a marble rolling around in a bowl before it inevitably comes to rest at the only point it possibly can.

[quidnunc]

It is interesting and I did overstate my case (tends to happen a lot as a student not versed in all the material).

Anyhow, it's hard for me to see the necessity of 'ought implies can'. To be a certain way rather than another presupposes structure, and structure implies constraints. If Osama Bin Laden had the amorphous ability to make decisions he wouldn't be anyone at all. There's something about Osama Bin Laden that makes us single him out as a terrorist. I can see the intuition that we are "pulled" along with the universe through the sheer quantity of causes, however, we often single people out for conflict precisely because they are undeterred and unmoved. In other cases social pressure will alter behavior because we have the ability to change through learning but we are also aware of constraints in the person and out in the world - no one is presumed to be infinitely malleable. When we say that person is in the grip of a dangerous ideology we recognize there are causes within or beyond the person that aren't easily changed; hence discussion about "root causes" of terrorism.

We are likely to feel more upheaval of anger when someone kicks a dog because they enjoy it in contrast to someone who kicks a dog because they're in a state of fear. First, because there is a logic for conflict in an innate moral sense; second, we assume something much more sinister about the general attitudes of the first actor toward violence. I wonder whether assigning more responsibility to the former hangs on "ought implies can" rather than a heuristic about the status and quality of a person. It seems to me that it's possible to preserve that judgement of responsibility while making a distinction in treatment if the former had selective brain damage because it hangs on shared standards in humans in the capacity of a human that are to some degree sensitive to facts about human welfare and concerns rather than a being halfway between the apes and the angels, having an ultimate ability to choose.

[Mike in MI]

Scroll up to what Mike T. (who is not me) said about "atomization". In a deterministic universe, pathological factors, normal evolved responses (such as empathy), culturally transmitted behavior patterns, environmental factors, etc., are all just so many vectors impinging on the path that human organisms trace between birth and death. In such a universe, to circumscribe certain of a human being's movements and put them under the rubric of "responsibility" is a mere linguistic convention, for there is no fundamental difference between any of them. There is no genuine agency in a deterministic universe. Genuine agency is the ability to *originate* some of those vectors. *That* is the fundamental difference that MR requires.

By "some" I mean "at least one". Assuming we live in an indeterministic universe, we still may make only a handful of free choices in a lifetime. Or maybe only one. (Or perhaps several tokens of a single fundamental choice-type.) These few choices may be enough to form (or break) the habits that determine all our subsequent behavior, but that is enough to make us free and MR. So indeterminism and MR do not imply an amorphous personality. MR is perfectly consistent with a stable and relatively predictable character.

(Here also is an indeterminist answer to the neuroscientist in the subsequent thread. And if any virtue theorists are reading this, there's an indeterminist interpretation of Aristotle in there, too.)