He Coulda Caught That Ball!

I suspect the following is confused in some way, so please let me know how.

Take some non-morally-loaded action, such as a boy dropping a baseball tossed to him or a dog missing a ball tossed to him. (Presumably) because the boy and the dog have made similar catches lots of times, we might say, "He coulda caught that ball." Now, we may be making some claim about the general abilities (or capacities) of the boy or dog. But it seems we might certainly be making a claim about the exercise of those capacities on this particular occasion. If so, do people (or quite different--should people) have the unconditional sense of 'could have done otherwise' (CDO) in mind here? That is, do they (should they) mean that the boy or dog could have caught the ball with (at some relevant point just prior) all conditions, internal and external, being as they were? That is, are they committed to believing indeterminism must be true when they say "He coulda caught that ball"?

Of course, ordinary people don't have a clear conception of determinism (or indeterminism), but perhaps we could, with the proper Socratic probing, discover what they think about such cases. My suspicion is that people do not have an unconditional sense of CDO in mind here, nor would they be inclined to give up their talk of CDO in such cases if they were convinced determinism were true. (My suspicion is backed up by a couple modest empirical surveys, since I don't think armchair speculation on folk intuitions or putting our well-honed philosophical "intuitions" into the mouths of ordinary people will be sufficient here).

Perhaps with simple mechanisms people would explicitly recognize that for it to have done otherwise (e.g. for the mouse trap in Mouse Trap to have fallen) some prior condition would have had to have been different, thus explicitly accepting a conditional analysis of CDO. And perhaps with complex systems like boys and dogs they would not recognize this explicitly (or be befuddled by which conditions might have had to be different).

What's the point? Well, if the above is accurate, I want to know exactly what it is that should lead us to believe that the CDO involved in moral actions should be interpreted differently.

Here's another way to put it. I suggest that if a Consequence-style argument has the conclusion, "If determinism is true, then the boy (or dog) could not have caught that ball", people would be (should be!) more inclined to reject some element of the argument (perhaps Beta principle) than to conclude that determinism must be false (or that we are always speaking falsely when we make such CDO claims). So, why shouldn't such a move apply to the Consequence argument, or more generally, to unconditional readings of CDO in the free will debate? I suspect the 'ought implies can' principle may come into play here (in a way it doesn't with non-moral actions). But I'd like to hear where exactly incompatibilists (or compatibilists) think this line of thought goes wrong.

Comments

[Neal]

Something I've always had trouble with is thinking about what exactly is the "relevant point just prior". I suspect, though be aware that I have no empirical study to back this up, that when most people say, "He could have caught the ball", there are two options for what they might mean, and whether it is the first or the second depends upon where you place the "relevant point just prior".

So, if the relevant point just prior is very very close to the time when the boy actually missed catching the ball (a second or two, perhaps), I suspect that people do not usually mean what Eddy has called the unconditional sense of CDO. Imagine someone saying, "He could have caught the ball", and me replying with, "How?" -- I'm inclined to think that the person would go on to say something like this: "Well, if he had just reached a little higher or had his arm at a slightly different angle, he would have caught it." If that's what someone would say, then it seems clear that they don't have the unconditional sense in mind.

On the other hand, if the relevant point just prior is more than a couple of seconds (perhaps even before the boy reached his hand into the air), then I'd be inclined to think that they do mean the unconditional sense of CDO. Even given everything that did hold up until that point, the boy could have caught the ball because he could have reached a little higher.

So, at some point, the time for reaching higher (say) has passed. If the relevant point is prior to that time, the boy could have caught the ball in the unconditional sense of CDO. If the relevant point is after that time, the boy could have caught the ball, but only in a conditional sense (given that he had reached higher).

I have the sense that this only pushes the problem back a little, though, so I'm not sure.

[Graham Lester]

Speaking as an ordinary person, if my daughter stole her brother's piece of cake I would tell her that she could have done otherwise.

However, since I don't believe in free will, I don't really believe that she could have done otherwise in that particular instance.

By telling her that she could have done otherwise (which I don't actually believe in the literal sense), I am simply manipulating the intellectual environment in the hopes of tipping the balance in favor of a morally approved outcome in similar future situations.

If the dog takes the cake, I also don't think that it could have done otherwise. But I might strike the dog in the hopes of preventing it from stealing future cakes.

[Mark Smeltzer]

Eddy,

You've raised some very good questions!

I actually have what I believe to be a very effective answer to this problem (which spans out to produce a general theory of responsibility similar, if not identical some levels, to the Fischer/Ravizza RR account). However, since I'm planning on publishing it, I don't want to spoil the fun.

My working draft actually covers cases very similar to the one you've presented here!

[Eddy Nahmias]

Neal, I agree that if people *are* thinking in terms of unconditional CDO even for ordinary actions, this would come out by their saying or suggesting that the agent could have done something that would be sufficient to cause the alternative outcome (e.g. catching the ball) at an appropriate earlier time, with all conditions being identical. Often, this alternative will be traced to something like an effort: "He coulda tried a little harder (and if he had, he woulda caught it" (think Bill Buckner--sorry Tamler). Perhaps people even think something like this about the dog. But I still wonder if this is the way people think about such cases? If not, is it the way they think about cases of moral actions? If there's a difference between the cases, what drives that difference?

Graham, what if it turns out that (whether they explicitly recognize it or not), most people have roughly the same idea as you when they reprimand their children (each other, etc.) by saying 'you coulda done otherwise (better)'--that is, they recognize people's general capacities and hope to influence the way these are exercised on future occasions? Surely, we think people and dogs have relevantly different general capacities to do otherwise (at least for what they can do the next time around).

Mark, you tease! I'd like to hear how you're developing these ideas (at least say what you take to be 'this problem'). Of course, as I've put them, they are nothing particularly new (e.g. see Joe's paper at this site), but my recent work on testing folk intuitions has made me curious specifically about how people think about the ability to do otherwise in non-moral cases of action (and how they think of this ability for non-human events (actions?)).

[Steven Youngblood]

I think it goes a little deeper than that. A lot of soft determinists are inclined to think that sometimes one freely does what one is predetermined to do; and in such a case that one is able to act otherwise. However, past history and the laws of nature determine that one will not act otherwise. So, when we say that somebody could have done otherwise, we are saying that had the laws of nature been different, or (a weaker claim) that had the history of the world been different, then they would have been able to do so.

This feels like a horribly fishy move to me though. It feels much like the kind of scenario I used to face with my grandmother: "Can I have another piece of cake?" "Yes you can, but you MAY not." The distinction here I think is plain, and while the analogy may not map exactly onto the above scenario, I think we can get the general gist. Whilst I could have a piece of cake, the situation is such that I can't. However, that's like saying that I COULD get into the castle, but because the drawbridge is up, I CAN'T.

So, had the world been different at time (T-1), then the boy could have caught the ball. However, since the world wasn't different at time (T-1), he couldn't. There is something that makes me very uneasy about that claim.

If the idea is that "if the world had been different, then he could have caught the ball" is sufficiant for CDO, then there is a definite sense of something missing. When I say that the boy could have caught the ball, what I mean is that he couldn't REALLY have caught it, because circumstances were such that he couldn't possibly have. What is meant is that there is a possible world out there in which he did catch the ball, and that that is enough to say that he could have done otherwise.

I'm not sure that this is enough to switch the burden of proof to the determinists, but I do think that there is a definite suspicion that they can't have their cake and eat it too.

[Mark Smeltzer]

Eddy,

I guess I'll have to check out Joe's paper! Thanks for the heads up.

[Mark Smeltzer]

Eddy,

I just finished reading Joe's paper, and I have to say that I think he's definitely on the right track (nice paper Joe!). My project, relative to Joe's paper, could be represented as a critical analysis of what Joe calls our "general abilities" in terms of our every day intuitions regarding responsibility, an analysis of how responsibility grounding abilities manifest in agents over a period of time, and what motivates incompatibilists to misrepresent these abilities, or ignore them all together, when discussing the nature of responsibility.

As I take it, the problem is to decipher what we mean when we use the CDO principle in evaluating an agent's ability to act in non-moral (yet normative) contexts – since I believe it is obvious that we do cling to the “could have done otherwise” principle even in those non-moral contexts (like the game of basketball); and whether the role of the CDO principle is the same regarding evaluations of an agent’s ability to act in the moral context (recognizing the fact that there are many perspectives on what counts as *the* moral context) as it is in other contexts (which I argue will is the case).

Or, to put it more succinctly, I think problem can be adequately expressed as the need to provide a sufficiently robust answer to this question: “With respect to the intuitive could-have-done-otherwise and the ought-implies-can principles in connection with the issue of responsibility, what the heck does ‘can’ mean?” (Assuming that ‘could’ is simply a past tense of ‘can’.)

[Joe]

In the paper noted by Eddy and Mark (thanks!) I argue for an ambiguity between all-in abilities and general abilities. (I also note that maybe the term 'ambiguity' isn't quite right but it'll do for now.) All-in abilities are related to Eddy’s unconditional CDO’s.

To explain the distinction, I use an example from Peter Unger's Philosophical Relativity about a pianist on a train who could, in one sense, play a piano (since he is a pianist) yet could not, in another sense, play a piano (since none is available). (Unger uses the example to make a different point.) I claim that the Frankfurt examples show that all-in abilities are not necessary for moral responsibility. I also think that the Consequence Argument shows that all-in abilities are not compatible with determinism. General abilities are more fundamental than all-in abilities, for if I do not have the general ability to play a piano, then I don’t have the all-in ability to play it now (given that there is no piano). But the reverse does not follow. There are lots of pianists who are sitting in rooms without pianos.

I think that general abilities are enough to ground the freedom-relevant condition for moral responsibility. I don’t think that having a general ability to do otherwise is a sufficient condition for being morally responsible but I do think that it is a necessary condition. I don’t see how anyone can argue this point. Where the difficulty comes in is providing an account of the other necessary conditions for moral responsibility. This is why in the manipulation cases discussed on this blog I have tended to argue that it wasn’t the freedom-relevant condition that was violated but some other condition was violated (usually the epistemic condition).

[Tamler Sommers]

I think the analogy is misleading because in general, there are more external factors at work when catching balls (wind, ground surface etc.) than when making moral decisions.

Take another non-moral case (I'm gritting my teeth as I say that): Grady Little coming to the mound and deciding to leave a fatigued Pedro Martinez in the 8th inning of game 7 against the Yankees. This seems like a better analogy to a moral case, because there were far fewer external determinants that could have affected the outcome. It's just a man making a decision, just like in many moral situations.

Does that change things? My guess is that most Bostonians think that Grady Little, at that moment, under identical conditions, could've taken the ball from Pedro, like any remotely competent manager would have done, rather than patting him on the shoulder and leaving him in. And they hold him responsible for failing to do that, for that monumental lapse in judgment.

[Robert Allen]

Eddy,

I think that you are right that it is usually lack of effort that people are thinking of when they assign blame. However, just to set the record straight, that wasn't poor Buckner's problem. (Can you believe that he once had to re-locate his family out west to avoid vituperation from Red Sox "fans"?) His ankle was injured and he should not have been in the game at the time. The Red Sox manager- John McNamara- should have replaced him with Dave Stapleton, who was a much better fielder. Thus, Red Sox fans who hold the above view would fault McNamara- and Grady Little, for that matter, for not 'having his head in the game'- that is, concentrating hard enough. But, of course, as Chisholm noted, that only begs the question could he have concentrated harder under the circumstances?

Re. your comment on Graham's post: I think that he would agree with you that there is a difference between a dog's and a child's ability to make behavioral adjustments. That is why he thinks that the latter should be chastised and the former punished corporally.

[Daniel Cohen]

The point that has come up about trying, as being the ultimate locus of unconditional ability, is appealing. However, don't forget Austin's famous counterexample to such views. When a golfer kicks himself for missing a putt, his thought is not that were he to have tried harder he would have holed the putt (the general ability), and that he could have tried harder (the all-in ability). After all, he really did try, but failed nevertheless.

I expect that when bluntly faced with the facts in such (after all, familiar) cases, most people will get pretty confused. Unlike philosophers, they won't really consider the possibility of withdrawing the blame or the self-reproach, or to think of the blame or self-reproach as being essentially pragmatic in justification. The experience (shared even by hard determinists) seems to be something like the following: Inductively speaking, it's really surprising that I missed the putt - I always (or usually) holed it before in similar circumstances. But I missed it, so I must have neglected to do something that normally guarantees success. I must have failed to try hard enough! Hold on, that's false. I really did try. Oh, I don't know, it's just the case that I could have holed the putt, but didn't.

This little monologue, drawn from introspection, is ultimately hard to interpret coherently. What do I mean, after all, in saying that I *just* could have holed the putt? At the very least, I know what I don't mean. I don't mean simply that I usually hole similar putts. And I don't mean that my holing the putt was not completely a function of my mental state prior to the putt, or that my mental state prior to the putt was consistent both with my holing and missing the putt (in fact, I would normally reject both of these claims). And, when pressed, I don't necessarily even think that I would have holed it if I had tried harder, because I might really think that I did try as hard as I could.

Eddy, I think it’s going to be hard to quantify such normal experiences in an experimental context. This is because it doesn’t simply seem to be a question of whether or not people implicitly believe determinism, or any other straightforwardly intelligible notion. Rather, I suspect normal thinking is much more agent-causal in nature. What you might look for, and what I predict you would find, is a lot of categorical table thumping (“I just could have got the putt, and that’s that.”) That, and a real resistance (and discomfort) at the prospect of analysing such claims.

[Eddy Nahmias]

Daniel, I agree that understanding how we think of the ability to do otherwise is difficult whether we rely on our own philosphically informed introspection or the phenomenological or intuitive responses of prephilosophical folk or whatever.

For the record, here's a little study I ran (with grad students Steve Morris, Thomas Nadelhoffer, and Jason Turner), reported in a footnote of our paper "The Phenomenology of Free Will" coming out in the latest issue of Journal of Consciousness Studies. Let me emphasize that I don't think our results show a whole lot, but they do offer an interesting ordinary folk response to the ordinary language philosopher Austin.

We asked subjects:
Imagine you are playing putt-putt and you have just attempted a short putt but missed it (perhaps you can actually remember an event like this). You think to yourself, “I could have made that putt.” Which of these descriptions best captures what you mean when you think, “I could have made that putt”:
1) I could have made that putt under the exact same conditions.
2) I could have made that putt under very similar conditions.
3) I could have made that putt only if something had been different.
4) I make putts like that sometimes and I miss them sometimes.
5) None of the above.

Results (33 subjects):
1) 24%
2) 6%
3) 43%
4) 24%
5) 3%

[Kip Werking]

Those numbers strike me as a resounding endorsement of incompatibilism, or at the least the need for revision with respect to our sense of fw/mr.

One nit-pick: using the word "could" within the question might be confusingly circular (I am thinking here of the literature where "could" is often defined in terms of itself, which is not helpful).

By the way, Eddy, I think experimental philosophy is fantastic, and I really appreciate the word you are doing to explore the folk or pre-philosophical intuitions we have about fw/mr.

[Eddy Nahmias]

Kip, I don't see how these results are a "resounding endorsement of incompatibilism" at all. Answers 3 and 4 seem to me to jibe best with a conditional reading of "could have done otherwise" whereas only 1 clearly suggests an unconditional (incompatibilist) sense. Though again, I'm not hanging too much on any of this. Please explain.

[Kip Werking]

Eddie, it is entirely possible that I am just reading my own incompatibilist leanings into the results. If that is the case, please forgive me--it's been a rough week.

But initially (and even now this impression seems reasonable), both answers 1 and 3 struck me as firmly incompatibilist. You write that 3--which is the most popular--is a compatibilist analysis, but there is an essential difference between the compatilist analysis of 'could' and 3!

Compatibilism: Only if things were different, I *would* have succeeded.
3: Only if things were different, I *could* have succeeded.

Note how relevant this difference is! It seems to me to shift the entire meaning of the phrase from compatibilism towards incompatibilism. Because note the necessary (and paradoxical?) consequence of these two premises:

A. Things were decidedly *not* different.
B. Only in if things were different *could* I have done anything else.

B is a conditional whose condition is never satisfied and so the conclusion can never follow: Nobody can ever do anything, because nothing is ever different than the way things are. This was the option that folk psychological or pre-philosophical intuitions supported MOST!

Likewise, option 1 is unambiguously incompatibist. But these two options account for 67%, or two-thirds, of the results. Hence my comment that these numbers are a resounding endorsement of incompatibilism. Please correct me if I have made a careless mistake--I have not slept in a while.

Again, I want to thank you for doing this empirical research, which despite its inherent limitations, is nevertheless important in my opinion, and quite interesting.

[Mark Smeltzer]

I'm a compatibilist, and I would have marked option #1 if I had taken that survey (which was, "I could have made that putt under the exact same conditions"). So, I agree with Kip's suggestion that the results are misleading because the answers incorporate the word "could". However, I disagree with Kip insofar as think the results of that particular survey do not provide clear support for the (tacit) acceptance of either position on the popular level.

[tnadelhoffer]

For what it's worth--here is my two cents worth (which are in agreement with Kip's two cents!):

We have to specify which type of incompatibilist we are talking about here. For example, the "hard determinist"--says that she "could have done otherwise" only if things had been different. But, given that determinism is true, things could not have been different (based on a Consequence Argument version of determinism whereby the physical state of the universe + the physical laws = one (determined) outcome). Hence, she could not have done otherwise. Given this view of determinism and the meaning of "could have done otherwise," if the hard determinist were asked whether she could have made the putt--she should answer "only if something had been different." After all, if things were exactly the same, the consequence argument tells us that only ONE outcome was possible--namely, missing the putt.
For these reasons, I think that answer #3--the most popular answer--is the answer that a hard determinist would give.

However, it is not the answer that a libertarian would give. This suggests that it is unhelpful to talk about "incompatibilist" intuitions in this context. After all, according to some libertarians (namely, agent causationists), #1 is the most obvious choice. Keep in mind that according an agent causationist, she could have made the putt even if everything had been the same insofar as she believes determinism is FALSE. The event causationist, on the other hand, may very well find #3 a better answer choice than #1 since she will need room for some indeterminism to enter into the picture. Either way, answers #1 and #3 are going to best settle with incompatibilist views--although we must be careful to distinguish which incompatibilist views we are talking about for the reasons I have already mentioned. Perhaps the most striking thing about this pilot study was that so few people picked 2 while so many picked 3--since #2 implies #3 and vice versa. After all, "very similar" (#2) implies "slightly different" which is precisely what #3 says.

In any event, I am interested to see what others have to say. I, for one, have politely disagreed with Eddy about the significance of the results of this study from the start--which may partly explain why it ended up in a footnote rather than in the body of the paper. But given that he and I are on different sides of the fence about compatibilism/incompatibilism, it is amazing that we were able to succesfully collaborate in the first place. Despite our differences, it was a rewarding and educational experience.