Blog Coordinator

« Free Will and Experimental Philosophy: An Intervention | Main | Free Will and Science in Padua »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Hi David,

Thanks for the thoughtful post! I have been thinking about cases along these lines a bit so let me throw in my two cents.

All attributability views say that an agent is MR for her attitude or action A only if A reflects (in the appropriate way) her underlying evaluative stance. But evaluative stances are complex things and needn’t be univocal. They might encompass a variety of principals, cares, commitments, concerns etc. that are in subtle (or not so subtle) tension with each other. Thus we need to distinguish two ways of understanding the ‘reflect’ relationship. On a *wide construal*, we look at the overall ‘resultant’ from summing over these heterogeneous evaluative elements and see if the action or attitude is an expression of this overall vector. On a *narrow construal*, we check the elements of the person’s evaluative repertoire one by one, seeing if any serves as the right kind of ‘anchor’ for the person’s action or attitude.

I think it’s plausible that the Daddy-issues agent (I assume its not really you!) has some deep seated concern over being abandoned as well as a need for safety. That is, he has elements of his ‘deep’ evaluative self that see the prospect of safety and avoiding abandonment as a reason to end a relationship, even if his other cares and concerns are in tension with this evaluative verdict. On a narrow construal of the reflect relationship, his spurning actions can indeed (narrowly) reflect his underlying evaluative stance, explaining why the actions are attributable to him.

It might strike one as strange to say that the ‘dysfunctional’ attitude (yearning for safety and fear of abandonment) is internal to the agent in the way that I am suggesting. But I think that this is right. In any case, it’s one way for the attributability theorist to address your very interesting case.

Hi David,

This is an interesting case, but it seems at present underdescribed to be a counterexample to the popular theory of attributability mentioned above. It seems we need to hear both about the nature of fear and the connection between the fear of being rejected and the evaluative judgments concerning the partner’s minor faults. For example, if one plumps for a cognitive theory of the emotions where emotions are (some special kind of) evaluative judgment, then this would not seem like a counterexample. I am suspicious of such theories of the emotions and I suppose at the very least it might seem like a kind of cost of this theory of attributability that it requires going cognitive with respect to the emotions to avoid the counterexample. Another question concerns the relationship between the fear of rejection and the evaluative judgments. What if the fear of rejection directs one to make these judgments? That is, perhaps the correct description of what the patient discovers in therapy is not that the evaluative judgments were not motivational or explanatory, but that they are not (in some sense) ultimately motivational or explanatory. This would not (at least obviously) make problems for the theory of attributability, would it? It would seem then that the case needs to be one in which emotions (as understood on non-robust cognitive views) moves one to break off these relationships without the pertinent evaluative judgments also playing any motivational role. Indeed, wouldn’t the case need to be one in which no evaluative judgments whatsoever plan any motivational role? This seems much harder to come by, though perhaps still possible.

I am wondering about dependence and its relata. Must it be dependence on evaluative judgments I actually make? Above I assumed that dependence at least was motivational: is this right? Is mere counterfactual dependence sufficient? Must some of the evaluative judgments I make actually causally produce the relevant action or attitude?

Why shouldn't we say that though the attitudes have a source of which I am unaware, and of which I would not approve, the attitude is nevertheless mine? Or at least that that story is plausible enough to explain the intuition that it is attributable to me? How I came to have an attitude is one thing; whether that attitude is now constitutive of my evaluative stance quite another.

Interesting question, David.

I'm thinking that the attributability view is more basic than the account of it in terms of judgement-sensitivity or dependence on evaluative judgments. So my inclination is to think that your spurning behavior is appropriately attributable to you, and that this shows a problem with the "normative" account of it, but not with the attributability view itself.

Attributability views of responsibility (or an aspect or "face" of it) must presumably rely on a conception of the "self" to which behavior is being attributed. But the "normative self" (as it were) is just one option, and it would be a mistake, I think, to tie the attributability view per se to the normative conception of the relevant "self". Gary Watson originally introduced the "attributability" idea, and he also defends a normative conception of the self, but the two issues can be prized apart, and one might embrace an attributability view but a different conception of the self. Or so it would seem to me. (I have benefited here from reading work by Ben Mitchel-Yellin, and also discussing these issues with him.)

While I'm getting e-mails with comments, none are showing up on my browser's feed of the blog, so I will simply respond to what I've read thus far and perhaps they will all show up in the right order.

Thanks for the replies, everyone. John has in mind precisely what I had in mind: such cases suggest a counterexample to the normative self view, according to which the relevant practical agent to whom actions and attitudes are attributable is more restricted than, say, the person, and it is defined in terms of evaluative judgings, from which that agent can't be alienated. I think it's right to expand the notion of the relevant agent here in light of Freudian style cases as well as others, so I was looking for ways to push back from the strict evaluative judgment camp.

I am looking, then, to widen the scope of the relevant practical agent even beyond Chandra's suggestion of a pretty wide notion of "evaluative stance." The agent may actually (and sincerely) judge that abandonment isn't in fact something to fear in the way he has, that fear expressed in this way is quite bad, and yet have been moved to act in this way such that the attitude is still attributable to him. In particular, I think it worth avoiding a view of attributability that talks only in terms of reasons being given for performing various actions from one's evaluative set. That is to say, I think it worthwhile to combine "authenticity" and "authority" based understandings of attributability (where the evaluative judgment view is in the latter camp). (And Chandra, if you're uncomfortable thinking of "I" referring to me, substitute "Tamler" for I).

Chris, you're right that the case is underdescribed and you raise some good issues. I think my target is the view that evaluative judgments are where the buck stops, so if it turns out that the evaluative judgments themselves are dependent on a noncognitive something or other that is more fundamental to agency, then I take it that the view will be incomplete (on the assumption that that something more fundamental could be the source of attitudes more directly).

Neil, I agree with your basic point, and it's making sense of what makes that attitude mine even when its source is not an evaluative judgment that I'm interested in discovering.


Could you say just a bit more about what in your view constitutes an evaluative judgment?

Hi David--interesting question. I'm wondering if we could not invoke some moral demarcation of evaluative competence mapped to more ordinary assessments of whether someone is competent to (say) stand trial or some such. Then the therapist stands in the same place as a judge who points out your "crimes" of rejection and asks whether you can understand them as transgressions of common social "law": most people would not reject lovers based on simple faux pas too critically assessed as something more than they are. If you can understand that and acknowledge the rationality of that, then you are competent to accept those rejections as attributable to you, even if you did not act maximally rationally in going through with your rejections. On the other hand if you simply reject the therapist's analysis, and refuse to see that you were in any way irrational in your rejections, then you have shown that you were indeed not competent in your evaluations and exhibiting that behavior, and thus it was not in any baseline rational way attributable to you. Mistakes are attributable to us, and maybe even deeply unreflective mistakes. But perverse character flaws linked to an inability to make rational evaluations in any case potentially are not. Just a thought; I'm not marrying myself to such an account.

One move for this theory of attributability would be to explain the case in terms of evaluative judgments. On this approach, the way the daddy issues affect one's romantic behavior is by shaping one's evaluative judgments.

This might involve understanding the category of evaluative judgment broadly, so that it includes emotions like anxiety that one will be rejected by one's lover. That sounds fine to me, but then I'm sympathetic to crude forms of compatibilism on which psychologically simple creatures like animals can be morally responsible for their actions, and I think people don't usually like that.

Another way to develop this line would be to have the emotions causally bring about whatever mental states are the evaluative judgments. Maybe the fear of rejection causes the evaluative judgment that the relationship isn't worth pursuing. This line might be more satisfying for people who don't want simple emotional responses to make us responsible for their consequences.

Here's a suggestion as to what makes an attitude mine: it's centrality to the web of attitudes. The more of the web that would unravel or swing loose were the attitude to be rejected, the more deeply it is to be attributed to me. This story makes belonging independent of origin, though not entirely independent of personal history.

Looking at some of the comments above, I wonder if it would help to make a clearer distinction between an evaluative judgment and an evaluative stance. An evaluative judgment is any judgment involving an evaluative concept. Narrowly, such concepts include ‘good’, ‘desirable’, ‘valuable’, etc… (graded concepts of axiology). But I think people often speak more loosely and allow any normative concept (‘right’ and ‘is a reason for’) to figure in an evaluative judgment as well.

The Scanlon/Smith attributability picture, as I understand it, makes a distinction b/w evaluative judgments and evaluative stances, but the picture is complex and the terminology confusing. Here is my take on their view (please correct me David or others if I’ve got parts of this wrong). Scanlon/Smith think pretty much any mental attitude (desires, intentions, emotions, unreflective conduct such as patterns of noticing) embody evaluative judgments because each of these involves (perhaps only implicitly) seeing certain considerations as counting in favor of certain responses. This comes out most clearly in Scanlon’s discussion of desires ‘in the directed attention sense’ as necessarily involving seeing certain considerations are reasons. But the evaluative judgments associated with our actions and attitudes (call these ‘attitudinal evaluative judgments’) might be false or unjustified. Luckily, we are rational creatures and we have more permanent ‘standing’ dispositions to make evaluative judgments. Thus we reflect on and can revise our attitudinal evaluative judgments. These standing dispositions to judge in certain ways collectively constitute an underlying ‘evaluative stance’.

It was suggested above in a couple of places that on the attributability view, you are MR if your action or attitude reflects some or other evaluative judgment. This is probably not a plausible view overall, and in any case is not the Scanlon/Smith view. Rather, on their view, you are MR in virtue of the fact that your actions and attitudes (and their associated attitudinal evaluative judgments) exhibit the right sort of dependence relation on your underlying evaluative stance. The motto for this kind of view is that your actions and attitudes are judgment-sensitive attitudes, where ‘judgment’ here refers to your underlying evaluative stance, rather than attitudinal evaluative judgments themselves. So when David (er, I mean Tamler) spurns these women, he is MR because the spurning is judgment-sensitive; not because the spurning reflects an attitudinal evaluative judgment.

That was an attempt at a clarification, but it seems pretty tangled in its own right. Help me out if I am wrong or if you see their picture differently.

I take it that you (David) are trying to find some notion of a self that is not so closely tied to evaluative stance. A person would be MR if an action or attitude is ‘self-sensitive’, and not necessarily judgment-sensitive. That seems very promising. Maybe you can fill in the details of how such an account would go down the line…

Josh: I have in mind what I think Watson has in mind: judgments of worth, things that give us reasons for pursuing courses of action grounded in a general evaluative commitment or stance.

Alan: I'm typically loathe to think we can get illumination from comparisons to the law, as I think there are lots and lots of subtle differences between legal and moral responsibility. Regardless, even if one rejects the analysis and "refuses to see" that one was irrational, I don't yet think that disqualifies one from attributability, especially if another structure were in place grounding the attitude in some key aspect of agency.

Neil S: I think your understanding of evaluative judgments is indeed overly broad, or at least goes far beyond what advocates of the view would be comfortable with. I've allowed for something like your second suggestion in responding to Chris, but again, if there's something more fundamental and inalienable to the practical self than evaluative judgments, the advocates of this view will likely balk. But I myself am quite happy with subsuming evaluative judgments under the rubric of certain sorts of emotional dispositions (cares) on the story of attributability toward which I'm inclined.

Neil L: This is the approach taken by Arpaly and Schroeder in their "Whole Self View." I think it is subject to difficult counterexamples, e.g., cases of radical conversion (I have a sudden and unintegrated desire to completely change my miserable self), and thoroughgoing ambivalence. I also have a hard time figuring out why an attitude's purely extrinsic relation (integration/centrality) is relevant for responsibility for that attitude.

I don't see ambivalence and conversion as particular problems for the view. The view very naturally allows for degrees of integration; it predicts ambivalence. Are genuinely radical conversions actually possible? It seems to me that they are not, for the reasons Dennett gives in his 'brain writing' paper. Suppose they are, though. Then I would say of them what Sartre says. He writes somewhere something like this: my sudden teenage religious fervor might best be seen as a sign of my vocation, or a symptom of adolescence confusion. Only time will tell. Why time? Because the sudden fervor needs to become strongly integrated into my mental economy to be properly identified with me.

What does it have to do with responsibility? Nothing, directly. The view is that I am responsible for actions that express my attitudes. The integration theory is a theory of what makes the attitude mine. We need a further story about expression of those attitudes to ground responsibility.

Chandra: I don't know how I missed your remarks yesterday. Thanks, they are indeed very helpful. I should say that in my initial case I explicitly had in mind the Watsonian version of attributability, which seems more clearly to render attitudes mine to the extent they are or depend on evaluative judgments (and in the beginning, these were judgments of what's best. You're quite right, the Scanlonian/Smithian view is more complicated, and Scanlon and Smith actually disagree, I think, over the scope of attributability now, which makes things even more complex (Scanlon accepts almost any element crossing one's psychic domain as one's own, whereas Smith excludes several types of items).

Indeed, Angie says explicitly ("Responsibility for Attitudes"), in talking about the evaluative stance Chandra mentions, that "'Judgments'... do not always arise from conscious choices or decisions, and they need not be consciously recognized by the person who holds them. Indeed, these judgments are often things we discover about ourselves" (252). Now while this might seem to cover the case I've given, I'm not so sure, as I'm uncertain whether the repressed fear in question reflects any sort of valuing, mistaken or not, which is the real issue.

At any rate, you're right, Chandra, that I am indeed looking for a construal of the self that extends more widely than that of "evaluator." What that consists in precisely, however, is, for now, a well-kept secret...even to me. I have some hunches, but I think there's a very complicated story here. I'll likely try out some ideas here in the coming months.

Neil, I'm interested here only in the case given above as a possible counterexample to a very specific evaluator view of attributability. The objections I tossed off to the integration view are two of five I've developed at length elsewhere, but going into any detail on them here would be a side trip I don't want to take right now.

What if instead of seeing a psychoanalyst, the patient was being counseled by a neuroscientist. What if the neuroscientist did up a quick history and explained all of the patient’s behaviors, not in Freudian terms, but through the lens of mechanism and genetic predisposition. Does this change the case any? I don’t think it does. In both cases, the agent, who seemed to meet all of the requirements of attributability at the time, turned out to be unconscious of the complex causal chains of his behavior. I think the takeaway from both cases would be that the seemingly reasonable prerequisites for attributability and responsibility are based on unrealistic/outdated conceptions of the human mind and models of agency (therefore they are not so reasonable).

The comments to this entry are closed.

Books about Agency

3QD Prize 2014: Marcus Arvan