Recent Work on the Concept of Responsibility

Hello all,

I'm working on an update of my Stanford entry on moral responsibility. Since other entries (e.g., on compatibilism and incompatibilism) deal with those theories that attempt to spell-out the conditions on being responsible, my task has been to focus on the concept of responsibility itself. Below is a draft of some paragraphs on recent work in this area that I'm planning to add to the end of the entry. Using a distinction Gary Watson drew in his "Two Faces of Responsiblity," I've tried to map what I take to be an interesting trend, namely the increased attempt to articulate different aspects of the concept of responsiblity. As I note at the end, I think there's a connection between the drawing of these distinctions and the move toward revisionism discussed here in an earlier post.

I'd be happy for any feedback/corrections from this very able group. Some of the parenthetical citations are not yet complete. Thanks!

**************************

Traditionally, most philosophers writing on moral responsibility have attempted to articulate an account of those conditions on responsible agency that mapped onto what was presumed to be a unitary and shared concept of moral responsibility. However, more recently a number of authors have suggested that at least some disagreements about the most plausible overall theory of responsibility might be based on a failure to distinguish between different aspects of the concept of responsibility, or perhaps several distinguishable but related concepts of responsibility.

Broadly speaking, a distinction has been drawn between responsibility understood as attributability and responsibility as accountability. (Terminology in marking this and similar distinctions varies. Here I adopt the terminology used in Watson 1996.) The central idea in judging whether an agent is responsible in the sense of attributabilty, say for an action, is whether the action discloses something about the nature of the agent's self (Watson 1996, p. 228). Some hold additionally that a judgment of responsibility in this sense includes an assessment of the agent's self as measured against some standard (though not necessarily a moral standard)-i.e., that our interest is in what the action discloses about the agent's evaluative commitments (Watson 1996, p. 235; Bok, p. 123, nt. 1. For reservations about whether responsibility in this sense entails an evaluative judgment, see Fischer and Ravizza,(?); Haji, p. 8; and Scanlon 1998, p. 248). Perhaps the clearest example of a conception of responsibility emphasizing attributability is the so-called "ledger view" of moral responsibility. According to such views, the practice of ascribing responsibility involves assigning a credit or debit to a metaphorical ledger associated with each agent (see Feinberg, pp. 30-1; Glover, p. 64; Zimmerman, pp. 38-9; and discussion of such views in Watson 1986, pp. 261-2; and Fischer and Ravizza 1998, pp. 8-10, nt. 12). To regard an agent as praiseworthy or blameworthy in the attributability sense of responsibility is simply to believe that the merit or fault identified properly belongs to the agent.

To be responsible for an action in the sense of being accountable (or "appraisable" according to the terminology of some) presupposes responsibility in the sense of attributability. However, to judge that an agent is responsible in the further sense of being accountable entails that the behavior properly attributed to the agent is governed by an interpersonal normative standard of conduct that creates expectations between members of a shared community (whereas the standard invoked above may or may not be thought to generate interpersonal expectations). In this way, the concept of moral responsibility as
accountability is an inherently social notion, and to hold someone responsible is to address a fellow member of the moral community (Stern, Watson 1987, McKenna). By emphasizing the way the reactive attitudes were tied to expectations of good will grounded in our interpersonal relationships, P. F. Strawson drew attention to this social aspect of responsibility. Recent attempts to further articulate how best to understand the relevant notion of holding responsible and its relation to being accountable reflect his on-going influence.

An agent is praiseworthy or blameworthy, in the sense of accountable, if one is warranted, or justified, in holding her responsible. On one popular view, holding someone responsible is interpreted as regarding him or her as an appropriate candidate for the reactive attitudes and possibly other forms of reward or censure based on what the agent has done (Fischer and Ravizza 1998, pp. 6-7; Wallace, pp. 75-77; Watson 1996, p. 235; and Zimmerman, ch. 5). On another view, holding someone responsible is fundamentally a matter of making a moral judgment accompanied by an expectation that the agent who performed the act
acknowledge the force of the judgment or provide an exonerating explanation of why she performed the action. To hold someone responsible is thus to be one to whom an explanation is owed. On this view, the reactive attitudes and associated practices are derivative from this more fundamental expectation (Oshana, p. 76-7; Scanlon 1998, pp. 268-271). Since the reactive attitudes and associated practices may have consequences for the well-being of an agent (especially in the case of those blaming attitudes and practices involved in holding someone accountable for wrong-doing), they are justified only if it is fair that the agent be subject to those consequences (Wallace, pp. 103-117; and Watson 1996, pp. 238-9). The fairness of
being subject to those consequences has often,in turn, be interpreted as the source of the idea that praise and blame are justified only if they are merited (Wallace, pp. 106-7; Zimmerman, ch. 5. For the claim that the relevant notion of fairness need not be understood as desert, see Scanlon, pp. 283-7).

The recognition and articulation of diversity within the concept (or amongst concepts) of moral responsibility has generated new reflection on the nature of and prospects for theories attempting to spell-out the conditions on being morally responsible. While some continue to believe that a plausible unified theory can be offered that captures the conceptual diversity sketched above, a number of others have concluded that the conditions for the applicability of our folk concept is either metaphysically indefensible or morally suspect (Bok, ch. 1; Double 1996a, chs. 6-7; Honderich, 1988, vol. 2, ch. 1; Nagel, and G. Strawson 1986, 105-117, 307-317). For example, some have argued that while a compatibilist sense of freedom is
necessary for attributability, genuine accountability would require that agents be capable of exercising libertarian freedom (Honderich, 1988, vol. 2, ch. 1). Others have granted that while the folk concept of accountability may rest on a notion of fairness as desert, the relevant notion of desert is itself morally objectionable (Scanlon, p. 274; Kelly). Of course, there have always been those--e.g., hard determinists--who have concluded that the conditions on being responsible cannot be met and thus that no one is ever morally responsible. However, a noteworthy new trend amongst both contemporary hard determinists
and others who conclude that the conditions of our folk concept cannot be satisfied has been the move to offer a revisionist conception of moral responsibility rather than to reject talk about responsibility outright (For this general trend, see Vargas 2004 and forthcoming). Revisionism is a matter of degree. Some revisionists seek to salvage much if not most of what they take to be included in the folk concept (e.g., Honderich 1988, vol. 2, ch. 1; Scanlon, pp. 274-277; and Vargas 2004), while others offer more radical reconstructions (see Pereboom, Smart,and Smilansky).

Comments

[John Fischer]

Andrew,

This is extremely thoughtful and insightful.
One point I'd like to mention: I think it is useful to distinguish between the "concept" of moral responsibility and its conditions of application. (Sorry, Quine.) Further, it may be helpful then to distinguish revisionists about the concept from revisionists about the conditions of application. I consider myself primarily a revisionist about the conditons of application of the concept, rather than the concept itself. Specifically, I deny the commonsense notion that moral responsibility requires freedom to do otherwise (understood literally--as the freedom to add to the past, given the natural laws.)

Manuel: do you make this distinction? (Don't kill me--I'm over fifty and thus should not be held responsible for lapses of memory.

[Andrew Eshleman]

Thanks John. That does sound like a helpful further distinction.

[Manuel Vargas]

Nice work, Andrew.

For what it is worth (and that may not be much!), I'm less persuaded than John that the concept/conditions of application distinction is especially useful. For example, it seems to me that if we change the conditions under which we call something a frog we've made a conceptual revision about what it is to be a frog. But there is room for plenty of intelligent disagreement about how to think of concepts, and you could undoubtedly pick examples that look friendlier to John's view.

I do think that if you think there is a difference between conditions of applciation and the concept, then revisionist about either or both are possible. I've acknowledged this possibility at various points (e.g., the dissertation), but I'm not sure if any of it has appeared in print. In print, I've tended to treat what John would call "revisionism about the conditions of application" as an instance of conceptual revision (I explicitly include the conditions of applications in my treatment of concepts in "Responsibility and the Importancce of History").

A further thing to keep track of is revisionism about what. John is perhaps moderately revisionist about free will (though I sometimes think he is closer to what I call "weak revisionism"- the view that says that our actual commitments (or conditions of application) don't really need revising, just our mistaken construal of them), but about moral responsibility I don't think he is revisionist at all, or if he is, it isn't more than "weak revisionism." For more on why I think that, see "The Revisionist's Guide" (posted in the papers blog). But maybe John disagrees?

[Andrew Eshleman]

Thanks Manuel.

Let me see if I can use your example to get at what I take John to be saying. Suppose that it is widely believed (I think it may be) that a necessary condition on being a frog is that it have moist slimy-looking skin while toads, in contrast, have dry warty skin (Set aside the fact that technically at a general level, all toads may be classified as frogs from what I understand). Now along comes a specialist who shows that while most frogs do happen to have moist slimy-looking skin, some have dry warty skin. He has, in effect, offered an error theory and revision of the commonly accepted conditions on being a frog. Suppose the ammendment is accepted. Has the concept of frog itself been altered? I think it's tough to know what to say. But if the test is whether practice would be altered, then it seems like one should say "yes" because I will not be so quick to say the next dry warty skinned hopping amphibian I see is a toad.

Similarly, John says, people tend to think that alternative possibilities are necessary for moral responsibility, but they are mistaken; all that's needed is guidance control. Again, does this mean the concept itself is altered? Am I right John that you want to say that practice is not altered so we should say that the concept is the same?

[Manuel]

If the "effect/no effect on practices" standard is what John has in mind by concept/conditions of application distinction, I suspect we can talk about all the revisions we need to talk about by distinguishing between the target of revision and the kind of revision.

Once you specify the general target of revision (e.g., moral responsibility), there are usually 3 more particular targets you could have in mind: (1) characteristic beliefs, including conditions of application- what I generally call "the concept" (2) practices, and (3) attitudes). As I alluded to above, I think there is room for three broad varieties of revisionism: weak, moderate, and strong.

So in the frog case, I say that what we have is moderate revision about the concept with moderate revision about the practices. In John's case, I think we have weak to moderate revisionism about the concept of responsibility (depending on how he thinks we have misconstrued our actual commitments or whether he thinks we have to revise our actual commitments, irrespective of how we tend to understand them) and non-revisionism about the practices.

Concept individuation is a tricky thing, though, and given a good example or argument, I'd be happy to accept that the framework I just sketched still needs to be supplemented with an additional distinction between the concept and its conditions of application.

[Barry]

Hi, I am an amateur reader on moral responsibility, so I have a clarificatory question.

Something I found missing from the article is something that I have always taken for granted when I think about moral responsibility. When someone acts wrongly, they are morally responsible by default. Whether this is put in terms of attributability or accountability, or the fairness of reactive attitude, it is always important that it is default reasonable, or default fair, that someone is responsible when they act wrongly. Exculpating explanations or excuses always bear the burden of proof. So it seems more natural to say that excuses need to be warranted or justified, rather than attributions or accountability.

Is this something that is controversial or so commonly held that it is not worth mentioning?

[Andrew Eshleman]

Hi Barry,

Good observation. I think what you're noticing is that the aspects of reponsibility philosophers are trying to conceptually pry apart may, do not in practice, present themselves sequentially. That is, philosophers are interested in the end in offering an account of when an agent actually is responsible, but in practice an agent is often held responsible and then we "ask questions later" (for someone who may share your sense that this is philosophically significant, see Austin's "Plea for Excuses"). So, for example, I may respond with resentment when shoved from behind and then retract my reaction when I discover that the person who pushed me is an emergency medical tech. trying to reach a cardiac arrest victim.

So, I think many would accept your observation but resist saying that the default response amounts to saying they are "morally responsible by default." Rather, I would say they are treated as if they are responsible, but whether they actually are is a matter of whether the conditions on attribulity and accountability are fulfilled.

[Barry]

Andrew, here is a follow-up.

I think I interpret everyday, sequential practices of responsibility attributions to be a lot more closely linked to a philosophical explanation of the conditions of attributability and accountability than you suggest. Let me put it another way. To me, it seems that when we come up with a philosophical theory, it would suffice to come up with the conditions of excuse or exculpability. Only if they are fulfilled is someone not responsible. Otherwise they are. To me, our ordinary practices of holding someone responsible seem to warrant this kind of methodological approach.

Maybe what it comes down to is this: the more pressing question is what in the world makes true claims of exculpability...not what makes true claims of responsibility. Our everyday practices suggest that we think of responsibility as the general case, applicable to every case of wrong action; we acquire (and require) knowledge about the specific cases of excuse, so we need conditions under which the specific case obtains. These practices reflect something deep about how responsibility works (not just attributions of them.)

[Manuel]

Barry- I think that you are certainly right that when we are generally dealing with normal adults we suppose that they are morally responsible unless we have good reason to think otherwise. But I wouldn't read too much into that for the following reason: we don't start out that way. When dealing with infants, we don't suppose that they are responsible. Something has to happen to us to move from not responsible to responsible, and so we need an account of what those things are that have to happen. You could, of course, argue that young children are exculpated, and that this merely reinforces your point. But it seems to me that this puts the cart before the horse, as no human has ever started off being fully responsible, so no human has ever started off with us assuming that he or she was responsible (though maybe you could find cultures where this isn't true). There is more to be said about all of this, of course, but I take it that something like this is the reason why most philosophers don't assume that the only thing we have to do is to provide an account of exculpation. Finally, for what it is worth, I'm inclined to think that we can't make much sense of exculpation without some idea about what makes ascriptions of responsibility turn out to be true.