This paper distinguishes between synchronic responsibility (SR) and diachronic responsibility (DR). SR concerns an agent’s responsibility for an act at the time of the action, while DR concerns an agent’s responsibility for an act at some later time. While most theorists implicitly assume that DR is a straightforward matter of personal identity, I argue instead that it is grounded in psychological connectedness. I discuss the implications this distinction has for the concepts of apology, forgiveness, and punishment as well as the way in which this distinction can be used to defend quality of will accounts of responsibility against objections involving psychological manipulation. I argue that the intuition that a manipulated agent is not responsible can be explained by appeal to the conditions of DR and, as such, does not unproblematically shed light on the conditions of SR.
This is a preview of subscription content, log in to check access.
Buy single article
Instant access to the full article PDF.
Price includes VAT for USA
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
This is the net price. Taxes to be calculated in checkout.
Connectedness should not be confused with continuity. Whereas continuity is not scalar and is transitive, connectedness is scalar and is not transitive. This is because continuity is the ancestral relation of connectedness; that is, continuity consists in overlapping chains of connectedness.
Consider French: “Suppose that Sebastian gets drunk and drives his car on to Quincy’s property. Sebastian had no intention of damaging Quincy’s property. After regaining sobriety and learning of his misadventures, Sebastian, who is not yet an alcoholic, subsequently and quite deliberately returns to the local pub and proceeds to get himself roaring drunk. Though we would have excused him from moral responsibility for the damage he did on the first spree had he subsequently modified his behaviour, he did not do so and he is making the crucial past behaviour, not an out-of-character happenstance, but very much in character and hence something for which, as Aristotle would say, he may be held morally accountable” (1984, p. 499).
For some, a case involving a good act and a subsequent worsening of character may have more intuitive appeal.
See, for example, Smith (2007).
Note, too, that there is some danger that the critic is simply using ‘blameworthy’ to mean ‘synchronically blameworthy’ and ‘appropriate to overtly blame’ to mean ‘diachronically blameworthy’. If this is the case then this is not an objection to my account but simply an objection to my terminology.
One might raise a related objection by appealing to Watson’s (1996) distinction between responsibility as attributability and responsibility as accountability in order to explain the example. For Watson, attributability concerns whether an action can be properly ascribed to an agent. To say that an agent is to blame for an action in the attributability sense is simply to say that the agent committed the action and that the action has particular faults. Accountability, on the other hand, is conceptually tied to the notion of response. It is responsibility as accountability that we are concerned with when we are concerned with whether an agent deserves to be punished for her action. Appealing to this distinction, one might claim that at t 1 the insult is attributable to Jane and that she is accountable for it. Later, at t 2, she is accountable for the insult only in the second scenario though the insult is still attributable to her in both scenarios. As will become evident, my claim is that responsibility as attributability varies across time. Though it is true that there is a difference in accountability at t 2 in the two scenarios this difference is explained by a difference in attributability. This is shown when we consider, as we will, cases of extreme psychological disconnectedness between the time of the act and the time of the responsibility ascription.
This is simply to say that DR can come apart from SR. I do not mean to deny that, for example, if an agent is blameworthy to degree d at t 1 for an act X that occurs at t 1 then it is true, at some later time t 2, that the agent was blameworthy to degree d at t 1 for the act X which occurred at t 1. I am pointing out the possibility that though it is true that the agent was blameworthy to degree d for X, she may not now be blameworthy to degree d for X.
Parfit (1984, p. 326) makes some brief but highly suggestive remarks that I attempt to develop in what follows.
See Frankfurt (1971). I also believe Strawson (1962) should be read in this way. See McKenna (2005). What have been called structural or mesh theories of responsibility are QOW theories on my usage. One could accept a broadly QOW based account of responsibility and yet still hold that SR is historical. For my present purposes I wish to simply stipulate that QOW accounts are ahistorical accounts of SR.
This language comes from Fischer and Ravizza (1998, p. 171).
Shoemaker (forthcoming) offers the following list of theorists who have accepted that “moral responsibility presupposes personal identity”: Butler (1736, p. 104), DeGrazia (2005, pp. 88–89), Glannon (1998, p. 231), Haskar (1980, p. 111), Locke (1694), Madell (1981, p. 116), Parfit (1984, pp. 323–326; 1986, pp. 837–843), Reid (1785, pp. 116–117), Sider (2001, pp. 4, 143, 203–204), Schectman (1996, p. 14). I add Haji (1998), and Mele (1995).
Shoemaker (forthcoming) convincingly argues that responsibility does not presuppose identity.
Admittedly, more would need to be said to establish this claim. I won’t take up that task here since what I have to say in what follows depends on the claim that personal identity is not a sufficient condition for responsibility to freely transfer through time. It can be assumed, if one likes, both that personal identity is a necessary condition of responsibility and that it holds in the cases I discuss. Nothing hinges on this.
As quoted by Parfit (1984, p. 271).
Parfit (1984) claims that identity matters only insofar as it implies psychological connectedness and continuity (the ancestral relation of connectedness). I follow Shoemaker (1999) who has argued that what matters is primarily connectedness and not continuity, at least in the realm of survival, anticipation, and responsibility.
Somewhat more precisely some agent S’s psychological state P1 is psychologically connected to S’s later psychological state P2 if and only if P1 caused (in the right way) P2 and P2 is similar to P1. Given that psychological connectedness is a relation that comes in degrees (since similarity comes in degrees), when I say that two states are directly connected I mean only to say that they are connected to a high degree.
It may be that “maximally connected” is a threshold concept with a relatively low threshold. We may, for instance, only require a small degree of psychological connectedness in order for blameworthiness to fully transfer across time. This view is defended by Glannon (1998). He also holds that “diminished psychological connectedness does not imply diminished responsibility” (p. 231), so long as the threshold is met. This, it seems to me, is mistaken. The mistake is to think that, with respect to some past act, an agent is either responsible for that act or not. That is, that DR is not scalar. One can be led to this view when one fails to distinguish between SR and DR. Glannon focuses on SR and it is this that leads him to deny that diminished connectedness implies diminished responsibility: “What makes a person partly or fully responsible for his behavior is not so much the strength or weakness of the connections between his earlier and later mental states, but more so his intention and his beliefs about foreseeable consequences at the time of action” (p. 232). Were Glannon discussing the nature of SR he would be exactly right. The problem with his view is that it implies that an insignificant change in connectedness, from the threshold to just below, could be the difference between being responsible and not being responsible. The better threshold view holds that when the connectedness threshold is met DR is equal to SR, but that when the threshold is not met DR is diminished rather than eliminated. Though I leave it open here whether a threshold understanding of connectedness with respect to responsibility is best, I’m inclined to think that though we may employ a threshold understanding in practice, this is merely the best we can do given our epistemic limitations but that in fact, DR is sensitive to subtle changes in connectedness.
The agent-causal libertarian denies that we are caused to act because of these psychological elements. However, the claim can be reformulated in a way that does not offend the libertarian’s sensibilities. When we act we do so for reasons and these reasons are provided by beliefs, desires, values and so on even though these reasons don’t cause us to act as we do.
Darabont (1994, pp. 111–112).
Compare Murphy (2003, p. 70): “If the wrongdoer is unrepentant, he generally does not (in my view) merit forgiveness.” The claim that appropriate forgiveness requires repentance is closely related to the claim that whether forgiveness is appropriate is partly determined by the agent’s DR (it will, of course, depend on other things, too, such as whether the person has the standing to forgive). This is because it is plausible that sincere repentance involves breaking the relevant psychological connections: “[Repentance] is surely the clearest way in which a wrongdoer can sever himself from his past wrong. In having a sincere change of heart, he is withdrawing his endorsement from his own immoral past behavior; he is saying, ‘I no longer stand behind the wrongdoing, and I want to be separated from it. I stand with you in condemning it’ ” (Murphy and Hampton 1988, p. 26, italics mine).
The example also shows why the objection considered earlier won’t do. According to the objection, there is no distinction between SR and DR. The cases I have been considering are merely cases in which, though the agent is just as blameworthy for the act at the later time, it would be less appropriate to overtly blame her at the later time. Cases involving extreme psychological disconnectedness show how implausible such a response is. For in such cases the later person is, for all practical purposes, simply a different person than the agent that committed the act. It would be inappropriate to overtly blame the later agent for the deeds of the earlier agent, but this is because the later agent is not blameworthy for the earlier deeds. And she is not blameworthy for those earlier deeds because they are not her acts; they are not attributable to her.
Consider Parfit: “Suppose that a man aged ninety, one of the few rightful holders of the Nobel Peace Prize, confesses that it was he who, at the age of twenty, injured a policeman in a drunken brawl. Though this was a serious crime, this man may not now deserve to be punished” (1984, p. 326).
This may also explain the Aristotelian view that virtuous action must proceed from a firm and unchanging character as well as the related Humean claim that a bad action provides reason to blame the agent only if it was the result of an enduring character trait.
Others think that the manipulation cases show that SR has an ultimacy requirement which is incompatible with determinism, and thus that compatibilist QOW accounts must therefore be mistaken.
Some hold that manipulations undermine the ownership condition on responsibility which is an essentially historical condition. For example, compatibilists such as Fischer and Ravizza (1998, Chap. 8) hold that in order for an agent to be responsible he must have “taken responsibility” for the mechanism that led to the action, and that a manipulated agent is not responsible because he has not taken responsibility for the relevant mechanism. Others (source incompatibilists such as Pereboom) hold that manipulations undermine responsibility because they prevent the agent from being the ultimate source of her action.
Mele also tells us that “readers who are still inclined to believe that Beth is morally responsible for the killing are encouraged to replace Beth in my story with the sweetest person they know” (p. 465). It is quite plausible that this request to replace the manipulated agent with someone we know and presumably care a great deal about (he mentions his grandmother) merely introduces bias due to special relationships. The fact that a parent finds it difficult to blame her child does not shed much light on the blameworthiness of the child. It is merely a reflection of the special relationship that the parent stands to the child. See Scanlon (2008, pp. 171–173).
A third possible source of the not-responsible intuition concerns the intuition that the manipulators are themselves responsible. This is obviously true, but there is plenty of blame to go around. The fact that one is not solely responsible for some bad act does not entail that one is not fully responsible for it (see Frankfurt 1971, p. 25. fn. 10).
Some theorists have noted the related claim that manipulations themselves undermine personal identity. Arpaly (2003) discusses this possibility: “But perhaps much of our tendency to look at HP [Hapless Patient, a manipulated agent] as exempt from blame comes from our response to a third kind of scenario: HP is changed into a murderer, and then changed back…I take it that a case involving puzzles about personal identity should not be used as a test case for theories about moral responsibility” (p. 168). Other theorists who have considered this possibility include Fischer and Ravizza (1998, p. 235 fn. 30), Haji (1998, pp. 6–7), Mele (1995, p. 175 fn. 22), Talbert (2009), and Vargas (2006, p. 359).
Because I hold that manipulations can result in responsible agency this is what McKenna (2008) has called a “hard-line” reply. Indeed, all QOW theorists and, I think, all compatibilists must take a hard-line reply towards the relevant manipulation cases. This simply follows from the claim that responsible agency can arise from deterministic causal chains.
And there do seem to be actual cases in which the manipulation lasts long enough that we are inclined to hold the agent responsible (cases in which we can’t downplay the existence of the blameworthy manipulated agent). This seemed to occur when Patty Hearst was convicted for her role in a bank heist. And it seemed that she was willing to be a part of the robbery only because she was brainwashed by her kidnappers. She was, of course, later pardoned. But this seems to be explained by the “deprogramming” that apparently occurred.
It is also the question at issue between QOW theorists and source incompatibilists.
This also applies to manipulation cases that do not involve a reversal of the manipulation. It applies as long as the pre-manipulation agent is a morally decent person who is manipulated to do something morally objectionable (or vice versa). Since the pre-manipulation agent (Agent1) is relevantly disconnected from the manipulated agent (Agent2), Agent1 won’t be responsible for the bad act of Agent2 (assuming that responsibility for future acts even makes sense). In this way, Agent1’s innocence is simply not relevant to Agent2’s responsibility. Thus, the intuition that the agent is not responsible in a manipulation case that does not involve a reversal is undermined to the extent that that intuition is a reaction to the moral qualities of Agent1. The intuition that the agent is not responsible in a manipulation case that does involve a reversal is undermined to the extent that the intuition is a reaction to the moral qualities of either Agent1 or Agent3 (the post-manipulation reversal agent).
The fact that the relation that warrants ascriptions of DR admits of degrees is what has been missed by the wielders of the manipulation cases who have actually addressed worries about the transfer of responsibility through time. This is because they assume that personal identity is all there is to DR. Their general strategy in responding to this worry is to argue that in the given case it is plausible to think that personal identity is held fixed across the manipulation. The problem is that this is irrelevant. Personal identity is relevant in these cases only insofar as it is relevant to the relation that warrants ascriptions of DR: psychological connectedness. The relation of personal identity is an all-or-nothing relation. It either holds or it does not. But this is not true of psychological connectedness. This relation is scalar. So in a given manipulation case it may be true that identity is preserved (on some particular account of personal identity) while it is also true that the relation that warrants responsibility through time is diminished by some degree. For example Haji says:
Assume, in addition, that the members of the minimum competing set, in conjunction with the shared elements, are the bare minimum of psychological elements required to preserve personal identity so that we can be assured that Jenny1 is identical to Jenny2 (1998, p. 6).
The “good” Manson case and others developed in this chapter might have prompted worries about personal identity. Is the transformed Manson the same person as the pretransformation Manson? Is Beth, after becoming a Manson “twin,” the same person as the earlier Beth? This is not the place to advance a theory of personal identity. But, surely, the pre- and posttransformation agents have much in common? Manson just before his transformation (t-Manson) is much more similar, on the whole, to Manson just after it (t*-Manson) than he is to neonate Manson or toddler Manson. Still, t-Manson is the same person as the neonate and toddler Manson, in a familiar “personal identity” sense of “same person.” So what is to prevent him from being the same person, in the same sense, as t*-Manson? It is worth noting, further, that t-Manson and t*-Manson may be strongly psychologically connected, in Parfit’s sense (1984, p. 206). They may be such that the number of direct psychological connections between them “is at least half the number that hold, over every day, in the lives of nearly every actual person” (1995, p. 175 n. 22).
Both of these passages illustrate that these theorists are implicitly thinking that the relation that warrants DR must either hold completely or not at all, because they assume that this relation is that of personal identity. But it does not follow from the fact that personal identity is held fixed that the conditions of DR are also held fixed. And if the relation that determines DR is not held fixed then we cannot be sure that the manipulation affects SR, which is what these theorists conclude.
Furthermore, it is not clear that Mele’s appeal to Parfit’s Psychological Criterion can help his case. This is because Parfit, in trying to develop the most plausible version of a psychology based account of personal identity, realizes that one must draw an arbitrary borderline in order to make a scalar and non-transitive notion (psychological connectedness) fit the logic of identity (which is not scalar and is transitive). It is this arbitrariness that led Parfit to famously conclude that identity “is not what matters,” and hence cannot be the basis for DR. And even granting that the pre- and post-manipulation agents are strongly psychologically connected, it is doubtful that they are strongly psychologically connected in the relevant respects. The connections that matter concern the motivational structure implicated in the action, and it is clear that the pre-manipulated agent is not strongly connected to that psychological structure (otherwise there would be little need for the manipulation in the first place).
Arpaly, N. (2003). Unprincipled virtue. Oxford: Oxford University Press.
Butler, J. (1736). Of personal identity. In J. Perry (Ed.), Personal identity. Berkeley, CA: University of California Press.
Darabont, F. (1994). The Shawshank redemption: The shooting script. New York: Newmarket Press.
DeGrazia, D. (2005). Human identity and bioethics. Cambridge,MA: Cambridge University Press.
Fischer, J. M., & Ravizza, M. (1998). Responsibility and control: A theory of moral responsibility. New York: Cambridge University Press.
Frankfurt, H. (1971). Freedom of the will and the concept of a person. In The importance of what we care about (pp. 11–25). Cambridge, MA: Cambridge University Press.
Frankfurt, H. (2002). Reply to John Martin Fischer. In S. Buss & L. Overton (Eds.), Contours of agency: Essays on themes from Harry Frankfurt. Cambridge, MA: The MIT Press.
French, P. A. (1984). A principle of responsive adjustment. Philosophy, 59(230), 491–503.
Glannon, W. (1998). Moral responsibility and personal identity. American Philosophical Quarterly, 35(3), 231–249.
Haji, I. (1998). Moral appraisability: Puzzles, proposals, and perplexities. Oxford: Oxford University Press.
Haksar, V. (1980). Equality, liberty, and perfectionism. Oxford: Oxford University Press.
Kane, R. (1998). The significance of free will. Oxford: Oxford University Press.
Locke, J. (1694). An essay concerning human understanding. Partly reprinted in J. Perry (Ed.), Personal identity. Berkeley, CA: University of California Press.
Madell, G. (1981). The identity of the self. Edinburgh: Edinburgh University Press.
McKenna, M. (2005). Where Frankfurt and Strawson meet. Midwest Studies in Philosophy, 29, 163–180.
McKenna, M. (2008). A hard-line reply to Pereboom’s four-case manipulation argument. Philosophy and Phenomenological Research, 77(1), 142–159.
Mele, A. (1995). Autonomous agents. New York: Oxford University Press.
Mele, A. (2009). Moral responsibility and history revisited. Ethical Theory and Moral Practice, 12, 463–475.
Murphy, J. (2003). Getting even: Forgiveness and its limits. New York: Oxford University Press.
Murphy, J., & Hampton, J. (1988). Forgiveness and mercy. Cambridge, MA: Cambridge University Press.
Parfit, D. (1984). Reasons and persons. Oxford: Oxford University Press.
Parfit, D. (1986). Comments. Ethics, 96, 832–872.
Pereboom, D. (2001). Living without free will. Cambridge, MA: Cambridge University Press.
Reid, T. (1785). Of Mr. Locke’s account of our personal identity. In J. Perry (Ed.), Personal identity. Berkeley, CA: University of California Press.
Scanlon, T. M. (2008). Moral dimensions: Permissibility, meaning, blame. Cambridge, MA: Belknap Press of Harvard University Press.
Schectman, M. (1996). The constitution of selves. Ithaca, NY: Cornell University Press.
Shoemaker, D. (1999). Selves and moral units. Pacific Philosophical Quarterly, 80, 391–419.
Shoemaker, D. (forthcoming). Responsibility without identity. Harvard Review of Philosophy, XVIII.
Sider, T. (2001). Four-dimensionalism. Oxford: Oxford University Press.
Smith, A. (2007). On being responsible and holding responsible. The Journal of Ethics, 11, 465–484.
Strawson, P. (1962). Freedom and resentment. Proceedings of the British Academy, 48, 187–211. Reprinted in G. Watson (Ed.) 1982.
Talbert, M. (2009). Implanted desires, self-formation and blame. Journal of Ethics and Social Philosophy, 3(2). www.jesp.org. Retrieved July 2010.
Vargas, M. (2006). On the importance of history for responsible agency. Philosophical Studies, 127, 351–382.
Watson, G. (1996). The two faces of responsibility. Philosophical Topics, 24, 227–248.
Wiggins, D. (1976). Locke, Butler and the stream of consciousness: And men as a natural kind. In A. O. Rorty (Ed.), The identities of persons. Berkeley, CA: University of California Press.
For helpful discussion and comments, I would like to thank Cheshire Calhoun, John Martin Fischer, Peter French, Michael McKenna, Douglas Portmore, David Shoemaker, and an anonymous referee at Philosophical Studies.
About this article
Cite this article
Khoury, A.C. Synchronic and diachronic responsibility. Philos Stud 165, 735–752 (2013). https://doi.org/10.1007/s11098-012-9976-6
- Free will
- Moral responsibility
- Personal identity
- Quality of will