Responsibility Without Identity

It’s taken to be a platitude of folk morality that I can only be morally responsible for my own actions.  Call this The Platitude.  Sometimes The Platitude is presented in a more expansive form: (a) I can be responsible for my own actions; and (b) I cannot be responsible for anyone else’s actions.  This platitude is then taken to entail what I’ll call The Slogan: moral responsibility presupposes personal identity.  Classical philosophers who have embraced The Slogan include Locke, Reid, and Butler, and contemporary philosophers who do so include DeGrazia, Glannon, Haksar, Madell, Parfit (on one reading), Schechtman, and Sider.  Nevertheless, The Slogan is false.  Responsibility doesn’t presuppose identity, even if The Platitude is true.

What I say here will be extremely compressed, so feel free to demand details.  In this post I will be concerned only with those authors who believe that responsibility presupposes numerical identity, that the earlier Φ-er and the later person responsible for Φ-ing must be one and the same person or individual.  This is in fact the view held by all the authors just mentioned.  Now some may object that Schechtman, for instance, holds a version of The Slogan with respect to narrative identity, a uniting sort of conception of identity (where various actions and experiences are united into the life of one person via a self-told narrative), but implicitly for her (as explicitly for DeGrazia) narrative identity actually presupposes numerical identity, insofar as hers is a theory about what makes experiences occurring at different times properly attributable to one and the same person or consciousness.  So I will assume here that all these theorists think that moral responsibility presupposes numerical identity.  In a later post I may show how moral responsibility doesn’t presuppose narrative-identity-sans-numerical-identity either.

            To see what’s going on, then, consider an outline of an argument that’s representative of these theorists’ general approach:

1.                  One is morally responsible only for one’s own actions.  (The Platitude)
_________________________________________________________

2.                  Thus, an action is one’s own only if one is numerically identical with the performer of the action.  (The Slogan)

3.                  X is numerically identical with Y iff X bears some specified numerical identity relation to Y.
__________________________________________________________

4.                  Thus, an action is one’s own only if one bears some specified numerical identity relation to the performer of the action.
__________________________________________________________

5.                  Thus, one is morally responsible only for those actions for which one bears some specified numerical identity relation to the performer of the action.

The general idea, then, is to apply one’s theory of numerical identity to the criterion of ownership of actions.  So an advocate of the Biological Criterion (like DeGrazia) might wind up saying that one is responsible only for those actions for which one bears the relation of unique biological (animal) continuity to the original agent.  Or an advocate of the Memory Criterion (like Locke) may say that one is morally responsible only for those actions for which one has a consciousness (memory) of the performance by the original agent.  And the same goes for the advocate of the Psychological Criterion: one is morally responsible only for those actions for which one is uniquely psychologically continuous with the original agent.

            The problem is that all these applications of specific criteria of personal identity succumb to powerful counterexamples.  One obvious one is to Locke’s Memory Criterion: if Mel is caught for drunk driving and then yells anti-Semitic epithets at the arresting officer but fails to remember anything of the episode the next day, it seems clear he’s still morally responsible for his action, that the action was indeed his own.  And one can think of counterexamples to the other criteria as well, examples where the action is the agent’s own action (and thus he’s responsible for it), even though the proffered criterion of identity doesn’t obtain.

            But the most comprehensive counterexample comes from the fission case: if I were to split in two and each half re-grew its missing half, I couldn’t be identical with either survivor (they’d be exactly similar and so there’d be no non-arbitrary reason for me to be one or the other), and I couldn’t be identical with both (2≠1), so I must not be identical with either.  Nevertheless, were I to perform some immoral action and then undergo fission, it seems our intuitions would be that both fission products are (at least to some degree) morally responsible for my actions.  But how could my actions be theirs if they’re not identical with me?

            Here is where someone like Sider steps in and suggests a four-dimensionalist ontology to save the general approach.  There were, we can say, two space-time person-worms that were entirely spatially coincident for the stretch of time up until fission, at which point they split.  So both fission products are morally responsible for their own actions insofar as they are both one and the same person as the original agent.  Indeed, says Sider, its ability to resolve puzzles like fission is one motivation for adopting a four-dimensionalist ontology generally.

            I think this move is unmotivated, however, simply because premise 1 (The Platitude) doesn’t entail premise 2 (The Slogan).  All I’ll do here is offer an argument by analogy, and then I’ll let you hash it out.  Here is a platitude about taxation: I can be legitimately taxed only on my own property, and never on someone else’s property.  In one sense this is true, but in another sense it’s false.  It is true that I cannot be taxed on your property if it is exclusively yours; it is false that I can’t be taxed on your property, however, if you and I share ownership of the property in question.  We can accept, then, that a person can only be taxed on her own property, and that one person can’t be taxed on anyone else’s property, without having to accept that taxation presupposes exclusive property ownership.

            Similarly, then, we can accept that a person can only be responsible for her own actions, and that one person can’t be responsible for anyone else’s actions, without having to accept that responsibility presupposes personal identity.  The reason is that there are two senses in which I can’t be responsible for someone else’s actions.  It is true that I can’t be responsible for actions that are exclusively yours; it is false that I can’t be responsible, however, for actions of which we share ownership.  The fact that one owns an action doesn’t necessarily imply that one owns it exclusively, and that, therefore, there is a one-to-one identity relation between the current owner and the original agent.  Instead, it just implies that one has a special ownership relation with the original agent’s action.  But given that ownership isn’t necessarily tied to exclusivity—a single piece of property may have many owners, after all—one may easily deny The Slogan: The Platitude simply doesn’t entail that responsibility presupposes identity.  Instead, it entails only the importantly different slogan that responsibility presupposes ownership.  But if this is true, then the motivation to go four-dimensionalist is undercut.  One can handle the fission case without violating The Platitude: both of my fission products can be responsible for my actions pre-fission, insofar as both share ownership of them.  This renders the identity of the fission products irrelevant, at least with respect to responsibility, and so renders the difficulties associated with the various criteria of identity irrelevant as well.  What we need now, of course, is an analysis of ownership of actions, but that’s for another day.

34 Replies to “Responsibility Without Identity

  1. Quick question: Why doesn’t the ownership relation presuppose, in the relevant sense, personal identity?
    If it does, then I’m unsure how the argument is supposed to go.

  2. Christian: What do you mean by “the relevant sense”? If it’s numerical identity, which is a 1-1 relation, then ownership doesn’t presuppose identity given that both of my fission products may own my actions — they’re properly attributable to them — despite the fact that they aren’t me.

  3. I’m not sure that the given counter-examples work against the platitude and the slogan. In these cases, we can either reject the platitudes or the suggested criteria for identity. The former option seems less revisionary and more plausible to me. What the cases then rather show is that the assumed criterion for personal idendentity does not match the ones we are using in attributing personal identity.
    In Mel’s case, it is difficult to see how any plausible criterion of personal identity could give the outcome that Mel was not the same person when drunk. He still had most of the same mental states and was the same animal. Similarly, the fission case seems to show that personal identity is not necessarily coextensive with numerical identity – there could be many me.

  4. David
    Very interesting post. After my initial reading I am wondering how you would handle someone who has been ‘brainwashed’ into performing an action? Based on the memory criterion it would seem to me that one would argue that Person A (PA) before being brainwashed is a different ‘person’ from Person A after being brainwashed (PAbw). However, on the biology criterion they would be one and the same. It would seem to me that they would be one and the same on the psychological criterion even though PAbw would not remember being brainwashed there would still be a psychological continuity.
    Regarding being morally responsible for performing an action, if PA performed an action then we would hold him responsible for the action if the reason for performing the action originated with him. However, if PAbw performed an action, I think we would not hold him responsible for the action because the reason for performing it did not originate with him. If this were correct, then the person who brainwashed PAbw would be the one responsible for PAbw’s actions because the reason for PAbw performing the action originated with the person who did the brainwashing. Thus, one can be responsible for the actions of others.
    One might argue that what PAbw did as a result of being brainwashed is not really an ‘action’ because for some event to be an action the explanatory and justificatory reasons for acting must originate with the person doing the action. Or, if it is an action, it is an action of the person who did the brainwashing and PAbw is simple the surrogate standing in for that person.

  5. Dave,
    I’m not sure that the issue here is essentially one of identity.
    The Slogan, I take it, can be stated as follows:
    Slogan. A person P is responsible for an action A only if there exists a person P’ such that P’ performed A and P = P’.
    But this is logically equivalent to the following:
    Slogan*. P is responsible for A only if P performed A.
    Notice, the identity statement, P = P’, has dropped out entirely. So it seems the Slogan doesn’t really say anything about identity.
    I don’t mean this to be major criticism. Your argument might be good; I just worry that framing it in terms of identity confuses matters.
    Jussi,
    I’m puzzled by this:

    Similarly, the fission case seems to show that personal identity is not necessarily coextensive with numerical identity – there could be many me.

    First, I don’t see why the fission case is required to show that personal identity is not coextensive with numerical identity. More mundane examples are sufficient to show this. My bicycle is numerically identical to itself. But my bicycle is not a person. So this is an instance of numerical identity without personal identity. So the two aren’t coextensive.
    Second, I don’t see how you could be many, with or without fission. I think the following is a logical truth: for any X and Y, if X = you and Y = you, then X = Y. (This is a straightforward consequence of the symmetry and transitivity of identity.) But this is just to say that there’s only one you.

  6. “It’s taken to be a platitude of folk morality that I can only be morally responsible for my own actions. Call this The Platitude. Sometimes The Platitude is presented in a more expansive form: (a) I can be responsible for my own actions; and (b) I cannot be responsible for anyone else’s actions.”
    Do we need to rely on fancy sci-fi counter-examples to question this? Suppose Jussi coerces or dupes David into assaulting me. It then surely makes considerable sense for me to hold Jussi responsible for what David has done…

  7. Thanks to everyone who’s commented thus far.
    Jussi: I don’t argue that we should abandon the platitude and the slogan. I argue merely that we should abandon the slogan (that moral responsibility presupposes PI), insofar as its application to the platitude has led to serious counterexamples and metaphysical extravagance. This is unnecessary, especially since (my main point) the slogan just isn’t entailed by the platitude. So we can preserve the platitude (that I can be responsible only for my own actions) without making the mistake of trying to apply a particular criterion of PI to moral responsibility.
    John: Interesting point. In the longer version of this argument I discuss brainwashing and how, as you say, such cases yield real problems for those who want to apply a biological or psychological criterion of PI to responsibility. I was thinking explicitly about the Patty Hearst case. Consider a case like that: someone is brainwashed into joining a cause, robs a bank while under the influence of her captors, and then eventually “comes to her senses.” My intuition is that the post-brainwashed Patty isn’t responsible for the bank robbery, despite her biological and psychological continuity with the bank robbery. Indeed, this is the sort of case a narrative identity theorist would find to motivate a move to narrative identity. There are problems with that view too, though I won’t go into them here.
    Campbell: You’re right. My target, though, is those theorists (and there are a lot of them!) who take the slogan to be entailed by the platitude, and so think it incumbent on them to then apply a criterion of PI to the case of responsibility. Nevertheless, I doubt your reformulation is entailed by the platitude either, so it would be hard to get any version of the Slogan motivated by the platitude (which I want to preserve).
    Jimmy and Robert: These are by no means uncontroversial cases of responsibility for others’ actions, and they would surely be rejected as such by the theorists in question. Instead, they’d say, while we may hold the parents responsible, for example, in light of the child’s action, the parents are actually responsible for their own actions, namely, a kind of negligence in watching over the child, or poor parenting in general. The coercer case might be a bit trickier, but a similar sort of reply could be presented, I think. But in any event, Jimmy, you seem to be objecting to the platitude, which I wasn’t doing. Indeed, it’s easy enough to see that your case actually isn’t an objection to the platitude, for it sounds as if you think the assault is properly attributable to Jussi after all, and attributability is the necessary condition of responsibility according to the platitude (not identity).

  8. For some reason my name isn’t showing up on the “Comments” list on the homepage. I’m adding this comment to see if it’ll list my name now.

  9. Campbell,
    I’m confused. Did I say that the example is required to show that personal identity is not coextensive with numerical identity? Yes, there are other examples where the two come apart. Fission case seems to be a more interesting sort of instance though.
    The logical truth you refer to goes only for numerical, token-identity. It doesn’t apply for type-identity. It could be that in the odd fission case I’m turned into a type of which there are many tokens which are identical in one sense but not in the numerical sense.
    David,
    I’m confused by the dialectic. In the reply, what do you refer to with ‘this’ in ‘this is unnecessary’? Also, what is the evidence that applying the slogan leads to counterexamples and metaphysical extravagance? It does so with some views about personal identity but as far as I can see this is just an argument against those views and not the slogan.
    Also, I would like to see a view about diachronic ownership that does not rely on identity of the same person owning the same property earlier. Even in the taxation case, the same person has to have had owned her share earlier. Saying this seems to require that we are using some criteria for her sameness throughout the time.

  10. Jussi: By “this is unnecessary” I meant that it’s unnecessary for us to go in a route that yields the serious counterexamples and requires metaphysical extravagance, for there’s a simpler way to preserve the platitude, namely, by adopting the tempered slogan that responsibility presupposes ownership.
    As for the counterexamples, in the longer paper from which this post is just a teaser, I survey the four main theories of identity on offer: (1) Locke’s memory criterion, (2) a biological/animal continuity criterion, (3) a psychological continuity criterion, and (4) a narrative identity theory. In applying them to the case of responsibility, each has counterexamples it’s easy enough to cook up: just imagine a case in which an action is properly attributable to an agent despite the fact that the particular identity relation doesn’t obtain. But all the theories stumble over the fission case, and so rescuing the methodology at issue (applying theories of PI to responsibility) is a major motivation for Sider, e.g., to go 4D. But of course 4D is very controversial, so I’m simply suggesting that we can avoid the problems and the metaphysical extravagance altogether by recognizing that the platitude from which we started doesn’t entail that moral responsibility presupposes PI at all.
    Finally, why think identity is being presupposed in the taxation case, that the person who’s now being taxed must have been identical to a person who owned the property earlier? I might have just inherited my father’s business, which he co-owned with you, and it’s still perfectly legitimate to tax me on what is both yours and my (now) property. The taxation case is really to reveal the thought that there’s no necessary 1-1, exclusivity relation between some piece of property and the owner of that property. For numerical identity to hold between an action and the owner of that action, however, there is a necessary 1-1 exclusivity relation.

  11. David: Is there anyway to access your longer paper. I am sure that there are othrs besides me that would like to read it.

  12. David: Is there anyway to access your longer paper. I am sure that there are othrs besides me that would like to read it.

  13. David: Is there anyway to access your longer paper. I am sure that there are othrs besides me that would like to read it.

  14. David,
    The problem I’m having is understanding why you say your platitude is a platitude. “I’m responsible for Joe’s actions” isn’t a contradiction or anything. Unless, of course, “my actions” are “the actions for which I’m responsible”. Off hand, it seems I’m responsible for all sorts of things that aren’t my actions.

  15. David,
    thanks. It isn’t very easy for me to come up with counter-examples to psychological connectness views or narrative views. But, if you do have ones in the paper that will be exciting. I guess my own view is that in attributing moral responsibility to others we are implicitly using some criterion of identity and that the counterexamples only show that this criterion is not captured by the suggested philosophical views. Attributions of responsibility seem to be the best access we have to our notion of personal identity. If we deny the connection – that a person is an unit of responsibility, then it is difficult to see what we would mean by being the same person.
    I don’t think the inheritance tax case works for you at all. It is not a case where you are taxed for a property that you did not own before. You are being taxed for property that is not yours. This is nothing like an action that was not mine earlier that I would be responsible for now.
    I think a lot of work in the argument (especially in the fission cases) against the slogan is done by the notion of numerical identity. This only works if those who defend the slogan read personal identity as numerical identity. I’m not sure all of them do.

  16. Jussi,
    On the first issue. I thought you would want to use the least controversial example necessary to make your point. But perhaps your point is more radical than originally stated. Perhaps you want to say, not merely that personal identity is not coextensive with numerical identity, but moreover that personal identity can be instantiated without numerical identity; or, to put it another way, that a person may bear the relation of personal identity to a person other than himself. (My mundane example went the other way: numerical identity without personal identity.) In the fission case, you want to say, the two post-fission persons are numerically distinct yet personally identical in virtue of being tokens of a common type. So, roughly, your idea is to read the ‘identity’ in ‘personal identity’ as qualitative rather than numerical.
    That seems odd. Normally we think of personal identity as the relation of ‘being the same person’. But being tokens of a common type seems insufficient for being the same person. You and I are tokens of a common type: we’re both philosophers. But we’re not the same person.
    On the second issue. Your suggestion that fission would transform you into a ‘type’ seems bizarre too. Do you think that every type of person is itself a person? To be a philosopher is to be a token of a certain type of person. Is that type itself a person? In the fission case, how many post-fission persons are there? It seems you want to say three: one type and two tokens. That seems like one person too many.
    In any event, let’s suppose that fission would transform you into a type. Then how many types would you be? Just one. So you would still be one, not many.
    Perhaps we’re talking past each other. When you said that you might be many, I understood you as saying that there might be more than one thing to which you bear the relation of numerical identity. For the reasons I gave before, I think this is a logical contradiction. But perhaps you meant something different, namely, that there might be more that one thing to which you bear the relation of type to token; you might be a type of which there are many tokens. This seems implausible, because it’s doubtful that you could be a type, but I guess it’s not a logical contradiction.
    Still, it seems to have absurd consequences. If by the number of things that a thing is we mean the number of things that are tokens of that thing, then there will many things that are zero things, because they are not types at all and so have no tokens.

  17. Campbell,
    I think this:
    “In the fission case, you want to say, the two post-fission persons are numerically distinct yet personally identical in virtue of being tokens of a common type. So, roughly, your idea is to read the ‘identity’ in ‘personal identity’ as qualitative rather than numerical.”
    is a nice way of putting what I had in mind.
    I’m not sure I see yet the absurd consequences of this view. Many of the oddities are created by the odd fission cases which as we know strech our ordinary concepts in any case. I don’t think I have to give up the thought that personal identity is a matter of being the same person as long as ‘the same person’ is read in a type sense.
    You are right that just being of any old same type is insufficient for being the same person. I would assume that one would have to have a view about the criteria of personal identity to restrict which types count – only the ones that satisfy those criteria.
    I think the number problem would be solved by disambiqueting ‘person’ – one type and two tokens, how many persons? Well, depends on which sense you use ‘person’, the type or the token one. There is no sense in which there are three.

  18. I’m not persuaded by the analogy. It seems to me that if I own a piece of property, and then undergo fission, this is going to give my state tax authority a lot of headaches. Conversely, if I co-own a piece of property, sure I can be taxed on it, but that’s because *I* co-own it.
    My impression is that David wants to say something like this: in the fission case, the two products (B and C) share responsibility for A’s previous actions (A being the source of the fission). But why would they? They weren’t around. *They* didn’t co-perform A’s action.
    I’ve always thought that our concepts of responsibility (and perhaps personhood) are not designed to handle cases of fission and fusion (because they don’t arise) so we shouldn’t lean very heavily on any intuitions we have in such cases.

  19. Robert: I wasn’t aware that justifications had to be given for platitudes. Aren’t they just asserted as such, then counted as data to be explained? 🙂
    At any rate, this is presented as an obvious platitude throughout the literature. To present a putative counterexample to it — that I can be responsible for Joe’s actions — thus is taken to require some detailed explanation. How does the case go, exactly? What kind of responsibility is at issue? Is this a case of someone taking on responsibility for something he didn’t do? Is it a case of being responsible for a subtly different action than Joe’s? And so forth. I’m actually agnostic about the platitude itself. My main point here was to show that, taking the platitude to be a platitude, as all the theorists under discussion do, it doesn’t yet follow that moral responsibility presupposes PI, and so the methodology of applying theories of PI to cases of responsibility is unmotivated (which also may explain their failure to present a counterexample-free theory).

  20. Jussi: Your view of identity, insofar as I understand it, is extraordinarily nonstandard. I wonder if what you have in mind is actually the view that personal identity presupposes moral responsibility! In other words, you’ve got some conception in mind according to which our practical concerns determine in some way the right view of personal identity. It’s possible that Locke’s view was like this, but that’s a very nonstandard reading, and anyway, for my purposes, all the theorists in question hold the opposite view.
    As for the tax case, well, I thought you said you’d wanted a case in which a current property owner wasn’t identical to the past owner of the exact same piece of property, and that’s precisely what I gave you.
    You say:

    I think a lot of work in the argument (especially in the fission cases) against the slogan is done by the notion of numerical identity. This only works if those who defend the slogan read personal identity as numerical identity. I’m not sure all of them do.

    What’s the basis for your uncertainty? Any texts I’ve overlooked or misread that you could draw my attention to? I’d sure hate to hold someone responsible for a view that wasn’t theirs.

  21. Heath:
    You say:

    My impression is that David wants to say something like this: in the fission case, the two products (B and C) share responsibility for A’s previous actions (A being the source of the fission). But why would they? They weren’t around. *They* didn’t co-perform A’s action.

    You’re right: much depends on our intuitions in this case. If you deny that we have any clear intuitions in the fission case, then perhaps my later moves are unmotivated. Nevertheless, reactions in this case from those I’ve polled are strong and unambiguous: both of my fission products would bear at least some responsibility for my actions. This reaction is even more pronounced when they say, “But I wasn’t around — I didn’t do it!” for it seems ridiculously disingenuous and, in fact, irrelevant.
    I don’t think our concept of responsibility was “designed” at all, let alone not designed to handle cases of fission or fusion. The structures of our practical concerns are actually remarkably dynamic and can/could handle all sorts of cases. And I don’t think the set-up of fission is at all difficult to imagine (in various versions), nor is it difficult to figure out how I’d react to it. But perhaps I’m just a philosophically tainted SOB.
    As for the taxation case, again, this was intended to make clear that ownership isn’t an exclusive relation: a piece of property doesn’t entail there’s just one owner. So too ownership of an action doesn’t have to presuppose an exclusive relation: an action doesn’t entail there’s just one owner, as it would have to if responsibility presupposed identity. For if I owned some past action and you owned that same action, and ownership presupposes identity, then we’d both be identical with the original owner, which would make us both identical to each other, which we clearly are not. So the fact that we can each rightly say, “I (co-)own that action” doesn’t at all mean that either of us is identical to the original agent.

  22. David,
    thanks. It could be that my view is non-standard. I did have something like inferential semantics for personal identity in mind in which one part of what gives meaning to our claims about the persons being the same ones is that from this one can infer moral responsibility. The fact that we use sayings like she is not the same person anymore as an excuse for responsibility seems to support this. Would such an excuse be available on your view?
    I’m not sure I wanted this:
    “a current property owner wasn’t identical to the past owner of the exact same piece of property”
    this sounds rather trivial when we sell and buy stuff from others constantly. I thought the point of the tax case was supposed to be something more substantial.
    The challenge I had in mind is that ownership (of even a share of x) through time seems to presuppose personal identity. You wanted to share the platitude. That implied that I can only be responsible for my own past actions. In the fission cases, it seems like a fission-product can come to share the ownership of a past action with the other fission-product. This is fine. But, there still needs to be a story of why the past action would be the fission-product’s own. I worry that this story will have to rely on some implicit criteria of personal identity. The product is in some relation R with the doer of the action. I cannot see how there could be an account of this relation that did not rely on our views about personal identity.
    I wish I had the texts with me. I’m sure you are right about DeGrazia. I would like to hear more about why Schechtman would be implicitly talking about numerical identity. Parfit is an interesting case. If I remember him right, he thinks that in the tricky cases our criteria for using personal sameness give up – there is no fact of the matter which one of the products is the same as x before. Instead, we should concentrate on psychological connectedness that can equally well in these cases connect many people after fission. It isn’t a big step from this to say that on this ground there can be cases where in future there could be many tokens that hold an equal claim for being the earlier person.

  23. Jussi, you say:

    The fact that we use sayings like she is not the same person anymore as an excuse for responsibility seems to support this [my view of identity]. Would such an excuse be available on your view?

    Sure, why wouldn’t it? For one thing, this can’t be literally meant without contradiction (that she, that continuing person, isn’t the same person as she was before), so it’s a gloss for something, usually a great psychological change. But then if it’s just the psychological change that’s the source of the excuse, that could be what undermines ownership as well (in certain cases), and so could easily count as an excuse on the view I’m proposing.
    Indeed, an important point here is that the argument I raise motivates greater exploration of the nature of ownership (which you’re calling for as well). But given that ownership (in the taxation case and also in the fission case) doesn’t presuppose exclusivity, it can’t presuppose numerical identity. Now you’re right, there’s got to be some relation between the current owner and the action. What is it? I have some ideas (which I ain’t yet sharing with this crowd), but it’s not numerical identity with the original agent that’s at issue.
    As for Schechtman, she’s talking about what it is that unites the various experiences of one and the same subject of experiences, and while the uniting aspect comes through a self-told narrative (and thus hers is a story about narrative identity), the identity of the subject itself (whose various experiences may or may not be united) is precisely an issue of numerical identity.
    As for Parfit, he actually does say (in his later work) what he should have said in his earlier work, namely, there is a fact of the matter in the fission case: I do not survive; the fission products are different people. But he argues that PI doesn’t matter anyway. Now in R&P, he says that there could be two attitudes to take. The Extreme Claim holds that only the deep further fact of PI matters for responsibility, and insofar as there is no such fact, there’s no such thing as responsibility. The Moderate Claim holds that what matters in PI — psychological connectedness and/or continuity — matters for responsibility, so perhaps responsibility presupposes psych. connectedness (Glannon accepts a view sort of like this). But this doesn’t get you anywhere near the claim that “there can be cases where in future there could be many tokens that hold an equal claim for being the earlier person.” It’s not a theory of identity, after all; instead, it’s just a theory about what matters in identity.

  24. David,
    Maybe I’m missing something, but AFAICT it’s consistent to say that
    (a) if S is responsible for action A, then there is a fact about S’s numerical personal identity over time, and
    (b) action A might have been performed by more than one person.
    I would think this is what we say about cases of conspiracy, collective actions, etc. E.g. we might both beat someone up, be charged with assault, and be held responsible for the crime and the hospital bills, without anyone making an effort to distinguish which damages you inflicted and which I inflicted. The point of “responsibility presupposes identity”, I take it, is to make sure you did the action for which you are being held responsible. That’s separate from whether anyone else did it too.
    Also, I agree that no one actually designed our concepts of responsibility. Obviously. More like, they evolved. My point was that they evolved for the needs of societies in which personal identity is not ordinarily a very hard problem. I would bet that if we did fiss or fuse, we’d have rather different criminal procedures, for example. Evidently you think not. I don’t know how to resolve that dispute. But here is a sample of my puzzlement:
    Suppose I commit a crime that deserves 30 years in jail. Between the commission of the crime and the sentencing, I split into 30 individuals. Do my descendants get one year each? That seems too trivial. (And what do we do if they split while in prison?) Do they all get 30 years? That seems like overkill. If half of them fuse back again, does one guy get 15? How could we possibly decide this? With current ways of assigning punishments, I have no idea. One non-current way would be to adopt a non-retributive philosophy of punishment. But then the metaphysics of personal identity has made a big difference in how we think about responsibility.

  25. Heath: Stipulate that S and T are two different persons. If S is responsible for A, then S owns A. (That’s what the platitude says.) If T is responsible for A, then T owns A. If ownership consists in the relation of personal identity, though, S=T (by transitivity), which is false by stipulation. So what you want in the collective action case (although I have my doubts about the case) is that it’s true of both criminals that that was their (i.e., each one’s) action, and to get that we don’t need it to be the case that each is identical to one or the other persons who performed the original action. I’m only going to hold you responsible for your actions, those actions that are properly attributable to you, and whether or not that means you’re in fact identical to the person who performed the action is irrelevant.
    As to the practical concerns you cite about fission-responsibility, I don’t see how that’s relevant to the original question, which is about whether or not our intuitions are that my fission products are (at least in part) responsible for my actions. Such an intuition is about a certain sort of legitimate attribution (expressed or not), but the details about what such products deserve in light of that attribution is a wholly separate matter, about which I too have no set views (but don’t need to for purposes of this issue).

  26. Jussi,
    I don’t know the relevant literature very well, but my impression is that ‘personal identity’ is a philosophical term of art. When philosophers say that personal identity is a kind of numerical identity (as they often do), they do not mean to make a substantive claim about the correct analysis of some folk concept; rather, they mean to stipulate the definition of a technical term. If you use the term ‘personal identity’ to mean something else, then you’re just changing the subject. You’re not disagreeing with those who hold the standard view; you’re just talking past them.
    Perhaps others, like David, who know the literature better than me could say whether this impression is correct.

  27. That seems right, Campbell. There are some who pay some passing attention to a more “folk”-based version of the concept of personal identity, but they either take it as metaphor or as really about some other concept altogether. Given that it’s typically metaphysicians who’ve delved into the notion of personal identity, the field is most often taken to be a specific application of the general metaphysical inquiry into the identity of objects, and so the same understanding of (numerical) identity is in play.

  28. My understanding is that “personal identity” is jargon for “numerical identity of persons” where this is an instance of “numerical identity of Fs”. Thus we could have discussions about highway identity, nation identity, or tree identity, it’s just that these topics carry less interest.

  29. Campbell,
    I think that’s right. I just wonder why it is me who is changing the subject. We start from a *platitude*. I take it that platitudes are something that belong to the realm of folk-concepts. The particular platitude we are interested in says that I can only be responsible for my own actions. Now, the question is whether this claim *entails* something about personal identity.
    Here’s one neutral way of phrasing what many have thought the platitude entails: I can only be responsible for an action if I am *the same person* as the one whose action is in question. Now, there are two ways to read that *the same person* part here in what is entailed.
    We could read it as a placeholder for what we have stipulated in the philosophical definition, x, about the numerical personal identity. I’m with David that it would be a bizarre coincidence if a platitude in the folk responsibility-language would presuppose a philosophical, technical notion of personal identity. It would be odd if folk were applying the concept of ‘responsibility’ on the basis of a tacit understanding of some sophisticated philosophical theory about personal identity. That is implausible in so many ways.
    However, if we understand the phrase ‘the same person’ in what the platitude we start with in the same way as those folks do whose platitude is in question, then I do believe the entailment is worth defending. It is not implausible that the folk has some non-philosophical conception of what is required for being the same person over time and that this conception underwrites their attributions of responsibility. I would be interested in seeing arguments against this entailment which I take to be more interesting.
    Anyway, I think this way of disambiqueting ‘the same person’ is real progress in our debate, so thank you.

  30. Now, I am commenting purely in a bystanding position, I am an undergraduate at Edinburgh and as such should be treating things with kid gloves. However, I would like to pose a couple of questions to the author, even if doing so will reaveal a large abyss of ignorance on my behalf.
    I am by no means in a position to discuss many of these issues in as capable a manner as yourselves, however in reading this I have a couple of points.
    Firstly, while adequately showing that the Platitude doesnt entail the Slogan, I see no reason to assume that the Slogan isnt itself true, but not as a consequence of the Platitude? In other words, can one have responsibility without some sense of personal identity?
    Secondly, can it not also be noted that the Kantian argument against an Ontology of God be applied here. In folk morality, as this post indicates, the Platitude is thought to entail the Slogan, yet invoking a Kantian sense of ‘existence preceding essense’ it would be impossible for the Platitude to lead to the Slogan, for the existence of the agent is necessarily prior to action which is necessarily prior to responsibility?
    If I am barking up the wrong philosophical tree please feel free to throw words or sticks to divert my attention.
    Matthew

  31. Hi, Matthew, and welcome. Regarding your first point, yes, The Slogan might be true independently of the Platitude. One of my points, though, is that the primary motivation for The Slogan is its presumed entailment from The Platitude, and if we can get all we want from The Platitude itself (that responsibility presupposes ownership), why try to defend the Slogan at all? What matters is whether or not some action is mine, not whether or not I bear the identity relation to some past person-stage who performed the action. Analysis should then focus on the nature of that ownership relation.
    As for your second, Kantian point, I’m not sure I follow. I haven’t been talking in terms of a priority relation, so I don’t yet see why the Kantian slogan would be applicable here.

  32. I defer to your response to my first question, I do now see what it was that you were aiming for, and thus how the possible indepedant truth of the Slogan, is not actually what you were focusing on, and as far as your line of argument goes is irrelevant.
    In point of my Kantian reference, I was eluding to the fact that the supposed entailment of Slogan from Platitude, was a priori unjustifiable.
    As you indicate, the Platitude is the statement of personal and exclusive responsibility, and the Slogan is claimed to be the derivational statement that responsibility presupposes identity.
    I was postulating that by applying the Kantian postulate of “existence precedes essence” that existence (and as part and parcel of this) identity, were naturally occuring before one could have any responsibility (essence).
    As such the Kantian postulate negates any beneficial power of the Slogan, which is demonstrated to be meaningless.
    I do hope this is a slightly better explaination of what I was intending to put forward before. I appreciate that you weren’t arguing with a mind to priority relations or such things, it was more a side note that occured to me as an additional way to distance the Slogan from the Platitude.

Comments are closed.