Abstract
Psychological Sequentialism holds that no causal constraint is necessary for the preservation of what matters in survival; rather, it is sufficient for preservation if two groups of mental states are similar enough and temporally close enough. Suppose that one’s body is instantaneously dematerialized and subsequently, by an amazing coincidence, a collection of molecules is configured to form a qualitatively identical human body. According to Psychological Sequentialism, these events preserve what matters in survival. In this article, I examine some of the main arguments for the view and argue that they fail to establish that no causal constraint is necessary. I also argue that Psychological Sequentialism yields implausible consequences that render it hard to accept the view.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
See, e.g., Parfit (1984: 215–217). ‘What matters in survival’ is a term that denotes one’s special prudential concern for some person. Typically, this concern involves the attitude one has toward one’s own future experience that one anticipates: e.g., we are concerned about our own future surgery in a way that is different from someone else’s. Sometimes this concern is about one’s own existence: e.g., we normally wish that we would still exist after an imminent battle. Hence, it can be said that what matters to one’s current self is preserved in one’s future selves. It has been argued that what matters to one can also be preserved in a distinct individual, as in a case of fission where a single psychologically continuous stream of consciousness divides into two or more branches.
It is important to note that their view does not aim to infer causality from mere succession of qualitatively similar mental states. The view is that causality is not necessary for maintaining personal identity or the preservation of what matters in survival; mere similarities suffice.
I follow Campbell (2005) in using this appellation.
Here I employ the notion of person-stages merely for convenience’ sake. Those who favor a three-dimensionalist ontology should take ‘a person-stage’ as used in this article to mean a person wholly existing at a time.
Temporal closeness may not be a part of the definition of ‘psychological sequentiality’ holding between two person-stages. See, for example, Campbell’s definition in his 2005: 381. However, many of the examples discussed by sequentialists suggest that temporal closeness is relevant to two person-stages’ being psychologically sequential, although spatial continuity is not.
I note that the perfect match required between psychologically sequential individuals is at the psychological level, and not at the neural level. How the psychologies are implemented at the neural level does not seem relevant to maintaining psychological sequentiality, so long as there is exact match between the psychologies thus implemented.
I find that having exactly the same reactive attitudes is less obviously relevant to the question of blameworthiness, because the default assumption is that the appropriateness of guilt, resentment and indignation is to be explained by the blameworthiness of the agent, and not the other way around. I take it that, on Kolak and Martin’s view, having a certain reactive attitude toward some virtuous or vicious traits is somehow indicative of being responsible for the actions related to them.
Nozick previously introduced a rudimentary version of an RME-type example in his 1981: 41, which Kolak and Martin utilize and refine. This is probably why they feature Nozick as the main character of their examples.
Here Kolak and Martin seem to presuppose the dubious principle that if one’s culpability is carried over to a later person, then they are numerically identical. (Notice, however, that the converse is plausible; cf. footnote 16.) Though this supposition may be cast in doubt, I will not press this point further in this article. Instead, I will try to show that in the case under discussion, the culpability of Nozick1 does not survive the breach.
One might point out that this argument conflates personal identity with what matters in survival, and is therefore fallacious. However, even among those psychological theorists who believe that personal identity is not what ultimately matters in survival, it is widely accepted that personal identity happens to preserve what matters in survival. This is because personal identity is constituted by a non-branching form of psychological continuity, and psychological continuity underlain by the right kind of cause is responsible for preserving what matters in survival. According to this view, in general, if a later person is identical to an earlier person, then the later person preserves what matters in the earlier person’s survival, but the converse does not hold. Hence, I do not think that Kolak and Martin conflate personal identity with what matters in survival; instead, their argument can be understood as inferring from the obtaining of personal identity to the preservation of what matters in survival.
This asymmetry in our attitudes has most famously been challenged by Lucretius. He writes, “Look back again—how the endless ages of time come to pass before our birth are nothing to us. This is a looking glass Nature holds up for us in which we see the time to come after we finally die. What is it there that looks so fearsome? What’s so tragic? Isn’t it more peaceful than any sleep” (2007: 101)? I will not address this issue here because doing so would be beyond the scope of this article. For some influential contemporary treatment on this issue, see Nagel (1970), Brueckner and Fischer (1986), Feldman (1991), Kaufman (1996), McMahan (2006).
This does not mean that our asymmetrical attitudes toward past and future nonexistence cannot be justified. One might, for example, try to justify the asymmetry as follows: we would be much less distressed in retrospect about a past surgery than we would be in prospect about a future surgery; this is because the past surgery is already over and done with, but the future surgery is yet to come. Similarly, regarding the harms associated with our past and future nonexistence, one is already over but the other is yet to come—hence the asymmetry in our attitudes toward them.
It should be noted that here Campbell relies on the counterfactual account of causation, which analyzes causation in terms of counterfactual dependence, where counterfactual dependence between events is roughly understood as follows: an event e counterfactually depends on an event c iff had c not occurred, e would not have occurred either. Pace Campbell, I will later claim that on a different version of the “indiscernible swap” argument, it is not clear that the occurrence of the psychological state of your psychological successor at t 2 is not caused by the occurrence of your psychological state at t 1 , on any plausible version of the counterfactual account of causation.
One might argue that Campbell’s argument is based on the functionalist theory of mind since the argument involves the matching of two brains existing in different substrates, and point out that functionalism is subject to some serious objections involving inverted spectra and multiple realizability, for instance. The possibility of an inverted spectrum may threaten the functionalist theory of mind because it suggests that two functionally indistinguishable individuals may have different internal phenomenology. I think sequentialists can stipulate that one’s psychological successor in RME-type cases ought to have experiences with exactly the same representational contents. As to the qualitative contents of their experiences, it may be necessary to stipulate that one’s psychological successor be able to enjoy nearly as much variety of qualitative experiences as one does (so, not too many missing qualia, though it might be okay to miss what it’s like to experience some nameless shade of grey and the like). In regard to the multiple realizability objection, I think sequentialists can point out that the default position in the philosophy of mind is that our mental properties supervene on properties of microphysical entities and arrangements thereof, so that if two individuals A and B are indiscernible at some appropriate microphysical level, then A and B are indiscernible at the mental level. So, sequentialists can stipulate that the exact match between two brains in Campbell’s argument occurs at the appropriate microphysical level.
Additionally, this could mean either that your psychological successor retains your old brain matter at t 2 (if the scientist is simply rearranging the parts of your brain just as one might shuffle a deck of cards), or that your old brain matter is replaced with new matter at t 2 (which seems likely if the scientist is using a quantum fluctuation generator to make your brain match the blueprint donor’s). The second alternative is closer to the RME-type cases discussed in the preceding sections. However, I do not think that the distinction between these two alternatives makes much difference since mere change of matter does not seem relevant with respect to what matters in survival.
Note that this is different from the dubious principle presupposed by Kolak and Martin in their argument involving the same psychological characteristics; cf. footnote 9.
There is another line of reasoning for rejecting the claim that what matters to you at t 1 in terms of survival is preserved in your psychological successor at t 2 . Campbell thinks that the relation that matters in survival holds here because the psychological state of your psychological successor would have been exactly similar at t 2 even if the swap had never occurred. In this sense, the swap is “trivial”—the world would not have been any different had it not occurred. However, it was recently pointed out that this sense of triviality (namely, making no “significant difference to the qualitative nature of the world”) does not entail that the trivial event in question does not affect what matters in survival. See Brueckner and Buford (2013: 99–101).
I have used the term ‘brain state’ to refer to the total global state of a brain in all its parts, including all the distinct individual states that the brain is in at some given moment. I do not mean by it, for instance, an individual state of the brain in some localized part of it, say the state of some individual c-fiber in the brain. When interpreted in the second way, my description of the Frankfurtian counterexample may seem to make the dubious assumption that there is a one-to-one mapping between an action and a brain state. For example, my raising an arm probably does not bear a one-to-one causal relation to any single state or event in some localized part of the brain, such as the firing of an individual neuron—so that if that neuron fires, then my arm is raised. My understanding of the Frankfurtian story does not make this assumption. I do think that the story presupposes some mapping of Jones’s brain states and his actions, but that mapping is not one-to-one. (Without the mapping, Black would not be able to predict Jones’s actions from a reading of his brain states.).
In fact, Lewis (2000) divides preemption cases broadly into two kinds, on the one hand “cutting” cases as discussed above, and on the other hand “trumping” cases that do not involve cutting of the preempted causal chain midway in its path. In this article, it is only the former kind that I refer to by ‘preemption’.
For instance, in the preceding examples, if Suzi’s rock had been blocked at any point in its itinerary, Billy’s rock would have broken the bottle instead. Likewise, even if the causal link between e 1 and e 2 were somehow severed, e 2 would still have occurred due to the causal process running from e b to e 2 . So the effect in each of these examples does not depend counterfactually on the preempting cause—i.e., it is not the case that if the preempting cause had not occurred, then the effect would not have occurred either.
For instance, in dealing with preemption, Lewis employs the notions of stepwise dependence (to handle “early” cutting) and quasi-dependence (to handle “late” cutting) in his 1986: 200–206. He later rejects the notion of quasi-dependence in his 2000, where he develops the concept of influence to handle the problems raised by the newly introduced trumping cases, as well as the “late” cutting cases discussed in his earlier article.
So far I have discussed the “indiscernible swap” argument only in light of the counterfactual account of causation because Campbell’s discussion was couched in terms of this account. However, other theories of causation may account for our intuitive judgment that there is a causal relation between e 1 and e 2 with greater ease. For instance, according to a simplistic regularity theory of causation, e 2 is caused by e 1 , because e 1 instantiates one event-type and e 2 instantiates another, these types being such that instances of the former are invariably followed by instances of the latter.
Here I say ‘many’ rather than ‘all’ because, if there are such things as universals as some have argued, or if God exists, my thoughts about them will be exactly the same as the corresponding thoughts of my twin in content.
One might argue that it is not at all physically likely that a successive series of sequential psychological states is instantiated in the manner I have suggested above. However, the realism of the example is not an issue here. The point of the example is that for most of us the prospect of a series of psychological states merely sequential to one’s own psychological states does not seem to provide the kind of relief typical to the preservation of what matters in survival.
References
Brueckner A, Buford C (2013) Against Psychological Sequentialism. Analysis 73(1):96–101
Brueckner A, Fischer JM (1986) Why is death bad? Philos Stud 50(2):213–221
Campbell S (2005) Is causation necessary for what matters in survival? Philos Stud 126(3):375–396
Dick P (2002) The minority report and other classic stories. Citadel Press, New York
Elliot R (1991) Personal identity and the causal continuity requirement. Philos Q 41(1):55–75
Feldman F (1991) Some puzzles about the evil of death. Philos Rev 100(2):205–227
Frankfurt H (1969) Alternate possibilities and moral responsibility. J Philos 66(23):828–839
Hirsch E (1982) The concept of identity. Oxford University Press, New York
Kaufman F (1996) Death and deprivation: or why lucretius’ symmetry argument fails. Australas J Philos 74(2):305–312
Kolak D, Martin R (1987) Personal identity and causality: becoming unglued. Am Philos Q 24(4):339–347
Lewis D (1986) Philosophical papers, vol II. Oxford University Press, Oxford
Lewis D (2000) Causation as influence. J Philos 97(4):182–197
Lucretius (2007) The nature of things (trans: AE Stallings, an introduction by R Jenkyns). Penguin Books, New York
McMahan J (2006) The Lucretian argument. In: McDaniel K, Raibley JR, Feldman R, Zimmerman MJ (eds) The good, the right, life and death: essays in honor of Fred Feldman. Ashgate Publishing, Burlington, pp 213–226
Nagel T (1970) Death. Noûs 4(1):73–80
Nozick R (1981) Philosophical explanations. Harvard University Press, Cambridge
Parfit D (1984) Reasons and persons. Clarendon Press, Oxford
Siderits M (1988) Ehring on Parfit’s relation R. Analysis 48(1):29–32
Van Inwagen P (1983) An essay on free will. Oxford University Press, Oxford
Acknowledgments
This work was supported by the 2010 Research Fund of UNIST. I also would like to thank Matt Hanser, Boram Lee, and anonymous referees for Axiomathes for their detailed and insightful comments on earlier drafts of this article.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Yi, H. Against Psychological Sequentialism. Axiomathes 24, 247–262 (2014). https://doi.org/10.1007/s10516-013-9221-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10516-013-9221-8