What we have so far is a narrative approach to personality. So, what is it good for with respect to the problems of personal identity? First of all, it offers a new line of reasoning to explain the significance of episodic memory for the persistence of persons. The point is that the raw material for those stories we tell about ourselves is provided by the faculty of episodic memory. More pre
137 It is a distinguishing feature of narrative accounts to put emphasis on the social constitution of the self; see e.g. Taylor (1989:I.2.2.) and Quante (2002:170).
138 We would just like to mention a few ways how a person may defend her self-concept once it is challenged in some particular aspect. Let's assume someone is pointing out an episode of my life that is at odds with how I claim to be like. First of all, I may deny the truth of his account of what I did. If I have to admit its truth, I may give some reason why I did not act in line with my general disposition in that particular situation. Or, I simply may try to outvote him by citing a number of counter-examples that support my claim.
cisely, the availability of episodic memories is a condition of the possibility of the kind of storytelling by which persons generate their self-concepts. Hence, if episodic memory is wiped out completely, as in severe cases of retrograde amnesia, then a sufficient condition for a change of personal identity to occur is fulfilled. This, of course, is perfectly in line with our earlier result that some sort of episodic memory continuity (as specified in Mem-Crit1) between person-stages is indeed a necessary condition for them to belong to the same person.
The episodic memories of persons provide the raw material for the communicative process whose narrative outcome is an intersubjectively approved account of their personalities - including a critical appraisal of their self-concepts. We consider it indispensable for an adequate understanding of persons' characters to make allowance for their narratively constituted self-concept. Only in this way one can make sure not to miss anything that is pertinent for deciding questions regarding the persistence of a particular person. By just following a highly schematised check-list in describing personalities one cannot do justice to individual variations of the contents of personal self-concepts. These variations are due to the fact that persons refer to vastly differing properties in composing their self-concepts. But even this does not account for all the variety: something quite pertinent is left out if one does not pay attention to a person's particular narrative style in expressing her self-concept. Some persons seem to follow the aesthetic ideal of a classic coming-of-age novel in that they see every event worth mentioning in the light of how it contributed to the eventual achievement of one overarching telos. Quite to the contrary, others may cultivate a thoroughly "episodic" stance towards life, avoiding the assumption of any overarching telos, seeing their lives as a conglomeration of independent phases which would have to be characterised by nothing so much as the contingent circumstances and tasks they were confronted with at the respective times.
It will be illuminating at this point to take a closer look at possible cases of change in a person's narrative style: Can a person who is there after a transition to an episodic outlook on herself still be considered as the same person who formerly used to think of her life as proceeding towards some telos? Well, why not? For instance, after some deep disappointment a person loses all her faith in a deeper meaning of life. As a consequence, she may begin to live her life from one day to the next, from year to year, thereby just changing her behaviour along with the circumstances, opportunities and roles presenting themselves to her. But it still would be her life. After all, she has got reasons for why her self-concept changed so radically. Others may question the consequences she has drawn from her experience of disappointment, but they will need to admit that she has got a story to tell that is worthy of consideration. As long as she cares a bit - maybe not for herself, but for others -, as long as she is ready to take responsibility for what she does and for what she did prior to her "identity crisis", there is no reason to doubt that she is still the same person, even though profoundly changed in her personality in general and in her self-concept in particular.
We now consider a more radical change in "narrative style", from telos-style narration to one which maybe could be labelled "radical deconstruc-tivist": When questioned about his self-concept, the radical deconstructivist will say "I don't have a self-concept, there is nothing but ever-fleeting images of myself." He considers "personal persistence" to be an illusion, too: Every morning he will wake up thinking "Let's see who there is today". Consequently, he will refuse taking responsibility for "his" actions. He may say, "I know someone did something yesterday, but that has nothing to do with...", and after a pause he may continue by saying "who cares!" Well, the others will certainly care! If this "deconstructivist stance" turns out to be nothing but a superficial "attitude", as some sort of transitory digestive trouble possibly caused by reading philosophical or spiritual literature of a certain kind, it has to count as nothing but a very weird and rather phoney sort of image in the end. The person in question will therefore be held responsible for everything he does (or did, earlier) - whether he likes it or not. If, on the other hand, his opinions are considered genuine (this presupposes that his behaviour is properly erratic), he most likely will be considered to suffer from a severe depersonalisation disorder and will be treated accordingly. This means that rather than describing the transition of the "coming-of-age person" into the "radical deconstructivist" as a change of personal identity, one would describe it as a person going so insane as to lose personhood. But why? The reason seems to be not so much that the deconstructivist story is not convincing (though in fact it is not, as it does not even try to account for anything, least of all the transition from the "telos-style" to the "deconstructivist" one). Rather, it does not count as a story in the first place (and thus the "deconstructivist style" in truth is not just one "narrative style" among others).
Quite obviously not every sequence of utterances can count as a story, but what the criteria for "story-ness" are is a harder question. There are some minimal requirements with respect to coherence that any narrative with a claim to truth (and so a fortiori any narrative defence or modification of a self-concept) needs to fulfil. These requirements are such that violating them in particular instances means that - with respect to the goal of intersubjective agreement - there is something not quite correct yet about the story, which has to be amended to get it "right". Violating them more often, however, does rather result in the breakdown of "story-ness" itself.
The first requirement we want to draw attention to is that inconsistencies with respect to the time-order in which a person places the events of her story cannot be tolerated. This does of course not mean that the order in which events are to be reported has to be the order in which they actually occurred. This would be a just demand on chronicles, but not on all sorts of stories with a claim to truth. Quite to the contrary, depending on context and purpose, such stories may present us with a whole lot of very helpful flashbacks or anticipations of things yet to come. What can be justly demanded, however, is that the order of events in a story with a claim to truth can be consistently deduced from it. In other words: If, according to the story, some event is both to happen and not to happen at a certain relative place in the time-line, then this is definitely something calling for amendment. It should be clear that if you have an increasing amount of inconsistencies in the time-line of a story, it will, at some point, eventually break down as a whole. Just as a bunch of "timeless" statements (like e.g. Euclid's axioms) doesn't make a story, a bunch of statements about events that cannot be ordered even partially is no story, either (even if terms like "post-modern", "deconstructivist" or whatever are invoked).139
The second requirement we would mention has to do with the observation that a mere concatenation of event-stating sentences does not make up a story either - even if the time-line is correct and the sentences are all true.140 The coherence of a story thus must consist in more than just a consistent time-line: There has to be some sort of "connectedness" of its statements beyond that. To be sure, a story must feature certain recurrent objects and "themes", but in general it would be quite futile to try getting a grip on the pertinent kind of connectedness in question by looking for purely syntactical or even semantical criteria. To get closer to the heart of the matter, it is crucial to realise in a first step that a story's claim to truth (if it has such claim) is never just "to report something that happened", nor "to tell it all" or "to tell it most exactly". Rather, for every story, there is an explicit or implicit background of goals against which objects, events and levels of descriptive detail and completeness are judged pertinent or irrelevant.141 So, ideally, what is told in a story is relevant with respect to "what it is about". This very general observation can be elaborated a bit further by taking into account that, even with respect to the specific goals which determine what the story is about (and what not), a story's claim to truth (again, if it makes such claim) is never just to report what happened, but also in a way to make sense of it. This in turn can be understood in a variety of ways, the most prominent being the intentional explanation of actions and the causal explanation of other events ("Why did Stalin offer the reunification of Germany in 1953?",
139 Here and in the following we would ask the reader to keep in mind that we are talking about stories that claim to be true. We do not want to say that e.g. decon-structivist movies are worthless as pieces of art.
140 See Danto (1985:117). His counter-example: "Naram-Sin built the Sun Temple at Sippar; then Philipp III exiled the Moriscos; then Urguiza defeated the forces of Buenos Aires at Cepada; then Arthur Danto awoke on the stroke of seven, 20 October 1961."
141 Again Danto (1985:131) has a nice example to illustrate the point: "Suppose I'd wish to know what happened at a court trial. [...] I should be dismayed if [...] he were to tell me how many flies there were in the courtroom, and show me a complicated map of the precise orbits in which they flew, a vast tangle of epicycles."
"Why did the dinosaurs become extinct?").142 For example, a self-conceptualising story should provide us with something which a simple listing of p-traits ("social, serious, open-minded ...") cannot provide: an answer to the question "Who are you?" in the sense entailing questions like: "How did you become who you are?", "Why did you change in this way?" Bearing this in mind, the relevance of parts of a story can be assessed through questions like "Does this set the stage for, or contribute to the understanding of something else which the story is essentially about?", "Is this something we can take for granted or rather something that should be explained?", etc. Again, "ideally", every part of a story would be relevant in the sense that it serves a function with respect to the descriptive and explanative goals associated with the story. The function of the respective part would be its significance, which is why we are speaking of the "requirement of significance".143 As with the requirement of consistency of the time-line, single violations of the requirement of significance, while sticking out like sore thumbs calling for band-aid ("Why did you tell us about that watch of your father, the one which you accidentally broke?"), will not threaten "story-ness" - but an increasing amount eventually will do.144
The third and last requirement to be mentioned here is, actually, a generalised version of the first one: Keeping a consistent time-line means an instance of following the more general requirement of keeping truth-claiming stories free from contradictions. Therefore, the addressees of self-conceptualising stories actually have the right to object to any inconsistency they notice. And again we can state that, whereas truth-claiming stories are
142 To be sure, a satisfying story does not have to (and usually in fact does not) give all these answers by way of mentioning explicitly all the pertinent maxims or causal laws. However, for the story to be satisfying, they must be, in principle, distillable from it - at least to the extent of providing an explanation schema.
143 To forestall any misunderstandings, it is extremely important to point out that the requirement of significance does not mean that, in their self-conceptualising enterprise, persons should try to give "meaning" to every miniscule episode or even to a major part of the episodes of their lifes (this could be called a "theodicy-style" narrative). Rather, the requirement just says that significance is to be given to every episode which is actually making the cut, showing up in the story constituting a person's self-concept. Note also that this does not demand or even favour telos-style narration (i.e. that the significance of any part of a story lies in its contribution to the same final goal).
144 Though we maintain that the goal of "making sense" and the requirement of significance also hold with respect to stories that do not have a claim to truth (especially novels), we of course acknowledge that this may interact with other goals here. For instance, it may be extremely boring and shallow if the significance of every scene is too obvious. It even may be especially praiseworthy if the significance of many events is left unclear until the last moment, or proves to be quite different than it seemed at first ("twist"), or if the story as a whole is open to more than one "interpretation". Again, not even a "deconstructivist story" is ruled out as a possibly subtle and worthy work of art - it is just ruled out as a story.
"robust" with respect to singular, isolatable contradictions, they can take only so much of these before eventually breaking down wholesale.145
Of course, the internal consistency of a story is not enough to make it a true story. To this end, a story also has to be "consistent with the facts". Regarding self-conceptualising stories, there can be more or less obvious conflict between what a person says and what she does. We already saw how to deal with "external" inconsistencies of this kind. If they can't be "explained away" (i.e. if it can't be shown that they just seemed to contradict the facts), they can be explained with reference to an image that a person may consciously create in order to fool others about her true self-concept, or with reference to a discrepancy between how a person thinks about herself and how she actually is. As a special variant of the latter case those numberless situations are worth mentioning in which persons fail to act in accordance with their "ideal self-image". The important thing here is, though, that no amount of external inconsistencies will ever lead to a breakdown of "story-ness". As long as it is internally consistent, a story may be as blatantly and impudently wrong as you dare to imagine - it will still be a story. So, contrary to what we found with respect to internal inconsistencies, the coherence of self-constituting stories, and hence the stability of self-concepts, is never threatened by external inconsistencies. Thus, the maniac who thinks he can take on the world all on his own has a self-concept, however "off-target", while the "radical deconstructivist" has none at all.
Having dealt with structural criteria which stories (with a claim to truth) must fulfil as such (and good stories in particular), we revert attention back to our example of a person going from telos-style to episodic style self-conceptualisation. What are the general lessons such examples may teach us about the distinction between personality changes and changes of personal identity? Well, we came to the conclusion that persons may persist through very profound changes of their personalities including their self-conceptualisations. This seems only to confirm our earlier result that it is not possible to establish adequate character-based criteria for personal persistence: If all we have got with respect to two person stages at different times are two most comprehensive descriptions of personalities, including the respective self-concepts in every detail, then it will not be possible to tell whether the two person-stages they describe belong to the same person. However, by putting things this way something quite essential is missed: As we have seen by now, a person's self-conceptualisation is not exhausted by opinions stating whether she has or has not certain p-traits. Rather, it is also constituted by narratively structured opinions146
145 In this we of course accept that standard logics are wrong in assuming that from a contradiction every proposition follows - in that case a single contradiction would suffice to make a story worthless.
146 This may be a good place to state that it does not matter so much whether the "stories" a person has to tell about herself are explicit in the sense that she actually regarding how the person acquired these traits and, consequently, by likewise structured opinions about how her self-concept itself has developed and about how it came to change, maybe profoundly, throughout her life. That is, a later stage of a person is connected to the earlier ones by way of narrative integration of the personalities of the earlier stages. Though this is only possible by recourse to episodic memory, it is nevertheless a much stronger link than mere memory continuity provides: In narratives, memories are arranged within a wider time-frame, and they are interlinked with respect to their content by recurrent objects and themes. By including episodic memories in a story-line or by leaving them out as "irrelevant", storytellers valuate their contents, and they give a certain "meaning" (in the sense of "significance") to them by putting them at certain, specific places within the story.
In this way a story can do several things a mere episodic memory, or even a string of such memories, cannot: Firstly, it can in a way incorporate stages which are not directly accessible by episodic memory anymore. For example, I may have no recollection at all of what I did on the 13th of February, 1984, but if I know where to place that date within the larger frame of "my story", I will nevertheless have a pretty good idea about what kind of guy I was back then: what my temper, goals, opinions and talents were - even what I used to think of myself. The second important thing a story (as against a string of memories) can do is to tell what happened and at the same time give an account of it, make sense of it, explain it. This is so strong a tool that it even can deal reflectively with, and account for changes occurring over time in the story itself. What we have in mind here goes way beyond those trivial changes due to the mere addition of ever new parts with the advance of time: For example, a story could change in that it picks up events which were left out as irrelevant in earlier versions, or leaves out or qualifies elements which were formerly considered of special import. Maybe some seemingly inconsequential action of mine proved to be one of earth-shattering consequence much later; or what I thought was the root of all my problems was in fact important only in that it diverted my attention from my real, however repressed conflicts. In both cases, the later story would in a way explain (or try to explain) what was wrong with the earlier version, and thus would incorporate the "story of the old story" into the new story-line ("I used to tell myself that my moodiness was due to too much stress at work until I eventually realised, when my uncle died, that it had much more to do with ..."). As we have seen in the example about the person turning from telos-style to episodic narration, a story could even told them to someone, or e.g. wrote them down in a diary, or rather implicit in the sense that they could be elicited by prodding her in the right way, asking the right questions etc. What counts is that she actually has a set of opinions about herself, about her life and development as a person, which together exhibit a narrative structure.
change its overall narrative style - and such a change, too, could be accounted for in the new story.147
We are now in a position to state more clearly what is required for a decision as to whether two person stages (at different times) that exhibit more or less different personalities do nevertheless belong to the same person at different times of her life: Two person stages x and y belong as temporal parts to some (the same) person p if and only if they are continuous and the narratively structured self-concept of the later stage y does plausibly integrate stage x. Before we pause to talk about a host of issues regarding this criterion (like the meaning of "plausibly integrate"), we first propose a generalised version148:
(PI-Crit) The members of a set of person-stages M all belong as temporal parts to some (the same) person p if and only if they are all continuous to one another and for every stage whose narratively structured self-concept does not plausibly integrate all earlier stages there is a later stage which does so.
The criterion has now the form of a biconditional ("if and only if"). That is, if we manage to hold on to it, we finally got the necessary and sufficient condition for personal identity we have been after for so long. So let's begin to discuss it.
Firstly, it may be questioned why the criterion is focussing on the self-concept of the individual. After all, didn't we make quite a fuss about the intersubjectivity of the stories about personality, speaking about "co-authorship" and such, denying the individual person epistemic authority? But our focussing on the individual's self-concept has nothing to do with epistemic authority. Rather, if the question of personal identity is put, one has to focus on certain capacities of the person (stages) under consideration, namely their capacities of memory and narratively structured self-conceptualisation. Thus, the criterion does not involve in any way that a person stage is, by definition, itself the ultimate authoritative judge about the question of what earlier stages belong to it or not, even though the objective (i.e. intersubjec-tively validated) answer to this question directly depends on its subjective capability of relating itself to those other stages.
147 According to Danto (1981), works of art are like symbols in that they have meaning, but - unlike other symbolic representations - manage to say something about their subject matter through their mode of representation itself. In this vein, the narrative style of a story itself "tells" something about how the storyteller conceives of the subject matter of his story. In other words: the narrative style of a story constituting a self-concept is itself expressing something about how the storyteller sees himself. The content expressed in this way is therefore itself part of his self-concept (provided the storyteller could, in principle, make corresponding explicit statements - otherwise we would have to treat his narrative style as a symptom for a p-trait which he is not himself aware of).
148 This is necessary for two reasons: Firstly, the case of two stages does not imply anything regarding the case of n stages. Secondly, only the generalised version will allow us to incorporate an important liberalisation.
Secondly, one may ask why the criterion does not include the liberalizing provision that later stages only have to integrate such earlier stages with regard to which they can directly quasi-memorise any episodes of experience from the first-person perspective. The answer to this was implicitly given above: Though episodic memory is a prerequisite for narrative self-conceptualisation, the latter transcends the former. I may not remember anything I experienced on a certain day, but as I can locate that day within the wider frame of my life, I nevertheless should have a concept of "who I was" at that time - what my temper, goals, beliefs and talents were like, what I thought of myself, and even which opportunities and hardships I was confronted with that I think were bound to have an impact on my further "story". This is a relation I should by some measure have to all earlier stages of mine, even to those with regard to which I am not able to directly recall any specific experiential episodes.
Thirdly, one may wonder whether, on the other hand, the criterion is too liberal. Why don't we require every stage to integrate all earlier ones? Why did we add the provision that, in case of failure, things are "saved" by a later stage doing the integrating? Now, let's assume we have consecutive person stages a, b, c belonging to the same human body B, and things be such that the person at b suffers a "crisis" in that she is not able to integrate the earlier stage a, but then overcomes the crisis at c so that she can integrate again both a and b. Now, had we formulated our criterion more strictly, we would face the following situation: b and c would belong to one person, as would a and c, but not a and b. Therefore, we would have not one person, but two which just share stage c (and all following ones) - a case of "psychological fusion". This would certainly be an artefact of the criterion being too strict, and it is proof of the adequacy of PI-Crit as given above that we can describe this example as the case of a single person recovering from a crisis.
For further illumination, let's compare the issue to the case where we have to deal with a phase of retrograde amnesia at b which is eventually overcome at c. Here, we may count the stage b as belonging to the same person as a and c without having to tamper with the definition of memory continuity, because the modal formulation "can (quasi-)memorise" provides us with the leeway we need: We already mentioned that while someone is asleep at night, he may be considered to be "able to memorise" episodes of the day before in the sense that ifwe wake him up and ask him, he will tell us something about it. In a very similar sense we may (and actually in practice do) treat a transient phase of retrograde amnesia: If the specific contingent obstacles that were eventually removed for c would have been already removed for b, then this stage would have remembered something. The point is that this strategy is only licit if we restrict ourselves to true conditionals which are not analytic trivialities (in the vein of "if we would have made him remember, he would have remembered") and, furthermore, if we have an "if"-clause which is not just logically, but also empirically possible (by this provision we exclude con ditionals such as "if we had a perpetuum mobile, then we could save big on resources"). In the case of the overcome amnesia, the mere fact that it was overcome already provides the required backing for the assumption that it was empirically possible to overcome the amnesia after all.149 It is important to note that this elegant way out is not transferable to the case where someone's narratively structured self-concept is not able to integrate his earlier self, for this would amount to translate "does plausibly integrate" into "would do so if it were different".
Now we address ourselves to the most critical question: What is "plausibly integrate" supposed to mean. First of all, the expression is to convey that it is not required for pieces of narrative self-conceptualisation to represent the "true story" in the sense that they would need to explain all the changes in personality in the most intersubjectively convincing way. Positively, and in the most general way, "plausibly integrate" means:
a) there is a story (i.e. a structure which satisfies at least the basic criteria for "story-ness") relating the earlier stages and the pertinent changes to the actually "present" stage and b) this story is sufficiently grounded in episodic memory to be counted as genuine.
Apart from this, the threshold for "plausible integration" will, however, not be the same tout court, but differ widely with context. The strongest demands will be put on the coherence of a story when there is reason to doubt that all pertinent person stages are represented by the same body. Even though conceptually we consider it to be possible that a person may wake up one morning in a different body, there certainly is no empirically accredited case of one person showing up in different bodies at different times. However, sometimes it so happens that a person claims to be someone who had disappeared for a long time. Especially in the past (before the advent of DNA-testing), it quite often was impossible to settle allegations of that kind by checking physical continuity between the person who had disappeared and the person claiming to be him. If the present person lays claim on the property of the person who had disappeared and if others who knew that person are affected by these claims, then we have a type of conflict on which various novels and movies are based. In situations of that kind, the claimant
149 That is why cases of ongoing amnesia have to be treated with some agnostic care. For example, it is reasonable that a court does, at the one hand, not subject a person suffering from total retrograde amnesia to punishment for a crime committed during her "former life". For maybe the amnesia is total, the episodic memory "wiped out" once and for all, the band of psychological continuity severed, never to be mended again. Such a person would in fact be "someone else". However, it is also reasonable that the court will not let the amnesiac simply go free either: he will still be submitted to punishment, should the amnesia be overcome, and as long as there is a chance for this, the accused also has to be treated like a person with a disorder rather than a "new" person.
needs to tell a particularly convincing story about how and why he disappeared and about where he has been and what he did in the meantime. On the other hand, those who take an interest in unmasking an impostor will give close scrutiny to the requirement of sufficient grounding in episodic memory - typically by asking tricky questions the correct answers to which an impostor would be highly unlikely to know ("Your great-aunt told our investigators that you had a nickname for that sleigh you were so fond of as a child - how did you call it?").150
If, on the other hand, physical continuity between two person stages is unquestionable, there is strong prima facie reason to suppose that both stages belong to the same person. Therefore, the demands on the plausibility of stories by which person stages are integrated are much lower here. Actually, the onus of proof is on those who want to argue that two person stages that are represented by the same body are not stages of the same person. This becomes obvious when one considers court cases where a person is accused of some felony. Here we have got a profound public interest that the person who is sentenced is actually the person who committed the crime. If a defendant's physical continuity with the criminal offender is established, then we may get one of the extremely rare instances where someone takes an interest in arguing that he emanated from a change of personal identity, so that "he is no longer the person" who committed the felony.151 Sometimes defendants try to argue in this vein if the crime has been committed long ago and if - in the meantime before their apprehension - their personality has changed very favourably, so that they wouldn't be likely to commit such a felony now.
150 Just for the record: Such scenarios of investigation would in principle also apply if it was (or became) empirically possible for persons to survive their bodies, be it by continuous transmogrification into different bodies or by more discontinuous means of replacement (including "reincarnation"). In the end, it all comes down to the question under what circumstances we would accept someone as someone. These circumstances are, we argue, related to someone's capability of memory-grounded storytelling. Assuming this is true, then if certain phenomena became rather common (instead of just being explored in novels and movies), society would eventually begin to treat them as cases of reincarnation or body switching. To make this more palpable, imagine someone would show up claiming he were in fact the reincarnation of your long dead brother. Of course, you would react most sceptically, to say the least. But now imagine he would not only show the characteristic personality traits of your brother, but would also be able to recount and relate to all the things you experienced together, and could do the same, consistently, for all other members of your family. Wouldn't you and your family eventually begin to accept him? At the very least - we dare to predict - you would do so if that sort of phenomenon, rather than being unheard of, would actually be a more common thing for which society had already developed certain traditions of coping.
151 The inverted commas express reservation against this wording, for if "he" is to denote the person that appears in court and if actually an identity change occurred in the meantime, then he never was the person who committed the felony.
However, as we saw in our discussions about "character stability" and "character continuity", character change alone - however drastic - is by no means a sufficient condition for a change of personal identity. Even a most radical and abrupt change in character may leave a person's capability of narratively integrating that change untouched.152 Consequently, on the basis of our narrative account of personal identity, the defendant's best chance would be that his lawyers provide convincing evidence that he suffers from complete retrograde amnesia. In this case he will not be able to accomplish narrative integration of the person stages before onset of amnesia, thus not satisfying PI-Crit with respect to those stages. And while he will153 not really be acknowledged as a "new person" because of this, but rather as a person with a certain form of dissociative disorder (dissociative amnesia), he would (or at least should) not be submitted to punishment until he does recover (if he ever does).154 Now if the defendant does not suffer from dissociative amnesia, but actually does remember the deed, he will have a comparably harder time proving that he is not the one who is to be held responsible for it. However, even this is not impossible in principle: What the defendant would probably claim in such a case is that he remembers that "someone else" with his body committed the felony, and his lawyers would try to rack up sufficient evidence that he is suffering from DID (dissociative identity disorder, previously also called multiple personality disorder). In this case, the set of person stages associated with one and the same body can in principle be regarded as divided into several mutually exclusive subsets which are characterised by the fact that every stage x is only able to integrate those prior stages which belong to the set x itself belongs to. The subsets correspond to the different (usually very different) "personalities" that take turns in "surfacing" for certain periods of time. It is even more clear here than in the case of dissociative amnesia that this situation will not in fact result in an official acknowledge
152 This also has implications for cases where, in countries with capital punishment, people sometimes change their character in a favourable way while spending years on "death row". It is not a good idea to argue, for someone of that kind, that he shouldn't be punished because he "is no longer the one who committed the crime". For when people change from "criminal" to "good citizen", they usually got that character change firmly integrated into their self-concept, and even make a special point of telling the story of that change. Thus, rather than taking recourse to an alleged "identity change", arguments for the reprieve of these people would have to draw on the inhumanity of delivering capital punishment after years of incarceration. (Here is a last comment on the inadequacy of trying to argue for identity change through character change: If drastic character change would indeed all by itself count as a change in personal identity, then this would have to be acknowledged also if such a change in character had taken an unfavourable direction!).
153 By most courts, in all countries, as far as we know, and according to extant law.
154 It is known that retrograde amnesia can be caused by traumatizing events, and the felony itselfcould be just such an event. However, in such cases there is also a good chance for the memory eventually to return.
ment of "several persons" - DID is a disorder after all. Instead it will (or should) result in suspension of punishment.155
The two examples given above (potentially different bodies with a claim to personal identity vs. one and the same body with a claim to a change in personal identity) were supposed to make clear that, and which way, the criteria for "plausible integration" of person stages in identity-preserving stories can systematically vary with the general type of situation and claim we are confronted with. Both examples were set in the context of a "court trial", though, because we wanted to show that our narrative account of personal identity is not in danger to relegate questions of personal identity into the realm of subjective opinion (like matters of taste), but, on the contrary, can be applied to and makes a difference in situations where we want to settle such questions according to objective (i.e. intersubjective) standards of truth. On the other hand, questions of personal identity obviously do not only occur and play a role in situations where public interest is at stake. For example, a marriage may come to an end because one of the partners insists (even in these words) that the other is "not the person anymore she fell in love with" - and maybe the partner even agrees to that. Or, to give another example, some listeners may find it fully acceptable when a person accentuates a profound change of his personality by changing his name from Saulus to Paulus, and explains the change with reference to the "epiphany" of another person that passed away some time earlier. These examples show that the narrative account is also at work if there is no public interest at stake - just that what is considered "plausible integration" is then defined to a large part by more "particularistic" standards agreed by storytellers and recipients in particular narrative situations.
This should suffice for now to establish the general framework of the narrative account of personality that in our view is most apt to resolve questions concerning personal identity. We don't need to summarise it at this stage, as this will be done in the next section. Also, its consequences for the normative issues at hand will receive more attention there.
155 In fact, things can be really complicated with DID. For instance, sometimes the different "persons" associated with the body of a patient with DID do not know of each other and, consequently, can't recall events that are experienced by those "other persons" - in which case we have to deal with forms of amnesia too. Sometimes they only partially or asymmetrically recall each other. If they do, they may have opinions about and emotions towards each other, and sometimes they seem to be able to "communicate". Furthermore, with the progress of time, some "persons" may dissolve and others "arise". If the courts would take it on themselves to treat DID-phenomena as anything other than a disorder, they would have a hard time to disentangle things. And even if they could, there would result a host of practically unsolvable problems. For example, if the person who committed the felony is still among the set of alternating persons, how could one punish this person without at the same time unduly punishing the others?
Was this article helpful?