Mulder: Ok, here are the details about John Perry's reconstruction of Grice. It is true that Grice was worried about the Alzheimer/ daydream types of cases. Another example he mentions would be repressed memories that no process of psychoanalysis or any other process could recover. Perhaps a diagram will help you.

Diagram. Aulisio/Alzheimer's/daydream cases

//////// y

/ R

x — R .....—R——z

For his theory, Grice's solution is to let the relevant R relation be symmetrical, which means simply if xRy then yRx. This feature gives the result that even though it may be that no later stages z whatsoever will or could contain memories of experiences in y (where y is the experiences that are never remembered etc), still y gets linked into the later chain if it is R-related to earlier stages x that are R-related to z. Then, since xRy we have (by symmetry) yRx, and since xRz, and since R is transitive by definition, yRz. Walla!! Now Perry's formulation Grice's criterion for copersonality of stages is such that any two stages are co-personal just in case both are in some "Grice set" where a "Grice set" is a set of stages which can be put into a sequence (not excluding repetitions and not necessarily ordered temporally) so that each member of the sequence is R related to the next member and to which no more stages can be added in the sense that there are no stages not in the set which are R-related to some members of the set (see pp. 135-137). This means that any two stages will be co-personal just in case Rxy, that is, R and copersonality are necessarily coextensive. Indeed "branching" of R is not possible on this account! "Fission" cases are the most plausible examples of branching but for Perry-Grice's symmetrical R we can always get a sequence of the necessary sort to get copersonality of all relevant stages. By the way, Parfit explicitly denied the symmetry of R, so that is the formal difference between these two approaches.--]

interrupt[ Scully: Yes, Mulder, I see your point. And indeed, it is evident from his initial discussion in Reasons and Persons that Parfit treats R as non-symmetrical because of his intuitions about how to deal with the R-branches. He sees that R by definition is transitive, and so if it were also symmetrical that would mean that in cases like the Kirk case, the two stages Y and Z would be R-related. That is why he denies symmetry of R. He denied it simply to make sure that branching of R would be possible!The main point to be drawn here, it seems to me, is that within the family of concepts that these philosophers are using to discuss personal identity, there is going to be some flexibility in how the concepts are defined. Nothing necessitates one view or the other about whether one should regard R as symmetrical, and as you point out, Grice had some pretty good reasons for treating R as symmetrical. --Mulder, don't you think we have enough for one night? --

Mulder: Ok. Two qualifications need to be made to what I have said about Perry's Grice. (a) Perry's version of Grice differs from what Grice actually wrote in an important respect, since Grice himself assumed no temporal overlap of stages of a single person and he assumed that the relevant sequence would preserve the temporal ordering of the stages. (So far as I can tell, Grice explicitly considered neither fission types of cases nor time travel, which is understandable since he was writing well before the discoveries in Star Trek and Back to the Future.) So Grice's unique sequence corresponding to an Alzheimer's example would just be the stages as ordered by their occurrence in external time.

(b) Secondly, while Perry shows that his definition of co-personality in terms of "Grice sets" is equivalent to the definition that Grice himself gave, Perry's equivalence proof works only on the assumption that there is no temporal overlap of the relevant stages (see footnote 7, 155). I suspect Perry focussed exclusively on his Senile General case (that is, an Alzheimer case) without explicitly considering other types of possible "branches" like that in Parfit's fission thought experiment. Actually, on Grice's own account (before being reformulated by Perry) branching is possible, since when stages co-occur there is no sequence that corresponds to some unique temporal ordering. However (i) Perry's reconstruction of Grice is true to the spirit of Grice's work, since the symmetrical R is at the heart of Grice's account, and (ii) Perry's reconstruction guarantees survival in fission.

Scully, I should also point out that the idea “no temporal overlap” of stages cannot be used (as Judith Thomson uses it) to explain precisely what the “no branching” proviso in Parfit's Psychological criterion means. The "no temoral overlap" idea cannot be used to do this for two reasons: (a) there can be branchin, of the relevant type, on Parfit's theory even without temporal overlap of stages (as in the Aulisio/Alzheimer's cases mentioned by the alien--the cases that worried Grice), and (b) there can be temporal overlap of stages without branching of R (as in the time travel example mentioned by the alien). Suppose, for instance, we have Zkirk produced instantaneously on Planet Z but Ykirk not produced on Planet Y until two hundred years later, well after the death of Zkirk.

Diagram. R-branching of Kirk with no temporal overlap

/////z_____.

//

x — R

\\.................R.....___ y_____ .

I am saying this is a genuine case of branching R (if R is assumed capable of branching, as Parfit assumes, even though none of the stages of Zkirk and Ykirk overlap in time and it would at least be odd if Parfit concluded that there could be co-personality in a fission case with no temporal overlap simply because there is no temporal overlap. As Unger argues, fission with or without temporal overlap should be treated in the same way (Identity, Consciousness, and Value, p. 88). In any case, the main thing here is that Perry-Grice's account does not need a “no branching” clause since it can simply treat R as the copersonality relation. Moreover, [edited out by space-saver FBI-SS-6582934*K37]--] [Mulder: Why did you zap me? --Mulder]

Scully: C'mon, Mulder. If there is a point to our investigation --which I doubt -- it cannot possibly involve abstruse details about one person time travelling repeatedly combined with reincarnation. I am concerned about you, Mulder. By the way, given what you just said, it appears to me that the Psychological account requires Perry's Gricean approach--or at least an account that grants survival in fission.

Mulder: Why?

Scully: Couldn't you have a sort of “Branchline” case, a variation on the “normal” Fission ones; Parfit himself mentions this type of case. Imagine a situation where Glenn, say, is teleported to Planet X but his body on the Shuttle is not disassembled and destroyed, as would be “normal.“ Maybe this diagram will help.

Diagram. Branchline (duplication)

//// X-Glenn (planet-X)

///

Glenn_______/_______Glenn

Now to most people it would seem obvious that Glenn survives this event and would be on the shuttle the whole time; moreover X-Glenn on planet X would appear to be a duplicate of Glenn. (I assume this in my labels in the diagram, just using the term `Glenn' to refer to the guy on the shuttle both before and after the duplication occurs; of course, I would want to leave open the possibility that X-Glenn = Glenn as on Perry's-Gricean account). As noted, everyone seems to agree that Glenn would survive in this case as the the guy still on the Shuttle because, after all, he'd have all of the normal physical and psychological continuities throughout the entire story--“and so what if a duplicate gets made?” --But this means there is a tough problem for a Psychological account which, like Parfit's, thinks that one does not survive in the original Fission example. The problem is to explain how this Branchline case differs from Fission. Any Psychological theory that denies survival in fission has to deny survival of Glenn in this “branchline” version of the story as well unless they can identify a relevant psychological difference in the two stories!!

Mulder: Good point, Scully! --Sydney Shoemaker and some other Psychological theorists recognized this problem-- as Carol Rovane put it, for examle, one needs a purely psychological explanation here of the difference between fission and branchline; but, so far as I know, nobody has given one. Nobody has explained why Glenn on the shuttle after the duplication event would be PSYCHOLOGICALLY closer to the pre-branch Glenn than is X-Glenn on Planet X--

Scully: Indeed, it might turn out that X-Glenn is even psychologically closer to Glenn-before, than is Glenn-after, since X-Glenn will bbe DOING the things on planet X that the earlier Glenn intended to do when he decided to teleport there--

Mulder: Yes, Mark Johnston points that out p. 163 of the book Dancy edited--]

Scully: ok, cool. --So you want to call it a night?

--interrupt[Mulder: Scully, ok, but what about the point Zoe Alexander made in her thesis about de re thoughts directed to one's body-- these might at least account for our intuition that Glenn on the shuttle is psychologically closer to Glenn-before than is X-Glenn? --

Scully: What? Why? --

Mulder: She tried to argue that since some of Glenn's de re body directed thoughts after the branch will be satisfiable or true whereas corresponding thoughts by X-Glenn will not be satisfiable or true. E.g., suppose each has an apparent memory of some experience of Glenn's before the duplication, involving, say, his heart, and each thought has the de re content about “this heart” referring to the heart each has at that point-- suppose the memory is about how his heart was pounding during the first space flight, and each has the content:

this heart was pounding so hard when I took off.

Where of course when each guy thinks this thought the reference of the phrase `this heart' is fixed by sensations they have in their chests. Ok. Let me use your diagram and suppose each has a token of this thought at a particular time t after the branch.

Diagram. Branchline (duplication)

////X-Glenn (on planet X)

t

Glenn___ ___ /_t_____Glenn

Her point is that the de re memory will be true or satisfiable for Glenn after the branch which is not the case for X-Glenn, since X-Glenn's thoughts with the content “this heart” will be about a heart that did not exist prior to the duplication! --

Scully: Mulder, this may be correct but it is irrelevant--it won't solve the problem -- that is why she wasn't satisfied with the work she did there. In any case, these body-based thoughts are irrelevant given the following point, which is that, as I already pointed out, nobody really has shown how Glenn after is psychologically more strongly connected to Glenn before the duplication than is X-Glenn --much less has anybody suggested psychological reasons why X-Glenn on X would not in any case be connected enough for strong connectedness --And notice that even if Glenn on the Shuttle were for some reason psychologically closer to the pre-branch Glenn--that is, even if the de re body-based thoughts were enough to establish that Glenn is more strongly connected than X-Glenn due to the role of such thoughts etc etc this still cannot save the Psychological account. For if X-Glenn is connected enough to pre-branch Glenn for strong psychological connectedness (even if not as strongly connected as is post-branch Glenn) then there still is a fission from the Psychological point of view, and (assuming with 1984-Parfit no survival of fission) Glenn still does not survive in the Branchline case.--And this is quite contrary to our intuitions. --

Mulder: Oh! I see your point, Scully! Now I see it! Its like what Peter Unger said in Identity, Consciousness, and Value, p. 86: given all this we should just forget about the Psychological approach.-- BUT WHOA BETSY! We would have to forget about it only WERE WE TO ACCEPT THE No-Survival INTERPRETATION OF FISSION.--

interrupt[Scully: Yes, Mulder, that is why I said the Psychological account depends on moving to Perry's-Gricean take on things. The cost of getting survival in Branchline is to grant survival in Fission. --Of course, keep in mind what the girl pointed out earlier: Parfit in 1971 granted "survival" in fission, even though he denied personal identity (he assumed that wedge so that survival wasn't regarded as entailing identity). And there also now is that “non symmetrical personal identity” theory Zoe and Carson just now concocted.-- Those also would be alternatives they could adopt to incorporate the relevant intuitions about branchline. That is, they also could be used in branchline to get an acceptable Psychological account, since they countenance survival of Glenn by virtue of each of the two post-branch persons (Glenn and X-Glenn). So the Psychological account is still on the table, but Branchline shows it has to be a form in which there is survival in fission. --Mulder, how about let's wrap this up? It's getting really cold out here.-- I really have no idea how these kids are managing to sleep in a tent out here tonight--]

interrupt[Mulder: Yes, Scully! You're right! Parfit especially in his 1995 article needs to see this since it looks as if on page 21 he's actually abandoning the psychological reductionism that in earlier years was so dear to his heart, and it appears to me he has been frightened by Unger! Scully, in this branchline case Perry's-Gricean theory --or true, one of those others -- can say that Glenn survives; of course, that theory will say that Glenn survives as each person, Glenn and X-Glenn! In general, it seems to me that the Psychological theory can get Glenn's survival on the shuttle here only by granting it via X-Glenn on X as well (unless somebody can give a relevant explanation of the psychological difference between the two). --]

interrupt[Scully: Yes, that was exactly my point.--]

interrupt[Mulder: True, this result may seem counterintuitive. But for a Psychological theory, the result is not inappropriate, given the mutual R-relatedness--the continuity of mind in each case --and, Scully, your brilliant insight is that in light of the branchline problem, the ONLY viable form of a Psychological account of personal identity is a theory like Perry's-Grice's or Parfit's 1971 view that countenances survival of fission --so thank goodness they ARE viable forms of the Psychological account. To repeat, the branchline thought experiment shows that the psychological theories get wiped off the map unless they move to a Survival interpretation of the fission cases. Otherwise they just succumb to Unger's objection--unless, of course, they can give a Psychological explanation of the difference between Fission and Branchline. But that is unlikely to happen.-- I hope you see this. --Mulder

Scully: Good heavens, Mulder, yes I do see it. That's why I explained it to you! --What are you drinking? --