**

First night (first part: 10 p.m. -2 a.m. PST)

1-22-99 FBI Electronic Listening Device auth. #53465792093:

Tent site #33

Sacajawea State Park, Cascade Mountains, Washington

[muffled]

[silence]

[inaudible]

[inaudible]

--ginger snaps. Do we have any coffee?

I'll never forget that. I didn't do anything!

I know, me neither. Wow. Actually we might.

Might what?

Forget.

Never!

Joe, stages come and go.

Stages?

A stage is a bunch of mental and physical events glued together. Like a set of events causally interacting.

What does that have to do with me?

That's all there is to you! As persons that is what we are --we are simply a series of mental and physical events glued together -- Could you pour me some coffee?

Glue? What on earth are you talking about?

The glue consists of the causal relations between all the various physical and mental states. That is all there is to you.

Huh. sounds scary.

True. It can difficult to think about. It can make us uncomfortable if it challenges ideas we have about ourselves.

So the glue is causal?

Causal connections between mental and physical states.

Like what?

Like planning something then doing it later on. Your planning to do it is a mental event, and it causes the later action. Talking about dinner one moment, then doing it a bit later. Our existence is constituted by the relationships between states and events like that.

Like what?

Like planning things, Joe, like I just said. When we deliberate, consider alternatives, choose, decide, plan, intend, and so forth. Those sorts of things. Or like remembering earlier experiences. Or simply identifying with experiences, past or expected, as one's own. Or identifying with the person having them, that is, seeing oneself as the same person who had or will have certain types of experiences. Ordinary mental events.

Like my thinking now that I am the same person who blew up the tent last week?

Exactly. And taking responsibility for it too. That sort of thing also figures in. Being the same person now as you were then just means there are these relations.

Am I the same person now?

Well, yes. Sure. You have changed, of course, so you are not “qualitatively” exactly the same. You have new properties. But you are “numerically” the same. --I mean, there's no reason to deny that. Being numerically identical--being numerically the same person-- just means being constituted by states and events that stand in the right sorts of relations. Of course there also is the continuity of various plans and projects; plus the normal physical continuity of our bodies. That is relevant too. The point is that's all there is to it--

Zoe, do you really think of yourself in this way?

Well--

It seems so abstract.

Well, yes it is abstract. But when we stop and think about it, it's the only way that makes sense of ourselves; I mean, since there is no simple, essentially unified soul or anything like that.

How do you know?

Joe, can you even make sense of the idea “same soul” at different times?

Why not?

If there is no way to identify it at one time, how could you treat personal identity as based on a soul? I mean, even if souls were to exist, the glue idea would still be important. John Locke made this point some three hundred years ago.

The same Locke who influenced Jefferson and the Declaration of Independence?

Yes. Very same guy.

So he talked about person glue?

Well he didn't say glue, I don't think. But he had the idea. He focussed mainly on the memory connection over time as the basis for personal identity.

Like remembering now what we just did.

Exactly. Our doing it 15 minutes ago is causing our remembering it now at this moment. It was great! --Want another cookie?

It was weird! --Thanks. --Zo, would just one memory link or intention link be enough to get the glue?

What?

To get the glue?

No.

How many do you need?

Derek Parfit has written a lot about it. He says that over one day you need at least half the causal links that occur in a normal adult during one day.

Mulder: Scully, FYI what we have here is a sloppy characterization of Parfit's theory, not uncharacteristic of Alexander's M.O., I might add. --where did she come up with glue?! -- The framework from Parfit is as follows. The details here matter. I revised his disussion a little bit, for the sake of clarity --for example, he doesn't use the notion "person stage" but it helps to present his view clearly. Please let me know if it makes sense.

(1) Let a "person stage" be: the full collection of mental and physical events occurring for that person during a certain period of time (such as over one normal day, or perhaps brief).

(2) There are significant causal and intentional relations between mental and physical events so that stages are connected when these relations hold between events within the stages. Examples are memory, intentional action, identifying-with -- just the normal physical and psychological continuities.

(3) Let us say two stages are strongly connected when there are enough direct connections between the events within the two stages. Any specific formula for what is “enough” is going to be somewhat arbitrary, so there will be borderline cases. By the way, Scully, that fact is enough to ensure the possibility of indeterminacy of persons -- quite independently, I might add, of any bizarre thought experiments!

(4) Finally we get to continuity which is the core of the reductionist conception of personal identity. It holds between two person stages just when there is a chain of strong connectedness linking them. There can be continuity between two stages even if they are not strongly connected or perhaps even if there are not any direct connections at all between them. By the way, continuity often is abbreviated as “the R-relation.” --]

Scully: Mulder, the what relation?

Mulder: The R-relation.

Scully: Why not just call it continuity of mind?

Mulder: Actually I don't know. Philosophers since Bertrand Russell started calling it R. So I called it R too.

Scully: Sort of a private language? --

Mulder: ok, Scully. In any case we finally get to (5) copersonality:

(5) two stages x and y can be said to be co-personal (stages of one and the same person) just in case there is non-branching R between x and y.

--Hope this helps.]

Scully: Hey, Mulder, why did you slip in the term “non-branching” into the definition of “copersonality” of stages?-- Why does it say copersonality is non-branching R?

Mulder: Ok, Scully. Good question. If R “branches” you can get two or more separate R-streams going. Stages on different streams wouldn't be copersonal.

Scully: I must grant you this is interesting, Mulder. In medical school while dissecting human bodies I often wondered what made a person different from a live body.

Mulder: I am pleased you finally see the point of our investigation--

Scully: I didn't say that. Mulder, could you please remind me-- why are we on this case?

Zo, ok. So you are saying to get enough of the person glue you have to have at least half of the mental connections that occur in one day for a normal adult?

Yes. That's what Parfit says. If from one day to the next there are at least that many connections then you have enough connectedness for identity.

Normal adult? He must've been joking.

Why?

What is a normal adult?

Well--

And what if I now am not strongly connected to me on some earlier days? I mean, there might not be enough of the causal links. For example, between me today and me on my 6th birthday there are not very many psychological links.

Probably not. But over all that time there are enough connections for strong connectedness from one day to the next. And so Parfit says your current stage today is psychologically continuous with your stage on your 6th birthday. You have this continuity if there is the right sort of chain from day to day.

So between me today and me yesterday there are enough connections, and between me yesterday and me the day before yesterday there are enough.

Right. Etcetera. Going all the way back to your 6th birthday. Just so long as there is that sort of chain connecting you today with your 6th birthday then you have continuity. And if you have continuity with that little guy then you have identity with him too, at least in most cases. That is how to make sense of how we continue to exist even while changing.

Zoe, isn't it arbitrary to treat the connections over one day as the standard? Why is one day the right interval to use? Why not use one hour intervals as the right period of time? Or five second intervals?

Good questions.

And how could you count the links anyway? And aren't some more important than others? And why does it have to go back just from one day to next? Today might be more closely linked to the last time we went camping then to yesterday.

Good questions, Joe. These could be investigated, but nobody has done it in any detail, at least so far as I know.

The details should be investigated. The details might matter.

I agree. They might. But for some purposes it is pretty clear the details are not very relevant.

Like what purposes?

Parfit mainly wanted to focus on the ramifications there would be of any theory like this no matter what the details. Especially the ramifications for morality and practical life. The really interesting thing is that no matter what what your favorite formula might be in spelling out the details, there are going to be consequences that seem contrary to common sense.

Like what?

Like the possibility of indeterminacy.

What is that?

It is conceivable, given this type of theory of personal identity, that you could get yourself in a situation in which something happens so that afterwards the question “does Joe still exist?" has no determinate yes/no answer.

You're kidding!

Mulder: There's an email about this earlier in the files--the one from Zoe Alexander wrote to Mark Epstein's lawyers. Here, -- its at

****>>>> Zoe's email re Parfit.

Zoe tried to explain Parfit's argument for the possibility of indeterminacy. Of course she is lying when she says the idea of “nondensity” was developed by her. True, it is in her thesis but obviously she stole it from Liddy's article. It puzzles me why students still think they can get away with stuff like this!

Scully: Lighten up, Mulder. Don't forget how I got you through the Academy. --Ok, I see it. It came through. Maybe I'll read it some time. --By the way, Mulder, isn't she dismissing a bit too quickly the idea that we as persons exist as unchanging, substantial souls?

Mulder: No. Scully, the great analytic philosopher, Gretchen Weirob, explained all this. She was a friend of mine. I was there when she died. A couple of nights before her death, she argued that even if souls happen to exist, they simply are not relevant to our existence as persons-- at least not if we have knowledge of our own identities over time. Actually, I reconstructed Weirob's argument that personal identity through time is not soul identity through time, based on the conversations that John Perry tape-recorded and published in the book, Personal Identity and Immortality. --By the way, in his transcript he refers to me as “Miller” rather than “Mulder” --Perry apparently was a bit hard of hearing.--Anyway, here--I'll post my reconstruction of Weirob's argument:

****>>>> weirobs argument

What it means is that even if souls exist, they cannot account for our identity through time as people. Weirob's point was that whether or not souls exist is a question completely independent of how we should think of ourselves as persons -- at least, given the type of ordinary knowledge of identity that we obviously possess. I hope this is clear.

Scully: Wow. Good for you, Mulder. I wasn't necessarily expecting an argument! Thanks. It looks interesting.. by the way, don't you think we have enough for one night?--

Look, Joe. The main thing here is that as persons we are not some simple, essentially unified, necessarily determinate type of entity. One could argue for this, if an argument were needed. It is something that western and buddhist philosophers generally agree about. I'm not saying that everybody agrees, of course, but there is a sort of consensus that we just do not have knowledge of anything like a soul, in any way, shape or form; and when we think about it, we see that it is very difficult to make sense of it in terms of our real experience of life.

So the consensus means there is no soul?

No, of course not. Maybe everybody is just wrong.

Ok, but why does no soul mean there is the possibility of indeterminacy?

Actually this follows from the points you yourself were making earlier -- there is going to be something arbitrary about any given formula for determining whether there is the right sort of connection between two person stages. So it is very unlikely that it will give completely determinate results about every possible case. Parfit argues for this, and I think he argues for it well. I tried to write about it in my thesis.

Isn't this what Liddy writes about too?

Yah, right. --He plagiarized my thesis.

What is plagiarized?

He stole it.

He did? What is wrong with that?

Joe, sometimes I think you are serious.

Huh? He tried to explain it to me too.

Explain what?

The possibility of indeterminacy-- if there is no soul. What “nondensity” means.

How did he do?

Not very well. He kept complaining because there weren't any pictures.

Pictures would be nice. It is very true that these ideas are quite different from what we normally assume about ourselves. For example, whatever else, we tend to assume that we necessarily are determinate entities --that is, whatever might happen to us, there still would be a determinate “yes” or “no” answer afterwards to a question about whether we would still exist. Yet, as I said, if there are no souls, if there is nothing like that in the picture, then indeterminacy is possible. Just understanding that is enough to get the feel for what “nondensity” means.

Huh.

Parfit tries to discuss how these points are radically contrary to common sense.

huh.

This is why Parfit finally concludes that personal identity over time does not even matter. He literally is claiming that my continued existence is not really what matters to me when I want to survive.

How does he get that conclusion?

Because the chain of relevant causal links --the continuity -- can branch. Parfit uses stories in which continuity branches to argue that identity doesn't matter. He tries to use these stories to explain how lots of our moral ideas are wrong or at least unsupported.

Branching? Like in the Star Trek episode when Kirk gets beamed twice to two different planets?

Exactamente. He becomes two different people. This shows why continuity between you now and somebody earlier doesn't quite guarantee your identity with the earlier person --there can't be an intervening branching of the continuity.

Like when Frost says two roads diverged in a narrow wood and sorry I could not travel both and be one traveller, long I stood--

yes! you know that poem?

I never understood it. Why does Frost say I couldn't travel both?

In a fission case like that or like in Kirk's neither person afterwards would be you.

Why not?

Because if there are two people afterwards, and if they are not identical to each other, and if each would have an equal claim to being you, then we couldn't identify either of them with you.

Whoa. you've got my head spinning, Zo.

Look. I don't know if you need to get too caught up in the argument here in order to see the main point. The value of these types of thought experiments -- for instance, imagining fission of Kirk or your own fission, standing in the middle of a forest, then walking down two paths-- is to make vivid how we are not essentially unified, as Mark Johnston puts it, and what types of problems and questions are at least coherent since we are not essentially unified.

Like if we were souls?

Yes. In the Kirk type case, Parfit says you would have WHAT MATTERS after the branch, because both the people would be continuous with you, and it is something like continuity that matters -- in fact you have what matters twice! But it is the very fact you'd have it twice that means nobody afterwards would be IDENTICAL to you.

Why not? Actually I don't have to just imagine fission. It happens all the time on our home planet. We have machines for it. Like phone booths.

Mulder: Scully, please take note!]

Oh, you don't say!

Actually I've already done that. My other me is back home.

Where might that be?

I can't talk about it.

Mulder: I hope you are paying attention! --]

Scully: For God's sake, they're just playing. --Mulder, have you never heard the word “joke”? --By the way, it's getting cold out here.--

How about just some clues?

Sorry, Zo.

Where is it?

Liddy says I'm not supposed to talk about it.

Who cares what Liddy thinks?

Zo, why does Parfit say you don't survive if continuity branches?

I can't tell you. Liddy says I'm not supposed to talk about it.

C'mon, please! What if I were to play a little bit with this little

No! Don't! That would be torture! In that case I am afraid I would be forced to talk!!

Ok, then.--

Are you a poet and don't--

Talk!

Branching of R just rules out survival.

What?

Branching of R just rules out survival.

What is R?

Oh, sorry, I got distracted. “R” is a shorthand way to talk about psychological continuity.

So today I am R-related to me yesterday.

Yes.

And I now am also R-related to me on the day before yesterday.

Right. Even if not strongly connected.

And so I now am also R-related to me on the day before the day before yesterday.

Right.

And dot dot dot all all the way back, so now I am R-related to me on the day when I was 6 years old.

Right. --Even if you are not very strongly connected. And Parfit argues that it is R, or perhaps R plus strong connectedness, that matters to us when we wish for continued survival. The branching cases show this because Kirk doesn't survive fission but still he has what matters because he is R-related to both guys later on.

So Kirk didn't survive the double beaming?

No.

I still don't see why not.

It's because of the transitivity of identity. If both the later people were identical to Kirk, one on planet Z, the other on Y-- they would have to be identical to each other too. But they're not. They're on different planets!

What is transivity?

Transitivity. A formal property of relations. Let's see. Any relation Q is transitive just in case for any x,y,z it is always true that if xQy and yQz then xQz. Like “being the sister of” is transitive. If Xena is a sister of Yada and Yada is a sister of Zora--

Then Xena is a sister of Zora?

Right.

But what if they are half-sisters--if Xena is a half-sister of Yada, and Yada is a half-sister of Zora--

Ok, good point Joe. Then it wouldn't necessarily follow that Xena is a half-sister of Zora. So my example only works for being a “full sister” of--but that is a transitive relation.

Ok.

Right.

Mulder: Are you getting this, Scully? The point is that identity is transitive, so if A=B and B=C, then A=C. Scully, this is to use “identity” to mean numerical identity --what it means to be one and the same person. The same word “identity” also can be used, of course, to talk about a person's qualities, how they would identify themselves, and so forth --such as in, “I'm searching for my identity,” meaning I'm looking to define myself, etc. But in this Kirk example they are only talking about the numerical sense of identity --just the conditions under which one would be one and the same person through time. To illustrate branching idea, here is a diagram.

Diagram. R-branching of Kirk

///// y

x — R

\\\\\ z

Let the names “ZKirk” and “Ykirk” refer to the people constituted respectively by the stages z and y. The question then is whether the original person Kirk (constituted by x) is the same person as ZKirk and YKirk. If so--that is, if Ykirk=Kirk and Kirk=Zkirk, then the transitivity of personal identity means ZKirk=YKirk. The obection to this idea is that ZKirk and YKirk are on different planets, so how could be they be one and the same person? That's what's going on here.--

Mulder: did you say something? I was polishing my nails.

Well, Zoe, why do you have to say that personal identity is transitive? Is there any law that says you go to jail if you say personal identity is not transitive?

No. Actually a great logician named Prior suggested giving up transitivity on personal identity when he heard about this type of thought experiment. Michael Hand argues for it nowadays but nobody listens-- It is just too extreme. Actually I tried to argue with Parfit about it, but I could never get anywhere with him.

Mulder: There is more going on here than meets the eye, Scully. According to a NY Times article (October 14, 1997)--you can find it earlier in the files--she was arrested after she punched Parfit in Washington Square while screaming (according to an eyewitness quoted in the police report) “I already did survive fission you asshole! My other me is a rock and roll star! And I'm good too!” In fact, that is why we started following this case, as you may recall.

Scully: How could I forget?

Well, it is the story that is odd, isn't it?-- I mean, from your guys's point of view.

What did you say?

It is your story and your theory that are odd. Not to mention you yourself.

True, Joe. Odd but true. The theory, I mean. Actually any interpretation of fission probably is going to be somewhat odd. Since fission is such an unheard-of possibility for us, we almost certainly would have to revise our conceptual scheme or our beliefs somewhere. And it isn't obvious how to do it. Parfit, for example, actually changed his mind about how to think about fission. --In one of his first articles, in 1971, he suggested using the word `survive' so that you would survive fission. But all the same, he held, nobody afterwards would be identical with you. So back then he suggested we place a wedge between the concepts of survival and identity.

Wedge?

A logical wedge: so that for example `Kirk survives as Xkirk' does not entail `Kirk is identical with XKirk'.

That sounds odd.

Right. It would be odd. And Parfit would've agreed it would be a revision in our use of the concept “survives”. But that was his suggestion in 1971 about how to handle the oddities of fission. But later he changed his mind --in his book, in about 1984, he does something different. There he assumes, as we normally would assume, that survival requires identity. I think he may have been influenced by Sydney Shoemaker's nice discussion of this. In any case, ever since then he's held that what matters in survival doesn't require literal survival.

You can have what matters in survival even though you don't survive.

Yes. That is what happens for Kirk in the fission case. He has what matters in survival -- in fact he has it twice, because of the two guys later on. But it is because there are two that he himself doesn't survive--the R-relation branches.

That's odd.

Well, right. That's what we've been saying. Either way you get something that sounds odd. Of course, Parfit himself celebrates this departure from common sense; and he goes on to argue that fission cases show we should re-examine lots of our attitudes about life.

So in 1984 the wedge is between what matters in survival and survival?

Scully: Mulder, what is going on here?

Mulder: Ok. Parfit argued in both 1971 and 1984 that identity is not what matters in survival. This is what she means about a “wedge” between identity and what matters. In the 1971 article, the wedge is placed between survival and identity (and Parfit does not envision a wedge between survival and what matters in survival): so what matters entail survival but survival does not entail identity. In his book in 1984, on the other hand, the wedge is placed between survival and what matters in survival (and in 1984 he does not even discuss the earlier wedge between survival and identity; and he just assumes there that “survival” entails identity). Let me diagram it for you. In the following diagram let > represent entailment. Here is where Parfit places the logical wedge in each case:

(1) what matters > (2) survival > (3) identity

The `84 wedge is between (1) and (2); that is, it denies that

(1) A has in B what matters in survival

entails

(2) A survives as B.

On the other hand, the `71 wedge denies that (2) entails

(3) A is the same person as B.

Scully: Why do you have to put a wedge in at all?

Mulder: Huh? --I don't know.

Yes. So even though he puts the wedges in different places, still in both cases he argues that identity does not matter.

Zoe, why put in any wedges at all?

Well, good question, Joe. I've never been totally clear about this. Actually David Lewis claimed you could have an account pretty much like Parfit's but without putting in any wedges at all between what matters and identity. It never seemed right to me but I couldn't figure out why not.

What did he say?

Ok. Remember relation R?

Psycholonical cogituity?

Psychological continuity. Right. Yes. To simplify suppose that R is a relation between person stages, where your stage now is just all the mental and physical events that constitute you right now.

So R is the glue between stages?

Right. So you persist through time because of stages at various times being R-related. Now Lewis assumes that R is what matters in survival; and then he pointed out that obviously there isn't identity of stages at two different times--

My stage now is distinct from my stage yesterday--

Right. They are two distinct sets of events. So Lewis said the R-relation should not be compared to identity, the way Parfit did it, but rather it should be comparied with what he calls the “I-relation”: this is the relation that holds between any two stages just in case they are stages of one and the same person.

Ok.

And then he says the R-relation just is the I-relation.

And so --in the branching case?

Yah. Ok. In branching Lewis says that there are two people involved in that story: and prior to the branching the two people just shared their stages!

Wow.

Mulder: Let me explain this, Scully. Recall the Kirk branching scenario.

Diagram. R-branching of Kirk

///// y

x — R

\\\\\ z

Let x,y,z be stages. Everyone agrees we have Rxz and Rxy. Lewis's innovation is to suggest that R=I, that is, the relations R and I are necessarily coextensive. So Ixz and Ixy here.

Scully: But what about the two stages z and y after the fission then--the ones on the two different branches? Are they related?-- Do we get Rzy and Izy?

Mulder: No. Neither Lewis nor Parfit want that result: neither want to say Rzy or Izy. Now Parfit avoided it by denying that R is symmetrical (that is, Rxy does not entail Ryx); and he and Shoemaker formulated the criterion of personal identity that requires no branching between stages --this is what I told you about earlier. And so, of course--to put it in Lewis's terms-- Parfit and Shoemaker deny Ixy: the post-fission stages afterwards are not stages of one person.

Scully: But what about Lewis?

Mulder: Well, now a small but important detail Zoe forgot to mention is that Lewis stipulates that R (which remember just is relation I for him) is symmetric but not transitive. So from Rxz and Ixz we can infer both Rzx and Izx (because of symmetry). However, since R is not transitive we cannot infer Rzy from Rzx and Rxy. And so he also denies Rzy and Izy.

Scully: I think something odd is going on here.

So, Joe --look. A stage x prior to the fission is R-related to stages on both streams afterwards; and of course x is I-related in just the same way, since for Lewis R and I are necessarily coextensive.

What does that mean?

Relations R and I are necessarily coextensive if any two stages are R-related just in case they are I-related.

Ok. But why doesn't Lewis get both Rzy and Izy here?

Uh. Ok. Good question. -- Ok, let's see. He wanted R to be symmetric in order to make it correspond with I, since obviously Ixy if and only if Iyx.

Why is that obvious?

Because x and y are stages of one and the same person just in case y and x are--it doesn't matter which way you say it.

Huh.

Ok. But he denied that R is transitive! Ok, I forgot to mention that. Ok-- that's it, I remember--his reason was suppose you have somebody like Methuselah in the Bible, who lived a long time, like 960-some years--

In the where?

In the Bible, Joe. --Jesus, you are illiterate, aren't you? --Anyway over such a very long period there might not be any connectedness at all. --right.-- So Lewis' defines his R so that it requires both continuity plus some degree of connectedness. So over a really long life we might have continuity --a chain of connectedness from one point to the next and so on, but near the two end points there might not be any direct connectedness at all. And since Lewis requires some direct connectedness for R, the two end points would not be R-related. And then since R=I, you don't have I-related either. So Lewis's position is at least coherent--since I guess its plausible this could mean you don't have the same person over such a long period either.

I'm not sure about that.

Ok. But just assume it for the moment to see Lewis's picture. Ok, then. Back to the fission case: Lewis gets Ryx from Rxy because his R is symmetric. But since his R is not transitive,we cannot infer Ryz from Ryx and Rxz. And so Lewis consistently denies Ryz.

And Iyz?

Yes. Remember R just is the same relation as the I-relation for him.

Ok. But I thought you told me Lewis was objecting to Parfit.

Yes, he was. He was replying to Parfit's 1971 article where Parfit said one could have survival without identity. Parfit said you could survive even though nobody afterwards would be identical with you. Parfit acknowledged this is an odd result, contrary to common sense, and Lewis said he wanted to reply on behalf of common sense. Lewis says the two concepts do not come apart, and if they did come apart, he would have to believe commonsense rather than his philosophy.

But I thought you told me identity is transitive.

Yes, it is.

But the I-relation is not transitive.

No. But remember the I-relation is a relation between stages of person whereas identity is a relation between continuant persons.

Continuant persons?

Just the series of interconnected stages. So for Lewis identity of persons is transitive. But the I-relation is not.

So Lewis agrees with Parfit after all?

No. Joe. I just explained.--

Zo, look. You said Parfit holds that identity does not matter. It is survival that matters, where survival doesn't entail identity.

Right. That was the 1971 position. Lewis was replying to that. He says you don't need to place a wedge between survival and personal identity.

Ok. Let's just focus on that. Now Lewis talks about relations between person-stages. Now it seems to me that since personal identity is both symmetric and transitive, the corresponding relation between stages also should be both symmetric and transitive.

Yah. Ok. Well, Parfit and Shoemaker certainly would buy that. You could call the relation “copersonality”. They would treat the “copersonality” relation as both transitive and symmetric, given that copersonality holds between stages just in case you have the non-branching R relation between them.

Ok. But, Zoe, it seems to me that identity matters just in case copersonality matters. But Lewis basically says copersonality does not matter. Copersonality is transitive, but the I-relation is not transitive; and for him you said it is the I-relation that matters. So it looks to me like he agrees with Parfit that identity does not matter.

Wow.

What?

Joe, you're right!

What?

That's what was wrong! Now I see it!

What?

What you just explained! --Lewis made it look like he was the common-sense conservative in the debate with the radical Parfit. But you're right! He simply assumes more or less the very thing for which Parfit was arguing-- that copersonality does not matter, since on his view it is the nontransitive I-relation that matters.

You didn't see this before?

No, I didn't see it. -- Ok, I see what's going on. This point gets hidden because in a case like the Kirk fission case, a statement like `Kirk is the same person as Xkirk' is ambiguous--

Why?

Because the term `Kirk' is ambiguous when we use it to refer to the person constituted by the pre-fission stage x--

--because two people share that stage?

Yes. Exactly. Prior to the fission two people share the stage, so even though it is natural to use a term like `Kirk' to talk about “the guy over there” before the fission in fact there are two distinct guys over there, sharing a body, sharing thoughts, and so forth--

There might even be more than two.

Right. It depends on how many fissions later on there will be! But let's keep it as simple as we can. The point is that Lewis doesn't have to reject transitivity on personal identity, the way Prior had suggested, but this is only because no statements like `Kirk is the same person as Zkirk' or `Kirk is the same person as Ykirk' are true.

Why not?

I just explained it: because any use of `Kirk' to make reference to a person, prior to the fission, will fail to make successful reference--remember there are going to have been two distinct people sharing stages all along. So a statement like `Kirk is the same person as Zkirk' turns out false or at least it lacks a truth-value. There are different theories about how to handle statements where a referring term like `Kirk' does not succeed in making successful, unambiguous reference. But the details don't matter because in any case the statement is not true.

Zoe, there's something wrong with Lewis's position.

So you don't like the stage-sharing idea?

No, that's ok. I mean, it's weird, but I could accept that. It's just that--

I agree. That's what I meant before. Actually Lewis performs a sort of amazing philosophical magic. He simply assumes that for which the revisionist Parfit was arguing --that copersonality does not matter-- but then makes it look as though he is defending the commonsense idea that survival and what matters do not diverge.

Hmm. That's not it--

What you've shown is that Parfit wins the debate hands down. Even if you grant stage-sharing, Lewis doesn't defend common sense since common sense would treat copersonality between stages as an equivalence relation.

A what?

Equivalence relation: both transitive and symmetric, where reflexivity follows from transitivity and symmetry.

Common sense says that?

Well, no. Ok. Common sense never really considered that, I suppose. But that is certainly what would be presupposed about the copersonal relation between stages just as certainly as that personal identity itself is an equivalence relation. So, good job, Joe! --You're pretty good at this.

Thanks.

Well, thank you. I was trying to figure this out when I wrote my thesis but I never got anywhere. So what I'd say now is: ok, technically Parfit wins the debate with Lewis after all. But still Lewis was doing the right thing: generally instead of just trying to spell out abstract revisionist doctrines such as “identity does not matter” maybe Parfit should emulate Lewis.

How?

Well, Lewis simply assumes that the I-relation diverges from “copersonality.” He was wrong to say he was defending common sense, which is why Parfit technically wins the debate. But the point is to develop theories that enable us to think clearly about ourselves. For instance, to make vivid how to think in terms of a conceptual framework in which survival does not entail personal identity and so forth like in the 1971 article. Maybe he should reconsider his 1971 position.

Ok, maybe. But, Zoe, I think there is a problem with Lewis other than the magic. Its something else.

What?

If R is not transitive due to Melusethah situations--

Methuselah.

Ok. Mesulethah.

No, Methuselah.

Ok. Whoever. If that is why R is not transitive, then how could Lewis defend the idea that Ryz does not hold in the Kirk fission case? That is, how could he explain why the two stages afterwards are not R-related?

What? Why not?

You said everybody agrees we have Rxz and Rxy.

Right, that's just the set up for the story.

And you said Lewis has Rzx because his R is symmetric; he needs that in order for R to jive with the I-relation. --Right?

Right. He assumes I is symmetric, and makes R symmetric too because he wants I and R to turn out to be necessarily coextensive.

So Rzx follows from Rxz because R is symmetric.

Right.

Ok, so Lewis has both Rzx and Rxy. And then you said that technically he can block the inference from Rzx and Rxy to Rzy because R is not transitive. Right?

Right.

But the reason R is not transitive had to do with the Melusethah--

Methuselah--

--cases. But the Kirk fission case is not like that one. In the Kirk case we do not have such a long period of time at all. It could be as short as you please!

Wow. Joe! You are absolutely right! --

So why should transitivity fail here?

Yes, I see your point! Lewis has no explanation for that!

Really?

Yes! This is where Parfit's rejection of symmetry of R becomes important-- He denied symmetry on R all along precisely to avoid getting Ryz in this fission case.

It looks to me like Lewis needs to deny symmetry on R as well.

Well, Joe, I think you are right --but wait. Then he has to say the I-relation isn't symmetric either since he wants to maintain that I and R are necessarily coextensive.

Yah.

Hmm, a copersonality relation between stages that is not even symmetric?

Yes, I think--

No, that's going too far--

I think that is what Lewis needs here.

--Wait --Joe, I just had an idea. Maybe that isn't so bad! Since it looks like one needs to deny symmetry on R, like in Parfit, why not do that? -- ok, wait. Just do that.--But then you could just forget about the stage sharing business. Right. just assume only one person beforehand; but after the fission there are two people. One person becomes two people. That sounds right.--Ok. One way to make this work is by rejecting symmetry on personal identity the way Prior and Hand reject transitivity. --But don't reject transitivity.

Ok. I didn't really like that anyway. I mean, in the Melasethup cases

Yah, right.

-- it seems to me that you'd still have one person throughout all that long period of time even without any direct connections just so long as you had the chain of strong connectedness.

Ok. Well, that is controversial. Mark Johnston said exactly the same thing.

Ok, good.

Ok, then. For present purposes, let's consider a view where we don't give up transitivity on the I-relation-. But we give up symmetry. We have a conception of personal identity that is transitive but not symmetric. It corresponds to a conception of copersonality that is transitive but not symmetric.

So copersonality is just the R-relation?

Yes! That's right, Joe. So copersonality is just R --I mean, it's the way Parfit has always thought about R. Copersonality is just the same thing as psychological continuity. And rejecting symmetry at least makes more sense than rejecting transitivity --at least if you are fiddling with the concept in order to handle the fission cases.

So what would it mean?

Ok, let's see. We can say `Kirk is the same person as Zkirk' because x is copersonal with z --there is Rxa. But it wouldn't necessarily be able to put it the other way around; it wouldn't necessarily follow that `Zkirk is the same person as Kirk' since z is not copersonal with x. So just like R, copersonality just goes one way.

That sounds odd.

I know it sounds odd. But like we said, with these cases there is going to be oddity somewhere. Copersonality between stages has a direction --its the direction of going forward in time.

Hmm.

This new idea places the logical wedge between personal identity and identity as such. Kirk becomes Zkirk, Kirk survives as Zkirk, Kirk just is the same person as Zkirk. But we don't say Kirk is identical simpliciter with Zkirk. --

Simpliciter?

Uh. Just the notion of absolute identity, pure and simple. --And of course, we can't go backwards with the various claims: Zkirk does not become Kirk, Zkirk does not survive as Kirk, -- those claims sound pretty good--

Maybe that is just the alternative theory Parfit should develop.

Maybe so. I like it. It is like his wedge in 1971, except it keeps the connection between survival and personal identity; it puts the wedge like Prior between personal identity and identity simpliciter. But of course on this view now we also are stuck with saying-- ZKirk is not the same person as Kirk. So it does matter how you say it!

Kirk is the same person as Zkirk, but Zkirk is not the same person as Kirk?

Exactly. That's funny. You have to stay on your toes. But remember this ties personal identity itself to becoming and survival. That's the nice part about this theory.

So I now am not the same person as anybody in the past.

Well, Joe, I guess that follows.

And I will become somebody in the future --but I am not yet that person.

No.

So I am only here right now.

Yes, I guess so. That is interesting, Joe. that is the “full presence” assumption. I never thought a reductionist picture could explain it, but maybe this version could.

Full presence?

That I do not have temporal parts. That I exist fully at any moment at which I exist at all. That intuition might underlie the idea that we are “dense”, soul-like. It is interesting to see that you can get something similar emerging out of a reductionist theory like this one.

But only because I don't even exist at other times?

Right.

That is pretty extreme.

Ok. True. But I like it.

Zoe, I don't know. It seems so odd to say that nobody in the past is the same person as me.

Well, I know. But, then again, remember that somebody from the past is you! Joe yesterday survives as you; that guy yesterday has become you now. --But ok, I agree it is odd. But I remind you again that something is bound to turn out odd here. And this oddity might be well worth developing.

Scully: Good Lord, Mulder! Am I getting this straight?

Mulder: I wish Weirob were still alive to sort it all out.

Scully: So in addition to the wedges you drew before we also have a new one now.

Mulder: What?

Scully: Well, we also have this new "Non-Symmetrical personal identity view" she's been talking about, where personal identity doesn't entail identity simpliciter. Like in Prior's suggestion. But she abandons symmetry rather than transivitivy on personal identity. She just explained it. Survival, personal identity, and the R-relation all turn out coextensive. Using your earlier chart we add another wedge:

(1) what matters > (2) survival >

(4) personal identity > (3) identity

So on this new Non-symmetrical view, survival entails personal identity, but personal identity doesn't entail identity as such; the reason is that personal identity is not symmetrical whereas identity is.

Mulder: I don't think Weirob would like this sort of fiddling around with the idea of “personal identity.”

Scully: She was pretty conservative?

Mulder: Oh, yes.

Scully: Well, I don't like it either. By the way, Mulder, why do you have to put in any wedge at all?

Mulder: I'm not sure.

Scully: Mulder, how about let's call it a night? I think we have enough for one day.

Mulder: Stay with it, Scully. We need more evidence.

continue first night (2 a.m. - 4 a.m.)

use web browser to go back

return to homepage The Z-Files