Perry On Personal Identity

What follows is a (long-ish!) summary of
Perry's A Dialogue on Personal Identity and Immortality

1. The First Night

    The Challenge

Weirob challenges her friend Miller to comfort her on her death bed by showing that there is, at the very least, the mere possibility of her surviving after her death. The next three nights, then, are spent arguing whether such a thing is possible.

    Numerical Identity vs. Qualitative Identity

Weirob is first careful to make the distinction between numerical identity and qualitative identity (in the dialogue, Perry calls the former "identity" and the latter "exact similarity"). Numerical identity is the relation that each thing holds to itself--e.g., I am numerically identical to myself, you are numerically identical to yourself, Jon Stewart is identical to himself, etc. Qualitative identity, on the other hand, is the relation that many things can have to many others, provided that they have the same properties in common. For example, in recitation I talked about how two pieces of chalk could all have the same properties--e.g., they could both be white, cylindrical, so many inches long, kept in a cardboard box, etc.--yet since they are two pieces of chalk they are not numerically identical. Rather, they merely share properties, but are not one and the same piece of chalk.

    The Soul View

Miller's first stab at proving that survival after death is possible involves claiming that people are identical to souls, not bodies. If this is right [so the argument would go] then survival after death is possible because even though your body dies, you--your soul--lives on.

Weirob challenges this in the following way: the soul is defined as something immaterial--something that cannot be seen or felt or touched or smelt, etc. Yet all we have access to are material bodies--things that can be seen or felt or touched or smelt, etc. Souls in principle cannot be seen or sensed in any way; that is, by their very nature they are inaccessible from the outside. So, even though I might want to conclude that you are the same person in class this week as you were last week, the only thing I have to go by in concluding this is what I see or sense. I cannot, for example, see or sense that your soul is here--indeed, souls are just the sort of thing that one cannot see or sense! So Weirob's objection to the claim that people are identical to souls is that there is a serious problem of accessibility.

In class we will discuss one of the flaws of this kind of argument: viz., that it might be committing the intensional fallacy.

Instead of finding fault with Weirob's reasoning, however, Miller instead claims that there is a correlation between bodies and souls, which is why we can conclude that a certain soul is around whenever a certain body is. But as Weirob is keen to point out, we aren't justified in making such claims of correlation if we don't have some other, independent way of showing that souls are around whenever we think they are. Since we can never see or sense that souls are around, then we can never justify the claim that souls are correlated with bodies.

    The Soul View Take 2

Miller attempts the soul view again, this time claiming that we can legitimately establish a correlation between souls and bodies. He claims that because bodies exhibit certain behavior that implies certain psychological characteristics--e.g., because someone may scream this or that, or argue in a certain manner, or be a happy or sad, or be really energetic or act like a drunken fool, etc.--we can infer from this that there is the sameness of soul, and then correlate this with the sameness of body.

Weirob objects that we cannot judge from the sameness of psychological characteristics that we have the sameness of soul. To make her point, she proposes the following analogy: If we wanted to test whether a certain river--say, the Ohio River--was the Ohio River, as opposed to any other river, we would check to make sure that the water was of a certain quality, that it flows in a certain place, that our fishing hole we went to the other day is still there, etc. If we found that the water of a certain river was of a decidedly different quality, or it suddenly had entirely different fish, or our fishing hole was no longer there, etc., we would more than likely conclude that the river we are at is not the Ohio. So: we judge that a river is the river it is because of the qualities that we expect it to have after getting to know it. [Notice the parallel to people: we judge that people are who they are because of the psychological characteristics we expect them to display after getting to know them.] However, a river, while perhaps exhibiting certain characteristics over time, is continually changing waters. That is, since rivers run, there will always be different waters flowing through the same river over time. Likewise, Weirob argues, souls might work the same way. In fact, because we can't be certain that this isn't how souls work, Miller cannot conclude the sameness of souls from the sameness of psychological characteristics.

Miller tries to respond that he at least knows he himself has a soul, and he can thus establish the correlation between soul and body in his own case. Then, he can generalize by analogy to other cases, resulting in the general conclusion that there is a correlation between souls and bodies. [For info on Arguments by Analogy, go to my logic page here.]

Weirob's response is that this move doesn't help him. For the river analogy still is still possible that psychologically similar souls could be flowing in and out of one's body, such that one could not detect a difference. The problem, Weirob summarizes, is that by the very nature of what a soul is--i.e., immaterial, un-see-able, un-sense-able, etc.,--one cannot have a legitimate principle of personal identity based on them.

2. Second Night

    Personal Identity without Bodily Identity

Miller begins the second night by proposing the following consideration: when we first wake up, we know who we are without opening our eyes and looking at our bodies. Indeed, as Descartes has shown us in the Meditations, we can close our eyes and introspect, and we can come to the conclusion that we exist without having to assume anything about having bodies at all. Miller's point is that we can affirm personal identity without having to affirm bodily identity, so the two must be different. His point, he thinks, is further supported by the fact that we can easily imagine waking up and finding ourselves in completely different bodies. (In lecture, I gave the example of swapping my current body for a younger, more fit one, thus showing that such a situation is at least imaginable.)

Weirob grants that all of this might be possible, but then what, she questions, does this show? For even if it shows that body identity is not necessary for personal identity, it still doesn't show that personal identity is soul identity. That is, it doesn't support Weirob's view, but it doesn't show Miller's original soul view either. So, Weirob asks, where does all of this leave us?


Miller explains that they had both been thinking about personal identity in the wrong way. Picking up Weirob's previous river analogy, Miller suggests that personal identity is more like the flow of a river or the expanse of a road, than it is like they were supposing the first night. Miller's idea is like this: Take College St., as an example. Now suppose we take someone new to Oberlin and show him College St. in between Java Zone and  Tappan Square. We point to the street and say, "This is College St." Then we walk west for awhile and point to the street between several houses and say, "And this is College St." Our visitor should not be confused, even if the same road in two different places looks a bit different; for something to be the same road, it need not be the same in all of its parts. Rather, what makes something the same road is that it is connected in the right way by continuous road bits--e.g., what makes College St. College St. is that each of the bits of road that make up College St. are connected in the right ways with the rest of it.

Likewise, Miller thinks, with personal identity. What makes a person the same person over time is that it is connected in the right ways by continuous conscious bits. That is, so long as there is psychological continuity--a continuous flow of consciousness or psychology--this is all that is need for a person to be the same person over time. In this way, Miller discards the the thought that we are identical to some immaterial floating soul, and dismisses Weirob's idea that we are identical to material bodies. Rather, personal identity is the whole stretch of conscious moments, connected together in the right psychological way.

Miller thinks that if personal identity is thought of in this way, then the possibility of survival after one's death should be an easy thing to prove. For all one need to imagine is that there is a Heaven, where some conscious being is psychologically connected to a being here on earth in the right wort of way.

   Memory View

Weirob remains unconvinced, for she demands that in order for this to be possible, Miller will have to be more specific about what being connected "in the right sort of way" is supposed to mean. He responds by saying that it all has to do with memories: so long as someone retains the memories his past self, one is psychologically connected in the right sort of way.

Weirob is careful to point out that there is a difference between actually remembering something and seeming to remember. To get a grasp on  the difference, think of any sci-fi movie such as Total Recall or The Matrix or Eternal Sunshine of a Spotless Mind, where real memories are erased (or never created to begin with) and fake ones are implanted. Also, any crazy nut can claim to remember being Abe Lincoln or Genghis Kahn, but even though this crazy nut might think he remembers being Abe Lincoln or Genghis Kahn, he doesn't actually remember unless he is indeed Abe Lincoln or Genghis Kahn. But if we can't tell the difference from actually remembering and seeming to remember, how are we to tell the difference when we are imagining that the former, and not the latter, situation is possible?

Cohen enters at this point and claims that the important difference between actual memories and merely apparent memories is that the former are causally connected in the right sort of way to the happenings of the world. That is, actual memories are true, and are thus connected to facts of the world; that is, the world causes real memories to be what they are.

Weirob grants Cohen that real memories can be distinguished from fake ones by appeal to the causal process that created them. But how, she wonders, does this help Miller prove that survival after death is possible? [Remember that this is the whole point of the dialogue--to convince Weirob that it is at least possible that she will survive after her death.]

Miller argues that the solution is to imagine someone who has Weirob's memories in heaven, and that these memories were caused in the right sort of way. That is, all we have to do to get Weirob to see that it's at least possible for her to survive after death, is for her to imagine someone in heaven who has her memories and is causally connected in the right sort of way. And we can imagine that God could create such a being in heaven--someone who had all of Weirob's memories caused in the right sort of way...

    Memories plus God view

Weirob responds as follows: Suppose God could create such a being in heaven--someone who had all of Weirob's memories caused in the right sort of way. Being God, couldn't he have also--if he had wanted to--created a second sort of being? That is, isn't it possible that God could have created TWO beings in heaven that had all of Weirob's memories that were caused in the right sort of way? If so, then there would be TWO people in heaven that are identical to her, but not identical to each other (since they are TWO and not ONE). So: either God cannot create someone identical to Weirob, or he is limited to making only one such person. Since God is all-powerful, it can't be that He is limited in how many copies of people he can make, so He must not be able to create someone identical to Weirob.

Cohen intervenes and claims that it shouldn't matter: if God makes one copy, survival after death is possible; if he makes more, it isn't. But since it's at least possible that he make one instead of many, this is enough to get that it's possible to survive after death.

Weirob correctly points out that this is a change in position, since now survival doesn't depend on memories that are causally connected in the right sort of way, but also on there being a lack of identical copies in heaven. This position is less defensible, Weirob thinks, because this makes identity dependent on something entirely external to the agent. That is, it is because of something outside of Weirob (viz., whether God creates more than one copy of her) that determines whether she will survive in heaven or not.

3. Third Night

Cohen begins that third night questioning Weirob's view of personal identity--i.e., the view that persons are identical to their bodies. If people are identical to their bodies, then what becomes of cases where someone's brain can be transplanted into someone else's body? [Cohen claims that just such a thing really happened in California, but noticed that even if this didn't really happened, the mere possibility of it would be enough to make his point.]

At this point, Weirob denies that the the person who would survive after a brain was transplanted into a different body would be the person who's brain was involved. That is, what's essential for personal identity is the body that is involved, not the brain. Weirob maintains that even if the person after the operation has no memories of the body its now in, this doesn't matter. For memories, she thinks, must be caused in the right way and she thinks that they would not be caused in the right way in this case. The idea is that in a choice between psychological continuity or bodily continuity as being responsible for personal identity, Weirob would choose the latter.

    John Perry, A Dialogue on Personal Identity, 1978 (Hackett Publishing).
    Lecture Notes in Phil 26, Professor CDC Reeve (Fall 2004) UNC-CH.

Page Last Updated: Aug. 28, 2009
Back to Meg's Teaching Page
Back to Meg's Main Page