The Problem of Personal Identity in Two Pages

This article was reprinted in the online magazine of the Institute for Ethics & Emerging Technologies, April 25, 2015.)

The problem – Is a person the kind of thing that can die on earth and be alive somewhere else? To understand this consider a thought experiment. If we make a perfect copy of you—complete with your thoughts and memories—is that copy really you or just a duplicate? (If you think the copy is you, then the waking up in heaven scenario is not problematic; if you think it’s just a copy, then the thing that wakes up in heaven isn’t you.)

Personhood at a moment – What is it to be a person at a particular moment? The bundle theory says you are a body and the mental events in your brain. There is nothing more to you than that. The soul theory says that there is more to you than this, there is a core to you that we might call your soul, ego, or self. So the bundle theory denies what the soul theory affirms—that there is some inner core that is the real you. David Hume, the Buddhists and others denies there is any such core.

Split brains – Modern science accords well with the bundle theory, but not the soul theory. For example, if one severs the brains hemispheres one seems to create different persons. This does not fit well with the idea of a unified soul; but on the bundle theory this is easy to explain—there are just two streams of consciousness. Moreover, if the bundle theory is correct—and science suggests it is—then the prospects of immortality seem bleak.

Personhood over Time – If soul theory were true we could say what personal identity over time means that you have the same soul in the past, present, and future. But on the bundle theory it is hard to see what accounts for personal identity.

Qualitative and Numerical Identity – By identity we might mean identity in the qualitative sense—qualities like patience, humor, honesty, etc.—or we might mean identity in the numerical sense—the same birth date, parents, etc.  So if you meet your old high school friend Jim Smith at a reunion, you might find his qualities have changed. He used to be carefree and now he’s serious. But he is still Jim Smith, the guy you went to high school with who was born in a certain year to Mr. and Mrs. Smith. He is not an imposter.

Now you want to be the same person in the future (or in heaven) that you are now;  you want there to be numerical identity. Otherwise you will no longer exist. But what theory might explain this numerical identity? Here are some theories.

Body Theory – x is the same person as y because they have the same body.

Problems – 1) this rules out post-mortem existence because your body will decay; 2) the prince who changes bodies with the cobbler suggests that identity is not tied up exclusively with bodies; and 3) the story of the “ship of Theseus” parallels the human body story—you do not have the same body that you used to have.

Same Brain Theory – x is the same person as y because they have the same brain.

Problems – One problem with this theory is one can be dead and have an intact brain. This suggests that the brain is not the locale of personal identity. Moreover, the brain’s physical structure changes over time even if the neurons are relatively stable.

Memory Theory– x is the same as y because they possess the same memories. (This explains the prince and cobbler, the prince is still the prince even in the cobbler’s body, and it appears to make post-mortem existence possible.)

Problem – Our memories are limited. So if memories make us who we are—we aren’t much. Furthermore, how can memory theory account for personal identity when being “the same person as” should be transitive across time? But memories aren’t transitive across time in this way. At 60 you may remember your 30 year old self, and at 30 you may remember your 10 year old self, but at 60 it is hard to remember your 10 year old self.

We might revise memory theory to deal with these objections by introducing the “memory-links theory.” In this theory x is y because there is a chain of memories linking a person; persons have identity based on psychological continuity. The problem here is how much psychological continuity there really is.

Moral responsibility – Another reason to accept the memory theory is that it fits well with our idea of moral responsibility. The argument is simple:

  • memories imply responsibility
  • responsibility implies identity
  • thus, memories imply identity

Problems – It seems you could remember a past action and no longer be responsible for it because you have changed. Why should I be responsible at 60 for something my 18 year old self did? So responsibility should depend on a person being the same person qualitatively, not just on being the same numerically.

Is the memory theory trivial? – If our memories are unreliable, then they can’t be the basis for personal identity; but even if our memories are reliable, that doesn’t say much about personal identity. Here’s why. If I say: “I am the same person as I was twenty years ago because I remember being the same person,” then I am just presupposing that I am identical with my past self. But that doesn’t show that I am identical.

Conclusion – Philosophers generally agree that soul theory explains nothing, but that bundle theory and some form of psychological continuity best explain personal identity. In addition, accepting the kernel theory gives us good reasons to be selfish; whereas the bundle theory may lead to more concern for others.

(This entry relies heavily on James and Stuart Rachels’ book: Problems from Philosophy.)

4 thoughts on “The Problem of Personal Identity in Two Pages

  1. Are the names you used for the ideas/theories what’s typically used in the community? I am familiar, for example, with what you call the “memory-links theory” as “psychological criterion over time” or “psychological continuity”… I just want to know the proper terms.


  2. You’ll like my upcoming paper. I’m literally planning on publishing it tomorrow to the various preprint servers. The timing is uncanny. Keep an eye out for a paper titled “The Fallacy of Favoring Gradual Replacement
 Mind Uploading Over Scan-and-Copy”.


  3. They are pretty typical, and used that way in many intro college textbooks. However sometimes different names can be used, which is typical in philosophy. For more, the entry in Wikipedia is very good on personal identity, as is the Stanford Encyclopedia of philosophy and the Internet Encyl. of Philosophy.

  4. Thanks for the info. And I’ve never had a problem with scan and copy uploading. I think worries that it isn’t really you are misplaced.

Leave a Reply

Your email address will not be published. Required fields are marked *