(or, How I found my personal identity)
In my last post, I described my first and only experience taking an actual philosophy course, how throughout that semester we took a quick journey through a good number of the major topics in philosophy, and how we were assigned a couple of short essays that semester apparently named after a marginalized demographic of the Harry Potter universe. I remember clearly that the first topic of the course was free will, and accordingly, this subject inspired the prompt for our first squib, which I displayed in the other post. Apparently one of the topics very soon to follow was the issue of personal (and perhaps also object) identity, as this prompted our second squib, which I’ve also found tucked away in the form of a Word document dated October 15th, 2007. It seems that either our main instructor was already getting tired of assigning squibs or the discussion leaders were getting tired of grading them, because as far as I can tell this was our last one even though the semester was only halfway over.
Again, I don’t know exactly what our prompt was. Clearly we were instructed to read an essay of John Locke and somewhat of a response essay by Thomas Reid, the first of which proposed a definition of personal identity and the second of which attempted to poke holes in that definition. It’s unfortunate that I didn’t include sources at the end of these essays. My best guesses for the Locke and Reid essays we read are (at least appropriate sections/excerpts from) this and this.
Unlike the problem of free will, the question of identity was not a favorite topic of mine going into writing this mini-essay; yet I clearly remember that I received an A on this squib, which didn’t happen with the free will squib. Moreover, having been asked to think critically on something that I hadn’t spent much time considering previously, I was able to form and articulate a position that was most likely fairly original among the group of students taking the course, and (perhaps unsurprisingly) it’s the position I still tend towards to this day.
Here’s the essay (once again, everything I originally wrote is in blue).
The papers of John Locke and Thomas Reid are a clear indication that the issue of defining personal identity is a very difficult one. In his paper on personal identity, Locke dismisses many candidates before finally settling on what he perceives as a satisfactory definition. Reid, in his turn, attacks Locke’s definition of personal identity from many angles by invoking thought experiments occurring in conceivable worlds. Both of these thinkers strive to come up with a single definition of personal identity which will apply in all situations for which personal identity is relevant, but both seem to have trouble with coming up with such a definition. In this paper, I suggest that it is not necessary to come up with a single definition of personal identity which applies to all situations, and I will use Reid’s Dreamless Sleeper thought experiment to demonstrate this claim. My object is not to argue that it is necessarily impossible to come up with a universal definition of personal identity, but to show that it suffices to use different definitions for different contexts.
In his paper, Locke creates a definition of personal identity which equates it to continuity of consciousness. A particularly powerful counterargument to Locke’s definition is formed by imagining a person who enters a dreamless sleep and later wakes up with memories from before the sleep but no memories of the sleep itself. We define P1 as being the person who goes to bed, P2 to be the person in the dreamless sleep, and P3 to be the person after waking up. By Locke’s criterion, it seems that P3 and P1 are the same person, but P2 cannot be said to be the same person as P1 and P3. Even if we modify Locke’s criterion to involve a chain of consciousness, we still cannot say that P2 is the same person as P1 or P3. This leads to such absurd conclusions as “No person can go to sleep.” Therefore, it seems that Locke’s conception of what it means to be the same person does not hold, not even for situations that may easily arise in our own universe. However, this does not imply that we cannot derive a definition of personal identity for the dreamless sleeper which will satisfy our practical needs for a particular circumstance, as I will now demonstrate.
First we must ask ourselves why we should care about personal identity. Several situations and contexts come to mind. One particularly important dilemma in which the issue of personal identity comes into play is in the appropriation of punishments. Other situations include the treatment of the dead and the comatose, and the treatment of a patient under anesthetic. Each of these situations can be applied to the Dreamless Sleeper thought experiment, and for each situation, one can arrive at an answer to the question of personal identity.
In every human civilization throughout history, there has been an instinct to punish those who violate the laws of that particular society. In order to carry out punishment, there clearly must be an intuitive notion of personal identity, so that one can ensure that the same person who committed the crime is punished. In order to decide on definition of personal identity that fits this circumstance, we should first ask ourselves what the purpose of punishment is. Let us assume that the purpose of punishment is 1) to deter a wrongdoer from continuing to break the law, and 2) to discourage all members of that society from committing the same crimes. (I will not argue in this paper that the above is indeed the purpose of punishment, but will assume it for the purposes of this example.) Now we imagine a human P1 who commits a crime and then falls into a dreamless sleep (P2). A punishment to fit the crime is agreed upon, and the remaining question concerns the manner in which the punishment should be inflicted.
The first motivation for punishment is to deter the criminal from continuing to commit his crime (or from causing harm in any other way). If this is our priority, then it is pointless to punish P2 except by containing him in custody while he sleeps so that when the physical body wakes up (P3) with the same personality and memories as P1, it will not be able to return to the streets to commit more crime. There is certainly no point in making P2 suffer while asleep, since he or she will not be conscious of it. Therefore, our treatment of the sleeping P2 revolves around the assumption that that physical body will eventually wake up and will need to be kept in custody. (In the scenario where the physical body is not expected to wake up eventually, there is no point in inflicting the punishment at all.)
The second motivation for punishment is to discourage members of the society (including the criminal himself, if he is let free) from committing the crime and thus doing further harm to the community. For the purpose of teaching the criminal a lesson, it is imperative that he have memory of both committing the crime and receiving the punishment. None of this is accomplished by punishing P2, because even if the physical body wakes up (as P3) with the memories of P1, it will not have memories of receiving the punishment. For the purpose of teaching other members of society a lesson, it seems that punishing P2 is similarly pointless, because the sleeping body is not aware of the punishment and therefore is not suffering. (For the purposes of this argument, I am excluding the case of doing something to the physical body which will cause it to suffer after it wakes up as P3.)
After summing up the above, it is possible to conclude that P2 should not be considered the same person as P1 in the context of meting out a punishment for a crime that P1 committed. More generally, in the context of appropriating punishments, person A is the same person as person B who committed the crime if the punishment inflicted on A will be experienced consciously by A and remembered in connection to the crime.
Other circumstances in which personal identity becomes relevant may yield different definitions of personal identity. For instance, we examine the case where P1 makes an appointment to undergo surgery and then is put under anesthesia (becoming P2). In this case, it is clearly absurd for the surgeon to treat P2 as a different person from P1. The purpose of surgery is to remedy a physical problem so that the physical body that wakes up with P1’s memories may have an improved life. The conclusion that P2 is the same person as P1 relies on the notion that P1, P2, and P3 are connected by the continuity of the functions of a physical body. As long as the dreamless sleeper P2 shares the same body as P1 (body bits doing the same corpse-y dance), P2 is the same person as P1 in this context.
A somewhat altered context is the case where P1 expresses a wish to be cremated (“I want (myself) to be cremated following my death”) and then dies. In this scenario, P2 is the dead body (assuming no afterlife, which implies that no entity has the memories of P1 after P1’s death). Then the wishes of the living P1 should clearly be met as though P2 is the same person as P1 (the “myself” in P1’s quote above refers to the dead body, P2). A variation on this case is that of a permanently comatose person, such as the Terri Schiavo situation (assuming that Schiavo was indeed completely unconscious and had no hope of recovering).
The strategy used above for each analysis of the definition of personal identity in the Dreamless Sleeper thought experiment is different from that of both Locke and Reid. In their papers, Locke and Reid strive to come up with a definition of personal identity which will lend an answer to the issue of whether P2 is the same person as P1 and P3 that makes intuitive sense for every situation in which personal identity is relevant. My approach is different. First, I choose one particular situation in which a choice must be made which depends on whether or not P2 is the same person as P1. Then I examine the criteria necessary for the choice to be right or wrong. From this, I derive a definition of personal identity which suffices for that context. This method yields a definition of personal identity for any given situation that comes up and thus, gives us the same practical advantage that Locke and Reid hope to attain. Therefore, although all definitions proposed by Locke may fail as universal definitions, we can still come up with circumstantial definitions which serve equally well.
 While I was already being an advocate of singular they by 2007 (I know I mentioned it in a Facebook note, the closest thing I had to a blog, before that time), apparently I didn’t feel comfortable using it an essay for one of my college classes. Perhaps I chose male pronouns out of carelessness — it’s mostly but not completely consistent, since at only one point later in the paragraph I refer to the criminal as “he or she”.
 Nowadays my choice of phrase would be “commit more crimes”. I’m easily able to believe that I used singular “crime” in the collective sense in that context nearly 11 years ago, though. Funny the little things that jump out at you when looking at old writing.
 One of my few distinct memories of the main lecture content for that course, apart from the relentless use of terms like Thinker and Decider, was a theory proposed (I’m not sure it was credited to any particular philosopher) during our unit on identity that maybe a dog doesn’t exist as a single entity but rather, as “doggy bits just doing their doggy dance”. I’m not sure exactly what this is supposed to mean, and I’m not entirely sure I did back then either. I’d like to think that I coined the phrase “body bits doing the same corpse-y dance” in order to allude to the dog thing in the squib, and that’s entirely possible, but it seems more likely that this was also a phrase handed to us in lecture.
 Here I was referring to this tragedy, which at the time of writing the squib was a fairly recent national controversy.
I guess I look back on this assignment as a happy moment in my development as a thinking person, because I actually surprised myself with the position I wound up taking. I expected (and probably others expected of me) that after some thought I’d be able to come up with some global definition of personal identity that would be an improvement over what both Locke and Reid had to offer. Instead, my answer was more of a meta-answer: it is not necessary to find a global definition (note that I stopped just short of declaring that one doesn’t exist!), but we can make do with a set of local definitions that cover the space of all possible situations in which for practical purposes we care about personal identity. What I just said about global and local definitions is inspired by a common idea in theoretical math, where “global” things are often obtained by patching together “local things”, but I don’t know how securely ingrained in me such concepts were at the time of writing the squib. I am fairly certain I remember likening my solution to a metaphor used by Steven Hawking in The Universe in a Nutshell (which I believe I had a copy of then) where he suggests that there may not be a single model of the universe but that it can be patched together from “local” models that explain phenomena on different physical scales.
I don’t have much more to add on this topic, as I think I gave a fairly good answer in the above mini-essay and haven’t felt the need to seek a more satisfying one. The one additional point I could make about my squib is that it conveniently skates over the rather sticky question of why it’s important to respect the wishes of the dead as though their corpses embody the same person as when they were alive. The whole “respecting the dead’s wishes” thing seems like a weak spot in utilitarianism at least, and while it should be explored, I think the appropriate place for that would be neither in that squib nor in this blog post.
I was pleased to see a similar point of view (but expanded to greater sophistication) expressed in this essay by John Nerst (where the definition of “existence” is discussed), which played to a similar theme as in this other essay also by Nerst (where the meta-answer of “global definitions are not necessary” is proposed, with some object-level discussion of object identity as well). All in all, I would like to see philosophy take a more pragmatic approach in its use of specific terms and in its study of specific profound questions.