Reasons and Persons by Derek Parfit
"Like my cat, I often simply do what I want to do." Derek Parfit
Unfortunately, people are not cats and we expect them to have reasons for performing certain actions, be those reasons moral, rational, or otherwise. The reasons we have for acting in a specific way is one of the major questions that Derek Parfit attempts to answer in Reasons and Persons, with the other being what is the nature of personhood. The book is divided into 4 parts, with parts 1 and 2 focused on reasons for actions, part 3 on the nature of personhood, and part 4 on the repugnant conclusion. I find Parfit’s arguments largely convincing, and the book is quite readable for a 500 page philosophy text. Others may be more shocked by Parfit’s conclusions and find them difficult to accept. Parfit believes that persons are fictional, or at least that personhood does not matter in any important sense. Parfit is concerned with personhood in the sense of “am I the same person I was as a child?” not in the sense of “are dolphins persons?” He also believes that in some cases determining whether someone is the same person as someone else is impossible and/or meaningless. The organization of the book is somewhat scattered; he will go off on tangents that will slowly work their way back to the main point of the chapter, but it usually easy to follow along. He also has nice summaries at the end of each section that are well organized and help re-focus the previous discussions on his central points. His jargon and theories are easy to understand, though it can be difficult to keep track of which is which when there are theories S, P, Q, and others floating around. I have read the book multiple times, which definitely helps in understanding and picking up smaller things you might have missed the first time. I would strongly recommend the book to anyone interested in ethics, personhood, or rationality. I would still recommend it, though less strongly, to anyone else.
Part 1 is probably the least interesting part of the book and the least organized, though not without its merits. Parfit frames the discussion in part 1 around a theory of rational self-interest that he calls S. S states that each person’s ultimate aim is for their life to go as well as possible. In other words, what is rational to do or what you have the best reasons to do is whatever would make your life go well. Although there is a lot of ambiguity in what “as well as possible” means, it does not affect Parfit’s discussion too much. Another theory of rational self-interest Parfit discusses is P. P states that each person has most reason to do whatever will fulfill their current goals and desires. Unlike S, P is only concerned with the present.
One of the more interesting side discussions that arises from Parfit’s evaluation of S is collective consequentialism, which sort of combines Kant’s universalizability (you should act in such a way that you could rationally wish that everyone would act in the same way) with utilitarianism. This moral theory states that one should act in such a way that if everyone acted in the same way, utility would be maximized. For example, under regular utilitarianism you might figure that you would maximize utility by donating 90% of your salary to people suffering from global poverty. This might make you very unhappy, but so many people in the 3rd world are helped that total utility is maximized. Under collective consequentialism you would calculate instead what would be the ideal amount for someone to donate in your position if everyone in a similar position donated the same. This number would be much lower than 90%, because in this ideal situation many more people are donating. It might be something like 50%. Although the world as it is would have less total utility by collective consequentialism, it has two major advantages over regular utilitarianism. It is more fair, in the sense that everyone has the same obligations and the fact that other people are awful and are not doing their fair share no longer means that you have to sacrifice even more to make up for that (If everyone was donating more in the first example, the ideal amount for you to donate would be less than 90%. For each person that opts out, the amount that you should donate goes up). The other advantage is that calculating utility becomes easier because you get to assume an ideal world. You no longer have to figure out how each person would act in the real world; you get to assume everyone acts the same and acts morally. Obviously, these assumptions are not going to be accurate in the real world, but at least it makes the very difficult problem of actually sitting down and calculating utilities of various actions more tractable.
Part 2 builds off part 1 and argues that S is not the correct theory of rationality. Morality would indicate that you have reasons to be concerned with other people, not just yourself, so S needs to make a case that ignoring other people in a moral sense is rational. However, S also needs to make the case that ignoring your future self is not rational, otherwise P would be true, not S. In other words, if you don’t care about other people, why should you care about your future self? The answer to this question may seem obvious, namely that you will never be other people but you will be your future self, but Parfit undermines the idea that one is the same person existing throughout one’s whole life in part 3. (There is a lot more discussed in parts 1 and 2, such as Parfit’s namesake hitchhiker, but I found part 3 to be by far the most interesting part of the book, so I am drastically simplifying and compressing the first 2 parts).
In part 3 Parfit discusses personal identity in what is likely the meatiest and most famous part of the book. He begins with the transporter thought experiment. A machine scans your body, and makes an exact replicate (quantum physics notwithstanding) on Mars while destroying the original. Parfit considers several alternative variants as well, which result in your original body being fine or your original body being sickened and soon dying instead. Which of you (or both or neither) is “you” in each instance? If you died, is your replicate “you?” How concerned should you be if the process sickened you and you will die shortly if you have a healthy replicant elsewhere? If you survived, are you both you? Is there a theory that can explain what happens in all 3 scenarios satisfactorily?
Parfit evaluates several theories of personal identity. Physical continuity is one possibility. You are the same person over time if you keep the same body. In this case none of the copies are you in the transporter thought experiment, because your body is still on Earth in whatever state it is in and your replicant is on Mars. The degree to which your body needs to be the same is debatable, but there must be some part that is identically or at least continuous with the other parts. Psychological continuity is another theory, one that is based on continuous memories instead of a continuous body. I remember being myself yesterday, yesterday I remember being myself the day before, therefore I am the same person today as I was 2 days ago. In this case the copy will be you, although it does run into some confusion when there are multiple people who are or were psychological continuous with each other running around. Parfit will argue for a version of the psychological continuity theory of personhood. These are both reductionist views, in that personhood is merely used as descriptions of a set of facts. Parfit believes that being a reductionist about personhood compels you to believe that personhood is unimportant, but other philosophers would disagree. There are also non-reductionist views on personal identity, such as that having the same soul over time makes you the same person. The soul exists independently of mere facts about the body or mind. Parfit also suggests that there can be materialist non-reductionist theories about personal identity, but I don’t know if there really can be.
For example, most people are reductionists about nations. Nations clearly exist, and consist of people and/or territory, but there is no deeper fact or soul of a nation beyond the people or territory. In addition, reductionists will be open to the fact that in some cases we will not and cannot know if the object in question is the same “thing” or not as it was before. Whether the Austrian Empire is the same country as the current Austria does not have a “true” answer. There may be arguments for saying that it is and arguments for saying that it is not, and it is possible to reach a consensus on the question, but there is no logical proof one way or the other. This seems all well and good in regards to nations, but in terms of personhood, being a reductionist means that questions like “Am I about to die?” and “Will I be the same person a year from now that I am today” may be answered “Maybe” “If you want to be” or “I have no idea.” Parfit holds that this is the correct view towards personhood. Going back to the teleporter experiment, there is no “true” answer to the question “am I about to die?” according to Parfit. There will no longer be a body and mind like mine on Earth, and there will be one on Mars. That’s it, that explains the whole story. You can call the one on Mars me or not, it’s just a matter of preference.
Under normal circumstances, answering “I don’t know” to the question if someone is the same person or not is absurd, but Parfit proposes another thought experiment that would make it appear to be a reasonable answer. If a mad scientist replaced your memories and personality slowly with that of another person, at what point do you “become” a new person? Answering that you are definitely yourself early in the process, definitely someone else later in the process, and you don’t really know who you are in the middle of the process seems reasonable, at least compared to the alternative were changing one tiny memory or trait changes you from 100% yourself into 100% a new person. Although this experiment has never been done and probably will never be done, we do have a lot of evidence from people suffering from brain trauma and mental illnesses that you can lose a piece or pieces of your memory or personality while keeping the rest intact (See Oliver Sack’s The Man who Mistook his Wife for a Hat for some extreme examples). Whether or not these are the same people as they were before is a difficult question, which Parfit would argue has no right answer. If instead we found that brain trauma would either do nothing or completely rewrite everything in your mind, that would have been evidence for a non-reductionist view of personal identity.
For another example where there appears to be no answer to if you are the same person or not, Parfit proposes that your brain could be split and transplanted into two separate bodies that belong to your other identical triplets. Parfit assumes that the human brain in this thought experiment has 2 independent and nearly identical hemispheres. I do not know if current neuroscience would support these facts. His argument maintains at least some force as long as the scenario is physically possible, even if it could never be done with human brains. Although you would lose your own body, both new people would be very similar to your original self. Any possible answer to the question which, if any, of the 2 new people is the same person as the original doesn’t work. That neither of them is the same person doesn’t make sense because people lose significant parts of their brain to strokes and the like and remain the same person as they were before. Your body is nearly identical to your original body in this scenario, and in real life people’s bodies change drastically over time. It would be impossible to select only one of the two as you, because they are identical after the surgery (or at least they are equally and mostly identical to the original). There could be no possible basis for choosing one over the other. Calling both of them the same persons seems to undermine some basic facts about personhood, like that persons are unique and do not have multiple independent bodies (I would be curious to see a theory of personhood that allows for multiple independent bodies). We have all the information, but still no idea if the same person still exists over time. From the point of view of the person whose brain was transplanted, both new “persons” would still have most of memories, personalities, skills, etc and definitely feel like they are still the same person as before.
“There will be two future people, each of whom will have the body of one of my brothers, and will be fully psychologically continuous with me, because each has half my brain. Knowing this, we know everything. I may ask, “But shall I be one of these two people, or the other, or neither?” But I should regard this as an empty question. Here is a similar question. In 1881 the French Socialist Party split. What happened? Did the French Socialist Party cease to exist, or did it continue to exist as one or the other of the two new Parties? … Even if we have no answer to this question, we could know just what happened.”
Parfit argues that division in this manner, as well as teleporting, is just as good as ordinary survival. Since questions of personhood are empty, the fact that the “people” in the scenarios have the same memories and personalities as before, and nothing feels that different internally, they are fine. Personhood does not matter; maintaining thoughts, goals, feelings, etc is what is important. If we consider these scenarios “from the inside” nothing seems wrong. Nothing important is missing. If you were to feel different, have confusing memories or a sudden change in personality, that would be a sign that something significant happened, but that is not the case in the relevant though experiments. I simply step into the teleporter or lay down on the operating table, briefly fall unconscious, and wake up on Mars or on a different operating table. It would feel no different than waking up from anesthesia after surgery. You wouldn’t notice anything wrong or missing. To say that you “die” when going through the teleporting is to say you die when you go under anesthesia. You could be destroyed and replicated constantly, and it would both make no difference and would be unnoticeable.
“I want the person on Mars to be me in a specially intimate way in which no future person will ever be me. My continued existence never involves this deep further fact. What I fear will be missing is always missing… Judged from my standpoint of my earlier belief, this not because teletransportation is about as good as ordinary survival. It is because ordinary survival is about as bad as, or little better than, being teletransportation. Ordinary survival is about as bad as being destroyed and replicated.”
This is the most important idea in the entire book, and one of oddest and most profound ideas I have ever encountered. Ironically it is almost identical to the Buddhist idea of no-self, which Parfit also acknowledges, although he arrived at in a very different fashion. Parfit even describes this realization in a way that if you squint enough looks like Enlightenment.
“I find it liberating and consoling. (Before) I seemed imprisoned in myself. My life seemed like a glass tunnel, through which I was moving faster each year, and at the end of which was darkness. When I changed my view, the walls of the glass tunnel disappeared. I now live in the open air. There is still a difference between my life and the lives of other people. But the difference is less. Other people are closer. I am less concerned about the rest of my own life, and more concerned about the lives of others.”
This seems like it should have profound consequences, which Parfit acknowledges though downplays. Does it make sense to do things like save for retirement, punish people and hold them accountable for acts they committed in the distant past, or even to keep promises or commitments made in the past if we can’t say we are the same person over time? Is worth it to start smoking now to look cool even if “I” get lung cancer 30 years from now? Should we even be afraid of death, if we are being “destroyed and replicated” all the time? Is there any reason to favor ourselves or our future selves over other people? Parfit does hold that memories, personalities, and goals are important and provide reasons for considering the future and not wanting to die, although we should be less concerned about these things than most people are now. I worry though that this is not enough; many people have completely different personalities and goals at 50 than at 20, and have forgotten a lot of memories and made a lot of new ones. How strong is this relationship between your 20 year old self and 50 year old self? And can it justify things like not dating that hot girl now because she would be a terrible mother and wife in 20 years? And if you barely resemble your 20 year old self at 50, should your 20 year old self regard this change as being as bad as dying? I apologize for the question overload, but these are things that I think we must think about if we take Parfit seriously, and I don’t really know how to answer them myself.
Parfit holds that it could be rational to give our future selves no special concern, although giving our future selves special concern is permissible as well. The best Parfit can do is give a moral argument for prudence. No one will be in a better position or have a greater ability to help your futures self then you, so you have a moral duty to do so, just like you have a lesser moral duty to help other people. You could also make the argument that you have a special duty towards your future self like you have a special duty towards your spouse or children. This is a lot weaker than traditional arguments for prudence, but at least it is something.
This argument has other interesting effects; we often don’t morally criticize people for actions that hurt only themselves, but by this view we should treat these kinds of actions in the same way we would if they were hurting someone else. By the same argument, paternalism and restricting others’ freedom for their own good would be easier to justify, and hurting someone for their own future benefit would become harder to justify. It also seems like an argument for effective altruism; how can you justify spending millions on your future self’s retirement instead of on people living in extreme poverty? How can the benefit of one person in the future (who isn’t you and at best you have some vague responsibility for) outweigh benefitting hundreds of much needier people now?
In regards to punishment Parfit similarly believes that punishment is not required but is still permissible under his view of personhood. He does not make many definitive claims, except that as someone changes over time away from the person they were when they committed the crime, they deserve less punishment, although perhaps not none. His example is that an elderly Nobel Peace Prize winner should not be punished if it comes to light that he started a drunken brawl in his 20s, even if it would have been correct to punish him at the time. This probably doesn’t fall too far out of the average person’s moral intuitions, but I can imagine more bizarre examples that would. And since Parfit is such a big fan of crazy sci-fi thought experiments, I’m sure he would not have minded if I do one myself. Imagine a killer is on the run and is about to be captured and severely punished. However, he has the ability to instantly change his personality and forget all details of his crime. He and no one else knows if the changes will be permanent. If he does changes himself, the police can’t punish him, since he is no longer connected in any important way to the criminal (according to Parfit). Maybe they could detain him until his original personality returns, but the longer this would take the less plausible it would become. We don’t preemptively detain people except in extraordinary circumstances, and imprisoning an innocent man because at some point he might become someone dangerous or someone who should be punished seems really weak. Theoretically anyone might become someone dangerous at some point in the future. It kind of becomes a get-out-of-jail free card if you are willing and able to drastically change yourself after committing a crime. (This thought experiment was inspired by Memento and Death Note). Parfit might be fine with this, as he considers drastically changing yourself in such a way to be just as bad as dying, and suicide already functions as a get-out-of-jail- free card of sorts. I don’t know how he would account for the chance that you would revert to your original self though. Parfit readily admits that connectiveness over time is something that you can easily influence, so you could choose to give yourself greater or less unity over time by how much you change yourself, although he does not consider anything as drastic as my thought experiment.
Part 4 deals with several things, but the primary one is the Repugnant Conclusion. The Repugnant Conclusion states that by utilitarianism, it would be better for a vastly greater number of persons to have lives just barely worth living by whatever measure of utility you chose than for far fewer people to have much better lives. In mathematical terms, a situation with 1 million lives each having a net utility of 1 is better than a situation where 1,000 people each have lives of net utility 100. 1,000,000 x 1 > 1,000 x 100. This conclusion can also apply over the course of a single life. Is it better to have a very long but mediocre life or a short amazing one? 100 years of utility 1 is greater than 30 years of utility 3. Parfit finds this situation counter-intuitive and disturbing, hence the name. On a smaller scale, questions of whether or not to have children (or have an abortion) follow similar logic. How many children is the “correct” number to have? As many as you can while maintaining their (and your) quality of life at a level that is worth living? Or the amount that will maximize the existing children and your quality of life, which is likely a much smaller number?
Parfit attempts a number of mathematical tricks to avoid the Repugnant Conclusion, including that quantity of people shouldn’t matter above a certain point, that suffering shouldn’t matter beyond a certain point, that happiness shouldn’t matter beyond a certain point, and that some very good lives might be infinitely more valuable than any number of lives barely worth living. He finds all of them implausible, since one could construct a similar thought experiment to the Repugnant Conclusion for each of them which is even more absurd and/or repugnant. The more obvious solution would be to reject utilitarianism, which Parfit does not consider. I am struggling to imagine how other moral theories would even tackle the question of which world is better. Which world does not use people as mere means? Which world would the man of virtue chose to bring about if he had the choice? (Maybe the smaller scale question of how many children to have could be answered by one of these theories, and then extrapolated). While worrying, I don’t think the Repugnant Conclusion is a knockout argument against utilitarianism, and I am not as repulsed by it as Parfit. Until we have a better handle on what exactly a life that is worth living, but only barely, would even look like, I don’t think it’s worth spending too much time on.
I would recommend this book to everyone reading this review. My complaints are ultimately minor compared to how enjoyable and enlightening the book is. The most serious of these complaints is that Parfit does not seem to show proper concern for all the bizarre implications that his theory of personhood would have. I would have liked to see more definitive ideas in this section, or an acknowledgement that the lack of any is a major problem. A more minor disappointment was that the question about whether your blackout drunk self is the same person as your sober self was never raised (I have been puzzling over this since I saw a Buzzfeed post where a person left a glass of vodka out while blackout drunk with a note saying it was water to prank his sober self the next morning. I cannot decide whether this works as a prank or not). I have omitted a lot of tangential content which was still interesting, and Parfit did a much better job explaining these concepts then I did. I am somewhat biased though, since I have a background in philosophy and Buddhism, I found Parfit’s arguments surprisingly easy to understand and accept, I imagine others might find him far less convincing or far more disturbing than I did. The idea of personhood being of extreme importance is probably very useful evolutionarily; it encourages activities like planning for the future and valuing your own life over others. It should not be surprising that it is difficult to discard, even if one accepts the idea that we should. I will end with the same quote from the Buddha that Parfit ends with:
“O Brethren, actions do exist, and also their consequences, but the person that acts does not. There is no one to cast away this set of elements and no one to assume a new set of them. There exists no Individual, it is only a conventional name given to a set of elements.”