With the rising trend of self-immolation in the East, particularly common in protests over the Chinese occupation in Tibet, I have chosen to examine from a philosophical perspective our views on whether suicide is morally permissible, and, within that, whether it can ever be rational.
For recent articles specifically on self-immolation in Tibet:
We have also recently seen self-immolation occur in both Syria and Lebanon:
Is suicide permissible?
Apart from those who endorse the “sanctity of life” argument, most people would agree that in some extreme situations, such as when a terminally ill person is in great pain, suicide is a permissible option. In this essay, however, I will argue that it is not just in these extreme cases where we can condone suicide, and that actually the permissibility of suicide is much more far-reaching than may be imagined on first glance. Throughout this essay, I would like to define suicide as “the intentional killing of self”.
My argument for the wide-reaching permissibility of suicide will be as follows:
(i) We should respect the rational choices of people, even if their choices are causing harm to themselves, providing that it doesn’t harm others to a significant extent.
(ii) Suicide can be a rational choice, in a surprisingly wide range of cases.
(iii) There are a lot of cases where suicide does not harm others to the extent that it should not be respected if rationally chosen.
(iv) There are a significant range of cases which satisfy both (ii) and (iii).
Conclusion: Therefore, in a significant range of cases, people’s choice of suicide should be respected.
I will now attempt a defence for each one of my premises.
Premise (i) is extremely similar to Mill’s Harm Principle, which is widely accepted as intuitively valid and is the basis for a lot of the things we condone in others even though we may not ever think they were a good idea for ourselves – such as others smoking habits, say. Our principle is, however, actually even more plausible than Mill’s, as the substantial difference is that it contains a premise of rationality – the lack of which causes Mill’s Harm Principle a lot of problem cases. Under our proposition, unlike Mill’s, we should still act paternalistically towards anyone who is mentally ill and is doing things that don’t necessarily harm others but are based on delusions or flawed thinking patterns. Our focus on rationality, then, actually makes our premise less controversial. A fuller defence of Mill’s Harm Principle can be found in Joel Feinberg’s “Harm to Self” as well as Mill’s own writings.
My second premise, then, will be entirely dependent on what I mean when I refer to a desire being “rational”. An interesting position for this topic is held by Devine, who proposes that rationality is based on a weighing of alternatives, and therefore contends that a decision not to live cannot be defined as “rational”, as we do not have suitable evidence about what will happen to us in the alternative scenario of death. Reasonably, he claims, we should have a natural desire towards self-preservation until something throws that desire into question. Death is surrounded by not so much ‘imperfect information’ but instead total ignorance, which therefore makes one’s death incomparable to any other either good or bad futures that one could imagine for oneself. It is this complete lack of comparability with alternatives that, to Devine, defeats any attempt to choose death rationally. If someone does choose death over life, then it is either from faith,of which they hold with enough conviction to believe that what happens after death will be better than their current situation,or as a massive gamble (analogous to the gamble made in Pascal’s Wager, where someone unsure about God and the afterlife weighs up their whole life situation based purely on supposed probabilities of possible outcomes and their ensuing consequences).A gamble to choose death over life could never be rational under Devine’s definition of rationality as a weighing of alternatives, as a person could not satisfactorily weigh up the alternative of non-existence, as they have no experience of it and cannot compare it analogously to the experiences they do have.
To Devine, then, suicide can never be rational because we have no idea what death implies. However, this definition of rationality seems a little counter-intuitive. For a start, to what extent can we really estimate what is going to happen in our future lives, anyway? We never really have sufficient evidence to weigh up alternatives to any substantial degree of accuracy. And, if Devine’s theory is dependent upon us weighing up all the options, yet we can’t weigh up the option of committing suicide, we therefore can in fact never weigh up all the options, and must sadly conclude that we can also never choose not to commit suicide. This is, clearly, not the conclusion that Devine would want to reach.
On the other hand, Brandt famously defends suicide, claiming a rational agent is someone who can make a sufficiently informed comparison between the likely utility of two possible futures, one in which he survived, and one in which he didn’t, and then make a choice that would be intelligibly rational – that is, comparable to any other decision we make which involves probabilities of possible future events occurring. A third perspective is outlined by Cowley, who proposes that suicide cannot be either rational or irrational. He claims that these terms actually make no sense when the suicidal victim has no future.
It seems that the definition of rationality is something that has a number of opposing convincing philosophical viewpoints behind it, and which one we choose will be the central feature of my argument. I will begin with the much less controversial claim of what exactly is irrational. In The Cambridge Dictionary of Philosophy, we find “To say of an action, belief or desire that it is irrational is to claim that it should always be avoided”. This seems pretty straight-forward. Under some theories of theoretical rationality, it is then claimed that any belief that is not irrational then can be defined as rational. Although this is in some ways an attractive definition, it is, however, quite far-reaching, and maybe too broad for a discussion on such a sensitive topic. Another way of positively describing rational beliefs or desires, then, is simply to suggest that the beliefs or desires simply have to be consistent with the overwhelming majority of one’s beliefs or desires. This definition of rationality still gives us wide scope but is not quite so all-encompassing, and it is this definition that I will be basing my essay on. For a desire or belief to be seen as rational, then, I will define it as being a comprehensive belief or desire which falls within a consistent and coherent belief system, and that allows for reasonable procedural thinking.
Within the topic of suicide, it is also crucial that we differentiate between fleeting emotional reactions and thought-out rational decisions, so I will also add in another constraint – that the belief or desire is maintained over time. My definition for rationality, then, has now become quite slim. What I am aiming to rule out, however, is temporary desires, relating to weakness of will, addiction or emotion. It may well be the case that a drug addict really, really wants to be cured and, when in a sensible or “rational” state of mind, genuinely believes freedom from addiction will highly improve their quality of life and general well-being. However, when going through withdrawal symptoms, and presented with the drug, they may well be overcome with a huge overriding desire for it. Satisfying this temporary physical desire for the drug, however, can be attributed to their addiction. We know it would not be conducive to their well-being, and, in a balanced state of mind, the addict would agree that taking the drug would definitely dramatically reduce their well-being in the long run, and that is not what they really desire. Similarly, someone on a diet who is ultimately aiming to lose weight may experience a craving for a particularly greasy burger, but we could attribute this to lack of self-control as opposed to them actually having any real desire for it or thinking that it would improve their well-being in any real way. Again, someone may lash out due to a strong rush of emotion, but, when calmer, reasonably regret the actions they undertook whilst they were experiencing the fit of rage. I want to count among irrational desires, then, temporary desireswhich can be attributed to addiction, weakness of will or emotion. In the case we are thinking about, namely suicide, it is of the highest importance to rule out actions which are taken when someone’s thought patterns are illogical. For instance, I am not intending to condone a case of suicide if the person involved has a mental illness like depression which is inducing particular thought patterns which we consider mentally flawed – and, importantly, curable. Another case which I do not want to condone is one where the person involved is acting emotionally in the spur of the moment – for example, a teenager impassioned by being rejected by their first love, or hormonally over-reacting to being grounded by their parents. Because it is so immensely important that these cases are not overlooked and accidentally found to be permissible, it is therefore of utmost significance that our argument contains a definition of rationality which does not allow for fleeting irrational reactions to situations.
As stated above, then, I freely acknowledge that not all suicidal behaviour is rational. In fact, a significant proportion of it is not. Some people really do have extremely debilitating mental illnesses which should be treated, not condoned. Other people have such strong emotional reactions to a life event, such as the death of a love one, or another loss (a job, a relationship, a reputation) to which they feel they cannot cope with, that they exhibit an extreme reaction.
There are a large range of circumstances, however, in which I believe suicide can be a rational choice. These situations range from ones involving, say, a terminally ill patient in great pain, which many people will intuitively accept that there can be a rational argument for, to a healthy, happy young individual with no apparent suffering in his life, which may be a significantly more startling conclusion. There are two arguments I would like to address to support my premise. The first is an argument from something I call the “Torture Thought Experiment”. My second argument will draw on historical, religious and cultural context.
The argument I would like to present as the “Torture Thought Experiment” looks at relatively uncontroversial cases in which we would accept that it would be a rational choice for someone to decide to commit suicide, and how it seems that there is no way we can draw an non-arbitrary line to say when someone’s decision that death is a better option for them than life becomes irrational or incorrect, and we must therefore accept that people must decide for themselves what life situation is unbearable enough to them such that death would be a preferable option. We can start by considering someone that knows they are going to be horrifically tortured for the next three days, and then killed. In this situation, if the person decided to take their own life before the torture began, instead of having to go through the torture and subsequent murder, it would be an understandable life choice that we could endorse. We can understand why a quick and comparatively painless suicide may be the preferable option for this person.
From here, we must look at what it is about the situation that makes us say that, under these circumstances, suicide would be a rational life choice. Is it the amount of time? If we endorsed three days, would we endorse four? What about a week? A month? A year? Ten years? Well, it seems that are reasons to prefer either option. Three days of torture may be preferable to ten years of torture, but, on the other hand, the person taking their own life before the torture begins is only knocking three days off their life, as opposed to ten years. It seems both of these options have their pros and cons. It doesn’t look like it is the amount of time that is involved in why we think this situation is permissible – either the amount of time that the person will be taking away from the life they had left, or the amount of time the torture would continue for.
Is it, then, the amount of torture faced? Well, firstly, that’s something that I didn’t explicitly specify in the original situation (any more than calling it “horrific”). But what do I mean by “horrific” torture? Many will presume that by this I am referring to physical torture, so this is the first thing I will consider. But what may be “horrific” to one person may be completely different to that which another considers horrific. It is common knowledge that people have different pain thresholds – a young child, for example, might feel excruciating pain at something that a hardened Royal Marine, say, would barely notice as an irritant. Whilst one person may claim they would not be able to withstand, for instance, the torture of all their fingers being systematically broken one by one, someone else may say that they could cope with that, but would not be able to put up with them being actually cut off one by one. But would the person with the higher pain threshold really be able to judge the other person for not being able to put up with as much pain? It seems obviously apparent that everyone has different physical pain thresholds, and it is therefore impossible to objectively judge what constitutes “enough pain” to respect someone’s preference for death, when everyone’s view on pain is so different.
We can also generalize this to emotional or mental pain. It might be equally, if not more, torturous for a mother to be forced to watch her own children physically tortured, or to have to harm them herself, than to go through the physical pain of the torture herself. Should she not equally, then, be allowed to choose suicide over this option, which for her is most horrific? Someone may be tortured by being raped every day, or degraded and humiliated. Should we tell this person that their mental anguish is not as potent as someone’s actual physical pain? Someone else may be forced to confront some of their deepest fears – for example, someone with a rat phobia being placed in a room full of rats – something not everyone would find traumatic, but that a particular person may find personally more horrifying than facing physical pain. Should we not also accept that people’s mental pain thresholds differ, and that they are equally as important to consider as people’s physical pain thresholds? In all these cases, we can sympathize with the victim’s pain, but we must accept different pain thresholds, both physical and mental. We can’t objectively specify an amount of pain which can be said to be too much pain.
Maybe what legitimizes this situation, then, is not the amount of time the torture the takes, or how physically or mentally painful we might objectively judge it to be, but the absolute certainty of it. In this situation we see a definitive end. We see no future positives. There are no future possibilities, because both the torture and the proceeding death are absolute certainties. In real life, however, nothing is known so definitively. Is this our problem?
Well, is it such a huge jump to suggest that we should also condone suicide at the knowledge of current torture enduring until you die, without being able to put a definite time on when that death would be? How about the current torture indefinitely enduring? Or what about the possibility of current torture continuing, or torture beginning in the future? Maybe it is possible that a percentage probability could be allocated so as to which if the chances of an event occurring were above a particular value one would still be rational in committing suicide. However, that might be 99% for some, 95% for others, or even 51% (more likely than not to continue). Again, we are back to the problem we were faced with before – it does not seem like we can objectively suggest what probability would be high enough to condone not taking the gamble.
To summarize, I believe these thought experiments show that we cannot suggest what amount of physical or mental pain is enough to condone suicide, or, equally, how high the probability of it actually happening has to be. There is no objective measure of what pain is too much to suffer through, or when a rational decision becomes an irrational one. We must allow for individual differences and for the fact that someone may not be able to tolerate a life that we judge to be tolerable. Because there is non-arbitary cut-off point, unless we want to bite the bullet and accept a random cut-off point, which is extremely unsatisfactory, we must either condone every suicide which conforms to our definition of rationality, or swing the opposite way and side with those who follow the sanctity of life argument, and say that suicide is never acceptable. I believe, however, that the very first scenario in the Torture Thought Experiment, alongside even less controversial cases like martyrdom, where, intuitively, we really do not want to condemn the agent for any kind of wrongdoing, immediately makes condoning suicides which conform to our definition of rationality jump out as a lot more accurate than condemning all suicide.
My second argument, then, comes from religious, historical or cultural context. Although we may, on first thought, believe that the vast majority of suicidal cases are irrational, we come at this by pre-supposing a set of beliefs that we hold and that are also held by a large proportion of people in the same cultural context as us are also held by the people who are committing suicide. There is a certain consistent and rational set of beliefs in our culture (which could include, say, that committing suicide is the cowardly way out, or that life should always be prolonged for as long as possible), which, given that set of beliefs, will indeed make suicide irrational, but these beliefs are contingent. Someone can hold another belief set with which suicide would not come through as irrational. Suicide is not intrinsically irrational.
Any religion where religious believers are able to have opinions on an extensive range of things without contradicting themselves, qualifies under my definition as “rational”. A religion like Jainism, however, endorses a practice known as “Sallekhana” – essentially a form of ritualised suicide (an extremely carefully structured fast to the death), as its ultimate aim. Strict Jains will give up their possessions, their family, their clothes and their home. Eventually, they believe, the next step is for them to also give up their body. As the body is viewed as just a shell in which their immortal soul resides, it is simply the next phase in their efforts to leave the constant cycle of re-incarnation. Their belief that a ritualised fast to the death is the only way to reach Nirvana is simply the next step to willingly give up everything that ties them down to the material world and so therefore fits in perfectly with their belief set, whilst it would not slot so easily into the belief set of many other religions. And this is not just a religious matter – across historical periods and ranging cultures we see a range of different attitudes towards suicide. Although our immediate thoughts and feelings about suicide may be necessarily negative because of the context we are used to hearing about it in and the culture we have been brought up in, it is not that suicide in itself is not irrational or intrinsically wrong, but the culture surrounding it that makes it so. Historically, if found condemned to death for committing a crime, Japanese samurais were allowed to commit suicide as an honour, instead of the humiliation of being killed by someone else. This represents a view of suicide as brave and correct as opposed to cowardly and wrong. Roman generals who had lost a battle fell on their swords as an honourable way of dying. In our current culture, however, we now condemn suicide as a cowardly or selfish act, and a suicidal person is often seen as someone to be pitied, which makes it much more difficult to see how a suicidal preference could cohere with a rational belief set.
Premise (iii), then, suggests that we can only condone suicide when the benefit to the individual’s well-being is significantly greater than the pain caused to other people by the act. I have already spoken about historical context and how it seems it may be highly dependent upon the society surrounding the suicidal person as to what moral bearing their actions are given and how upset those around them are ( Jains, for instance, have already given up their family connections). Secondly, and more importantly, we must not overlook an individual’s well-being for the sake of others around them. An individual should not have to completely sacrifice their own well-being for others well-being – that is simply too demanding. Even if there is some harm caused by someone’s decision to commit suicide, then, it would have to be shown to be significant harm to stop the individual pursuing his rational desire. A terminally ill man in excruciating pain, say, should not have to continue to undergo it just because his death will cause his wife sorrow, particularly as she will face the same sadness when he is dead a month later anyway. This is obviously an extreme case, but I am prepared to admit that situations should be looked at on a case-by-case basis. Maybe there is a person whose responsibilities and duties to others put them in such a place that it would be impermissible for them to commit suicide (for example, debatably, a parent). However, this is not much of an admission, seeing as I am already looking at all suicidal cases on a case-by-case basis, having already ruled out many circumstances such as mental illness and emotion.
Finally, then, on premise (iv), it has already become apparent that there is a surprisingly large range of cases in which both premises (ii) and (iii) are fulfilled – cases where the participant doesn’t have to be terminally ill or facing definite and obvious physical torture. Although it may seem a shocking conclusion, as with the Jainism case, a young, healthy and fundamentally happy young person may have a rational desire which slots in with a particular belief set and will therefore make their desire to commit suicide rational.
In conclusion, I believe that the argument set out above and my defence of it shows that there are a significantly larger number of cases than may initially be presumed in which suicide can be considered a rational decision, and therefore should be condoned.
 Mill, John Stuart – I9 in “On Liberty” (1859, Liberal Arts Press Inc), namely ‘The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant.’ (68).
 Devine, Philip – “On Choosing Death” in “The Ethics of Homicide” (1990, University of Notre Dame Press)
 Brandt, Richard – “The Morality and Rationality of Suicide” in “A Handbook For The Study of Suicide” (1975, Oxford University Press)
 Cowley, Christopher – “Suicide is Neither Rational nor Irrational” (2006, Ethical Theory and Moral Practice Journal) (http://link.springer.com/article/10.1007/s10677-006-9031-9/fulltext.html )
 Audi, Robert – “Rationality” in “The Cambridge Dictionary of Philosophy” (1995, Cambridge University Press, p674-675)
Audi, Robert – “Rationality” in “The Cambridge Dictionary of Philosophy” (1995, Cambridge University Press, p674-675)
Brandt, Richard – “The Morality and Rationality of Suicide” in “A Handbook For The Study of Suicide” (1975, Oxford University Press)
Cowley, Christopher – “Suicide is Neither Rational nor Irrational” (2006, Ethical Theory and Moral Practice Journal) (http://link.springer.com/article/10.1007/s10677-006-9031-9/fulltext.html )
Devine, Philip – “On Choosing Death” in “The Ethics of Homicide” (1990, University of Notre Dame Press)
Feinberg, Joel – “Harm to Self” (1986, Oxford University Press)
Griffin, James – “Well-Being: Its Meaning, Measurement and Moral Importance” (1988, Clarendon Press)
Mill, John Stuart – “On Liberty” (1859, Liberal Arts Press Inc)
Nozick, Robert – “Anarchy, State and Utopia” (1974, Basic Books)
Stanford Encyclopaedia of Philosophy – “Suicide”, “Autonomy in Moral and Political Philosophy” (links: http://plato.stanford.edu/entries/suicide/, http://plato.stanford.edu/entries/autonomy-moral/ )
I spent a month living in a Buddhist nunnery in Northern India this summer which gave me lots of opportunity to talk about a number of issues with practicing Jain and Buddhist monks and nuns (before I knew I was writing this essay), which may have been what gave rise to my original ideas on things like the differing views on suicide in Eastern cultures. Unfortunately they have requested to remain unnamed. I also benefitted greatly from suggestions and additions from my supervisor, Sebastian Nye (University of Cambridge).
Amery, Jean – “On Suicide: a Discourse on Voluntary Death” (1999, Indiana University Press)
Bennett, Jonathan – “Rationality: An Essay Towards An Analysis” (1964, Routledge and Kegan Paul Limited, London)
Bentham, Jeremy – “An Introduction to the Principles of Morals and Legislation” (1907, Oxford, Clarendon Press)
Blackburn, Simon – “Rationality” in “The Oxford Dictionary of Philosophy” (2008, Oxford University Press)
Bowker, John – “Jainism” and “Sallekhana” in “The Oxford Dictionary of World Religions” (1997, Oxford University Press)
Brown, Harold – “Rationality” (1988, London, Routledge)
Cherniak, Christopher – “Minimal Rationality” (1986, Massachusetts Institute of Technology)
Graber, Glenn – “The Rationality of Suicide” in “Suicide and Euthanasia: The Rights of Personhood” (1981, Knoxville, University of Tennessee Press)
Hassoun, Nicole – “Raz on the Right to Autonomy” (2011, European Journal of Philosophy)
Honderich, Ted – “Rationality” in “The Oxford Companion to Philosophy” (1995, Oxford University Press)
Hume, David – “On Suicide” in “Selected Essays” (1755, published 2005 Penguin Books Ltd)
Luper, Steven – “Philosophy of Death” (2009, Cambridge University Press)
Mill, John Stuart – “Utilitarianism” (1861, published 1998 Oxford University Press)
Moore, George Edward – “Principia Ethica” (1903, published 1993, Cambridge University Press) and “Ethics” (1912, published 2005, Oxford University Press)
Nagel, Thomas – “Death” in “Mortal Questions” (1979, Cambridge University Press)
Parfit, Derek – “Reasons and Motivations” (1997, Proceedings of the Aristotelian Society, Supplementary Volumes, Vol. 71) and “On What Matters” (2011, Oxford University Press)
Raz, Joseph – “The Morality of Freedom” (1986, Oxford University Press)
Rescher, Nicholas – “Rationality: A Philosophical Enquiry into the Nature and the Rationale of Reason” (1988, Clarendon Press)
Scanlon, Thomas – “What We Owe To Each Other” (1998, Harvard University Press)
Szasz, Thomas – “The Myth of Mental Illness” (1974, HarperCollins)
Wolff, Robert Paul – “In Defense of Anarchism” (1998, University of California Press)