Guest Blogger – Lizi Atkinson: Should we respect the desire of a deaf couple who wish their child to be deaf?

We already use genetic engineering to create things like crops that can resist changes in climate, and bacteria that can produce drugs. But in March last year, a group of leading geneticists called for a temporary prohibition on gene-editing in embryos. However, just a month later, researchers in China published the results of a controversial experiment – where they had modified the DNA of human embryos. Though the embryos they worked on were ‘damaged’ – that is, they could not have grown into living babies – it was a huge step for the scientific community. And, in the UK, a governmental body got ready to hand out licenses for creating a particular kind of genetically-engineered human – using a technique the US already tried and banned 13 years ago. It seems more and more likely that genetic enhancements to humans will become a reality – and with this new step in the UK, 2015 could well have been a historic year for the science.

Our decisions on what we feel is morally right with regards genetic enhancements in humans will affect generations to come. My friend Lizi Atkinson, a Philosophy graduate from the University of York, presents her essay contemplating whether or not we should respect the desire of a deaf couple to have a deaf child.

Genetic Engineering Baby

Do we have a reason to respect the desire of a deaf couple who wish their child to be deaf?

In this essay I will argue that we do have a reason to respect the desire of a deaf couple who wish their child to be deaf although this does not necessarily mean that we will agree with their action. The first step I will take in order to support my answer is to show that the initial intuitions that may make a person answer ‘no’ to the essay title are based on the individual model of disability, which I believe to be mistaken. I will then endeavour to explain Hull’s model of disability which I deem to be stronger and show why this is the case. With a new model of disability in place, reasons to respect the couple’s desire, for example inclusion in the deaf community, become clear and I will explicate these reasons. I will finish by explaining why although we can respect the couple’s desire to have a deaf baby, this is ultimately idealistic and as such we cannot agree with their action.

Background of the argument

The focus of this essay is based on a real-life case study; I will briefly explain this example in order to give a deeper understanding of the context of this essay. In 2002 a lesbian couple in America sparked controversy by seeking out a sperm donor with a family history of deafness to maximise the chances of having a deaf child. This decision was greeted by a cold reception, as the majority of people could not understand the couple’s motivation for deliberately trying to bring a disabled child into the world (Sparrow 2002: 11).

Initial intuitions about the issue

In order to delve deeper into the debate I must first show where initial negative responses to the question could come from. I must first emphasise that there are a number of different ideas about what ‘impairment’ is, for example Tremain who said that knowledge and power are the causes of impairment (2002: 34). Despite the importance of a discussion about impairment I do not have the scope in the essay to focus on these and will assume a biological account of impairment, which claims that impairment is a biological functional limitation. It could be argued that we do not have a reason to respect the desire of a deaf couple to wish their child to be deaf because the couple is intentionally bringing their child into the world disabled. The child in virtue of the disability will have fewer opportunities than children who are able to hear and this seems plainly unfair. The child will be disabled if born unable to hear and thus we have no reason to respect the deaf couple’s desire and their decision would be wrong.

The hostility to the couple’s decision assumes the correctness of the individual model of disability, this is the definition of The World Health Organisation (in Hull 1998: 207-8). According to this model the problems that disabled people face are caused primarily or solely from their impairment, disability is simply the consequence of impairment. It would seem therefore in the case of the deaf couple that by trying to maximise the chances of having a child with an impairment they are maximising the chances of having a disabled child.

Problems with the individual model of disability

I will now explain why the individual model of disability should not be employed and thus why the initial negative intuitions about the decision of the deaf couple are based on false assumptions. The individual model of disability fails to take into consideration social factors involved in disability. Although there are some functional limitations in all disabled people this does not necessarily that all impaired people are disabled. As Hull (1998: 200) argues functional limitation does not guarantee disadvantage, for instance using a wheelchair for movement would not be problematic if all the streets and buildings had step free access, thus a person in a wheelchair would not, in this case, be seen as disabled. Another example is people who have a mild functional limitation with their vision, these people are not seen as disabled because they have glasses which stop them from suffering because of their impairment, thus social factors are sometimes involved in instances of disability. The importance of social factors in disability clearly shows that the individual model of disability cannot be an adequate account of disability, and thus claims about the deaf couple based on this model of disability are based on false assumptions and are themselves false.

Now that I have shown that the initial intuitions about the deaf couple’s desires are irrelevant I will endeavour to show which view of disability I see to be the strongest and how applying this model of disability to the case study, which is the focus of this essay, can show we do have a reason to respect the desire of a deaf couple who wish their child to be deaf.

Further discussion in the definition of disability and a stronger model

On the other end of the spectrum to the individual model of disability is the social model; a movement of disabled people led by Oliver promoted this model (Terzi, 2004:141). It argues that disability is a purely social construct. Disability is a relation between an individual and her social environment, those viewed as being ‘disabled’ are deliberately excluded, and this exclusion comes from both people’s actions and the built environment. In contrast to the individual model this definition denies any causal role between impairment and disability, leaving the source of disability to oppression and segregation.

I do not find this model of disability to be adequate either, as it ignores the importance of impairment in disability, as Terzi stresses (2004: 150) it is important to have a functional limitation in a model of disability because otherwise any arbitrary cases of social exclusion could see somebody as deemed disabled. Terzi expanded her argument with a thought experiment that imagined how a person with a physical limitation would be seen in a world with no social barriers or restrictions, if they would face no difficulties then disability could be seen as socially constructed. Terzi explains how a blind person in this world would still struggle with understanding non-verbal communication, they would also face the disadvantages and problems that come from the fatigue and discomfort that blindness can cause. From this it appears obvious that impairment does have a place in some instances of disability.

A model, which involves both social, and biological factors in disability, is Hull’s model. This model says that we should not overlook the physical reality of functional limitation but that many of the disadvantages faced by disabled people are “social disadvantages of a remediable kind” (1998: 199). For Hull, to be disabled is to be disadvantaged, and thus unable to do something, this could be from either a social or physical limitation. If the disadvantage comes from social factors then society is to blame for the disability, but there are cases where the functional limitation causes the person to be unable to do something no matter the provisions put in place by society. I feel this model to be the strongest as it avoids the problems of both the social and individual models. Now that I have defended what I deem to be the strongest model of disability, I will apply this model to the case of the deaf parents to show that we do have a reason for respecting their desire to have a deaf child.

Applying this definition of disability to deafness

Although, as Hull argues, there are examples where not all disability is social I believe that in the example of deafness the disadvantage comes from society. Deaf people would not be disabled in the world of Terzi’s thought experiment; this can be partly demonstrated by the strength of the deaf community. The deaf community do not see themselves as disabled, rather as a minority group, they have their own language and culture and can survive perfectly well amongst each other. (Lawson, 1981: 31). Another example demonstrating the social aspect in deafness is the story of Martha’s vineyard, a real place where a genetic mutation meant that a high percentage of the population were born deaf. The amount of deaf people on the vineyard inspired the residents to develop their own sign language that was spoken by both hearing and deaf inhabitants. Deafness was never considered to be a handicap, and the residents lived happily and safely (Groce 1985: 1-5). It seems therefore from these factors that deafness is a socially constructed disability.

So, applying Hull’s disadvantage criteria to deaf people shows that they are disabled (as they are disadvantaged), but this disadvantage primarily comes from society, we have not put in place enough measures to cater for deaf people, and if we had then they would not longer be deemed disabled.

Why do we have a reason to respect the couples desire to have a deaf child?

Now that we can see that deafness is a disability primarily because of society and that the disability does not spring merely from functional limitation, I will lay out what I see to be some convincing reasons why we should respect the desire of a deaf couple to maximise the chances of having a deaf child.

As I mentioned earlier, there is a very strong deaf community, which deaf people see as a minority culture. The couple could be seen as wanting their child to be able to get the most out of this culture which they deem to offer the most valuable opportunities for their child. The deaf community have a language, so the child is not missing out on communication, thus as Sparrow argues (2002: 14) the couple see themselves as giving their child the best opportunities by enabling them to be part of the deaf community. We do not judge parents who wish for their child to be part of other minority cultures, for instance religion, so maybe we do have a reason to respect the desire of the couple to have a deaf child.

In response to this claim it could be argued that the child could be part of the deaf community without actually having the functional limitation, why subject the child to not being able to hear when they could benefit from both cultures?

It appears this is not as simple as it seems, although in some cases a hearing child could be included, the deaf community can be hostile to those who can hear, and there are conditions that must be met in order to be a member of the community (Higgins, 1991: 24-9). These are identification, shared experience, participation and membership. A hearing child with deaf parents may not be able to identify with the deaf community, they may struggle in sharing the full experience as they do not share the impairment. Therefore, a hearing child in the deaf community would not be wholly involved and accepted. The child may feel alienated from both the deaf and hearing cultures.

If society banished all barriers, by learning sign language, and making the world a more accessible place for deaf people then a deaf child would not be seen as disabled.

From this point of view it can seem we do have a reason to respect the desire of a pair of deaf parents who want a deaf child, they see their community as providing the best life and opportunities for their child, the child is not disabled within the deaf community. As Sparrow argued (2002: 15) the parents may feel they can be more effective as parents to a deaf baby. A hearing child growing up in a deaf family may face problems with communication, and be limited in their ability to communicate orally, putting them at a disadvantage at school. All parents want the most for their baby and this requires participation from parents. If they do not see their impairment as a disability, and feel they could raise their child better if it was deaf then it appears we do have a reason to respect and empathise with their desire.

What next?

Although the existence of the deaf community and Hull’s definitions of disability give us reason to respect the desire of a deaf couple who wish their child to be deaf, I do not believe this implies that this course of action should be taken.

I see that although the reasons for respecting the desire are correct, they are too idealistic. The child is seen as disabled because of society and although this doesn’t seem fair the way things are now aren’t going to dramatically change overnight. The child may have a wealth of opportunities from within the deaf community but at certain points in their life they will have to venture outside of this setting, and at these points the fact that the child cannot hear will be problematic.

I understand the couple’s desire to have a deaf baby, and in the world of Terzi’s thought experiment I would agree with the course of action, as it could be the same as choosing to teach your child to live in a religious community. Despite this, the fact is that if the child decides to leave the religious group they won’t find themselves all of a sudden disabled, whereas a deaf child leaving the deaf community would be at a loss. In the unfortunate event of something happening to the parents the deaf baby may be taken away from the deaf community, and really struggle with disability.

We can respect the desire of parents to let their children play in the park unsupervised but this doesn’t mean that the child will be perfectly safe, there exist people who could put the child in danger. All parents must make decisions about how to raise their child in response to the world we live in, the deaf couple must see that it would not be maximally beneficial to deliberately try to bring a deaf child into the world.



I have therefore shown that although we do have a reason to respect the desire of a deaf couple to want a deaf baby this does not mean that we can agree with the action of trying to find a deaf donor in order to maximise the chances of fulfilling this desire.

On Hull’s view of disability the child would be disabled because of social factors, and thus initial intuitions about the wrongness can be overcome. Despite this, and the existence of the deaf community, a child born deaf will struggle to live a maximally beneficial within the current society, therefore I cannot agree with the decision to try to maximise the chances of having a deaf baby.



The Frustrating Fallacy of the Wasted Vote

Please note: this piece was written as a political opinion piece for a light-hearted left-wing student newspaper, so it has a more light-hearted, informal and humorous tone than the other work I display on here, which is more philosophically rigorous and academic in nature.

The Golden Age in engagement with politics comes round like clockwork every five years – it is those all-encompassing 2-3 weeks before the general election. At this time and this time only, hordes of people who for the past few years have thoughtlessly flicked past the Politics sections in their newspaper of choice, hastily changed the channel whenever a boring politician has come on the news, and quite frankly ignored every change in policy that wasn’t linked to their tax code all suddenly, as one, develop unshakeable political opinions apparently linked to their very identity. They swot up on everything evil the enemy party have done, quote endless facts and figures as if they’ve known them for years, relentlessly argue with people who they normally find most pleasant about major political differences that simply haven’t been brought up before and develop an insatiable appetite for heartfelt Facebook statuses demonizing those who voted differently.

And then it is over. #GE2015 stops trending; your interest in it has waned. And, in the week or so after the election, as the right-wing media stops shouting in our ears about Ed Miliband’s bacon sandwich, many electoral fallacies are retrospectively cleaned up. It turns out the world’s leading economists are as settled on austerity being a bad thing as leading scientists are on climate change being a bad thing[1][2][3][4][5]. It turns out the Tories have maybe not been so successful over the last five years as they were claiming[6]. It turns out David Cameron wants to repeal the Human Rights Act[7][8]. Sure, there were plenty of people who knew all this stuff before the election, and voted because of it or in spite of it, but many people didn’t. There have been brilliant articles written and shared which have genuinely served to make some people think again and – I know even in some of my friend’s cases – regret the party they voted for.

But there has been one fallacy left largely unchallenged – a fallacy that seems to be as common amongst lefty discourse as it is amongst righty discourse. It surrounds the mysterious phenomenon of tactical voting. If you read the title of this before you clicked on it, you may well have correctly guessed that I have chosen to call this a remarkably snappy name if I say so myself: it is The Frustrating Fallacy of the Wasted Vote.

general election.jpg

The Frustrating Fallacy of the Wasted Vote

I voted Green on the 7th May 2015. The Greens have policies on immigration and housing and CrossRail and taxation and the foreign aid budget and education and the NHS; policies that some people agree with and some people disagree with. Yet the most common response I have had when I share my voting preference with friends, family and colleagues is the same whether their views are remarkably liberal or quite terrifyingly right-wing: “Oh – but that’s a wasted vote.”

A wasted vote.

Let’s just clear up what people mean when they say “a wasted vote”. The reason people identify a Green vote (or any other minority party vote) as a “wasted vote” is because the party in question is extremely unlikely to win the seat. In my constituency of Twickenham, some of my more left-leaning friends voted Lib Dem. This was a “tactical vote” to keep out the Conservatives. Labour would be “a wasted vote”, as they had little to no chance of winning in the borough. It’s pretty simple terminology, and we’re used to people using it all the time. And, somehow, when talking about the election, it is terminology that we seem to accept it as a given.

So, I think what people mean when they say “a wasted vote” is that it is a vote that doesn’t make a difference to who wins a seat in that constituency.

Hang on a minute, though. No single vote ever makes a difference to who wins a seat. Since 1945 (post-war politics) there has never been a constituency where it has been down to one vote.

In fact, in the entirety of UK history, there have been just two cases of constituencies where just one vote was relevant. These were in 1886 and 1910 – and on turnouts of 6,099 and 9,553 respectively. Now, an average constituency receives more than 70,000 votes. And, these outcomes were then ultimately decided undemocratically anyway – in the 1886 case it was down to the Returning Officer to decide who was the winner, and in the 1910 case, the courts selected a winner.

The chances of there being a single vote in it are, even in a particularly close constituency, still astronomically small. Estimates on just how small vary quite dramatically, but at the most generous you’re looking at about a 1/10,000 chance of your known-to-be-close constituency actually being a 1-vote-in-it constituency. In other constituencies we’re talking of chances at worse odds than winning the lottery. (An interesting consequence of this is that it leads to the old adage “if you didn’t vote, you can’t complain about politics!” being comparable to telling a homeless person “if you don’t play the lottery, you can’t complain that you’re poor!”)

So, whether you vote for the winner, the runner-up or the unheard of independent candidate, your vote is a ‘wasted vote’ by this definition – as it doesn’t make a difference to who wins in the constituency. (And, of course, this is all without even getting started on how little difference that makes to who actually forms a government and which policies actually get passed).

Of course, if you can get thousands of people to change their vote, then tactical voting does work. But let’s be realistic here. In my “very close” seat, the Conservatives knocked out Vince Cable by 2,017 votes. For me to have persuaded 2,017 people – about double the number of Facebook friends I have – to vote for the Lib Dems would be unlikely. For none of them to have been voting for the Lib Dems anyway would have been even more unlikely. And for all those people to all have lived in the same constituency would be even more unlikely still!

But let me humour you for a moment. Let’s imagine I was famous, and people knew who I was, and they really respected me and listened to me, and I didn’t have a job but I did have loads of money so I spent all my time campaigning, and I had brilliant resources, and due to the most amazing PR strategy the world had ever seen I had personally managed to persuade 2017 would-be Labour voters in the Twickenham constituency to vote Lib Dem (or – even better – only a measly 1009 would-be Conservatives, as their vote would be a -1 to the Conservatives as well as a +1 to the Lib Dem vote, so we’d have only needed half the amount). I’m sure we can all accept this is an extremely unlikely scenario. But – and here’s the thing – even if this was the case, it would still make no logical sense for me as a singular voter to vote tactically when I was actually in the booth – even if I believed I had persuaded thousands others. Because, without any actual inside knowledge of the voting numbers, the chances of the result coming down to 1 vote would still be so slim. So, it is still illogical for you, personally, to vote tactically.

So, just to confirm, whether you voted for the Cannabis Is Safer Than Alcohol party, or whether you voted for the Conservatives, and whether you voted in the close constituency of Thurrock or in the landslide constituency of Sevenoaks, your vote was just as wasted as mine.

So, I guess it now looks like I’m going down to head down the hard-line behavioural economics line[9], which says that, logically, because the probability of your vote making a difference is so amazingly slim, it is in fact never worth your time to even bother turning up to vote – even if it’s only five minutes to register online and five minutes’ walk to your nearest polling booth.

Well, no, not exactly. Although it is a position I sympathize with, for now I will favour a slightly less dramatic but no less important conclusion.

What I’m saying is, to think a vote counts if it goes towards the winner – or towards a candidate that doesn’t win but had a good chance – isn’t a wasted vote, but that a vote for a party that were never going to win is a wasted vote, is a completely erroneous judgement.

Maybe a more positive note to finish on. Here are some positive reasons that votes can make a difference, and how voting for a minority party can help.

Firstly, voting for a minority party really helps the push for electoral reform. The more people vote for minority parties, the more outdated and ridiculous the FPTP system looks. (Note: if you don’t want electoral reform, you won’t see this as a positive point. Obviously.)

Secondly, voting for a minority party can encourage the major parties to change their policies. Parties create policies that they think they will win votes on. 1,157,613 people who may well have otherwise voted Labour chose to vote Green on 7th May. Hopefully, this will encourage the Labour party to make their policies more environmentally friendly, for instance – in order to entice those naughty hippie outliers back into voting Labour. Similarly, the large UKIP vote will encourage the Conservatives to take a tougher stance on immigration and the EU than they may have otherwise. Voting for a minority party is a way of making your dissent with the major parties known, but, alternatively to a spoiled ballot paper, it’s clearer why you’re dissenting and what you want instead.

Of course, I accept on both of these points one could well logically argue that your vote is wasted too: 1,157,613 voting Green will make no actual difference to whether Labour decides to change a particular policy compared to 1,157,612 voters. Essentially, this is also true, but only in the same way it is with many things that are achieved collectively (for instance, it is highly likely that a demonstration of 3 million people wouldn’t have a different outcome depending on whether you came or not, but generally you would attend a demonstration not because you believed you personally would be the agent of change, but because you believed in something so strongly that you wanted to be part of the collective that did bring change). Also, straw that broke the camel’s back, and all that.

In conclusion, I didn’t write this article to advise you not to bother voting. I wrote it to advise you not to bother voting for someone that you don’t agree with under the irrational banner of ‘tactical voting’. If you are going to bother turning up, then you may as well vote for someone you actually agree with. The lesser of two evils is still evil.

Or, how about this. Seeing as it doesn’t really matter who you vote for anyway (unless this article suddenly goes viral in a particular constituency in #GE2020 – which, despite being quite personally ambitious for the reach of my work, I highly doubt) just stop propagating the fib that a vote for a party that got 1 MP or less was any more wasted than yours was.










Is a Consequentialist Theory of Punishment Plausible?

In the last few months, Secretary for Justice Michael Gove has announced various plans for prison reforms. He advocates various policies to shorten sentences and deport foreign prisoners in order to reduce the overall number of inmates. Alongside this, he plans to give more power to prison governors, and put a much greater focus on the education of prisoners. This looks to be the biggest reform of the penal system in the UK in a generation.

Of course, alongside this, it goes without mentioning there are numerous stories in the news every day of people facing or appealing various punishments for various different crimes. In all of these cases, we may often hear others expressing views that punishments are too harsh, or too lenient. But what are they judging this on?

Quite a few people see the worth or the suitability of a particular punishment by the supposed consequences of it – normally that it reduces and prevents crime. The idea that the rightness or wrongness of something is determined by its consequences forms the basis of a moral theory called ‘consequentialism’. I’m going to look at whether there is a plausible consequentialist theory for punishment.

Prison reform

Is there a plausible consequentialist theory of punishment?

Punishment is a phenomenon that is a social norm in every civilized society that we see today, and indeed, it is the view of all but a few extreme thinkers that societies would not be able to function without it (Skinner, for example, thought that punishment was not a necessary pursuit, and only positive reinforcement should be encouraged). Although it is possible to criticize the legitimacy or appropriateness of various punitive acts – many are no doubt excessive, brutal and undeserved – the practice of punishment is widely accepted to be clearly justified by the norms of a liberal, constitutional democracy (although this is not a fool-proof argument – it wasn’t so long ago that slavery was the norm in liberal, constitutional democracies around the world). Not only is punishment something imposed by law, and the governments that control society on a large scale, but also on a smaller, more personal basis – something used by teachers in school, parents on their children – all the way down to petty punishments amongst friends and acquaintances, who may with-hold their friendship or act a bit colder towards someone, to “teach them a lesson”. Everywhere we look we see countless examples of punishment in practice. But what is punishment exactly?

Before we define punishment, I must point out an important distinction that can sometimes be forgotten – that the definition of punishment is an extremely different notion from the justification of punishment. There is such a thing as an unjust punishment.

With that clarified, the Stanford Encyclopaedia of Philosophy defines punishment as having seven qualities. The first of these is that it is an authorized act, not an incidental or accidental harm. Secondly, it is constituted by imposing some burden or form of deprivation or by withholding some benefit, which could also be seen as a deprivation of ‘rights’ (the definition of rights is an essay of its own). Thirdly, punishment is a human institution. If a wrong-doer breaks his leg, this is not punishment, as it is not afflicted by human agency. Fourthly, punishment is only imposed on people who are believed to have acted wrongly. Fifthly, no single explicit purpose or aim is built by definition into the practice of punishment. Nietzsche theorized that punishment is consistent with several functions or purposes, yet not consistent with having no function or purpose. Sixthly, not all socially authorized deprivations are punishment – only those which are imposed in the consequence of finding criminal guilt (e.g. a fine, as oppose to a rise income tax). And, lastly, ‘authorized deprivation’ must be taken to include parental or other forms of punishment, not just punishment by law.

The two main theories on punishment are known as “consequentialism” and “retributivism”. This is, fundamentally, a deontological vs. teleological debate – is punishment good, or necessary, within itself – a practice required by justice? Or is the point of punishment simply to increase the overall net social welfare by reducing and ideally preventing crime?

In this essay, I will be assessing the arguments for and against a consequentialist theory of punishment. Duff describes the main principle which consequentialism rests upon as being “a practice is justified if and only if its consequences are at least as good as any alternative.” So, the practice of punishment achieves whatever end-state justifies it, which consequentialists may believe to be public interest, general welfare, or the common good. In this case, we may see punishment being a necessity in crime prevention, or that private retaliation may be pre-empted by confidence in punishment by law, or even simply to protect and vindicate rights of individuals. It may even be useful in itself as a technique of social control.

There are some major downsides to a consequentialist theory of punishment. Firstly, there are the complications of making consequentialist calculations. Taking into consideration all the variables, and therefore working out which overall outcome and therefore which punishment is most preferable, is currently impossible. There are simply too many variables, and too many unknowns, to ever know how a situation will turn out, not mind judge from that which outcome would be the best. A situation with an immediately positive outcome might later turn hugely negative – for example, people might rejoice and feel safe again if they thought the rapist and murderer terrorizing their street had been imprisoned, but when they found out he was in an innocent man, they would be shocked and outraged, and lose all faith in the government – they may not be so quick to rejoice if the real culprit really was caught. Therefore it may be said that a consequentialist theory is simply too ambitious to apply to punishment.

The unjust punishment of an innocent person as a scapegoat is also a problem for consequentialists. Hard consequentialists would have to concede that, if the overall good was an increased, an innocent man should be punished for a crime that he did not commit. This seems extremely counter-intuitive. However, the consequentialists have something on their side here – that which was their flaw in my last point. The complexity of consequentialism means that a consequentialist could say ‘in theory, you should kill the innocent person, but that would never happen in real life.’ As there are no certainties in how a real-life situation will turn out, it is always better not to risk the punishment of an innocent, because of the horrendous impacts it would have upon society if the truth was revealed – and there is never any complete inevitability that it won’t be.

On a similar note, it appears that consequentialism also justifies the excessively harsh punishment of the guilty, if this is what the majority of society calls for. This could happen in the case of a well-known criminal receiving a lot of media attention and therefore a particular government being swayed by public opinion in punishing the accused more harshly than they deserved. This may also happen if the government is trying to deter others from acting in the same way, or trying to send out a particular message in society. For example, in the 2011 riots around England, people were punished for minor crimes like petty theft, or minor cases of vandalism, with punishments that did not fit the crime, and were even as extreme as imprisonment. This was not due to the nature of the individual crimes, but instead the message of absolute intolerance that the government in power wanted to portray.

There is a semantic point to be made here, that we can call a punishment “unjust” even if it is consequentially beneficial – for example, scapegoating an innocent man. If to justify a punishment is for it to cause a primarily positive outcome, then this statement would be a contradiction in itself. We may use this as an argument to suggest that a solely consequentialist theory of punishment may not stand up – it may be necessary, but not sufficient.

There are also some obvious benefits to consequentialism. Firstly – the reduction of crime. This would be the ultimate aim of punishment, to a consequentialist, so any system that worked best for this would be acceptable. Punishment can reduce crime by incapacitating, reforming, or deterring potential offenders. However, the success of punishment is nearly impossible to test, as it all works upon theories, such as who would probably have committed a crime if they hadn’t been punished – there are no certainties. Also, if we do take evidence to be convincing, it doesn’t look like punishment really does serve as an adequate deterrent. For example, the murder rate is just as high in countries where capital punishment is practiced as where it isn’t – there is no evidence to suggest it is a deterrent at all. And there are no examples where everyone who doesn’t murder is rewarded, as opposed to everyone who did being punished (this would be positive reinforcement taken to the extreme). For all we know (as unlikely as it seems) this system might work equally well, if not better. Aside from deterring crime, there is another issue consequentialists face with the aims of ‘incapacitation’ and ‘reformation’. Would it not be a more effective method of crime prevention, if that is the ultimate aim, to punish those who are judged most likely to commit crimes if left free – whether this is due to their characteristics, background or past behaviour? Yet this seems to go against everything that we feel is ethical about punishment. It appears that we think the right of the innocent not to be punished for crimes they have not committed is an intrinsic and fundamental demand of justice.

To conclude, I do not think a consequentialist theory of punishment can be complete in itself. It appears that there is no plausible consequentialist theory of punishment that does not have a flaw somewhere within it, or that goes in the opposite direction to our instincts. I do not, however, think consequentialist theories and retributivist theories should stand alone and opposing, and in the case of punishment, I think there is something to be said for both of them, and the best theory of punishment would accept that both are necessary, but not sufficient alone.

The Moral Importance of Intention

The one story that has consistently dominated the headlines for the past few weeks has been that of the trial of Oscar Pistorius, the famous Paralympic athlete accused of murdering his girlfriend, Reeva Steenkamp. There is no doubt that the bullet that killed Steenkamp was fired by Pistorius. The outcome of the trial will rely instead on Pistorius’ intention – did he intend to kill/injure Steenkamp or not?

Within philosophy, the criminal justice system’s focus on intention brings up a number of interesting questions. Some philosophers argue that intention has no real moral relevance at all. In this essay, I want to defend what I find to be the imperative moral importance of intention, and look at the additional clauses we will need in order to produce a full and complete theory of intention which is not tripped up by various problem cases.


The Moral Importance of Intention

The question of whether intention is of any real moral relevance is one that has yielded a wide range of philosophical views in recent years. In this essay I will argue that, on the basis of intuitions (having made a preliminary methodological assumption that moral intuitions are our best evidence if we want to find a satisfactory moral theory for something that is sensible, practical and fits in with what we really do in the world) that intention plays a crucial part in what moral weighting we give to an agent’s actions, and that we can look to The Doctrine of Double Effect (DDE) to distinguish between the rightness and wrongness of cases with regards to good moral agents. I believe a complete theory of DDE accurately represents all of our intuitions about a situation, and can therefore distinguish between the permissibility and impermissibility of a certain action. My essay will look into criticism of intention as a whole –  that we can never actually know what someone’s intentions are – and also into problem cases for DDE, and whether DDE simply re-describes a moral situation in a different way.

Before I begin, I will define what I mean by “intention”. We normally use intention in one of two ways – firstly, an agent-based notion, such as “I didn’t intend to hurt anyone”, and secondly, an action-based concept, such as “I intend to go into town today”. This essay is solely talking about the first, agent-based notion of intention. In “Actions, Reasons and Causes” Donald Davidson says “Giving the reason why an agent did something is often a matter of naming the pro attitude (a) or the  related belief (b) or both; let me call this pair the primary reason why the agent performed the action”[1]. Although Davidson himself does not explicitly use this is a definition for intention, I would like to. Intention, therefore, does not refer to an event, or state of the agent, but instead to a way of re-describing what an agent is doing in terms of what can be referred to as a “primary reason”. This primary reasonfor an action A involves a desire (or, as Davidson says, a pro-attitude) towards some goal G, and an instrumental belief that performing action A is a means to attaining G. For example, someone’s primary reason for wearing a coat on a cold day might be that they want to stay warm, which they hold alongside the belief that wearing a coat is a means to stay warm that day. It is this notion of intention that I will be talking about in this essay. Other reasons the person may hold for wearing the coat may include things like that it goes with the rest of their outfit, or that they haven’t worn it much yet, but these reasons are secondary reasons insofar as they would not constitute any action being taken if they did not come alongside the primary reason.

It is immediately obvious to us, then, that in day-to-day life, an agent’s intention is carefully considered when we decide whether to attribute praise or blame, giving us morally-weighted terminology like “it was an accident”, or “he did it deliberately”. Although some may argue that someone’s intention is morally insignificant, and it can be avoided when looking at whether an action is permissible or impermissible, in the face of something else more concrete (most commonly, perhaps, the actual consequences of the action they have undertaken), this does not seem like an intuitive suggestion.Intention is highly important – it would be a harsh judge who ruled that the same amount of blame should be attributed to a person who accidentally started a fire through carelessness or oversight, which then eventually led to the death of two children, and a callous, deliberate arsonist who burnt the house down with the ultimate end goal of burning the children alive. Treating situations like these, so obviously completely different, as the same, goes against all our moral intuitions. Having made the assumption that we want to find a moral theory that fits in with our moral intuitions, then, it seems like we can rule out the position of the extreme sceptic, who suggests that intention has no relevance to the way we understand and interpret morality. But how much weight we attribute to it?

Well, things are not so simple. There are many cases where an agent’s action produces multiple outcomes, where, if we are to judge morality on someone’s intention, we need to carefully consider whether their intention was for the good or the bad consequences. If it is to be the case that we attribute so much moral weight to intention, we must look more carefully at problem cases and distinguish what it is for someone to intend something (and not merely foresee it). Therefore, the theory that I will be concentrating on in this essay is DDE. I believe DDE can help us make sense of all the problems we have with intention.

DDE is often invoked in order to condone an action which causes harm as a side effect of promoting a general good. There is an essential difference within DDE which captures the permissibility (all other things considered) of between bringing about some harm as a side effect of a good, and the impermissibility of bringing about the same harm as a means to the good. DDE is, then, specifically focussed on the difference between intention and foresight. A classic case used to explain how there can be such a significant difference between these two fundamentally similar things is the case of the doctor and the pregnant woman. The woman has been in a dreadful accident, and, to save her life, the doctor must perform an operation which removes a part of the woman’s womb, which will undoubtedly kill the unborn child. If he fails to act, both the mother and unborn child will perish of natural causes. Therefore, the doctor goes ahead with the operation, despite knowing that the unborn child will not survive it. The death of the foetus, in this situation, is a foreseen but not intended side-effectof the doctor’s actions. This is an entirely different scenario to one in which the doctor is performing an abortion, and the removal of the foetus is the intended consequence of the doctor’s actions. Consequentially similar acts, then, can have different intentional structures and therefore should be treated ethically differently. DDE, if found to be correct, has extremely wide-reaching applications in condoning controversial actions in a range of cases including military and medical ethics.

Thomas Aquinas is credited with being the first philosopher to introduce this distinction in his work ‘Summa Theologica’, as a justification of homicidal self-defence[2]. His criteria for a harmful action being morally justifiable as a foreseen side effect incorporated that (a) the nature of the act itself is good (or at the very least morally neutral), (b) the agent does not intend the bad effect, either as an ends in itself, or as a means to the good (c) the good effect outweighs the bad effect to the extent where causing the bad effect is justifiable (now known as the proportionality condition) and (d) the agent attempts to minimize the harm caused by the bad effect. It is important to note that these are all vital components of a complete theory of DDE, and that it is not the intention/foresight distinction alone.

I have chosen my essay to focus on DDE as I believe it accurately captures all our intuitions about intentions, and, although later I will look at “problem cases”, I consider there to be reasonable explanations as to why these do not create the problems they claim to for DDE. DDE also captures what we recognize as important about intention a lot better than the alternate explanations – of which the most commonly cited is the Doing/Allowing Distinction (DDA).

DDA suggests that what makes situations morally different is whether we are actively doing harm, or merely allowing harm to happen. However, although DDA is an important philosophical difference and is not something that should be overlooked, it unfortunately cannot do all the moral work that DDE does. For example, within our first case of the doctor saving the mother’s life but unintentionally killing the foetus, the doctor is actively removing a part of the woman’s womb, so it cannot be said that he is simply allowing the foetus to perish. However, we want to condone his actions. Similarly, in problem cases which I will go into in more depth on later, such as the famous Trolley Problem, DDE is simply a more accurate representation of our intuitions. In the original Trolley Problem case, where we press the lever to change the track, and therefore kill one instead of five, the active action is made in changing the track – we do not just allow the trolley to continue on its path and kill the five. It seems that DDA does not quite pinpoint what exactly our relevant concerns about the case are.

Another reason that I am determined to focus on DDE instead of DDA rests on my previous discussion that our intuitions insist that intention must be morally relevant. It may even be the case that the best way to understand “doing something” is that we are intentionally doing something. Despite all the interesting things DDA has to say, we cannot overlook the issue intention – and the problems within it are best captured by DDE. These two distinctions are both philosophically informative, but DDE is ultimately better at unpicking moral cases.

It seems clear, then, that there is an easily intelligible case for DDE. But there are also two key worries surrounding it. The first is an epistemic worry – how can we ever really know what someone’s intentions are? The second worry is that under DDE a seemingly morally insignificant change to a situation, or a trivial re-description of an agent’s intentions, can seem to make the difference between permissibility and impermissibility.Looking first at our epistemic concern, then – how would we have known if that doctor really did intend the unborn child to not survive?

Well, whilst allowing for a few exceptions, I would like to claim that, in the majority of cases, we can make at least a very educated guess as to what the person’s intention would be. In this particular situation, we have no reason to sensibly believe that the doctor would have a desire to cause the unborn baby to die. However, we can clearly see why he would want to save the woman’s life – even if we took it to be for completely egoistic reasons, such as his medical prowess bringing him respect amongst his work colleagues and possibly furthering his career. Of course, there may be a hypothetical case where a doctor does want to kill an unborn baby – having, for example, some vendetta against the family. We are presuming, however, that when making a judgement on an agent’s intention, we have the facts available to us such that we can know whether or not this is the case.

Although we may not be able to satisfactorily make the leap from this case to saying that we can always definitely tell for sure what someone’s intentions are, judging someone’s intention is still going to be much more difficult than other, alternate theories. The main form of resistance to DDE comes from act-consequentialists, yet it seems that act-consequentialists are in just as bad, if not a worse situation with regards to epistemic worries than proponents of intention. With time being infinite, the consequences of every action an agent takes spans infinitely onwards in time, with each event their action has caused in turn causing another. Judging the entire state of the world and how it would have differed from what would have been if an agent did not perform that particular action therefore makes act-consequentialism comparatively infinitely more difficult to judge than an agent’s intention, which is only a primary reason at a certain time, ‘t’. In both cases, it is plausible that our objective judgement may not always be infallible, but that does not mean that the theory itself is intrinsically flawed.

Let us then move on to our second worry. Firstly, what if by simply re-describing an agent’s intentions we can legitimize an action we wouldn’t otherwise condone? A famous example of a case in which DDE appears to fall down in this way is that of a terrorist bomber[3]. Obviously, we may say, a terrorist bomber whose intention is to kill hundreds of innocent civilians is wrong – his intention is evil (this is in contrast to the case of the allied bomber, whose actions we may condone, as his intention is to end the war). The suggestion, however, is that the terrorist’s intention may not be for the innocent civilians to die, but instead, for them to appear dead, so as to instil the fear and panic which will ultimately further the terrorist’s desires. Unfortunately for the bomber, he has judged that the only way in which these civilians will convincingly appear dead is if they actually are dead. Their deaths, therefore, are a foreseen but unintended side effect of them actually appearing dead.

Understandably, this case is extremely worrying for many philosophers. Surely it should not be possible for the permissibility or impermissibility of an action to be able to change simply because you have been able to tweak your intentions in a particular way?

Well, although this conclusion would be extremely worrying, this re-description could not stand alone – it would have to come with further evidence. If the bomber truly did not intend for the civilians to die, there should surely be other evidence to back this up. For example, we can ask whether or not they looked for other solutions to achieve their aim? Did they feel any sadness and regret about this terribly unfortunately unintended side effect?

This case also does not fulfil all the conditions for DDE. We must remember that DDE is not solely concerned with the intention/foresight distinction, but there are also other important conditions within it, which make this situation not as problematic as it may first appear. For a start, it could be claimed that the bomber is, in fact, still using the civilian’s deaths as a means to an end,something that is strictly ruled out by DDE. Another consideration which this situation does not seem to fulfil, at least until we have more information, is that of the proportionality condition, which, after the huge amount of harm done (lots of innocent people’s deaths), will have to be filled with quite a large humanitarian gain if the terrorist achieves his aims (even more innocent lives saved.) However, especially with the use of the emotionally loaded word “terrorist” we instantly do not consider that the bomber’s aim will be to save innocent lives – instead, we imagine him wanting to further some aims that are ultimately bad in themselves – especially as this is followed by us not having enough information to see the bomber’s intention as anything but “causing fear and panic”. We need to know more about the whole situation in order to determine whether the terrorist fulfils the proportionality condition, whether he tried to minimize the harm done, and whether he felt any regret and sadness about this unfortunate side effect, before we can judge whether or not the situation could be permissible.

In the terrorist bomber case, nothing has changed apart from how the terrorist’s intentions. In some cases, we accept that this may be enough to change the permissibility of the situation. But there is another, more worrying instance, where a physical change in the situation, in a way which does not seem morally relevant, appears tochange an action’s permissibility under DDE.

This brings us onto our second worry – the idea that the distinction between foresight and intention may simply be too contrived. We do not want to end up with a scenario where a tiny change to the physical layout of a situation sees the permissibility change alongside it. There are some famous problem cases that have been proposed which claim to show DDE is not an accurate description of what we want to endorse or condone with regards to morality. Philippa Foot, for instance, came up with a hypothetical “trolley case” in which your trolley would either continue to run on the track it was currently on, killing five people, or, if you made the active decision to flick the switch, would run onto the other track, killing one person[4]. Although some people may claim the bad situation was already in place and you should not get involved under any circumstances, to most it seems initially apparent that the morally correct thing to do would be to take the track which only killed one person, and the death of this one person would be a foreseen but unintended side effect. We could prove that it was unintended by suggesting that the one person may, in fact, escape. In this case, you would prefer for the person to escape and for no-one to die, which suggests you never intended their death.

However, Thomson came up with a further suggestion – that the track now loops around and joins up with the other track, which has the five people on it.[5] The only way in which to stop your trolley making the full circuit and ending in killing the five people is for something to stop it: namely, the one person. Now, then, Thomson argues that your preference has now become that you do not want them to escape, as the trolley would then kill the five people, and you therefore now intend for the one person to be harmed. Therefore, the death of this person has now become a means to an end, as opposed to a foreseen but unintended side effect – something that DDE cannot condone. But surely the trivial difference of a few metres of train track cannot make the difference between right and wrong? Yet the death of the one is now part of the plan to save the five, meaning that the permissibility has changed since we allowed for the changing of the track in the previous case. It is suggested, then, that the Loop Case shows that using someone as a means to an end does not make a difference to moral permissibility (all other things being equal), and DDE should therefore be abandoned.

Solutions to this have been suggested – Michael Costa’s article “Another Trip On The Trolley”[6] advises that, seeing as the harm is already in place, there is no way to escape the “using a person as a means” clause – either the one person will save the five, or the five will save the one (by stopping the trolley circling round and killing the one). You are forced into using a person or people as a means to save others, so it may as well be that you minimize the damage. However, Costa’s argument unfortunately collapses if a minor adjustment is made to the track so that the one person is on higher ground, such that the trolley would never circle around and hit them if it took the track with the five on. Again, a minor adjustment to the logistics of the situation seems to stop the argument in its tracks.

We also do not want to fall into the trap of attributing rightness and wrongness based on a trivial re-description of the agent’s intention. Something that has been claimed is that the agent’s initial intention is to stop the five being hit from the front – an aim which would be defeated by the trolley hitting the five from the back (as it would do, without the presence of the one). Matthew Liao, however, argues that it seems implausible that the agent’s true intention to prevent the trolley hitting the five is concerned with the clause “from the front”, instead of simply intending to prevent the trolley hitting them at all –similarly as with the few extra metres of track, why should we care if the five are hit from the front of the back? [7]

Frances Kamm proposes that we should add an extra clause to DDE, based on the recognition of an important difference between doing something to bring about harm, which should be condemned, and doing something because harm will occur, but not in order to bring that harm about. The latter of these should also not be confused with a situation where one merely foresees harm. Proponents of DDE have overlooked the possibility that one may do something because, but not in order that, harm may occur, and have therefore fallen foul of miscategorising this case.

The example Kamm gives to clearly showcase this difference is as follows:

“I intend to give a party in order for me and my friends to have fun. However, I foresee that this will leave a big mess, and I do not want to have a party if I will be left to clean it up. I also foresee a further effect of the party: If my friends have fun, they will feel indebted to me and help me clean up. I assume that a feeling of indebtedness is something of a negative for a person to have. I give the party because I believe that my friends will feel indebted and (so) because I will not have a mess to clean up. These expectations are [necessary] conditions of my action. I would not act unless I had them. The fact that they will feel indebted is a reason for my acting. But I do not give the party even in part in order to make my friends feel indebted nor in order to not have a mess.”[8]

Kamm refers to DDE plus this extra amendment as “The Doctrine of Triple Effect”. The significance of this in the Trolley Loop Case is clear. The trolley has not been diverted in order to hit the one, but because it will hit the one. This amendment may not have been recognized before simply because it is only necessary in very small number of extremely specific cases, like the Party case and the Trolley Loop case, but using it allows us to adequately draw all the conclusions we want from the problem cases without simply re-describing an agent’s intentions.

In conclusion, I believe that intention has imperative moral relevance that we cannot overlook, and that a full and complete theory of DDE can fully and accurately summarize our intuitions with regard to intention, as long as we remember that it is not simply an intention/foresight distinction, and ensure we also include vital ingredients such as a proportionality condition, a “means to an end” condition and also Kamm’s distinction between doing something to bring about harm, and doing something because harm will occur, but not in order to bring that harm about.

[1] Davidson, Donald, “Actions, Reasons and Causes” in “Journal of Philosophy” Vol.60, Chapter I, p.686

[2] Aquinas, Thomas – “Summa Theologica”, 1273, (II-II, Q64, A7)

[3] Bennett, Jonathan – “The Act Itself” (1998, Oxford University Press) discusses a huge range of “bomber” cases in his chapter on “Intentions”. The particular case I am focussing on, however, may be better attributed to McMahan (McMahan, Jeff – “Revising the Doctrine of Double Effect” (1994, Journal of Applied Philosophy, Volume 11, No.2))

[4] Foot, Philippa – “The Problem of Abortion and the Doctrine of the Double Effect”, (1967, Oxford Review, Number 5)

[5] Thomson, Judith Jarvis – “The Trolley Problem” (1985, Yale Law Journal, Vol. 94, p.1402-1406)

[6] Costa, Michael – “Another Trip On The Trolley” (1987, Southern Journal of Philosophy)

[7] Liao, Matthew – “The Loop Case and Kamm’s Doctrine of Triple Effect” (Nov 2009, Philosophical Studies)

[8] Kamm, Frances – “Intricate Ethics: Rights, Responsibilities, and Permissible Harm” (2006, Oxford University Press), p95



Main Sources

Aquinas, Thomas – “Summa Theologica” (1273, published 1988 in “Basic Writings of Saint Thomas Aquinas”, Anton Pegis (ed), Indianapolis/Cambridge, Hackett Publishing Company)

Bennett, Jonathan – “The Act Itself” (1998, Oxford University Press)

Costa, Michael – “Another Trip On The Trolley” (1987, Southern Journal of Philosophy)

Davidson, Donald – “Actions, Reasons and Causes” (1963, The Journal of Philosophy, Volume 60, American Philosophical Association, Eastern Division, Sixtieth Annual Meeting)

Foot, Philippa – “The Problem of Abortion and the Doctrine of the Double Effect”, (1967, Oxford Review, Number 5) and “Morality, Action and Outcome” in “Morality and Objectivity: A Tribute to J. L. Mackie” (ed. Honderich, Ted) (1985, London, Routledge)

Kamm, Frances – “Physician-Assisted Suicide, the Doctrine of Double Effect, and the Ground of Value”, in “Ethics”, Volume 109 (1999, University of Chicago Press), “The Doctrine of Double Effect: Reflections on Theoretical and Practical Issues”, in “Journal of Medicine and Philosophy” Volume 16, pages 571-585 (1991, Netherlands, Kluwer Academic Publishers), and “The Doctrine of Triple Effect and Why a Rational Agent Need Not Intend the Means to His End” (2000, Proceedings of the Aristotelian Society, Supplementary Volumes, Vol 74), “Intricate Ethics: Rights, Responsibilities, and Permissible Harm” (2006, Oxford University Press)

Liao, Matthew – “The Loop Case and Kamm’s Doctrine of Triple Effect” (Nov 2009, Philosophical Studies)

McMahan, Jeff – “Revising the Doctrine of Double Effect” (1994, Journal of Applied Philosophy, Volume 11, No.2)

Olver, Ian – “Is Death Ever Preferable to Life?” (2002, Dordecht, The Netherlands, Kluwer Academic Publishers)

Otsuka, Michael – “Double Effect, Triple Effect and the Trolley Problem: Squaring the Circle in Looping Cases” in “Utilitas” (2008, Cambridge University Press)

Quinn, Warren – “Actions, Intentions, and Consequences: The Doctrine of Double Effect” (1989, “Philosophy and Public Affairs”, Volume 18, p.334-351)

Stanford Encyclopedia of Philosophy – Doctrine of Double Effect, Doing vs. Allowing Harm, Intention (Links: , , )

Thomson, Judith Jarvis – “Killing, Letting Die, and the Trolley Problem” (1976, The Monist, Vol. 59), and “The Trolley Problem” (1985, Yale Law Journal, Vol. 94)

Unger, Peter – “Living High and Letting Die” (1996, New York, Oxford University Press)

I also benefitted greatly from suggestions and additions from my supervisor, Sebastian Nye (University of Cambridge).

Other Sources

Boyle, Joseph – “Who Is Entitled to Double Effect?” (1991, The Journal of Medicine and Philosophy, Volume 16, p.475-494)

Donagan, Alan – “Moral Absolutism and The Double-Effect Exception: Reflections on Joseph Boyle’s ‘Who Is Entitled to Double Effect?’” (1991, The Journal of Medicine and Philosophy, Volume 16, p.495-509)

Nagel, Thomas – “Agent-Relativity and Deontology” in “The View From Nowhere” (1986, New York, Oxford University Press)

Quinn, Warren – “Actions, Intentions, and Consequences: The Doctrine of Doing and Allowing” (1989, “Philosophical Review”, Volume 98, p.287-312), “Morality and Action” (1993, Cambridge University Press), and “Virtues and Reasons: Philippa Foot and Moral Theory: Essays in Honour of Philippa Foot”, also edited by Rosalind Hursthouse and Gavin Lawrence (1995, Oxford University Press)

Is Suicide Permissible?

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[1], 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[2], 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[3]. A third perspective is outlined by Cowley, who proposes that suicide cannot be either rational or irrational.[4] 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”.[5] 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.


[1] 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).

[2] Devine, Philip – “On Choosing Death” in “The Ethics of Homicide” (1990, University of Notre Dame Press)

[3] Brandt, Richard – “The Morality and Rationality of Suicide” in “A Handbook For The Study of Suicide” (1975, Oxford University Press)

[4] Cowley, Christopher – “Suicide is Neither Rational nor Irrational” (2006, Ethical Theory and Moral Practice Journal) ( )

[5] Audi, Robert – “Rationality” in “The Cambridge Dictionary of Philosophy” (1995, Cambridge University Press, p674-675)



Main Sources

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) ( )
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:, )

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).


Other Sources

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)