The Permissibility of Saving the Fewer.

University of St Andrews

Matriculation ID: 120016390

Module Code: PY4626

Title of Assessed Work: When faced with a choice between saving the lives of one group of people or another group of people, is it ever permissible to choose the group with fewer people?

Tutor’s Name: Dr. Lisa Jones

Number in sequence (e.g. essay 1 of 2): 2

Date Submitted: 12.12.14

Word Count: 3660

Declaration

By entering my matriculation number above:

• I confirm that I have read and understood the University’s policy on Academic Misconduct including Plagiarism;

• I hereby declare that the attached piece of written work is my own work and that I have not reproduced, without acknowledgement, the work of another.

• I confirm that this piece of work has not previously been submitted for assessment on another course.

In a situation where an agent can save either a larger or

smaller group of people, a question arises as to whether

it is, all things equal, morally permissible for them to

save the group with fewer people. On the one hand, number

skeptics believe that we are permitted to save the fewer,

and on the other, consequentialists and contractualists

believe that we are never permitted to do so. The

standard example that I shall use in this essay is that

of a boat captain, who is capable of saving either a

group of five persons on one island, or one person on

another island. The assumptions involved in all things

being equal are as follows: the captain has no special

relations to any individuals involved, and stands to gain

nothing from the choice he makes; he cannot choose to do

nothing; no individual has any more right to the

captain’s resource than another; all individuals are

going to die unless saved, and no individual’s death will

be worse than another. In this essay, I will argue that

the best theory to use in this situation is Jens

Timmermann’s individualist lottery, and thus, that it is

permissible to save the fewer. This is based on two

essential claims. Firstly, consequentialists ground their

view that it’s only permissible to save the larger number

on a sub-theory of aggregation, which John Taurek shows

to be false. Secondly, all theories apart from the

individualist lottery fail to treat people as ends,

equally respecting their preference to live, with

consequentialists and contractualists not even giving the

fewer a chance to be saved. For the sake of clarity in

this essay, I will proceed in a chronological way, with

the exception of giving Scanlon’s argument before Kamm’s.

To begin, I will analyse a consequentialist view,

followed by Taurek’s rejection of it, and his counter

theory. I will then present Thomas Scanlon and Frances

Kamm’s views against Taurek and critically assess them.

Finally, I will present Timmermann’s position.

From a consequentialist viewpoint, the permissible action

will be the one that brings about the best consequence,

or prevents the worst. In the case of the boat captain,

he will always choose to save the greater number, because

the death of five persons is worse than the death of one,

and the captain has no knowledge of the islanders that

would suggest otherwise. The problem for

consequentialists is the first premise, that the death of

five individuals is worse than the death of one, is

actually false. Taurek argues that we cannot aggregate

individual claims into a group claim. We must think of

the group of five as five individuals, each with their

own preference to be saved and live. Furthermore, we must

focus not on the loss of individuals, but on the loss to

individuals. The loss to each individual is the same:

their life. Any of the five individuals in the larger

group of the island cannot claim that his loss is any

greater than the loss of the single man on the other

island. Therefore, we cannot say that the five persons

losing their lives is worse than the one man losing his.

As Taurek writes, “Five individuals each losing his life

does not add up to anyone’s experiencing a loss five

times greater than the loss suffered by any one of the

five.”1 We cannot aggregate the loss to each of the five

individuals into a greater loss than that of the loss to

the single person on the other island. Similarly,

Elizabeth Anscombe has argued that no single individual

of the larger group can declare that they are wronged if

we save the single person. Anscombe asks how it is so

that a group of single people can claim that they

together are owed rescue as opposed to the single man,

and that they are wronged if the single man is saved

rather than them.2 The idea is the same: we cannot

aggregate.

Taurek’s argument against aggregation makes it impossible

for the consequentialist to argue that it’s only

permissible to save the greater number. Their view is

based on the claim that the death of five individuals is

worse than the death of a single man, and Taurek shows

1 John Taurek, “Should the Numbers Count?” Philosophy and Public Affairs 6, No. 4 (1977), p. 3072 Frances Kamm, Morality, Mortality. Volume I (New York: OUP, 1993), p. 119

that this is false. Consequentialists might argue that

even though any single individual cannot claim he has

suffered a loss five times worse than the single man,

there is still more overall loss in the larger group. For

example, say we have five oranges versus one orange, all

of which have a taste value of one. Taurek’s point is

that no one of the five oranges has a taste value of

five, simply by virtue of being with four other oranges.

However, though we may not be able to aggregate in this

way, consequentialists could still say that there is an

overall value of five taste points against one. But this

type of aggregation requires me as the judge of how much

the oranges mean to me. This is not possible in the cases

of life, as the boat captain cannot aggregate lives based

on their value to him. A further criticism of the

consequentialist position is that it doesn’t treat people

as ends, defined by Timmermann as, “independent

existences that deserve to be respected.”3 The captain

simply saves the greater number, not respecting the

preference of the single man on the other island to live.3 Jens Timmermann, “The Individualist Lottery: How People Count, but Not Their Numbers”, Analysis 64, No. 2 (2004),p.110

The single man stands no chance of being saved, as though

he was not on the island at all. Indeed, a

consequentialist may simply say that equality doesn’t

really matter. They are after all consequentialists, thus

the morality of the action itself is not necessarily

important in judging its permissibility. Despite this,

the consequentialist position of saving the greater

number ultimately fails because it falsely aggregates

individuals’ claims into a group claim.

For Taurek, the way we should proceed in the boat

captain’s situation is to allow chance to determine who

is saved. In allowing chance to determine the action of

the captain, we must accept that it is permissible to

save the group with fewer, though more specifically it is

permissible to save either group. Having established that

consequentialism fails due to aggregation, we should

focus on a solution that aims to be as fair as possible

to all the individuals involved. The way in which we

should determine the chances of being saved is by the

number of outcomes. In the island example, there are two

outcomes: either the captain saves the five, or he saves

the one. Each outcome should be given an equal chance,

and so, in such a situation, the captain should toss a

coin to decide whom to save. This ensures that the fewer

have just as much chance of being saved as the larger

group.4 Additionally, in different number cases,

consequentialists often posit that were the numbers to be

more extreme, we would be more inclined to save the

greater number. For example, were there to be a million

on one island and a mere handful on the other, the

disparity in numbers would naturally push us to save the

million. However, Taurek states, “I cannot see how or why

the mere addition of numbers should change anything.”5 In

essence, we should still flip a coin, giving the handful

the same chance at being saved as the million.

Taurek encounters a serious objection here, both from

Kamm and Timmermann. Taurek leaves us not treating

persons equally by focusing on giving the fewer just as

much chance of being saved as the many. Say we start off

4 Taurek, p. 3065 ibid.

with a situation where there are five on one island and

one on the other. Here we should flip the coin to decide

who is saved. There is a fifty-fifty chance for both

groups. If we then add a thousand people two the island

with the five, such that the captain now saves a thousand

and five, or just one, Taurek believes that we should

still flip a coin. This suggests that the preferences of

the additional thousand people are totally ignored, or at

least, in combination with the five already present, only

accumulate to a single man’s preference. Seeing as

additional people’s preferences seem to not matter at

all, it seems that Taurek’s policy cannot be one that

treats each individual’s preference in the situation with

respect, which is the ground for adopting a method that

permits us to save the few. There is a big difference

between giving the fewer an equal chance of living, and

giving them a fair shot at being saved. We want the fewer

to have their fair share of chance to be saved, not

necessarily an equal chance. Giving the fewer an equal

chance pushes the equality and fairness objection back

the other way, and fails to treat those in the larger

group as ends with preferences that need to be respected.

Following Taurek’s argument, Thomas Scanlon, a

contractualist, affirmed that there is a way for it to

only be permissible to save the greater number, without

the need to aggregate. Importantly, although it advocates

the same conclusion as consequentualism, Scanlon does not

argue the permissibility of saving the greater number

because of its consequences. To understand the

contractualist position, we must adapt our situation

slightly. Imagine the same boat captain situation, but

this time with an equal number case, such that there is

only one person on each island, A and B. In this case, it

would be permissible to save either A or B. When another

person, C, is added to the island with B, the tie between

A and B is broken in favour of saving C. What matters to

Scanlon is that each islander must be given positive

weight and each person’s life must be given the same

importance,6 meaning that Scanlon’s approach is one that

6 Thomas Scanlon, What We Owe To Each Other (Massachusetts: HUP, 1998), p. 233

fundamentally aims for maximal fairness to each

individual. For Scanlon, the idea that we save the

greater number because of the presence of additional

people means that we are not aggregating, but simply

acting to give proper weight to each individual.

However, Michael Otsuka has correctly argued that Scanlon

is actually guilty of covertly aggregating numbers.

Otsuka states that Scanlon’s argument is grounded in an

appeal to the claim that that the captain should save the

greater number because the preference of a group of

individuals to be saved outweighs the opposing preference

of a single individual to be saved. Remember that

aggregation involves grouping individuals preferences.

Otsuka states, “a claim is that of a group of individuals

when it is the claim of individuals considered together

or in combination rather than one by one.”7 Although

Otsuka makes a strong case against Scanlon’s

contractualism, involving Kamm’s argument alongside

Scanlon’s contractualism here is wrong. Kamm makes a

7 Michael Otsuka, “Scanlon and the Claims of the Many versus the One”, Analysis 60, No. 3 (2000), p. 292

similar objection against aggregation in Volume I of

Morality, Mortality. If we take three people (A, B, and C),

Scanlon’s argument can be seen as suggesting the

following syllogism: B and C dying is worse than B dying;

A dying is equally bad as B dying; Therefore, B and C

dying is worse than A dying. What Kamm objects to here is

the conclusion. Originally, we have a moral tie between A

and B. When we add C, this breaks the tie before we have

settled the conflict between A and B. As she states, “If

we then save the greater number, this means severing our

tie to A before giving him an equal chance.”8 I think its

clear here that Kamm believes in assessing group claims

in a one-by-one style, not in combination as Otsuka

suggests.

The above rejection from Kamm demonstrates that Scanlon’s

methodology totally fails to respect A’s preference to

live, which is contradictory to his proposal at the heart

of his theory to give the life of each islander positive

weight. Seeing as C is situated on the same island as B,

the tie is broken in favour of saving C, and thus B as 8 Kamm, p. 87

well. In this scenario, A has no chance of being saved by

the captain. In fact, B can even somewhat complain that

his preference to live was totally ignored, as B is only

saved by virtue of the captain’s duty to save him once he

has chosen to save C. In response, Scanlon can only point

to a notion of aggregation such that A’s life was given

its proper weight, but it was outweighed by B and Cs’

lives. However, admission that he was aggregating would

take us back to Taurek’s denial that it is even possible

to combine B and Cs’ life against A’s.

Kamm’s balancing argument, like Scanlon’s contractualism,

is directed at achieving the goal of it only being

permissible to save the greater number, avoiding

consequentialist aggregation, and promoting the idea of

giving each individuals interests equal consideration.

Take the island situation, but with one person on each

island. All things equal, Kamm believes, “Since their

interests are opposed and of equal weight, it might be

suggested that they cancel each other out.”9 Seeing as we can do

this for an equal number case, when the numbers are 9 ibid, p. 116

different, we should simply cancel out equal and opposing

claims until we are left with a majority in one

situation. In a case where there are five individuals

stranded on island A and three on island B, we cancel out

the equal and opposing claims of the three on B with the

five on A, leaving us with two unopposed preferences on

island A. Therefore, by majority rule we should save

those on A. Clearly, this method will always lead the

agent to save the greater number, but not by aggregation,

as individuals interests are assessed one by one, not

collectively. There is room for misconception here. It is

not that the individuals’ preferences are cancelled and

not considered, but “that neither of two equal and

opposing claims can finally decide an outcome”. Each

individual person’s preferences are thus considered,

unlike Taurek.

The first objection of Kamm’s balancing argument is that,

like Taurek, it seems to ignore extra people. Lets take a

situation where there are ten people on one island, and

five on the other. It will only take six of the ten

people to balance out the five on the other island, which

leaves four individuals whose interests have not been

considered in determining who gets saved by the captain.

This would, like Taurek, mean that Kamm does not actually

treat all people’s interests equally. However, Kamm does

respond to this potential objection by stating that,

“each ‘excess’ member of the majority knows that if his

side had not yet won, he would have been used to balance

an opponent.”10 Those individuals who are ‘excess’ to the

balancing procedure are therefore only excess after they

have already had their interests fulfilled.

Furthermore, although Kamm’s theory is not

consequentialist, it does come under scrutiny from an

associated problem, presented by Bernard Williams.

Williams argued that consequentialist theories,

especially utilitarian ones, fail to take the integrity

of individuals seriously, treating those involved not as

individuals, but rather as mere units of value to be

balanced against other units.11 This is an issue that

10 ibid, p. 11711 Lecture slides, Week 8, slide 11

applies to Kamm, in as much that she does use people as

units of value that are balanced off against each other.

However, although it is an issue, especially for Kamm,

all theories involved in resolving this situation are

guilty of somewhat compromising the integrity of

individuals. In Taurek’s coin toss, we are somewhat

guilty of treating people as units of value, just that

its always one, and in Timmermann’s lottery, persons’

lives are essentially reduced to a segment on a wheel.

My major issue with Kamm’s theory is that it gives no

chance to the fewer to be saved. On majority rule, the

larger group will always be saved, yet it’s like there

was no other group. Yes, Kamm has considered each persons

interest to live, but she has not given them any chance

to be saved. It’s not good enough to simply consider them

by cancelling them with others. Treating all the

individuals in the scenario equally, and as ends doesn’t

mean that we simply need to consider them, but that we

give them a chance to be saved. On the one hand, Taurek’s

method for deciding who gets saved is too generous to the

fewer, which makes it unfair on the bigger group. On the

other hand, Kamm’s proposed balancing argument is too

focused on saving the many that the smaller group don’t

have any chance in being saved. It is all to easy to get

confused into thinking that a number skeptic is simply

trying to advocate that we should save the fewer, but

this is far from the truth. The number skeptic believes

we are permitted to save either the larger or smaller

group. To be totally equal to all individuals involved,

we need a theory that sits in between Taurek and Kamm’s.

We cannot give the group with fewer no chance of

survival, but we must also be careful to not give them

too much chance of being saved, as this compromises the

equality given to the larger group.

The perfect solution to this issue is Timmermann’s

proposal of an individualist lottery. The captain can

either save five people on one island, or one person on

another. What the individualist lottery allows the

captain to do is create a wheel that is divided into six

equal segments, one for each individual involved in the

situation. A chance spin will decide a segment and the

captain will direct his boat to save the individual that

has been chosen by chance. If this is the lonely man on

the one island then he is saved and the five on the other

island will die. If it is one of the five on the other

island the captain will first save the person who won the

lottery, and then incurs a duty to save the other four

islanders on that island. Mathematically speaking, the

one and the five have the same chance of being saved in

both an individual lottery and a weighted lottery. But in

the weighted lottery, the lonely man is given less chance

of being saved (one in six) compared to the five (five in

six), since the chances for the five have been

aggregated. However, in the individualist lottery each

individual has a one in six chance of being saved, which

accomplishes the goal of equally considering everyone’s

preference to live, and giving each individual an equal

chance at being saved. Of course, depending on the

numbers involved in the situation the larger group will

always have a greater chance of being saved. In a two

versus one situation, it’s slightly more likely, and in a

million versus one scenario, it’s extremely likely that

the larger group will be saved. However, there is a

chance for the fewer to be saved, and it would be

permissible to do so. As Timmermann writes, “Being stuck

on an island and losing the lottery, whoever perishes

will undoubtedly bemoan his ill fortune; but he cannot

complain about unfair treatment”.12 Timmermann’s

individualist lottery is the only example of a solution

where no individual involved can genuinely complain that

they have not been given their fair chance at surviving,

and thus should be adopted as the go-to method for

situations such as the captain’s dilemma.

To conclude, in situations where we must either save one

group of people or another, there are various methods

that present themselves. Firstly, there are those that

advocate it is only permissible to save the group with

the greater number of persons. One is the

consequentialist proposal, which simply guides us to save

the greater number because that will bring about the

better consequence. It is worse that five die than one. 12 Timmermann, p. 111

However, Taurek has shown that this aggregation of

individual claims to a group claim cannot be done,

rendering this proposal redundant. Another promoter of

saving the greater number is Scanlon and his

contractualism, yet in a way that does not involve

aggregation. However, this view falls short on two

counts. Firstly, Otsuka shows that Scanlon does actually

end up aggregating. Secondly, Scanlon’s view, which is

supposedly grounded in giving each individual’s life

‘positive weight’, simply gives the fewer no chance of

surviving at all. Kamm finally presents us with a way to

justify saving the greater number that doesn’t involve

aggregating, yet her theory also suffers in it’s horrific

unfairness, not giving the fewer any chance of being

saved. We must turn to theories that at least make it

permissible to save the few. Of these, I consider two.

Taurek’s argument that we should flip a coin turns the

argument somewhat upside-down, in that it ignores the

preferences of all-but-one in the larger group. We do not

want to toss a coin to decide the fate of millions versus

one. Timmermann’s individualist lottery accomplishes

total equality in treating all individuals as ends, both

equally respecting their preferences, and giving them a

fair chance of survival. It is a theory where no

individual involved can complain, and is such the best

proposal regarding these situations. Thus, we are

permitted to save either the greater or the fewer. What

we actually do will depend on chance determining which

individual wins the lottery.

Bibliography:

Kamm, Frances, Morality, Mortality. Volume I (New York: OUP,

1993)

Lecture Slides, PY4626, Week 8, Slide 11

Otsuka, Michael, “Scanlon and the Claims of the Many

versus the One”, Analysis 60, No. 3 (2000), pp. 288-293

Scanlon, Thomas, What We Owe To Each Other (Massachusetts:

HUP, 1998)

Taurek, John, “Should the Numbers Count?” Philosophy and

Public Affairs 6, No. 4 (1977), pp. 293-316

Timmermann, Jens, “The Individualist Lottery: How People

Count, but Not Their Numbers”, Analysis 64, No. 2 (2004),

pp. 106-112