Some people believe that when facing difficult decisions we should give priority to those who are worst-off. In ‘Prioritarianism and the Measure of Utility’, Michael Otsuka argues that this is only true in some situations.
Two Difficult Decisions
Think about how you would respond to the difficult decisions posed by the following two scenarios.
Scenario 1. You have two identical twin sons, Bill and Ted. Bill has just been diagnosed with a serious and debilitating medical condition and will soon require regular hospital treatment in the city. Ted is in good health, plays sport and loves the outdoors. The countryside is his ideal environment. To make things a little more concrete, let’s set their well-being on a scale from 0 to 50: Ted’s well-being would be 40 in the city and 45 in the countryside. Bill’s well-being would only be 16 in the countryside but 20 in the city.
You are required to move and you have to choose between a house in the city and a house in the countryside. Which would you choose?
Scenario 2. Imagine now that you have just one son, Rufus. A medical test has shown that there is a 50% chance that he will develop the same condition as Bill, and a 50% chance that he will remain healthy and be like Ted. The same scales apply: if he remains healthy, Rufus’ well-being will be 40 in the city and 45 in the countryside. If he develops Bill’s condition, his well-being will be 16 in the countryside and 20 in the city. This is all that you know, and you must decide now where to move. Which option would you choose?
When faced with decisions like these, it helps to have a framework available to explain why one choice is better than another. Here’s one:
Prioritarianism claims that the people who are the worst off get priority. If somebody is worse off, our moral obligation to help increase their level of well-being is stronger – even when helping them prevents us from providing a larger increase to the well-being of somebody who is better off.
On this view, the right choice in both of these scenarios is to move to the city. Why?
In Scenario 1, Bill is worse off because of his medical condition, and so he gets priority. We have a greater obligation to help him out, even if this means that Ted’s well-being suffers by a larger amount. In Scenario 2, Rufus gets priority in the event that he turns out worse off and so you should move to the city just in case he becomes unwell.
Prioritarianism claims that we should give priority to those who are the worst off in absolute terms; it is not concerned with how well off people are in relation to one another. It shouldn’t make any difference to a prioritarian’s decision, therefore, whether they are considering one person’s level of well-being or comparing the well-being levels of two people.
Given that the number of people in a scenario is irrelevant to prioritarianism, if we are convinced by the prioritarian response to Scenario 1 then we should also be convinced by what the prioritarian has to say about Scenario 2. However, some people, such as Michael Otsuka, are not.
Otsuka poses the following challenge to prioritarianism. When faced with a decision where the choice we make will affect the same person differently depending on how their future pans out, why should we favour the choice that will only benefit this person should the worst happen? Especially when this choice is made at the expense of a greater increase in well-being in the better scenario? If you were Rufus would you not want to maximise your potential well-being by moving to the countryside? Surely this would be the most rational thing to do.
To support this claim, Otsuka appeals to an influential account of how rationality operates in these kinds of situations – an account based upon the Von Neumann–Morgenstern (VNM) utility theorem:
VNM Utility Theorem: When faced with a difficult decision, a self-interested, rational person will prefer a certain option if and only if this is the option from which they expect to gain the most.
As we have seen, the greatest expected gain for Rufus would come from a move to the countryside and so if Rufus were to act rationally he would choose the countryside.
In response, the prioritarian may say something like this: although it’s true that prioritarianism shouldn’t go against what it would be rational for an individual to choose for themselves, the truly rational choice to make in this kind of situation would be the prioritarian one. What should we make of this? Otsuka’a answer is that the prioritarian clearly hasn’t understood his challenge.
For Otsuka’s argument, the crucial feature of the VNM utility theorem is its use of ‘if and only if’; this is the reason the prioritarian response fails. Why? Because saying ‘p if and only if q’ is equivalent to saying ‘if p then q, and, if q then p’. So, as well as saying that if an option offers the largest expected gain then a rational individual will prefer it, the VNM utility theorem also says that if a rational individual prefers an option then this is the option that offers the largest expected gain.
Otsuka’s insight is that the size of the expected gain to be had and the rational choice to make cannot come apart in the way that the prioritarian hopes. As well as describing what choice a rational person will make, the VNM utility theorem also specifies how the expected gains themselves are to be measured.
An Alternative Framework
As long as something like the VNM utility theorem is correct, prioritarianism is in trouble. Unless the prioritarian is prepared for some of her choices to clash with rationality, she’ll be unable to defend her decision to move Rufus to the city.
Recommending the same choice in both of our scenarios was meant to support the prioritarian claim that we should consider well-being in absolute instead of relative terms. Otsuka’s challenge has seriously undermined this. If we’re prepared to accept the prioritarian choice in Scenario 1 but not in Scenario 2 then this must be because of some difference between the two cases. The most obvious difference is that in Scenario 1 we are considering two people’s well-being whilst in Scenario 2 we are considering only one. If the number of people in a scenario is significant, then this suggests an alternative, relative, way of interpreting ‘worse-off’ and an alternative framework for difficult decisions.
Otsuka’s alternative framework is this: when faced with a difficult decision involving more than one person, we should give priority to those who are worst off in relative terms. Such a choice can be justified by appealing to equality. In decisions involving only one person, where there is nobody to be equal to, we should maximise well-being. This explains why we should move Bill and Ted to the city but move Rufus to the countryside. Excellent.
By Ewan Rodgers
Michael Otsuka is Professor of Philosophy in the Department of Philosophy, Logic and Scientific Method. His paper ‘Prioritarianism and the Measure of Utility’ has been been published in the Journal of Political Philosophy, 23 (2015): 1-22.