Article Text

Download PDFPDF
Why sore throats don't aggregate against a life, but arms do
  1. Alex Voorhoeve
  1. Correspondence to Dr Alex Voorhoeve, Department of Philosophy, Logic and Scientific Method (LSE), London School of Economics and Political Science, Houghton Street, London Wc2A 2AE, UK; a.e.voorhoeve{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Part IV of Bioethical Prescriptions 1 masterfully joins philosophical imagination and rigour in its discussion of moral questions that arise in allocating scarce healthcare resources. I shall focus on a question on which Kamm's analysis yields remarkable insight, even though I disagree with some of her conclusions. The question is: Suppose that one must either (a) save all members of a group of A-people (who are otherwise fine) from an identical individual loss, short of death or (b) save a single young person, B, from a terminal illness, thereby restoring him to good health for a normal lifespan. What ought one to do?

In Bioethical Prescriptions and elsewhere, Kamm argues for the following two-part answer:

  1. If the loss to each person in the A-group is very small, then one must save B's life, no matter how numerous the A-group.

  2. If the loss to each person in the A-group is close enough to B's loss, then for a very large number of people in the A-group, one is permitted to save the A-group.

Kamm offers the following principle underlying (i): Each of us who is otherwise fine has a duty to suffer (at least) a relatively minimal loss (e.g., a sore throat) in order to save another person's life. So long as suffering the small loss is a duty for any given person, no number of the small losses can be aggregated to outweigh saving the life (p. 369).1

This antiaggregation principle has an interesting rationale (chapters 8–102).3 An important part of our distributive morality involves placing oneself in a person's shoes and assessing how what is at stake for her compares to what is at stake for a single other with a competing claim. When one places oneself in a person's shoes, one takes on her permissible self-concern. For example, when one takes up the perspective of a member of the A-group (call her A1), her well-being takes on special importance compared with B's well-being. Up to a limit, such concern for her dear self is permissible. It is also permissible for her to act on it when no other moral considerations (such as rights or special relationships) stand in her way. For example, in a one-to-one situation, A1 has a prerogative to avert a moderate loss to herself (say, losing an arm) rather than save B's life when she cannot do both. But when what is at stake for A1 is very minor (eg, a sore throat), then she is obligated to avert B's death. In a one-to-one comparison of competing claims, from A1's permissible personal perspective, B's claim then takes priority. Given that, by assumption, all A-people face the same loss, this is then also true from every other A-person's perspective. The same is true from B's perspective, of course. It follows that when what is at stake for each A-person, taken separately, is very small relative to what is at stake for B, then from each person's perspective, when one compares competing claims one-to-one, there will be unanimous agreement to prioritise B's claim. In sum, a rationale for Kamm's antiaggregation principle is this. As an impartial distributor, one ought to respect a form of unanimity that emerges when one takes up each person's perspective, one at a time, and compares what is at stake for her with what is at stake for a person with a competing claim.

Let us now turn to the question of when aggregation is permissible, on Kamm's view. The aforementioned one-to-one perspective is but one element of distributive morality. Another element recognises that numbers count. The more claims of a given strength one satisfies, the more good one does. When the number of A-people one can save is sufficiently large, one may do more good by saving them than by saving B. Kamm allows the pursuit of this greater good only when each A-person's claim is ‘close enough’ in strength to B's. But when is this so? Bioethical Prescriptions lacks a clear answer, but the following proposal fits elements of Kamm's outlook. It is acceptable to aggregate the A-people's claims when the aforementioned antiaggregation principle is respected. That is, it is acceptable to aggregate the A-group's claims when from an A-person's permissible personal perspective, her loss may take priority over B's life in a one-to-one comparison. In such situations, every A-person can permissibly prioritise herself alone over B. By contrast, B will permissibly prioritise himself over any A, taken separately. In such cases, the pairwise comparison of claims from each person's perspective therefore does not resolve the conflict of interests. The proposal is that it is permissible to resolve this conflict by an appeal to the greater good.

In some passages, Kamm comes close to endorsing this. She suggests that an individual does not have a duty to give up an arm in order to save a stranger's life and wonders whether arms might therefore aggregate against a life (pp. 170 and 182–3).2 As she writes: for macro decisions—for example, whether to invest in research to cure a disease that will (…) deprive a few people of ten years of life, or in research to cure a disease that will only whither an arm in many—[we] might permit aggregation of significant (…) lesser losses (p. 370).1

However, in other places, Kamm appears to reject this idea. In common-sense morality (and on Kamm's view), it is clearly permissible for a person to save herself from lifelong paraplegia rather than save a stranger from death. On the proposed principle for aggregation, it would follow that one is allowed to save a very numerous A-group from paraplegia rather than save B from death. But in Intricate Ethics, Kamm argues that even 10 000 people's claims to be cured of paraplegia cannot jointly outcompete one person's claim to be saved from death. “In a context where a life is at stake”, she writes, “saving [a multitude] from paraplegia is not appropriate (because paraplegia is not relevant to death)” (p. 485).i 4

This judgement cannot, of course, be defended by an appeal to the perspective of each person concerned. It therefore lacks the rationale of Kamm's more modest antiaggregation principle. It is also very implausible. When, with limited resources, one can either save 10 000 from paraplegia or instead save one from death, it seems straightforwardly permissible to do the former. The view that a public entity such as the National Health Service in Britain or a private donor such as the Gates Foundation ought to do otherwise strikes me as absurd. A more plausible view allows aggregation of weaker claims against a life just in case, in a one-to-one contest, a person with a weaker claim could permissibly prioritise her claim over a stranger's competing claim to life. When one must choose whether to save one from death or many from lesser harm, this view would permit saving the many whenever this is obviously morally right. It would also place a plausible constraint on the aggregation of lesser claims out of respect for the person whose life is at stake. If Kamm were to endorse it, she would give a bioethical prescription worth following.


I thank Luc Bovens and Joseph Mazor for comments.



  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

  • i See also p. 298.4

Linked Articles

  • Author meets critics: precis
    Frances Kamm
  • Author meets critics: response
    T M Scanlon
  • Author meets critics: response
    Julie Tannenbaum
  • Author meets critics: response
    Frances M Kamm

Other content recommended for you