Rahul Kumar, from “Contractualism on the Shoal of Aggregation” in Reasons and Recognition: Essays on the Philosophy of T. M. Scanlon, pages 148-150:
Recall, for instance, a case mentioned earlier concerning the choice of locating a chemical research facility in a large urban center or in a sparsely populated rural area: intuitively, the facility ought to be located in the sparsely populated rural area, though the consequences of an accident are much worse for any of the local population than it is for any of those living in the urban center. Structurally, however, this type of case is different from that of Rocks: in that case, the implications for each of the rescuer’s choice are the same. But in this case, the potential burdens are not equal. And intuitively, it seems that it is the greater number, each [standing] to have to bear the lesser, though comparatively morally relevant, burden that can complain of being wronged if exposed to the risk that the chemical research facility creates.
Though a detailed exploration of how to account intuitions of this type of case using contractualist resources goes beyond what I can undertake here, it is worth briefly considering what kinds of resources might be marshaled for the task. First, it is relevant that the above chemical factory case is one that involves the imposition of a risk of harm rather than the imposition of harm. Those living in the rural area may, therefore, stand to suffer a greater harm if the risk materializes as harm yet be less likely to suffer harm than those living in the urban setting. This may have to do in part with a sense that those living in the rural rather than the urban setting have a better opportunity to avoid being harmed (it being easier, for instance, for them to get out of harm’s way).
Second, though those living in the urban center will not each be as seriously harmed by an accident at the chemical research facility, the significance of their being harmed may not be exhausted by the harm each incurs. The proper functioning of a large urban center could be considered an irreducible, non-exclusive public good: one that cannot in principle be realized without a great many people doing their part and contributed to the lives of both those who participate in realizing it and those who do not. Appealing to the possible implications of a proposed principle for the availability or accessibility of a public good, which may supervene on the fact that a great number of individuals stand to be harmed, does not violate the individualist restriction in the way appealing to the mere fact of the number who stand to be harmed clearly does.
Such a strategy might be used to understand the basis of an intuition to which Scanlon draws attention in his discussion of aggregation, that it might be wrong to save a life rather than saving millions from paralysis or blindness. Depending on how the details of the case are filled out, however, a further, third strategy is available. Say that the one will be saved or millions saved from paralysis or blindness depending on how certain public resources are deployed by state officials. A case might be made on grounds of the duty of the state to concern itself with public health for state resources being deployed in a way that balances the importance to those subject to the state’s authority of having the resource in question deployed in a certain way with the number of citizens who will be benefitted from the resource being deployed in this way. A principle governing state action that is sensitive to the number of persons who stand to be affected is compatible with the individualist restriction as long as the justification of the principle relies not [on] the fact that a great many stand to be benefited, but on an account that connects the interests of a citizens with the state attending to public health concerns in a way that is sensitive to the numbers of persons in question.
[…]At the heart of deontological or non-consequentialist ethical theory, as Christine Korsgaard puts it, is the thought that the ‘subject matter of morality is not what we should bring about, but how we should relate to one another.’ In this respect, Scanlon’s contractualism is avowedly deontological, emphasizing the importance to the morally motivated of standing in a particular type of relationship to all others, one of mutual recognition.
It is one thing, however, to note the general relational character of the contractualist outlook but quite another to keep the importance of the point in view when working through substantive moral questions in a contractualist mode. This is particularly true, I believe, in the case of the discussion of how to make sense, on contractualist terms, of the intuitive relevance of aggregative considerations. In thinking about the relevant kinds of cases, we are naturally drawn to focusing on the possible outcomes of different choices, and from there it is easy to find oneself thinking that what is of central moral importance is what outcome obtains.
[…M]any of the challenges to the plausibility of the contractualist account in its current form that the intuitive relevance of aggregative considerations are thought to present are rooted in attaching an importance to facts about outcomes that is at odds with contractualist thinking. Careful attention to the relational character of the contractualist account […] allows us to both make better sense of contractualism’s rejection of the relevance of aggregative considerations and start to see how the seeming relevance of such considerations in certain cases can be explained using contractualist resources.