John Broome argues that when all claims cannot be perfectly fairly satisfied in outcome, the contribution to fairness from entering claims into a lottery, and so providing them some surrogate satisfaction, ought to be weighed against, and can outweigh, what fairness can be achieved directly in outcome. I argue that this is a mistake. Instead, I suggest that any contribution to fairness from entering claims into a lottery is lexically posterior to fairness in outcome.

If pleasure is more open than pain to a double definition, first as a real sensation, second as a more indirect impression, it is clear that the calculus – the advantages of which Bentham praised so fulsomely − cannot be identical for pleasure and pain alike. Sensations may be combined in the infinitesimal calculus in a substantive way, but this is impossible for the more indirect reflective impressions, which require other sorts of mathematics. For Bentham, it is not a question of eschewing calculation, but of facilitating it, perhaps through a probability calculus in a Bayesian or subjective style. The theory of fictions permits the combination or substitution of the two aspects of pleasure, so that what seems to be an ambiguity in Bentham's approach to pleasure is really an attempt to render the concept useful, that is, capable of utilization in calculations bearing on important areas of practical policy.

While many theories of well-being are able to capture some of our central intuitions about well-being, e.g. avoiding alienation worries, they typically do so at the cost of not being able to capture others, e.g. explaining deprivation. However, both of these intuitions are important and any comprehensive theory of well-being ought to attempt to strike the best balance in responding to both concerns. In light of this, I develop and defend a theory of well-being which holds that our well-being depends, in part, on the nature of our well-being qua person, a class whose members are defined by their possessing certain cognitive and volitional capacities including those capacities constitutive of autonomy. I argue that this ‘person-centred’ theory of well-being is able to address concerns about alienation and deprivation, along with capturing the importance of autonomy to well-being, better than many popular subjective and objective theories of well-being.

Interpersonal aggregation involves the combining and weighing of benefits and losses to multiple individuals in the course of determining what ought to be done. Most consequentialists embrace thoroughgoing interpersonal aggregation, the view that any large benefit to each of a few people can be morally outweighed by allocating any smaller benefit to each of many others, so long as this second group is sufficiently large. This would permit letting one person die in order to cure some number of mild headaches instead. Most non-consequentialists reject thoroughgoing interpersonal aggregation despite also believing it is permissible to let one person die in order to prevent many cases of paraplegia instead. Non-consequentialists defend this asymmetry largely on the basis of intuition, and some rely on the notion of relevance to formalize the grounding intuitions. This article seeks to clarify and strengthen the non-consequentialist notion of relevance by engaging with three objections to it.

The received view that John Stuart Mill opposed the use of violence to attain desirable political goals has been undermined by authors stressing Mill's defence of revolutionary causes during his lifetime and his efforts to outline a justificatory theory of political violence. In light of this scholarship, claims of Mill's ostensible ‘gradualism’ with regard to the appropriate methods and pace of social progress may merit reassessment. At the same time Mill's account appears to sanction violence that respects criteria of justice but not of expediency and vice versa, making it untenable as a cogent guide for carrying out or evaluating acts of violence. That this tension is analogous to tensions elsewhere in Mill's writings provides more evidence for the view that his theoretical project was a systematic one, and raises new questions about his philosophical enterprise.

In this article I attempt to show conclusively that the apparent intrinsic difference between causing collateral damage and directly attacking innocents is an illusion. I show how eleven morally irrelevant alterations can transform an apparently permissible case of harming as a side-effect into an apparently impermissible case of harming as a means. The alterations are as obviously irrelevant as the victims’ skin colour, and consistently treating them as relevant would have unacceptable implications for choices between more and less harmful ways of securing greater goods. This shows not only how the principles philosophers have proposed for distinguishing between these cases cannot withstand scrutiny, but how we can be sure that there are no relevant differences yet to be discovered. I conclude by considering reasons to think that there are deontological constraints against harming, but that they apply just as forcefully against collateral harms as they do against intended harms.