Traditional debates about scientific realism tend to focus on issues concerning scientific representation (broadly speaking) and de-emphasize issues concerning scientific intervention. Questions about the relation between theories and the world, the nature of scientific inference, and the structure of scientific explanations have occupied a central place in the realism debate, while questions about experimentation and technology have not. Ian Hacking's experimental realism attempts to reverse this trend by shifting the defense of realism away from representation to intervention. Experimental realism, according to Hacking, does not require us to believe that our theories are true (or approximately true), nor does its defense depend on inference to the best explanation. For Hacking, the strongest proof for realism is that we can manipulate objects: 'So far as I'm concerned, if you can spray them, then they are real' (ibid., 23).

There is a puzzle about how to understand the conclusion of a successful instance of practical reasoning. Do the considerations adduced in reasoning rationalize the particular doing of an action, as Aristotle is sometimes interpreted as claiming? Or does reasoning conclude in the formation of an attitude – an intention, say – that has an action-type as its content? This paper attempts to clarify what is at stake in that debate and defends the latter view against some of its critics.

Williamsonian modal epistemology (WME) is characterized by two commitments: realism about modality, and anti-exceptionalism about our modal knowledge. Williamson's own counterfactual-based modal epistemology is the best known implementation of WME, but not the only option that is available. I sketch and defend an alternative implementation which takes our knowledge of metaphysical modality to arise, not from knowledge of counterfactuals, but from our knowledge of ordinary possibility statements of the form ‘x can F’. I defend this view against a criticism indicated in Williamson's own work, and argue that it is better connected to the semantics of modal language.

Leading accounts of personal autonomy are content-neutral: they insist that there are no a priori constraints on the content of the desires or values that might motivate an autonomous action. In Gerald Dworkin's provocative words, ‘the autonomous person can be a tyrant or a slave, a saint or sinner, a rugged individualist or champion of fraternity, a leader or follower.’ ‘There is nothing in the idea of autonomy that precludes a person from saying, “I want to be the kind of person who acts at the command of others. I define myself as a slave and endorse those attitudes and preferences. My autonomy consists in being a slave.” ’ John Christman similarly claims that ‘any desire, no matter how evil, self-sacrificing, or slavish it might be’ could be autonomously formed. The same seems to apply to Harry Frankfurt's view, that actions are autonomous if they stem from second-order volitions that reflect what the agent cares about; it puts no constraints on the content of what a person might care about. All of these accounts hold that the mere content of a desire or value is never sufficient to rule out that it might be autonomously acted on by someone.

Analysis of the nature of voting is fundamental to an understanding of the nature and value of democracy. Three questions ought to be distinguished concerning voting. First, how are we to conceive of the activity of voting? What is its nature? Second, what is it that people aim at when they vote? What are the grounds on which people vote, the reasons with which they justify their vote? And third, what ought people to aim at when they vote? What are the relevant grounds for voting in one way or another? These last are explicitly normative questions. At times individuals vote on the basis of concerns which are not appropriate to the issues involved. The first question is not a question of psychology or of value; it is a metaphysical question. What is the nature of voting? How is it possible to count votes? Two answers are widely given in the literature. The first is that voting consists in the expression of preference. On this kind of view, either a vote is a preference or it is a fairly direct and unambiguous expression of the preference of the voter.

By a potential person I shall mean an entity which is not now a person but which is capable of developing into a person, given certain biologically and/or technologically possible conditions. This is admittedly a narrower sense than some would attach to the term ‘potential'. After all, people of the twenty-fifth century, if such there will be, are in some sense potential people now, even though the specific biological entities from which they will develop, i.e. the particular gametes or concepti, do not yet exist. For there do exist, in the reproductive capacities of people now living and in the earth's resources, conditions adequate to produce these future people eventually, provided of course that various possible catastrophes are avoided. Indeed, in some sense of ‘potential’ there have been countless billions of potential people from the beginning of time. But I am concerned not with such remote potentialities but with currently existing entities that are capable of developing into people.

Intellectualists tell us that a person who knows how to do something therein knows a proposition. Along with others, they may say that a person who intends to do something intends a proposition. I argue against them. I do so by way of considering ‘know how ——’ and ‘intend ——’ together. When the two are considered together, a realistic conception of human agency can inform the understanding of some infinitives: the argument need not turn on what semanticists have had to say about (what they call) ‘the subjects of infinitival clauses’.

The distinction between occurrent and non-occurrent mental states is frequently appealed to by contemporary philosophers, but it has never been explicated in any significant detail. In the literature, two accounts of the distinction are commonly presupposed (and occasionally stated explicitly). One is that occurrent states are conscious states. The other is that non-occurrent states are dispositional states, and thus that occurrent states are manifestations of dispositions. I argue that neither of these accounts is adequate, and therefore that another account is needed. I propose that occurrent states are active states.

At the beginning of Book X of his last work The Laws Plato turns his attention from violent and outrageous actions in general to the particular case of undisciplined and presumptuous behaviour in matters of religion: “We have already stated summarily what the punishment should be for temple-robbing, whether by open force or secretly. But the punishments for the various sorts of insolence in speech or action with regard to the gods, which a man can show in word or deed, have to be proclaimed after we have provided an exordium. Let this be it: ‘No one believing, as the laws prescribe, in the existence of the gods has ever yet performed an impious action willingly, or uttered a lawless word. Anyone acting in such a way is in one of three conditions: either, first, he does not believe the proposition aforesaid; or, second, he believes that though the gods exist they have no concern about men; or, third, he believes that they can easily be won over by the bribery of prayer and sacrifice” (§ 885B).

Galileo and his fellowers discovered, and subsequent centuries have amply confirmed, that you get much better predictions by thinking of things as masses of particles blindly bumping each other than by thinking of them as Aristotle thought of them — animistically, teleologically, anthromorphically. They also discovered that you get a better handle on the universe by thinking of it as infinite and cold and comfortless than by thinking of it as finite, homey, planned, and relevant to human concerns. Finally, they discovered that if you view planets or missiles or corpuscles as point-masses, you can get nice simple predictive laws by looking for nice simple mathematical ratios. These discoveres are the basis of modern technological civilization. We can hardly be too grateful for them. But they do not, pace Descartes and Kant, point any epistemological moral.

Seventeenth century discussions of materialism, whether favorable or hostile towards the position, are generally conducted on a level of much less precision and sophistication than recent work on the problem of the mind-body relation. Nevertheless, the earlier discussions can still be interesting to philosophers, as the plethora of references to Cartesian arguments in the recent literature makes clear. Certainly the early development of materialist patterns of thought, and efforts on both the materialist and immaterialist side to establish fundamental points in the philosophical analysis of mind, have considerable historical interest at the present time. This paper attempts to clarify the significance of some of leibniz's views in connection with the materialist thesis. I do not have in mind his rather notorious parallelism, though some of the points made below bear indirectly on the character of this position (or perhaps on the question whether he held it consistently). Instead, I will examine his approach to arguments against materialism.

This article examines the complaint that arbitrary algorithmic decisions wrong those whom they affect. It makes three contributions. First, it provides an analysis of what arbitrariness means in this context. Second, it argues that arbitrariness is not of moral concern except when special circumstances apply. However, when the same algorithm or different algorithms based on the same data are used in multiple contexts, a person may be arbitrarily excluded from a broad range of opportunities. The third contribution is to explain why this systemic exclusion is of moral concern and to offer a solution to address it.

It is becoming more common that the decision-makers in private and public institutions are predictive algorithmic systems, not humans. This article argues that relying on algorithmic systems is procedurally unjust in contexts involving background conditions of structural injustice. Under such nonideal conditions, algorithmic systems, if left to their own devices, cannot meet a necessary condition of procedural justice, because they fail to provide a sufficiently nuanced model of which cases count as relevantly similar. Resolving this problem requires deliberative capacities uniquely available to human agents. After exploring the limitations of existing formal algorithmic fairness strategies, the article argues that procedural justice requires that human agents relying wholly or in part on algorithmic systems proceed with caution: by avoiding doxastic negligence about algorithmic outputs, by exercising deliberative capacities when making similarity judgments, and by suspending belief and gathering additional information in light of higher-order uncertainty.

In this discussion I shall argue that some fairly widely held views about human habits are mistaken. These misconceptions are important because of the pervasiveness of the habitual in human behavior and because it is the concept of habit that has served as the prototype of various conceptions of conditioned response which are used in psychological explanation. One major task of this analysis is to show that accounts in which actions are explained by reference to rules are not incompatible with accounts in which the same behavior is seen as the product of habit.

John Rawls's famous difference principle is capable of at least four distinct statements, each of which occurs in A Theory of Justice. According to what I shall term the Crude Principle it is a necessary and sufficient condition for the justice of an institutional scheme which expands social and economic inequality that, subject to the satisfaction of more weighty principles, it increases the level of advantage of the least advantaged. Expressing this principle Rawls writes that,

There are decision problems where the preferences that seem rational to many people cannot be accommodated within orthodox decision theory in the natural way. In response, a number of alternatives to the orthodoxy have been proposed. In this paper, I offer an argument against those alternatives and in favour of the orthodoxy. I focus on preferences that seem to encode sensitivity to risk. And I focus on the alternative to the orthodoxy proposed by Lara Buchak’s risk-weighted expected utility theory. I will show that the orthodoxy can be made to accommodate all of the preferences that Buchak’s theory can accommodate.

1. Introduction

A paternalistic act is one in which the protection or promotion of a subject's welfare is the primary reason for attempted or successful coercive interference with an action or state of that person. My aim in this paper is to determine the conditions under which such acts are Justified. The route I take is through the concept of consent, with actual consent providing the foundation for a rather complex condition which I claim is necessary and sufficient for the Justification of paternalism.

Because a paternalistic act apparently violates a subject's right to non-interference, it constitutes a problem in the general area of rights. Ideally, then, one should have a developed and convincing theory of rights — that is, a theory which contains an analysis of the concept of a right, and offers solutions to the central problems, e.g. who has rights and why, what considerations override prima facie rights, and how the relative importance of prima facie rights is determined — from which the solutions to various problems about paternalism can be derived.

Automated Influence is the use of Artificial Intelligence (AI) to collect, integrate, and analyse people’s data in order to deliver targeted interventions that shape their behaviour. We consider three central objections against Automated Influence, focusing on privacy, exploitation, and manipulation, showing in each case how a structural version of that objection has more purchase than its interactional counterpart. By rejecting the interactional focus of “AI Ethics” in favour of a more structural, political philosophy of AI, we show that the real problem with Automated Influence is the crisis of legitimacy that it precipitates.

The idea of equality exerts considerable influence on our moral imaginations, yet it has remained philosophically elusive. Although men and women have thought equality worth dying for, philosophers have been largely unable to give any systematic account of its importance as a moral ideal, or of its function in moral and political theory.

It is widely held that we can obtain beliefs and withhold believing propositions directly by performing an act of will. This thesis is sometimes identified with the view that believing is a basic act, an act which is under our direct control. Descartes holds that the will is limitless in relation to belief acquisition and that we must be directly responsible for our beliefs, especially our false beliefs, for otherwise we could draw the blasphemous conclusion that God is responsible for them. For Descartes and his followers judgment and assent are acts of the will which may be made both when they ought and when they ought not to be made. They are expressions of freedom of the will and as such we are directly responsible for the beliefs we acquire. Other philosophers who seem to espouse volitionalism include Aquinas, Locke, Kierkegaard, Newman, James, Pieper, Chisholm and Meiland.