Derek Parfit

(1942 - 2017)

This third volume of this series develops further previous treatment of reasons, normativity, the meaning of moral discourse, and the status of morality. It engages with critics, and shows the way to resolution of their differences. This volume is partly about what it is for things to matter, in the sense that we all have reasons to care about these things. Much of the book discusses three of the main kinds of meta-ethical theory: normative naturalism, quasi-realist expressivism, and non-metaphysical non-naturalism, which this book refers to as non-realist cognitivism. This third theory claims that, if we use the word ‘reality’ in an ontologically weighty sense, irreducibly normative truths have no mysterious or incredible ontological implications. If instead we use ‘reality’ in a wide sense, according to which all truths are truths about reality, this theory claims that some non-empirically discoverable truths — such as logical, mathematical, modal, and some normative truths — raise no difficult ontological questions. This book discusses these theories partly by commenting on the views of some of the contributors to Peter Singer's collection Does Anything Really Matter? Parfit on Objectivity.

This book has four loosely connected parts. Part One discusses some ways in which theories about morality and rationality can be self‐defeating. Such theories give us certain aims, but also tell us to act in ways that frustrate these aims. If these theories are revised, these objections can be partly met. Part Two discusses the relations between what a single person can rationally want or do at different times, and what different people can rationally want or do. I also discuss the rationality of four attitudes to time: temporal neutrality, and the three kinds of bias towards the present, the near, and the future. Part Three discusses personal identity, or what is involved in our continuing to exist throughout our lives. Most of us, I argue, have certain beliefs about this subject that are false, but hard to give up. If we accept these claims, they may change some of our emotions or attitudes, and we should revise some of our beliefs about both rationality and morality. Part Four discusses our obligations to future generations, and some related questions about what would be better or worse futures for mankind. The most difficult question here, which I fail to answer, is about the relative importance of the number of people who will exist, and the quality of life of these people.

Discusses numerical identity, or being one and the same, qualitative identity, or being exactly similar, personal identity, or what is involved in our continued existence over time. According to the Physical Criterion, our identity over time consists in the continued existence of enough of our brain. According to the Psychological Criterion, our identity consists in overlapping chains of psychological continuity and connectedness. The chapter discusses how we are inclined to believe that, even in purely imagined cases, our identity must be determinate. When we ask – Would I still exist? Would that future person be me?, it seems that it must always have an answer.

According to the Self‐interest Theory, or S, our own well‐being is the supremely rational aim. According to Consequentialism, or C, the ultimate moral aim is that things go as well as possible. The chapter explains how these theories can be indirectly self‐defeating, in the sense that our trying to achieve these aims may cause them to be worse; how it can be rational to cause ourselves to be irrational, and how it might be right to cause ourselves to be disposed to act wrongly; how these theories might be self‐effacing by telling us to believe other theories; and how these theories do not fail in their own terms.

Discusses Autonomy and Paternalism; becoming and ceasing to be a person, or human being; whether reductionism about persons undermines desert. It examines personal identity and commitments; the separateness of persons and principles of distributive justice – whether we should extend the scope of these principles, and give them less weight, whether the units for distributive principles should be lives, successive selves, or people at times, and how a reductionist view gives some support to the utilitarian rejection of distributive principles.

Explains why C cannot be directly self‐defeating. Theories are agent‐relative if they give different agents different aims. Two such theories are S and Common Sense Morality, or M. It is often true that, if each of several people does what would be best for themselves, that would be worse for all these people. In such cases, S is directly collectively self‐defeating. In moral analogues of such cases, M is similarly self‐defeating. The chapter describes how these problems can have political, psychological, or moral solutions.