Eight Lectures, weeks 1-8
Faculty Board Room, Faculty of Philosophy (3rd floor, Raised Faculty Building, Sidgwick Avenue)
Students in Part II of the Philosophy tripos; MPhils, graduates and other visiting students welcome
This is a course of eight lectures on themes in recent (analytic, English-language) metaethics, delivered principally for students studying for Part II Paper 3 Ethics in the Philosophy tripos, University of Cambridge.
These lectures are not introductory and will not be suitable as someone’s first encounter with philosophy. While I shall try to explain all technical terminology and references to historical figures, I shall be presupposing some previous philosophical background: roughly, a good understanding of basic concepts in logic, metaphysics and ethics, and a broad-brush knowledge of the history of western philosophy. Anyone attending who is looking for an elementary introduction to metaethics should attend the IA lectures on Metaethics instead.
Different lecturers have different styles. My own is heavily influenced by an essay of Henry Sidgwick’s, ‘A Lecture Against Lecturing’, especially by his view that any lecture that simply replicates material that could just as well be printed as a book is useless. Moreover, I will not be delivering lectures that can act as substitutes for private reading and study. Anyone aiming to do exams on this course should be prepared to master a much wider range of material, and in greater depth, than I shall be covering in these lectures.
The point of these lectures is different. Firstly, they’re meant to be synoptic rather than fine-grained: that is to say, they aim to provide a survey of a large and growing literature and to make connections between different arguments and themes. They’re designed to help you to place the things you’ll be reading into a wider (philosophical, historical, intellectual) context, not to offer potted summaries of the things you’ll be reading. Secondly, they’re designed to raise questions rather than provide answers: a good part of learning to do philosophy involves getting puzzled by things, and seeing how deep some problems can go, and just what stands in the way of solving them. At the conclusion of each lecture, I’d like you to leave not with the answer, but with a sense of what needs to do to come up with a satisfactory answer. There will be many opportunities to ask questions during the lectures.
There is, by now, an enormous literature on metaethics. No series of eight lectures can possibly address all of it. I’ve been highly selective, choosing debates that still have some life to them, i.e. are still raging, and for good reason. Other things being equal, I’ve gone with questions I’m actually bothered by personally and philosophers I myself engage with. This has meant leaving a good deal out, but so would any other series of lectures on this topic.
Schedule with set readings
Papers marked § are required; other papers are important background or classic texts.
Week 1. Morality, reasons and rationality
Peter Strawson, ‘Freedom and Resentment’, in Proceedings of the British Academy, Volume 48: 1962, ed. Gary Watson (Oxford: Oxford University Press, 1962), 1–25. [URL]
Bernard Williams, ‘Internal Reasons and the Obscurity of Blame’, in Making Sense of Humanity (Cambridge: Cambridge University Press, 1995), 35–45. [URL]
Week 2. Companions in Guilt
§ Hallvard Lillehammer, ‘Companions in Guilt: Entailment, Analogy and Absorption’, in Companions in Guilt Arguments in Metaethics, ed. Christopher Cowie and Richard Rowland (Abingdon: Routledge, 2019). [PDF]
J. L. Mackie, ‘The Subjectivity of Values’, in Ethics: Inventing Right and Wrong (New York: Penguin Books, 1977), 15–49.
Terence Cuneo, ‘The Parity Premise’, in The Normative Web: An Argument for Moral Realism (New York: Oxford University Press, 2007), 89–114. [URL]
Week 3. Genealogical Debunking
Sharon Street, ‘A Darwinian Dilemma for Realist Theories of Value’, Philosophical Studies 127, no. 1 (2006): 109–166. [JSTOR]
Louise Hanson, ‘The Real Problem with Evolutionary Debunking Arguments’, The Philosophical Quarterly 67, no. 268 (2016): 508–33. [URL]
Week 4. Styles of constructivism
§ Sharon Street, ‘Constructivism in Ethics and the Problem of Attachment and Loss’, Aristotelian Society Supplementary Volume 90, no. 1 (2016): 161–189. [URL]
Christine M. Korsgaard, ‘Realism and Constructivism in Twentieth‐Century Moral Philosophy’, in The Constitution of Agency (New York: Oxford University Press, 2008), 302–25. [URL]
John Rawls, ‘Kantian Constructivism in Moral Theory’, Journal of Philosophy 77, no. 9 (1980): 515–572. [JSTOR]
Sharon Street, ‘Constructivism about Reasons’, in Oxford Studies in Metaethics, ed. Russ Shafer-Landau, vol. 3 (Oxford: Oxford University Press, 2008), 207–46.
Sharon Street, ‘What Is Constructivism in Ethics and Metaethics?’, Philosophy Compass 5, no. 5 (2010): 363–384. [URL]
Week 5. Styles of realism
§ Claire Kirwin, ‘The Metaethical Significance of the Transparent First Person’ (unpublished ms: contact lecturer for a PDF)
David Enoch, ‘The Argument from the Deliberative Indispensability of Irreducibly Normative Truths’, in Taking Morality Seriously: A Defense of Robust Realism (Oxford: Oxford University Press, 2011), 50–84.
Iris Murdoch, The Sovereignty of Good (London: Routledge, 1970).
John McDowell, ‘Values and Secondary Qualities’, in Morality and Objectivity, ed. Ted Honderich (Boston: Routledge and Kegan Paul, 1984), 110–29.
Kieran Setiya, ‘Murdoch on the Sovereignty of Good’, Philosophers’ Imprint 13 (2013).
Week 6. Styles of Expressivism
§ Huw Price, ‘From Quasirealism to Global Expressivism – and Back Again?’, in Passions and Projections: Themes from the Philosophy of Simon Blackburn (Oxford: Oxford University Press, 2015), 134–52.
AJ Ayer, ‘Critique of Ethics and Theology’, in Language, Truth and Logic, 2nd ed. (New York: Dover Publications, 1952 ), 103–25.
Simon Blackburn, ‘How to Be an Ethical Antirealist’, Midwest Studies in Philosophy 12, no. 1 (1988): 361–375.
Richard Rorty, ‘Private Irony and Liberal Hope’, in Contingency, Irony, and Solidarity (New York: Cambridge University Press, 1989), 73–95.
Week 7. Thick concepts
§ Debbie Roberts, ‘Shapelessness and the Thick’, Ethics 121, no. 3 (2011): 489–520.
Bernard Williams, ‘Knowledge, Science, Convergence’, in Ethics and the Limits of Philosophy, Reissue (Abingdon: Routledge, 2006 ), 132–155.
Hilary Putnam, ‘Objectivity and the Science—Ethics Distinction’, in The Quality of Life, ed. Amartya Sen and Martha Nussbaum (New York: Oxford University Press, 1993), 143–57.
Week 8. Moral Deference
§ Andreas L. Mogensen, ‘Moral Testimony Pessimism and the Uncertain Value of Authenticity’, Philosophy and Phenomenological Research 95, no. 2 (2017): 261–284.
Karen Jones, ‘Second-Hand Moral Knowledge’, Journal of Philosophy 96, no. 2 (1999): 55.
Alison Hills, ‘Moral Testimony and Moral Epistemology’, Ethics 120, no. 1 (1 October 2009): 94–127.
Paulina Sliwa, ‘In Defense of Moral Testimony’, Philosophical Studies 158, no. 2 (2012): 175–195.
Lecture I: Morality, reasons and rationality
1. Themes of the lecture
– What is the subject matter of metaethics?
– What is the place of metaethics in the history of western philosophy?
– What explains the pervasive focus in the metaethical literature on rationality and reasons?
2. Why ask these questions?
– To show the continuity of contemporary metaethics with questions that have been asked going back to Plato
– To show how we can motivate the highly complex questions in the contemporary literature on metaethics from ordinary human concerns (esp modern concerns)
3. Ethics and metaethics
– How to mark the distinction? ‘Higher-order’ versus ‘lower-order’; ‘Inside-out’ vs ‘outside-in’; ‘questions in ethics’ vs ‘questions about ethics’; ‘engaged stance’ vs ‘disengaged/alienated stance’; ‘internal perspective’ vs ‘sideways-on perspective’, etc
– Ethics: discourse, thought, language, sentiments, practice
– Two tasks for metaethics: Description (what are we doing when we engage in ethical practice); Evaluation (is our ethical practice justified/rational; are its presuppositions true; are its statements meaningful…)
4. Sources of scepticism in metaethics
– Naturalism: Ethical practice seems to presuppose the existence of entities – goodness, rightness, justice – that don’t obviously belong in a naturalistic ontology, i.e. a scientifically respectable account of what exists in reality. E.g. Darwin, Freud (?), JL Mackie
– Responses: Give up on ethics; deny naturalism; qualify naturalism; give a revised account of ethics consistent with naturalism
– Prudential rationality: If rationality requires prudence (i.e. pursuit of self-interest), then ethical practice (which generally demands the sacrifice of self-interest) seems to be irrational. E.g. Plato’s immoralists (Thrasymachus, Gorgias); Nietzsche; Freud (?)
– Responses: Give up on ethics; deny the primacy of prudential rationality; show ethics to be demanded even by prudential rationality; qualify ethics so that its demands are consistent with those of prudential rationality
5. Metaethics and the study of reasons
– Two reasons to be interested in reasons:
(1) We need to understand what a reason if we are to say if we have reason to be moral (i.e. give a response to the moral sceptic)
(2) Ethics is normative (‘should’, ‘right’, ‘ought’, ‘good’ etc); plausibly, all these other notions can be reduced to the notion of a reason. A reason-for-action could well be a ‘normative primitive’
6. Williams on reasons
– Bernard Williams (‘Internal and External Reasons’, ‘Internal Reasons and the Obscurity of Blame’) moved by both prudential and naturalistic considerations: what, if anything, of our ethical practice will survive if we insist that (1) ethical practice is consistent with our best scientific understanding; (2) ethical practice can be shown to be a requirement of rationality?
– Conclusion: Something will survive, but it won’t be the entirety of what is conventionally called ethics
– A popular view implicit in much ethical practice is the idea that there are universally binding, objective principles that govern the conduct of all human beings and do so irrespective of any particular agent’s beliefs or desires. Williams calls this the view that there are ‘external reasons’. Williams rejects this, claims that all reasons are internal.
– For something to be a reason for an agent to do something, it must have some link to that agent’s motivations.
– Specifically, Williams is trying to give an account not of the metaphysics of reasons (‘what are these things called reasons?’) but rather about the semantics of reason-statements (‘what is the meaning of a statement of the form “A has a reason to F”’? ‘what are the truth conditions of such statements?’)
– He proposes that ‘A has a reason to F’ can be translated to: ‘if A deliberated soundly from his existing motivations, i.e. had access to all relevant empirical facts and didn’t make any mistakes of inference, then A would be motivated to F’.
– Claims that this is a plausible account of what one thinks when one thinks ‘I have a reason to F’ – I mean, ‘If I thought about things properly, given my previous motivations, this is what I’d want to do’.
7. Internalism: costs and benefits
– Internalism is consistent with naturalism: reasons are not some special, ‘queer’ category of things that no scientific instrument can detect; they emerge out of facts about human psychology (general human psychology as well as the particular psychology of particular humans)
– Internalism can explain why most people have reason to be moral, i.e. to take the interests of other people seriously. It’s because most people have motivations (e.g. natural sympathy for other humans, commitments to particular people, internalised sense of shame and guilt, etc) such that they also have reason to do what morality, as conventionally understood, demands.
– But is this enough? Arguably not: internalism cannot give you complete objectivity. It allows for the possibility that some people, given their psychologies, have no reason to be moral. For those people, morality is not a requirement of rationality.
– If there are only internal reasons, morality can’t be everything it seemed to be. For Williams, this is not an objection, just a consequence, and one we have to learn to live with.
8. Manne on internalism
– Kate Manne (‘Internalism about Reasons: Sad but True’) defends internalism using different arguments. Thinks it can better account for how we understand the practice of reason-giving.
– Uses the distinction in Strawson (‘Freedom and Resentment’) between the ‘interpersonal stance’ and the ‘objective stance’. What’s the difference between trying to persuade someone (which involves treating them as potentially persuadable by rational argument) and simply trying to manage their behaviour (when they’re being unreasonable, or are drunk)?
– Manne proposes that we stop trying to give reasons beyond a certain point; some people are beyond the reach of reasons. If we acknowledge this in our practice, then this supports the contention that we are already internalists. We acknowledge that some people, at least some of the time, have no reason to do other than they do, or be other than they are. This is sad, but that doesn’t stop it being true.
– This means that with certain people, attitudes like blaming making no sense (they assume that they had reason to act differently, but if her argument is sound, they don’t). But one can still criticise them in other ways, or condemn them, or ignore them, or ostracise them, or (in extreme circumstances) punish or incarcerate them. But this is because we have reason to do these things, not because they have reason to do what we’d like them to do.
– Reason-giving with such people may serve a practical purpose, e.g. it expresses our values, it puts our objects ‘on the record’, sometimes it may even be self-fulfilling (by falsely asserting that they have a reason, we may end up transforming their psychology so that our assertion becomes true)
9. Internalism and naturalism
– Internalism is, in the end, a particularly sophisticated form of subjectivism. It effectively rules out the possibility that there are objective and binding moral truths that apply to everybody. It may be naturalistically acceptable, but it seems to have won this at the cost of betraying the objectivity many people think essential to morality. To Williams (etc), that’s the point; that kind of objectivity is, for other reasons, not worth having anyway.
– Next week, we’ll consider an influential set of arguments for thinking there to be something suspect about the idea of objective moral demands, reasons that apply to everyone. JL Mackie is broadly motivated by a naturalistic worldview, but develops it into specific arguments, which are designed to show that the idea of objective value, absolute moral demands, etc, all presuppose the existence of facts or entities that are multiply ‘queer’: they have a combination of features that nothing can have.
– Against Mackie’s argument, we’ll consider a growing body of literature that holds that Mackie’s arguments prove too much. In trying to undermine ethics, they end up showing that there is something equally suspect about mathematics, logic, epistemology, etc. But of course, it’s absurd to give up on mathematics and logic on such flimsy grounds, so we should be equally wary of giving up on ethics on the basis of metaphysical arguments. These are called ‘companions-in-guilt’ arguments and we’ll be discussing some of the most sophisticated versions in the current literature.
Lecture II: Companions in Guilt
Questions in metaethics arise from looking at ethical thought, discourse, practice from the point of view of theses in other parts of philosophy. E.g.
Ethics and metaphysics: the nature (if any) of moral facts, mind-dependence, consistency with naturalism or other outlooks
Ethics and epistemology: the possibility of moral knowledge or justification, the nature and reliability of moral intuition, debunking arguments
Ethics and the philosophy of mind: the nature of moral judgement: cognitive or conative, representational or expressive, direction of fit; thick and thin concepts
Ethics and the philosophy of language: the semantics and pragmatics of moral utterances; (more than minimally) truth-apt? emotive? imperative? otherwise expressive? thick and thin terms
Ethics and the theory of rationality: are ethical judgements derivable from more basic principles of (practical) reason? necessary, or contingent on conative states of particular agents?
Ethics and axiology: how is ethical value/evaluation (i.e. goodness) like/unlike other kinds of value/evaluation (e.g. epistemic, aesthetic)
Ethics and normativity: how is ethical normativity (i.e. reasons, obligation) like/unlike other kinds of normativity (e.g. epistemic, logical)
Some questions in metaethics are distinctively about ethics (or morality) and posed while presupposing the legitimacy of other kinds of evaluation/normativity; others are questions about all forms of normativity/evaluation.
The debate about the ‘queerness’ of moral facts, inaugurated by JL Mackie, was originally a debate about the metaphysical status of ethics in particular; ‘companions in guilt’ strategies force us to ask the broader question. Does the supposed queerness of moral facts generalise to other kinds of value/normativity and lead us into a radical scepticism (about epistemic reasons, logic, etc)? If not, then why not use the legitimacy of other kinds of value/normativity to cast light on the queerness strategy itself, and thereby, vindicate the metaphysical status of ethics?
3. Master argument for error theory
JL Mackie’s argument involves two steps:
(P1) Interpretative: Moral claims presuppose the truth of judgements of mind-independent moral facts [i.e. of value not conditional on any psychological fact about some particular agent; in Mackie’s lingo, objective values].
(P2) Queerness: There are no such things as moral fact [e.g. because they’d be ‘queer’].
(C) Error Theory: All moral claims [insofar as they presuppose the existence of objective value] are false.
Mackie takes this argument to be obviously valid, and P1 to be extremely plausible. The action for him is entirely in the argument for P2.
4. Notes on Mackie’s argument
One might challenge P1 by adopting various less metaphysically demanding views on the semantics of moral claims, e.g. expressivism (Blackburn, Price), hermeneutic fictionalism (Kalderon, cf Stanley, Yablo). The former, nowadays, hold that moral judgements are only minimally true, in a way that involves no metaphysical commitment; the latter claim that moral judgements involve some kind of pretence, so can’t be falsified by proof of the non-existence of moral facts.
Even if one thinks the argument is sound, there are more and less radical ways of taking the conclusion. The radical interpretation would be abolitionism: let’s stop engaging in moral discourse. The modest interpretation would be some kind of reformism, i.e. to reinterpret our own discourse in some other way, e.g. expressivist or fictionalist.
5. Argument from queerness: rough formalisations
(P1) Nothing in the universe has the property of being objectively prescriptive. [Naturalistic premise (?)]
(P2) Moral facts, if they were to exist, would have to be objectively prescriptive. [Interpretative premise]
(C) Moral facts do not exist.
(P1) If facts exist, it must be possible to know them. [Knowability premise]
(P2) If moral facts exist, it must be possible to know them. [From P1]
(P2) Moral facts are not knowable through the senses. [Plausible, common ground]
(P3) If something cannot be known through the senses, there must be some extrasensory perceptual faculty by which it can be known. [Plausible: how else?]
(P4) There is no extrasensory faculty that gives us knowledge of moral facts. [Naturalistic premise]
(C) Moral facts do not exist.
6. Styles of counterargument
(1) Naturalism: Moral facts are natural facts; some natural facts can be objectively prescriptive (denies P1 of the metaphysical argument)
(2) Subjectivism: Moral facts don’t have to be objectively prescriptive, e.g. internalists about reasons, ‘hypotheticalists’ (denies P2 of the metaphysical argument)
(3) Rationalism: Moral facts can be known by reason; nothing queer about the extrasensory faculty of reason (denies P4 of the epistemological argument)
7. Companions in guilt: analogy
(P1) Moral facts are, in some relevant sense, like epistemic [or logical or mathematical] facts. [Analogy based on shared normativity]
(P2) Either: moral facts and epistemic facts are both queer or neither moral facts nor epistemic facts are queer. [Symmetry]
(P3) Epistemic facts are not queer. [Innocence of one companion]
(C) Moral facts are not queer. [Vindication of other companion]
(1) Irrelevance: The similarity between moral and epistemic claims is not of a relevant kind (denies P1).
(2) Asymmetry: Moral facts could be queer without epistemic facts being queer (denies P2).
(3) Generalisation: Epistemic facts are also queer (denies P3; requires some further strategy like fictionalism about epistemic facts, or some kind of naturalistic reduction of epistemic facts to facts about evidential support, etc).
8. Companions in guilt: entailment
(P1) If moral facts do not exist, epistemic facts do not exist. [Entailment]
(P2) Epistemic facts do exist. [Indispensability arguments; self-defeating for error theorist to deny epistemic facts]
(C) Moral facts exist.
(1) Epistemic facts are not really indispensable (cf Quine on mathematical abstracta). We can take a fictionalist or reductionist view of them (denies P1).
(2) Even if epistemic facts are indispensable, it does not follow that moral facts are. Argument shows not that all normativity is okay, only that we should only accept those normative domains about which it can convincingly be claimed that they’re indispensable. Moral claims are dispensable, in that we might be able to live, and think coherently, without them.
(3) Equally, it’s somewhat plausible that error theory for epistemic normativity is self-defeating — how else can error theorists argue for their claims unless that can say that we should accept the conclusion once we’ve accepted the premises. But is this equally plausible for moral normativity?
Companions in guilt strategies try to show that arguments like Mackie’s prove too much. In trying to undermine the claims of one aspect of discourse, they inadvertently (and unwisely!) undermine the claims of other aspects of discourse that we cannot do without.
Responses to such strategies try either to deny the analogy, or deny that the analogy does much to show what defenders of moral realism/objectivity need.
General (methodological) moral: We need to think carefully, and separately, about both (1) the ways in which morality is like other normative discourse and (2) the ways in which it is an entirely distinctive kind of discourse. We cannot hope to redeem it (or indeed, undermine it) otherwise.
Lecture III. Genealogical anxiety
To be uploaded after lecture.