Lecture 1. Scepticism
Lecture 2. Nature
Lecture 3. Existence
Lecture 4. Perspective
Lecture 5. Rationality
Lecture 6. Language
Lecture 7. Practice
Lecture 8. Politics
Metaethics I: Scepticism
- The lower order and the higher
The constituents of morality/ethics (should we use them synonymously?):
– Thought: concepts, conceptual schemes, outlooks, theories
– Phenomenology: experience, sentiments
– Discourse: language, vocabulary, phrasing, idiom, metaphors
– Practice: actions, behaviours, habits, traits, childrearing, socialisation, disciplinary mechanisms
Styles of metaethical reflection: introspection, observation, interpretation
Sources of metaethical questioning:
– Metaphysics: Are there such things as moral properties? Are they real? Are they mind-independent or mind-dependent? Are they natural properties, or reducible to or supervenient on natural properties? Are they supernatural? Are they non-natural?
– Epistemology: Can there be moral knowledge or expertise? Can we acquire moral knowledge by testimony? Can moral judgements be shown to be unjustified because of the conditions in which they were acquired?
– Philosophy of mind: Are moral judgements (more like) beliefs or desires, cognitive or conative mental states? What’s their direction of fit? Must moral judgements motivate?
– Philosophy of language: Can the propositions expressed by moral sentences be true or false? Do moral terms refer? How can we explain the characteristic functions of moral discourse? What can we learn about moral terms by comparing them with (e.g.) slurs?
– Philosophy of logic: Can we legitimately derive an ought from an is? If moral propositions are not truth apt, how can we make inferences from one moral proposition to another?
– Philosophy of science: Is the subject matter of ethics like the subject matter of the natural sciences? Can there be ethical truths independently of perspective? Is the history of ethics like the history of science? Is there ethical progress, and should we conceive of this as the elimination of error?
– Ethics: Does adopting some metaethical position have any ramifications for how we live? E.g. by making us more or less confident in our moral convictions, more or less tolerant, more or less inclined to find life meaningful.
– Political philosophy: Do metaethical theories have ramifications for politics, and vice versa? Could our metaethical views command more or less (say) tolerance of disagreement? E.g. is there a distinctive liberal or Marxist or feminist metaethical position? Do our metaethical views rule out (or in) certain political theories?
- Scepticism about morality
It is some years now since I realized how many false opinions I had accepted as true from childhood onwards, and that, whatever I had since built on such shaky foundations, could only be highly doubtful. Hence I saw that at some stage in my life the whole structure would have to be utterly demolished, and that I should have to begin again from the bottom up if I wished to construct something lasting and unshakeable in the sciences. (Descartes, First Meditation)
Say we begin from the phenomenology of morality (or, indeed, from observing our discourse and practice), and then ask whether we have incurred philosophical commitments just in virtue of having that phenomenology, or of engaging in that discourse and practice. We may end up with the following interpretation:
Moral utterances express propositions; as such, they are truth-apt, i.e. can be evaluated for truth and falsity. There are inference rules governing their proper use. Moral expressions try to refer. Moral predicates pick out properties of actions, people and states of affairs. Moral propositions have truth conditions. Sometimes, those truth conditions are satisfied because the properties picked out by the predicates of moral propositions exist and are instantiated in the action or person mentioned in the subject. These properties exist mind-independently, and the demands expressed by such propositions are thus categorical. So at least some, possibly very many, moral propositions are true.
Several questions arise:
– Is this the correct interpretation of the phenomenology/practice/discourse of morality? Must we incur such heavy philosophical commitments?
– If it is, are those commitments true, or justified?
Assume for the moment that we answer ‘yes’ to the first question (we’ll return to it later in the series). What would it be to answer ‘no’ to the second? Consider this sceptical voice (one of many such voices we might channel):
Some of our beliefs about morality – our metaethical beliefs – are just explications of aspects of our phenomenology. But this phenomenology is non-veridical, illusory. It points towards an order of reality that just isn’t there, however compelling, attractive or comforting it may be to think that there is. That phenomenology may well be the product of an Evil Demon – or something importantly like it, e.g. ruling-class ideology, patriarchy, Christianity, colonialism, a vindictive superego…
- Moral error theory
Consider the position termed ‘moral error theory’. The central idea is that our ethical discourse (etc) is just what it seems to be: we have ethical beliefs, they can be true or false. The thing is: they’re all false.
Is this ridiculous? Consider some analogies. How seriously would we take a scientific or mathematical error theory? Not very, says David Lewis, in this striking passage.
I’m moved to laughter at the thought of how presumptuous it would be to reject mathematics for philosophical reasons. How would you like the job of telling the mathematicians that they must change their ways, and abjure countless errors, now that philosophy has discovered that there are no classes? Can you tell them, with a straight face, to follow philosophical argument wherever it may lead? If they challenge your credentials, will you boast of philosophy’s other great discoveries: that motion is impossible, that a Being than which no greater can be conceived cannot be conceived not to exist, that it is unthinkable that anything exists outside the mind, that time is unreal, that no theory has ever been made at all probable by evidence (but on the other hand that an empirically adequate ideal theory cannot possibly be false), that it is a wide-open scientific question whether anyone has ever believed anything, and so on, and on, ad nauseam? Not me!
What is Lewis’s point here (apart from the rhetoric)? And does it apply to morality? Two sets of questions we may ask:
(1) What, if anything, makes it ‘presumptuous’ for philosophers to go around telling mathematicians, or for that matter scientists, they must change their ways? Suppose we distinguish here between philosophers trying to correct mathematicians’ first-order mathematical judgements and trying to change their self-understanding. The former may well be presumptuous, but is the latter equally so? And in the case of morality, why may philosophers not challenge either ethical practice (in general, or some particular practice) or some theoretical understanding of it?
(2) What exactly is a ‘philosophical’ reason, and what could constitute a ‘philosophical reason to reject morality’?
- Responses to error theory
Reaction: Moral error theory is false.
Conservativism: Moral error theory is true, but not very important. We can carry on doing what we’ve been doing, though we may have to give a slightly different account of what that is.
Reformism: Moral error theory is true, but only a little important. We can mostly carry on doing what we’ve been doing, but we’ll have to change a few things here and there, and we certainly have to give a different account of what we’re doing and why.
Radicalism: Moral error theory is true. Life will never be the same. How could it?
Next week: Naturalism and its discontents
– Bernard Williams, ‘Naturalism’ in Truth and Truthfulness (Princeton: Princeton University Press, 2002), 22–27. [PDF]
– Stephen Darwall, Allan Gibbard and Peter Railton, ‘Toward Fin de siècle Ethics: Some Trends’, The Philosophical Review, Vol. 101, No. 1, Philosophy in Review: Essays on Contemporary Philosophy (Jan., 1992), pp. 115-189. [JSTOR]
Metaethics II: Nature
- Naturalism in the history of metaethics
The distinction between ethics and metaethics is relatively new in the history of western philosophy, but this doesn’t mean it’s always a mistake to read this distinction back into the history of philosophy. The question ‘What is (e.g.) Aristotle’s metaethics?’ is not obscure.
Throughout this history, one position – in its many varieties – has been dominant: naturalism. Aristotle, Hume, Bentham, Mill, Nietzsche – all can be labelled naturalists of one kind or another. As opposed to this, we can count certain other positions: the divine voluntarism of any number of medieval and early modern philosophers; and certain forms of rationalism, most powerfully developed in Kant.
- Defining naturalism
Naturalism, if it is to include even the five thinkers above, will need to be a broad church indeed. Let’s adopt, as a starting approximation, a formulation given (and possibly endorsed?) by Bernard Williams:
Naturalism is a general outlook which, in relation to human beings, is traditionally, if very vaguely, expressed in the idea that they are “part of nature” — in particular, that they are so in respects, such as their ethical life, in which this is not obviously true. [Truth and Truthfulness (Princeton: Princeton University Press, 2002), 22]
Why insist on naturalism about some domain even if it’s not ‘obviously true’? What, in other words, is the motivation for being a naturalist? Simon Blackburn sets out the most common motivation succinctly:
To be a naturalist is to see human beings as frail complexes of perishing tissue, and so part of the natural order. It is thus … to refuse any appeal to a supernatural order… So the problem is one of finding room for ethics, of placing ethics within the disenchanted, non-ethical order which we inhabit, and of which we are a part… ‘Finding room’ means understanding how we think ethically, and why it offends against nothing in the rest of our world-view for us to do so [Ruling Passions: A Theory of Practical Reasoning (Oxford: Oxford University Press, 1998), 49]
Philip Kitcher has a slogan that tries to capture one upshot of this outlook:
There are to be no spooks. [Philip Kitcher, Preludes to Pragmatism: Toward a Reconstruction of Philosophy (New York: Oxford University Press, 2012), 304]
And the slightly more detailed, and less inclusive, account given by Kitcher, of what it is to do philosophy in a naturalistic way:
Naturalistic philosophy should begin from the picture of the world collectively provided by these diverse investigations [zoology, anthropology, art history, etc], and, where it extends that picture, it should do so in ways that accord with the methods and standards of current inquiry, or with methods that would convince current investigators that they improved on the contemporary versions. [Ibid]
Note the care of Kitcher’s formulation. First, it isn’t reductive: it doesn’t say, for instance, that everything comes down to physics. Second, it isn’t mindlessly conservative: it allows for reform under certain conditions. Third, it gives us a basis for assessing proposed reforms: do these reforms ‘accord with the methods and standards of current inquiry’? Or, do they accord with some putatively improved methods, as long as those methods could ‘convince current investigators that they improved on the contemporary versions’?
Naturalism as we’re construing it is, then, a metaphysical thesis combined with an epistemological one. More specifically, it’s an ontological claim, about what exists (or what sorts of things exist), and a methodological claim about how we should think about ontological questions and the relationship between philosophy and other areas of human thought.
[NB. If you’re interested in this area more generally, the Stanford Encyclopedia entry on ‘Naturalism‘ by David Papineau is extremely helpful, as is the entry on ‘Naturalism and Physicalism’ [PDF link] in Robert Barnard and Neil Manson, eds., A Companion to Metaphysics, Continuum Publishing, 2012.]
- Expanding ontology: Moore and the Open Question argument
Part of the reason for being careful in how the central naturalistic commitments are described is to avoid the now well-known arguments against less well-armoured formulations. The best known anti-naturalist arguments in twentieth-century metaethics are those in GE Moore’s Principia Ethica, trained at the multiple claims he convicted of ‘the naturalistic fallacy’. The style of naturalism he was attacking was (probably) the one presupposed in the utilitarianism of Bentham, Mill and (maybe) Sidgwick, identifying the good with pleasure – where one of the advantages of this identification is that pleasure is (supposedly) readily explicable in naturalistic terms.
This style of theory is naturalistic in two ways: it falls out of a broader, naturalistic worldview; all three figures, recall, were atheists of one sort or another, unusually for their time. The beginning of Bentham’s Principles of Morals and Legislation is revealing: ‘Nature has placed mankind under the governance of two sovereign masters, pain and pleasure. It is for them alone to point out what we ought to do, as well as to determine what we shall do.’ It also makes a very specific claim identifying goodness with, or reducing it to, some natural property. The refutation of the latter doesn’t by itself undermine the whole naturalist outlook. It could, instead, undermine or unsettle ethics (as many naturalists in fact want to do). Or, less frighteningly, it could point towards a reinterpretation of ethics – specifically, of ethical experience and discourse – so that it’s naturalistically innocuous.
The central argument against ‘moral naturalism’, naturalism in the narrow sense, is what has come to be called the Open Question Argument – already anticipated briefly in Sidgwick’s The Methods of Ethics but made in detail in GE Moore’s Principia Ethica. To state the argument in its crudest form: take the naturalist claim to be x is good iff x is N (where N is some natural property). It is an open question about any x that is N (grant this, arguendo) whether x is good. The question has an ‘open feel’. Cut to conclusion: it cannot be the case that the good is identical with or reducible to any natural property.
The upshot of this is that we have to choose between three possibilities: either the good (and with it, every other ethical property) is nothing, because only natural entities are anything. Or it is something, but not, in the strict sense, a (mind-independent) reality. Or it is real, but non-natural. The third option is the one Moore himself took.
Non-natural is not the same as (what is pejoratively called) supernatural. All non-naturalists need to make something like this distinction, though it is surprisingly hard to do so in a principled way, because the matter comes down to that vexed question of exactly what nature is. (Good luck with answering that! There’s an excellent and somewhat dispiriting list of attempts catalogued in the introduction to Michael Ridge’s Stanford Encyclopedia entry on ‘Moral Non-Naturalism‘.)
A popular way of marking the difference is to make it a matter of their role in causal explanations. In claiming that there are non-natural entities, one isn’t claiming that they have any part to play in causal explanations. But someone claiming there are supernatural entities (e.g. ghosts) typically does claim that. Admittedly, this is mostly a matter of terminological fiat, and it’s not clear how much rests on it. Nevertheless, it’s clear enough that non-naturalism on any construal involves expanding ontology beyond the natural. Does this meet the standards Kitcher (et al) have set for this?
Most philosophers in the decades after the Principia didn’t think so. So the course of mid-century metaethics was not Moorean. His kind of non-naturalism has only very recently come back into fashion, with the work of Derek Parfit crucial in reviving it as a viable position. The views that proliferated in the interim were ‘anti-realist’ positions, which took Moore to have shown that good is not a natural property, but accounted for this in terms of its not being a property at all, and therefore, not the kind of thing one could have beliefs or knowledge of.
This is one way of making sense of the dominance of ‘non-cognitive’ approaches to metaethics in this period: the ‘emotivism’ of AJ Ayer and CL Stevenson, the ‘prescriptivism’ of RM Hare, and later, the ‘expressivism’ of Simon Blackburn and Alan Gibbard, and then a variety of styles of ‘fictionalism’ and ‘pragmatism’ associated with figures such as Richard Joyce (fictionalism), Huw Price, Robert Brandom, Richard Rorty (pragmatism). In a different way, it also points us towards the ‘Kantian constructivism’ of Christine Korsgaard and the ‘Humean constructivism’ of Sharon Street. The labels can be overwhelming, but it can be helpful to think of them as a series of unfolding responses to the slow rise to dominance of broadly naturalistic styles of thinking on the one hand, and the way in which everyone agreed that Moore was ‘on to something’, however unclear it was what his arguments really showed. What, if anything, was Moore on to?
4. Consequences of non-naturalism
Moore’s OQA has generated a vast discussion, but this isn’t the moment to go into it. Let’s instead consider what it would mean if Moore is right about ‘non-naturalism’. This term could denote the (only slightly interesting) semantic thesis ‘evaluative predicates can’t be analysed into non-evaluative ones’. It could denote the epistemological, and methodological thesis, that fundamental moral truths are self-evident and can be gained by ‘intuition’. It could also, and usually does, denote the metaphysical thesis that moral properties are real, and neither identical with nor reducible to any natural property or set thereof. One could believe all three, but the metaphysical thesis is the most interesting; what are its implications?
One methodological principle that follows naturally (if not logically) from non-naturalism about ethical metaphysics is this: moral philosophy is autonomous. It needn’t, indeed shouldn’t, defer to the natural sciences, however widely conceived. It is sui generis, and (surprise!) philosophers – as opposed to zoologists, or ethologists, or psychologists or anthropologists – are uniquely well-positioned to do it. At any rate, no one else is, in virtue of their natural or social scientific expertise any better placed to do it. (A mischievous thought: Does this fact itself constitute a reductio of non-naturalism?)
Next week: The ontology of morality; companions in guilt and innocence
– J. L. Mackie, ‘The Subjectivity of Values’ in Ethics: Inventing Right and Wrong (New York: Penguin Books, 1977), 15–49. [PDF]
– John McDowell, ‘Values and Secondary Qualities’, in Morality and Objectivity, ed. Ted Honderich (Boston: Routledge and Kegan Paul, 1984), 110–29. [PDF]
Metaethics III: Existence 
- Naturalism and ontological parsimony
Recall Nelson Goodman’s dictum:
You may decry some of these scruples and protest that there are more things in heaven and earth than are dreamt of in my philosophy. I am concerned, rather, that there should not be more things dreamt of in my philosophy than there are in heaven or earth. [‘The Passing of the Possible’, in Fact, Fiction, and Forecast (Cambridge, Mass.: Harvard University Press, 1983), 34.
Let’s take the underlined part of the quotation as a provisional statement of a principle of parsimony. Note that one of the casualties of that formulation may be ‘heaven’ itself. (If you’re interested in this area in its own right, Alan Baker’s Stanford Encyclopedia entry on ‘Simplicity’ is a good place to start.)
Now consider the following quotation from the first chapter of JL Mackie’s Ethics: Inventing Right and Wrong:
[…] ordinary moral judgements include a claim to objectivity […] [T]his assumption has been incorporated in the basic, conventional, meanings of moral terms. Any analysis [naturalist, non-cognitivist, etc] of the meanings of moral terms which omits this claim to objective, intrinsic, prescriptivity is to that extent incomplete [Harmondsworth: Penguin, 1977), 35]
Mackie starts from a feature of our ordinary moral thought, discourse and phenomenology. He proposes an interpretative hypothesisis: this thought includes a claim to objectivity. This comes out in our approach to moral disagreement, the pattern of our moral sentiments (blame, outrage, resentment), and is written into the very nature of our use of moral language, which has exactly the same structure as can be seen in our non-moral discourse. Like our non-moral discourse, it comes with ontological commitments, which can be true or false. They are false, and so, therefore, is the rest of our moral discourse.
- JL Mackie’s argument(s) for error theory
The so-called argument from queerness is only one of several arguments Mackie gives for doubting the existence of ‘objective value’. The shape of his master argument is nicely summed up in his conclusion, and can be given briefly here:
(1) Interpretation of moral discourse. Ordinary moral discourse involves a claim to objectivity.
(2) Rejection of charitable interpretation of this discourse. This claim to objectivity is not captured in either naturalist analyses (on which the properties denoted by our moral language are, or are reducible to, natural properties) or in non-cognitive analyses (on which no properties are denoted by our moral predicates whatever).
(3) Burden: Anyone claiming that ordinary moral discourse is systematically false must needs argue for this claim as it is so radically at odds with common sense. This burden can be met with multiple arguments.
(4) Relativity: The starting points of moral thinking seem to depend on actual ways of life.
(5) Queerness: Moral properties of the kind presupposed in ordinary moral discourses would have to be intrinsically action-guiding and motivating; these are metaphysically peculiar features to possess.
(6) Supervenience: It is unclear how objective values could supervene on natural features.
(7) Epistemology: It is difficult to account (in naturalistic terms) for our knowledge of value entities.
(8) Alternative explanations of moral belief: The objective commitments of moral discourse can better explained in terms of familiar patterns of ‘objectification’ than by accepting the truth of those commitments.
The best, simplest, most parsimonious explanation of our moral experience is one in which its central commitment (to the existence of objective value) is unnecessary. There is, therefore, no need to accept it.
Of these it is Mackie’s ‘argument from queerness’ that has most interested his readers. The rough shape of the argument is as follows:
If there were such things as objective values (rightness, goodness, etc – moral or otherwise), they would be things of a very strange (‘queer’) sort, utterly different from anything else in the universe. There is no adequate account of the existence of such things, nor of how we could come to know about them. We should not believe in objective values.
Objective values are values that transcend the contingencies of human ends, institutions and practices; they are supposedly, as we say, mind-independent.
Here is one go at reconstructing (one of) Mackie’s arguments as a valid syllogism:
P1. If some putative entity would not figure in the best, i.e., naturalistic, account of the world, then we should not believe in its existence.
P2. Objective values would not figure in the best, i.e. naturalistic, account of the world.
C. We should not believe in the existence of objective values.
To introduce the first of many ‘isms’ we shall encounter in these lectures: if we take the term ‘realism’ – specifically, ‘moral’ or ‘value realism’ – to be the view that ‘Objective values exist (and we should believe in them)’, then Mackie, in denying this view, is defending one of many possible varieties – perhaps the strongest and most provocative variety – of ‘anti-realism’.
Suppose we conceded P1 for argument. There would still be the task of showing that objective values, and similar entities, couldn’t be accommodated in a naturalistic worldview. How does Mackie try to do this? By drawing attention to the allegedly essential features of objective values that make them incompatible with (‘no-spooks’) naturalism.
The term ‘queer’ is a bit of a red herring. Mackie has nothing against ‘queer’ entities as such: indeed, science and other naturalistically respectable inquiries such as mathematics are full of them (quarks, numbers, spin…). But he thinks there is reason in each case to admit those intuitively strange entities into our ontology. The problem with objective value is that it’s irredeemably queer, queer in a way that has no counterpart in a naturalistic worldview.
Lots of people – even lots of non-philosophers – feel this sense of ‘queerness’ acutely. There’s something very odd about the kinds of things we seem to be positing in our moral discourse. One of the many virtues of Mackie’s presentation of the argument is that he doesn’t stop at saying this. He tries to state as explicitly as possible what’s odd about the idea of objective value, and the naïvely realist picture of morality that depends on it. This is good, because this allows to engage with specific allegations on their merits, rather than (the philosophical equivalent of) a whispering campaign.
- Four strands of queerness: three metaphysical and one epistemological
The general metaphysical strategy: pick up on some putatively essential feature of objective value; show that nothing that exhibits such features exists, or is likely to exist.
(1) Argument from the impossibility of being attitude-independent and necessarily motivating
Two allegedly essential features of objective values. (1) Being objective, they exist independently of our attitudes. (2) To judge correctly what is objectively valuable entails that one is motivated to, e.g., pursue it.
P1. Objective values must exist independently of our attitudes and be necessarily motivating.
P2. There is nothing that is both necessarily motivating and independent of our attitudes.
C. There are no objective values.
(2) Argument from the impossibility of objective prescriptivity
An alleged feature of objective values: Objective values are objectively prescriptive – they give us reasons independently of our contingent desires and ends.
P1. Objective values must be reason-giving independently of our contingent desires and ends.
P2. Nothing is reason-giving independently of our contingent desires and ends.
C. There are no objective values.
(3) Argument from supervenience
Naïve observation: If I judge one action to be right and another to be wrong, then it makes sense to ask ‘what makes this action right/wrong?’ To put it in less naïve language, objective values seem to supervene on natural or descriptive facts.
From this, we can formulate a (sort of) trilemma:
(A) Supervenience. Objective values require a descriptive base. (Why? If not, there is no plausible explanation for why situations differ in value.)
(B) No entailment. There is no logical entailment of the objective values by the descriptive base. (Why? If there were, the values would be explanatorily redundant.)
From (A) and (B), we can derive:
(C) The same natural facts that make an action (say) obligatory in the actual world could make that action impermissible in a different possible world.
(C) is extremely unattractive. But it seems to be implied by (A) and (B). So what should we do?
– Reject (A) and be left with an explanatory hole?
– Reject (B) and render values redundant?
– ‘Bite the bullet’ and just accept (C)?
(4) Argument from the unknowability of objective values
The general strategy of the epistemological strand of the argument from queerness is to pick up on an allegedly essential feature of objective values and say that given those features, we have no way of knowing about such values, and thus shouldn’t believe in them.
Objective values are allegedly irreducibly distinct from the natural entities posited by the natural sciences; they are not part of the ‘causal nexus’ dealt with by the sciences. Question: how then could we know about them?
The senses? No: have you ever seen/heard/smelt/touched/tasted an objective value? Inference? But inference from what? They are, by stipulation, causally inert. Then do we have some special faculty – something like ESP? – that acquaints us with these entities? Maybe: it was believed for a considerable part of European history that we did (‘conscience’, ‘intuition’). But isn’t the idea of such a faculty somewhat – dare we say – queer?
P1. If we are to have knowledge of objective values, we must have some special faculty (‘evaluative intuition’) that connects us with them.
P2. We do not possess any such special faculty.
P3. We cannot have knowledge of objective values [even if they do exist]. (From P1 and P2)
P4. We should not believe in the existence of entities we cannot have knowledge of.
C. We should not believe in the existence of objective values. (From P3 and P4)
- Responses to arguments from queerness: companions in guilt strategies
The arguments above, once they’ve been laid out explicitly, offer dozens of opportunities for disagreement, either with the validity of the inferences they involve, or with one or more of their premises. We can’t go into all of them today, but we can look at one general strategy philosophers unmoved by Mackie have used to deal with his kind of argument. Call this, following Mackie himself, ‘companions in guilt’ strategies.
The basic structure of such arguments, expressed loosely, is this:
P1. If we reject objective values, morality etc, then parity of reasoning requires us to reject objectivism about one or more of the following: epistemology, mathematics, the mind, logic, philosophy…
P2. We should not (cannot?) reject objectivism about those things.
C. Therefore, we shouldn’t reject objectivism about value/morality either.
Obviously, the prospects of arguments of this kind depend on the companion chosen, and the extent to which that companion is in fact innocent: otherwise P2 simply won’t be plausible.
Example 1: Take a particular argument above, the argument from the impossibility of objective prescriptivity. Roughly: Objective values must be objectively prescriptive; nothing is objectively prescriptive; therefore, there are no objective values.
By parity of reasoning, someone making this argument would have to say: ‘The laws of logical inference are objectively prescriptive; nothing is objectively prescriptive; therefore, there are no laws of logical inference.’
But this looks like a reductio. Of course there are laws of logical inference! If that’s so, then there’s at least one set of objectively prescriptive things: laws of logical inference. In which we must reject the premise of the original argument which said that there aren’t any. If there can be one set of objectively prescriptive things – viz. laws of logical inference – then why not think there could be one more – viz. objective values? Maybe objective prescriptivity is ‘queer’, but if this argument succeeds, it isn’t queer in the really objectionable sense that requires us to exclude anything claiming to exhibit it from our considered ontology.
Example 2: Take the epistemological argument in the previous section. If objective values are causally inert, we would have to possess a special, and unlikely, faculty of intuition to have knowledge of them.
But here’s a companion in guilt: logical truths. They’re just as causally inert as truths about objective value. Does that mean we can’t have logical knowledge? Or that we can only have such knowledge through some special faculty of ‘logical intuition’? Of course not. We can have logical knowledge because we can recognise the truth of elementary logical statements as self-evident. Maybe truths about objective value are similar: indeed the central argument of Sidgwick’s The Methods of Ethics (1874) is that there are self-evident moral truths from which a version of utilitarianism can be deduced.
- Value as a secondary quality?
One potentially objectionable aspect of Mackie’s naturalism is his commitment to regarding as ‘objectively real’ only those properties that are mind-independent. But we needn’t follow him along this path. Objectivity and mind-independence are not the same thing: moral judgements might be objective even if moral properties aren’t mind-independently real.
Mackie’s expectations of moral objectivity are, more or less, of the worldview Hilary Putnam characterised as ‘external realism’. This worldview can be expressed – as it is helpfully summarised in the first chapter of Tim Button, The Limits of Realism (Oxford: Oxford University Press, 2013) – in three core commitments:
The Independence Principle. The world is (largely) made up of objects that are mind-, language-, and theory-independent. (8)
The Correspondence Principle. Truth involves some sort of correspondence relation between words or thought-signs and external things and sets of things. (8)
The Cartesianism Principle. Even an ideal theory might be radically false. (10)
But should we accept these background assumptions? And should they be the test of whether our moral discourse is either systematically false, or radically deceived about its own nature?
Take the old Lockean distinction between primary and secondary qualities. The distinction is one of mind-, or more precisely response-dependence. Mackie wants to deny to value the status of a primary quality. Suppose we just concede that, and not just for the sake of argument? Could value not be a secondary quality, like – to take the classic example – colour? Mackie needn’t deny it; it may even be his basic point, as long as we agree with him that value being a secondary quality means it can’t be objective.)
John McDowell denies this. As he puts it:
A secondary quality is a property the ascription of which to an object is not adequately understood except as true, if it is true, in virtue of the object’s disposition to present a certain sort of perceptual appearance: specifically, an appearance characterizable by using a word for the property itself to say how the object perceptually appears. [John McDowell, ‘Values and Secondary Qualities’, 111]
What more is there to being red (say) than appearing red in standard conditions? Similarly, we might ask, what more is there to being objectively valuable than seeming valuable – or being the object of a pro-attitude – to well-placed or competent adjudicators in circumstances favourable to forming appropriate attitudes? (This last phrase does a lot of work for McDowell.)
McDowell denies that objectivity requires realism or response-independence; what he thinks it needs is a robust distinction between the illusory and the veridical, a ‘seems-is’ gap. If we have that, we have all we need to vindicate morality against a sceptic in Mackie’s mould.
This can work as a response to Mackie’s epistemological concerns as well: the way in which we can know about objective values is by knowing about how competent judges would respond to the world in favourable circumstances. For McDowell, this is objectivity enough. Is it really?
 I have gratefully drawn in several places in this lecture on the notes and handouts of my predecessor in this role, Dr Christopher Cowie.
Next week: Perspectival and non-perspectival knowledge; science and ethics
– Bernard Williams, ‘Knowledge, Science, Convergence’ in Ethics and the Limits of Philosophy (Abingdon: Routledge, 2006 ), 132–155. [PDF] See also the helpful short commentary on that chapter by AW Moore, pp. 216–19 of the 2006 Routledge edition. [PDF]
– Hilary Putnam, ‘Objectivity and the Science–Ethics Distinction’ in Nussbaum and Sen (eds.), The Quality of Life (Oxford: Clarendon Press, 1993). [Oxford Scholarship Online]
– Simon Blackburn, ‘The Absolute Conception: Putnam vs. Williams’ in Practical Tortoise Raising (Oxford: Oxford University Press, 2010). [Oxford Scholarship Online]
Metaethics IV: Perspective
- Recap: Naturalism and ontology
So far, we have, with JL Mackie, been thinking about the basic questions of metaethics in terms of whether or not there exists such a thing as ‘objective value’. Mackie powerfully argues for the view that we shouldn’t believe in such a thing on broadly ‘naturalistic’ grounds: there is no room for such entities in a naturalistic worldview.
In response, we’ve considered three lines of thought. First, that this argument may well end up proving (or rather, disproving) too much. Its bar for what’s naturalistically respectable may well end up excluding such eminently respectable, and philosophically indispensable things, as logical inference rules and reasons for belief. Second, this argument tends to conflate (or wrongly associate) objectivity with mind-independence. This may be a mistake: maybe, as John McDowell – who may be a termed a ‘sensibility theorist’ – suggests, value could be objective and mind-dependent in the manner of such ‘secondary’ qualities as colour. Third, this argument ignores the possibility that ethics could be objective in the (lesser?) sense that there is still a robust distinction between veridical and illusory perception (cognition?), and so, something could still count as misperceiving, and therefore, as perceiving aright. Isn’t this objectivity enough?
- Ethics compared to science
Pessimism: If anything is objective, science is. If anything is real, it is physical reality. This indubitable objectivity shows up ethics for what it is: irreducibly perspectival, and to that extent, less objective. Ethics and science are, in other words, importantly disanalogous. This is the truth in what used to be called the ‘fact-value distinction’. (Very roughly, the position defended by Bernard Williams.)
Optimism: Science isn’t as objective as all that, but that’s fine. If ethics isn’t objective either, it doesn’t count any more against ethics than it counts against science. We need be more no more anxious about ethics than we need be about science. Ethics and science are, in other words, importantly analogous. There is no fact-value distinction, really. Values, like other facts, are made no less objective by their dependence on facts about us, our minds, our psychologies, our concepts, our perspectives. Science is ethics’s companion in guilt, that is to say, its companion in innocence. (Less roughly, the position defended by Hilary Putnam and McDowell.)
- The ‘absolute conception’: Williams on the aspirations of science and ethics
Williams’s central thought is this: both to claim that ethics (or more specifically, ethical thinking) fails to meet some standard of objectivity and to deny this, we first need some idea of a kind of thinking that does meet that standard of objectivity. Then we can ask whether ethics could ever hope to meet this standard.
Williams thought that science aspires, and may in some conditions achieve, a representation of the world ‘as it is anyway’: anyway, that is, independently of us. More precisely, science aims at a representation of the world minimally dependent on the perspective from which it is produced. Moreover, this representation may be capable of explaining the possibility of that perspective itself, as well as other perspectives on the world. It is, in a sense, a secularised version of an old Cartesian idea: that the very idea of knowledge, and of science as a unified body of knowledge, requires the idea of the world as it appears to God.
To use his own words:
we should form a conception of the world that is “already there” in terms of some but not all of our beliefs and theories. In reflecting on the world that is there anyway, independent of our experience, we must concentrate not in the first instance on what our beliefs are about, but on how they represent what they are about. We can select among our beliefs and features of our world picture some that we can reasonably claim to represent the world in a way to the maximum degree independent of our perspective and its peculiarities. The resultant picture of things, if we can carry through this task, can be called the “absolute conception” of the world. In terms of that conception, we may hope to explain the possibility of our attaining the conception itself, and also the possibility of other, perspectival, representations. [Ethics and the Limits of Philosophy (Abingdon: Routledge, 2006 , 138–9]
As Simon Blackburn puts it, you could imagine that ‘rational Martians would, if intelligent and scientific enough, come to share with us the scientific framework we employ, deploying thoughts about spatial configuration, temporal passage, velocity, mass, energy, electric charge […].’ But it’s not safe to assume that ‘they would taste as we do, smell the same smells, feel heat or cold as we do, or respond to colours in our way.’ Still less can we assume ‘they would have anything like our moral sensibilities or our political or normative sensibilities […] Everyone knows that in these areas variations of sensibility are to be expected.’ [Simon Blackburn, ‘The Absolute Conception: Putnam vs. Williams’, in Practical Tortoise Raising (Oxford: Oxford University Press, 2010), 247]
- Objectivity and the possibility of convergence
A simple way into Williams’s subtle and demanding argument is this: for our thinking in some area to be objective, there must be some prospect of our reaching principled agreement in that area. We could all come to think the same thing, for the right reasons, and those who failed to think the same way would be, to put it bluntly, wrong. Williams’s word for this is ‘convergence’, specifically, the convergence of beliefs (or more generally, of our judgements – where these also cover non-belief attitudes).
The crucial passage from Williams is this:
In a scientific inquiry there should ideally be convergence on an answer, where the best explanation of the convergence involves the idea that the answer represents how things are; in the area of the ethical, at least at a high level of generality, there is no such coherent hope. … The point of the contrast is that, even if this [convergence] happens, it will not be correct to think it has come about because convergence has been guided by how things actually are, whereas convergence in the sciences might be explained in that way if it does happen. [Ethics and the Limits of Philosophy, 136]
The mere fact of convergence isn’t enough. We could all come to (say) love something – mint chocolate chip ice-cream, Paul Rudd – without thinking this evidence of their objective loveable-ness. Similarly, English has become over the last century the language of international science; no one thinks this evidence that English is the language of scientific reality itself.
There is a social-scientific dimension to all this. Williams popularised the term ‘thick ethical concept’, a concept that is, in his definition, ‘at the same time world-guided and action-guiding’ (Ethics and the Limits of Philosophy, 141). Here are some such concepts: ‘honourable’, ‘treacherous’, ‘chaste’, ‘scab’, ‘bluestocking’, ‘uppity’, ‘cringe-making’, ‘mansplaining’.
Williams think we can (sometimes) converge in our ethical beliefs, and these beliefs may even constitute knowledge, when they involve thick ethical concepts. (‘I believe, I know, that he’s unchaste, etc.) There can be disagreement, and agreement, on whether someone’s behaviour falls under one of these concepts.
Among people who have the concept of chastity, there may be a correct answer to the question ‘Was he/ his behaviour chaste?’ and this would involve a combination of description and evaluation. But this doesn’t mean that people must have the concept of chastity in the first place. A person, or culture, could well be able to carry on without it.
It’s not impossible that there could be a world in where everyone had the concept of (say) chastity – as, for instance, may well be true of the concept of a ‘human right’. But the explanation for this, plausibly, would be historical, political and sociological. ‘Why do these people converge in their beliefs about chastity?’ we may ask. A plausible answer would be, ‘Because they live in a society informed by such-and-such religious tenets.’ A much less plausible answer would be, ‘Because they have shared insights into the nature of chastity.’
The general explanatory principle here is that any explanation for converge in ethical beliefs in some community cannot invoke concepts which are themselves (partially) constitutive of being in that community. The explanation needs to be from a vantage point outside of the social world and its assumptions.
But here we have a contrast with scientific explanation. Why is there so much convergence about (say) the structure of the oxygen atom? Here, Williams thinks, the answer could plausibly be: ‘Because the people converging are suitably sensitive to truths about oxygen.’
- Science as a companion in guilt?
One way of responding to Williams on this point is by trying to show that the perspectival-aperspectival distinction doesn’t really map on to the ethics-science distinction. One way of doing that is by showing how much scientific knowledge is perspectival. Putnam makes some important observations in this connection.
First, he stresses just how much science is contested – this as a way of weakening the assumption that there is some easy way for scientists to reach perspective-independent representations of reality. Second, they point out that scientific evidence, which is often theory-laden, frequently underdetermines theories, and that available bodies of evidence are consistent with multiple theories. There is no simple route in science from evidence to theoretical truth without recourse to notions of ‘theoretical virtue’: simplicity, elegance, practicality, etc. In other words, science – less fancifully described – displays many of the features that ethics displays.
One representative passage that illustrates both Putnam’s argument and the style of his rhetoric:
[…] without the postulate that science converges to a single definite theoretical picture with a unique ontology and a unique set of theoretical predicates, the whole notion of ‘absoluteness’ collapses. It is, indeed, the case that ethical knowledge cannot claim ‘absoluteness’; but that is because the notion of ‘absoluteness’ is incoherent. [Hilary Putnam, ‘Objectivity and the Science–Ethics Distinction’ in Nussbaum and Sen (eds.), The Quality of Life (Oxford: Clarendon Press, 1993), 149–50]
Why should only ethics be found to fall short of true objectivity?
- Rational underdetermination?
John McDowell thinks that the epistemologies of science and ethics are sufficiently similar for their twin claims to objectivity to stand or fall together; and of course, he thinks they should both stand. As he puts it:
The only similarity there needs to be between ethics and science, for ethical realism properly understood to be acceptable, is that in both of them it can be rational to say of a conclusion that logos [reason?] itself compels it […] [John McDowell, ‘Two Sorts of Naturalism’, in Mind, Value, and Reality (Cambridge, Mass.: Harvard University Press, 1998), 186]
To think otherwise is to be in thrall to a crude scientism that still hasn’t shaken off the silliness of logical positivism and its rhetorical cousins. (‘Realism’ in McDowell’s – confusing? polemical? – usage may be an attempt to shake off some of the term’s misleading ontological commitments.)
Say we grant that there are in ethics, as there are in science (and more to the point, mathematics), truths that are ‘compelled by logos’. As David Wiggins puts it, we understand the idea that ‘There is nothing else to think’ about some matter. The worry is that this doesn’t get us very far against the style of scepticism it is supposed to answer.
Say we grant to Putnam and McDowell that there is no such thing as ‘the world’ independently of any evaluative perspective. Can’t there still be such a thing as the world as it appears independently of any historically contingent evaluative perspective? (And wasn’t this what Williams was trying to say in the first place?)
Marco Polo on his travels saw many things – some social practices must have puzzled him, others less so, but he must have seen the same sky as the Chinese, the same birds, the same trees. And when he met other people, he could recognise their different bodies as human bodies, their distinctive political system as political systems, their unfamiliar languages as languages. And no doubt, he could recognise their unusual ethical systems as ethical systems, with a (to him) strange way of looking at the world. But what kind of pressure did their ethical conceptions place on him?
Did he observe some Chinese custom, or judgement, or concept, and conclude that those judgements were ‘compelled by logos’? Or – if he could have formulated the thought in these terms – that logos seriously underdetermined any interesting ethical claims? It just doesn’t look like the beliefs involved any concrete form of ethical life are externally constrained by how the world is, but there is such an external constraint on scientific beliefs.
- The slide to relativism
If the Chinese he met ate (say) snakes, it would be hard to sustain his Venetian beliefs that snakes were inedible – even if he didn’t then want to eat them. If the Chinese he met practised, and preached the virtues of, footbinding (as Marco Polo in fact claimed), he may well have rejected the idea, but would have been hard-pressed to refute his Chinese interlocutors in ways that they could recognise.
More precisely, as Hallvard Lillehammer puts it,
[…] the concern is that two different cultures may come to describe and evaluate the social world in different and conflicting evaluative terms, with neither perspective having the intellectual resources to debunk the descriptions and evaluations of the other in ways that participants in the conflicting perspective could reasonably come to recognize as compelling [Companions in Guilt: Arguments for Ethical Objectivity (New York: Palgrave Macmillan, 2007), 126]
[T]here is no suggestion that we should try to describe a world without ourselves using any concepts, or without using concepts which we, human beings can understand. The suggestion is that there are possible descriptions of the world using concepts that are not peculiarly ours, and not peculiarly relative to our experience [Descartes: The Project of Pure Enquiry (London: Penguin Books, 1978), 244]
Putnam and McDowell favour a view of ethical thought on which it is, like all other forms of thought including the scientific, is ‘inescapably ours’. As Lillehammer summarises their view:
In ethics, as in science, it is impossible to get outside one’s own skin and view one’s responsiveness to the world from some entirely external, or ‘sideways on’, perspective. […] there is no plausible inference from the mere fact that ethical thought is essentially perspectival to the claim that all ethical thought is inescapably parochial or biased. We can therefore recognize that our ethical conception is inescapably ‘ours’ while remaining confident in its virtue as reasonably informed and reflectively stable. [Companions in Guilt, 125]
This is all very well. But the big question now becomes: who are ‘we’? How specifically, or generally, should this category be conceived? Are we modern western liberals? Human animals? Rational beings? Something yet more general?
- Motivations for constructivism
One family of views in metaethics embraces the Putnam-McDowell view of ethics as ‘inescapably ours’, but gives that idea a distinctive spin. In one prominent case, the spin is Kantian, in the other, Humean. This family of views is ‘constructivism’, and it is to it that we shall turn next week.
Next week: Two eighteenth-century approaches to the problem: Kantian and Humean constructivism
– Sharon Street, ‘What Is Constructivism in Ethics and Metaethics?’, Philosophy Compass 5, no. 5 (2010): 363–84. [PDF]
– Christine Korsgaard, ‘Realism and Constructivism in Twentieth-Century Moral Philosophy’, Journal of Philosophical Research APA Centennial Supplement (2003): 99–122. Reprinted in The Constitution of Agency: Essays on Practical Reason and Moral Psychology (Oxford: Oxford UP, 2008). [PDF]
Metaethics V: Rationality
- Recap: Realism, naturalism, perspective
Realism: The world is made up of objects that are mind-, language-, and theory-independent; truth involves some sort of correspondence relation between words (etc) and external things; even an (epistemically) ideal theory might be radically false.
Naturalism: We ought not to include entities in our ontology unless they are already used in explanations in natural (or maybe even social) sciences, or can be shown to help us improve on the explanations available in our best science.
Ethical ‘properties’ don’t seem to (need to) be part of scientific explanations. Why not? A hypothesis: they are irreducibly perspectival. In that sense, they are not – in the strong sense – real. Does that mean they can’t be objective? Maybe because our ethical judgements will always be ours, and merely ours, and objectivity is all about getting away from what is ‘merely ours’ to a point of view that could potentially be anyone’s.
- The dialectical position
Bernard Williams: In the sciences, we can coherently imagine ourselves coming to agree in all our scientific judgements, and further, that we are doing is converging on a description of the world as it is ‘anyway’ – an ‘absolute conception’ that is minimally dependent on a perspective peculiar to ‘us’. Our colour concepts are peculiar to us, given our perceptual faculties; but the concept of light-wavelengths is less so. We and some hypothetical Martians (who, say, hear or smell light-wavelengths) may be unable to share colour concepts, but we could still share the concept of light-wavelengths.
Hilary Putnam / John McDowell: Even if we grant that, why think that such convergence couldn’t occur for some special set of ethical concepts, and for some beliefs involving those concepts? If such convergence were to occur, wouldn’t that be objectivity enough?
Williams: Maybe, but the best way to understand these beliefs being objectively true would be ‘they were the beliefs that would help us to find our way around in a social world which […] was shown to be the best social world for human beings.’ [Ethics and the Limits of Philosophy, 155] Without this, the perspectival nature of ethics and ethical conceptual schemes means we are doomed to relativism – variation without the possibility of principled convergence on something we can regard as ethical truth.
- Normative concepts as solutions to practical problems
Christine Korsgaard: Williams’s ‘view of the relations between science and ethics represents a rather deep confusion about the difference between knowledge and action — something almost amounting to a failure to tell them apart.’ [‘Realism and Constructivism in Twentieth-Century Moral Philosophy’, 110] In other words, Williams is faulting practical reason for not being like theoretical reason. But maybe that’s the point: they’re essentially different.
Williams/Putnam/McDowell: How so? Surely in both cases, our concepts purport to represent a reality, and some of them do so better than others.
Korsgaard: No, that is not (really) so. Ethical, and more generally normative, concepts are not best thought of as ‘the names of objects or of facts […] that we encounter in the world. They are the names of the solutions of problems, problems to which we give names to mark them out as objects for practical thought.’ 
Think of John Rawls’s distinction between the concept of distributive justice and competing conceptions of distributive justice. The concept arises out of this fact, that ‘people join together in a cooperative scheme because it will be better for all of them, but they must decide how its benefits and burdens are to be distributed. A conception of justice is a principle that is proposed as a solution to the distribution problem, arrived at by reflecting on the nature of the problem itself. The concept refers to whatever solves the problem, the conception proposes a particular solution.’ [115–6]
Williams/McDowell/Putnam: But this is just another way of describing the worry, not a solution to it.
Korsgaard: Look at it this way. You all agree that we can’t really step out of our skins in ethics. Ethics is essentially perspectival. The worry is how it could be both perspectival and objective, how our ethical judgements and concepts could be ours without being ‘merely’ ours. To this I propose that once we conceive of ethical concepts as solutions to a practical problem, our practical problem, we have a way of understanding objectivity. ‘If you recognize the problem to be real, to be yours, to be one you have to solve, and the solution to be the only or the best one, then the solution is binding upon you.’ [116, emphases added]
This gives us what we were seeking, an idea of what objective truth in ethics could be: ‘the correct conception of a concept will be a guide to its correct application, and when a concept is applied correctly, what we get is truth. But what makes the conception correct will be that it solves the problem, not that it describes some piece of external reality. […] our use of the concept when guided by the correct conception constructs an essentially human reality […] that solves the problem from which the concept springs. The truths that result describe that constructed reality.’ Hence the label ‘constructivism’.
- Constructivism as proceduralism?
Sharon Street: ‘Normative truth consists in what is entailed from within the practical point of view. The subject matter of ethics is the subject matter of what follows from within the standpoint of creatures who are already taking this, that, or the other thing to be valuable. In response to the question “What is value?” constructivism answers that value is a “construction” of the attitude of valuing.’
The ‘bumper sticker’ version of constructivism: ‘no normative truth independent of the practical point of view’. This is importantly distinct from the popular but misleading alternative, ‘no normative truth independent of procedure’. Where certain constructivists – Rawls, TM Scanlon – use procedures, they are not meant to constitute ethical truth; they are heuristics, epistemic devices, to reveal what does or doesn’t follow from within ‘the practical point of view’.
- Two styles of constructivism: Kantian and Humean
Kantian (metaethical) constructivism (e.g. Korsgaard): ‘no matter what the particular substantive content of a given agent’s starting set of normative judgments, moral values are entailed from within that agent’s standpoint, whether the agent ever recognizes that entailment or not.’
Humean (metaethical) constructivism (e.g. Street): ‘the substantive content of a given agent’s reasons is a function of his or her particular, contingently given, evaluative starting points. […] “pure practical reason”—in other words, the standpoint of valuing or normative judgment as such—commits one to no specific substantive values. Instead, that substance must ultimately be supplied by the particular set of values with which one finds oneself alive as an agent—such that had one come alive with an entirely different set of evaluative attitudes, or were mere causes to bring about a radical shift in those attitudes, one’s reasons would have been, or would become, entirely different.’
Kantian constructivism, if it can be made to work, can vindicate universalism about (moral) reasons, from a fairly minimal, formal conception of practical reason. Humean constructivism doesn’t claim this for itself. To the extent that it can vindicate universal reasons, this will be a contingent matter.
- The Caligula Problems
Street: ‘[…] consider the hypothetical case of an ideally coherent Caligula. At least on the face of things, it seems easy to imagine someone who values above all else, and in a way that is utterly consistent with his own other values plus the non-normative facts, torturing others for fun.’
We may want to say that even if there could be such a one, he would be acting wrongly, i.e. in ways he has (moral) reason not to act. Can we? It depends on how we answer the more basic question: ‘Does normative truth outrun entailment from within the practical point of view?’
Moral realist: Yes. And that explains how Caligula could be acting wrongly.
Kantian metaethical constructivist: No. But there cannot be an ideally coherent Caligula. Despite appearances, he is forgetting something, or ignorant of some relevant non-normative facts. So he is acting wrongly by his own lights, though he may not know it.
Humean metaethical constructivist: No. So if there could be an ideally coherent Caligula, he would not be acting wrongly by his own lights (and there are no other lights).
The realist has to face up to all the problems with realism we’ve discussed so far. The Kantian constructivist has to find some principled way of arguing against the possibility of an ideally coherent Caligula. The Humean constructivist has to tell us why their deeply counterintuitive and revisionary commitment doesn’t constitute a reductio of their view – maybe by ‘debunking’ the intuition that seems to tell against their position.
Next week: Expressivism, fictionalism
Simon Blackburn, ‘Antirealist Expressivism and Quasi-Realism’ in David Copp (ed.) The Oxford Handbook of Ethical Theory (Oxford: Oxford University Press, 2007). [URL]
Metaethics VI: Language
- Interpreting moral discourse
Metaethics, among other things, is concerned with the status of moral discourse, phenomenology and practice. To keep things simply, let’s just consider the first: moral discourse (broadly, the things we say in morality). Two questions:
(1) What does what we say in morality commit us to saying about morality? (The hermeneutical question)
(2) Are these commitments true? (The metaphysical question)
Some possible replies to (1):
(3) Our moral discourse presupposes moral realism (i.e. is committed to the mind-independence of moral properties; a correspondence theory of moral truth; the possibility of radical error on moral matters). (Hermeneutical realism)
(4) Our moral discourse, properly interpreted, presupposes no mind-independent entities. It is a kind of make-believe, though one we take very seriously, and which doesn’t wear its make-believe character on its sleeve. (Hermeneutical fictionalism)
(5) At a deep level, our moral discourse does not even claim to state facts – despite the surface of moral discourse. None of it is really truth-apt, just the expression of non-cognitive attitudes or commands or something else of the kind. (Hermeneutical emotivism/prescriptivism)
(6) Our moral discourse does claim to state moral facts, attribute properties, etc and is therefore truth apt. But there is not very much to being a moral fact, and consequently, not much to our claims being true. (Hermeneutical minimalism)
From (3) we might go in at least two directions in answering (2):
(7) Moral realism: Yes. The presuppositions of our moral discourse are true; so is the discourse.
(8) Moral error theory: No. The presuppositions of our moral discourse are false; so is the discourse.
From (6) we might go in at least two directions, especially in light of a further question, i.e. does the lack of ontological commitment in moral discourse make it like, or unlike, other kinds of discourse?
(9) Local moral expressivism: Our moral discourse is minimally, and only minimally, truth-apt; some of it is true. The rest of our discourse, preëminently, scientific discourse, is more robustly truth apt.
(10) Global expressivism: All discourse, including the scientific, is only minimally truth-apt. There is no ‘out-there’ for any of our discourse to correspond to, a fortiori for moral discourse to correspond to.
- (Local) expressivism: a short history
Moore’s Open Question Argument seems to show that if good is a property, it must be a non-natural property. Some people accept that. Others reject the very idea of a non-natural property. In which case the good must be nothing (no thing?). Where does that leave the discourse that speaks of the good? Maybe it isn’t in the business of stating facts or attributing properties at all – despite appearances.
Enter: AJ Ayer and CL Stevenson (‘emotivists’), RM Hare (‘prescriptivist’). Moral discourse should be given a non-truth-apt interpretation; consequently, the mental states involved in moral judgement are not beliefs or belief-like cognitive states (i.e. they don’t aim to represent the world) but are more desire-like (conative or non-cognitive states). This would, if true, explain something non-naturalism renders obscure: that moral judgements tend to motivate those who hold them.
Worries: These views are committed to denying the existence of moral disagreements (as not really disagreements, as they are not disagreements over matters of fact) or make them impossible to resolve rationally.
Further, if moral judgements aren’t truth apt, they can’t be part of a truth-functional logic, and a number of everyday uses of truth-functional connectives (negation, conjunction, disjunction, entailment…) are no longer permissible. It also makes what seem like unobjectionable moral inferences invalid (the Frege-Geach problem).
How to solve these problems? First, don’t get misled by the surfaces of things. Recall Wittgenstein’s remark: ‘We remain unconscious of the prodigious diversity of all the every-day language games, because the clothing of our language makes everything alike. (Philosophical Investigations, 224)
What are the rules of the language game that is morality?
- Simon Blackburn’s Quasi-Realism
Some desiderata for a metaethical theory:
(1) Either doesn’t posit ‘queer’ (non-natural) entities, or justifies positing them
(2) Can explain moral motivation
(3) Consistent with the phenomenology of moral discourse and practice, and able to explain any departures from the naïve interpretation of this phenomenology
(4) Can allow for genuine moral disagreement, and give a principled basis for rejecting certain disagreements as genuine
(5) Can allow for the fact-stating, property-ascribing surface of moral discourse
(6) Can vindicate the use of moral statements in logical arguments, and allow these statements to retain their meaning in embedded contexts (e.g. in conditionals)
Simon Blackburn proposes that ‘the realistic surface of the discourse does not have to be jettisoned. It can be explained and defended […]. [T]houghts about fallibility, objectivity, independence, knowledge, and rationality, as well as truth and falsity themselves, would be available even to people thinking of themselves as antirealists.’ [‘Antirealist Expressivism and Quasi-Realism’, 154]
Quasi-realism does well on (1) – it is ontologically parsimonious – and on (2), insofar as it is a non-cognitivist view. It is compatible with (‘no spooks’) naturalism in the broadest sense, though it is different in letter and spirit from the view termed ‘moral naturalism’ (it does not claim that moral properties are, or are reducible to, natural properties).
On (3), Blackburn proposes that we understand the ‘mind-independence’ seemingly inherent in moral discourse as not a higher-order claim, but as itself a first-order claim. And the naïve idea that one may be puzzled about a moral question can be accommodated as a pro-attitude to continuing to think about the empirical facts in light of which one forms one’s attitude. Similarly, the thought that there is only one right answer to a moral question can be understood as an (optimistic) attitude to continuing to argue about something. Etc.
Some of these revisionary interpretations are more plausible than others. As you may have noticed, they have the same structure as the anti-realism of a John McDowell. We can ‘recreate’ or create a simulacrum of objectivity without the realism. The hope is that if these interpretations are right, then we can have something very close to genuine disagreement, thereby succeeding on (4).
On (5), it simply has to take a deflationist about properties. Take a ‘property’ like ‘disgusting’. When we say ‘Slugs are disgusting’, we seem to be attributing a property to the slug. But ‘disgustingness’ isn’t a property like ‘being 5 cm long’ is a property. It is plausibly understood as a ‘projection’ of our attitudes on to the slug. The quasi-realist can say something similar about moral properties. (A quasi-realist can also be called, in virtue of this commitment, a ‘projectivist’, though the terms pick out different and separable aspects of the position.)
This leaves the notorious Frege-Geach problem. Can the quasi-realist deal with it without sacrificing the most important part of the view: the denial that moral judgements are truth apt?
- Fractured sensibility
Cut to the chase: the expressivist has two viable options. The second – adopting ‘minimalism’ about truth – will be the subject of next week’s lecture. The first, Blackburn’s earlier proposal, is to try to construct a ‘logic of (non-cognitive) attitudes’ that will – if not exactly mimic the old logic of propositions and their relations – give us a way in which to vindicate moral reasoning without using a truth-functional logic.
Take the standard case: can an expressivist make sense of the validity of the following (innocous!) modus ponens?
(A) Lying is wrong.
(B) If lying is wrong, then getting your little brother to lie is wrong.
So: (C) Getting your little brother to lie is wrong.
Blackburn has a proposal: we should take
the conditional (B) as itself expressing something about the interplay of two first-order attitudes, in this case the disapproval of lying and the disapproval of getting your little brother to do it for you. In effect, the conditional voices a disapproval of any moral system, or sensibility, that contains the first but not the second. Coupled with premise (A), construed as voicing disapproval of lying, a sensibility must then contain second, on pain of being so badly ‘fractured’ that one would not know what to make of the overall combination. The vice seemed sufficiently parallel to classical inconsistency to deliver a satisfactory expressivist theory of the inference, since logically valid inference and avoidance of logical inconsistency are generally regarded as coming to the same thing. [‘Antirealist Expressivism and Quasi-Realism’, 156–7; emphasis added]
Problem: A fractured sensibility sounds like a moral fault, not a logical mistake; this isn’t sufficiently parallel to classical inconsistency.
A lot of ingenious work has gone into coming up with better expressivist versions of logical inconsistency (etc). E.g. Blackburn’s attempt in the late 1980s at drawing on ‘conceptual role semantics’, i.e. understanding the meaning of conditionals in terms of their inferential role, and then showing that his simulacrum plays the same role and should thus be taken to have the same meaning.
Even more elaborately, Allan Gibbard’s work in the early 1990s introduces the idea of a ‘factual normative’ world and gives a complicated account of what is wrong with certain combinations of attitudes.
General worry: So much fuss, when we could just abandon the whole project altogether.
Reply: (1) It’s worth it to have a general theory of reasoning, of which classical inconsistency (etc) is just a special case. (2) It’s this or the error theory. And there lies nihilism!
Another possibility: Minimalism?
Next week: Global expressivism?
– David Macarthur & Huw Price, ‘Pragmatism, Quasi-realism and the Global Challenge’ in Cheryl Misak (ed.) New Pragmatists (Oxford: Oxford University Press, 2007). [PDF]
– Huw Price, ‘From Quasi-Realism to Global Expressivism—and Back Again?’ in Johnson & Smith (eds.), Passions and Projects: Themes from the Philosophy of Simon Blackburn (Oxford: Oxford University Press, 2015). [PDF]
Metaethics VII. Practice
- Minimalism to the rescue
Minimalism: To assert that a statement is true is just to assert the statement itself.
If this is right, then the expressivist can agree with everything the realist says, and even say of them that they are true. To say this is just to express the attitude again.
A question: Does this apply across the board? Shouldn’t we just be expressivists and minimalists about everything?
Blackburn’s reply: This is risky for the expressivist. If we become minimalists about everything, then we lose out on being able to distinguish between mental states that ‘represent’ and ones that don’t. If everything represents, minimally, then there is no substantive distinction between (e.g.) science and ethics. And this is a serious cost to anyone whose basic motivation for expressivism was to be able to make this distinction in a systematic and principled way.
2. Globalising expressivism?
A question: Does this apply across the board? Shouldn’t we just be expressivists and minimalists about everything?
Blackburn’s reply: This is risky for the expressivist. If we become minimalists about everything, then we lose out on being able to distinguish between mental states that ‘represent’ and ones that don’t. Everything represents, minimally. Then there is no important distinction between (e.g.) science and ethics. And this is a serious cost to anyone whose basic motivation for expressivism was to be able to make this distinction in a systematic and principled way, with moral judgements coming firmly on the non-representational side.
Price’s suggestion: ‘the apparent uniformity of assertoric or declarative discourse may well mask a multiplicity of different functions.’ [‘Semantic Minimalism and the Frege Point’, emphasis added]
He suggests we ask some of the following questions: ‘What does assertoric discourse do for us? […] What are the concepts of truth and falsity for? (What function do they serve in the lives of a linguistic community?)’ [‘Semantic Minimalism and the Frege Point’]
3. From Hume to Wittgenstein
From the basic Humean points (we project value into the world and then think we discovered it there; moral judgements are expressions of (conative) attitude), we’re moving to an even more radical late-Wittgensteinian one. Some relevant quotations:
Think of the tools in a tool-box: there is a hammer, pliers, a saw, a screw driver, a rule, a glue-pot, glue, nails and screws. – The functions of words are as diverse as the functions of these objects. [Philosophical Investigations, 11]
We remain unconscious of the prodigious diversity of all the every-day language games, because the clothing of our language makes everything alike. [Philosophical Investigations, 224]
Blackburn’s mistake? As Price has it, Blackburn’s (old) view wrongly concedes that assertion is standardly ‘fact-stating’, and that we need to give an account of moral judgements as being only superficially fact-stating (deep down, they’re attitude-expressing). But suppose we deny the assumption, that there is any one thing that assertion standardly does. That makes the problem disappear before it even gets off the ground!
Price’s alternative. Assertion is a multi-purpose tool; there is no one thing to be said about assertion in general. This means that with morality, as with everything, we should be asking not ‘what does our moral discourse mean?’ (if that involves giving truth conditions in the old-fashioned sense) but rather, ‘what does our moral discourse, with all its assertoric elements, do for us?’
A proposal: ‘[…] “representational” language is a tool for aligning commitments across a speech community’ [‘Pragmatism, Quasi-realism and the Global Challenge’] If this is right, it explains what assertoric/representational language does, and why moral discourse would involve such language, but without incurring any ‘queer’ metaphysical commitments.
It’s not hard to see why Price thinks of this view as a variety (or descendent) of pragmatism (the view of William James, John Dewey, CS Peirce, etc):
Pragmatism thus begins with […] phenomena concerning the use of certain terms and concepts, rather than with things or properties of a non-linguistic nature. It begins with linguistic behaviour, and asks broadly anthropological questions: How are we to understand the roles and functions of the behaviour in question, in the lives of the creatures concerned? What is its practical significance? Whence its genealogy? [‘Pragmatism, Quasi-realism and the Global Challenge’]
The question may well be: ‘Sure, pragmatism’s true. But does it work?’
Next week: Metaethics and politics: Richard Rorty’s ‘ironism’
Recommended reading: Richard Rorty, ‘Private Irony and Liberal Hope’ [PDF]
Metaethics VIII: Politics
Huw Price considers a possible concern that (his, and other varieties of) pragmatism
requires us to distance ourselves from our own linguistic practices, to such an extent that we are able to ask […] why we use those words and concepts in the first place. Many philosophers seem to be troubled by this detached perspective, feeling that it threatens our right or ability to continue to engage in these practices in a meaningful way. The attempt to regard one’s practices ‘from the outside’ thus engenders something akin to agoraphobia—a fear that one is losing touch with one’s values and community. [‘Two Paths to Pragmatism’]
(1) This is not a problem with pragmatism as such; this is a problem we, post-Darwinians, have to face anyway. How do we reconcile the external (naturalistic, etc) image of ourselves with other, internal, perspectives?
(2) The problem is, given our historical situation, inescapable; the only thing we can do is ‘to try to alleviate the symptoms’ – we need to acquire ‘the ability to be untroubled by a certain cognitive distance between one’s theoretical standpoint and one’s everyday linguistic activity.’ [‘Two Paths to Pragmatism’]
(3) Our behaviour in other areas of life give us hopeful models: pleasures and pains are no less pleasurable/painful just because we know their biological basis. We know how to live with dissonance already; the complaint is ‘theatrical’.
We can be anti-realists at the second-order level, but this needn’t unsettle our practice or sentiments at the first-order level. Life will go on pretty much as it always did.
- Richard Rorty and the consequences of pragmatism
Others in the pragmatist tradition have said more on these wider political questions, preëminently, Richard Rorty (1931–2007). Rorty claims to be a descendent of the 19th-century tradition of American pragmatism, particularly as embodied by John Dewey. His work, since his influential Philosophy and the Mirror of Nature (1980), sets itself (like Price’s) against ‘representationalism’. He denies that it is philosophy’s (or indeed science’s) aim to represent the world accurately. There is no ‘vocabulary’ in which the world ‘demands’ to be described (he is fond of such locutions). That means that there is no standpoint from which one can judge whether one is using the correct ‘final vocabulary’, by which he means simply, ‘a set of words which [people] employ to justify their actions, their beliefs, and their lives’. What makes them final? The fact that ‘if doubt is cast on the worth of these words, their user has no noncircular argumentative recourse.’ [‘Private Irony and Liberal Hope’]
However, it is part of the modern condition that we are painfully aware that not everyone’s final vocabulary is the same. Once we become aware of the existence of living alternatives to our final vocabularies, we cannot be bluffly commonsensical and ‘unselfconsciously describe everything in terms of the final vocabulary to which [we] and those around [us] are habituated.’ We can become ‘metaphysicians’, and insist that ‘what matters is not what language is being used but what is true.’ Or – and this is the path Rorty prefers – we can become ‘ironists’. An ironist fulfills these three conditions:
(1) She has radical and continuing doubts about the final vocabulary she currently uses, because she has been impressed by other vocabularies, vocabularies taken as final by people or books she has encountered;
(2) she realizes that argument phrased in her present vocabulary can neither underwrite nor dissolve these doubts;
(3) insofar as she philosophizes about her situation, she does not think that her vocabulary is closer to reality than others, that it is in touch with a power not herself. [‘Private Irony and Liberal Hope’]
- Rational argument versus ‘redescription’
Rorty is not very keen on what is generally thought central to the tradition of western philosophy: the giving of rational arguments. By this stage in his life, he isn’t even all that keen on philosophy – except as one more body of literary texts. We read them as we read other literary texts, and our commentaries on them constitute literary criticism. What we shouldn’t do is to ask such questions as ‘Are these claims true?’
‘Literary citicism does for ironists what the search for universal moral principles is supposed to do for metaphysicians.’ The ironist ‘takes the unit of persuasion to be a vocabulary rather than a proposition. Her method is redescription rather than inference. Ironists specialize in redescribing ranges of objects or events in partially neologistic jargon, in the hope of inciting people to adopt and extend that jargon.’
- Private irony, public liberalism
Irony is all very well in private; but in public life, we must still be liberals. Rorty’s (somewhat idiosyncratic) idea of the liberal is someone who thinks that the worst things human beings do to each other are cruelty and humiliation (the latter is something to which humans are uniquely vulnerable).
Rorty’s worry is that it may be hard to live life with this liberal conviction while being an ironist in private. Is that state of mind sustainable? ‘Believing’ that cruelty and humiliation are bad, while also holding them to be part of one final vocabulary among others? Here, the metaphysician may have the edge:
Metaphysicians tell us that unless there is some sort of common ur-vocabulary, we have no “reason” not to be cruel to those whose final vocabularies are very unlike ours. A universalistic ethics seems incompatible with ironism, simply because it is hard to imagine stating such an ethic without some doctrine about the nature of man. Such an appeal to real essence is the antithesis of ironism.
It gets worse. Even the ironist’s preferred tool, redescription, can itself act as a device of humiliation:
Ironism … results from awareness of the power of redescription. But most people do not want to be redescribed. They want to be taken on their own terms […] The redescribing ironist, by threatening one’s final vocabulary, and thus one’s ability to make sense of oneself in one’s own terms rather than hers, suggests that one’s self and one’s world are futile, obsolete, powerless. Redescription often humiliates.
Part of Rorty’s response here is to challenge the metaphysician’s concern with reasons. Why is it so important to have a rational basis for not being cruel to other human beings? Isn’t it enough to know what things are cruel? The expectation of justifications of such things is, Rorty thinks, a consequence of a ‘metaphysical upbringing’. The pragmatist has to, and can only, replicate the authority of the metaphysician piecemeal:
novels and ethnographies which sensitize one to the pain of those who do not speak our language must do the job which demonstrations of a common human nature were supposed to do. Solidarity has to be constructed out of little pieces, rather than found already waiting, in the form of an ur-language which all of us recognize when we hear it.
- Questions for Rorty
(1) Can Rorty coherently say what he’s saying? Isn’t he just occupying the very transcendental standpoint whose existence he denies? It may be more coherent for Rorty to try (like the Wittgenstein of the Tractatus?) to show something that, by his own lights, cannot be said. Of course it is possible that he simply lacks the literary talent of a Nietzsche. He sees what redescriptions need to be effected; he is simply not the one to effect it.
(2) Is Rorty’s hope of liberal irony merely complacent? Can we now (rather than in the decade of Gorbachev) sustain a private irony along with liberal, or indeed, some other non-liberal, political commitments? Specifically, what exactly does pragmatism imply for confrontations between liberal and non-liberal societies?
(3) The paradox of pragmatism: ‘Of course pragmatism is true; the trouble is that it doesn’t work.’ (attrib. Sidney Morgenbesser) If pragmatism is true, and what matters is what works, then how should pragmatists respond to the fact that what works is realism, metaphysics, etc? If we’re to be effective pragmatists, does that mean being robust realists?
- Confidence and contingency
Rorty thinks we can acknowledge the contingency of our moral and political values without our commitment to them weakening. But it is an open question – and largely an empirical one – to what extent this is possible. Bernard Williams takes it to be a problem about whether our practices can be ‘sustainable under reflection’, whether they can survive if they’re made transparent to us. Does an understanding of why we give our practices authority undermine or strengthen our sense of their authority? (See, esp, Chs VIII, IX of Ethics and the Limits of Philosophy and Ch IX of Truth and Truthfulness.)
This points towards a more critical, rather than complacent, understanding of liberalism, one that has more room in it for such regrettably metaphysical notions as truthfulness. If any kind of liberalism is worth believing in (and maybe, in 2017, it no longer is), Williams suggests that it will be because it is more capable of withstanding critical reflection than anything else. He suggests that what’s needed isn’t irony but a deeper self-understanding, so that we can say, confidently, of our outlook (liberalism or whatever takes its place),
this outlook is ours just because of the history that has made it ours; or, more precisely, has both made us, and made the outlook as something that is ours. We are no less contingently formed than the outlook is, and the formation is significantly the same. We and our outlook are not simply in the same place at the same time. [‘Philosophy as a Humanistic Discipline’]