Meta-ethics

Non-Cognitivism

The three cognitivist positions we looked at differed in what they thought ethical language was about. But what all three cognitivist positions have in common is their central claim that moral judgements are statements expressing (or attempting to express, in the case of error theory) beliefs about the world, and these statements can therefore be true or false.

Non-cognitivism rejects this claim; it rejects moral realism and also rejects Mackie's error theory, because non-cognitivism holds that ethical language is not fact-stating, moral terms do not attempt to pick out anything in the world, and so they are not truth-apt. Ethical language performs some other function entirely.

EmotivismEmotivism can be seen as a reaction against intuitionism. Remember that Moore thought that we knew what was 'good' through intuition, for example the love of friendship and beauty. Now, among a certain intellectual elite, Moore's examples of 'goods' as intuitive knowledge seemed obviously correct. But to others, what seemed obvious was that what Moore thought of as 'goods' were simply an expression of his own feelings and attitudes. This ‘emotive’ approach was adapted by later philosophers – one of the most well-known of which was A.J. Ayer.

Ayer and logical positivism

In the 1930s, a school of philosophy emerged called logical positivism. The cornerstone of their beliefs was the verification principle. This claims that a statement only has cognitive meaning (i.e. can be said to be true or false and is therefore useful in discussion) if it is either analytic or empirically verifiable.

The roots of emotivism lie in logical positivism. For Ayer, only two types of statement make genuine truth claims (claims that are true or false):

Analytic

“Bachelors are male and unmarried.”

Empirically verifiable

“It’s snowing outside.”

If your statement is neither of these things, then it’s cognitively meaningless. Now, if we apply this to moral judgements we can start to see a problem:

1. Moral judgements aren’t empirical, we can’t prove or disprove them using sense experience. We can indeed show that murder causes grief and pain, or that it is often done out of anger. But we cannot demonstrate, in the same way that it is wrong.

2. Moral judgements aren’t analytic (true by definition). If I say ‘murder is wrong’, there is nothing about the definition of ‘murder’ that entails ‘wrongness’.

If we accept both of these statements as true, then by the Verification Principle moral judgements are meaningless. They do not express truth-apt information.

Ayer agreed with philosophers like Moore, who claimed that moral claims including terms like 'right' and 'good were unanalysable. But unlike the intuitionists and other non-naturalists, he said this is because there is nothing to analyse. 'Good' and right' are what Ayer calls 'pseudo-concepts': they don't refer to anything at all. Ayer instead concludes that moral terms are simply expressions or exclamations of our emotions, like going 'Boo!' (at things we don't like) or 'Hurrah!' (at things we do like).

NB. Don’t confuse emotivism with subjectivism. Emotivism holds that moral judgements express feelings but don’t report truths about feelings. Emotivists are not arguing that moral judgements are our way of saying ‘I disapprove of this action’, something that may be true or false, they are saying moral judgements literally express our feelings or emotional response towards a particular event. It is effectively (as above) the same as ‘Boo’ or ‘Hurrah’, or pulling a particular face when I say something – none of these things hold truth value.

SUBJECTIVISM “I DISAPPROVE OF STEALING”

REPORTING FEELINGS

FACTUAL

HAS TRUTH VALUE

(EITHER TRUE OF FALSE)

COGNITIVE

E.G. “STEALING IS WRONG”

EMOTIVISM

(radical subjectivism)

“STEALING BOO!!” EXPRESSING FEELINGS

NOT FACTUAL

HAS NO TRUTH

VALUE (NOT TRUE OR FALSE)

NON-COGNITIVE

For example - Suppose that you say “Brrr!” as you shiver in the cold. “Brrr!” isn’t literally true or false; it would be out of place to respond to it by saying “That’s true”.

Now suppose that you say – “I feel cold”. Here you’re saying something true (or possibly false) – since you do feel cold. A moral judgement for emotivists is like “Brrr!” (which just expresses your feelings), and not like “I feel cold” (which is a truth claim about your feelings).

The persuasive character of moral judgements

The American emotivist philosopher C. L. Stevenson went a step further than Ayer in analysing the emotive meaning of moral judgements. He argued that moral judgements which employed terms like 'good' and 'right' were not only expressions of a feeling. More importantly, thought Stevenson, they were also attempts to influence other people, to persuade them to feel as we feel and to have the same attitude that we have. So Stevenson might say that when we claim 'Abortion is wrong' we effectively mean 'Boo to abortion. You should feel this way too'. In saying this Stevenson is able to give an account of how moral terms motivate or guide action - they do so like someone shouting or urging us to do something, they motivate through the power of the emotion behind the words.

Attractions of emotivism It does away with the mysterious 'non-natural' properties of intuitionism. It also

connects moral judgement with conduct in an intelligible way, which, as Mackie’s arguments for error theory showed, intuitionism cannot.

It explains why there is such diversity of moral viewpoints among human beings –morality is all to do with feelings, and feelings differ from person to person.

It could lead to greater tolerance, because if there is no right and wrong in moral questions we will learn to respect the views and moral judgements of others.

Criticism - The verification principle is unverifiableThe main difficulty with logical positivism is that according to the principle of verification, the principle of verification itself is meaningless. The claim that ‘a statement only has meaning if it is analytic or can be verified empirically’ is not analytic and cannot be verified empirically. But if the principle of verification is meaningless, then what it claims cannot be true. So it does not give us any reason to believe that the claims of ethics are meaningless. We will discuss this in more detail during A2 Religious Language.

Prescriptivism

R. M. Hare’s (1919-2002) prescriptivism can be seen as a development of emotivism, in that it further explores the uses and purposes of moral judgements. Like emotivism, prescriptivism is a non-cognitivist, anti-realist theory which denies that values are types of facts, and denies that moral statements have truth-value. But rather than expressing feelings, the purpose of value judgements is to tell other people how they ought to act. When a doctor gives you a prescription they are recommending a course of action, and similarly our moral judgements are a form of prescription to a certain behaviour.

For Hare, moral language can be broken down into two ways of prescribing action:

Imperatives to act – Most common terms are ‘right’ and ‘wrong’.

Using moral language like this implies that someone should do something or act in a particular way. For example ‘Eating meat is wrong’ entails the imperative ‘Don’t eat meat’. If you accept the moral judgement then you will act accordingly. This means if you ask ‘should I eat meat?’, and I answer ‘eating meat is wrong’, I have answered your question.

Value Judgements – Most common terms include ‘good’ and ‘bad’.

Using moral language like this is intended to commend or criticize something. Although this commendation is not explicit about what we should do, it does serve as guidance for our choices later.

So when I say ‘Ameerah is a good student’ I am commending Ameerah for something that she has done and implying she should probably continue to do it in future.

By commending or prescribing something we are making a value judgement, we are saying the thing we are commending is praiseworthy in some way. This is different from describing something – Suppose I say ‘This is a good strawberry, because it is sweet and juicy’. If we think ‘good’ as applied to strawberries is just a matter of whether they have particular qualities (sweet and juicy) then all I have said is ‘This is a sweet and juicy strawberry because it is sweet and juicy’.

But this does not seem to be what I mean when referring to the strawberry as ‘good’, I am not merely describing some extra quality it has instead I am commending it for having that particular quality.

This difference between describing and commending is what Hare thinks Hume got right with his Is-Ought gap and Moore got right with the naturalistic fallacy.

Moral judgements are very different types of thing than factual statements about the world, they operate in a different way and are intended to do something different. Realists (naturalists in particular) are wrong to claim them as simple facts.

Standards

Another key point of Hare’s theory is that he believes we make moral judgements (for example calling something ‘good’) based on whether or not the thing in question adheres to a set of standards. ‘Good’ chocolate for example might be of a certain quality and taste, whilst ‘Good’ teachers might be ones that enthuse the students and get them interested in the subject.

For Hare, a good person (as in morally ‘good’) is just someone who is the way we think we ought to be as people. They have achieved a standard that we consider to be important.

These standards though are not absolute, they’re not something that everyone has access to and should follow (as a virtue ethicist might claim) instead we have to adopt them. Your idea of a ‘good’, commendable person might be different to mine.

Universalisability

Since, in describing something as good, I am appealing to a particular set of standards, it would be illogical if I had two identical things to call one ‘good’ and the other ‘not good’. Either they both achieve the standards, or neither does. If I say one chocolate is good and the other is not, I must be appealing to some key difference between the two.

This also extends to moral imperatives – terms like right and wrong are intended to guide someone towards a particular standard. Two actions, in similar circumstances must both therefore be right or wrong. To have one right and the other wrong, with no relevant difference between the two would be odd and contradictory. Hare says we must therefore be willing to ‘universalise’ our moral judgements.

Universalisability may sound complicated but what it comes down to is the simple idea that in morality it is wrong to make an exception for oneself or for others: if you ought to do some action then so ought I and everyone else to do it if we find ourselves in the same situation.

Persuasion vs prescription

There is a key difference in the way emotivism sees moral language as an attempt to influence others, whereas prescriptivism sees it as guiding action. So for emotivists, if Theresa May tells us 'it is wrong to join terrorist organisations fighting in other countries' then she is trying to affect our attitudes and behaviour so that we don't join a terrorist organisation. But, for the prescriptivist the essence of moral language is

not to influence but to guide: May is actually saying 'Do not join terrorist organisations'.

The emotivist doesn't think Theresa has much more to add, as for them ethical language is non-rational. However, Hare's prescriptivism accounts for our expectation that Theresa would be able to offer reasons for her instruction, and be able to enter into a discussion with us about these reasons.

So for Hare, when I express a moral judgement, I am not trying to influence or persuade you, nor am I expressing my feelings. I am prescribing what you ought to do. Whether, as a result, you act as I prescribe is a different matter.

Attractions of prescriptivism

As with emotivism, prescriptivism avoids the problem of having to explain the existence of mysterious non-natural facts, or the relationship of natural facts to evaluative judgements.

It also has an advantage over emotivism in that it enables us to avoid the conclusion that moral discourse is fundamentally non-rational.

In summary, according to Hare’s prescriptivism, the main features of moral language are:

1. It is used to commend, to provide guidance for choosing what to do.2. It assumes a set of standards, features in virtue of which something counts as

e.g. good / bad / right / wrong etc.3. These standards are moral standards but they are adopted, rather than being

true or false.4. Two relevantly similar things must both be good / bad / right / wrong etc. or

not. To think otherwise is logically contradictory.5. When we engage in ethical discourse we are attempting to show that our own

prescriptions are consistent (not logically contradictory) and prescriptive.

Issues for Anti-Realism

Criticism 1 – Can Anti-Realism account for how we use moral language?

Emotivism and Moral languageIs emotivism’s analysis of the meaning of moral language correct? We can argue that it is not on three accounts:

First, being emotive and influencing people’s attitudes is something that lots of non-moral language does as well, e.g. advertising. So we will need to say more to distinguish morality from advertising.

Second, does moral language always function to influence others? We may express our moral attitudes to others who already agree with them or that we know to be indifferent to our views – so influencing their attitudes is not the purpose. But this doesn’t mean that we aren’t expressing a moral judgement.

Third, we don’t want moral language to always be emotive. The key moral terms ‘good’, ‘right’, ‘wrong’ and ‘bad’ may arouse emotions in others or express ours, but again, this depends on context. We would not argue that it is always good to arouse emotions in others on moral issues, especially by using emotive language. Moral discussion can be, and sometimes should be, dispassionate.

Evaluation

But how strong are these objections? The purpose of moral language, says emotivism, is to influence what people do. Without this, we would have no moral language or judgements at all. However, that doesn’t mean that it always has to be used for this purpose. Many types of language can be used in ‘non-standard’ ways in different situations. For example, it is possible to use fact-stating language to insult someone; e.g. ‘You have a big nose’. Likewise, language which is standardly emotive can be deployed without the intention to arouse emotion or influence action. The objections don’t show that moral language isn’t ‘essentially’ emotive, only that it isn’t always emotive.

Emotivism on Moral DebateIf I say ‘abortion is wrong’ and you say ‘abortion is right’, according to emotivism, it seems that I am just expressing my disapproval of it and you are expressing your approval. I’m just saying ‘Boo! to abortion’ and you’re saying ‘Hurrah! for abortion’. I am also trying to influence your attitudes, and you are trying to influence mine. But we are not doing so rationally, or by appealing to facts about what is good or bad. We are doing little more than if we had a disagreement over the best flavour of ice cream. Trying to influence people without proper reasoning is just a form of manipulation or propaganda. While sometimes moral argument might take this form,

we do usually take ourselves to be reasoning about what is right, not simply mouthing off.We can put the point another way. If, as emotivism claims, moral judgements and arguments are about influencing people’s attitudes, then a good moral argument will be one that is effective. That is all. There is no other, e.g. rational, criterion by which we might judge that it is a good or bad argument. Whatever I appeal to, to make you change your mind, no matter how irrelevant or far-fetched, if it makes you change your mind, it is a good argument. This is highly unsatisfactory.

Response

This objection, however, need not be fatal. Ayer argues that moral disagreements do not actually exist. For Ayer, what appear to be moral arguments are actually disagreements about matters of fact (for example, whether or not a 20-week-old foetus can feel pain, or can survive outside of the womb), and over the definition of terms (for example, whether a foetus is a person).

So if we are in disagreement over some issue, it is rational to argue so long as our disagreement concerns something objective, such as a factual belief about the world, or the meaning of the terms we are using. Solving this disagreement may also depend on the facts we are discussing. If I can demonstrate that the foetus is a human being and that it can be harmed, then the other person may come to agree with me on this argument and alter their emotional responses / moral judgements accordingly.

Evaluation of Emotivism on Moral Disagreement

Ayer’s defence of emotivism may show that particular moral judgements can be a matter for rational debate (when they depend on facts). However ultimately, for the emotivist, we are still arguing over the correct emotional response to an event. Any reasons I may offer for why something is wrong can only reduce to some gut feeling for which no justification can be offered. Therefore any sense that there is a rational basis for moral dispute is really an illusion.

Prescriptivism on Moral Language

Hare's prescriptivism gives a detailed account of what it is to make a moral judgement, and his theory initially seems to be more plausible than the 'Boo/Hurrah' theory of emotivism. But we may argue that Hare's account of moral language and the way we use it still seems to be limited. Warnock illustrates this when he asks if it is really plausible to suppose that all moral discourse is essentially concerned with telling people what to do. Surely, as well as prescribing, we may deploy moral terms in order to resolve a conflict, complain about something that has happened or even confess to our wrongdoings (amongst other uses). In other words, Hare’s analysis of moral language seems too narrow, only applying to situations in which moral judgements are intended to dictate action.

Response

This objection, however, is based on a misunderstanding. What the prescriptivist intends to stress is not the link between morality and specifically prescribing behaviour but the connection between moral discourse and any kind of action. So, while moral statements clearly do not always tell someone what to do, it is still plausible to hold the view that accepting a moral proposition means acting in a certain way if the appropriate circumstances arise.

Nonetheless, while the prescriptivist is right that ethical pronouncements are closely linked to behaviour, we can still raise questions about their explanation of this link.

Prescriptivism on moral reasoning

We’ve already seen how Emotivism deals with the problem of moral disagreement, but how does Prescriptivism approach the same issue? If I say ‘abortion is wrong’ and you say ‘abortion is right’, according to prescriptivism, it seems that I am just prescribing that you and I should not abort while you are refusing the prescription. But are we doing so rationally, or by appealing to facts about what is good or bad?

Hare argues that prescriptivism can explain moral reasoning. First, we can ask about someone’s reasons for prescribing what they do (remember, unlike Emotivism Hare thinks prescribing an action should come with suitable reasoning. The prescription on it’s own is not enough). Second, morality involves consistency (i.e. we must be able to universalise our standpoint) and we can debate whether or not a particular standpoint achieves this. For example, Singer claims there is no relevant difference between the suffering of people and the suffering of animals. If we wish to argue that causing the suffering of people is wrong, but that animal suffering is fine, we need to find a relevant difference between the two; a way of showing inconsistency in

Singer’s view. Moral disagreements can be about the consistency in applying certain standards, and reason can help resolve this.

Third, we can infer prescriptions from other prescriptions. We might form an argument against abortion using prescriptive imperatives: ‘Do not take innocent human life. Abortion is the taking of an innocent human life. Therefore, do not commit abortion.’ In this case our prescriptions are logically related to one another. So we can also use reason and debate to discuss these relations, are they correct or have we made a mistake?

Criticism 2 - Does moral anti-realism become moral nihilism?

Moral nihilism is the rejection of all moral values and principles. It is the view that nothing is of moral value, that we have no moral duties. Moral anti-realism claims that there are no mind-independent moral properties, no objective moral truths, and non-cognitivist forms of anti-realism claim that morality is an expression of our emotions or attitudes. We may object that if this is so, then really there are no moral values – we invent them. Really, we have no obligation to be moral, because we have no obligation to have certain emotions or adopt certain standards of value. If moral properties are not objective, if moral judgments are not objectively true, then why accept morality at all?

Moral anti-realists can argue that this is either an unfair simplification of their theories or a straightforward misunderstanding. Error theory allows that we can have subjective moral values, and all three theories will argue that living without moral values is itself a choice or expression of feeling, and one that moral people will disapprove of morally. The theory that moral values are a reflection of our feelings does not imply that we should stop having moral feelings. The emotivist may still show disapproval of anyone who advocates that morality doesn’t matter or is just a matter of taste. Similarly, the fact that we must adopt standards of value doesn’t show that we should stop making prescriptions, and we may prescribe that people live according to particular standards.

Response

The moral nihilist can respond that this is unjustifiable. Can we really justify interfering with how other people behave – when they behave ‘immorally’ – just because their actions don’t accord with our feelings or choices? This seems very petty. But this isn’t the reason we are interfering, claims the anti-realist. It is not because it offends us, but because they are being racist or cruel or cowardly or whatever.

The difficulty here is that my taking racist discrimination as a good reason to prevent an action is itself an expression of my feelings or the standards on which I make prescriptions. It is ultimately my subjective point of view. For the moral realist, by contrast, that racist discrimination is a good reason to interfere with someone’s action is a moral fact, based on the moral properties of racism. The moral realist claims to have the backing of objective reality whereas the anti-realist is stuck with subjective feelings and emotions.

Criticism 3 - how can anti-realism account for moral progress?

If there is no moral reality, we can argue, then our moral views cannot become better or worse. Obviously, they have changed – people used to believe that slavery was morally acceptable and now they do not. But how can we say that this is progress if there is no objective moral truth?

Response

There are two responses moral anti-realists may offer here:

First, they can claim that there can be very real improvements in people’s moral views if they become more rational. Some examples of how this may happen include:

1. People coming to know certain facts that they didn’t know before. In the case of slavery, people believed many things about slaves that were not true (one popular false belief was that they were stupid).

2. People becoming more consistent, more willing to universalise their principles. For example, Singer argues that if we were consistent in our feelings about preventing suffering, we would not eat meat.

3. People becoming more coherent in their moral judgements. Many of us have moral feelings that come into conflict with each other that we should be aiming to solve, e.g. over lying.

Because people are ignorant, do not always think logically, and have not resolved the conflicts between their different feelings and conventions, there is plenty of room for moral progress. But moral progress just means becoming more rational in our moral thinking, not becoming more ‘correct’ in our moral judgements.

The second response moral anti-realists can give is this: if we disapprove of past moral codes and approve of our own moral code, then we will say that we have made moral progress. Society has moved from moral principles that were bad (i.e. principles we disapprove of) to moral principles that are good (i.e. principles we approve of). That is what moral progress is.