Meta-ethics means "beyond ethics", in other words, rather than asking whether an action is good or bad (normative ethics), meta-ethics asks "what does goodness or badness mean?" In meta-ethics we discover how language functions, and whether moral language is a special type of language.
Moral theories divide into two branches. Naturalists argue that goodness is a natural feature of the world because it's an observable property, a moral fact if you like, subject to empirical research and observation. This was the position of the utilitarian thinkers. Non-naturalists argue that moral goodness is not a natural feature of the world, but a subjective intuition, or an expression of emotion (emotivism), or a logical property of how we reason (prescriptivism and the ethics of Kant).
The debate really began with Hume, who argued that it is problematic moving from a descriptive statement to a normative statement, an "is" to an ought. Although Hume was himself an ethical naturalist (one of many paradoxes in this branch of ethics), he is the father of logical positivism, which itself produced the theory of emotivism in the 1920s and 1930s. Emotivists such as A.J Ayer are building on Hume's insights and those of the Vienna School of empiricism, known as logical positivism. By concentrating on the language of morality these theorists hope to uncover something about the grounds of morality - and their hidden agenda, if you like, which turned their task into something of a crusade, was to destroy the meaningless (but dangerously authoritarian) grip of metaphysics.
Go to the weblinks section for a fascinating interview of the great emotivist A.J.Ayer by Brian Magee.
Go to the handout for an analysis of the fact/value (is/ought) problem and how this spawned emotivism (Ayer and Stevenson), intuitionism (Moore, Prichard and Ross) and prescriptivism (Hare).
Meta-ethics may seem at first an irrelevance, a strange obsession for the intellectuals in their ivory towers. But on closer inspection it is actually fundamental - if we are talking and thinking nonsense, or if our arguments are fallacious (invalid) then the human race really is doomed to lurch from one tyrannical thought system to another, or from one superstition to the next.
To be liberated to redeem our future we have to reason - but to reason we must be sure we know what we mean and believe something meaningful.
The best introductory treatment is found in Jones et al chapter 3 - see the biblio graphy (and be warned, A level textbooks in their desire to keep things nice and simple can be prone to serious error in this part of the course). The best slightly more advanced treatment is found in Pojman's outstanding book (2006, chapter 11).Add a comment
For twenty-five multiple choice based on Louis Pojman's chapter 11 (the Fact/Value Problem), which includes a dsicussion of all our meta-ethical theories, go tot his link and find chapter 11 in the chapter menu. Then click on the "tutorial quizz" on the lefthand menu. There are also excellent questions for reflection and further readings (extracts from primary sources).
Harry Gensler's thirty self-test questions which usefully are marked for you (on prescriptivism):
For A video of Brian Magee interviewing A.J.Ayer on Logical Positivism.
For a good itnroduction to emotivism go to:
For a very clear discussion of Ayer and Stevenson, with useful quotes and a critique, go to:
Charles Stevenson's chapter on an emotive theory of language is available here:
For criticisms of emotivism go to:
Harry Gensler gives a succinct one page summary of prescriptivism:
An obituary of RM Hare reviews his life and work:
Here is a thought-provoking couple of sides on prescriptivism and torture;
A photocopied article from Philosophical Quarterly 1986 discusses the link between prescriptivism and preference utilitarianism:
Excellent notes on Moore's open question argument go to:
Intermediate websites:Some clear notes froma University Professor:
A photocopied chapter of a meta-ethics textbook:
An American academic, Daniel Fincke, gives an excellent short introduction to Hare and emotivism:
Add a comment
Meta-ethics examines the language behind ethical theories and moral beliefs. Instead of asking "is stealing wrong?", the meta-ethical philosopher asks: "what is meant by saying stealing is wrong?" So a theory of metaethics is a theory of meaning. Consider these three statements:
· Stealing hurts people.
· Stealing is wrong.
· You ought not to steal.
What does each statement mean? The first statement is a description of what happens when you steal. We could test whether it’s true or false by asking you how it feels to have something stolen from you. This is a naturalistic statement, as it describes a natural feature of stealing, that it causes pain.
The second and third statements are normative rather than descriptive, as they have a value-judgement within them. Stealing is bad. But notice how the third statement is going further, it is a prescriptive statement, prescribing or strongly advising a course of action: not to steal.
And what of the second statement? Is it merely an expression of a strong feeling, as the philosopher A.J. Ayer (1910-88) suggests, equivalent to making a grunt of disapproval, such as “stealing, yuk!”? Or is it based on an intuition, as the philosophers Moore and Ross suggest?
Three branches of metaethics
There are three branches of metaethics which we need to consider.
The starting point of emotivism is David Hume’s analytic/synthetic fork, so called because it gives us two alternative types of language, which states that statements about the objective world can only be of one of two sorts.
The Scottish philosopher David Hume (1711-76) argued all statements that are either true or false are either analytic or synthetic. An analytic statement is true by definition: "all bachelors are unmarried". The truth or falsehood of this statement is contained in the very idea of "bachelorhood". A synthetic statement, in contrast, can only be verified by sense experience. My brother is a bachelor is synthetic because I can check whether he is or is not married (it's a statement of fact).
Moral statements are neither analytic nor synthetic, argued Hume, so they're an expression of emotion or sentiment. The fork therefore has two "prongs", the trouble is, moral statements don't fit either the anlaytic or synthetic "prong", and so are pronounced objectively "meaningless".
Don't overstate this though: moral statements are still subjectively meaningful - meaningful to me.
The Emotivism of Ayer and Stevenson
AJ Ayer (1910-1989) argued that statements about reality needed to be verified true or false according to sense experience. Ethical statements had no factual content as they could not be verified true or false.
Take, for example, the utilitarian proposition that things are good or bad according to the pleasure or pain produced.
So “goodness” here is a natural property of an action because it can be measured by consequences, (the "naturalistic fallacy" below).
So, argues Ayer, if we can still ask the question “is it good?” after we have asked “is it pleasurable?”, then goodness or badness must mean something else other than the pleasure or pain produced.
Ayer goes on to argue:
So Ayer concludes that moral statements are primarily expressions of emotion, hence “emotivism”, although he goes on to hint at something else as well: “it is worth mentioning that ethical terms do not serve only to express feeling, but to arouse feeling, and so to stimulate action” (1971:143)
CL Stevenson (1908-79) builds on this last point by arguing that ethical statements do two things:
1. They express an attitude or feeling
2. They seek to move people to behave in a certain way
This second persuasive function indicates that when two people disagree about what is right or wrong, they are showing a disagreement in underlying principles and values.
If I say “stealing is wrong” I am effectively saying “I feel strongly that stealing is bad, and you should think so too!”
Wittgenstein would say that Ayer has made a category error in failing to realise that language is like a game where anything can be ruled in as analytic, synthetic or anything else – stealing is wrong does mean something because of the context in which it is defined – it is more than an expression of opinion. There is also a logical problem with the analytic/synthetic distinction. Ayer's verification principle which states that all statements of fact must be provable true or false, is itself neither analytic nor synthetic: it's not true by definitiona nd it certainly can't be proved true or false by sense experience. So is it therefore meaningless?
Test yourelf on emotivism with twenty-two multiple choice questions below:
What might we mean by saying “you’re a bad person”?
If Wittgenstein is right, that the logical positivists make a category error int heir misunderstanding of the natire of language, then what do I think I’m doing by saying “you are bad?”
Here are some possible meanings:
1. You are acting against the consensus view. Most people disapprove of your action.
2. Your action is against the good end which is what most people pursue ( a natural law answer).
3. Your action has bad consequences (ie causes pain).
Note: all these three are natural or observable features of your action. They correspond to the views of an ethical naturalist, a view which Ayer explicitly rejects. But just because he rejects it, it does not follow that I don’t mean something like this when I say “X is wrong” or “Y is bad”.
The naturalistic fallacy
Philosophers like GE Moore (1873-1958) claim that you cannot make a value judgement from a factual observation.
· You can’t derive an "ought" from an "is" – “to define an ethical judgement as a statement of fact is an error”. So saying "he has stolen my money" is not equivalent to saying "you ought not to steal".
· “In trying to define goodness as a natural property, naturalists confuse the property of goodness with some other non-moral properties good things happen to possess” (Robert
· Happiness and goodness are not intrinsically linked. We can always ask, as Ayer points out in the quote above, it may make you happy, “but is it good?” This is sometimes called the "open question argument" because the fact of happiness doesn't automatically entail moral goodness (smoking dope makes me happy. But is it good?).
Good is just good: we know this, says Moore, by intuition.
Both A.J. Ayer and R.M. Hare (1919-2002) are non-cognitivists. This means that neither think that saying something is good can be true or false. G.E. Moore would disagree (making him a cognitivist) because he thinks we know what is itnrinaically good by intuition and then we can calculate empirically what the right thing to do might be by examining the likely consequences (he shares this view with utilitarians generally). In other words as a cognitivist
A.J. Ayer’s theory of emotivism seems to have two problems:
1. Moral arguments are pointless because all we are really saying is “stealing, boo” or “generosity, hooray”.
2. It reduces moral statements to almost nothing (it is reductive). There seems to be little difference between saying “I don’t like peaches”, and “stealing is wrong”. But surely there is a difference?
Can R.M. Hare’s prescriptivism solve these problems?
R.M. Hare agreed with A.J. Ayer that when we make a moral statement we are just expressing our own attitude, but he thought that we were doing other things too. He thought that when we make moral statements we are also saying what we think other people in similar situations should do. We are prescribing (telling them) what they should do. This is why the theory is called prescriptivism.
Let’s consider again what the basic difference is between an intuitionist and emotivist.
I say: “you shouldn’t steal”
G.E. Moore (intuition ism) says I mean: I have a feeling which I can’t define that stealing is wrong, so you shouldn’t steal.
A.J. Ayer (emotivism) says I mean: I just don’t like stealing, so I don’t want you to steal.
R.M. Hare (prescriptivism) says you mean: I just don’t like stealing, you shouldn’t steal, and I don’t think that anybody else should steal either.
Three essential features of prescriptivism
For a statement to be prescriptive, rather than descriptive, it must prescribe, or strongly advise, a course of action. So although Hare accepts that there is no natural feature of an action (such as pleasure or pain) that determines whether it’s good or bad, nonetheless when I say “stealing is wrong” I am meaning by this “it’s wrong, and I don’t want you to steal!”.
When I say “murder is wrong” my statement means something entirely different from a purely descriptive statement like “lemons are yellow”.
So, says Hare, moral statements are commitments to action , which is something more than just a feeling (as emotivists argued).
Hare was a Kantian to this extent: he argued that the nature of moral statements meant they were always universalisable (as Kant’s first formulation of the categorical imperative suggested).
Hare believed we could reason our way to a moral decision. The reasonableness of saying “murder is wrong” lies in the fact that I don’t want you to murder me. My statement “stealing is wrong” means that everyone in similar situations ought not to steal.
Hare believed that two things guided my moral decision: my inclinations, what I felt I wanted, and my understanding of the consequences of my actions.
Both required moral imagination, the ability to put myself in another person’s shoes, and say, as we say to children “how would you feel if I did that to you?”
Hare was reacting to Ayer’s belief that moral statements were only expressions of opinion, and Ayer;s argument that there could be no such thing as rational moral discourse.
It’s true that moral statements are expressions of opinion, but so much more than this: they are universalisable, reasonable calls to behave in a certain way.
Thirdly, the moral “ought” is overriding. It’s a very strong form of “ought”.
“I ought to get my hair cut today”.
“I ought to take a holiday”
“I ought to help my sick mother”.
These three examples show how “ought” can have different intensity: if I say “you ought not to mix purple and yellow” this is an aesthetic judgement about beauty.
If I say “you ought not to steal” I am giving a strong prescription: the strongest ought there is.
I am saying in effect: “you must not steal, it’s the strongest advice and admonition I can give you”.
So R.M. Hare thinks that when we make moral statements we are not talking about ourselves but everybody. If I say “stealing is wrong” I mean: “you shouldn’t steal” and “nobody should steal”.
Because Hare believes that when we make moral statements we are saying what should be true for everybody, he believes that before we make moral statements we should think about what the world would be like if everybody had to follow our rules. In fact he is a preference utilitarian with a Kantian view of moral language! Peter Singer, the famous preference utilitarian, was his student.
Although R.M. Hare does not believe that anybody can be correct or incorrect when they say something is good (he is a non-cognitivist). He does believe that before we say something is good we should follow a particular method.
This method is universalisability. He borrowed this idea from Immanuel Kant (1724 -1805). What universalisability means is that before you say something is good you should imagine what the world would be like if your statement became a rule and everybody had to follow it.
Imagine you think: “Rory shouldn’t steal that phone”. R.M. Hare says that we should imagine a world where nobody in Rory’s situation can steal phones. Would we like that? Only if we like the idea of a world where nobody like Rory can steal phones should we say that Rory shouldn’t steal that phone.
In other words we have to universalise (universalise means apply to everybody) our moral statements. We also have to live by them ourselves.
Test yourself on prescripitivism by clicking this link and trying this 22 question multiple choice:
Prescriptive and evaluative meaning
Lots of words contain prescriptive or evaluative meaning, not just "good" and "bad", "right" and "wrong".
With virtue ethics, for example, we found that Aristotle commended various habits of character which lay between the vices of deficiency and excess: courage, self-control, moderation, and humour, for example.
If we produce a more Christian list, which words would be in it? Words like loyalty, obedience, meekness, generosity or immorality, lust, greed, drunkenness.
Now a word like meek has descriptive meaning. We can describe a meek person as "one who has their passions under control, and so will not be easily provoked", or "someone who is too nervous to act", depending on whether you take the positive or negative connotation of the word (the English language being rich in connotations!).
Hare argues that with our list above of virtues and vices, the descriptive meaning is primary and the evaluative secondary. In fact, as I've just shown with the word "meek" the evaluation could be good or bad, depending on the meaning I take of meekness (meekness = dormat or maturely controlled).
But with good the evaluative meaning is always primary.
"When we call a car or a watch or a cricket-bat or a picture "good," we are commending all of them. But because we are commending all of them for different reasons, the descriptive meaning is different in all cases".
Task 1A What is the descriptive meaning of good in these examples?
1. This meal's good!..........................................................................................
Complete the following sentence. According to Hare, the descriptive meaning of good...................... whereas the evaluative meaning ................................
So good has a particular force in language because...................................
Hare is saying that there is more to saying something is good than just our attitude. We are also saying what is good for the world. Because we are saying what is good for the world; before we say something is good or bad we should think about what would happen if everybody had to follow our rule.
Hare believed that once we have decided on our principles we should try to live by them. But we can make two criticisms of this theory.
1. R.M. Hare’s theory permits all kinds of terrible moral theories. A Nazi could universalise his ideas that Jews should die and it wouldn’t be wrong. This is because Hare does not believe any moral statements can be true or false. As long as the statement has been universalised then it is acceptable.
2. Hare’s theory allows any pointless theory to be moral. We can universalise nose picking on Wednesdays if we want to.
Task 1B Which has the greatest moral force?
1. You shouldn't have sugar in your tea.
Now, write down your conclusion from this exercise.
The word "ought" in English is used ..........................................................................................................................................................................
Intuitionists argue that when I say “stealing is wrong” I mean “I have a moral intuition that stealing is wrong”. An intuition is like a hunch, something I know innately.
W.D.Ross (1877-1971) argues that we know by intuition the idea of prima facie duties. Prima facie means “at first appearance”. Ross also calls them “conditional duties” because they apply as long as there are no stronger obligations present. For example, Ross would argue we should keep a promise unless stronger moral obligations arise, such as the need to protect someone’s reputation, or to save a life.
Our prima facie duties are given by intuition, and where they conflict, or when we make a choice, we have an actual duty.
Moral intuition or perception has three functions in this approach:
1) It tells us when one prima facie rule, which at first seems to apply, does not apply because another overrides it. In other words, moral insight tells us when we have exceptions to specific guidelines. This requires us to be sensitive to the situation.
The other two functions are related not to the situation directly but to the general rules.
2) Moral intuition tells us what the prima facie duties are. We just see, by moral intuition, that generally, non-injury is a good rule to follow.
3) Moral intuition tells us what the priority rules are. We just see, by moral intuition, that generally non-injury takes precedence over beneficence (doing good).
Note that the moral intuition is not the same as perceiving a colour, a sound, a taste, a texture, or a smell; nor is it the same as perceiving physical thing. It is a grasping of a truth. When it picks out morally relevant parts of a situation, it makes use of perceptions of the nonmoral kind, but it goes beyond them to certain features as morally relevant, features that call for applying a prima facie duty to the situation. When moral intuition grasps the prima facie duties themselves, it is a grasping of a moral general truth.
The simple theory explained above leaves unanswered the question where these moral intuitions come from. This question can be answered in part by the theory of Virtue Ethics, for example. Our abilities to have correct moral perceptions depend upon our moral upbringing, the moral habits we have formed. Have we formed virtues or vices?
Moral perception can be corrupted or distorted. We can imagine people who always follow a distorted version of the duty of self-improvement and ignore the other principles; for instance, they promote their own pleasure (taking that to be the essence of happiness) and do not care whether they injure others or are unjust in their dealings with others. One might plausibly say that these people have formed defective moral habits, vices.
Exercise: follow this link for a short quiz on intuitionism
|POWERPOINT Three issues in meta-ethics|
|Category: ETHICAL THEORY 8 Meta-ethics|
|Published on Tuesday, 16 October 2012 20:25|
1. The issue of naturalism: is there such a thing as a moral fact observable in the world? What might this fact be?
2. The issue of facts and values (the "is/ought problem"). Can you derive a value from a fact, or make a prescription from a description? If so, how?
3. The issue of the meaning of "good' and "right". Is goodness an expression of a feeling, or an intuition, or evidence of a special type fo language with a logic all of its own?
Add a comment