In this section we will be probing a more rarified set of questions. In the previous sections (on normative ethics) we have been looking at questions of what persuades us to act in ways that are right or wrong, good or bad. Here we will be looking at whether there is such a thing as truth in ethics; what is meant by ‘moral sense’; the meaning or significance of calling something right or wrong, good or bad; whether only humans can be called moral agents.
Perhaps the most pressing meta-ethical question is whether we can talk about ‘truth’ in the context of our moral behaviour. To highlight this as a difficulty, consider the following two sentences:
We can test the first statement for its truth-value: if there is a boy, a chair and he is on it, then it is true, if not, it is false. Given the facts of the case, not only to we get to the truth, everyone with the same facts will get to the same truth about the truth or falsity of the statement. Further, the statement must either be true or false and not some mix of the two. The second sentence is quite different: we would agree about the fact of his being a boy or not, but we might disagree quite profoundly about whether he is good: a Believer might say ‘yes, because he has done God’s will’; a utilitarian say ‘yes, because he has increased happiness’; a deontologist say ‘yes, because he has done his duty’; a virtue ethicist ‘yes, because he has behaved virtuously’.
What allows us to establish truth in the first statement is that it involves entities (boy, chair) and relations (on) that can be verified using our senses. This is the case for facts such as this which occupy the physical world. However, it is a much more open question as to whether values (such as judgements about the boy being good or not) can be verified in anything like the same way – do we have a ‘moral sense’ analogous to our ‘visual sense’ so that we can ‘perceive’ goodness as we analogously perceive boys and chairs?
It was the philosopher David Hume (1711–76) who first set out the problem. He observed that ethical systems tell us what we ought to do. They base these instructions on the facts of the world – how something is. A simple example would be:
What Hume argued was that it was wrong to derive statements containing ‘ought’ from statements containing ‘is’ because they are different sorts of statement altogether: there is no logical connection between them. We might feel that we can fill out the sentence above (say from a utilitarian perspective):
Although this looks a bit more convincing, Hume would point at the claim ‘you ought to increase happiness’ and say that this is not derived from anything in the rest of the argument. Hence, you are not justified in the conclusion drawn. The conclusion that Hume did draw was ethics (which includes all ethical statements) cannot rest on the empirical facts of the world: there is a gap between them.
At this point, before dealing with Hume’s analysis directly, we might consider where in the world ethical statements might be. A first possibility is that Hume is wrong and that the ‘ethical world’ of claims about how humans ought to behave meshes perfectly with the physical world. There are two standpoints within this position: supernaturalism which identifies our knowledge of what is right and wrong with a God-given moral sense together with God-given norms; naturalism which claims there is substantial objective knowledge to be discovered in ethics just as we have discovered objective knowledge about the physical world.
A second possibility is that there are objective facts about ethics but they are not a straight-forward part of the physical world. We get at these facts in the same way as we get at facts about numbers and geometry, for instance – through intuition. This is called intuitionism (or sometimes non-naturalism).
The third possibility is that there is no such world: ethics is an invention of humans for their convenience alone. Within this position are the standpoints of emotivism and prescriptivism.
All these ‘-isms’ are, certainly for me, a bit off-putting so let’s see if we can quickly sort out and eliminate a few of them.
This is the claim that there is nothing factual about ethical statements. When I say something like ‘he is a good boy’ what I am really saying is that I feel positive about him. The feeling (or emotion) is separate from any of the facts about the boy: if both you and I know all the facts we might feel differently (you might say he can’t be good because he’s an atheist whereas I might say he can’t be good unless he’s an atheist, for instance). Emotivists stress that some of our claims about the world are based on beliefs (such as those about boys, chairs and so on), whereas others are based on our attitudes (feelings or emotions). Our attitudes are neither right nor wrong, they are just the way things are inside us at the time.
You will see that this means that ethics is wholly subjective – only I know how I feel about things in the world and that knowledge is denied to anyone else. And the same goes for everyone else: objective knowledge is impossible to obtain about ethical claims. From this, it is a very short step to say that ethical claims add nothing to our knowledge at all – they are meaningless.
To many people, this is a very naïve position (one which we have already alluded to critically in the section on Ethical egoism). The first objection is that our attitudes towards situations which have an ethical dimension can be changed by objective reasoning. I might say that, in a certain situation, it is right to do one thing, you might say it would be wrong. What we can then do is give our reasons for our respective claims and these reasons can be weighed and judged objectively. Not only that, our reasons can be universalized which, again, argues for objectivity. If objectivity is possible then ethics cannot be wholly subjective as emotivism demands.
This position accepts that there are no ‘moral facts’, no ‘moral knowledge’, but develops the notion mentioned in the last paragraph about the universalisability of moral judgements. Though this position is most often associated with R.M.Hare (1919–), you can see that it derives from Kant’s deontological system which deals with imperatives in a very similar way. What Hare emphasises is that moral judgements prescribe behaviour of a certain sort: they tell us to do things. There are other prescriptive statements (‘Shut up!’ for instance) but ethical prescriptions differ in that they involve reasoning – and because we can reason about them this means that they must involve some sort of logical relations. It is this logical aspect of the reasoning which makes them universalisable.
Prescriptivism is more attractive than emotivism because it introduces a greater subtlety which ethics seems to demand (it is not simple). What it points out is that when we make value judgements there is something over and above mere descriptive claims. So, for example, if I say ‘this is a good book’ we might agree on any number of descriptive characteristics (the book’s structure, plot, characterization, length, etc.) but still disagree about whether it is ‘good’. This ‘over and above’ characteristic of value judgements is sometimes called supervenience. What this extra does (or what it seeks to do) is influence the choice we make once we have seen or heard it.
A further attraction is that moral judgements demand commitment (due to this aspect of their universalisability): once we have judged we must commit to that judgement. This commitment to observe the universal principles of moral judgements fits in nicely with the way that humans seem to learn: we generalize from principles so that we can perform specific acts in the right circumstances. This seems to confirm that there are principles to be had.
Hare starts with Kant but does go further. You will recall that a difficulty with Kant was what to do when duties conflict (in moral dilemmas). Hare’s solution is an existential one: there is no ‘true’ answer (remember that prescriptivism rejects the possibility of ‘moral facts’ and ‘moral knowledge’) but, once you have made your choice, you must commit yourself to the principle which provides the justification for universalizing it.
Prescriptivism was given short shrift in 1958 by Philippa Foot (1920–2010) through its inadequacy on four fronts. She said that a) it was too broad and would allow claims like ‘All Jews should be killed’ to count as morally right; b) it was too trivial, allowing such things as ‘Everyone should stick out their tongue on Monday mornings’ as moral judgements; c) that it misses the point of ethics since it could ‘miss out’ moral principles that most people feel are important such as ‘torturing for fun is wrong’; and d) that it is endlessly revisable (since there are no objective facts) and so denies the stability that ethical systems require for social stability.
There are very few emotivists or prescriptivists lurking about the place nowadays.
You will remember that Hume insisted that ‘you cannot derive an ought from an is’; ie he said there is a logical gap between factual statements and moral statements. G E Moore (1873–1958) thought that he put his finger on just where the gap can be found when he turned to an analysis of what the terms used in ethics mean.
The term that Moore focused on was ‘good’. We all use this term in a variety of ways – ‘a good film’, ‘a good road’, ‘a good idea’, ‘a good boy’ – and all know what the term means when it is used in this way. What Moore criticized was the notion that this term (and, by implication, all such ethical terms) was connected to the natural world: he thought this naturalism was making a mistake, was committing what he called the naturalistic fallacy.
Using utilitarianism as an example (though we could easily substitute any other system of normative ethics), here is how his argument goes.
‘Naturalism claims that ethics is a part of the natural world and, as such, can be observed and measured in some way. Happiness is a part of the natural world.’
So far, nothing obviously wrong.
‘Being morally good (for the utilitarian) means doing whatever increases happiness.’
Again, this is a fair summary. But having set this up, he then points out what he sees as the fallacy.
‘The following claim is trivially obvious: ‘It is good because it is good’. But if ‘being good’ means ‘increasing happiness’ then the following should be equally trivially obvious: ‘It is good because it increases happiness’. This is not the case because when we are faced with the second sort of claim we might agree that, yes, some action or other increases happiness but still ask is it good? In other words, ‘good’ refers to something other than (or as well as) the natural property of happiness. Claiming that ‘good’ is a natural term is a fallacy.’
We’ll come back to this in a moment, but first let’s see how this argument led Moore into his intuitionist position. His argument (along with Hume’s observation) establishes that ethical terms cannot be defined in a naturalistic way since they are not empirical properties. Nonetheless, the term must refer to something in our minds otherwise we would not be able to appreciate what it means.
Now, he goes on, there are other terms that we know in our minds but which are not natural properties of things ‘out there’ in the natural world. Prime examples of this come from the world of arithmetic and geometry. Once we have appreciated the proof, for instance, we know that all plane triangles have internal angles whose sum is 180°. We do not know such things through our senses – it is a part of our intuitive knowledge.
Moore then points out that our knowledge of ethical terms, since it is intuitive like our knowledge of geometrical relations, can be either right or wrong: ethical statements, like geometrical statements, can be checked for their truth-value: we can use our intuitive knowledge to establish whether behaviour is right or wrong. Thus, for the intuitionist, objective knowledge of ethics is possible, but such knowledge is not to be found in the natural world. Our ‘moral sense’ consists in the intuitive grasp that we have of ethical concepts and the relations between them.
Moore’s view was tremendously influential in the early part of the twentieth century, leading (for many philosophers) to a re-routing of ethics. The way to develop a knowledge of ethics was thus to develop a knowledge of the meaning of ethical terms – an emphasis on the language we employ rather than anything else.
For other philosophers, this was a mistake. One thing they particularly didn’t like was Moore’s ‘world of the intuition’, a ghostly sort of place where things have a different metaphysical status and where (as in Plato’s real world) reside numbers, shapes, knowledge of the Good. If this world is allowed, then it ushers in the problem of how we can know this intuitive world – the traditional answer of the ‘soul’ or ‘spirit’ does not fit comfortably with hard-headed naturalism. Had Moore gone wrong with ‘good’ being a non-natural term? Was Hume up the creek in claiming there was a logical gap between ‘ought’ and ‘is’?
This ethical position insists that morality is as much a part of the natural world as boys, chairs, colours, language and mathematics. Ethical terms are natural terms in that they can be meaningfully examined and agreed upon rationally. Morality has a content and this, for the naturalist, is anchored in the human predicament which, naturally, we do not wish to see deteriorate.
So, for example, naturalists claim that goodness is a natural property. It might be happiness, eudaimonia, or whatever, but that any of these natural things (which, of course are observable and measurable things in our natural world) include within them (as a whole or a part) the notion of ‘good’. They point out that Hume and Moore are mistaken in their analysis and that there is no ‘naturalistic fallacy’. Let us look at some of the example statements we have already been considering to see why they point this out:
Hume said there is a ‘logical gap’ in a). Moore says there a mistake in b). Both Hume and Moore would agree that c) is true. Naturalists say that statements a) and b) are not substantially different from statement c) and this is why Moore and Hume are wrong.
Take statement c). What we have here is a statement of a relationship between ‘plane three-sided figures’ (p) and ‘internal angles adding up to 180°’ (q): p = q. The reason why we know this is true is because we have had it demonstrated to us by a series of reasoned steps in a geometrical proof. There is no reason to suppose that statements such as a) and b) cannot have their p and q elements linked in just the same way – all it needs is a series of reasoned steps. Of course, it may be difficult to discover this series, but there is nothing in either Moore or Hume which demonstrates that this is logically impossible.
A further objection to Moore is that he treats ethical terms as if they are things in themselves rather than properties that might ‘emerge’ from (or be supervenient on) things in the world which is why he sees them as ‘non-natural’. Just as ‘smoothness’ is a property of a table-top but not of the atoms in the table, so ‘good’ can be a property of the boy but not of his atoms.
One attraction of naturalism is that it allows for objective knowledge of ethics and, like scientific knowledge, can be discovered (and is definitely not simply invented to suit current tastes). If we allow that ‘moral facts’ are mere inventions then we are inexorably led to moral nihilism (as advocated by the contemporary philosopher Gilbert Harman, for example) where moral knowledge is an illusion and ‘anything goes’. We can take a little detour here to see how naturalists resist the claims of ‘moral anti-realists’ who say that there are no moral facts, truths or knowledge.
Like the relativists we considered in the normative ethics section, anti-realists point out that there is no universal moral code – different cultures and different ages have different moral codes and this shows that such codes are invented for our convenience. In reply, realists (including naturalists here, of course) say that simply pointing out that variation is a fact does not imply there is no moral truth. The differences in norms may be due to ignorance, immaturity, superstition or irrational authority in just the same way as beliefs about the world may differ due to these things. If another society believes sacrificing a goat will make it rain it doesn’t mean we have to abandon truths about meteorological physics. If another society routinely tolerates rape we are not thereby forced to accept that it is not true that rape is wrong. In addition, there actually does seem to be a consensus about moral behaviour amongst peoples: not torturing for fun, not lying and cheating, being compassionate (at least to friends and family), being just.
Another attraction of naturalism is that it provides a natural basis for our learning of ethical principles and their application to situations in the world as we encounter them as first mentioned in the section on prescriptivism above. A natural basis is one that is open to revision and development which, of course, holds out hope for an improved moral education for humanity.
However, unlike (natural) scientific knowledge, ethical knowledge is not simply a matter of checking how the world is. Naturalists submit that ethical and scientific knowledge are different aspects of the natural world. The difference lies in what is called ‘direction-of-fit’. In science, we make a claim in words and see if the world fits the claim – if it does our claim is true, if not it is false. If our words turn out to be false, we must change them to fit the world as it is. On the other hand, an ethical claim in words seeks to fit the world to its truth. If ‘it is right to reduce suffering’ is our ethical claim and we found that it is a fact that there is increasing suffering in the world, we seek to change the world and not our ethical claim. What ties the two aspects together is reasoning where justifications are provided for establishing how the world is as well as for how the world ought to be. These justifications have to be anchored in the world and in our nature if they are to coincide with the truth.
Before leaving this section it should be emphasized that this area of philosophy is currently very lively with disputes over moral realism, anti-realism, (even ‘quasi-realism’), naturalism and the rest, all being picked over fruitfully. Thus, do remember that this is a mere brush with the issues and you shouldn’t see it as in any way definitive.
In the ‘Core’ section we have looked at the idea of one of the distinguishing characteristics of humans being moral agency. What we will briefly explore in this section is where we might have derived this moral nature and whether it might belong to us uniquely.
As before, we can look to God first. Believers think that humans are not the only moral beings. God (and his angels? And/or other gods?) are moral beings too. General accounts for our morality then link our nature with God’s which shows where it came from. Also generally, this sharing of nature with God is denied to other beings – we are unique as the only natural (rather than supernatural) moral beings. This approach sweeps in a host of fascinating theological questions including how we can know that our nature and God’s nature are similar in this regard; what justifications God might have for His apportioning of morality to only a small part of his Creation; why God has not made his presence or intentions clearer (why are many of us so ignorant of Him that we worship another God or none? Why do we interpret what we ought to do differently even within the same religion?); why have Evil in his Creation in the first place? All of these questions (and many more) are beyond our scope here which is limited to mere philosophy where we seek reasons which are open and acceptable (in principle) to all human beings rather than a subset of humanity who Believe. Thus, reluctantly, we turn from God and see if we can find other possible sources for our moral values.
In the preceding section we have already looked at the possibility that our values are all simply invented for our own convenience which changes with age and culture. Since this position has been fairly unconvincing, we will now look at the possibility that values are a natural part of us derived in a natural way (without any supernatural intervention).
As you will appreciate, we have been emphasizing reason as a human characteristic that is central to matters of ethics. If reason is supreme in ethics, or of greatest importance, then if we are the only species capable of reasoning we are indeed unique. This approach is typical of, say, Kantian ethics where any other human attribute is marginalized.
It is interesting to note that Kant was prompted (at least in part) into developing his moral philosophy by reacting to Hume’s insistence that reason has nothing to do with ethics. Hume (in his Treatise of Human Nature) observed that when we consider any action for virtue or vice there is never anything to discover in the action itself. If we see a bully beating up a small child there is nothing observable that is wrong: what makes it wrong is our feeling of outrage. Hume insists that not only does ethics reside in the subject rather than ‘out there’ in the actions, but that our initial reaction is a sentiment (an emotion, or feeling, or, as Hume says, a passion). Far from reason driving us to an ethical position, it is sentiment that provides the ethical position. Reason comes afterwards in justifying the initial sentiment from which our sense of virtue and vice derives. Hence his slogan: ‘reason is, and should be, the slave of the passions’. (From this you can understand Kant’s reaction since Kant puts reason above all.) For Hume, what motivates our actions in the world are our sentiments – reason has no motivational force. This is demonstrated by the smoker who cannot give up: they reason perfectly that it is physically harmful, repulsive, expensive, and so on, but reason doesn’t work – they still feel like a cigarette and are driven to smoke.
At first glance, Hume’s moral philosophy is rather chilling and pessimistic, especially if we link it to another aspect of human nature – our overwhelming self-interest. No wonder Kant sought to put reason in a supreme position since this insists that other humans matter. However, Hume was rather more optimistic than the first glance might have revealed. He indicated that although it is the case that self-interest is never far from our actions, there is another aspect of our nature which influences what we do: sympathy for others. It is sympathy which enjoins us to discount (at least to some extent) our own selfishness and to do things which benefit others – perhaps even at our loss.
What we need to address here is whether we can link sympathy and ‘good behaviour’. First, it seems to be a brute fact that we are sympathetic. We can appreciate the plight of other humans (and even other animals) and share their grief, or pain, or joy and so on. Empirical facts such as tears during a romantic film, wincing on seeing someone struck in the face are evidence of our sympathetic nature. That said, the question still remains that, if it is my self-interest to punch you in the face to get more food for myself, why don’t I over-ride my sympathy for your pain and go right ahead? We are still back at the question why be moral?
A clue to how this can be answered goes back to the virtue ethics of Aristotle. We cannot simply be sympathetic individuals. For sympathy to work we need at least two of us. This obvious observation derives from the fact that the self is not simply wrapped up in the individual: there is a social nature to the self. This is naturalistic because, of course, humans belong to a natural social group. It is through our interactions that ethical behaviour might emerge.
An illustration of how this works stems from David Gauthier’s observations on the theme of what is known as the prisoner’s dilemma. This famous thought-experiment to investigate the basis for morality goes like this. Imagine that two spies, Andy and Bill, are caught and put in prison. If both admit to nothing, they will both be kept in prison for 1 year. However, if they both spill the beans about the other being a spy, both will get 5 years in prison. But, if one spills the beans while the other keeps silent then the spiller gets to go free immediately as a reward, the non-spiller gets 10 years a) for being a spy and b) for not owning up.
The dilemma is to think what you would do in their position. We can take either since they are both in the same position. Andy will think to himself that Bill can do one of two things: stay silent or spill the beans. If Bill stays silent then what is best for Andy is to spill the beans (since he gets 0 years in prison). If Bill spills the beans then what is best for Andy is to spill the beans too (otherwise he gets 10 years in jail rather than 5). So, whichever Bill does, Andy should, in his own interests, spill the beans. Of course, Bill will reason in exactly the same way and so they both end up in jail for 5 years rather than the 1 year that they would have been in prison if they had both kept silent.
The prisoner’s dilemma shows that self-interest will not necessarily guarantee the best outcome for the individual. This is reinforced in social situations where you are likely to re-encounter people with whom you have interacted before. This (more natural) pitching of self-interest against sympathy is illustrated in a game devised by Axelrod. Two players are given two cards each, one labeled cheat, the other cooperate. The banker tells them that they must choose one card and lay it out simultaneously with the other player. If both players put down cooperate the banker gives each £5. If both players put down cheat, both are fined £1. If one player puts down cheat and the other cooperate, then the ‘cheater’ gets £10 and the ‘cooperator’ is fined £5. The game will continue until the banker says quits.
Let’s assume that there are 100 games and you are playing me. Obviously, you could get maximum returns by always playing cheat so long as I am sucker enough to keep playing cooperate: you would get £1000 while I would be short £500 (and vice versa). If we both played cheat a lot of the time, we would end up with less money than if we both played cooperate more often.
Now, the obvious objection to this is that it is all very artificial: in life we don’t have clear and obvious rewards for cooperating with others (giving up a little self-interest to maximize returns) rather than cheating (going all-out for number one which may result in our being worse off in the long run). That said, Axelrod’s game is extremely robust in that there is one clear strategy that maximizes results so long as the ‘rewards’ and ‘fines’ are within reasonable bounds – and this is found whether played on computer-devised trials of strategies, or when actually playing the game with humans irrespective of their age, colour, creed, etc. This optimal strategy is to play cooperate first and then, after the second game, play the same card as your opponent last played. Thus, if they played cheat in game 2, you would play cheat in game 3. This ‘tit-for-tat’ strategy means that cheating is not rewarded whereas cooperation is.
This analysis provides support for the mutual cooperation that is seen in natural groups of animals. If grooming the fur of others in the group is common, obviously one animal’s best strategy is to allow another one to groom and then refuse to groom them back (the sucker!). However, these ‘suckers’ will then become reluctant to groom the ‘cheats’ and will just groom other ‘suckers’. Cheats will get lousy and be worse off than if they had cooperated. This gives us a justification for the evolution of the attribute of sympathy that is a part of human nature: it helps the group as a whole to maximize the ‘happiness’ (health, well-being, success, lack of suffering, such notions) of every individual.
This brings us back to human uniqueness. Clearly, we are not the only intelligent social animal on the planet. When we look at chimps, for example, we see similar attributes with respect to cooperation and sympathy. Can we call their activity moral?
This is resisted for one main reason. This lies in the wider scope of our moral behaviour than that found in the rest of the animal kingdom. Sympathy only really makes sense for the well-being of a group if it is confined to just the members of that group – there is no point being kind to someone or something that you will never encounter again. Humans typically sympathise with people they haven’t met, will never meet again, will never ever meet (such as future generations). This sort of sympathy is, perhaps, an extension of our capacity to reason from the individual to the universal. If so, then, as the only species capable of reason, we are unique in our morality.
Finally, if our morality is part of our nature, can we identify a difference in another part of our nature, our gender? Carol Gilligan (in 1982) pointed out that when presented with the same moral problem, girls and boys generally differed in their responses. Boys tended to seek and defend principles and rules whereas girls looked for more relational solutions. Thus, when presented with the dilemma of whether a poor man should steal a drug from the chemist to save his dying wife, boys argue that ‘life’s worth more than money’ or ‘he has a greater duty to his wife than to the chemist’. Girls’ answers are generally such as ‘his wife won’t respect him if she finds out he steals’ or ‘they should discuss what to do first’. This observation has led some to posit a ‘feminist ethics’. However, it has not really caught on. This is largely due to the fact that this separation is only a generalization – some males look for relational solutions, some girls for principles – and that most of us agree that both aspects of ethical problems need to be taken into account when deciding moral questions.