In the series of extracts from my almost-finished book on the  history of moral thought, I have reached Chapter 17, which looks at the subjective turn in analytic philosophy, and the unravelling of morality in the twentieth century, from the intuitionism of GE Moore’s Principia Ethica to JL Mackie’s ‘error theory’ and moral nihilism. This extract begins with Moore and looks at how intuitionism gave way to emotivism.

GE Moore’s Principia Ethica, published in 1903, came to be both one of the most famous ethical work of the twentieth century and one of the most troublesome. It was a work whose arguments were extraordinarily flimsy and highly dubious and yet, as Mary Warnock observed in her study of twentieth century ethics, has come to be regarded ‘as the source from which the subsequent moral philosophy of the century has flowed, or at least as the most powerful influence upon this moral philosophy’. The publication of the Principia Ethica was, John Maynard Keynes wrote, ‘exciting, exhilarating, the beginning of renaissance, the opening of a new heaven on a new earth’. The influence and excitement and exhilaration of Moore’s book lay less in the lucidity of its moral argument than in its ability to locate a fundamental shift in the character of moral thought. If the eighteenth century had seen the triumph of the human in moral thought, and the nineteenth had wrestled with the moral implications of the death of God, the twentieth had to grapple with the consequences of the growing disaffection with human agency.  One expression of this was, paradoxically perhaps, an increasingly subjective view of morality. In the Anglophone world that view found a grounding, in part at least, in the Principia Ethica.

George Edward Moore was born in 1873, trained in Cambridge where he eventually occupied the chair of mental philosophy and logic. He was, with Bertrand Russell, Ludwig Wittgenstein and Gottlob Frege, one of the founders of the analytic school in philosophy, which came to dominate the Anglo-American world. He was also, alongside Virginia Woolf, John Maynard Keynes, EM Forster and Lytton Strachey, an important member of the Bloomsbury Group, and while not as well known to the public as other leading intellectuals, nevertheless wielded considerable influence through his social networks.

Moore wrote the Principia Ethica, so he tells us, in order to ‘distinguish clearly two kinds of question, which moral philosophers have always professed to answer’ but which ‘they have almost always confused  both with one another and with other questions.’  The two questions are: ‘What kind of things ought to exist for their own sake?’, and ‘What kind of actions ought we to perform?’.

Moore’s answer to the first question is that things that ought to exist for their own sake are things that are intrinsically good.   It is, however, he insists, impossible to define what it is to be ‘good’. Goodness is the name of a property that is simple and beyond analysis. We intuitively recognize that which is good, but we cannot analyze it in terms of anything more fundamental. Nor can any evidence be adduced to show that something is intrinsically good. All we can do is acknowledge it when we encounter it:

If I am asked ‘what is good?’ my answer is that good is good and that is the end of the matter. Or if I am asked ‘How is good to be defined?’, my answer is that it cannot be defined, and that is all I have to say about it. But disappointing as these answers may appear, they are of the very last importance.

Moore compares the concept of goodness to that of ‘yellowness’. Yellow is, like good, simple and incapable of analysis, a property we understand by being directly acquainted with it, but which cannot be described or defined to anyone who has never seen that colour. ‘Yellow and good’, Moore wrote, ‘are notions of that simple kind, out of which definitions are composed and with which the power of further defining ceases.’ The fact that ‘yellow’ is indefinable does not prevent us from being able to say what things have the property of being yellow. Similarly with goodness: the fact that it is indefinable does not mean that we cannot say what things possess the property of goodness.

To show that goodness cannot be equated with any non-moral property, Moore developed what came to called the ‘Open question argument’. If the property X is good, then the question ‘Is it true that X is good?’ is meaningless. But the question ‘Is it true that X is good?’ is not meaningless. It is, in Moore’s words, an open question. Suppose X is pleasure, and that goodness can be defined as pleasure. The question ‘Is pleasure good?’ is not meaningless.  It is an open question whether pleasure whether pleasure is good. Therefore, Moore concludes X, whatever property it may be, cannot be equivalent to the good.

The trouble is Moore’s claim is less an argument than an assertion. Indeed, given that goodness is indefinable, it is difficult to know how one could construct an argument. Moore himself later acknowledged that ‘I did not give any tenable explanation of what I meant by saying that “good” was not a natural property’.

Moore’s answer to the second question that he raised – ‘What kind of actions ought we to perform?’ – was more straightforward, to the point, indeed, of being trite. ‘Duty’, Moore wrote, is defined as any action that ‘will cause more good to exist in the Universe than any possible alternative.’ ‘Right’ is ‘identical with “useful”’ and so ‘it follows that the end always will justify the means’ and that ‘no action which is not justified by its results can be right’. Moore, in other words, was a utilitarian, but one who thought that goodness could not be measured, or even defined, but was simply recognized, and intuited.

* * * * *

If the Principia Ethica were to be remembered for one idea only, it would be that of the ‘naturalistic fallacy’, which has become one of the defining ethical concepts of the twentieth century, though not necessarily in the sense that Moore had meant it. For many, Moore dealt a devastating blow to ‘naturalistic’ theories of ethics. Even those who have never been near the Principia routinely dismiss ethical arguments as being invalid because they exhibit the ‘naturalistic fallacy’.

The ‘naturalistic fallacy’ is, for Moore, any attempt to define the ‘good’, in other words, any attempt to define the indefinable. Moore calls the fallacy ‘naturalistic’ because he thinks it has been most committed those by those who have tried to define the good by equating it with some natural property.  He gives two examples. One is John Stuart Mill, who, like many utilitarians, equates the good with pleasure and with that which is desired.  Pleasure might be good, and the good might be associated with pleasure, but, Moore argues, pleasure cannot be a definition of good. Good and pleasure are distinct properties. Yellow is associated with electromagnetic wavelengths of between 570 and 590 nm.  But a particular wavelength is not a definition of ‘yellow’. So it is with pleasure and good.

Moore’s second example is that of the nineteenth century social evolutionist Herbert Spencer, the man who coined the phrase ‘The survival of the fittest’ to describe ‘the preservation of favoured races in the struggle for life’, a phrase that came for many to define the newly-discovered Darwinian process of evolution. Spencer, Moore suggests, claimed that more evolved species, and more evolved humans, are more ethical. He failed, in Moore’s eyes, to recognize the distinction between ‘higher’, ‘better’ and ‘more evolved’. And this led to the assumption that ‘good’ meant ‘more evolved’.

Many critics have pointed out that Moore misread both Mill and Spencer. Mill does not claim that good means pleasure, simply that pleasure provides the only criterion for goodness. Similarly, Spencer did not claim that ‘good’ meant the same as ‘more evolved’; rather he suggested that societies evolved in the same way as species did and that a more evolved society exhibited a better understanding of morality. Both Mill and Spencer were wrong in their understanding of the good, though not in the way that Moore assumed.

The problem is, however, deeper than a misreading of Mill and Spencer. The naturalistic fallacy, the philosopher Bernard Williams observes, is strictly neither a fallacy nor a critique of naturalism.  It is not a fallacy because, while those who try to define the good may, in Moore’s eyes, be mistaken, but they are not committing an error of reasoning, which is what is normally meant by a ‘fallacy’. It is not a critique of naturalism because, for Moore, the fallaciousness consists not in attempting to define the good in terms of a natural property, but in attempting to define the good at all, whether that be in terms of natural or non-natural properties. Moore’s naturalistic fallacy suffers, in other words, from the same incoherence as his concept of the good.

Previously, ‘naturalism’ had been contrasted with ‘supernaturalism’ and expressed the idea that ethics had to be understood purely in worldly, as opposed to divine, terms. In Moore’s view, however, Augustine and Aquinas were as guilty of committing the naturalistic fallacy as Spencer and Mill. The belief that the good can be defined as God’s word is as erroneous as the belief that good can be defined as happiness or evolutionary progress. Yet the idea of the ‘naturalistic fallacy’ has come to be wielded by theologians as much as by their critics, largely because of a one-sided view of the ‘fallacy’. In the wake of Moore, the naturalistic fallacy came broadly to be a way of expressing David Hume’s warning of the difficulty in deriving ‘ought’ from ‘is’, values from facts.  Moore’s criticism of naturalism is, in fact, very different from that of Hume. Hume was a naturalist in the old-fashioned sense of rejecting all and any divine explanation. But he also rejected the idea that there were any such things as ‘moral facts’. Moore accepted the existence of moral facts but did not consider them to be of the sort that could be established either through empirical observation or conceptual analysis.  Instead they were ‘self-evident’, or ‘intuitions’, beliefs that one knows to be true but for whose truth there is no evidence or reason.

* * * * *

In 1912 the Cambridge philosopher Henry Prichard published an essay in the journal Mind called ‘Does Moral Philosophy Rest on a Mistake?’.  The mistake, in Prichard’s eyes, was the belief that argument could settle moral questions. All demands for proof, empirical or rational, that something is a good or some act is a duty was an error. Hence moral philosophy, at least as traditionally understood, is a subject with no subject matter.

Pritchard, like Moore, saw moral truths as self-evident intuitions, though Prichard himself seems not to have recognized the similarity between his argument and that of Moore’s. Even more than Moore, Prichard disdained to make an argument in defence of his case. Like moral truths themselves, Prichard clearly saw his case as self-evident and intuitive. The idea of moral truths as intuitions harked back to the English Platonists of the eighteenth century. Prichard’s essay helped give those ideas new traction, launching the Cambridge Intuitionist school, that included WD Ross, EF Carritt, WHB Joseph and CD Broad.

For each of the Intuitionists the good was self-evident. The trouble was that the goods that were self-evident were not the same to all of them. Moore had, for instance, argued that one’s duty was to produce as much good as possible. Not so, responded Ross in his book The Right and the Good.  The right is like the good: a concept that is unanalysable, but intuitively recognized. Certain kinds of actions are right and wrong in themselves, without consideration of whether their consequences increase the amount of good in the world. Such duties, for Ross, included duties of fidelity (keeping promises, telling the truth, paying debts), of reparation (compensating for harm done),  of gratitude  (repaying kindness), of non-maleficence (not injuring others), of beneficence (improving the condition of others), of self-improvement and of justice (distributing goods rightly).  These duties are ‘self-evident’ in the sense of being ‘evident without any need of proof, or of evidence beyond itself’.  For Moore, however, not only were such duties not self-evident but they may not have been duties at all. If telling the truth creates more harm than good, then it is a duty to lie.

Since no empirical fact or rational argument could settle this debate, as the different claims were deemed to be simply ‘self-evident’, so the very notion of moral truth began to disintegrate. Many philosophers now followed through the logic of this. If moral truths cannot be verified, they suggested, perhaps they are neither self-evident nor truths, but merely expressions of personal preference, of feelings and emotions, of individual likes and dislikes. ‘Questions as to “values”’, Bertrand Russell wrote, ‘lie wholly outside the domain of knowledge.’ For Russell, ‘when we assert this or that has “value”, we are giving expression to our own emotions, not to a fact which would still be true if our personal feelings were different.’

So arose ‘emotivism’, first sketched by AJ Ayer in his groundbreaking 1936 book Language, Truth and Logic, a book through which he established himself as the leading English representative of logical positivism. Ayer argued, with other logical positivists, that there are two types of statements that convey meaning. Empirical statements express matters of fact whose truths can be established through observation and verification.  Analytical statements, such as mathematical truths, are necessarily true, though they cannot be verified empirically. Ethical propositions do not fall into either category. They are literally meaningless because they convey no meaning. ‘If I say to someone “you acted wrongly in stealing that money”’, Ayer wrote, ‘ I am not saying anything more than if I had simply said “you stole that money”. In adding that this action is wrong I am simply evincing my moral disapproval of it. It is as if I had said, “You stole that money” in a peculiar tone of horror, or written it with the addition of some special exclamation marks.’  The words ‘right’ and ‘wrong’ express not information but feelings. They also aim ‘to arouse feelings, and so to stimulate action’. Like Hume, Ayer insisted that when we talk of right and wrong we are not directly referring to things in the world but to our own attitudes towards these things.

The American philosopher Charles L Stevenson developed the emotivist argument, especially in his 1944 book Ethics and Language. Stevenson distinguished between what he called ‘descriptive’ and ‘emotive’ meaning. Factual statements have descriptive meanings.  They make claims about the world, can be true or false and which, if true, these claims constitute knowledge. Ethical statements, in contrast, possess emotive meaning; they do not make statements or convey beliefs or knowledge, and cannot be true or false. The meaning of ethical statements consists in their capacity to express and arouse feeling.  Stevenson is developing here an almost Sophist view of ethics, a view of ethics as essentially a technique of manipulation, as a form rhetoric, even of propaganda.

GE Moore was no emotivist, nor thought that values were simply subjective. Yet the argument he set running in the Principia Ethica led inexorably to Stevenson’s emotivism. The problem with this whole approach is the belief that the claim that ‘murder is wrong’ or that ‘one should tell the truth’ has no more force than the observation that ‘I like ice cream’ or ‘I think the Black Keys are cool’. I might think it odd if someone hated ice cream, or preferred Barry Manilow. But I recognize that these are simply personal preferences. In using terms such as ‘ought’ and ‘good’ I am, however, appealing to a standard that has greater authority, or at least that I want to have greater authority. For someone to think that it is right to murder people at will is qualitatively different from that individual thinking that Barry Manilow has a great voice. To suggest that slavery is a good would be more than simply ‘odd’. The trouble with emotivism is that it finds it difficult – nay, impossible – to capture this distinction.