Are moral statements merely an expression of feeling?

In Greek, the word ‘Meta’ means above/beyond. Therefore, meta-ethics is concerned with questions that differ from normative studies of what particular courses of action are right or wrong. It is not, for example, concerned with whether or not rape is wrong, but what it might mean to say ‘rape is wrong’. As the scholar Andrew Fisher put it, it takes a “birds eye view” of ethics. If, in this analysis of ethical language, moral statements are concluded merely to only be an expression of feelings, this seems to render any notion of universality unfeasible. Emotions are inherently variable and so the stated point in the question arguably leads to the contradictory, normative realm of moral relativism and subjectivism. Instead of a central focus on rationality and deriving a collective, coherent ethical basis, unreasoned emotions, on their own, lead us nowhere. In society, a common ground of shared language, meaning and morality is essential for its just and proper functioning. The key question is ‘What right do you have to force your emotions upon me?’. If moral statements are merely an expression of feeling, then the answer is ‘no right at all’.

Despite various more detailed theories which I will come onto, views over ethical language can generally be split into two types; cognitivism and non-cognitivism. For cognitivists, moral statements describe the world – consequently, to have meaning, a statement must be able to be verified. This is inextricably linked with the philosophical doctrine of verificationism which, unsurprisingly, states that a statement or question is only legitimate if there is some way to determine whether the statement is true or false, or what the answer to the question is. Cognitivism is seen to fit in with realism; the view that moral truths are independently objective, transcending culture or language (however, Wittgenstein makes this comparison difficult). This is because once something is verified it is either true or false, and so the created connection between empiricism and morality means that we can categorise morality as either true/false. On the other hand, non-cognitivism is an anti-realist approach, postulating that moral statements cannot be verified as true/false, maintaining that they are instead based on more subjective factors such as coherence with other statements or emotions. It is this side of the two-speared debate that will, in some parts, tend to argue for the statement given in this question. In assessing the sub-divisions of these two strands, we can decide whether moral statements truly are merely an expression of feeling and emotion.

Firstly, there is the cognitivist, meta-ethical theory of ‘ethical naturalism’. This states that morality is not merely an expression of feeling but instead, ethical statements are objectively true/false due to being supported by, or by the lack of, facts. They are therefore propositional and so no different from non-ethical, natural conveyances. This notion was in part developed from empiricism and David Hume (despite the fact that Hume was not by any means an ethical naturalist). Naturalists develop their ideas with non-moral evidence. If we define goodness as pleasure, we may look at evidence of pleasure and pain in actions. If we say that goodness is whatever God desires, we will look for evidence of God’s purposes in the natural world. The issue is factuality and observability. A famous ethical naturalist, F.H. Bradley, argued that goodness is a natural aspect of society, as people reach “self realisation within the community”. By acknowledging our objective position in society, we can come to carry out particular duties. For ethical naturalists, moral conclusions need to be supported by facts in the following way: ‘abortion is wrong because it kills a fetus’. By giving a verifiable basis to the statement, it has meaning and so we can empirically show whether the statement is true or false. Consequently, ethical naturalism would lead us to believe that moral statements are not merely an expression of feeling, but relate to empirical, verifiable observation of the world around us.

However, ethical naturalism faces a major problem. ‘Abortion is wrong’ and ‘because it kills a fetus’, it is argued, are two different kinds of statement. One is an ethical statement and the other is merely a factual observation. One could easily just say ‘abortion is right because it kills a fetus’ and because it is still factually supported, it seemingly has no more/less moral viability. This criticism is known as ‘The Naturalistic Fallacy’, most prominently espoused by G.E.Moore, but developed from what was known as David Hume’s ‘guillotine’; a clear cut attempt to distinguish between facts and ethics. For Moore, “you cannot derive an ought for an is”. This is supported by what has come to be known as the ‘open question argument’ – something directed at any ‘naturalistic’ moral statements. For example, if someone were to claim that hedonistic pleasure is a naturally good thing, we could always ask ‘Is pleasure good?’. But if pleasure really were naturally good, this question would make no more sense that asking ‘Is pleasure pleasurable?’ (i.e. it should be self-evident but it is not). Consequently, Moore stated that you could not draw a connection between a moral axiom and a natural relation to it. As we will soon come to see though, this is not as obvious a flaw of naturalism as it may at first seem. Yes, the ethical statement is supported by facts but there is then the potential for these facts themselves to be justified in a more normative way. For example, ‘abortion is wrong because it kills a fetus’ and killing a fetus is wrong due to the categorical imperative and the universalisation principle. Consequently, the fact at hand can be ethically ‘right or ‘wrong; by reference to normative justifications. Later in this essay this will be explored more, when it will become evident that even an apparently non-cognitivist theory can lead to the use of facts within moral statements and therefore correspond to some naturalistic account of ethical language. Despite this though, Moore argued that our morals are non-natural (not apparent in the world around us). In ‘Principia Ethica’ (1903), his argument was that good “cannot be defined”, offering a meta-ethical alternative known as ‘Intuitionism’.

Moore argues that “good is good, and that is the end of the matter”. It is an indefinable, completely simple and unanalysable idea. Moore compares this to the colour yellow – ‘yellow’ can only be known directly through intuition - yellow is just yellow, and that’s all there is to it. We don’t explain it in terms of something else due to the fact that it is irreducible by its very nature. Moore claims that goodness is just the same; you can’t explain it any further (“it cannot be defined, and that is all I have to say about it”). It then follows that we have an intuitive sense of right or wrong: almost by instinct we know what is valuable (relationships, beautiful things). Moore did not think that we can prove an intuition, and it may be that our intuitions are wrong, consequently leaving ethical debate in quite a weak position. This though, once again, is linked to realism. Whether these moral intuitions come from evolution and genetics, a transcendent source of innate command, or our societal subconscious, they are still classified by Moore as independently objective moral truths, despite not in themselves being justifiable via reasoned argument.

In a similar vein to Moore, the philosopher H.A. Pritchard argued that moral obligations form immediate apprehensions, a bit like mathematics. However, Pritchard thought that our intuitions involve more than just goodness, but also a sense of obligation.  There is a gap between saying that something is good and saying that one ought to do it. Just like Moore’s understanding of goodness, Pritchard states that ‘ought’ has no definition; it is merely recognisable to all of us. So, resolving a moral dilemma involves weighing up contrasting obligations and trying to work out which is most important. It is our reason that looks at the facts of a certain situation, and our intuition that decides what to do – moral differences are explained by varied levels of ethical development, relating to the psychological, conscience-based theories of figures such as Piaget and Kohlberg. Strangely though, postulating the idea that our reason looks at the situation to simply be followed by an intuitive decision of then what is right after drawing from those conclusions sounds suspiciously like ‘ethical naturalism’ (i.e. basing moral judgements on a verification and reasoning of particular facts). This, another similar problems are connected to the inherent ambiguity and variable nature of basing ‘goodness’ on intuition. Also, it is very hard to be sure with Pritchard if there is a ‘right way’ to solve moral dilemmas, leaving the notion of goodness in no-mans land. It is therefore a third intuitionist scholar, W.D.Ross, who tries to add a bit more weight to the theory. Relating to his more modern formation of Kant’s Categorical Imperative, in which he tried to deal with the issue of conflicting moral duties, he set out his 7 ‘Prime Facie’ duties, which are apparently right via intuition – these are fidelity, gratitude, justice, self-improvement, reparation, non-maleficence and beneficence. Ross thought that we would choose between these on the basis of judgement; we do not have a logical system for working it out. In this sense, his theory is truly intuitive. Therefore, even though in a very different sense to the concept of ‘emotivism’, which I will soon come on to, the meta-ethical theory of ‘moral intuitionism’ does seem to suggest that on some level, moral statements are merely an expression of feeling. They are not necessarily as arbitrary and discrete as some of the non-cognitive approaches, but by being so indefinable, intuitions face many of the same irresolvable criticisms.

Intuitionism fails by its own standards though. It is meant to be a cognitivist theory, linking itself to realism and verificationism. However, intuition is not a proof of any sort – you have merely stated that it has come from your intuition and there is no due process of objective verifiability. Also, intuitions alone are not a reliable step towards the necessary academic conflict and reasoned debate needed to progress. Unlike the colour yellow, which is a uniform intuition, experienced to be the same by everyone, this is not the case with ‘goodness’. If what is ‘good’ is based purely on my intuitions, then this leads to the solipsistic tendencies of individualistic subjectivism because my inclinations are subservient to external factors. Intuitions may tell us something about morality but what is ‘good’ must be complemented by something far more concrete and universal, leading to cosmopolitanism; the ideology that all human ethnic groups belong to a single community based on, what is to some degree, a shared morality. It sets out no way to deal with moral disputes.

Many would subsequently argue, that from our evaluation, it is necessary to therefore move away from cognitivism and its focus on morality as an essentially verifiable fact.    Due to the fact that G.E.Moore’s response to naturalism fails by its own standards, does this not suggest that ethical language is non-cognitive and anti-realist? Firstly, we need to consider the realm of emotivism and primarily A.J.Ayer– the approach that most agrees with moral statements merely being an expression of feelings.

As a meta-ethical scholar though, A.J.Ayer’s context is slightly more complicated. Emotivism has its roots in the Vienna Circle of the 1920s and in the doctrine of ‘logical positivism’ – which holds roughly that for any truth claim to be meaningful, it must be verifiable. In his book, ‘Language, Truth and Logic’, written at the young age of 25, A.J.Ayer said that there were only two types of meaningful statement. Firstly, analytic statements such as ‘all bachelors are unmarried men’; their truth is derived from their very own definition and an understanding of the terms occurring in the particular claim. Secondly, there is the synthetic category such as ‘it is raining outside’ which is checkable and requires empirical evidence to be able to tell whether it is right or wrong. In this sense, with Ayer’s supreme focus upon the verification principle, it seems that his basis is not much different from the realm of ethical naturalism and its descriptions of ethical language.

However, A.J.Ayer differs greatly – for him, moral statements are neither analytically true nor synthetically verifiable, and so are consequently meaningless. Instead, he sees them merely as an emotional statement and an expression of preference, attitude, feeling etc… over the issue at hand - therefore, as already mentioned, it is emotivism which is best placed to agree with the statement in question. “Ethical terms do not serve only to express feelings. They are calculated to arouse feeling, and so stimulate action” – but, crucially, they are not ‘true’ in any sense of the word (for Ayer at least). It is for these reasons that the meta-ethical theory of emotivism has also been labelled the ‘boo-hurrah theory’. If I say ‘murder is wrong’, all I am really saying is ‘boo to murder’ (expressing my personal disapproval) and so the theory can really just be classed as ‘simple subjectivism’. As Ayer put it, “in saying a certain type of action is right or wrong, I am not making any factual statement…I am merely expressing certain moral sentiments.” However, in this theory it is not the case that all emotive statements are equal. Moral statements arouse feelings, but with three different ‘strengths of command’. Implying a duty is the strongest form, whereas saying that one ‘ought’ to do something is less wrong and finally, merely stating that something is good/bad is particularly weak. Despite these different intensities as documented by Ayer, they are all apparently still the result of emotional stimuli, and so do nothing to appeal to reason.

C.L.Stevenson, another emotivist scholar, adds to Ayer’s work. For Stevenson, emotions are not as fickle and meaningless as implied through the rigid, logical positivist work of his predecessor. He says that for non-cognitivists, moral meaning is, and should be, located in individual subjective truth, not in objective truth/fact and so consequently, our emotions have their own virtue. Moral words and statements are not arbitrary but based on our own unique experience of the world and how we want it to be. They are not based on a “fleeting mood of the day” as many people often claim emotivism claims, but instead, are built up through formed life experience and so are used to try and persuade others to have a similar emotional response. Once again though, even if moral statements are based on genuine emotions, this still does nothing (like Ayer’s version of the theory) to lead us towards universalisation of morality, with my emotions being no more ‘right’, coherent or correct than yours. I cannot judge you or try to ultimately convince you that my views are more ‘just’ or in line with the human condition – so society and the very premises of our existence fall apart.

Despite trying to avoid the problem of speculative and metaphysical morality, as James Rachels points out so well, moral judgments appeal to reason and the way in which we establish beliefs by determining that one doctrine is more acceptable or coherent than another; not via mere emotions and feelings (if anything, the place of emotions in ethical decision-making is derived from prior moral reasoning). Surely something like genocide is ‘wrong’ due to its inability to fall in line with a moral code. As Stephen Pinker has said, reason and rationality inevitably leads to some form of collectivity and cosmopolitanism in our moral outlooks. Even though a meta-ethical theory, it seems that emotivism, as well as intuitionism, render what I see as the incredibly important normative debates of moral philosophy, meaningless. What makes Mill more right than Kant? Nothing does, for both are ‘just expressions of feeling and emotion’. It is primarily for this reason that I do not think moral statements are merely an expression of feeling, and it is for this same reason that Peter Vardy has described emotivism as an “ethical non-theory”.

R.M.Hare exposes these flaws well by rejecting what he sees as ‘emotional reductionism’. Morality is just too complex to reduce it to statements of ‘I believe killing is wrong’ and leave it at that. This is not a proper account of ‘goodness’. It is for this reason that Hare suggests that instead, moral statements are prescriptive instead of subjective. ‘Prescriptivism’ can still be classed as an anti-realist approach for it does not postulate the existence of independent moral truths, but instead it appeals to inter-personal argument and debate, whereby moral statements can be described as imperatives. By saying ‘abortion is wrong’, I am implicitly implying that in this moral situation/dilemma, you should not have an abortion. It is attempting to achieve consistency in moral judgments – something necessary for a modern humanistic and collective morality. In this sense, it subsequently avoids the dangerous territory of subjectivism that accompanies emotivism, but does not stray too far into the rigid notion of verifiability that encompasses cognitivism – this is why I think it is by far the best attempt at a meta-ethical account that delivers a basis on which we can revert back to the deeply important questions of normative ethics. Even though some have critiqued prescriptivism on similar grounds to emotivism, stating that we have no reason to accept one person’s prescription over someone else’s, this does not work. Prescriptions are varying moral outlooks and reason determines the winner. As Hare said in ‘Freedom and Reason’ in 1963, showcasing his version of preference Utilitarianism, one should put themselves ‘in another’s shoes’ in condemning certain moral statements and judgments such as prescribing that all individuals of a certain race should be exterminated. Prescriptivism does not therefore support the notion that moral statements are merely an expression of feeling.

However, as I earlier mentioned, there is an extra twist with the notion of ‘prescriptivism’. For Hare, we used ‘good’ prescriptively in relation to standards – for example, a ‘good’ chair is one that tends to support your back. When we use the word ‘good’, we are therefore referring to a set of external standards, derived normatively. Crucially here though, even though the standard we reference can go on to be justifiable via normative debate, we still originally justify our prescription with a fact. The chair is good ‘because it supports your back’; fact. Is this not ethical naturalism? There are two levels of verification here then – simply categorising moral statements as being about the natural world, being ultimately empirically verifiable, is not enough. The nature of moral statements is, and must be, prescriptive, and so the facts need to be justified as representing a moral ‘good’/’bad’ – they are not enough on their own.

In conclusion, what this issue has shown is the complexity and inter-connectedness of the depths of meta-ethics. Despite this, I hope to have done away with the individual, subjective tendencies of approaches such as intuitionism and emotivism that do nothing to account for the vast complexity and inherent collective nature of morality. If ethical language/statements are not reasoned and prescriptive, but merely expressional and personal, what basis is there to say that my moral compass is any more justifiable than the regressive doctrines of Nazism? The difficulty is finding a morality that we can class as ‘true’ for humanity, but one that should be followed without appealing to realism that can both overly-empiricise morality as well as often relate to metaphysical explanations. R.M.Hare seems to come the closest though. With moral statements being an expression of feeling, lets abandon society, our justice system and every single shared moment of reason that has ever existed.