Emotivism and Internalism: Ayer and Stevenson
James Edwin Mahon (Washington and Lee University)
(Published August 2005)
Printer-friendly PDF version
(34 pages, 216Kb)
1. Introduction
In this article I am concerned with the nature of the relationship between the main non-cognitivist movement in ethics in first half the twentieth century, namely emotivism, according to which ethical judgments are the expressions of emotions, and internalism, the position in metaethics that ethical judgments are necessarily motivating in directed ways, such that an agent who judges that an action is ethically right or good is necessarily motivated to some extent to perform that action, and an agent who judges that an action is ethically wrong or bad is necessarily motivated to some extent to refrain from performing that action.
Today it is commonly assumed that the emotivists were internalists. It is also commonly assumed that they were prompted to choose emotivism over other cognitivist positions in ethics because of their commitment to internalism. Finally, it is also commonly assumed that they used an internalist argument to argue for emotivism.
In this article I shall argue that the connection between emotivism and internalism is far more tenuous than is commonly assumed. I shall focus on two of the most important and influential emotivists, A. J. Ayer and Charles L. Stevenson. While it is true that Stevenson was an internalist, Ayer refused to endorse internalism. Neither of them was originally prompted to choose emotivism over cognitivist positions in ethics by internalist considerations. Finally, Ayer never used an internalist argument to argue for emotivism, and while Stevenson did once borrow an internalist argument from another philosopher in order to argue for emotivism, he later abandoned it.
In order to make this argument, it is necessary to be clear about what constitutes internalism. In order to do this, it is necessary to say something about W. D. Falk and William Frankena, who together are responsible for coining the term ‘internalism’.
2. Internalism: Falk and Frankena
In his 1958 essay “Obligation and Motivation in Recent Moral Philosophy”,[1] William K. Frankena characterized internalism as the position that “motivation [is]… internal to obligation”, that is, that it is “not logically possible […] for an agent to have or see that he has an obligation even if he has no motivation, actual or dispositional, for doing the action in question” (Frankena, 1958, p. 40-1). Externalism was simply the denial of internalism—the position that “it is in some sense logically possible” (Frankena, 1958, p. 40) for an a person to have or see that he/she has an obligation to act and to have no motivation, actual or dispositional, to so act. In particular, internalism held that “judgments of obligation must be practical in the further sense that their being efficacious in influencing behavior is somehow logically internal to them”, that “motivation is somehow… “built into” judgments of moral obligation” (Frankena, 1958, p. 41).[2]
Frankena claimed to be borrowing the labels of internalism and externalism from W. D. Falk’s 1948 talk “‘Ought’ and Motivation”.[3] However, the terms “internal” and “external” came originally from Falk’s 1945 article “Obligation and Rightness”[4] (to which Frankena also referred (Frankena, 1958, p. 60, n. 24)), and they meant something slightly different. Since Falk’s contribution to metaethics has largely been ignored, it will be necessary to say something further about these writings by Falk.
In the earlier 1945 article Falk attacked the deontological intuitionists H. A. Prichard, W. D. Ross and C. D. Broad. These deontological intuitionists provided an account of moral obligation according to which “to “have an obligation” is to be faced with a need or demand for action purely external to ourselves” (Falk, 1945, p. 139). Falk provided at least five arguments against their “external claim” (Falk, 1945, p. 139) account of moral obligation. One of these arguments was that it made nonsense of the commonly accepted belief that judgments of moral obligation are essentially, and not merely accidentally, connected to the motives of the person who makes them. Anyone who makes a (sincere) judgment that he/she is morally obligated to act, according to Falk, is necessarily motivated, or necessarily has an incentive, at least to some extent, to act:
It is an implication of the “external fact” analysis of obligation that it is not essential for judgments about moral obligations to have an influence on him whose actions they concern, and who accepts them as true. Discussions about duty would be about no more than the existence of claims, or the characteristics of hypothetical actions; and even if we convinced another of his duty, we should only convince him of the existence of a state of affairs external to himself. We should, therefore, not have necessarily affected his inward attitude or brought him one whit nearer to feeling an incentive to do what he ought to do. But in fact when we try to convince another that he ought to pay his bills, we expect our argument if accepted to effect some change of heart in him, though it may still not change his outward actions. Discussions about moral problems are commonly carried on in the belief that in proving our point, and having it assented to, we shall provide one another with motives or “exciting reasons” (Hutcheson). We should think it odd to receive the answer: “Yes, I know now that and why I ought to pay my bills, but I am still without any incentive for doing so, and I, therefore, have as little cause for paying them as I had before I knew I ought to.” (Falk, 1945, p. 141 [emphasis in original])
Against this “external fact” account of moral obligation Falk provided his own account, according to which moral obligation is a dispositional state of person. On this account, “for an agent to be morally obliged is to be prompted in a given situation by the thought of the action which, on full reflection and when he is honest with himself, proves unalterably more impelling than the thought of any rival one.” (Falk, 1945, p. 144 [emphasis in original]) On this account of moral obligation, as he argued in the later 1948 talk, a person who makes the judgment that he/she has a moral obligation to act is necessarily motivated to act, at least to some extent:
Nor would we here allow that the questioner who while acknowledging a duty requested in addition a motive had a real axe to grind, even if his own good was at stake. His very request would be absurd as what he was overlooking was not some matter of psychological fact, but the logical implications of his saying that he had a duty. What might be called the “purist” view of the connection of “ought” with motivation would be justified, but only because of the special meaning assigned to the term. (Falk, 1948, p. 131)
Falk distinguished, then, between an account of moral obligation according to which moral obligation was a fact “external” to the person, and an account according to which moral obligation was a fact “internal” to the person—namely, a particular kind of motivational state of the person. In Falk, ‘internal’ and ‘external’ apply to the relationship between the obligation and the person, rather than to the relationship between the obligation and the motivation.
Nonetheless, Falk did indeed claim that it is “essential for judgments about moral obligations to have an influence on him whose actions they concern”, and argued that one reason why the alternative account should be rejected is that it made the connection between judgments of moral obligation and motivation contingent. It was his claim that the connection between judgments of moral obligation and motivation is a necessary connection—which his own account of moral obligation, in terms of a certain kind of motivational state of an agent, could explain—that Frankena usefully isolated, and christened ‘internalism’.
It is important to note that Frankena argued that the distinction between internalism and externalism “cut across… cognitivism versus noncognitivism” (Frankena, 1958, p. 42). One of the first examples of externalism that Frankena provided was “any form of noncognitivism which identifies moral requirements with social or divine imperatives” (Frankena, 1958, p. 43). Meanwhile, the first two internalists Frankena discussed were, in fact, cognitivists: G. C. Field (Moral Theory, 1921) and W. T. Stace (The Concept of Morals, 1937) (cf. Frankena, 1958, p. 47 n. 47).
With respect to the two emotivists I am concerned with, Ayer and Stevenson, Frankena said different things. He did identify Stevenson as an internalist, on the basis of his 1937 article ‘The Emotive Meaning of Ethical Terms’.[5] In this article Stevenson said the following:
In the second place, “goodness” must have, so to speak, magnetism. A person who recognizes that x to be “good” must ipso facto acquire a stronger tendency to act in its favor than he otherwise would have had. (Stevenson, 1937, p. 16)
On Frankena’s reading, Stevenson treats this as a conceptual claim, and not as a psychological claim:
He must, then, be thinking that assenting to a moral judgment in some sense logically entails its having a tendency to affect one’s action—that an analysis of a person’s moral judgment or recognition that something is obligatory must in some way involve a reference to his tendencies to do the action in question. That is, motivation must be “built into” the analysis of ethical utterances. (Frankena, 1958, p. 54)
However, in the case of Ayer, whom he groups with Alf Ross, R. C. Cross, Hans Reichenbach and R. M. Hare, Frankena said that “they seem to be insisting not so much that moral judgments are motivating, as that they are prescriptive. Hence they may not be internalists.” (Frankena, 1958, p. 47, n. 11)
It is worthwhile noting here why Frankena did not insist that prescriptivists were internalists. Prescriptivism holds that moral judgments are universal prescriptions or imperatives that express the decisions or wishes of those who make them. Reichenbach,[6] for example, held that “the linguistic expressions of ethics are not statements” (Reichenbach, 1954, p. 280), but are “moral directives… that express volitional decisions on the part of the speaker” (Reichenbach, 1954, p. 291). If a person says that “stinginess is bad” then “What you mean is a directive and thus an expression of your will, namely, you say: I wish that there were no stinginess.” (Reichenbach, 1954, p. 284) Although, on this account, ethical judgments are expressions of wishes, Reichenbach is unconcerned with the speaker’s own motivational state.
Hare[7] held that “no moral judgment can be a pure statement of fact” and that it is “part of the function of a moral judgment to prescribe or guide choices” (Hare, 1952, p. 29). He argued that “to guide choices or actions, a moral judgment has to be such that if a person assents to it, he must assent to some imperative sentence derivable from it” (Hare, 1952, p. 171). However, Hare did not discuss motivation. He did say that “It is a tautology to say that we cannot sincerely assent to a second-person command addressed to ourselves, and at the same time not perform it, if now is the occasion for performing it and it is in our (physical and psychological) power to do so.” (Hare, 1952, p. 20) For this reason, Hare can be counted as an internalist. However, motivation was not Hare’s concern either. Hence, while Frankena may be wrong in doubting that Reichenbach and Hare were internalists, it is true that motivation, and hence, internalism, was not of central importance to them, and not something for which they argued.
As his hesitation over prescriptivism indicates, Frankena did not assume that anyone who was a non-cognitivist was an internalist. However, since he argued that Stevenson, whom he identified as an emotivist, was an internalist, and since he classified Ayer as a prescriptivist rather than an emotivist, it may be that Frankena assumed that anyone who was an emotivist was an internalist. For arguments for this conclusion, however, it is necessary to turn to other commentators.
3. Meyerhoff, Nagel, Urmson, Martin and Darwall
Perhaps the first commentator to argue that the emotivists were internalists was Hans Meyerhoff. In his 1951 article “Emotive and Existentialist Theories of Ethics”,[8] Meyerhoff argued, without using the terminology of internalism and externalism, for the truth of internalism. He argued that emotivists were committed to internalism, albeit unconsciously:
I think it can also be shown that the same idea, namely, that a moral judgment differs sharply from a purely descriptive statement by virtue of the fact that the speaker is invariably involved in or committed to a certain course of action in the former, but not in the latter, emerges within the framework of an empirical theory, even though the proponents of the emotive theory do not always say so explicitly. A sentence expressing an attitude, emotion, or volitional decision differs from a purely descriptive statement not only by virtue of its concealed imperative nature, but also by virtue of its (at least potential) commitment to action. It is not only an emotion in verbal disguise — as Mr. Ayer generally insists — but also a potential or incipient type of action. (Meyerhoff, 1951, p. 777)
Some years later, in 1970, Thomas Nagel discussed internalism in The Possibility of Altruism.[9] Nagel claimed that, according to internalism, “motivation must be so tied to the truth, or meaning, of ethical statements that when in a particular case someone is (or perhaps merely believes that he is) morally required to do something, it follows that he has a motivation for doing it.” (Nagel, 1970, p. 7) The appeal of internalism, Nagel said, “derives from the conviction that one cannot accept or assert sincerely any ethical proposition without accepting at least a prima facie motivation for action in accordance with it.” (Nagel, 1970, p. 7) Nagel argued that “emotivism can be counted as an internalist position of sorts” (Nagel, 1970, p. 7), and offered what may be taken as an explanation of how emotivism is a form of internalism:
Philosophers who believe that there is no room for rational assessment of the basic springs of motivation will tend to be internalists, but at the cost of abandoning claims to moral objectivity. One way to do this is to build motivational content into the meaning of ethical assertions by turning them into expressions of a special sort of inclination, appropriate only when that inclination is present, and rooted only in the motivations of the speaker. The result is a basically anti-rational ethical theory, having as its foundation a commitment, inclination, feeling, or desire that is simply given (though the superstructure may be characterized by a high degree of rational articulation). Motivational content is thereby tied to the meaning of ethical utterances—what the speaker means or expresses—rather than to the truth conditions of those utterances, which are left vague or non-existent. (Nagel, 1970, p. 7-8)
Nagel did not name names, or quote from any emotivist. He may have been claiming that emotivism is a form of internalism unbeknownst to emotivists themselves. However, he did hold that emotivists had the conviction that “one cannot accept or assert sincerely any ethical proposition without accepting at least a prima facie motivation for action in accordance with it”.
For an argument that names names, it is necessary to turn to J. O. Urmson’s The Emotive Theory of Ethics in 1968.[10] Urmson argued that “the original ground for the proposal of the emotive theory was the need to find some way out from the unacceptable dichotomy of naturalism and non-naturalism” (Urmson, 1968, p. 18). That is, “the emotive theory of ethics has its origin in epistemological despair”, since “there is no account of the meaning of ethical utterances hitherto proposed which is epistemologically acceptable, since naturalism is unfortunately false and non-naturalism abhorrent.” (Urmson, 1968, p. 19) Urmson, however, contended that there was at least one “further ground for emotivism” (Urmson, 1968, p. 19) that was not epistemological in nature and that was a “more positive” (Urmson, 1968, p. 19) consideration in favor of emotivism.
Although Urmson did not have that name for it, his further ground for emotivism was internalism. He appealed to Hume’s contention that “Only a passion could directly move us to action” (Urmson, 1968, p. 19), and argued as follows:
To say that something is bad or wrong is to commit oneself to an attitude and to give what, if not disputed, is an utterly sufficient reason for avoiding that thing; similarly we have already embraced an attitude if we say that something is right or good, and we have given a good reason for its pursuit. If a drink is poisonous it is irrational to take it only if you have the will to live; without the will to live one can rationally say that the drink is poisonous but there is no reason for not drinking it. But one cannot rationally say that it would be wrong to take the drink but that there is no reason for not drinking it. If we accept both this and the previous argumentation we can conclude only that neither naturalism nor non-naturalism does justice to the nature of value judgements; it is their failure to account for the distinctive features of value judgements, and the role they play in our life and thought, that is now invoked, rather than any general epistemological contentions about the conditions of significant discourse. […] We now have a positive reason, connected with the observed powers of ethical judgements, to look for another sort of analysis that will explain these powers. (Urmson, 1968, p. 20-1)
Urmson’s use of “reason” in the passage quoted above, in phrases such as “reason for its pursuit” and “reason for not drinking it”, is best understood normatively, in terms of justification. However, his reference to the “powers of ethical judgements” is a reference to something else. It is a reference to motivation. The claim is that one of the “powers of ethical judgements” is that an agent who makes an ethical judgment is motivated to act in the relevant directed way, and that emotivism explains why this is so. What is interesting about Urmson’s invocation of this other ground for emotivism is that he highlighted the fact that this was not a consideration invoked by any of the emotivists before Stevenson:
In the writers whom we have so far considered there is certainly nothing approaching a full-scale statement of this argument. The nearest approach to anything more than the view that in making ethical judgments we are venting our emotions and thus letting off steam is to be found in such remarks of Ayer as: ‘It is worth mentioning that ethical terms do not serve only to express feeling. They are calculated also to arouse feeling, and so to stimulate action.’ We need not suppose that they were all unsympathetic to, or ignorant of, the line of argument that we are now considering. The truth is that they were all exclusively interested in epistemological questions of a very general character and would have regarded any detailed discussion of ethical matters for their own sake as being, in the context, a mere digression. (Urmson, 1968, p. 21)
Stevenson proves to be the exception to this rule about the emotivists not invoking the internalist argument for emotivism. Urmson claims that “The first occasion known to me on which anything like the positive argument for emotivism now under consideration was deployed in print” was in Stevenson’s 1937 paper “The Emotive Meaning of Ethical Terms” (Urmson, 1968, p. 22), and here he quotes the passage, referred to already by Frankena, about the “magnetism” of “good”. Urmson then says:
Ethical utterances, he [Stevenson] holds, are to be interpreted according to the emotive theory of the meaning of ethical terms not only because of the technical difficulties raised against the rival descriptive theories but because ethical judgements are seen to be dynamic; the emotive theory gives a positive explanation how ethical judgements can be what Hume called ‘influencing motives of the will’. At last we have some positive ground for ascribing an emotive meaning to value judgments. (Urmson, 1968, p. 23)
Urmson is correct that none of the emotivists before Stevenson, including Ayer, used this internalist argument in arguing for emotivism. He is also correct that in his 1937 article, Stevenson does present an internalist argument for emotivism. However, Urmson’s account of the relationship between internalism and emotivism has at least two implications. First, it implies that the other emotivists, including Ayer, would have endorsed the internalist argument for emotivism if they had been made aware of it. Second, it implies that the internalist argument was important to Stevenson in arguing for emotivism.
Recently, other commentators have argued for these claims. For example, in his On Ayer,[11] Robert M. Martin, in defense of Ayer, has offered the following internalist argument for emotivism:
The aspect of Ayer’s ethics that seems so unbelievable to many philosophers is that on his view ethical utterances are neither true nor false; they never state facts. But perhaps the following considerations might make this more plausible. It appears that believing a fact is one thing, and being motivated to do an action is something else. Coming to believe a fact sometimes produces motivation, but it never constitutes motivation all by itself. So if holding an ethical view were merely believing something, it could not by itself be a motive for action. But if, instead, holding an ethical view is having an attitude, it is being motivated. (Martin, 2001, p. 73)
According to Martin, although he never actually made the argument, Ayer would have used this internalist argument for emotivism, had he been made aware of it.
Stephen Darwall, meanwhile, has claimed that internalism – or as he calls it, judgment internalism—loomed large in arguments for emotivism:
What I call judgment internalism is the position that it is a necessary condition of a sincere or genuine ethical or normative utterance, thought, or conviction—for instance, that one should x—that one would, under appropriate conditions, have some motivation to x. According to judgment internalism, nothing counts as a genuine normative or ethical thought or utterance unless it has the connection to motivation. Internalism of this sort has loomed large in the twentieth-century arguments for ethical noncognitivism—that is, for the view that ethical judgments have no cognitive content and so cannot literally be true or false, but rather express noncognitive (in this case motivation-laden) mental states. (Darwall, 1995, p. 9-10)
In a footnote to this passage Darwall cites Stevenson’s 1944 book Ethics and Language (Darwall, 1995, p. 10 n. 22). The claim, then, is that internalism loomed large in Stevenson’s arguments for emotivism. Indeed, elsewhere[12] Darwall has argued that internalism is precisely that which constitutes the ‘emotive meaning’ of ethical terms:
He [Stevenson] found the crucial element of his account in a feature of ethical concepts that both Ross and Moore had largely ignored, what he called their ‘magnetism’… Ethical assertions have a distinctive force to them; if we sincerely and fully accept them we characteristically find ourselves moved. Indeed, Stevenson argued, it is precisely this connection to motivation and feeling, ‘the emotive meaning of ethical terms’, that explains what Moore had discovered but could explain only with metaphysical and epistemological mystery: the irreducibility of ethical concepts to natural concepts. (Darwall, 1989, p. 389)
Both the claim that Ayer would have endorsed the internalist argument for emotivism, if he had been made aware of it, and the claim that the internalist argument was important to Stevenson in arguing for emotivism, I want to argue, are false. First, it is false that Ayer would have used an internalist argument for emotivism, if he had been made aware of it, because Ayer refused to endorse internalism. He remained suspect. Secondly, it is false that an internalist argument was important to Stevenson in arguing for emotivism. Stevenson originally argued for emotivism without making any to appeal to internalism. The internalist argument for emotivism that he used in his 1937 article was borrowed from another philosopher, and Stevenson abandoned this argument in his later writings. In order to make these arguments, it will be necessary to examine the writings of Ayer and Stevenson.
4. Ayer and Internalism
Emotivism did not originate with Ayer. In The Meaning of Meaning in 1923[13], C. K. Ogden and I. A. Richards advanced a distinction (which itself was not wholly original[14]) between the symbolic, or referential, use of words and the emotive, or non-referential, use of words. They argued that the peculiar use of the word ‘good’ highlighted by G. E. Moore in Principia Ethica in 1903, according to which the word referred to a single, unanalyzable, non-natural property, was in fact a use of the word in which it did not refer to any property.[15] Rather, this use (and only this use[16]) of the word ‘good’ was an exclusively emotive use. The word was used to express an agent’s attitude, and to evoke a similar attitude in another agent, or to incite another agent to act:
But another use of the word is often asserted to occur… where ‘good’ is alleged to stand for a unique, unanalysable concept. This concept, it is said, is the subject-matter of Ethics [footnote reference to Moore’s Principia Ethica]. This peculiar use of ‘good’ is, we suggest, a purely emotive use. When so used the word stands for nothing whatsoever, and has no symbolic function. Thus, when we so use it in the sentence, ‘This is good’, we merely refer to this, and the addition of ‘is good’ makes no difference whatever to our reference… it serves only as an emotive sign expressing our attitude to this, and perhaps evoking similar attitudes in other persons, or inciting them to actions of one kind or another. (Ogden and Richards [1923] 1989, p. 125)
Although Ogden and Richards claimed that in the purely emotive use (and not meaning[17]) of ‘good’ in ethics, the speaker expressed an attitude, and evoked a similar attitude in their listener, they made no mention of motivation to act on the part of the speaker. There was no invocation of internalism, much less an argument for internalism. Thus the original emotivists, at the very least, were not consciously internalists, and were not prompted to argue for emotivism because of their commitment to internalism.
Ogden and Richards’s brief account of the purely emotive use of ‘good’ in ethics was not a comprehensive metaethical theory.[18] However, it did inspire Ayer to develop just such a theory. As Ayer said later in his 1984 essay ‘Are there Objective Values?’:
This [chapter 6 of Language, Truth and Logic] was one of the earliest attempts to develop in any detail what came to be known as the emotive theory of ethics. I say ‘to develop in detail’ because the theory had already been suggested by C. K. Ogden and I. A. Richards in their book The Meaning of Meaning, of which the first edition appeared as early as 1923. It was from them that I borrowed the word ‘emotive’ which they used to cover all aspects of language other than what they termed referential or symbolic… I must confess that I had read The Meaning of Meaning some years before I wrote Language, Truth and Logic, but I believe that my plagiarism was unconscious. If I had realized how closely I was following in the footsteps of Ogden and Richards, I think it very unlikely that I should not have acknowledged my debt to them. (Ayer, 1984, p. 28-9)[19]
Ogden and Richards were not the only people whom Ayer “plagiari[zed]”, however. In his autobiography Part of my Life in 1977,[20] Ayer said that “I put forward the view which had been suggested to me by Duncan-Jones that moral pronouncements were expressions of emotion rather than statements of fact.” (Ayer, 1977, p. 155) In a talk to the Aristotelian Society in 1934, “Is Goodness the Name of a Non-Natural Quality?”,[21] C. D. Broad summarized Austin Duncan-Jones’s views (Duncan-Jones himself did not publish them)[22] as follows:
Now it has been pointed out by Mr. Duncan Jones that it is not safe to let this assumption [that ‘good’ is the name of a property] pass without question… [W]e must remember that a sentence, which is grammatically in the indicative mood, may really be in part interjectional or rhetorical or imperative. It may be in part the expression of an emotion which the speaker is feeling. In that case to utter the sentence ‘That is good’ on a certain occasion might be equivalent to uttering a purely non-ethical sentence in the indicative, followed by a certain interjection. It might, e.g., be equivalent to saying: ‘That’s an act of self-sacrifice. Hurrah!’ Similarly, to utter the sentence ‘That is bad’ on a certain occasion might be equivalent to saying: ‘That’s a deliberately misleading statement. Blast!’. Again, a sentence may be used to evoke a certain kind of emotion in the hearer. In that case to utter the sentence ‘That is good’ might be equivalent to uttering a purely non-ethical sentence in the indicative in a pleasant tone and with a smile. To utter the sentence ‘That is bad’ might be equivalent to shouting a purely non-ethical indicative sentence at the hearer with a frown. Here the use of ethical words ‘good’ and ‘bad’ is merely a stimulus to produce certain emotions in the hearer, as smiling at him or shouting at him might do. In this case the sentence might be called ‘rhetorical’. Lastly, such sentences might be used to command or forbid certain actions in the hearer. To utter the sentence ‘That is good’ might be equivalent to uttering a purely non-ethical indicative sentence followed by a sentence in the imperative. It might, e.g., be equivalent on a certain occasion to: ‘That’s an act of self-sacrifice. Imitate it!’. To utter the sentence: ‘That is bad’ on a certain occasion might be equivalent to saying: ‘That’s a deliberately misleading statement. Don’t do that again!’. (Broad [1934] 1971, p. 107-8)[23]
Ayer developed, then, in greater detail, a metaethical theory that was provided earlier by Ogden and Richards and Duncan-Jones. However there remained one vital difference between Ayer’s emotivism and theirs. According to these earlier versions, ethical judgments were expressions of emotions that evoked the same emotions in others. But according to these earlier versions, the emotions expressed and evoked were common-or-garden emotions. They were the sort of emotions that were expressed by such interjections as ‘Blast!’ and ‘Hurrah!’. This is not what Ayer argued in 1936 in Language, Truth and Logic.[24] Ayer argued that ethical judgments were expressive and evocative of ethical or moral emotions. Ethical emotions were sui generis emotions, entirely distinct from other kinds of emotions. An ethical emotion could not be reduced to a non-ethical emotion, such as aesthetic pleasure, or anger.
Although Ayer began chapter 6 of Language, Truth and Logic, entitled “Critique of Ethics and Theology” by saying that judgments of value were “expressions of emotion” (Ayer [1936], 1946, p. 103), when he discussed ethics in particular (as opposed to aesthetics), he abandoned the term ‘emotion’ and instead used the terms ‘sentiment’, ‘feeling’ or ‘attitude’. The reason is that only terms such as ‘sentiment’, ‘feeling’, and ‘attitude’ are qualified by ‘moral’ or ‘ethical’ in plain English. Ayer was arguing that ethical judgments express and evoke ethical emotions that are different in kind from non-ethical emotions. Hence his use of the expressions “moral sentiment”, “moral feeling” and “moral attitude”:
For in saying that a certain type of action is right or wrong, I am not making any factual statement, not even a statement about my own state of mind. I am merely expressing certain moral sentiments. And the man who is ostensibly contradicting me is merely expressing his moral sentiments. (Ayer [1936] 1946, p. 107)
What we have just been saying about the symbol “wrong” applies to all normative ethical symbols. Sometimes they occur in sentences which record ordinary empirical facts besides expressing ethical feeling about those facts: sometimes they occur in sentences which simply express ethical feeling about a certain type of action, or situation, without making any statement of fact. (Ayer [1936], 1946, p. 108)
Ayer never explained precisely how ethical emotions differed from non-ethical emotions. He didn’t do this because this was not his theoretical concern. He may have even believed that this was the task of science. However, it is not at all clear that there is any empirical way to differentiate ethical from non-ethical emotions, and if there is not, then the distinction between the two is literally meaningless, by Ayer’s own lights. As Alexander Miller[25] has recently argued:
But are there observable behavioural occurrences which would constitute the expression of this special sort of moral or ethical emotion? It is difficult to see how Ayer could answer this in the affirmative: we can perhaps imagine patterns of observable behaviour which would express disapproval, but what observable behaviour could possibly manifest the presence of a distinctively moral or ethical sort of disapproval? (Miller, 1998, p. 111)
Miller argues that there cannot be any empirical test, in terms of observable behavior, to distinguish between ethical emotions and non-ethical emotions, and hence that Ayer must relegate ethical judgments “to the category of nonsense and verbiage along with the sentences of metaphysics” (Miller, 1998, p. 111).[26]
It is not my concern here to defend Ayer’s emotivism from Miller’s objection. However, it might be argued that if indeed Ayer was an internalist, he has a reply to Miller. Ayer could argue that ethical judgments have the special feature of being necessarily motivating. He could then argue that the empirical test, in terms of observable behavior, for the presence of ethical emotions in people, is that people who make ethical judgments, that is, express ethical emotions, are always motivated to act accordingly. He could distinguish between ethical emotions and non-ethical emotions, and hence defend emotivism, by way of an internalist argument.
The problem with supplying Ayer with this internalist defense of emotivism, however, is that Ayer refused to endorse internalism. In Language, Truth and Logic Ayer argued against one version of naturalism, what may be called ‘individual subjectivism’, as follows:
And we reject the alternative subjectivist view that a man who asserts that a certain action is right, is saying that he himself approves of it, on the ground that a man who confessed that he sometimes approved of what was bad or wrong would not be contradicting himself. (Ayer [1936], 1946, p. 104)
By “a man who asserts that a certain action is right” was meant a person who asserts that according to him/her a certain action is right, and not a person who asserts that according to the majority a certain action is right (Ayer had already rejected what may be called ‘group subjectivism’). Furthermore, by “approves” was meant morally approves. So Ayer’s claim was that a person who asserts that he/she morally approves of an action that according to him/her is bad or wrong is not contradicting himself/herself.
Worried that this claim is patently false, Ayer reconsidered it some years later in “Are there Objective Values”,[27] and argued as follows:
It is now not entirely clear to me that a man who said that he approved of what was wrong would not be contradicting himself. He plainly would not be if he were using the word ‘wrong’ in a descriptive sense as referring to the violation of some set of moral standards, which it would be open to him not to accept; but the context indicates that I intended his use of the word to be understood as evaluative, and surely when the word is so understood, and also when it is granted, as it was meant to be, that the approval in question was moral approval, there would be a logical contradiction in claiming to approve of what was wrong. I am inclined to agree, but still feel some hesitation. I am not fully persuaded that when Milton represents Satan as saying ‘Evil be Thou my good’ he is not making him voice a coherent moral policy; and one that consists in the violation of principles that Satan himself acknowledges. Diabolism would be a much tamer exercise if it consisted merely in violating a moral code which one did not share. (Ayer, 1984, p. 26)
Thus Ayer was not prepared to say that “Evil be Thou my good” is an incoherent moral policy. He was not prepared to say that diabolism was impossible. He was not prepared to say that it was impossible for a person to violate moral principles that he/she acknowledged, without practical contradiction. He was not prepared to say that it was impossible for a person to be motivated to do what according to him/her was wrong or evil or bad, because it was wrong or evil or bad. If Ayer was not prepared to say this, however, then he was refusing to endorse internalism, the position that ethical judgments are necessarily motivating in directed ways.[28]
It has already been argued that at no point in any of his writings did Ayer appeal to an internalist argument to argue for emotivism. Now it can be said that, since he did not endorse internalism, it is false that he would have used an internalist argument to argue in favor of emotivism.
5. Stevenson and Internalism
The intellectual background to Stevenson’s emotivism is the same as that of Ayer. In his doctoral dissertation at Harvard in 1935, entitled The Emotive Meaning of Ethical Terms[29], Stevenson said that he was led to emotivism by reading Ogden and Richards: “The suggestion came from Ogden and Richards, and from a discussion with Mr. R. B. Braithwaite, of King’s College, Cambridge.” (Stevenson, 1935, p. 196) He also acknowledged Broad’s article “Is Goodness the Name of a Non-natural Property?” (1933-4), containing the summary of Duncan-Jones, as well as W. H. F. Barnes’s article “A Suggestion about Value” (1934) (Stevenson, 1935, p. 196). About these authors he said: “I trust neither Mr. Broad nor Mr. Duncan-Jones are acquainted with the work of Ogden and Richards”, and “I am indebted to Mr. Broad and Mr. Barnes not for suggestions, since I did not become acquainted with their articles until quite recently, but wish to express my gratification that others should have come to the same conclusions from apparently different sources.” (Stevenson, 1935, p. 197)
Stevenson, then, developed his emotivism from the same source as Ayer—Ogden and Richards’ The Meaning of Meaning. In Ethics and Language in 1944, Stevenson quoted the passage from The Meaning of Meaning about the ethical use of ‘good’. Later in the book, in discussing a variety of non-cognitivists, he named Ogden and Richards as the original emotivists:
In the past decade or two philosophical analysts have given increasing attention to emotive meaning, and several have emphasized it in ethics. Their views have found some adherents, but a rather larger number of opponents. A. J. Ayer’s analysis has provoked heated discussion. Bertrand Russell, whose most recent views on ethical analysis are almost identical with Ayer’s, has not, in this connection, been so widely discussed—perhaps because the old-school moralists, after their previous attacks on him, have become too exhausted to repeat them. Rudolph Carnap has devoted several brief pages to the imperative and expressive functions of ethical statements; and C. D. Broad has summarized an emotive view, drawing from an unpublished paper of A. E. Duncan-Jones. An earlier account was given by Ogden and Richards in The Meaning of Meaning, and several others have contributed to the view in this or that detail. (Stevenson, 1944, p. 266)
It has already been shown above that Ogden and Richards were not concerned with motivation, and hence, with internalism, in arguing for emotivism. It should also be said that the other emotivists whose views Stevenson became familiar with—such as Duncan-Jones and Barnes—were also unconcerned with motivation, and hence, with internalism, in arguing for emotivism. Stevenson, as Urmson has argued, was supposedly different from these early emotivists insofar as he was concerned with motivation, and hence, with internalism, in arguing for emotivism. The evidence for this is the passage concerning the “magnetism” of ‘good’ in the 1937 article. However, it is false that internalism was important to Stevenson.
Stevenson’s claim in his 1937 article that goodness has a “magnetism” has a footnote reference to G. C. Field’s Moral Theory from 1921. Field, who was a cognitivist, advanced an internalist argument against Moorean intuitionism:
As we have just seen, Mr. Moore is compelled to say that the goodness of a thing must be thought of as a reason for aiming at it. But on his theory how can this be so? How can it be a motive for action? We are told that it is a simple quality which we perceive immediately. But our mere cognition of it cannot move us to action... On his theory, we can perfectly well desire the good, and therefore act so as to attain it. But it is the desire for it which moves us to action, and not at all the goodness of it. The goodness of it and our desire for it are two entirely different things and have no necessary connexion. It becomes, therefore, merely a matter of taste whether we desire what possesses this simple, indefinable quality or not, just as it is whether we like a particular colour or not. The mere fact of a thing being good can never by itself influence us to aim at it or move us to action. In fact, it is not necessarily of any interest to us at all. So that Mr. Moore’s account of it fails to fulfil one of our essential conditions […] (Field, 1921, p. 56-7)
In his 1937 article, then, Stevenson simply helped himself to Field’s argument, rendering it as the claim that goodness has a “magnetism”.
However, there are two things that should be noted here. First, Stevenson did not advance the internalist “magnetism” of goodness argument in his 1935 dissertation. The reason for this, I conjecture, is that Stevenson had not read Field at this stage. Second, Stevenson never advanced the internalist “magnetism” of goodness argument again. He did not advance this argument in Ethics and Language, for example. There is no reference to Field in that book.
The conclusion that should be drawn from these facts is that internalism was not all that important to Stevenson. Internalism was not an impetus for arguing for emotivism originally, and the internalist argument for emotivism that he borrowed from Field was only used once.
This does not mean that Stevenson was not an internalist. I believe that he was. The evidence for this is his claim about the “magnetism” of ‘good’ in the 1937 article, as well as the following passage from Ethics and Language:
Suppose that we are trying to convince a man that something he did was wrong. He replies: “I fully agree that it was, and for that very reason I am all the more in favor of doing it over again.” Temporarily puzzled to understand him, we shall be likely to conclude, “This is his paradoxical way of abusing what he considers our outworn moral conventions. He means to say that it is really all right to do it, and that one ought to do it flagrantly in order to discredit the many people who consider it wrong.” But whatever we may make of his meaning (and there are several other interpretations possible) we shall scarcely take seriously his protestations of agreement. Were we not trying all along to make him disapprove of his action? Would not his ethical agreement with us require that he share our disfavor—that he agree with us in attitude? (Stevenson, 1944, p. 16)
It is clear from this passage that Stevenson disagreed with Ayer about the possibility of diabolism. Stevenson did not accept that a person could consider an action to be wrong and “for that very reason” be in favor of performing the action. However, it is important to see that even in this passage Stevenson fell short of making an internalist argument. Stevenson’s argument was that if I convince another person that an action is wrong, I have managed to get the other person to have a disfavorable attitude towards that action. Hence, if the other person claims to agree that the action is wrong, but nevertheless claims to have a favorable attitude towards the action, I will consider him/her to be speaking insincerely. This is because one cannot sincerely hold that an action is wrong and also lack a disfavorable attitude towards this action. In making this argument, however, Stevenson did not go the extra step and argue that one cannot have a disfavorable attitude towards an action and not be motivated to refrain from performing that action. Hence he did not advance the internalist argument that one cannot sincerely hold that an action is wrong and not be motivated, at least to some extent, to refrain from performing it. Although Stevenson would endorse this argument, he remained unconcerned with it.
Consequently, even if Stevenson was an internalist, it is not true that internalism was important to him and that he was prompted to choose emotivism over other positions in ethics because of his commitment to internalism.
Conclusion
The connection between emotivism and internalism is far more tenuous than is commonly assumed. Although Stevenson was an internalist, Ayer refused to endorse internalism. Neither Ayer nor Stevenson was originally prompted to choose emotivism over cognitivist positions in ethics by internalist considerations. While Stevenson did use an internalist argument borrowed from another philosopher to argue for emotivism, he later abandoned this argument, and Ayer never used an internalist argument to argue for emotivism.[30]
Notes
[1] William K. Frankena, “Obligation and Motivation in Recent Moral Philosophy”, in Essays in Moral Philosophy, ed. A.I. Melden (Seattle: University of Washington Press, 1958), p. 40-81.
[2] It is interesting to note that Frankena says “I owe this use of the phrase “built into” to my colleague C. L. Stevenson” (p. 41, n. 2).
[3] Proceedings of the Aristotelian Society, New Series XLVIII (1947-1948), p. 111-138.
[4] Philosophy 20 (1945), p. 129-147.
[5] Charles L. Stevenson, “The Emotive Meaning of Ethical Terms”, Mind 46 (1937), p. 14-31.
[6] Hans Reichenbach, The Rise of Scientific Philosophy (Berkeley: University of California Press, 1954).
[7] R. M. Hare, The Language of Morals (Oxford: Clarendon Press, 1952).
[8] Hans Meyerhoff, ‘Emotive and Existentialist Theories of Ethics’, Journal of Philosophy 48 (1951), p. 769-783.
[9] Thomas Nagel, The Possibility of Altruism (Oxford: Clarendon Press, 1970).
[10] J. O. Urmson, The Emotive Theory of Ethics (London: Hutchinson University Library, 1968).
[11] On Ayer, Robert Martin (Belmont, CA: Wadsworth, 2001).
[12] Stephen Darwall, “Moore to Stevenson”, in Ethics in the History of Western Philosophy, ed. Robert J. Cavalier, James Gouinlock and James P. Sterba (NY: St. Martin’s Press, 1989), p. 366-397.
[13] C. K. Ogden and I. A. Richards, The Meaning of Meaning ([1923] London: Routledge and Kegan Paul, 1949).
[14] Ogden and Richards were influenced by Anton Marty, who in his Untersuchungen zur Grundelgung der allgemeinen Grammatik und Sprachphilosophie (Halle, 1908) divided sentences into “statements which express and evoke judgments, and emotives, which express and evoke desires, intentions, emotions and states of interest.” (Stephen Satris, Ethical Emotivism (Dordrecht: Martinus Nijhoff), 1987, p. 18).
[15] Frank Ramsey’s dictum that “Theology and Absolute Ethics are two famous subjects which we have realized to have no real objects” (“Epilogue”, in The Foundations of Mathematics and other Logical essays, edited by R. B. Braithwaite (London: Kegan Paul, 1931), p. 289) may have been inspired by his reading of Ogden and Richards. See his review of The Meaning of Meaning, where he says that “the distinction between the symbolic and emotive functions of language… seems to me to be of great importance, and in the emphasis laid on it lies the chief value of this book.” (Review of The Meaning of Meaning, by C. K. Ogden and I. A. Richards, Mind 33 (1924), p. 109)
[16] Stanley Cavell and Alexander Sesonske are anxious to point out that, according to Ogden and Richards, it was only Moore’s peculiar use of the word ‘good’ that was exclusively emotive, and not any other use of the word ‘good’: “The sort of ethical statements that are branded as meaningless are those which purport to denote Moore’s “simple quality”. Ogden and Richards indicate that it is this particular use of ethical terms that gave rise to the emotive theory and to the “purely emotive use” so often referred to by the positivists… [B]ut later emotivists have continued to use Ogden and Richards’ terminology without indicating, or perhaps without realizing, that it is just Moore’s peculiar usage which is being referred to.” (‘Logical Empiricism and Pragmatism in Ethics’, Journal of Philosophy 48 (1951), p. 10 n. 10)
[17] Richards emphasized this point in a later paper: “My distinction was between two different uses of language.” (“Emotive Meaning Again”, Philosophical Review 57 (1948), p. 147 n 4 [emphasis Richards])
[18] As Stevenson said, “In The Meaning of Meaning the emotive theory of value was stated with extraordinary brevity: it was outlined rather than developed.” (“Richards on the Theory of Value”, in I. A. Richards: Essays In His Honor, edited by Reuben Brower, Helen Vendler and John Hollander (NY: Oxford University Press, 1973), p. 121)
[19] See also his “Reflections on Language, Truth and Logic”, where he says “[T]he emotive theory which I put in its place was not my own invention. I was reminded quite recently that it had been advocated by C. K. Ogden and I. A. Richards in their book The Meaning of Meaning, which was published as early as 1923. Since I made no acknowledgment to them, this is a fact that I must have forgotten when I espoused the theory, though I was aware that the use by myself and others of the word ‘emotive’ to cover the aspects of meaning that were not ‘literal’, in the sense of issuing in truth or falsehood, was due to them.” (“Reflections on Language, Truth and Logic”, in Essays on Language, Truth and Logic, edited by Barry Gower (Totowa, NJ: Barnes and Nobel, 1987), p. 26) The influence was noted by E. W. F. Tomlin, one of his early reviewers, however: “Of My Ayer’s distinction between ‘emotive’ and ‘scientific’ terminology (derived ultimately from I. A. Richards)” (E. W. F. Tomlin, ‘Logical Negativism’, Scrutiny 2 (1936), p. 216).
[20] A. J. Ayer, Part of My Life (NY: Harcourt Brace Jovanovich, 1977).
[21] C. D. Broad, “Is Goodness the Name of a Non-Natural Quality?”, Proceedings of the Aristotelian Society, 34 (1933-34), p. 249-68, reprinted in Broad’s Critical Essays in Moral Philosophy, edited by David R. Cheney (London: Allen and Unwin, 1971), p. 106-123.
[22] Duncan-Jones did later review Stevenson’s Ethics and Language, and there appeared to be critical of Ayer’s interest in emotivism: “At last an able writer [Stevenson] has thought it worth while to produce a detailed study of the expletive theory of morals, which was set adrift on the tides of philosophical discussion by writers who were not strongly interested in ethics.” (Review of Ethics and Language, by Charles L. Stevenson, Mind 54 (1945), p. 362)
[23] Oswald Hanfling remarks (as many others have done) that “Ayer’s account of moral statements… became known as ‘the boo-hooray theory’.” (Oswald Hanfling, A. J. Ayer (London: Phoenix, 1997), p. 46) However, it is to be wondered whether it was Broad’s summary of Duncan-Jones’s position that inspired this appellation for Ayer’s position.
[24] A. J. Ayer, Language, Truth and Logic ([1936] London: Victor Gollanz (2nd. edition, 1946)).
[25] Alexander Miller, “Emotivism and the Verification Principle”, Proceedings of the Aristotelian Society 98 (1998), p. 103-124.
[26] See also Alexander Miller, An Introduction to Contemporary Ethics (Oxford: Polity, 2003), chapter 3, “Emotivism and the Rejection of Non-Naturalism”, p. 26-51.
[27] A. J. Ayer, ‘Are there Objective Values?’, in Freedom and Morality and Other Essays (Oxford: Clarendon Press, 1984), p. 17-34.
[28] What Ayer is considering here might be called the ‘immoralist challenge’ to internalism (as opposed to the amoralist challenge – see David Brink, Moral Realism and the Foundations of Ethics (Cambridge: Cambridge University Press, 1989, p. 46f; also Michael Smith, The Moral Problem (Oxford: Blackwell, 1994), p. 66f)). According to the immoralist challenge, an immoral person may be motivated to do what, by his own lights, is evil or wrong or bad, because it is evil or wrong or bad.
[29] Charles L. Stevenson, The Emotive Meaning of Ethical Terms, Harvard University, 1935.
[30] This article derives from my doctoral dissertation, Motivational Internalism and the Authority of Morality (Duke University, 2000). I would like to thank my dissertation committee, and in particular my adviser, Alasdair MacIntyre, for earlier comments. Research for this article at the Harvard University Archives was made possible by a Glenn Grant from Washington and Lee University; I would like to thank the university for its support, as well as the staff of the Archives. A draft of this article was completed on a pre-tenure leave as a Visiting Scholar in the Faculty of Philosophy at the University of Cambridge, and as a Visiting Fellow of Clare Hall, Cambridge, in the Michaelmas Term of 2003. I would like to thank the Faculty of Philosophy, including the graduate students and staff, and the Fellows and staff of Clare Hall, for their support. Versions of this article were read at a meeting of the Moral Sciences Club at the University of Cambridge in November 2003, and at the University of Virginia in November 2004. I would like to thank audiences on both occasions for their comments, and in particular Simon Blackburn, Hallvard Lillehammer, Jane Heal, Brian King, Daniel Elstein, Loren Lomasky, Talbot Brewer, Dan Devereux, James Cargile, Brie Gertler and Antonia Lolordo. Finally, I would like to thank two anonymous referees for Studies in the History of Ethics for their comments.