Metaethical Subjectivism

Richard Double

• Publisher: Routledge
• Year: 2017
• eBook ISBN: 9781351918046
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

Metaethical subjectivism, the idea that the truth or falsity of moral statements is contingent upon the attitudes or conventions of observers, is often regarded as a lurid philosophical doctrine which generates much psychological resistance to its acceptance. In this accessible book, Richard Double, presents a vigorous defense of metaethical subjectivism, arguing that the acceptance of this doctrine need have no deleterious effects upon theorizing either in normative ethics or in moral practice. Proceeding from a 'worldview' methodology Double criticizes the rival doctrine of metaethical objectivism for lacking both 'completeness' and 'soundness', argues that a defense of metaethical subjectivism requires no special semantic analysis of moral language and defends the plausibility of metaethical subjectivism as explaining key intractable disagreements in moral philosophy. Double concludes by suggesting that the acceptance of metaethical subjectivism is better for constructing theories of normative ethics and moral practice than is metaethical objectivism.

Chapter 1
Metaethical Subjectivism and Metaethical Objectivism

The mathematician John Allen Paulos (1988, pp. 124–25) provides a problem he credits to Pascal. Two men bet $100 on who will win six coin tosses first. The game is interrupted after eight tosses with the first man winning five times and the second man winning three times. The contestants had made no agreement as to what would be done if the game were interrupted. So, how much money should change hands? One view is that the first man should win the $100 because he was leading when the game ended. Another view is that no money should change hands because the game was not completed. A third view is that the second man should award the first man ⅝ of the wager, which represents the percentage of coin tosses the first man won. A fourth view, which Paulos claims that Pascal favors, is that the second man should award the first man ⅞ of the wager: Had the game not been interrupted, the probability that the first man would have won is ⅞. (The second man needs to win three tosses in a row, which has the probability of ½ × ½ × ½ = ⅛, thereby making the first man ⅞ to win.) So, which of these four options is the most fair? I believe that there is no answer to this question, just as there are no objectively true answers to moral questions per se.

I shall argue that if metaethical objectivism were true, there would be answers to all sorts of moral problems that we acknowledge to be not only unascertainable, but to lack answers altogether. Examples of such substantive moral claims include:

• ‘In situation X, we should do a, all things considered.’
• ‘Reducing pain (to some specified degree) is more intrinsically valuable, ceteris paribus, than increasing pleasure (to some specified degree).’
• ‘Being merciful to a certain degree is a better character virtue than being just to a certain degree.’
• ‘It is worse (better) to deceive oneself than to deceive another.’
• ‘Preference act utilitarianism is a better moral theory overall than Kantian deontology.’
• It is a character flaw to be able to watch a sports event only if one roots for one of the teams or contestants.’
• ‘S deserves blame in this instance for doing a.’
• ‘Famous minority athletes should speak out in issues concerning social injustices within their sports.’
• ‘One should not try to be all things to all people.’
• ‘One may never do evil, even if doing so is stipulated to prevent a greater amount of evil of the same kind.’
• ‘Prosperous nations have a greater moral obligation to help their least prosperous citizens than they do to help much needier persons in less prosperous nations.’
• ‘It is better (worse) for our children to be overly optimistic, poor epistemic agents, who are energized and happy than it is for them to be good epistemic agents, whose accurate cognitions make them depressed and unsuccessful’ (Seligman, 1998).

These are examples of the many problems in normative ethics that I believe we can parlay into three strong arguments for metaethical subjectivism.

In section 1 of this chapter, I preface what I mean by ‘metaethical subjectivism.’ In section 2, I define ‘subjectivism’ and ‘objectivism’ for the three main areas of morality that I emphasize in this book: theory of moral obligation, theory of intrinsic value, and theory of character virtues. In sections 3 and 4, I discuss two types of meta-level methodological principles – metaphilosophies and intermediate-level philosophical principles – which motivate my overall argument. In section 5, I make explicit the presuppositions underlying my main argument of the book. In section 6, I anticipate an objection to that argument.

1 Metaethical subjectivism

Given the meta-level assumptions I elaborate in this chapter, the rest of the relevant beliefs I think are justified, and our conflicted moral intuitions over a multitude of moral issues, it is plausible (more than 0.5 per cent likely) to think that there are no objectively true answers to the main question for normative ethics, namely, ‘What should we do, all things considered (taking into account moral obligation, intrinsic value, virtue, and all other relevant considerations)?’ Although the familiar philosophical theories of moral obligation, intrinsic value, and character virtues have much, and sometimes decisive things, to say regarding the primary question, they speak in contradictory, conflicting, and incomplete voices, and, therefore, the most reasonable conclusion is metaethical subjectivism.

If we look at the situation from far enough away (from Thomas Nagel’s view from nowhere), we see that the main components of normative ethics – the theories of obligation, value, and virtue – are likely to be all objectively or all subjectively grounded. Questions about the reality of minds, personal identity, physical objects, abstract entities, and the existence of God are logically separate issues, which allow for creative metaphysicians to put together answers in various ways. The three major components of normative ethics, on the other hand, are intimately linked due to the contributions they make to the main moral question, ‘What should we do?’ Therefore, morality is more likely than nonmoral areas collectively to be entirely objective or entirely subjective. For example, if there were little reason to think that ‘What should we do?’ takes an objective answer, there would be little reason to think that the theories of moral obligation, value, and virtue are metaethically objective either. On the other hand, if obligation, value, and virtue were taken to be ontologically subjective, there would be little reason to think ‘What should we do?’ would take an objective answer. Even within the three components, it is difficult to see how, for example, two of the components might be objectively grounded and one of them subjective, or vice versa. The upshot is that if we can find some plausible reasons for subjectivizing any part of moral theory, we will thereby have a plausible reason to subjectivize the whole of moral metaphysics.

Mine is not the first argument from normative ethics to metaethical subjectivism.¹ J. L. Mackie (1977) provides two well-known arguments: the argument from the multiplicity of theories of moral obligation (the relativity argument) and the argument that moral properties would have to exemplify the strange combination of existing, yet be essentially motivating, a combination we posit nowhere else in our ontologies (the argument from queerness). Unfortunately, most metaethical objectivists think they have adequate ripostes to Mackie’s arguments. My argument in this book is in the spirit of Mackie’s, but it differs. Because I believe that moral obligation contributes only a small part of the answer to the what-should-we-do question, to find a metaethically objective moral theory, we should also use considerations from the theories of intrinsic value, character virtues, and other sources. When we attempt this holistic project, I think it becomes evident that any theory that there are moral truths would have to be ‘queer,’ not because they must satisfy Mackie’s internalism, but because their disorderliness makes subjectivism a better explanation for them than objectivism.

An obvious question to ask the author of any book such as this is how damaging to metaethical objectivism are the much-discussed standard objections to moral theories. (See Slote, 1992 for objections to commonsense, Kantian, and utilitarian theories; Mason, ed., 1996, for discussions of moral dilemmas; Stocker, 1990, for an extended consideration of moral pluralism.) I think such objections are strong reasons for adopting metaethical subjectivism. Nonetheless, I shall not rely on that literature. First, such objections are almost always treated as threats to normative theory, not metaethical objectivism, which is my target. (Bernard Williams, 1973 offers some suggestions about the metaethics of ‘ought’ statements, but does not push very hard.) Second, such objections, when typically examined in narrow detail, inevitably invite the objectivists’ stock reply that the fact that we do not know what the correct answers are does not mean there are no correct answers – as if this truism rebuts the subjectivist challenge. Third, I propose three different challenges to metaethical objectivism that I believe are broader and more difficult to stalemate than are the more familiar subjectivist arguments, good as I think many of them are.

2 Objectivism and subjectivism in three areas

In order to pin down my talk about ‘objectively true,’ here is how I understand ‘metaethical objectivism’ when applied to the three main parts of normative ethics I consider in this book:

• Objectivist Moral Obligation: ‘Actions may be permissible even if all persons believe them to be forbidden, and may be forbidden even if all persons believe them to be permissible.’
• Objectivist Intrinsic Value: ‘Actions (states of affairs and ways of life) may be intrinsically valuable even if all persons believe them not to be, and may lack intrinsic value even if all persons believe them to be intrinsically valuable.’
• Objectivist Character Virtue: ‘Character traits may be virtuous even if all persons believe them not to be, and may lack virtue even if all persons believe them to be virtuous.’

Metaethical subjectivism (moral nonrealism)

As the logical contradictory of metaethical objectivism (moral realism), I also equate in a single view the separately labeled theories of metaethical subjectivism or moral nonrealism. For me, metaethical subjectivism is the ontological position expressed by John Steinbeck’s preacher in The Grapes of Wrath: ‘There ain’t no sin, and there ain’t no virtue. There’s just stuff people do,’ and by Hamlet: ‘There is nothing either good or bad, but thinking makes it so.’ Nietzsche expressed metaethical subjectivism this way: ‘[T]here are altogether no moral facts. Moral judgments agree with religious ones in believing in realities which are no realities’ (Nietzsche, 1954, p. 501). My version of metaethical subjectivism proffers no account of what we mistakenly take moral facts to be, but simply denies that there are objective moral truths as defined above.

Metaethical subjectivism is different from noncognitivism, the philosophy of language thesis that moral judgments lack truth-values because they are, for example, expressions of emotion or approval of norms, attempts at persuasion, or prescriptions. Although all the noncognitivists I know turn out to be metaethical subjectivists in my sense, I view noncognitivism as a way to argue for subjectivism, to be distinguished from subjectivism-the-metaphysical-theory, which is my thesis. To see the difference we need only consider the two conceptual possibilities that each thesis is true and the other is false. First, subjectivism might be true, even if our best philosophy of language decides that most moral judgments (those other than pure emotive ejaculations) must be treated as indicative statements and, thus, all general noncognitive analyses fail. I think ‘X is wrong’ means ‘X is forbidden’ or ‘Not X is obligatory’ or some other real synonym – it means X is really wrong, just as common sense assumes.

Moreover, although moral language is used to express feelings, to recommend, to alter attitudes, to make prescriptions, and to perform other illocutionary acts, the very multiplicity of these functions seems to me strong reason to think that the meaning of moral utterances cannot be reduced to just one of these various roles. (Recall the Wittgensteinian/Austinian point that language has many functions.) Thus, it seems to me implausible on the face of it to try to show that the body of ostensibly indicative moral judgments is really so many commands, prescriptions, and so on. To use a Wittgensteinianism, only someone deeply under the spell of a philosophical program (for example the remains of a logical positivist methodology) would believe this.

The second conceptual possibility is that there might exist objective moral facts, even if some variety of noncognitivism fully accounts for the meanings of moral judgments. For example, it is conceivable that God might exist, even if some linguistic philosophers were (contrary to fact) correct in thinking that all apparently referential statements such as ‘God loves us’ really mean statements such as ‘I have confidence in the future.’ These two possibilities show that metaethical subjectivism, as I portray it, and noncognitivism are not equivalent and, indeed, do not belong in the same logical categories; subjectivism belongs to the metaphysics of ethics and noncognitivism belongs to the philosophy of language.

Besides not being committed to any noncognitivist theory of moral language, metaethical subjectivists in my sense are not committed to any specific cognitivist account either. For anyone who distinguishes between the metaphysics of morals and the philosophy of language, J. L. Mackie was right to hold ‘The denial that there are objective values does not commit one to any particular view of what moral statements mean’ (1977, p. 18). Cautious metaethical subjectivists will resist being dragged into the latter issue. Thus, subjectivists need not hold that moral claims mean the same thing as autobiographical or sociological claims, a point A. J. Ayer emphasized in Language, Truth, and Logic (1952, p. 107). For me, subjectivism does not claim that ‘X is wrong’ means or has the same truth condition as ‘I (or my society) disapprove of X.’ Nor, on my account, do moral judgments such as ‘X is wrong’ mean or even have the same truth conditions as more subtle claims such as ‘X has a feature that has the disposition to elicit the disapproval of normal humans’ or ‘Normal humans disapprove of X.’ Because metaethical subjectivism is the contradictory of metaethical objectivism, it simply denies that ‘X is wrong’ is capable of being objectively true.

Regarding the distinction between noncognitivism and error theory, the former holds that moral judgments can be neither true nor false because they make no statements that are capable of being true or false. So, Ayer’s emotive account of moral language, though consistent with metaethical subjectivism in my sense, holds that moral judgments lack truth-values. A view such as Mackie’s, which offers an ontologically subjectivist metaethics but not a noncognitivist theory of moral judgment, holds that moral terms purport to refer but fail. This is an error theory, which holds that all moral judgments are false in the way that ‘God is merciful’ is false, according to the atheist. Although I accept some noncognitivist accounts of some moral language, I do not endorse noncognitivism, and, thus, am an error theorist.

One may object: ‘If we accept subjectivism’s error theory, we see ourselves as speaking falsely when we seem to morally disagree; recognizing that this is the case, why do we continue to worry about moral disagreements?’ My answer is two-part. First, it is easy for even a subjectivist to get caught up in the heat of moral discourse and temporarily talk with the objectivists as if moral judgments are true. Subjectivists care passionately about moral issues, too. Second, despite our tendency to treat subjective judgments as objective, accepting subjectivism can help us to see that our passionate disagreements are not disputes over truth. Suppose a theist were deeply worried over how God, a morally perfect Being, could be both supremely just and supremely merciful? Here a subjectivist could reduce the passion by pointing out that the question presupposes metaethical objectivism: If we are subjectivists, then saying one of these qualities is morally better than the other must be an error. Although many persons prefer one trait to another, this preference is not grounded in moral facts about God, but rather in facts about the inquirers’ psychological constitutions.

3 Metaphilosophies

Philosophers try to accomplish different things by their moral theorizing. Some want to provide theoretical support for a moral theory (Kagan, 1989). Others wish to provide theoretical support for moral institutions (Gauthier, 1986; Rawls, 1971). Others combine moral theory and practice, trying in their philosophical work to make audiences morally better persons (Bernstein, 1998; Noddings, 1974; Singer, 1972). Some want to find a place for reason in ethics (Rachels, 1998). Some want to make sense out of their moral feelings by systematizing their moral intuitions, attitudes, and beliefs (Nagel, 1979, 1986, 1991). Others want to contribute to the critical acumen of moral theorizing without promulgating any single normative theory (Williams, 1981, 1997). Others, whose aim I share, want above all else to construct a picture of normative ethics and metaethics that integrates their vision of moral theory into the rest of their most plausible nonmoral theory (Harman, 1977; Mackie, 1977).

I use ‘metaphilosophy’ to mean a view of what philosophy is and what it is used for (Double, 1996). Any view of the purpose of a complex human activity (art, science, Little League baseball) is normative, because it reveals not only what we believe the activity can do, but what we desire the activity to accomplish or what goals of the activity we value. Metaphilosophies underpin our philosophical theorizing. Different philosophers bring different metaphilosophical stances to the problem of metaethics.

Although a larger taxonomy might be more useful, I limit my discussion to two. The metaphilosophy I endorse is Philosophy as Worldview Construction. Worldview metaphilosophy tries to construct an overall view of reality that is most likely to be accurate given our most reliable sources of evidence. The particular shape that Worldview metaphilosophy takes owes to the intermediate-level methodological principles we add to our metaphilosophies. Generally speaking, I take what I regard as the best truth-finding methods of non-positivist-science and enlightened common sense as the models for philosophical theory construction: letting entities into our worldview by inference to the best explanation and excluding them by Occam’s razor. Philosophy as Worldv...

Table of contents

• Cover
• Half Title
• Dedication
• Title
• Copyright
• Contents
• Acknowledgements
• 1 Metaethical Subjectivism and Metaethical Objectivism
• 2 Requirements for Metaethically Objective Moral Theories
• 3 What Metaethical Subjectivism Does Not Need to Provide
• 4 Moral Intuitions
• 5 Impartiality
• 6 Partiality
• 7 Maximization Troubles for all Moral Theories
• 8 The Fragmentation of the Moral
• 9 Moral Practice, Normative Ethics, and Metaethical Subjectivism
• Notes
• References
• Index