Ethics in the Wake of Wittgenstein

Benjamin De Mesel, Oskari Kuusela, Benjamin De Mesel, Oskari Kuusela

• Publisher: Routledge
• Year: 2019
• ISBN:
  9781351721530
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

This book brings together essays from leading scholars who, rather than taking a strictly exegetical approach, attempt to show how discussions in moral philosophy can benefit from Wittgenstein's later philosophical work. The essays in this volume make the argument that Wittgenstein's relevance for moral philosophy depends not only on his views about ethics, but also on the methods he introduces, on his views on the nature of philosophy and philosophical problems, and on the insights into language developed in his philosophy. They also focus on the 'Wittgensteinian tradition' in moral philosophy and its relation to more mainstream analytic moral philosophy, addressing how several prominent philosophers use these ideas and methods in their work. Ethics in the Wake of Wittgenstein seeks to answer the following question: Can we apply Wittgenstein's ways of dealing with problems in logic, philosophy of language, epistemology, philosophy of mind, and philosophy of mathematics to moral philosophy as well? It will be of interest to Wittgenstein scholars and those working on current debates in moral philosophy, metaethics, and normative ethics.

1
Logical-Linguistic Method in Moral Philosophy

Resolving Problems From Iris Murdoch and Bernard Williams With Wittgenstein

Oskari Kuusela

This chapter discusses the logical role or function of philosophical accounts, with the purpose of explaining how Ludwig Wittgenstein’s later method of the use of philosophical clarificatory models can help to resolve certain problems that arise in connection with more traditional conceptions of the function of moral philosophical theories. With regard to the problems in question, Iris Murdoch criticized moral philosophy and the linguistic method of analytic philosophy of her time for narrowness and ahistoricity, for imposing a false unity on the phenomena of morality and for presenting unacknowledged evaluative views in the guise of neutral analyses. Similar criticisms were later made by Bernard Williams. I agree with the critical points raised by the two philosophers. But although their work contains responses to the mentioned problems, neither spells out explicitly or in detail the solution they think is required, or explains how moral philosophy, as they propose to reconceive it, avoids collapsing into empirical moral anthropology, sociology or psychology. This is where Wittgenstein’s later philosophy of logic and philosophical methodology can help.

1. Murdoch and Williams on Logical-Linguistic Method in Moral Philosophy

Murdoch and Williams are highly critical of what they describe as the linguistic approach in analytic moral philosophy, although neither regards the linguistic approach problematic as such. Rather, the problems they raise pertain to how this approach has actually been pursued. As Williams writes:

Moral philosophy is one area of philosophy in which the “linguistic turn”, as it has been called, has not helped to give problems a more tractable shape. This is not to deny that moral philosophy, like other parts of philosophy, is properly concerned with reflection on what we say. Indeed, at one level it might have done better than it has if it had been more concerned with what we say.

(Williams 1985, 127; for the notion of linguistic turn, see Rorty 1992; Hacker 2013)

Similarly, Murdoch comments on moral philosophy in the 1950s:

[T]he main task is the task on which moral philosophy is in fact engaged—the analysis of contemporary moral concepts, through moral language. I have suggested that this task has been too narrowly conceived. We have not considered the great variety of concepts that make up morality.

(Murdoch 1998, 73)

The point, therefore, is not that philosophy should not concern itself with reflecting on what we say, and that this would not be helpful in moral philosophy. Both philosophers agree that more reflection on language, rather than less, and with a focus on a wider range of moral concepts, would be beneficial. Indeed, as Murdoch suggests, taking the linguistic method seriously would require adopting such a broader focus. “Philosophers have usually tended to seek for universal formulae. But the linguistic method, if we take it seriously, is by its nature opposed to this search” (Murdoch 1998, 74). Likewise, Williams remarks, in a context where he comments on the focus of moral philosophy on the so-called thin, abstract moral concepts, good, right and ought, and on moral philosophy’s attempt to explain reductively, in terms of thin concepts, the function of the more specific thick moral concepts, such as brutality, courage or kindness (Williams 1985, 128; see ahead):

Its prevailing fault, in all its styles, is to impose on ethical life some immensely simple model, whether it be of the concepts that we actually use or of moral rules by which we should be guided. One remedy to this persistent deformation might indeed have been to attend to the great diversity of things people do say about how they and other people live their lives.

(Williams 1985, 127)

Relatedly, both find problematic the ahistorical and supposedly morally neutral character of linguistic moral philosophy, conceived as logical analysis of universally assumed moral notions, as well as the failure of linguistic philosophy to take into account the social character of morality (Murdoch 1998, 74, 84; Williams 1985, 131).

As this summary indicates, Murdoch’s and Williams’s criticisms of the logical-linguistic method in moral philosophy are remarkably convergent. So much so that it is possible to present their criticisms as centred on four common, interconnected themes: (1) A problem with the alleged moral neutrality of proposed analyses of moral concepts. While presenting their accounts of morality in the neutral guise of analyses of any morality whatsoever, philosophical analyses of ethical concepts have been influenced by the philosophers’ own “local” moral views, ignoring other possible ways to conceive of and explicate morality. (2) The use of simplistic models to explain morality and the imposition of false unity on the phenomena of morality through these models, including a simplistic account of conceptual unity. (3) Narrow focus on certain moral concepts at the expense of others. (4) Ahistoricity, the failure to recognize or acknowledge the historical character of moral concepts and the historical variety of moral outlooks. Let us look at each of these criticisms in turn.

The first criticism seems to echo Nietzsche’s point that, whilst seeking to ground morality, philosophers have merely found a new expression to the dominant morality (Nietzsche 2002, 75–76). As Murdoch observes, expressions of different attitudes towards the world and concern with different aspects of moral life are found in the work of moral philosophers before linguistic moral philosophy, for example, in H.W.B. Joseph, A. E. Taylor, W. D. Ross, and H. A. Prichard. But, she asks:

Can it be safely assumed that linguistic philosophers are immune from such partiality, being able to derive from the study of language some sort of initial definition and subsequent analysis of morality which shall have the prestige and neutrality of logic?

(Murdoch 1998, 76)

Although linguistically oriented moral philosophers have indeed professed such neutrality, that is, “to analyse the essence of any morality, to display the logic of any moral language” (Murdoch 1998, 65), according to Murdoch, they have failed to achieve it. Instead, “[w]hat the modern moral philosopher has done is what metaphysicians of the past have always done. He has produced a model. Only it is not a model of any morality whatsoever. It is the model of his own morality” (Murdoch 1998, 67). Consequently, “[…] philosophers have done their moralizing unconsciously instead of consciously” (Murdoch 1998, 74; cf. point 4 ahead). Similarly, Williams believes that the focus of moral philosophers on the thin moral concepts has partly contributed to their own ethical assumptions informing their allegedly neutral analyses or explications:

A […] reason why the concentration on these general terms has done no good to the linguistic philosophy of ethics is that the theorist, in trying to sort out the relevant uses of them, brings to the inquiry presuppositions that are not only already theoretical but already ethical. The results are usually bad philosophy of language.

(Williams 1985, 128)

Second, regarding the use of simplistic models, Murdoch emphasizes the importance attending to “the initial delineation of the field of study”, and of considering the relation between the phenomena selected for study and the philosophical technique used to describe them. Moral philosophy, she points out, risks running in a circle whereby a narrow selection of phenomena for study suggests certain particular techniques of describing them, which, in turn, make other phenomena of moral life inaccessible to philosophy (Murdoch 1998, 76; for discussion, see Diamond 2010). An example is R. M. Hare’s view of moral principles as universalizable prescriptions of which Murdoch says that this “‘universal rule’ model” blurs a “real difference” between different moral attitudes that respectively emphasize or deemphasize the uniqueness of an agent’s personality or position. From a perspective that emphasizes the uniqueness of the agent—for example, that no one can have their destiny—considerations of universalizability might emerge as pointless (Murdoch 1998, 86; for another example, see 87–88).¹ Regarding Murdoch’s own moral philosophy, her account of moral differences as conceptual differences, relating to different morally relevant ways of conceptualizing reality, and her emphasis on the inner life of consciousness, as opposed to the emphasis of moral philosophy on external action and choice, provide examples of different ways of delineating the field of study. In this way, Murdoch’s own work and engagement with the moral philosophy of her time testifies to how the latter is in danger of eclipsing important moral phenomena by imposing a simplistic model on moral life, with Murdoch herself trying to draw attention to such neglected phenomena of moral life or aspects of morality (Murdoch 1998 77ff., 89, 2001, 2). She comments on this: “The current model illuminates and describes only a certain type or area of moral life, and […] if we attempt to construe all moral activities in terms of it we are led to ignore important differences”. (Murdoch 1998, 92). And:

Philosophers have been misled, not only by a rationalistic desire for unity, but also certain simplified and generalised attitudes current in our society, into seeking a single philosophical definition of morality. If, however, we go back again to the data we see that there are fundamentally different moral pictures which different individuals use or which the same individual may use at different times. Why should philosophy be less various, where the differences in what it attempts to analyse are so important?

(Murdoch 1998, 97)

Similarly, Williams characterizes the so-called fact-value distinction, that is, the view (as it may be described from the logical-linguistic perspective) that the evaluative and descriptive functions of language are distinct and can be sharply distinguished, as having been read into the use of moral language by linguistic philosophers. The distinction, in other words, is not something to which language itself gives a clear and unambiguous expression so that philosophers need only to register or record it. Rather, to use Murdoch’s terminology, the distinction constitutes a particular way of delineating and organizing the field of study, that is, moral concepts as expressed in moral language use, which, if it is imposed on the phenomena, may obscure the clear perception of the field, and cover important differences.² As Williams explains:

What has happened is that the theorists have brought the fact-value distinction to language rather than finding it revealed there. What they have found are a lot of those “thicker” or more specific ethical notions I have already referred to, such as treachery and promise and brutality and courage, which seem to express a union of fact and value. The way these notions are applied is determined by what the world is like (for instance, by how someone has behaved), and yet, at the same time, their application usually involves a certain valuation of the situation, of persons or actions. Moreover, they usually (though not necessarily directly) provide reasons for action. Terms of this kind certainly do not lay bare the fact-value distinction. Rather, the theorist who wants to defend the distinction has to interpret the workings of these terms, and he does so by treating them as a conjunction of a factual and an evaluative element, which can in principle be separated from one another. The clearest account, as so often, is given by Hare: a term of this kind involves a descriptive complex to which a prescription has been attached, expressive of the values of the individual or of the society.

(Williams 1985, 129–130)

As Hare explained, thick moral concepts contain two elements: an evaluative one to be understood in terms of the abstract thin moral concepts and a neutral descriptive element which anchors the evaluation in specific situations in reality. Here, it is then assumed that the abstract evaluative notions are contained in the more specific evaluative concepts which do not have any independent evaluative force (Hare 1963, 21ff.). Evidently, however, rather than stating something about the logic of moral language that moral agents would readily recognize as true, Hare is putting forward a controversial thesis that aims to explain reductively the evaluative function of thick concepts in terms of thin ones.³

Third, regarding the narrow focus of linguistic moral philosophy on certain moral concepts as opposed to others, primarily at issue is the occupation of philosophy with the thin or abstract moral concepts as opposed to the thick or specialized ones. (Murdoch’s notion of a specialized moral concept is wider than that of a thick moral concept but closely related.) Both Murdoch and Williams set themselves against this, regarding the emphasis on the abstract concepts as misleading. Accordingly, both conceive moral life as possible without thin concepts, with thick specialized concepts doing the evaluative work (Murdoch 2001, 40–41). But if this is possible, it cannot be taken for granted that morality overall must or can be explicated in terms of an account of one or more of the abstract or thin concepts. It cannot be assumed, in other words, that justice can be done to the phenomena of morality by explaining them in terms of an overarching account of goodness or right, or the notion of ought or duty. Indeed, it is not evident that the uses of the word “good” in a moral sense even constitute a single unified concept. Murdoch comments on the focus of moral philosophy on the concept of goodness and the word “good”:

We were too impressed by words when we assumed that the word “good” covered a single concept which was the centre of morality. We were not impressed enough when we neglected less general moral words such as “true”, “brave”, “free”, “sincere”, which are the bearers of very important ideas.

(Murdoch 1998, 73)⁴

Moreover, she suggests that, if we stop conceiving of moral evaluations as commendations attached to neutral descriptions of facts, as Hare proposes, but

consider rather the way in which a moral outlook is shown in ramifications of more specialised concepts which themselves determine a vision of the world, then the prohibition of defining value in terms of fact loses much of its point.

(Murdoch 1998, 94)

I return to this shortly, when discussing the fourth issue.

Relatedly, Williams writes:

The prescriptivist account claims that the value part of these thick terms is entirely carried by the prescriptive function, which can be analyzed in terms of ought. […] Other theorists say similar things about other general terms: some general, abstract term could do all the work. […] If this kind of analysis proves to be a mistake, and more generally the impulse to reduce ethical language to such abstract terms is misguided, then they are not doing all the work, and this will leave room for an idea I have already suggested, that a society in which ethical life is understood and conducted in such general terms is socially different from one in which it is not, and the differences require social understanding. If that is a fact, the linguistic approach certainly does not help us to recognize it. It encourages us to neglect it even as a possibility.

(Williams 1985, 130–131)

As Williams suggests, a society’s morally orienting itself in terms of thick as opposed to thin concepts might be reflected in differences in the social organization of the two societies. The significance of this is that the possibility suggests a connection between morality and sociality which is absent when morality is conceived in abstract terms such as those of Kant’s moral philosophy. In Kant, the fact that human beings are embodied and social beings plays no role when it comes to the foundations and fundamental concepts of morality as spelt out in what he calls “pure moral philosophy” (Kant 1997). Rather, for him, embodiment and sociality can enter moral philosophy only as an afterthought, so to speak, with human morality constituting a special case of morality. Anything specifically human, however, is irrelevant with regard to the essence and foundations of morality, merely contingent and accidental.

Fourth, regarding the ahistoricity of moral philosophy and of the logical analyses of moral concepts, this can be elucidated with reference to the fact-value distinction, conceived by linguistic philosophers as a timeless logical distinction.⁵ As Murdoch points out, however, rather than seen as an evaluatively neutral logical distinction, the distinction can also be construed as a moral one, that is, as part of a certain moral outlook that urges one not to be complacent about morality, not to take supposed moral facts as given (e.g., that something is “unnatural” or a sin and, therefore, wrong), and which emphasizes the importance of moral reflection (Murdoch 1998, 66, 93). In this capacity, as Murdoch points out, the distinction safeguards a certain notion of freedom shared by protestant Christian and atheist liberal outlooks, and which also is exemplified by Hare’s account of moral principles as freely chosen by the moral agent, rather than necessitated by any facts (Murd...

Table of contents

• Cover
• Half Title
• Series Page
• Title
• Copyright
• Contents
• Abbreviations of Wittgenstein’s Works
• Introduction
• 1 Logical-Linguistic Method in Moral Philosophy: Resolving Problems From Iris Murdoch and Bernard Williams With Wittgenstein
• 2 Moral Concepts, “Natural Facts” and Naturalism: Outline of a Wittgensteinian Moral Philosophy
• 3 Boundless Nature: Virtue Ethics, Wittgenstein and Unrestricted Naturalism
• 4 Between Tradition and Criticism: The “Uncodifiability” of the Normative
• 5 Rule-Following, Moral Realism and Non-Cognitivism Revisited
• 6 Are Moral Judgements Semantically Uniform? A Wittgensteinian Approach to the Cognitivism–Non-Cognitivism Debate
• 7 Truth in Ethics: Williams and Wiggins
• 8 Reasons to Be Good?
• 9 Hitting Moral Bedrock
• 10 Our Fellow Creatures
• 11 Comments on a Contested Comparison: Race and Animals
• 12 The Ethical and the Political in the Dilemma of Winch’s Vere
• Notes on Contributors
• Index