Reflective Equilibrium: Essays in Honour of Robert Heeger (Library of Ethics and Applied Philosophy)

A C.I.P. Catalogue record for this book is available from the Library of Congress.

ISBN 0-7923-5304-8

Published by Kluwer Academic Publishers, P.O. Box 17,3300 AA Dordrecht, The Netherlands. Sold and distributed in North, Central and South America by Kluwer Academic Publishers, 101 Philip Drive, Norwell, MA 02061, U.S.A. In all other countries, sold and distributed by Kluwer Academic Publishers, P.O. Box 322, 3300 AH Dordrecht, The Netherlands.

The publication of this book was made possible by financial support from The Netherlands Organisation for Scientific Research (NWO)

Printed on acid-free paper

All Rights Reserved © 1998 Kluwer Academic Publishers and copyright holders as specified on appropriate pages within. No part of the material protected by this copyright notice may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without written permission from the copyright owner. Printed in the Netherlands

Introduction Wibren van der Burg and Theo van Willigenburg

1

Reflective equilibrium: An attractive approach to moral thinking

Since John Rawls, in A Theory ofJustice, elaborated the method of reflective equilibrium, it has gained much attention and popularity. Rawls introduced the method as a way of developing a general theory of justice, Le. as an argumentative method for developing and justifying theories and principles to evaluate institutions and practices (Rawls 1971, 1985, 1987, 1993). Subsequent work by others has shown that a reflective equilibrium approach may be successfully used for various other purposes as well. It offers a promising coherentist account of justification, but it also gives guidance to philosophical thinking in more specific, practical contexts (e.g., Daniels 1979a, 1988; Heeger 1990a, 1992a and b; Kymlicka 1993; Carens 1995). Reflective equilibrium has especially proved to be a useful method for tackling practical moral problems (Beauchamp and Childress 1994). The basic idea behind reflective equilibrium is that, in developing and justifying moral theories and in seeking answers to practical moral problems, we bring to bear - in some ordered way - all kinds of moral and non-moral beliefs and theories. A reflective equilibrium process pays attention to our moral and non-moral beliefs at various reflective levels (particular intuitions, moral principles, abstract theories), and 'tests' various parts of our belief .system by revising and refining beliefs at all levels. In a process of mutual adjustment, we seek coherence among the widest possible set of beliefs that are arguably relevant in establishing a moral theory, in selecting moral principles or in deciding a specific moral problem. The method of reflective equilibrium is clearly coherentist in that it does not favour a specific type of belief (Lyons 1975; Daniels 1979a, 1980b and c; Raz 1982; Hanen 1983; Sayre-McCord 1985,1996; DePaul 1986, 1993; Brink 1987; Nielsen 1993).1 There are no axioms upon which other beliefs can be based. Critical scrutiny is reached by questioning the tenability and relevance of all types of beliefs, none of which is immune to revision. If a person's judgement regarding what is to be done in a particular case nlns counter to what However,'some authors, e.g. Ebertz (1993), have argued that Rawls's version of reflective equilibrium is less coherentist than he claims it to be; Ebertz calls him a moderate foundationalist. l¥: van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 1-25.

©

1998 Kluwer Academic Publishers. Printed in the Netherlands.

2

Wibren van der Burg and Thea van Willigenburg

is to be done in that case on the basis of a general principle he is committed to, then, according to the method of reflective equilibrium, it is an open question whether the judgement or the principle should be retained. So, in Rawls's famous words: 'Justification is a matter of the mutual support of many considerations, of everything fitting together into one coherent view.' (Rawls 1971, 21 and 579) When we talk about reflective equilibrium, we can refer both to a thinking or an argumentation process and to the product resulting from the process (Wellman 1971; Norman 1998). Product and process can never be separated because they presuppose each other. The process is one of a mutual adjustment of beliefs with the aim of attaining a full reflective equilibrium, a coherent set of beliefs, often in the form of a normative theory (Jamieson 1991). Only by looking at the process can we find out whether this equilibrium deserves the tenn reflective. If we merely look at the product in isolation, we can judge whether it is coherent, but a reflective equilibrium method requires more than simply coherence. This basic idea of a process of mutual support and mutual adjustment between various beliefs is quite natural and intuitively appealing. Such an approach roughly approximates the way in which many of us tend to think when we are dealing with practical moral problems (Dworkin 1978; Jamieson 1991). According to Norman (1998), the method amounts to little more than a codification of common sense. It is not surprising, therefore, that the reflective equilibrium method is especially popular in various fields of applied ethics. Some authors have even argued that there is no real alternative (Grice 1978; Rtiikka 1996-97). Moreover, the idea of going back and forth between more general convictions, such as principles and laws, and more concrete elements, such as considered judgements, judicial decisions or empirical data, is also common in other practices like law and science (Dworkin 1978, 160 ff.; Hanen 1983).

2

Reflective equilibrium: A controversial approach

The reflective equilibrium method seems to be based, then, on a very natural idea, practised in many contexts of reflection and inquiry. But this does not mean that the idea is unproblematic. On the contrary, reflective equilibrium is a quite controversial method. Norman Daniels characterised the method as a 'process of bringing to bear the broadest evidence and critical scrutiny' (Daniels 1996, 2-3; our emphasis). It is precisely these two characteristics of (1) 'broadness' in evidence and (2) critical scrutiny by way of mutual refinement and revision of beliefs which have raised serious criticism and controversy. We will start with the second point: the method for critical scrutiny. Opponents of reflective equilibrium argue that a coherentist method will never

Introduction

3

result in sufficient justification for our moral judgements or theories (Hare 1971, 1996; Lyons 1975; Little 1984; Sencerz 1986; Haslett 1987; Timmons 1987; Stein 1994). And if we cannot claim sufficient justification for our moral judgements, how can we be justified to act on them? Critics particularly point to the risks of relativism and subjectivism. Some suggest that reflective equilibrium brings us nothing more than a neat systematisation of our prejudices (Hare 1971; Brandt 1979; Blackburn 1993). The reply to such criticisms is quite straightforward (Nielsen 1977, 1982a and c, 1988, 1993; Swanton 1991). Perhaps the most annoying thing about the reflective equilibrium method is that it even takes pride in its anti-foundationalist way of justifying judgements, principles and theories. It does not even try to seek a more certain basis for normative statements, in contrast to the vast majority of philosophical (and theological) theories about morality that claim to find a basis in some ultimate principle or metaphysical claim, in divine commands or in the logic or pragmatic of normative language. 2 Neither does reflective equilibrium find a locus of certainty in particular or general moral intuitions revealing moral truth or in the wise man's appreciation of the moral character of a problem situation. The reflective equilibrium method tells us that it is· the process of broad reflection as a whole that is important, rather than one of the epistemic elements in this process. Moreover, reflective equilibriunl does not pretend to produce certainty; it aims at most at some weaker form of warrant. Any equilibrium attained is only a provisional equilibrium; it is always open to revision (Nielsen 1994b, 112). This openness to revision is an important advantage of reflective equilibrium in dealing with criticisms and alternatives. It means that these should be seen as challenges, as a critical input into the reflective equilibrium process. Some critics have argued that it is not clear why having reached such a provisional equilibrium should make us think that we are any closer to knowing what is morally right or true (Hare 1971; Haslett 1987; Gibbard 1995). The natural answer of the reflective equilibrium theorist to such a criticism will be that the method invites us to throw in any consideration that makes one think that a provisional equilibrium does not bring us closer to moral truth, and use this consideration to test and adjust the equilibrium. This will at least lead to an improved equilibrium set of beliefs (Daniels 1979; Nielsen 1988). Reflective equilibrium can incorporate whatever partial truth - in the form of criticisms, arguments, sets of beliefs - that traditional foundationalist moral approaches would be eager to bring to the fore. It will, therefore, often present a kind of middle ground between various competing alternatives (Daniels 1996, 2

As Nielsen phrases it (1994b, 90), it 'travels metaphysically and epistemologically light'. Cf. Norman (1998).

4

Wibren van der Burg and Theo van Willigenburg

339). It may use Kantian arguments or utilitarian principles to criticise and to strengthen its analysis (Rawls 1980; Korsgaard 1996a). It may incorporate the insights of modem casuistry and virtue ethics (Hampshire 1983; Jonsen and Toulmin 1988; Van Willigenburg 1998b). In theory, it can include every line of argumentation and every type of belief; it need not exclude anything. This strength, however, is also a major weakness, and here we find the second major criticism. If the method aims at being all-inclusive (aiming at bringing to bear 'the broadest evidence'), it will soon be unworkable. Perhaps a Hercules .or an archangel could combine all possible elements into one reflective equilibrium process and test and adjust them until some equilibrium is reached. For ordinary humans, this is impossible. It seems, therefore, that reflective equilibrium thinking requires some form of selectivity (Nonnan 1998). Though reflective equilibrium as a general approach is all-inclusive, every attempt to use it as a practical method must be selective. Choices must be made as to the types of beliefs, arguments and methodological criteria to be included, and how. As soon as we make such choices, we exclude other elements and criteria, and are thus vulnerable to criticisms of bias. This need not be a problem that frustrates the method. Different versions of reflective equilibrium - all more or less biased in some sense - may serve different purposes; it depends on the purpose which biases are acceptable and which are not. For exanlple, a methodological instrument like the Rawlsian veil of ignorance excludes knowledge of concrete facts and could thus lead to unrealistic outcomes (Miller 1992; Klosko 1993). As long as we focus on abstract principles for a political system, however, the resulting biases may remain tolerable. In other contexts, for example decision making in medical ethics, it would be absurd to exclude knowledge of concrete facts. This shows that a version of reflective equilibrium using an argumentative device such ~s the veil of ignorance can be acceptable for some specific purposes and completely unacceptable for other purposes. It appears, then, that we should not think of one all-purpose method of reflective equilibrium, but that we have to develop various versions of the method to suit different purposes (cf. Holmgren 1989, 60).

3

Two related perspectives on reflective equilibrium

From the previous section two related perspectives on reflective equilibrium and its history can be derived. Someone who is primarily interested in issues of moral epistemology will take the two major theoretical criticisms mentioned above as a starting point and analyse how various authors have tried to respond

Introduction

5

to them. On the other hand, someone who is primarily interested in a practical use of the method will analyse the different versions of the method and the way in which they have been elaborated in order to serve specific purposes. In this introduction, and in this book at large, both perspectives, the epistemological and the pragmatic, are taken into account. It will become clear that the epistemological and pragmatic questions concerning the wide reflective equilibrium method are intertwined. This has to do with the nature of the method, which claims to show us a systematic manner of conducting our moral inquiries. Reflective equilibrium is meant to structure our moral arguments when we try to assess acts and policies or to develop and justify theories and principles for evaluating institutions and practices. Justification along the coherentist lines as sketched by the reflective equilibrium method should be understood as a process of persuading others (and ourselves), whereby 'others' may include any reasonable challenger. The method aims at providing guidance in concrete moral decision making or in building theories that can be taken as reliable guides. This goal partly explains the specific epistemological characteristics of the method, like the important role it assigns to moral intuitions or considered judgements (cf. Rawls 1985, 228 f.). The best way to convince reasonable challengers is to show that the results of reflective equilibrium match their considered judgements, or to show - if the results do not match the initial judgements of the challengers - that, because of other beliefs they endorse, they would come to find these results convincing. A coherentist approach like wide reflective equilibrium best suits the way in which moral inquiry proceeds in its mundane and day-to-day form. This strongly pragmatic focus raises epistemological doubts and questions which cannot be ignored, however. What reasons do we have to think that a method like this will really deliver on its promise of progress and convergence in moral argument and theory? And if there is convergence of belief, what basis do we have for thinking that this has brought us closer to what we were looking for: a reliable moral assessment of acts, policies and institutions? In the next section, we will take up these epistemological questions. Surprisingly enough, it will appear that - in the end - these questions can only be fully answered by choosing again a more pragmatic perspective. If wide reflective equilibrium is best conceived as a model of persuasive argumentation - justification being the condition of standing up to all reasonable challenges - then the epistemic value of the result of reflective equilibrium thinking will strongly depend on the context and goals of argumentation.

6

4

Wibren van der Burg and Theo van Willigenburg

Reflective equilibrium method and credibility

In 1971, Rawls used the reflective equilibrium method to develop a theory of justice that was to describe and systematise our 'sense of justice'. The theory he aimed to construct was to contain a core set of principles, which together would build a 'conception of justice' that characterises our moral sensibility with regard to questions of justice - a sensibility which is part of our 'moral capacity'. Given this idea of the theory of justice looked for, a number of rival theories were tested by applying their principles and seeing whether the results match our 'considered judgements'. The latter are those 'judgments in which our moral capacities are most likely to be displayed without distortion' because they are given 'under conditions favorable for deliberation and judgment in general' (Rawls 1971,48). There is a certain ambiguity here. On the one hand, Rawls insisted that considered judgements are not immune to criticism and testing, which means that our moral capacity may itself alter under the influence of the theory and its principles. 3 On the other hand, reflective equilibrium was not presented as a full-blown coherence method, but was modelled on scientific method, which describes theory construction as proceeding from observation reports (English 1978; Nielsen 1982a; Gutting 1982).4 Just like in science, theoretical conjectures are tested against the 'facts' as they appear in observation reports. '[T]here is a definite if limited class of facts against which conjectured principles can be checked, namely our considered judgments in reflective equilibrium' (Rawls 1971, 51). This analogy with scientific method is even more explicit in Rawls's early article 'Outline of a Decision Procedure in Ethics' (1951). In this article, he articulated a method for validating and invalidating proposed moral principles and decisions made on the basis of them in concrete cases. The suggested decision procedure consisted of three stages: (1) a procedure for isolating a set of moral 'data', (2) a procedure for discovering and formulating a set of principles which would satisfactorily explicate these data, and (3) a discussion of the reasons one might have for accepting these principles as rationally justifiable. The main part of the 1951 article consisted of an elaboration of criteria on the basis of which a judgement may be called 'considered' and, thereby, function as one of the moral 'data' against which principles are tested. Rawls not only fonnulated various conditions on the nature and emergence of

3

4

This line of argument corresponds with what Dworkin (1978) called a 'constructive' model of reflective equilibrium, which he regarded as the best interpretation of Rawls (1971). This line of argument corresponds with Dworkin's 'natural' model of reflective equilibrium, which aims to describe an objective moral reality (Dworkin 1978).

Introduction

7

considered judgements, but also defined a limited set of 'competent judges' (characterised by a certain requisite degree of intelligence, knowledge, rationality, sympathy and 'moral insight'), whose considered judgements are the only ones which count as reliable data for decision making and theory construction. In A Theory ofJustice (1971), this idea of a limited set of competent judges is abandoned. An appeal can be made to the considered judgements of 'bienpensants' generally, and Rawls even indicates that 'for the purpose of this book, the views of the reader and the author are the only ones that count' (Rawls 1971, 50). The criteria for calling a judgement considered have also changed and have become more general: 'Considered judgments are simply those rendered under conditions favorable to the exercise of the sense of justice, and therefore in circumstances where the common excuses and explanations for making a mistake do not obtain' (Id., 47). The reason for this shift is that in A Theory ofJustice considered convictions are touchstones in a more modest sense than in Rawls's earlier description of his method. Rawls now more openly embraces a coherentist epistemology, according to which the data for which the theory has to account themselves may be adjusted, rectified and even rejected because of some convincing principle (Delaney 1977). There is no one-way accommodation but 'mutual adjustment' , a concept which Rawls adapted from Goodman's philosophical account of inductive inference (Goodman 1955 - see Rawls 1971,20 n. 7). Rawls (1971, 579 n. 33) also refers to the conception of justification found in the work of Quine (1960), where Quine stresses the role of consensus in the definition of 'observation' . Even so, Rawls still does not seem to endorse a radical coherentism. He compares the reflective equilibrium method with Chomsky's attempt to formulate principles which describe 'the sense of grammaticalness that we have for the sentences of our native language' (Rawls 1971, 20). Analogously, moral philosophy aims at the formulation of principles which account fo~ a person's sense ofjustice (Goodman 1955,65 f.; Daniels 1980a; Singer 1986). This 'sense of justice' will be subject to scrutiny because the judgements based on it will be confronted with principles building a conception of justice, but it is hardly imaginable that this scrutiny will lead to radical alterations if the analogy with grammar holds. Rawls's seemingly half-hearted coherentism has led to a long-standing debate about the status and nature of considered moral judgements (Carr 1975; Nielsen 1977 and 1982c; Grice 1978; Noble 1979; Sencerz 1986; Barry 1989a, 271 f.; Nelson 1990; Audi 1993; Ebertz 1993; Kekes 1993). Rawls seems to award a special epistemic status to considered moral judgements in order to give the

8

Wibren van der Burg and Theo van Willigenburg

process of reflective equilibrium some reliable starting points. This may be an effective way of answering those critics who think that pure coherence arguments will not help us in making progress in moral thinking as they simply move us in circles around our current considered opinions and the principles they supposedly manifest. But what could give considered moral judgements their epistemic credentials? There is a decisive difference here compared with scientific observation reports. We can explain why observation reports are generally reliable and these (causal) explanations do not completely rely on the credibility of the body of theoretical beliefs which provide their explanation (Dancy 1984; see also Dancy 1985 and Haslett 1987). There is, however, no evidence for the reliability of considered judgements independent of the principles and other theoretical elements they are supposed to support and manifest (Sencerz 1986). In Richard Brandt's terms: the initial credence level of considered moral judgements, i.e. the level to which we are committed to the beliefs involved in these judgements, does not tell us anything about their credibility (Brandt 1959, 1979, 1989, 1996). The fact that we trust our well-considered judgements does not add any evidence to their trustworthiness. Nothing prevents our considered judgements from expressing just the arbitrary commitments and sentiments of a prejudiced viewpoint (Singer 1974; Blackburn 1993). Increasing the credence level of our considered judgements by making them cohere with principles that provide generalisation and systematisation does not move us one inch closer to credibility. And there is nothing in the further process of mutually adjusting considered judgements and principles that could give us reason for thinking that we had moved to a more credible viewpoint instead of having refined and confirmed our initial prejudiced outlook. In answer to this no-credibility objection, Norman Daniels developed the idea - already manifest in Rawls's work - of a wide reflective equilibrium. The idea is to seek coherence among the widest set of moral and non-moral beliefs by revising and refining them at all levels (Daniels 1979a; for a critical view, see Holmgren 1989). Reflection is not limited to principles and considered judgements. Reflection also has to include relevant background theories like a theory of the person or a general social theory, specific thinking formats like the original position, and specific concepts like the concept of a social contract. These background theories, thinking fonnats and concepts may provide evidence for the credibility of a set of moral principles and the moral conception they embody. Explicitly involving background theories in the method of'reflective equilibrium provides an answer to the often heard criticisnl of Rawls's theory ofjustice that it implicitly presupposes some theory of the person (e.g. a theory which emphasises the separateness of free and equal persons), some theory about

Introduction

9

the role of justice in society (including the ideal of a well-ordered society) and various social theories necessary to back up Rawls's assumption about the feasibility and stability of a system based on procedural fairness. (See various contributions in Daniels 1975; Hampton 1989; Raz 1990; see also Sandel 1982; Rawls 1985 and 1987; Barry 1989a, 320f.; Nielsen 1994b). Wide reflective equilibrium explicitly includes all these background theories in the process of mutual reflective support and adjustment and, thus, expands the kinds of considerations that count as evidence for or against our moral views at all levels of generality. By paying attention to all kinds of moral and non-moral considerations that bear on a moral issue, it presents a model of moral thinking which, in its idealised form, may provide an answer to the criticism that coherence of beliefs does not lead to credibility. The moral thinker using the wide reflective equilibrium method has a simple question to put to the sceptic who wonders whether the established wide equilibrium of considerations does not just build a coherent set of prejudices: What consideration makes you think that this is just an amalgam of cooked-up prejudices? What consideration gives you reason to be suspicious? And when the sceptic comes up with this consideration, the moral thinker will invite her to throw these critical considerations back into the ring and seek a new wide reflective equilibrium. It seems plausible to suppose that enriching wide reflective equilibrium in this way contributes to the credibility of the judgements contained in it (Daniels 1979a). What could the sceptic say to this? There are at least two types of criticism that the sceptic could bring which are worth mentioning here because they have inspired the further development of the wide reflective equilibrium approach. The first type of criticism focuses on the conservative tendency inherent in any method of thinking that starts with our actual considered moral judgements. 5 The more epistemic value is bestowed on the considerations which are taken as starting points of reflective equilibrium thinking, the greater is the danger of including biases which can never be traced nor erased. Widening the circle of considerations to be included in reflective equilibrium does not bring us closer to moral truth if the point where we actually start drawing the circle is simply given with our initial judgements, even if these are fonned in optimal epistemic circumstances. In answer to this line of criticism, Michael DePaul (1993) proposed a radical conception of reflective equilibrium, which allows initial beliefs and degrees of beliefs to be altered in ways that go beyond what is required to make

5

Singer (1974); Brandt (1979); Aronovitch (1996); cf. also Cladis (1994); Walker (1995); Raikldi (1996-97, 185). Nielsen 1991 argues that reflective equilibrium is historicist and contextualist but not relativist.

10

Wibren van der Burg and Theo van Willigenburg

a person's initial belief system coherent. In wide reflective equilibrium as conceived by Norman Daniels, the raw materials (initial judgements) play an important role in determining the nature of the final coherence set because refinement and revision are only called for in order to resolve conflicts where these initial beliefs are involved or to extend the system of beliefs where the extension is always extension beyond the initial beliefs. Instead, a method of radical reflective equilibrium may result in a coherent set of beliefs which is inconsistent with any of the initial beliefs with which the thinking process started. The radical method may lead, so to speak, to radical conversions in one's thinking. Initial beliefs may start the process of reflection, but they will in no sense determine the direction this reflection may follow. Very different starting points may lead to the same equilibrium point of thinking, but the same starting point may also bring different people to very different outlooks on an issue. There is, therefore, no need to be worried about the credibility of the initial moral judgements. DePaul has an interesting suggestion about how to promote such a radical revision process. He makes clear that the highly intellectualist focus on moral judgement, construction of arguments and development of theories of standard reflective equilibrium approaches is one-sided. It leaves out important questions about the character and development of the inquirer's abilities and faculties for making judgements, constructing arguments and building theories. Typically, a person can acquire the ability to make relevant discriminations in judging and arguing only after a considerable amount of experience and training. Philosophical deliberation along the lines of the reflective equilibrium method should not only be thought of as affecting our beliefs and arguments, but should also be expected to cause 'changes in a person's judgmental faculties, so that these faculties no longer function in the same way, yielding the same beliefs and theories, as they previously did' (DePaul 1993, 211). This means that reflective equilibrium thinking may demand the expansion of one's range of experiences - be they 'real-life' or vicarious experiences through, e.g., literature and theatre - in order to test, refine and possibly revise one's judgemental faculties. DePaul thus draws our attention to an important point which, in the Rawlsian original position, is deliberately ignored as much as possible, but which should not be neglected: the character and experience of the person who goes through the reflective equilibrium process. Moreover, he gives an interesting suggestion about how to counterbalance possible conservative tendencies. The question remains, however, whether this will indeed bring us closer to moral beliefs which are true, or at least warranted..

Introduction

11

The second type of criticism of the wide reflective equilibrium approach points to some strong reasons to limit our expectations with regard to the results attained by using a method of philosophical thinking like wide reflective equilibrium (Baier 1985; Williams 1985; Clarke 1987; Raz 1990; Gibbard 1995). Even the most reasonable and most experienced people are likely to develop quite varied answers to the complex moral issues that humans seek answers to. This pluralism is a pervasive and deep fact; in his recent work, Rawls admits that, in A Theory of Justice, he was insufficiently sensitive to it (Rawls 1993, 54 ff.). Because of this pluralism, it is unrealistic to expect philosophical reflection - e.g. along the lines of reflective equilibrium - to move everyone who can think clearly and rationally, and who bases his judgement on a rich and uncorrupted experience, to convergence on one moral outlook vis-a.-vis certain matters. People may, for instance, have very different views about the priority of a value like justice. For a religious person, a relationship with God may be much more important than distributive justice. Others may think that being part of a caring community is much more important than a just society. Why would these people have to sacrifice their important moral concerns in order to abide by the principles of justice? Is the wide reflective equiliblium theorist who thinks that, in the end, the views of people will converge on some overall coherent set of moral considerations, not blatantly naive? In answer to this type of criticism, John Rawls tries to show that apolitical conception of justice may be reflectively endorsed even if there is no general convergence on a wide reflective equilibrium with comprehensive commitments that support this political conception (Rawls 1985, 1989, 1993). Philosophical, moral and religious controversies should be left aside as much as possible in 'a method of avoidance' (Rawls 1985, 283; Cladis 1994).6 Daniels calls this a 'political reflective equilibrium' in which, next to philosophical reflection, a major role is assigned to shared institutions. These institutions make, in the course of a shared history, groups with different comprehensive views accommodate themselves to justice as fairness (Daniels 1996, 160). These shared institutions have a distinctive political character. They make it possible to create a public or political realm of reasoning about matters of justice. In this realm, people share the key ideas of justice as fairness, like the idea that citizens are free and equal - even if they support these ideas for very different reasons. Different comprehensive views may lead to different justifications for the principles, reasons and ways of argumentation that are internal to the political conception of justice; yet, there is an overlap in the outcomes. They will even

6

Dworkin's constructivist interpretation of reflective equilibrium and the justification he gives for this interpretation in terms of a public standpoint already suggested some of the ideas elaborated in Rawls (1993). Cf. Dworkin (1978,163).

12

Wibren van der Burg and Theo van Willigenburg

imply different views on where to draw the line between the political and the non-political. Despite these differences, Rawls assumes that there is enough 'overlapping consensus' to create a commonly shared 'political reflective equilibrium' (Rawls 1987, 1989, 1993; see also Nielsen 1991 and 1994b; Cohen 1994; Greenawalt 1994; Barry 1995). 7

5

The pragmatics of reflective equilibrium

The various elaborations on reflective equilibrium - whether they constitute wide, radical or political conceptions - have certainly led to improvements. Nevertheless, they do not provide knock-down arguments to prove that it is the best method of moral reasoning, let alone that it leads to moral truth (if there is such a thing as moral truth). We can never prove a priori that it is so; we can only demonstrate that it does well in comparison with other methods. This brings us to the pragmatics of reflective equilibrium because we cannot compare argumentation methods in general, only in connection with the purposes they are meant to serve. A reflective equilibrium process may deliberately include or exclude certain types of considerations, depending on the issues at stake and the purpose aimed at. The way in which methodological requirements, such as the ideal of impartiality, are interpreted differs similarly. So there may be different types of reflective equilibrium. This pragmatic selectivity in the light of specific purposes can be an adequate response to the danger of all-inclusiveness, which could make a reflective equilibrium method unworkable. Moreover, it is also a good method to counter some items of epistemological criticism mentioned earlier. Rather than attempting to construct one general method, one can carefully analyse which biases are likely to occur under specific circl!mstances and design methodological devices or include specific elements in the process to counter these biases most effectively. We can structure the possible varieties of reflective equilibrium methods by asking three questions: 1. What are the purposes the method is meant to serve? 2. What are the considerations used in the process? 3. What are the methodological requirements to be observed in the process? 7

Klosko (1993,353), however, argues that the idea of an overlapping consensus on liberal ideas fails on empirical grounds, at least for concrete issues: 'The unavoidable conclusion of decades of empirical studies, then, is that significant majorities of Americans, like citizens of other liberal societies, do not uphold basic liberties in the strong sense Rawls requires.' For a similar criticism see Walker (1995, 102).

Introduction

13

The variation in purposes of reflective equilibrium and thus in the method itself can, to a certain extent, also be found in the work of John Rawls. In 1951, he regarded the method as a decision procedure and its product as a set of moral principles. The methodological requirements focus on the criteria for considered judgements and on the qualification of 'competent judges'. The considerations used in the process were restricted to considered judgements ('moral data') and moral principles. In 1971 the method was used for theory construction: its aim was, to be more specific, to arrive at a nonnative theory (and especially the basic principles) for the basic structure of an almost ideal (nearly just) society. Methodological devices such as the original position and the veil of ignorance were introduced and controversial religious views and knowledge of concrete facts were excluded from the argumentation process (cf. Daniels 1975). In 1993 there was another change of purpose because the aim no longer was a theory for an ideal society but a nonnative political theory for a modem, deeply pluralist society. In the method, the idea of an overlapping consensus was given a central role, but controversial religious views were still excluded from the reflective equilibrium process (cf. Barry 1995). Despite the variation, the primary purpose of the Rawlsian versions of reflective equilibrium is theory construction and the construction of moral or political principles; they tend to focus on the more abstract theoretical levels (Barry 1989a). Even the 1951 version only results in moral principles which still have to be interpreted and applied to concrete problems, whereas the other versions need various steps of transfonnation to be useful for practical problems. Therefore, they are less adequate for the purpose of concrete problem solving; we should look elsewhere for versions that may be useful in applied ethics. To get a new start, we should return to the minimal idea of reflective equilibrium (as in the 1951 article), which has been the basis for much work in applied ethics. Two elements are at the core of every version of reflective equilibrium: considered moral judgements and general moral principles or general beliefs. A method which consists of no more than these two elements (and some methodological requirements) is usally called a narrow reflective equilibrium (Daniels 1979a). Some authors have, for their purposes, been satisfied with this minimal version. It is, e.g., used by Glover (1977, 26 f.), who speaks about an interplay between 'responses' and 'general beliefs'. In the early versions of their textbook on biomedical ethics, Beauchamp and Childress (e.g., 1983, 12) similarly described their procedure as a dialectical reasoning between principles and judgements about cases. s 8

In the fourth edition of their book (1994, 20-37), however, they explicitly subscribe to a very wide version of reflective equilibrium.

14

Wibren van der Burg and Theo van Willigenburg

Most authors in applied ethics have nevertheless gone beyond the narrow reflective equilibrium and added or refined some elements. If one takes considered judgements as having rather specific ideas and particular situations or cases as their object, and principles as the more general normative propositions, one could introduce 'morally relevant facts' as a separate element in reflective equilibrium. This will be particularly useful if one uses reflective equilibrium as a method for analysing concrete moral problems. For, in order to judge a situation, we must distinguish the morally salient features of that situation, i.e. we must point out some natural properties of the case as a reason to evaluate the situation in one way or another. Whether one thinks it possible or even necessary to introduce morally relevant facts as distinct elements into the reflective equilibrium process, seems to depend partly on whether one believes that moral salience will 'simply' appear if one is sensitive to the situation (morally salient facts 'leaping' to the eye) or whether one believes that salience is brought in by more general moral considerations, that is, considerations that can be formulated independent of the concrete case. In the first line of thought, Robert Heeger and Theo van Willigenburg have developed a 'network model' for dealing with concrete moral problems (Heeger 1990a; Van Willigenburg and Heeger 1991). This network model consists of three basic elements: moral intuitions, moral principles and morally relevant facts. They also suggest various methodological requirements regarding these elements (such as the principles' universality and simplicity) and regarding the ethical practitioner (such as an open mind and moral sensibility). This is a very simple model, basically the narrow reflective equilibrium to which morally relevant facts have been added. It is important to notice, however, that the facts are not merely regarded as passive objects to which moral judgements or principles have to be applied, but also as a factor in the equilibrium process itself. The (Rossian) idea is that, being confronted with a morally problematic situation and having some idea of relevant prima-facie duties, people naturally fonn intuitions about the overall moral story to be told about the situation, which includes some initial intuitive view on what is morally required. There is thus a genuine interplay between facts, principles and intuitions. This network model has subsequently been elaborated and refined in a number of doctoral theses and other publications. It can be argued that methods such as the network model (or narrow reflective equilibrium) are most productive and least vulnerable to criticism when applied to fields in which there is already a relatively stable (though still partly incomplete) moral consensus both at the level of moral principles and at the level of concrete issues (Norman 1998). In such contexts it may be very useful for problem solving. (For an example of such an application of the network model, see Rutgers 1993.) However, in fields in which there is less consensus, in which moral opinions are in rapid

Introduction

15

change and moral philosophers are still searching for defensible formulations of basic principles and values, such a model is less adequate. When we need to develop moral views about radically new developments such as modern biotechnology, we need to broaden the scope and explicitly refer to background theories. It may even be unavoidable to include controversial background views such as religious views in order to deal with these issues (cf. Heeger 1997). The network model should perhaps be further broadened if we want to deal with practical issues which have broader theoretical implications or if we want to develop 'modest theories' or guidelines. One change may be necessary if we want to construct, e.g., a practically useful democratic theory for the real society we live in; it may then be useful to add the basic values or ideals that are implicit in the public culture of our society. We thus get an 'extended network model' in which five elements are to be included: moral intuitions, moral principles, morally relevant facts, background theories and moral ideals (Van der Burg 1991; Brom 1997). Another change may be needed if we use reflective equilibrium to develop a set of guidelines for concrete decision making. Hans van Delden, e.g., uses it to construct a set of guidelines on do-not-resuscitate decisions (Van Delden 1993). He makes an interesting change by not starting from the intuitions and principles as formulated by the philosopher but from the intuitions and principles of the practitioners; he incorporates the results of empirical research on moral opinions into the model. 9 Byway of this strategy, he increases the chances that his recommendations will be recognised by doctors and nurses as a critical reconstruction of their own views and, thus, that they will be accepted. Furthermore, as his purpose is effectiveness of his recommendations in practice, he suggests that these guidelines be supplemented with a virtue theory. Again, the purpose (in this case the development of a practically effective set of guidelines) determines both the elements and the method of reflective equilibrium and, in a sense, even requires going beyond the equilibrium method as such. Thus, different versions of the network model may be used for different purposes in the context of applied ethics. Yet other varieties of reflective equilibrium could be useful if the focus is more on intersubjective discussions and public debates. Usually, reflective equilibrium is merely a method for

9

The idea that empirical research on moral views is relevant to reflective equilibrium methods has been suggested by various authors (Ebertz 1993, 208 ff.; Miller 1994), yet attempts to integrate them are still scarce. Another example is Klosko (1993).

16

Wibren van der Burg and Thea van Willigenburg

intrapersonal argument. tO It may, nevertheless, be used as a model for interpersonal argumentation, as a real-life framework for structuring discussions. For example, we could structure a discussion in ethics committees along the lines of a reflective equilibrium process. It may also be regarded as a framework for structuring political debates in parliament or even in society at large. Each of these purposes will bring the need to modify the model, either by adding or ignoring certain elements in the process, or by constructing new methodological requirements. For instance, if we want to use reflective equilibrium as a model for committee discussion, we must see how the idea of impartiality can be guaranteed in a committee, both by the process of selecting members and by structuring the discussion and the attitude expected from the mernbers. Reflective equilibrium, thus, need not only be a method for philosophers - it may even be more attractive and effective for the general public. It may also be a method for moral education and, in connection with this, a model for moral development. The student can be confronted with concrete moral problems and be asked to solve them, starting with her moral intuitions, reflecting on them, trying to find principles that do justice to them, and so on. (In our experience, this basic structure proves to be very useful in teaching ethics to students in the various biomedical departments.) There are even other, quite different purposes for which a version of the reflective equilibrium method may be helpful. We need not only think of reflective equilibrium as a normative model; it may also be useful as a descriptive model. This may be so at the level of the moral development of one person, but also at the level of society or societal subgroups. If it is a good method in nonnative argument, the hypothesis is warranted that at least sometimes moral developments in society follow the lines, of a reflective equilibrium process (cf. Brom 1997). We may conclude that there is a broad variety of purposes for which reflective equilibrium methods may be used and, corresponding with this variety, there are many different versions. A systematic comparison between these versions could be a further step in the development of reflective equilibrium. For example, detailed enlpirical studies of the structure of public debates may be helpful to test and refine the method in normative contexts. Seeing how a method works in the context of theory construction and which biases result from

10

Of course, Rawls (1971) presents his theory as one in which various persons convene to decide on principles of justice, but many critics have convincingly argued that, in fact, there is only one - imagined - person doing the reasoning. Aronovitch (1996,401) argues that is the 'typical person' rather than a special person.

Introduction

17

specific methodological requirements may make us sensitive for similar biases in models developed for problem solving. A further refinement of reflective equilibrium methods may also be inspired by studies in other disciplines such as science and grammar. One practice where reasoning can be structured adequately along the lines of reflective equilibrium is law (Dworkin 1978; Hanen 1983; Van der Burg 1991 and 1993; for a critical view see Sunstein 1995). Haslett (1987, 309) has even argued that this method is more promising in legal than in moral reasoning because in case law and statutes, law has at least some elements with initial credibility. Specific characteristics of the versions in law may also be of heuristic value for a refinement of the method in ethics. For example, the virtue of integrity, which Ronald Dworkin has suggested is a central ideal in law and which may be a fundamental reason for striving towards coherence, could also be central to reflective equilibrium in (political) ethics. A comparison between how integrity functions in law, in morality and in politics may therefore result in a better understanding of the role of coherence in political and moral argument (Dworkin 1986; Van der Burg 1993).11

6

The contributions to this volume

The contributions in Part I, The Elements of Reflective Equilibrium, illuminate and assess the nature of the various types of considerations or belief'elements' which playa role in wide reflective equilibrium thinking. Marcel Verweij discusses the nature of moral principles by asking how principles can retain their character as more or less stringent and durable moral requirements - the cornerstones of one's system of moral beliefs - and still be open to revision or even rejection in a reflective equilibrium thinking process. Verweij argues that if reflective equilibrium is employed as a method for theory fonnation, one may distinguish (i) initial principles, (ii) principles during the process of reflection and adjustment and (iii) principles that are the product of this process. Initial principles may have a specific appeal and, therefore, authority that cannot be reduced to the inquirer's commitment to his concrete judgements. (Principles are more than summaries thereof.) The authoritative appeal of initial principles may, in a sense, be retained during the reflective equilibrium process even if they are subject to revision and adjustment. The situation is different when reflective equilibrium thinking is employed to assess concrete moral problems. Principles seem to function, then, more as a kind of containers of a wide range 11

Dworkin (1978, 163 and 1986); cf. Shiner (1992,229). Compare also a similar point in Glover (1977,28), on scientific integrity as the basis for coherence.

18

Wibren van der Burg and Theo van Willigenburg

of possibly relevant moral considerations. Still, some of these considerations will build the authoritative core of the principle (having, therefore, clear binding force), while others may be more in the penumbra of the meaning of the principIe, requiring specification and thereby interpretation in the light of the problem at hand. But here too principles may retain some of their authoritative appeal and kernel even in a process of change and adjustment. Theo van Willigenburg asks what makes certain facts of a case morally relevant, Le. bearing moral weight in judging the case. What makes certain facts count as reasons for or against a particular moral judgement? He argues that case features may appear to have moral relevance prior to and independent of having generalisable considerations like moral principles or rules pointing out and explaining that relevance. The reason is that the relevance of certain facts may dramatically vary according to circumstances, because of the strong context dependence of what counts as morally relevant. Facts, therefore, can have moral salience sui generis in a certain case, which makes it impossible to fully 'catch' moral relevance by invoking general moral considerations. The recognition of sui generis relevant facts and the understanding of the way they play their reasongiving role in reflective equilibrium thinking demands a kind of non-inferential, intuitionist rationality (based on a particularist epistemology). Van Willigenburg claims that intuitionism may very well be combined with inferential reasoning involving the various generalisable considerations which play their role in reflective equilibrium thinking. Reflective equilibrium is often understood as a method for testing our ordinary considered moraljudgements by bringing them into critical interaction with more general moral and non-moral considerations. The main question to be posed against the method is what makes us think that adjusting and pruning our judgements following the method of reflective equilibrium will free us from moral bias and bigotry? Anton Vedder argues that the specific coherentist method of moral justification as exemplified by the reflective equilibrium method makes it suited - par excellence - to confer justificatory force on our considered moral judgements. Wide reflective equilibrium functions as an instrument for unravelling the web of moral and non-moral meaning embedded in the shared experience and knowledge of communities and traditions. By exploring and formulating the connections between various moral and non-moral notions, reflective equilibrium thinking enhances our understanding of the way our moral judgements are embedded in the contexts of meanings, which together build a life fonn of which we are part. This does not exclude controversy and difference within a tradition or community, but it makes clear how understanding and justification always are fundamentally context bound - a deep fact acknowledged by the method of reflective equilibrium thinking.

Introduction

19

The contributions by Van Willigenburg and Vedder both stress the importance of a particular life form as the source of understanding moral notions and as the background of the way 'we' intuitively grasp the moral meaning of certain constellations of facts. Ton van den Beld analyses the role views of life - fundamental to a person's outlook and biography - should have in wide reflective equilibrium thinking. Are views of life to be included in the set of background theories and considerations which Daniels introduced into wide reflective equilibrium? Van den Beld shows that the metaphysical beliefs which a religious person may endorse, may have a direct influence on the way this person comes to weigh and balance various moral considerations in trying to reach an equilibrium of convictions with regard to a problematic issue. Having fundamental views of life' at the background' of reflective equilibrium thinking does not mean, however, that the reflective equilibrium method would be structured in a more foundationalist way. Religious beliefs and other background theories and views may be subject to scrutiny in the course of wide reflective equilibrium thinking, just as other belief 'elements' like principles and considered moral judgements. Van den Beld argues that reflective equilibrium thinking aimed at reaching interpersonal moral agreement and convergence in moral enquiry - for instance in the political sphere - need not at the outset exclude 'controversial' fundamental views of life from the agenda. Rawls's idea of an 'overlapping consensus' of views leaves enough room for having religious beliefs play their role as part of the background beliefs· in wide reflective equilibrium. Important to any life form are values that are implicit and latent in the public and moral culture. Values which are thus grounded in reality but which are oriented towards the future are the ideals or normative ambitions of a group or person. Such ideals usually cannot be fully realised and they partly transcend contingent, historical formulations and implementations in terms of rules and principles. In his contribution, Wibren van derBurg clarifies the role and character of these ideals and argues that reflective equilibrium thinking should account for their normative force and their specific role as continuous sources of new critical input into the reflective equilibrium process. As ideals are, on the one hand, immanent in our culture or experience and, on the other hand, transcend this culture or experience, they may form a point of critical reflection because their endorsement provides access to an indefinite surplus of normative meaning and sensitiveness to unforeseen possibilities. Having ideals as reflective 'elements' in our thinking process may, thus, countervail the conservative tendency of the (narrow) reflective equilibrium method. Part lIon The Methodology of Reflective Equilibrium takes up some intricate epistemological problems involved in reflective equilibrium as a method of moral

20

Wibren van der Burg and Theo van Willigenburg

enquiry. According to the method, it is rational to endorse a moral belief if this belief is sufficiently supported in a network of connected and coherent other beliefs. Usually, rationality is considered to be a person-related, epistemological criterion - what should be (ir)rational for you to know, given your set of beliefs, need not be identical to what should be (ir)rational for me to know. It is usually contended, therefore, that next to rationality we need another epistemological criterion, warrant, which introduces a general standard or procedure which may authorise the step from the evidential grounds to a claim, i.e. a procedure which explains why one may trust the conclusion reached. Warrant is truth conductive in the sense that it turns mere belief into knowledge. Eberhard Herrmann argues, however, that warrant in the truth-conductive sense is, just as rationality, an inevitably person-related criterion. Still - he says - it is possible to have a difference between rationality and warrant if one conceives - in a pragmatist way - truth as warranted assertability. In this view, knowing is not mentally mimicking what is 'out there', but is a constructive conceptual activity that guides our pragmatic interaction with the world. According to Herrmann, Putnam's internal pragmatic realism - a realism concerning what humans can conceive of as ideal verification - gi ves us enough epistemological foothold to avoid the person relativism inherent in the concept of warrant. To warrant a statement is to consider our 'epistemic circumstances' as ideal as, within our human situation, we could hope to have them. This explains why, according to Hernnann, it is necessary to be in an optimal 'condition' not only cognitively but also emotionally, and have a broad range of so-called formative experiences which help us develop a sensitivity for good and evil, right and wrong. Therefore, Hemnann says, moral beliefs may become warranted through moral experience developed not only by philosophical reasoning but also by way of art and the way different views of life express themselves. Marian Verkerk further develops Herrmann's point about the importance of formative experience. She argues, however, that we must not only pay attention to the specific sensitivities (our moral 'faculty') which we need in morally judging situations and balancing considerations; we also have to pay attention to the social and political interpretative framework of our thinking. Feminist moral epistemology shows that whether someone has rational and warranted beliefs is not only a question of having adequate reasoning capacities, but also, and more importantly, a matter of her having a sense of her 'self' that would support a full sense of flourishing. Not only personal, but also social and political transformations and sensitivisations are necessary to allow ideologically oppressed people to acquire new interests, desires, beliefs about the world and understanding of their own position. Literature and other forms of art may be helpful in developing one's capacities of moral understanding, but this should not distract our attention from the political aspects of defining and acquiring

Introduction

21

an adequate sense of personal integrity, which is a basic condition for sound moral judgement. Bo Petersson provides a critical analysis of the way coherence is employed as a justificatory criterion in wide reflective equilibrium thinking. There may be more ways to exclude incoherencies and to establish coherence of considerations, so there may be more than one equilibrium to be attained. Petersson shows that ideas of 'maximal coherence' and 'best fit' of beliefs are yet too unclear to provide sufficient guidance when it comes to a more systematic and detailed comparison between coherent sets of beliefs. It is, for instance, not clear what should count as decisive in judging 'maximal coherence' . Is strong mutual support through entailment relations of a small set of beliefs more important than supportive relations of a weaker kind (inductive or probabilistic or explanatory 'entailment') in a wide set of beliefs? Coherentist justification alone is not enough to attain warranted assertability. Petersson suggests that we further develop the main idea behind wide reflective equilibrium: our beliefs about the world, man, society and morality must fit together in a continuous and connected whole. This implies holism rather than coherentism, the difference being that (i) - if necessary - holism intends to endorse very different types of beliefs, and (ii) holism does not exclude the possibility that some kinds of belief have more weight than others without assuming foundationalisffi. Not only is it possible that there is more than one reflective equilibrium point, i.e. more than one way to reach coherence, in moral thinking very different persons may be involved with very different intuitions. Which considered judgements need adjustment and correction, and which intuition is to be upheld against conflicting principles or theories? 'Your Intuition or Mine?' is the provoking title of Inez de Beaufort's contribution. She is deeply suspicuous of the role of so-called 'yuk feelings' - strong feelings of 'this is wrong' - which seem to inspire many of the debates in medical ethics. Why would gut feelings have any authority in moral thinking? Why take intuitions seriously? Using the (up-to-now fictitious) example of male pregnancy, De Beaufort shows what the reasons could be for following and trying to understand our gut feelings about controversial issues (Are these feelings not grounded in the firmly held beliefs within a moral tradition?), and what reason there could be for mistrusting people's gut feelings (unjustified conservatism). The arguments De Beaufort brings forward for and against trusting our intuitions put some of the main features of reflective equilibrium thinking under serious criticism. The contributions in Part ill, The Scope ofReflective Equilibrium: Fundamental Views of Life and Religion, discuss the relations of moral commitments with religious beliefs and - correspondingly - the place of views of life in reflective equilibrium thinking. Vincent Brummer argues that we could express our moral

22

Wibren van der Burg and Theo van Willigenburg

commitments either in terms of moral principles or in terms of models and metaphors derived from religions or views of life. According to the first approach, religious beliefs provide believers with reasons for accepting certain moral precepts (principles) and with a powerful motivation for living according to those precepts which can, however, be expressed independently of one's worldview and also be endorsed by non-believers or adherents of completely different worldviews. According to the second approach, there is an internal relation between one's moral commitments and the (metaphysical) beliefs, which are part of one's (religious) worldview. It would then be logically incoherent to commit oneself to the way of life expressed in a religious model, Le. a specific morality, while refusing to believe that what is expressed in this model is somehow factually true. Views of life not merely have a motivational relation to our moral commitments but also determine the content of these commitments and the nature of the actions to which we are committed. This would, for instance, mean that adherents of different views of life or religions are only superficially committed to the same moral actions. Moral principles may prescribe a certain form of observable behaviour, but - as we come to intentions and internal understanding - what is done may be different according to different religions models. This does not make the talk of moral principles inert, however, especially in situations where people have to co-operate in spite of their different worldviews. Though religious beliefs will constitute an essential component within the 'backgrounds theories' used to justify moral commitments, reflective equilibrium thinking may take a more or less pragmatic direction in which one aims at reaching moral consensus in terms of shared moral principles and not in terms of the ideological models and metaphors through which people understand the meaning of their lives and experiences. Carl Reinhold Brakenhielm further elaborates the relation between what he calls 'cosmological claims', Le. general beliefs about the ultimate framework of human existence, and the moral and non-moral considerations which play a role in wide reflective equilibrium. He goes into a point that is also touched upon by Ton van den Beld and Vincent Brummer: cosmological claims cannot be immune to adjustment and modification in the course of the reflective equilibrium process. Though cosmological claims will be clearly underdetermined by other beliefs endorsed in reflective equilibrium (if only for their generality), they are certainly not immune to critical scrutiny. As far as cosmological views claim to make sense of the other beliefs entertained in the equilibrium, their moral and factual implications may be criticised and the cosmological model they stem from can be the object of rational inquiry. The problem remains, however, that even the 'best methods we have for an assessment of cosmological claims do not give us determinate answers about their truth, so that an element of relativism and perhaps unfounded commitment remains.

Introduction

23

This insufficiency of rational argument is an element of any reflective equilibrium, then. It also presents us with the problem of how to give a rational account of major changes in worldviews and major values if these views are seriously underdetermined by argument. In his contribution, Anders Jeffner distinguishes a general social constructivist, a materialistic constructivist, an idealist and an objectivist explanation of changing values and worldviews. He argues that the general social constructivist and the objectivist explanation each have a part of the correct story, which can only be understood if one sees that values are different from valuations. Valuations change, that is for sure, but of values it would be better to say that they are differently and perhaps better understood given changes in valuations. Jeffner argues that a coherentist model of moral thinking and of changes in moral beliefs is very well compatible with a kind of metaethical objectivism.

Part IV, The Purposes ofReflective Equilibrium: Modelling Public Debate and Dialogue, discusses various uses of the reflective equilibrium method: as a tool for modelling public debate, an instrument for developing public morality or a way of structuring a (Socratic) dialogue on core values in business. Jan Vorstenbosch discusses the possibility of using the reflective equilibrium method for analysing and organising public debates, Le. the process of exchanging usually via the media - opinions, arguments and general ideas about a specific issue whereby this process is accessible, at least in theory, to every citizen. One of the problems of using the method in this way is, of course, that reflective equilibrium is usually thought of as a model of moral thinking by an individual, whereas public debates are essentially collective. Given a plurality of opinions, whose considered moral judgements are such that they have to be taken as the starting point of the reflective equilibrium balancing process? Moreover, in public debate not only discussion on merits but also special interests and partiality are the driving forces. What about the conditions for impartiality which are usually considered to be constitutive of moral thinking? Vorstenbosch elegantly shows that the display of a wide range of opinions and arguments in public debates may countervail one-sidedness and overt partiality, and that one could think of a kind of division of tasks for the participants in the debate: intuitions are expressed by lay people, more precisely stated by opinion leaders, corrected by scientists, connected to moral principles formulated by ethicists and brought into interplay with background theories by those who present general views in the course of the debate. According to Vorstenbosch, the method of wide reflective equilibrium has mainly a heuristic function in bringing to light the variety of views on a particular issue (pointing out considerations that otherwise might have been left out of the debate) and making clear how

24

Wibren van der Burg and Theo van Willigenburg

arguments and considerations of the various positions are related, thereby creating balance and order without excluding viewpoints. Public debates may playa role in the development of common public morality, Le. shared convictions on how to interact with the non-specific members of the communities we live in. Frans Brom argues that the wide reflective equilibrium method can be used to understand the emergence and change of this 'morality of the public sphere'. In a pluralistic society, a common public morality is the result of an overlap between the different moralities for the public sphere that are endorsed by different (groups of) citizens. This overlap should not be conceived as a static set of beliefs but as the result of a continuing process of discussion within and between the various moralities, as an overlapping discourse. This overlapping discourse can be characterised as a deliberative interaction between positive social morality (de facto agreement) and critical morality (proposals for amendments and refinement of this social morality). This deliberative interaction may very well be modelled along the lines of an intersubjective reflective equilibrium process. In the above contributions, reflective equilibrium is presented as a method for pointing out and analysing the conceptual elements of moral thinking (private or public) and the way these elements are related. Jos Kessels argues that we also need to know more about the lnethodology as such, that is, about actual steps, interventions and procedures needed to bring about a reflective equilibrium of considerations. Kessels shows how a modem version of the classical Socratic dialogue can be applied as a tool for achieving reflective equilibrium. He does this by describing an example in which the Socratic dialogue method is used in a concrete case (value conflicts in a business firm). The example illuminates various methodological insights, like the importance of elenchus, moments of refutations and embarrassment, and aporia, moments of perplexity and deliberative despair, and it shows the fruitfulness of limiting fundamental questions to single examples and narrowing down the thinking process to specific judgements before developing statements of increasing wideness and generality. In the final part, Case Studies in Bioethics, four examples are given of how the reflective equilibrium method can be deployed in so-called 'applied' ethics. Egbert Schroten analyses the case of the bull 'Herman', in which a gene construct, coding for lactoferrine (a protein with possible anti-infectious effects), was successfully introduced. The question is whether it would be morally acceptable to have female offspring from this genetically modified animal in order to obtain lactoferrine in the mammary glands of cows. Schroten was the chairman of the Dutch governmental commission which had to advise on the moral admissibility of the genetic project, and he describes how the various

Introduction

25

elements and dynamics of wide reflective equilibrium thinking are visible in the deliberation and advisory work of this (multidisciplinary) commission. Bart Rutgers applies the reflective equilibrium method to two concrete cases in veterinary ethics: the declawing of cats and the routine practice of Caesarean section in beef cattle. He shows the importance and meaning of three fundamental moral principles that should guide our relationship with animals: non-maleficence, beneficence and the principle of respect for animal integrity. Decision making according to the reflective equilibrium method bears a strong resemblance to the way veterinarians try to solve clinical problems. Therefore, it should not be difficult for a veterinarian experienced in clinical thinking to make himself familiar with this way of moral thinking. Goran Collste discusses the dramatic issue of infanticide in the case of a mentally retarded, physically suffering newborn infant. He reconstructs various lines of argumentation both in favour of and against the view that the infant should be killed, and focuses on the role of background theories concerning the nature and value of human beings. He argues that also on the level of these fundamental theories and views rational argumentation is possible - e.g. on the importance of human dignity and the value of human life and on the moral importance of suffering - though, finally, the assessment of the arguments may be related to some personal point of view. Hans van Delden and Ghislaine van Thiel present a normative-empirical model of bioethics which acknowledges the relevance of practice-internal nonns for moral reasoning. They show that reflective equilibrium can be used as a method to elicit these internal norms and to integrate the considered moral judgements of professionals with insights and approaches stemming from ethical theory. Their case is the way the principle of respect for autonomy functions in Dutch nursing home care, and how this principle could be interpreted in order to be relevant in the particular context of care for the elderly in nursing homes. While starting from empirical data on the moral optnions of professionals, their normative goal is to suggest a better interpretation of respect for autonomy in nursing-home care. This project suggests fruitful new perspectives for the practical use of reflective equilibrium models and raises complex methodological questions. It is thus a good illustration of a common theme running through many contributions to this book: the fruitfulness of an integrated approach to methodological and pragmatic issues. By trying to deal with practical issues, we learn more about the problems and possibilities of reflective equilibrium methods and, conversely, by using the method of reflective equilibrium we can learn to cope successfully with practical problems.

Part I The Elements of Reflective Equilibrium

Moral Principles: Authoritative Norms or Flexible Guidelines? Marcel Verweij

1

Introduction

A man of principle is a man who acts according to his principles even if he has strong inclinations not to do so. Apparently, principles - at least moral principles - involve requirements that must be satisfied and that cannot be simply set aside. A man of principle will be convinced that he must do what his moral principles prescribe. To him, principles determine what is morally right, and they should therefore be obeyed. In this paper, I discuss the nature of moral principles in the context of moral reasoning along the lines of wide reflective equilibrium. In this context, principles seem to be not as fixed as 'the man of principle' assumes. After all, '[i]f a person's judgement regarding what is to be done in a particular case runs counter to what is to be done in that case on the basis of a general principle he is committed to, then, according to the method of reflective equilibrium, it is an open question whether the judgement or the principle should be retained' (Van der Burg and Van Willigenburg, this volume). The central question of my paper is in what sense principles can still be principles if they are that much open to revision or even rejection. In order to answer this question, I should first give an account of what principles are.

2

The nature of moral principles

Unfortunately, the views on the nature and status of moral principles differ widely in ethical theories. In some theories, principles are conceived of as indubitable foundations of morality. An example is the principle of utility in Mill's utilitarianism (Mill 1863/1969, chapter 2; Hare 1996, 197). In many other approaches, for example Beau.champ and Childress's biomedical ethics (1994), principles are not indubitable foundations, but general norms on a mid-level between ethical theory (for example utilitarianism) and concrete moral judgements. Such principles are not considered indefeasible; they can be overridden by other principles in cases of conflict. Yet, if a theory allows that a principle is overruled by other considerations, it is not implied that principles 29 HZ van der Burg and T. van Wi/ligenburg (eds.), Reflective Equilibrium, 29-40. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

30

Marcel Verweij

are downgraded to the status of merely rules of thumb. A rule of thumb is not a stringent norm but only a time-and-thought-saving device for decision making; in contrast to moral principles the breach of a rule of thumb does not excite compunction (Hare 1981, 38). If the meaning of the term 'principle' in philosophical ethics should not be too distant from how the term is used in COInmon language (as in the expression 'a man of principle'), then one should see principles as stringent and durable moral requirements. Even though principles have a different status in different ethical theories, I think it is possible to mention some features a norm should have if it is to be called a moral principle. Davis (1995, 88) stipulates that moral principles provide prescriptive action guides for moral conduct; they require that agents do certain actions when certain conditions obtain. This first stipulation should be further elaborated. First, as is implied in Davis's account, principles involve moral requirements of strong binding force. They do not just guide action by showing what would be good to do; they imply obligations that must be fulfilled. This is recognised by the 'man of principle', and I think this idea of 'principle' also reflects the way the term is mostly used in common language. If one fails to do what a principle requires, this may rightly evoke criticism from other people, or at least it should cause one to feel remorse. In this sense, principles can be contrasted with moral ideals. Second, moral principles are action guides that are universal in form. In their fonnulation, principles should not olention particular persons. An appeal to a moral principle therefore enables a person to attain an impartial perspective on a moral problem (Van Willigenburg 1991, 180). Third, moral principles summarise 'our own moral experience together with that of a whole community' (Heeger 1992b, 97-98). This characteristic comprises in fact two different features. On the one hand, principles summarise many experiences, which means that they are relatively general. 1 On the other hand, principles reflect moral experiences that are shared in a community. This latter feature explains why an appeal to a general moral principle can be an attractive strategy if one tries to convince other persons that they should agree with certain moral beliefs. Fourth, I think principles do not just summarise our numerous moral beliefs. They are often considered to be the cornerstones of a person's or a society's system of moral beliefs. In other words, they are fundamental (Heeger 1992b). 2 This feature has two aspects. On the one hand, principles are fundamental in

2

I think most authors assume this feature. An exception is Hare, who contends that principles at a critical level of moral thinking can be of unlimited specificity (Hare 1981,41). This does not necessarily imply that principles are also foundational.

Moral Principles

31

the sense that, if someone came to the conclusion that one of his moral principles is mistaken, this would probably cause a crisis in his system of moral beliefs. On the other hand, principles are fundamental because they cannot be reduced to some other moral concept (cf. Applbaum 1998,340). This may explain part of the special appeal principles have. Moral principles grasp our moral beliefs in such a comprehensive way that they do not just summarise moral experiences, but, moreover, hold these together. If this description is correct, moral principles have a special status within a system of moral beliefs. Even if they are not indubitable, they have a strong normative authority for the people who endorse them. Philosophical ethics may confirm (or reject) this authority by testing whether or not the principles are justified in the context of a wider substantial ethical theory, which in turn may or may not find firm ground in some 'Archimedean point' (e.g. human nature, or rationality).3

3

Reflective equilibrium as a method of theory acceptance

According to the method of reflective equilibrium, theory formation is a dynamic process; it does not start from a fixed, Archimedean point. Theories are developed by mutual testing and adjusting of all relevant types of moral and non-moral considerations. None of these elements (moral judgements, moral principles or any of the other elements) has a privileged status. If this process of·mutual support, critique and adjustment converges towards a reflective equilibrium in which 'all considerations are fitting together in one coherent view' , this point of coherence may be considered to be a provisionally, not absolutely, fixed point. Now, can principles still be endorsed as principles, Le. as authoritative moral norms if they are denied a privileged status? For an adequate answer to this question, it is necessary to distinguish between the initial principles, the principles during the process of reflection and adjustment and the principles that are the product of this process.

3.1 The initial set ofprinciples What is the status of moral principles at the start of reflection? According to, for example Rawls (1951), Daniels (1979a) and DePaul (1993), an inquirer should start with a set of considered moral judgements. The next step would be to formulate a set of principles that can explicate these considered moral

3

In this article I focus on the normative status of principles (their authority as binding norms) rather,than on their epistemological status, although I assume that these features cannot be completely separated.

32

Marcel Verweij

judgements. Strictly spoken this means that those principles, when applied to the cases that evoked the initial judgements, will yield the same conclusions as those judgements (Rawls 1951, 184-185). Consequently, the principles an inquirer starts with do not necessarily have an appeal that is independent of his commitment to his considered judgements. Thus, the initial set of principles does not consist of norms that have all the features I mentioned in section 2: They are presumptive principles rather than 'real' principles. Ronald Dworkin (1978) and Wibren van der Burg (1991) presuppose an alternative view of the status of principles. These authors point out that, in a method of reflective equilibrium, the initial set of principles should not just explicate judgements but also support these. Therefore, the principles should have a moral appeal that cannot be reduced to the inquirer's commitment to his concrete judgements. This latter view acknowledges that reflective equilibrium thinking (or any other coherentist method) will never start from a vacuum. Any inquirer will have an initial set of considered moral judgements but will also subscribe to a number of moral principles. These principles can already have the features I mentioned above, and have an initial appeal to the agent. This initial appeal may be explained by pointing out that they have their origin in common morality (Beauchamp and Childress 1994, 104) or that they are the result of prior theoretical reflection. For example, for many inquirers it would be impossible not to include a principle of non-maleficence and a principle of utility in their initial sets of principles. After all, they know in advance that these principles provide a satisfactory explication of many of their considered judgements.

3.2 The principles' status in the process of mutual adjustnlent Second, during reflection, the inquirer should criticise every element (for example a principle) in the light of the other considerations (for example judgements and background theories). He should be prepared to revise or reject judgements, principles and background theories in order to get a coherent set of considerations. In this process, none of the elements has a privileged status. However, this does not imply that every element is equally weak or strong. For example, a principle that conflicts with other consideratioI!s will probably not be revised as easily as certain moral judgements. Principles are uni versal statements, and when a principle is revised, the implications are much more comprehensive than when a concrete moral judgement is adjusted. This story is somewhat more complex if the inquirer's initial principles are 'mere' explications of his judgements. In that case, the inquirer starts with presumptions rather than 'real' principles. One may expect that at the beginning an inquirer will adjust the presumptive principles rather than the judgements he holds. But when, after some time, the principles seem to account for a large number of

Moral Principles

33

judgements, then the inquirer will tend to accommodate inconsistencies by revising judgements rather than principles (DePaul 1993, 19).

3.3 The principles that are the product of reflective equilibrium thinking Third, if this process of mutual criticism, adjustment and justification results in a stable reflective equilibrium, the coherence among all beliefs provides the justification for the product: a moral theory that contains the set of principles that have been developed. This justification is not absolute and timeless. If one is confronted with new evidence or radically new moral questions, this may be a reason to reconsider the conclusions reached earlier, and renewed reflection may shift the point of equilibrium. However, this does not imply that the principles are not reliable. The justification process may provide the principles with adequate support, so that they can fulfil their role as binding and fundamental moral norms. For example, Rawls's principles of justice are indeed to be understood as fundamental principles that should be applied and followed in designing and criticising the basic institutions of a well-ordered society. Now, this statement may be too optimistic because it presupposes that a stable reflective equilibrium can indeed be attained. Unfortunately, this can be doubted. According to DePaul (1993, 22), it is 'a state which no one could seriously hope to attain'. Rawls (1971, 51) acknowledges that wide reflective equilibrium is a philosophical ideal. Yet, if it is an ideal rather than a realistic state, there will be only limited support for a set of principles that has been developed through the process of mutual adjustment. More precisely, there will always be reason to continue the process of revising principles and judgements. Consequently, the basis to consider these principles as authoritative moral nonns seems to be rather weak. Does this imply that reflective equilibrium thinking is incompatible with such a view of principles? Not necessarily, I think. If the process of mutual adjustment is started with a set of principles that one endorses and not with mere presumptions of principles, then one will probably endorse even stronger the principles that, after revisions, have survived a process of scrutiny.4 The fact that the process has not (yet) resulted in 'one coherent view' is no reason to reject the principles that have been developed so far.

4

The presupposition here is that there is a substantial overlap between the implications of some initial principles and the implications of some of the final principles. An inquirer will have good reasons to strongly endorse the principles that reflect this overlap. If there would be almost no resemblance at all between the initial principles and the final principles and if the inquirer has not reached a stable equilibrium, then he will have little reason to endorse the final set of principles.

34

Marcel Verweij

4

Reflective equilibrium as a method of moral decision making

Robert Heeger and others have argued that the idea of reflective equilibrium also offers a fruitful model for moral decision making concerning concrete problems (Heeger 1988; Van Willigenburg and Heeger 1991). Moral reflection and judgement should not follow the lines of mere application of a received moral theory to the facts of a case and a subsequent deduction of moral conclusions. If we are confronted with a moral problem, we should test against each other our spontaneous intuitions, the relevant facts of the case and the principles we subscribe to. Through mutual criticism and adjustment of these elements we can hope to reach a point of equilibrium that offers sufficient moral reasons for a decision in the case at hand. If the method of reflective equilibrium is followed in theory formation as well as in moral decision making, the dynamics of moral theory is further increased. Moral practice and moral theory become strongly intertwined. Decision making in concrete cases may result in an adjustment of considered moral judgements as well as of our moral principles, thus shifting our 'fixed' point. This interwovenness of theory and practice is a welcome idea, as it may answer many of the critiques of the last decades on the 'application model' of applied ethics (Heeger 1993). Yet, it also invites philosophical problems regarding the status of moral principles. If in the analysis of any moral problem case, the justification and content of moral principles is (or even: should be) debatable, the authority of moral principles will inevitably be weakened. It becomes less self-evident that moral principles can be invoked to justify moral decisions. After all, other considerations about the case may point into quite other directions than a principle does. Upon reflection, the strong appeal of direct, concrete moral intuitions, supported by other principles (competing with the first principle) and the relevant facts of the case may be seen as a good reason for qualifying certain moral principles. Yet, the principle which is now to be adjusted or even rejected had been arrived at after thorough reflection, and it represented a point of coherence among vast bodies of morally relevant information concerning a large range of cases. It seems strange that critical reflection on a single problematic case can result in the adjustment or rejection of such a wellconsidered principle. If this is possible, principles appear to be rather flexible norms, which apparently lack the power to give rise to binding obligations. If a moral principle loses its fundamental credibility and power to support obligations, then what is left is a rule of thumb that, if followed, offers one a good chance to pick the right answer to moral questions. Is this concept of a moral principle correct? Or is it possible to safeguard a certain moral authority of principles even if these are open to criticism and adjustment in concrete moral decision making?

Moral Principles

5

35

Two levels?

A first possible answer I will mention only brietly. In this answer, a certain division of labour is proposed, like Richard Hare does in his Moral Thinking. Although Hare despises methods of reflective equilibrium because, in his view, these depend too much on intuitions (Hare 1981,75),5 his idea that moral thinking can take place at two different levels may be adopted by proponents of reflective equilibrium. In such a view, a first level of moral thinking is meant for normal, daily life, in which there is no need and no time to discuss every moral choice in depth. At this level of thinking, principles are assumed to be prima-facie binding and beyond doubt, and these may 'simply' be applied. We trust these principles as adequate moral norms that can be central in, for example, the education of our children. However, moral dilemmas cannot be solved by simply applying these prima-facie principles. Such cases require reflection on a critical level. While in Hare's view, critical thinking is thinking along the lines of universal prescriptivism, one could also see critical thinking as reasoning along the lines of the method of reflective equilibrium. At this level, moral principles are not simply applied; they are scrutinised against all other relevant considerations, and, if necessary, they are revised or rejected. Such a two-level approach is attractive because it avoids the predicament that one's principles should be questioned any time one is confronted with a moral problem. However, this approach is not without problems. For example, it may be argued that almost every moral question requires critical reflection (Van Willigenburg 1991,62). At least in every case in which one principle conflicts with another (which is not uncommon in practices such as health care), the process of mutual adjustment should start again. If so, the problem that principles appear to be 'flexible' rules rather than fundamental principles that an agent can endorse as binding norms, is not yet solved.

6

Principles specified

What is needed for reflective equilibrium as a method for moral decision making, is that principles can be adjusted, but that the side-effect is avoided that this weakens their authority to guide action. One answer to this problem is that' ... [principles] are usually not revised, as in common conceptions of WRE, but made more specific'. The idea of specification of norms has been developed

5

Although Hare's method is certainly not a method of reflective equilibrium, he considers his approach to be a coherentist one, or rather a reconciliation of coherentism and foundationalism (Hare 1996).

36

Marcel Verweij

by Henry Richardson (1990) and elaborated by David DeGrazia (1992). According to Richardson, this notion can 'explain how a moral theory can remain the subject of a more or less stable attachment despite the sort of revision that moral conflicts engender' (Richardson 1990, 284). When, in a concrete case, two or more principles conflict, this problem may be resolved by making one of these principles more specific. This specification should not be the introduction of an exception to the principle; the principle should be qualified substantially so that it is tailored to the case at hand. Richardson lays down some conditions on the relation between the initial principle and its modifications or specifications that explain that the revised principle or better, the specification - is in some significant sense the same as the initial principle. According to Richardson, if nonn p is a specification of nonn q, then p qualifies q by adding clauses about the scope of the norm or the nature of the act or end enjoined or proscribed. None of these clauses in p should be irrelevant to q and p should not simply be q-with-some-exceptions. 6 Specifying a principle means that one is getting a better understanding of how the initial principle bears on the case at hand. Thus, specification reinforces one's commitments to the initial principles. Although Richardson focuses on conflicts between principles (or, more generally, moral norms), specification may also be important when there is incoherence between moral principles and moral intuitions. For example, one may question whether it is justified to include in a phase-l trial an incurable cancer patient whose prognosis has been judged poor. The aim of such a trial is to assess toxicity of a new drug. Suppose an oncologist knows some patients who are willing to try every possible experimental therapy, even if the chances that the therapy will have beneficial effects are almost nil. Respect for autonomy requires that potential subjects are adequately informed, and that they voluntarily give their consent to participation in the trial. Now, suppose that one patient is well informed about the trial. It has been explained to him that the only aim of the trial is to test the toxicity of the new drug, and that the chances that participation in the trial will increase his life expectancy are considered to be negligible. Yet, the patient is highly motivated to participate because he has nothing to lose and he wants to try every possibility to fight his disease. Suppose his consent is indeed well informed and voluntary. Is it right to include this patient in the trial? We may have a strong moral intuition that doing so is still a sort of abuse of this patient and his predicament. Although 6

Richardson emphasises that the initial norms should not be absolute with respect to the domain of acts, but rather have the form 'Generally, one ought to ... '. This is necessary to avoid that specifications either are simply implied by the initial norm or that they are logically inconsistent with the initial norm (Richardson 1990, 292). According to DeGrazia, Richardson's assumption is more or less compatible with seeing moral norms as prima-facie binding.

Moral Principles

37

the normal requirements of respect for autonomy have been fulfilled, this moral intuition of patient abuse may enable us to see another element of respect for autonomy, namely that persons should never be used solely as a means for our ends, but also as an end in themselves. If a patient wants to participate in order to aspire the almost unattainable end of improving his health, this end strongly diverges from the physician-researcher' s end. Only if the patient freely decided to participate/or the sake a/the scientific objectives a/the trial, this divergence is settled. Only then the possibility of abuse of his predicament is ruled out. So, it may seem reasonable to specify the principle of respect for autonomy somewhat differently if patients in such a predicament are sought to function as experimental subjects. A well-informed voluntary consent would not be enough to satisfy the principle of autonomy. (It would be in most other cases of medical experimentation.) What would be needed in such a case is a specific motivation of the subjects: they should be willing to participate for the sake of the study objectives. This specification, 'Subjects should be motivated to participate in a trial for the sake of the scientific objectives of the trial', is at least justified in the case of medical experiments on incurable patients. It is not irrelevant for other cases of experiments on human subjects, although in other cases there will be fewer reasons to consider it to be a binding moral requirement governing the enrolment of research subjects. In this example, the principle of respect for autonomy is not revised; the moral intuitions and morally relevant facts of the case are considered to be the reason to propose a specific understanding of the principle. This specification may strengthen rather than weaken one's commitment to the initial principle of respect for autonomy. So, a specification of principles may be one way out of the predicament that the tests and revisions of principles, as needed in reflective equilibrium thinking, will weaken one's commitment to principles. Ultimately, endorsing the method of specification means that one acknowledges that principles can be interpreted or specified in several, if not numerous, ways. Explicating a moral principle then involves laying out the various specifications and implications of the principle for diverse cases. One of the problems of the method of specification is that in many problematic cases several competing specifications will be possible, 'thereby returning us to conflicts of the sort that drove us to specification in the first place' (Beauchamp and Childress 1994,31). However, Richardson did not conclude that his method would always yield one resolution to a problem. Likewise, the method of reflective equilibrium cannot rule out the possibility that reasoning about a moral problem can result in more than one equilibrium point.

38

7

Marcel Verweij

Principles as clusters or containers

Specification leads one to seeing the initial comprehensive principle as a large set of specified norms, each of which expresses one understanding of the initial principle. Specified norms will often have a strong binding force, but some specified norms will be binding in one situation, and non-binding in another. My example of medical experiments on patients illustrates this thesis. The specified norm 'subjects should be motivated to participate in a trial for the sake of the scientific objectives of the trial' can be considered binding in some cases of medical experiment, but not in others. Apparently, then, principles are chapter headings for a wide range of moral considerations concerning, for instance, autonomy. Remarkably, in the latest edition of Principles ofBiomedical Ethics, Beauchamp and Childress present their framework as four clusters of principles (Beauchamp and Childress 1994, 37). Indeed, these principles can be invoked to support all kinds of moral conclusions. For instance, the principle of autonomy is often invoked in order to support obligations of informed consent, but also to encourage physicians to improve patient autonomy (an ideal), to stress that coercive measures in health care are immoral, to emphasise the importance of physician-patient talks (also an ideal), to justify the idea that physicians should counsel their patients in a non-directive way (again an ideal), to explain that patients should never be treated solely as a means but also as .an end in themselves, etc. But, if principles are clusters of norms and if these norms can have a varying binding force in different cases, this will, again, raise questions concerning the status of the 'initial', or 'over-all', principle. If a principle is to be a fundamental moral norm of strong binding force, one should beware not to make it all-embracing. The more a moral concept, for example autonomy, can be extended through diverse specifications, the less it may be capable of guiding action. On the one hand, as stated above, a principle may yield competing specifications in one situation. On the other hand, if many specifications are not binding, agents may doubt whether a principle yields obligations at all. Can such a principle still be an authoritative and fundamental moral norm for a moral agent? It is important to stress that principles are not fundamental because they are comprehensive and 'cover' many moral intuitions orjudgements. Principles function as 'cornerstones' in a system of moral beliefs because of their ability to express, in an illuminating way, certain moral beliefs. They do so by means of abstract concepts such as 'autonomy' or religious concepts such as 'the sanctity of life' . These concepts enable us to grasp (in bothe senses of 'to catch' and 'to understand') our central moral beliefs. Yet, the strong appeal principles have for an agent does not only depend on these (abstract) concepts. The core of a moral principle can often best be explicated by referring to paradigmatic

Moral Principles

39

situations or acts in which the principle is evidently violated - in other words, situations in which certain obligations the principle gives rise to are clearly not fulfilled. For example, the core of the principle of respect for autonomy becomes most apparent if we imagine concrete examples of a person who forces other human beings to act for his sake. The core meaning of a principle of beneficence can be well explained by imagining a situation in which people stare at a drowning child and do not attempt to rescue the child. The clear core of moral principles consists of strict and binding moral requirements. The further a moral problem diverges from such paradigmatic situations, the less clear, binding and certain the implications of the norm will be. So, around the clear core of a principle, a penumbra of related - though less strong - moral considerations may be imagined. If principles are interpreted in this way, then the application of principles may sometimes involve a sort of casuistry, in which the implications of the principle are assessed by comparing the case at hand with paradigmatic cases that can illuminate the principle's clear core.? Casuistry may help one to explore specifications within the penumbra of the principle. If, in a moral case, the penumbra of a principle is at stake and not its core, the role of the principle seems to be primarily a heuristic one. It helps one to choose the morally right thing to do. Principles do not determine what is the right thing in every situation in which they are relevant. However, as principles have a clear core of binding moral requirements, they can still be endorsed as authoritative moral norms.

8

Conclusion

Principles are general, universal and fundamental norms that summarise moral experience and give rise to binding moral requirements. Principles therefore have strong normative authority for the people who endorse them. In a reflective equilibrium approach, the authority of principles is put under pressure. In this paper I have argued that, in various contexts, principles should have additional characteristics in order to safeguard their special nature. First, in a theoryoriented reflective equilibrium process one should start with initial principles that have a moral appeal of their own - they should not simply be generalisations of considered judgements or intuitions. Second, if the method of reflective equilibrium is followed in decision making about concrete moral problems, one should acknowledge that principles are mostly not revised during the process, but specified. Third, if principles are seen as clusters of specifications rather 7

The idea that casuistry can operate within a method of application of principles is elaborated by DeGrazia (1992).

40

Marcel Verweij

than as singular norms, one should recognise that the clear core of a principle consists of binding moral requirements.

Morally Relevant Facts: Particularism and Intuitionist Rationality Theo van Willigenburg

1

Introduction

(Wide) reflective equilibrium has gained popularity especially as an argumentative method for tackling practical moral problems. It is employed as a framework of guidance to philosophical thinking aimed at answering hard moral questions which arise in practical contexts like medicine and business. In thinking over practical problems all kinds of considerations may be facts of a case, that is, the facts that have to be taken into account in coming to a moral judgement. When Robert Heeger and I introduced a 'network model' of practical moral thinking, we suggested that to reach a justifiable moral judgement in a moral case at least four 'operations' were required: (1) searching for morally relevant facts of a case; (2) making explicit the moral intuitions one has with regard to the case; (3) searching for possibly applicable moral principles and elucidating the case from the perspective of these principles; and (4) balancing morally relevant facts, intuitions and principles until an equilibrium is reached (Van Willigenburg and Heeger, 1991). However important the recognition of morally relevant facts may be as a distinct 'element' in the reflective equilibrium thinking process, there are considerable difficulties in making clear what is meant by calling something a 'morally relevant fact'. What is a morally relevant fact? What makes facts morally relevant? In this contribution, I will show that if one wants to answer these questions, one will have to take position in the current debate in moral epistemology on 'generalism' and 'particularism'. I will argue that if one endorses a generalist picture of what makes facts morally relevant, one will have considerable difficulties in assigning an epistemically independent role to morally relevant facts. I will defend, therefore, a moderately particularistic picture of moral relevance, in which recognizing case features as morally salient is a more or less independent 'operation' in reflective equilibrium thinking. This position has some far-reaching consequences for the model of rationality endorsed in a wide reflective equilibrium approach. At the end of my contribution I will spell out some of these consequences.

41 ~

van der Burg and T. van Wil/igenburg (eds.), Reflective Equilibrium, 41-54. @ 1998 Kluwer Academic Publishers. Printed in the Netherlands.

42

2

Theo van Willigenburg

Facts as reasons

What is a morally relevant fact? I will start my analysis and discussion with a well-known parable: There was a landowner who went out early one morning to hire labourers for his vineyard; and after agreeing to pay them the usual day's wage (one denarius each) he sent them off to work. Going out three hours later he saw some more men standing idle in the marketplace. 'Go and join the others in the vineyard', he said, 'and I will pay you a fair wage'; so off they went. At midday he went out again, and at three in the afternoon he made the same arrangement as before. An hour before sunset he went out and found another group standing there; so he said to them, 'Why are you standing about like this all day with nothing to do?' 'Because no one has hired us', they replied. So he told them, 'Go and join the others in the vineyard.' When evening fell, the owner of the vineyard said to the steward, 'Call the labourers and give them their pay, beginning with those who came last and ending with the first.' Those who had started work an hour before sunset came forward, and were paid one denarius each. When it was the turn of the men who had come first, they expected something extra, but were paid the same amount as the others. As they took it, they grumbled at their employer: 'These latecomers have done only one hour's work, yet you have put them on a level with us, who have sweated the whole day long in the blazing sun!' The owner turned to one of them and said, 'My friend, I am not being unfair to you. You agreed on the usual wage for the day, did you not? Take your pay and go home. I choose to pay the last man the same as to you. Surely I am free to do what I like with my own money. Why be jealous because I am kind?' (Matthew 20,1-16)

Wittgenstein wrote that the world is the totality of facts, not of things. He meant that facts seem to be shaped just like sentences. 'It is a fact that p' is the same as 'it is true that p' and this can be reduced simply to: p. Facts are not some kind of atomic, independent structures or 'things' to be found 'out there', but are part of our description of the world. Facts are features of what we encounter in the world; features which are isolated in a manner functional to the particular way of encounter. Morally relevant facts are features of situations which are relevant in our moral involvement with those situations. Features of a situation may be relevant if they can figure as a reason in a moral judgement on that situation, that is, if they can give a reason for a particular judgement on the moral rightness of actions and policies or on the moral goodness of attitudes, states of affairs or results. If a fact can justifiably be called upon as a reason to support a particular moral judgement, it is morally relevant. So the fact that some labourers worked

Morally Relevant Facts

43

a whole day and others only one hour is morally relevant in judging the fairness of the amount of money they are paid. In answer to the grumbling workers of the first hour, the landowner points out another morally relevant fact which, in his eyes, provides a conclusionary reason in favour of his system of distribution of income: there was a clear agreement concerning the payment they would get, and the landowner simply sticks to this agreement. What's wrong with being as good as one's word? There are two ways of interpreting the landowner's argument. He may be understood to say that the fact that some worked the whole day in the blazing sun is not morally relevant at all, because the only thing that morally counts is whether he keeps to the agreement. This would amount to saying that there is only one (set of) fact(s) that is morally relevant: the fact(s) which provide(s) a conclusionary reason for judgement. As they do not provide such a conclusionary reason for judgement, other facts are denied moral relevance or their moral relevance is effectively 'silenced' by the fact(s) providing conclusionary reason. However, it is not very likely that particular facts (some worked a whole day and others only one hour) would have to be regarded as fully irrelevant because other facts prove decisive. It seems to be more obvious to think that the landowner does not deny that the difference in hours worked is morally relevant in judging the fairness of the wages, but that he insists that standing by an agreement to which every labourer has freely consented has greater moral weight. The fact of the agreement does not annul or silence the relevance of other features of what is going on; in other words, such a fact has no 'preemptive' or 'exclusionary' force against other facts (cf. Raz 1978). For instance, in answer to possible criticism on the content of the agreement ('it is unfair to agree to pay the workers of the first hour the same amount of money as the workers of the last hour'), the landowner points out some other morally relevant facts in support of his policy: it is his money and he is free to do with it as he likes. Moreover, why blame the landowner for being kind, Le. for agreeing to pay the labourers of the last hour the same amount of money as the first ones? It seems that the argument is not about what facts are morally relevant or not, but about the reason..giving force of these facts (cf. Harman 1978). The labourers could argue that fairness of income is more important than keeping to the letter of an agreement or being kind. Or they could try to show that in designing a fair distribution of income it is more important to use comparative criteria than absolute ones (fair minimum). The disagreement is about the relative weight of various reasons, that is, about their direction (Does this fact support or weaken the justifiability of a possible moral judgement?) and about their force (How strong is the reason this fact is giving us in comparison with other reasons?).

44

Theo van Willigenburg

Not all facts of the case are morally relevant. The fact that the landowner is growing Grenache in his vineyard instead of Sauvignon is probably not morally relevant. I say 'probably', as even facts that usually have no moral import whatsoever (How was the weather?) can become morally important in a specific context. In the parable the weather is surely not completely irrelevant: sweating a long day in the blazing sun is hard work and would probably deserve a better wage than work under more comfortable conditions. Of greater interest than facts which at face value have no moral import (the age of the landowner, the name of the vineyard, etc.), but which could in specific circumstances gain moral relevance, are facts which should not have moral relevance, like the skin colour of the labourers. Race, gender or sexual orientation are regarded as facts which should bear no moral weight at all. Here we have explicit reasons for not assigning moral weight to a particular fact, which is different from not having reasons to assign moral weight (cf. Raz (1978) on 'exclusionary' reasons, which are reasons for disregarding other reasons). A morally relevant fact is a fact that justifiably bears moral weight. It justifiably counts as a reason in arguing for a moral judgement. This means that what counts as morally relevant is subject to reflection and deliberation. In deliberation one points out other reasons which support counting a fact as morally relevant or assigning it a particular weight. These other reasons may be of a more general nature, like the reason expressed in moral rules (e.g., 'keep your agreement') or moral principles (e.g. principles of fair income distribution or a principle of beneficence). These more general considerations provide arguments in assessing the direction and force of the moral reasons provided by the facts of a case. Thinking over the moral relevance of facts naturally involves a person in a process of wide reflective equilibrium thinking in which facts, rules, principles and theories playa deliberative role. What is not clear, however, is the exact relation between facts as reasons and the reasons provided by more general considerations like moral rules and principles. Though the moral relevance (and particular moral weight) of a fact may be backed up with other, more general reasons, the 'facts of a situation' can be thought of as providing complete reasons. If the landowner points out that it is 'his money' and that he is free to do with it what he likes, the statement of this fact need not be supplemented by a whole catalogue of statements on the meaning of private property and the force of property rights in order to count as a reason which may play its role in the balance of reasons. Still, the more general background considerations on private property seem to be presupposed in our thinking that the fact that it is 'his money' bears moral relevance. Are more general considerations decisive, then, in assigning moral relevance and specific moral weight and direction to facts? Or can the recognition of moral

Morally Relevant Facts

45

relevance of facts be thought of as in some sense sui generis, that is, independent of general reasons as expressed in rules or principles? I will go into these questions and distinguish two main positions: generalism and particularism. These positions give different answers to the question of what makes a fact morally relevant, and this has direct implications for the role morally relevant facts play in wide reflective equilibrium thinking. Is this a more or less independent role - that is, does the recognition of morally relevant facts have a self-reliant epistemic function in moral thinking? Or is the endorsement of the moral relevance of facts epistemically indistinguishable from the reflection upon moral rules, principles and theories? I will argue that the answer to these questions not only tells us about the role of morally relevant facts in wide reflective equilibrium thinking, but that it also has direct implications for the nature of rationality ascribed to moral enquiry.

3

Rules as sources of moral relevance

As has been argued by Quine (1960), words acquire meaning in the context of a language and language only has meaning in the context of particular practices. The isolation of certain events as 'facts' is practice dependent. For instance, an exchange of certain sequences of utterances in the market place creates an agreement because of a particular economic practice. Similarly, pointing out certain facts as morally relevant is dependent on moral practice. The amount of money paid to the workers of the first and of the last hour is a morally relevant fact because we regard it as morally appropriate to pay wages according to effort or merit. If what is morally appropriate can be expressed in terms of rules or principles which lay down the morality built into a certain practice, then these rules and principles are detenninative in pointing out morally relevant facts. This means that a property of a situation or a case to be assessed derives its moral relevance from principles or rules which in general tenns define a moral practice. In Richard Hare's words: 'In general, to treat a feature of a situation as morally relevant is to apply to that situation a moral principle which mentions that feature.' (Hare 1981, 63) This is a so-called generalist position on what makes facts morally relevant. According to the generalist a property, feature or aspect of a case can only be morally relevant if it is picked out by some moral rule or principle. This means that a property can only be morally relevant if it is invariably relevant, that is, if it will make the same kind of contribution to the moral judgement in other cases that have that property. Because it makes the same kind of contribution, there can be a rule or principle (which is some general statement) that picks it out as an aspect that should carry relevance in moral thinking.

46

Theo van Willigenburg

According to the generalist, moral rules or principles, or whatever general considerations on the level of moral or background theory, play an indispensable role in any well-founded moral explanation of the bad- or good-making relevance of case features. In this picture general considerations are indispensable in pointing out which facts may justifiably carry moral weight in assessing a case. A good example of such a position, which illuminates the generalist idea of rationality related to the justifiability of moral relevance, is Barbara Herman's reconstruction of Kant's moral theory (Herman 1996). Kant's categorical imperative functions as a test for moral maxims that express the agent's conception of what is morally required in particular circumstances. A maxim summarizes the agent's balanced reasons for action, Le. his sense of the relative weight of the different moral features present and his idea of a suitable moral response. Final moral judgement is made by bringing this maxim to the categorical imperative procedure, Le. by uni versaiizing it and seeing whether one could reasonably will to have this maxim tum into law. Herman argues that the Kantian categorical imperative cannot be effective as a practical principle of judgement unless agents can formulate sensible maxims, and this requires that they have some prior moral understanding of actions and situations. They should be able to 'to pick out those elements of the circumstances or of proposed actions that require moral attention' and they should be able to discern a 'configuration of moral features'. To do this, the agent needs - Herman says - so-called 'rules of salience' , especially in those cases where it is not clear what is at stake, morally spoken, and what should be done. 'Rules of moral salience' , Herman says, 'enable an agent to appreciate what is at issue in hard cases by making perspicuous the morally significant features that make them hard.' (Herman 1996, 79) Moral awareness or sensitivity is needed to discern morally salient features. But why should moral awareness be guided by rules of salience? Surely, Kant did not introduce such rules into his system. Why are rules needed? According to Herman, rules of salience 'structure an agent's perception of his situation so that what he perceives is a world with moral features'. But why use rules to structure and enhance an agent's perception? In order - says Herman - to ensure that the objectivity of the categorical imperative procedure will not be threatened by 'too much detail and insufficient uniformity of (morally relevant) description', rules are needed 'to prevent an agent from thinking of (and so describing) his action in morally idiosyncratic ways' (Herman 1996, 76). According to the generalist position, case features can only be morally relevant if and only if they are invariably relevant, that is, on a general level. For, if we had to concede that a feature can be morally relevant in one case but have no relevance in another case or even have the opposite moral relevance (instead of a bad-making characteristic it is now a good-making aspect), we would

Morally Relevant Facts

47

abandon the possibility of rationally explaining and justifying why we thought that the features had that particular moral relevance in the first case. Explaining moral salience would be impossible if a feature could assume moral importance in one case, without having a similar role when exemplified elsewhere. This is why the discernment of moral salience needs to be a rule-guided activity. The generalist position on the moral relevance of facts is motivated by a strong idea of what is required by rationality in moral thinking. The generalist position on moral rationality, however, has a perhaps unforeseen consequence for the status and role of morally relevant facts in moral thinking. If facts can only be morally relevant if they are pointed out as such by rules and principles, then it seems that morally relevant facts will play no independent and distinct role whatsoever in moral deliberation. For to know that a fact is relevant is to know a moral rule or principle - Hennan even thinks that special rules of salience are needed - which picks it out as relevant. The reason for action provided by a fact is already fully expressed by a rule of principle that is known independently of the fact. It seems, then, that case features having moral relevance adds nothing to the thinking process, Le. nothing that could not already have been known at the moment one endorses a moral rule, principle or piece of theory. In the generalist picture the endorsement of the moral relevance of facts is epistemically indistinguishable from the reflection upon moral rules, principles and theories.

4

Particularist moves

I think that the recognition of the moral relevance of facts may have a self-reliant epistemic function in wide reflective moral thinking - Le. a function which is initially independent of more general considerations as expressed in rules and principles. In this I take a particularist position with regard to moral salience. According to this particularist position, facts (like properties of cases) can be morally relevant without this relevance being explicable by reference to uniformly valuable properties (enumerated in some rule). That is, particularists will allow for a moral relevance of facts sui generis. Such facts provide particular moral reasons not (fully) explicable in more general terms. A good example of the difference between particular and general reasons is provided by the story of the man who can only save one of several people in a burning house and wants to save his wife Rose. What reason could he give for this choice? He could say that as a husband it is his duty never to let down his spouse. Or, he could point out the long-time relationship they have had: 'Rose has given me so much, I owe it to her to risk my life saving her.' Or, perhaps, he may point to his promise

48

Theo van Willigenburg

20 years ago in front of the altar 'always to sustain his spouse' and, as you know, an honest man's word is as good as his bond! These are all general reasons (marital duty, reciprocity, promise keeping), and nobody will deny that they are good reasons, though the feeling is there that something is missing, some core reason: he loves his wife and dearly wants to save her. Why does he love his wife? Not because he has promised to do so. Not because she is a beautiful brunette (there are more beautiful brunettes). Not because of any set of generalizable qualities she might have. He loves his wife, because it is she. The final reason for wanting to save her is that she is Rose. This is the most precious reason, because it is the most particular one, not explicable in more general tenns. It is a reason which counts only for the husband. The fact that it is Rose who is in danger creates for him a particular and decisive moral presumption in favour of saving her. One could argue that 'it is Rose, my beloved' does not provide a complete reason. The complete reason would be 'it is Rose my beloved, and one has a duty to risk one's life in order to save a beloved person', which introduces a general reason (the duty to risk one's life to save a beloved person). If this is not specific enough, one may specify one's generaJ reason, e.g. 'one has to risk one's life to save a beloved person like Rose' , and include this as a general consideration in order to have a complete reason. However, the problem with specifying the general reason is that one may end up with a particular, agentrelative reason disguised as a general one, which does not add anything to 'because it is Rose'. The idea of the counterargument I am discussing here is that a complete reason will always involve some genuinely general moral consideration. I think, however, that this' generalist' reconstruction of the reason the man has for saving Rose does no justice to his thoughts and intentions. He must save Rose, because it is she. Even if he would not acknowledge a more general duty to risk one's life in order to save beloved persons, he could be fully convinced that he has to risk his life to save his wife because it is she. If asked why he thinks so (why not save the other persons), he would say: because it is Rose, because it is she (and the others are not Rose). Why not acknowledge this as a complete reason? Particular reasons are provided by morally relevant facts sui generis, not explicable in more general tenns, that is, they are not 'backed up' by some general reason-giving consideration. Because of their 'particularist' nature these facts may sometimes, but not always, create some moral presumption, or they may only create a moral presumption for some specific person involved in the situation (and not for others involved). In explaining moral relevance sui generis particularists usually do not point out examples like the one above about love and special duties, because there is a long-standing debate in moral philosophy whether or not duties of love should be counted as moral duties given their partial

Morally Relevant Facts

49

nature (see, e.g., Wolf 1992; Archard 1995). If impartiality is a defining feature of the moral point of view, then the fact that 'it is Rose' may be a good reason for the man to save her, but this would not be a moral reason, and no example of a 'particularist' moral reason would have been given. I believe, however, that there are good arguments for defending the position that partial duties should count as moral ones (Van Willigenburg 1998a; Kamm 1992). But I will tum to another - very influential - way of explaining and defending particularism in the field of moral reasons. Judging an act or policy in a certain situation to be morally right or wrong is resultant on weighing a variety of reasons for or against such a judgement. These reasons are provided by the facts of the situation. According to the particularist, the weight or direction of the moral presumption provided by a fact is strongly dependent on the context, that is, on the other facts of the situation. Whether or not a fact - say being frank - creates a positive moral presumption ('that is the morally right thing to do') depends on numerous other facts: one's intention (Is one trying to be honest to a good friend or does one intend to hurt someone else's feelings?), history (Has one promised to be confidential about things or may others rightly expect you to be frank?) and future (What will be the consequence of being frank?). So, the way in which a fact or consideration functions in one situation will or at least may be affected by other facts present in that case. This implies that the same fact or consideration in another case may provide a reason in favour of a completely different moral judgement, or may even provide no reason at all because of the other facts and considerations present in that case. Changes in the attendant circumstances can alter or nullify the moral tendency of a particular feature. Facts which in one case give reason in favour of performing an action may in another case give no reason at all, or even a reason against. This 'holism in the theory of reasons' (Dancy 1993, 60) entails a more or less strong form of contextualism which can explain sui generis moral relevance: some facts will have particular relevance only in the context of a certain constellation of facts. Such facts may be important in judging a case, but the generalist will miss them because he only acknowledges features which are universally and invariably relevant. According to the generalist, the assignment of moral relevance must be backed up by rules or principles, which excludes the possibility of having the moral relevance of the same fact vary from one case to another. There are two ways for the generalist to respond to the contextualism which particularism brings to the fore. The generalist could deny that it is the same fact which creates a positive presumption in one case and a negative one in another. Speaking out frankly what one thinks because one is an honest and straigthforward person is different from being frank with the intention to hurt someone else's feelings. In 'isolating' a morally relevant fact one has to include

50

Theo van Willigenburg

the agent's intention, and this shows that in the two cases we have to reckon with very different facts, not one and the same fact which has a radically different import because of attendant facts. The problem with this generalist move is that it makes it impossible to take the facts of a case at face value. Surely, what counts as a fact is determined by the various concepts provided by language 'games' in which we are involved when we think over a case. But it would be strange to have one language 'game' - the moral one - fully determine what may be seen as a 'fact'. At face value the frank utterances of the agent count as a fact building the story of the case, understandable as such, Le. independent of the intentions of the agent, though the moral import of that fact will strongly depend on those intentions. And even if one were to include intentions in the description of facts (as, for instance, a judge does in distinguishing murder from manslaughter), it would be strange to include also all other relevant attendant circumstances that might alter the moral import of the same fact. This is what I meant when I said that it would make it impossible to take the facts of a case at face value. In order to isolate what counts as a fact one would first have to come up with a full moral assessment of the case. This it not the way in which we normally think about individuating facts. Still, the generalist could, when trying to respond to the contextualism and holism of the particularist, include in rules or principles the countervailing reasons - if present in a case - that might radically alter the moral import of a fact: 'Being frank is morally right, except if it is done with the intention to hurt someone else's feelings or if one has promised confidentiality or if ... ' The problem with this strategy is that it seems to provide the agent with the wrong reason for acting. I am not obliged to help this boy with his homework because he is my foster son consigned to my care and because it is not the case that I have promised to visit my mother instead, and because it is not the case that I have the intention to manipulate the boy's thoughts and feelings so that he will choose to study philosophy though he really wants to become a medical doctor, and because it is not the case that the polders around Utrecht are flooded and my help as a volunteer in fighting the water is wanted. The absence of all possibly countervailing reasons is not a reason for action in this case. Only in the case of countervailing reasons which in some sense are ready at hand, it could be relevant to point out their absence: in judging the frank behaviour of a rude person it could be relevant to know that it was not this person's intention to hurt another's feelings. Some countervailing reasons may be present so often that they are commonly mentioned as an exception to a role. However, explicitly including all possibly countervailing reasons in a rule is not only absurd but also impossible. And adding phrases to a rule or principle like 'in the absence of any countervailing reasons' does not liberate the generalist from the strong contextualism revealed by the particularist. On the contrary, it confirms that

Morally Relevant Facts

51

moral judgement is deeply context dependent and that it is the unique configuration of facts in this case which gives reason for judgement. Contextualism in the theory of reasons does, however, not fully exclude invariability of reasons. I am in favour of a moderately particularist position, according to which some features of a case will have sui generis salience, but there will also be features which are relevant because they are universally and uniformly relevant and, therefore, can be pointed out by a rule. The strong particularist thinks that the moral salience of case features always depends on the circumstances and, therefore, always is variably relevant. (This is - roughly .- Jonathan Dancy's position, see Dancy (1981) and Dancy (1993), especially chapter4 'Why particularism?' .) Moderate particularism acknowledges, however, that there are facts which always bear a positive or negative moral import and, therefore, can be captured by a rule, but it denies that moral rules can identify every moral feature. I understand, for instance, W.D. Ross's (1930) list of general prima-facie duties (duties of fidelity, reparation, gratitude, justice, beneficence, self-improvement and non-malificence) as an enumeration of fundamentally relevant characteristics of actions, or, in McNaughton's words, as 'features of actions which are right or wrong-making characteristics and which always carry weight when we are considering whether a particular action is right or wrong' (McNaughton 1996, 435). In other words, the list points out properties of acts and situations which have invariant moral salience. But, apart from these, there may also be properties the moral import of which cannot be predicted from their role elsewhere, because in one case they may function as a reason in favour of performing an action, in another as no reason at all and in still another case even as a reason against. Therefore, (prima-facie) moral rules cannot point out every morally salient aspect of a case or action. This is the reason why the recognition and acknowledgement of moral salience is a distinct epistemic function in wide reflective equilibrium thinking. General considerations can be a reminder of morally relevant properties that may be present in a case,>Le. a reminder of the moral importance that a case fact could have. But the recognition of morally relevant facts is only partly dependent on the invocation of general considerations. It also requires detailed attention for what is going on in a case, trying to find a comprehensive view of unique constellations of facts in the case. Note, that the sui generis salience of features which particularism acknowledges is not purely accidental. There may be good reasons to think that a particular feature has important moral salience. But this reason need not be a general one (to be stated in terms of a rule). It may be based on a holistic view of the very particular case.

52

Theo van WiLligenburg

The interesting result of this is that in wide reflective equilibrium thinking, where morally relevant facts play an epistemically distinct role, rationality - What makes us think that a fact justifiably bears moral weight? - cannot just be what generalists like Barbara Hennan want it to be. According to the generalist we can only rationally explain and justify why we think that a case feature is morally relevant by pointing out this feature's invariable relevance. Coherence and consistency in thinking would be impossible if a feature could be morally relevant in one case but have no relevance or even an opposite relevance in another case and we could not give a general reason for this difference. According to the generalist the notion of sui generis moral relevance is mysterious and confers no explanatory power (Shafer-Landau 1997, 595). For the particularist, however, coherence and consistency in thinking not only has to do with invariability of reasons but, first of all, with the ability to get things right case by case. Rationality in moral outlook demands a range of sensitivities, 'so that no relevant feature escapes us, and we do not mistake its relevance either' (Dancy 1993, 64). Generalist rationality demands the search for general considerations, Le. considerations that carry weight here and elsewhere. Particularist rationality drives moral deliberation in another direction. It demands not to look away for considerations from elsewhere, but to look harder at the details of the case before one, and to discover what moral shape the facts are taking (see also McDowell 1981, 1985). According to my moderately particularist position, the shape of a case is not detennined by general considerations, though rules and principles may function as an important reminder of some (perhaps invariably) relevant aspects of the case. Some things will only be perceived and 'get shape', however, if one develops an intricate and holistic view on the unique character of this case. Allowing moral relevance of facts to play an independent role in wide reflective equilibrium thinking means that one takes seriously a (moderately) particularist model of rationality which takes the fonn of some kind of intuitionism. Again, this position does not claim that pointing out what is morally relevant is completely unconstrained by general considerations. It does claim, however, that thinking in tenns of general considerations is not all there is. Judging on the moral character of an act, situation or policy and providing answers to moral questions demand more than 'going backward and forward' between various generalizable considerations. It demands also a kind of attentiveness and apprehension of what is going on in the particular situation. This amounts to some kind of intuitionism because the appreciation of 'what is morally going on' will involve some sort of noninferentiality or directness of judgement. Some will ask what is 'rational' about this kind of intuitionism (see, for instance, Reader 1997), but this question may be the result of a confusion

Morally Relevant Facts

53

familiar in discussions about intuitionism. Noninferentiality or directness of judgement is often confused with immediacy ofjudgement. (In Van Willigenburg and Heeger (1991) we are confused ourselves.) Immediacy of judgement precludes reflection and considerate inquiry, and seems to conflict with basic requirements of rationality. As Robert Audi has argued, however, there is a sense in which a moral intuition can be a conclusion formed through rational inquiry, though not a conclusion that is an inference from one or more evidential premises. Non-inferential reflection on a case has the form of thinking over what is going on and coming to respond to an overall pattern: 'It is more like a response to viewing a painting or seeing an expressive face than to propositionally represented information. (...) One has not added up evidences and formulated their implications; one has obtained a view of the whole and characterized it overall.'(Audi 1996, 112-113)

5

Intuitionist rationality and wide reflective equilibrium

Allowing for sui generis moral relevance of facts which playa role as 'particularist' reasons in wide reflective equilibrium thinking, amounts to an idea of intuitionist rationality which demands a kind of comprehensive but noninferential understanding of situations. Noninferentiality means that the intuitive convictions formed are not dependent on general considerations and theories. This does not mean, however, that intuitionist rationality presupposes that one intuitively - as in a glimpse - grasps self-evident truths. Firstly, moral intuition is 'more like a belief based on careful observation than like an impression fonned from a glimpse.' (Audi 1996, 111). Noninferential observation is not identical to immediate grasping and may, therefore, involve careful inquiry and reflection. Especially when a case or situation is complicated, an intuition may not emerge until reflection has proceeded for some time. Secondly, the results of noninferential observation - moral intuitions are not at all infallible, self-evident truths. They are defeasible convictions which may be criticized in the course of further wide reflective equilibrium thinking by invoking principles, rules and theories. Also, a finn, intuitive 'particularist' conviction may find further inferential support by bringing it into reflective interplay with generalizable considerations. That an observation is intuitively 'evident' does not preclude its being or becoming 'evident' in other ways. Inferential justification may thus support intuitive justification. This means that intuitionist rationality (based on a particularist epistemology) may very well find its place in the broader context of wide reflective equilibrium thinking. Noninferentially pointing out the moral relevance of facts may go hand in hand with inferential reasoning involving generalizable considerations. This means

54

Theo van Willigenburg

that in the model of wide equilibrium thinking there is room for an independent epistemic role and function of recognizing facts as morally relevant.

Considered Judgements: Meaning, Community and Tradition Anton Vedder

1

Introduction

The method of reflective equilibrium (RE) - and, subsequently, the method of wide reflective equilibrium (WRE) - has been advocated as a device for theory acceptance in ethics, the justification of moral beliefs and testing the adequacy of conceptions of morality and of moral conceptions, respectively. 1 In this article, I shall primarily address the application of (W)RE to the question of the adequacy of moral conceptions. Nevertheless, the views which I shall put forward also have important consequences for the application of (W)RE to matters of moral justification and the testing of moral theories and conceptions of morality. My main interest in all this will be with the role played by considered judgements or intuitions, as some authors prefer to call them, in (W)RE when applied to questions of conceptualisation. John Rawls (1971, 47-48) takes considered judgements to be judgements 'rendered under conditions favorable to the exercise of the sense ofjustice, and therefore in circumstances where the more common excuses and explanations for making a mistake do not obtain [...]', while the person making the judgement is presumed 'to have the a~ility, the opportunity, and the desire to reach a correct decision [...]'. These are often quoted - and criticised - requirements. The controversy mainly relates to the vagueness of these requirements. Rawls nowhere explains what circumstances exactly give rise to excuses and explanations for mistakes, nor does he state what it is to have a desire to reach a correct decision. However, the formulas, because of their moralistic ring, are highly suggestive. According to the requirement to exclude situations in wh~ch excuses and explanations for mistakes obtain, the person must probably be unaffected by hallucinations, sub- and superliminal manipulation, severe suffering, time limitations, sensory handicaps, mental illness, distracting disturbances and stupidity in general. According to the requirement of the pursuit of a correct decision, the person must also have the will or the intention to become or remain unaffected by such misfortunes. With respect to the latter condition, it is difficult to understand what else Rawls may have meant by

See, for instance, respectively: Daniels (1979a); Heeger (1992b); Nielsen (1982a) and Swanton (1991). 55 ~

van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 55-72.

© 1998 Kluwer Academic Publishers. Printed in the Netherlands.

56

Anton Vedder

desiring to reach the correct decision, where the content of the correct decision cannot already be known to the person desiring to reach it. Apparently, the simple absence of the distortive conditions is not enough. In a rather Kantian vein, Rawls wants us to believe that persons should want them to be absent, as well: they should long for truth. However this may be, let us, as so many do, assume that according to these requirements considered judgements are not arbitrary, ill-considered or made under any kind of duress. 2 And let us, with others such as Aronovitch (1996), assume that these judgements are a kind of pre-theoretical intuitions which may be about particular cases or more general ideas. Then, a question comes to mind which has received far less attention, Le., whose considered judgement are we talking about? This question, of course, should be placed in the broader context of the question of which arrangement of principles, background theories and considered judgements exactly is being described when we say that a reflective equilibrium is reached. Are we primarily talking of (W)RE as a psychological category, i.e., as something which can exist in one specific person's mind or in the minds of a group of specific persons? Or are we talking predominantly of (W)RE as an epistem,ological category, i.e., as something which is rather an ideal dimension of an arrangement of propositions? Although, to my mind, no one has ever clearly and explicitly defended the psychological perspective, there is nevertheless a certain tendency among important proponents of (W)RE to think of them as psychological categories. It is astonishing to find how little effort has been made even by Rawls and Daniels to avoid an empirically or factually fashioned way of describing (W)RE. They present to us persons who reason under favourable circumstances, who have an urge for correctness and who' go back and forth, pruning and adjusting' principles, reasons, conceptions, intuitions and theories. Nowhere are we admonished to think of them and their activities as just hypothetical arrangements or, rather, as normative proposals. It should come as no surprise, then, that, for instance, Kai Nielsen and others have advocated (W)RE as a method of reaching consensus, which is a matter of factual psychological convergence rather than a matter of justification or even truth. 3 The problem with this matter-of-fact way of talking about (W)RE is that it endorses well-known criticisms of (W)RE holding that these methods are not much more than ways of bringing the elements of an individual's or a group of individuals' moral thought into (a kind of) coherence, scarcely excluding

2 3

The formulation is Nielsen's (1982a). Note that Nielsen (1982a) talks of consensus as a result of applying the method ofWRE, whereas Rawls (1971, 580-581) had already pointed to some common starting points as a necessary preliminary for making the application of RE work.

Considered Judgements

57

or adjusting extremely subjective views or prejudice. The latter are not avoided by the requirements vis-a.-vis considered judgements stated earlier on. A judgement may be completely subjective and even turn out to be a prejudice by some criteria, but need not at the same time be ill-considered or arbitrary. If, as the factual way of talking about (W)RE suggests, the input qua considered judgements in a piece of (W)RE reasoning comes mainly from a specific individual or a specific group of individuals here and now, then the critical force against subjectivism and prejudice must come from the elements which may be expected to be less affected by strictly personal preferences, e.g. principles and background theories. But, then again: what exactly permits us to cherish the hope that an individual's or a group's principles and theories are unaffected, and, in the famous process of adjusting and pruning required by the methods of (W)RE, will remain unaffected by subjectivism and moral bigotry? For, as is so often stressed by proponents of (W)RE, the emendation process of (W)RE works back and forth, not necessarily only changing and adapting the considered judgements. Emphasising the hic et nunc character of (W)RE certainly will not do to enhance confidence in this respect. In an attempt to amend the inherent liability to subjectivism, Nielsen (1982b) suggested to let judgements actually agreed upon within a community take the place of considered judgements. However, Christine Swanton pointed out that regarding questions of conceptualisation, where so-called contested concepts of morality are concerned, not much help is to be expected from judgements on which individuals explicitly agree: controversy abides even over paradigm cases. Instead, Swanton (1991) suggested to use, what Aristotle called 'endoxa' .4 Endoxa can be characterised as the judgements and conceptions of the many and the wise, expressed in pieces of common knowledge and statements by writers, scientists, theorists, etc., not necessarily only contemporary but historical as well. These are - I think - about the same pieces of common knowledge and statements which Nielsen (1993, 317) called 'considered judgements given in the traditions which are part of our culture'. Neither of the authors seems to think that these judgements introduce a kind of objective nonnative basis to (W)RE that would nln counter to the alleged anti-foundationalist character of the method. This contention is true when the idea of an objective normative basis is interpreted very narrowly. Including endoxa or the conceptions and statements from a tradition as considered judgements in (W)RE will only be likely to reduce the possibilities of subjectivism, because some contingent limitations of actual individuals or groups of individuals are transcended. However, if (W)RE were related to a view or 4

See also: Aristotle, Nicomachean Ethics VII, 1, 5, 1145bl ff.; Nussbaum (1986, 240-263).

58

Anton Vedder

theory of moral language and morality, which I shall provisionally call a functional-contextual approach, then (W)RE would be vested with an additional justificatory force. This would enhance its possibilities to be defended against allegations concerning its subjectivist taint. Let me elaborate this point in the next sections.

2

The meaning of complex notions

There is a way of looking at morality and language which suggests that we should assign a certain authority to shared experience and knowledge in communities and traditions. It is a view in which the rigid distinction between the moral and the non-moral, which is so typical of modem philosophy, is approached with scepticism, and in which moral and linguistic intuitions seem to coincide or, at least, to intertwine. In order to explain this, I have to elaborate at some length on the meaning of complex notions and some claims of the later Wittgenstein and Julius Kovesi. How are we able to understand complex notions like 'individual freedom'? Notions like this are concepts which have both a normative and a descriptive meaning. Mostly in combinations with expressions like 'respect for', 'restrictions on', etc., they are first of all used in a normative way. Thus, they can be used to advocate or recommend a conceivable state of affairs as well as to make clear that an actual or conceivable state of affairs does or does not conform to this recommended conceivable state of affairs. In both ways the notion also has a descriptive meaning in that it is used to describe an actual state of affairs or a conceivable state of affairs. Now, how are we able to understand words like this? How can we know their specific content? More specifically: what exactly is needed to know how a notion like this can be applied correctly to particular states of affairs and not to others? In replying to this question, one may feel tempted to define the meaning of a word on the basis of its descriptive meaning in terms of the resemblances or similarities of the phenomena to which the notion may be applied. However, in the case of a complex notion like freedom - but the same holds true for more down-to-earth notions like 'table', etc. - the relevant resemblances and similarities are not easily found, nor are they easily stated. This is so because these notions refer to supervenient properties, properties which for their occurrence in actual situations ultimately depend on the occurrence of different phenomena in compositions or compounds which can vary from situation to situation and from context to context. Because of this, definitions of such notions in terms of similarities tend to be too general. A striking example of this

Considered Judgements

59

deficiency is the definition of individual freedom, proposed by MacCallum (1967), as a relationship between three elements, viz., (a) a person (b) the absence of restrictive conditions and (c) actions, conditions of character or states of affairs. The problem of this definition is that it does not specify the kind or kinds of relationship which should be present for freedom to occur, where this is exactly what we want to know. In this respect the definition is too general. Nevertheless, even this definition is, at the same time, still too specific. It overlooks the possibility, pointed out by Isaiah Berlin, of freedom occurring in the binary relationship of a person and the absence of restrictive conditions, without it being the case that this person has the intention of performing any action or bringing about a condition of character or state of affairs. 5 One can adjust the deficiencies of definitions like these in part by invoking Wittgenstein's idea of 'Familienahnlichkeiten' or 'family resemblances'. According to Wittgenstein (1967, 32, remarks 66-67), when we look for the unifying element of the phenomena to which a notion can be applied correctly, we often will not find one particular similarity or one specified group of similarities. Instead, we will find an unspecified group of resembling characteristics and relationships. Wittgenstein compares this kind of similarities and typically intransitive relationships with the kind of resemblances in appearance and character existing between the members of a family. Wittgenstein's approach clarifies our understanding of the ways in which we use our notions by no longer explaining conceptual unity on the basis of one unique characteristic or well-defined group of characteristics shared by different phenomena, but on the basis of a set of resemblances and relationships not necessarily shared by all phenomena. In this way, he extends the possibilities of correctly applying the notions and shows the enormous complexity of the ways in which completely different and, at first sight, even opposing phenomena may be brought together under one term without rendering it ambiguous. Nevertheless, Wittgenstein (1967, 34-36, remarks 71,77) seems to have taken a pessimistic attitude toward the possibilities of expliciting the conceptual unity of complex notions in such a way that it would become a little clearer which resemblances and relationships would allow us to correctly apply a notion to a phenomenon. For him, the notions remain concepts with blurred edges. And only by giving illustrative examples of the ways in which we may apply them can we hope to clarify the rules for their correct application. 6 5

6

Berlin's critique, cited in Swanton (1992, viii-ix). For a thorough critical, though constructive, review of philosophical contributions on the subject of freedom during the last decades, see Swanton (1992). Wittgenstein's pessimism in a sense is shared by Austin (1979, 180) in his views on so-called 'defeasibility notions': complex notions, the meaning of which can only be found by investigating occasions and situations in which we think that

60

Anton Vedder

Perhaps Wittgenstein's pessimism can be partially explained by a latent tendency to think of family-resemblances as the constituents of the unity of the various applications of a notion. The possibility that things are in fact the other way around - family resemblances themselves being a result of this unity - seems to have escaped his attention. However, when we seriously ask why the occurrence of certain phenomena, and not others, in certain arrangements, and not in others, in certain contexts, and not in others, must be present in order to apply a term correctly, then this question can be answered more satisfactorily than by giving examples of the ways in which the notion is applied and by pointing to certain similarities and relationships between the phenomena to which it is applied. For, although giving such examples may give some vague idea of the general rules for correct application, we will still remain in doubt as to whether the resemblances and relationships hinted at are essential or typical for this notion, or not. The similarities or relationships may be coincident, having little to do with conditions allowing us to use the notion. Evidently, then, there is something else which predominantly determines the conceptual unity of notions and restricts the kinds of resemblances and relationships allowing us to gather different phenomena under one notion. Wittgenstein (1967, 10, 20, remarks 20, 43) hints at this unifying factor by identifying the meaning of words with the ways in which they are used in language. Now, identifying the meaning of words with their use seems not much of a help for illuminating the rules for the application of notions. What we seek to find are directives for the correct use of words. For Wittgenstein (1967, 8, 11, remarks 19,23; 88, remark 241), however, language is not an amalgam of personal and arbitrary instruments for communication. He characterises language as a form of life. The meanings of words are not based on personal convictions and choices, nor are they the result of agreements between persons or groups. Wittgenstein implicitly suggests that language, including the nleaning and use of words, is based on conventions. This suggestion should not be taken as necessarily implying conservatism. Wittgensteinnowhere gives the impression of considering language as a congealed, solid set of words and meanings in which the niles for the correct use of notions lay statically as sediments of the times before ours, nor does his suggestion necessarily imply that language and meaning are the results of an arbitrary historical process. Rather, he seems to maintain a certain aloofness to questions concerning the reasons for our linguistic

the notion certainly cannot be applied.

Considered Judgements

61

conventions' being as they are, because their sense or nonsense must in tum be judged and formulated within that same language, which is our form of life.? Wittgenstein, because of his self-imposed reserve, did not, at least not explicitly, go much further in clarifying meaning in terms of use. I think that this, however, is exactly the point where Kovesi's theory of meaning should be invoked as an implementation. Kovesi (1967) is of the opinion that the meaning or use of notions (or, for that matter, the conventions related to this use) in tum depends on the functionality of the phenomena, or constellations of phenomena, to which these notions refer. 8 Using the classical Aristotelic terminology, Kovesi (1967,3,15,61-62) claims that with respect to the meaning of complex notions, a distinction should be made between a material and a formal element. The material element is the whole range of the particular identifiable properties, respects and contexts of the phenomena to which the notion may point. The formal element is that which determines which properties (in which respects, in which contexts) may belong to the material element of the notion. The formal element in fact reflects the point of the notion. It states the reason why we can apply the notion correctly to phenomena, exhibiting certain properties, in certain contexts, in certain respects. This reason lies in the function of the phenomena to which the notion can be appplied. We can find it by meticulously uncovering the purposes for which we need, are interested in or value those phenomena. 9 When proposing his distinction between the formal and the material element of the meaning of a notion, Kovesi was mainly interested in bringing about an adjustment in the discussion among philosophers about the difference between moral and non-moral notions. This difference, according to him, is often confused with the difference between descriptive and evaluative terms. In Kovesi's view descriptive terms are as evaluative as moral ones are, and, conversely, moral notions are as descriptive as others are. The difference between moral and non-moral notions, however, lies in the different kinds of functions which make up the formal element of these different kinds of notions. According

7

8 9

Although Wittgenstein (1967, 174) thinks that certain phenomena, such as hoping and promising, are only possible through language, he nevertheless supposes that there is a reality outside language (1967, 230) and that non-linguistic, or rather pre-linguistic, thinking is possible (1967, 106-110, remarks 327-340); talking about the latter, however, would make very little sense, according to him. For a critical discussion of Kovesi' s main ideas as well as a reply to these criticisms, see Graham (1975) and Shiner and Bickenbach (1976). The pair of material element and formal element should not simply be equated with the pair of extension and intension of a notion, because the formal element always includes a description of the functionality of the phenomena to which the notion refers, whereas the intension of a term does not necessarily do so.

62

Anton Vedder

to Kovesi (1967, 12-13, 63), the different kinds of functions are symptomatic of a different point of view. To this claim, I will return.

3

Adjusting and refining the Kovesian view

Although Kovesi's explanation for the conceptual unity of complex notions is more plausible than explanations in terms of similarities or family resemblances taken on their own, his views should be adjusted and refined in some important respects. First of all, it should be noted that, of course, resemblances and family resemblances as such do playa role in the development of our knowledge of the meaning of words. Doubtlessly, in practice, when we have to explain notions or when we have to find out whether a notion can be applied, we will look for resemblances and relationships without much further ado. These resemblances and relationships, however, can also account for many errors. And it is exactly when the need for correction is felt or when we are confronted with hitherto unknown situations that we must invoke reflection on the formal element or point of the notion. Again, this process of discovery of the point of a notion in the purpose of the phenomena to which the notion refers, does not start from scratch. When Kovesi proposes that, in order to find the point of a notion, we look for the purpose of the phenomena, he seems to contradict himself. One might reasonably ask: what kind of phenomena should we be looking for? Is not that exactly one of the questions we would like to have answered when looking for the point of the notion? In other words, it seems as though the Kovesian explanation begs the question. In order to know what the notion means we should know what it means. I do not at all think that the Kovesian explanation in the end is selfdefeating. In a way, yes, in order to know what a certain notion means, we already need to have at least some idea of it. In looking for meaning, we do not come empty-handed like complete strangers speaking an entirely different language. When we look for the point of a notion, we, as it were, plunge into the midst of things, on the basis of some part of meaning of the term, which we already know. From here we go on, on the basis of some resemblances, relationships and partially uncovered functionalities, until we have adjusted and improved, for the time being satisfactorily, our knowledge. This brings me to a further respect in which the Kovesian explanation could be refined. Because of the distinction between the function of words and the function of phenomena to which these words point, the explanation - as well as the ways in which Kovesi himself applies it to particular notions - not only suggests that there is a gap between language and reality, but also that there

Considered Judgements

63

is somehow an easy way of bridging this gap. In this, he seems to be unconscious of the fact that this assignment of functions to phenomena again is something of language or, for that matter, of the knowledge and beliefs that are possible through language. The formal element of a meaning must not only be stated in language or its notions, but also be understood as part of a larger network of such notions. The question whether it is this net or framework of knowledge and beliefs or rather something that transcends it, which ultimately constitutes the function of notions, should only be answered with great reserve within that language. Here, I will follow Wittgenstein's example and not try to answer it. A third way in which the Kovesian model might be improved is to allow notions to have more than one point. With this I do not only suggest a way of explaining equivocality or ambiguity of notions. What I have in mind, specifically, is something which can be explained best by the example of the notion of individual freedom.

4

Freedom, for example

Christine Swanton (1992) - to my mind in part correctly - stipulates the formal element of individual freedom as the optimal functioning of the human individual's potential in practical activity, that is to say: the forming of desires and wants, deliberation, the formation of practical judgement and intentions, and the execution of intentions through actions. Swanton further adds in a Kovesian vein - and to my mind again correctly - that this good functioning should be thought of as a multidimensional, formal property, the material conditions of which can vary from context to context. She suggests, heavily leaning on Austin's concept of defeasibility notions, that the idea of the optimal functioning of this individual practical potential somehow has evolved from experiencing all kinds of flaws and limitations, both from within and from without, relating to (parts of) the process of agency, as well as to the (availability, eligibility and significance of the) varying objects of (parts ot) that process. Now, I think that all this may perfectly well be the case. Nevertheless, some features of the notion of freedom can be better explained if we allow there to be at least two other points in the formal element of the notion of individual freedom, namely: the purposes of establishing authenticity of preferences and moral responsibility. The features of the notion of freedom, which they explain best, are concerned with the significance that must be assigned to specific restrictions on freedom. Swanton somehow seems to confuse the significance of restrictions on freedom with the relevance of the significance of options to freedom as such. In Swanton's view, the significance of restrictions on freedom is ultimately a

64

Anton Vedder

completely subjective matter, depending on the affected individual's preferences and attitudes concerning his own freedom. 10 This need not be true, even if the optimal functioning of the individual's practical potential were the whole story of the point of freedom. For even then we might say that the affected individual's preferences and attitudes are not the only or necessary measure for the significance of restrictions. One might, for instance, also try to use the idea of optimal functioning of an individual's practical potential as such as a measure, and for instance state the significance of restrictions in tenns of the consequences they have for this optimal functioning as a whole. However that may be, we mostly do not assess the significance of restrictions on freedom in terms of the involved individual's subjective preferences concerning freedom, nor do we invoke straightaway the construction of optimal functioning of the individual's practical potential. What we do most often, when we say something concerning the significance of restrictions, is in fact even more complicated. Let me explain this by first saying something about the situations or contexts in which we are mostly or typically interested in the significance of restrictions on freedom. These are mainly situations of two types. The first is the kind of situation in which we want to know whether an individual's preferences and desires are truly his. We are concerned about the authenticity of his preferences and attitudes because we are afraid that this individual by expressing his preferences or by acting upon them might harm himself, or bring about something which in some respect is suboptimal for himself in comparison with what he might bring about if he were to do nothing or something else. The second type is one in which we want to know whether we can hold an individual responsible for what he has done. We want to know whether we can hold him responsible because we want to be just and fair in our distribution of reward and punishment. In both types of situations we will ask whether the individuals in question were sufficiently free as to their preferences, attitudes and actions. Or, to put it differently, we want to know whether the restrictions on freedom possibly present were so significant that they allow us to think of an individual's preferences as not truly his, or to think of an individual as not actually responsible for what he did. The significance of restrictions, which is the object of our interest then, is a significance measured against a background of conditions of normality, expectancy and propriety. I I We apply such standards of normality specifically

10

11

This is a simplification of her view. From the complicated ways in which she expresses herself, it is clear that Swanton (1992, 162-190) is wrestling with the problem, and allows of other solutions. Cf. Vedder (1995,57,63,67-69,81); Feinberg (1986, 115, 118-122) and Benn (1988, 132, 136).

Considered Judgements

65

to situations in which we suspect there to be restrictions on the range of available options, on the eligibility of those available options and on the individual's perception of the significance of those options. (And here once again, I must warn that significance of options is not the same as significance of restrictions on freedom.) What we try to establish is not whether there were any restrictions as to these aspects or dimensions of the freedom of the individual involved tout court. We try to assess whether there were restrictions surpassing the restrictions . that we would normally expect there to be in the total setting of the situation and in the individual involved. It is difficult to state the details of these standards, but if they were to be spelled out, they would certainly include norms concerning nonnal physical, biological and meteorological circumstances, as well as nonnal psychological and social behaviour of persons involved (including the agent's own) and even the normal functioning of social institutions. On the basis of these we decide whether restrictions are relevant and significant enough, in order to think of a person's preferences as genuinely his own, or to hold him responsible or not responsible. And, even more interestingly, as long as the restrictions present do not in fact deviate from standards of normality, we have a tendency not to think or talk of them in terms of restrictions on freedom at all, notwithstanding the possibly imperfect functioning of the individual's practical potential or the individual's attitudes towards his limitations. Mostly, therefore, we evaluate and describe situations in terms of freedom when we are motivated by the purposes of establishing authenticity and responsibility. Only occasionally, and perhaps most of all in the context of philosophical exercises, are we motivated by further purposes (which exercises, however, need not necessarily be senseless).

5

A functional-contextual approach

I have shown how, in the case of a notion such as individual freedom, it may be enlightening to accept the possibility of having more than one point to a notion. This feature of having more than one point does not render the notion ambiguous. It only explains how the notion can be used in ways which do not exclude each other but largely overlap. This overlap may, in its tum, be explained by connections between the points in question, such as good functioning of the individual's practical potential, authenticity of preferences and attitudes and moral responsibility, in the case of individual freedom. To this phenomenon, however, I will return later on. To take up the thread of adjusting the Kovesian model of explanation again: I think that investigating the linguistic descriptions of the formal elements of complex notions in general can endorse the plausibility of the Kovesian model.

66

Anton Vedder

Particularly, investigating the ways in which one may try to complete these descriptions contributes to our understanding of the organisation of knowledge, and moral knowledge in particular. As Kovesi himself already suggests implicitly, descriptions of the formal element will always be only partial descriptions. 12 Nevertheless, they can be complemented - be it not perfectly completed. The ways in which this completion will proceed reveals some remarkable features of complex notions. For the sake of convenience, I will focus again on notions commonly used in the field of morality. In describing the formal element of a complex notion, one has to use complex notions again which, in tum, for the sake of completion stand in need of clarification of their formal element, for which one must again use complex notions, etc., etc. One may call these different stages at which further formal elements must be explained in order to explain the formal element of the notion with which one began, the different levels of description. Perusing all conceivable levels of description of a formal element of a notion would be practically impossible for a human being. Nevertheless, one can go on for quite a number of levels. Even in an incomplete description, at some level it will be necessary to invoke notions which do not, or at least not clearly, belong to the same category as the notion with which one started. When, for instance, we try to complete the description of the formal element of the notion of individual freedom (a value word, that is) by invoking the notions of 'authenticity of preferences', 'moral responsibility', 'good functioning of the individual's practical potential', etc. (Vedder 1995, 70-103) and, at some other level, notions like action, will, person, individuality, etc., then we go in fact beyond the range of value words. Notions like moral responsibility, action, will, person, etc. cannot be considered to be values, at least not or not exclusively in the same way as we might consider freedom to be a value. These are words which refer to the framework of preconditions for morality as a whole, and in that sense belong to the constituents of morality as such. Furthermore, most of them as to their usage are not exclusively bound to the domain of the moral as it is traditionally conceived of. Conversely, it may very well be the case that in explaining notions which are not considered to be strictly moral notions, sooner or later one will have to seek recourse with what traditionally are called strictly moral notions or value notions. The fact that moral and non-moral notions playa part in each other's formal element suggests that the distinction between these categories is only a superficial one. If one descends to deeper levels of description of the formal element of these notions, the difference evaporates. This phenomenon, to my 12

Kovesi (1967,14-15,17).

Considered Judgements

67

mind, weakens Kovesi's claim that moral notions and others differ qua point of view. At least things seem to be more complicated than Kovesi suggests. A second remarkable feature is the phenomenon of apparent self-referral. Again already in descriptions with a limited number of levels it will soon become evident that a description of the formal element of a notion in a way presupposes itself. This dependency of the meaning of a notion on other notions and itself can be compared to the position of a thread in a spider's web. The thread can only stay in its position through its connection with other threads, which in tum . can only stay in their position and hold the first thread in its position through their connection with other threads, among which, of course, is the first thread. In this way, the thread depends upon others as well as itself in order to hold its position. When we, for instance, describe the formal element of the notion of individual freedom in terms of 'authenticity of preferences' , 'moral responsibility' and 'flourishing of the individual's practical potential', 'person', 'individuality' and 'action', then the description of the formal element of these notions at some stage of the description, and perhaps even more often and at different levels, requires the description of the formal element of individual freedom. Put differently and by way of example: in order to know what freedom really is, one must, inter alia, know what moral responsibility is; in order to know what moral responsibility really is, however, one must in a sense know what freedom is. This does not mean that the description of the formal element necessarily ends up in a vicious circle. 13 In a set of levels of description needed for clarification of the point of one notion, a description of the formal element of a first notion enters at another, different level of the description of the formal element of a second term than the level at which the description of the formal element of that second term enters into the description of the formal element of the first. These differences qua level of description of the formal element slightly modify the perceived meaning of the notions. In this way, each level of description reveals only a certain part of the formal element of the notion. Apparently, the different notions used at one level of description by their combination nuance the perspective from which the point of the term is presented. This implies that in the whole process of the description of the point of a term on different levels there are, in away, slight shifts in the meaning of 13

One may think that we are able to avoid circularity by maintaining that the apparent self-referral is in fact only referral to (part of) the material element of the notions. This, to my mind, is not true because for us, in order to understand the material element, we need a formal element which organises the material element of the notion. This condition also applies during the process of description of the formal element.

68

Anton Vedder

the notion. We can, for instance, sensibly describe the formal element of individual freedom partially in terms of responsibility, and responsibility, in tum, partially in tenns of freedom, because within the process of description, at different levels of description, the meaning of the notion is perceived slightly differently. However, understanding the complete meaning of the notion, Le., the complete description of the formal element thereof, would include simultaneous perception of all conceivable levels.

6

Coherence in a web of meaning

To sum up, the meaning of a moral notion can only be understood by grasping its point or points. The point of the notion in tum can only be understood through its connections with other notions, both what we traditionally might call moral notions and non-moral notions. For the reasons stated earlier on, one cannot conclude from this that it is exactly and exclusively this nestling in a whole web of notions, knowledge, beliefs and motivations that gives the notion its point and sense. What one may conclude is that the whole web of notions and knowledge, beliefs and motivations behind them constitutes a coherent system with that notion. Viewed this way, a particular notion may be considered to be one of many focal points of one larger framework comparable to a worldview. The coherence and specific connections within this framework limit the possible applications of the notion. At the same time, the framework seems to hide an unexpected reservoir of possibilities to extend the actual knowledge of meanings of terms with maintenance of coherence. I will illustrate all this once again with the notion of individual freedom. In doing so, I will first take up again the question of plurality of points and ambiguity which I left unanswered when talking about the three points of freedom earlier on. Next I will say something about the ways in which these points are nestled into something that resembles a worldview. As I suggested earlier on, I think that the meaning of individual freedom is best explained by reference to three points, which together make up at least a partial description of the fonnal element of the notion, namely: good or optimal functioning of the individual's practical potential, authenticity of preferences and moral responsibility. Envisaging these 'points' explains how and why we apply the notion of freedom. Now, I think that the good functioning of the individual's freedom is at stake in every situation in which we may apply the notion. But often it is at stake in a somewhat meagre or thin sense. What I mean is that we need this point to describe and evaluate situations in a rather uninteresting way in tenns of freedom. In order to make more sense of the notion, literally, as to its more significant usages, one or both of the others must come

Considered Judgements

69

to the fore. The latter two do not come into play necessarily or simultaneously. Whether they do, depends on whether the perceived circumstances occasion the need to establish authenticity of preferences or moral responsibility. When there is more than one point to the application of the notion to a situation, then these points do not, to my mind, constitute entirely different meanings of freedom. In cases where they underlie the application of the notion on one and the same occasion, they seem to show us the same phenomena from only slightly different angles. The ways in which they are connected to each other, to my mind, prevents them from constituting entirely different meanings. In the case of the formal element of freedom, ambiguity is avoided because good functioning itself is part of the description of the formal elements of moral responsibility and authenticity of preferences. The remaining slightly different angles from which we view the same phenomena under the one heading of freedom in such cases can be explained by the different other connections that the three points, when taken separately, have to other notions. With authenticity and responsibility, as I have shown, there is another way of interpreting the significance of restrictions on freedom than with the good functioning of practical potential. This difference, however, can in the end be explained by the connection which the former have to other notions such as respect for an individual's conception of the good and fair distribution of appraisal and blame, respectively. Although these different connections account for the different ways in which we may use the notion of freedom, these differences as such must not be overemphasised. There is unity in the different uses of freedom, thanks to the fact that the three points of the notion are interconnected through one of them. But there are more and less oblique ways in which these points are interconnected, directly or indirectly through mutual relationships with other notions. For instance, I think that we may safely say that all three of them again are tangled up with a certain conception of the human individual and individuality. This conception of the individual and individuality, in tum, has further connections to, or to put it more accurately: has connections within, a larger outlook. In this outlook, the concern for the individual through appreciation of the individual's interpretation of the good and through fair judgement on individual merits and demerits is predominant. It receives expression in many diverging ways ranging from subjectivist views of values to certain principles . of privacy, honesty and justice. But the concern for the individual and individuality is not limited to such traditionally moral items. It even pervades the ways in which we preferably explain and describe the phenomena of the world by making us cling to methodic individualism, and causing unease at the thought of attributing causal or even moral responsibilities to collectivities.

70

Anton Vedder

I think that, in general, investigating and making explicit the direct and indirect connections between our notions such as those at which I have just hinted can contribute to our understanding of the limits of our outlook (and especially our moral outlook). It may, however, also help us in refining that outlook by producing possibilities of articulating views and claims that are only latently present in that outlook and can only be found by meticulously analysing the internal connections of the notions and the knowledge, beliefs and motivations that go with them. Such an undertaking, however, is not an easy one. Apart from the natural limitations of our knowledge capacities, there are some special difficulties in uncovering the whole network. These complicating factors lie in habit foonation and standardisation through practices and theory. Kovesi (1967,13-14) already seems to have seen that in social practices such as the law and economic activities, the operations of institutions, the practising of professions and in theoretical work, consciously and unconsciously, the functionalities of phenomena are becoming articulated more and more in specific forms. The articulation of meanings of certain notions are consequently standardised. These practical modifications and standardisations influence, of course, our possibilities of uncovering meaning because they restrict - one is tempted to say: at a pre-linguistic level- the ways in which we can conceive of and fonnulate functionalities of phenomena. Only investigating and trying to make explicit the factors of habit and standardisation, however, offer the opportunities of unveiling hidden parts of meaning and of broadening our knowledge. ]4

7

The functional-contextual approach and (W)RE

In this last section, I shall return to the point of departure for this article. I shall explain how (W)RE insofar as it is regarded as a method for justification, may be considered to receive justificatory force from the fact that it explicates a worldview-like web of meaning. Finally, I will touch upon some diverse questions concerning our inability to spell out and understand this framework completely, the hidden treasure of meaning inside it and the ways in which we should conceive of moral autonomy in the light of the public character of this worldview. (W)RE may be viewed as an instrument for unravelling a web of meaning existing in the tradition of a community. Listing considered judgements about certain controversial notions or phenomena stated by 'the many and the wise' 14

Cf. Vedder (1997) for an example of the ways in which legal and ethical theory have narrowed the meaning of privacy.

Considered Judgements

71

from a certain cultural tradition or tradition of thought, and pruning and adjusting these by confronting them with each other and with theories and theoretical conceptions relating to those notions or phenomena - all this can be a fruitful way of exploring and formulating the connections between notions within a public framework of knowledge. Such explorations may be fruitful for various reasons. Not only do they enhance our understanding of ourselves as knowing creatures and moral beings. As I showed in section 6, they may also be profitable for our capacities to articulate and to solve hitherto unknown moral problems. The coherence and specific connections within our framework of knowledge limit the possible applications of the notion. At the same time, however, the framework hides a whole range of possibilities to extend the actual knowledge of meanings of tenns while coherence is maintained. This range of possibilities may be systematically uncovered by using the method of (W)RE. More importantly, however, where (W)RE is used as a method of moral justification or of acceptance of theories or conceptualisations, the ideas about a web of meaning resembling a worldview can provide (W)RE with additional justificatory or persuasive force. As I stated in section 1, including endoxa or the conceptions and statements from a tradition as considered judgements in (W)RE will at least reduce the possibilities of moral subjectivism because the spatiotemporallimitations of the actual individual or groups of individuals are transcended. Now, if these endoxa are understood as expressions of parts of a web of meaning existing in the tradition of a community, then including them as considered judgements in (W)RE will also show and ensure that a conception or practical judgement resulting from a piece of (W)RE reasoning fits with the past and present linguistic intuitions in a community. A set of different judgements and statements taken from a larger community and tradition is more likely to represent these intuitions than judgements of one individual or some group of contemporaneous individuals. Including them because they are considered to be expressions of parts of this larger net of knowledge and intuitions will further reduce the possibilities of linguistic subjectivism. In addition to this, however, if a certain moral authority were assigned to the shared experience and knowledge in a community and its traditions, stored in such a web of meaning, then including the conceptions and statements from a tradition would provide us with a morally nonnative basis. Now, it is exactly this functional-contextual conception of morality and moral language which suggests that we should indeed assign such an authority to shared experience and knowledge in communities and traditions. Of course, it is somewhat misleading to say that this view suggests to assign authority. In fact, the view in tenns of a web claims that this shared experience and knowledge stored in our language cannot but exercise it. It does claim this dominion amongst others by denying the possibility of rigidly distinguishing between the moral and the i

72

Anton Vedder

non-moral (identifying moral and linguistic intuitions) and by its aloofness vis-avis the extra-lingual world. The possibility of finding and explicating sensibly other sources of such an authority is thereby excluded. The authority is there already, so to speak; it is only to be acknowledged. Coming to the end of my article, I would like to say just one more thing on the functional-contextual approach, moral controversies and moral autonomy. What I have said so far in order to characterise the functional-contextual approach of moral language and morality does not entail that accepting this approach binds one to believe that all who speak the same language or use the same notions actually are in possession of the same knowledge, beliefs and motivations. The natural restrictions on human capacities for knowledge leave room for ignorance and error in different degrees and of different kinds in different persons. This opens the possibility of controversy. The part of the meaning of a notion, known to one person, need not be completely the same as the part known to other persons. As long as there is, however, some overlap between them, while they disagree for the rest, they can have a controversy about the meaning of the notion. According to the functional-contextual approach, the possibilities for deviant use and disagreement, however, are restricted as long as the users of the notions want to remain intelligible to others. 15 In the functional-contextual approach, the same kind of restrictions apply to the possibilities of moral autonomy. In the strict sense, an individual cannot be the creator of values. The individual can get to know values partially, identify with them, and in this way make them his own. Furthermore, he can uncover parts of the meaning of a value, hitherto unknown to him and to others. However, he cannot produce values out of nothing. The connections of moral notions within a public framework of knowledge, beliefs and motivation do not permit this. Conversely, however, these connections and our capacity to explore and formulate them, for instance by applying the method of (W)RE, also offer us the opportunity of solving disagreements and improving our insight when we are confronted with moral problems. A disagreement is not just replaced by explicit agreement or psychological consensus of opinions, but by agreement in language or, to use Wittgenstein's words, by agreement in form of life.

15

The meaning of a word can be subject to controversies. However, against GaBie (1955-56) I would argue that because of the public character of our language not even moral concepts are essentially contested.

Background Theories and Religious Beliefs: Their Role and Relation in Reflective Equilibrium Ton van den Beid

1

Introduction: The questions

Robert Heeger and I were trained in moral thinking at a time when it was generally thought among experienced practitioners that (informal) logic and relevant facts were sufficient for doing the job. It has become clear since then that this picture is too simple. The idea that a moral position can be built upon the logical properties of normative, language and ascertainable facts has given way to a widely shared belief that more elements are called for. The method of reflective equilibrium (henceforth RE) reflects this development. Even in its narrow version it is already an improvement upon the simple picture. Wide RE goes further in expanding the class of relevant considerations that bear upon moral thinking. From this perspective, the questions assigned to me for this Festschrift gain significance: Can the relation bet'ween fundamental views of life (religious beliefs) and moral beliefs be accounted for in terms of the RE model? And: What is the function of background theories in wide RE? Are fundamental views of life among them? What I shall do in this contribution is, firstly, argue that beliefs of a religious nature should not be denied a role in the process leading to RE, particularly in its narrow representation. Secondly, I shall look into the nature and function of background theories in Norman Daniels's conception of wide RE. A leading question in the investigation will be whether (systems of) religious beliefs do or could playa role in it. Thirdly, I shall give some reasons why fundamental views of life, whether of a religious nature or not, should have a place in wide RE.

2

Religious beliefs in narrow RE: The case of Jeanie Deans

RE is a method ethicists employ in conducting moral enquiry. As such, it distinguishes itself from other professional methods of moral thinking, at least initially, by keeping close to the way ordinary people try to solve moral problems. They too engage in moral thinking. Their enquiry is directed at the solution of first-order practical questions, most of the time of a specific nature. Socrates' question: How should one live, is very general, quite unlike that of the man who 73

w: van der Burg and T. van WiUigenburg (eds.), Reflective Equilibrium, © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

73-88.

74

Ton van den Beld

wonders whether he should commit suicide now that an aggressive cancerous tumour has been spotted in his brain. Likewise, the question whether it is morally permissible to commit perjury is not liable to elicit much of a moral enquiry among ordinary moral thinkers (nor, for that matter, among professional ethicists). Things are different, however, when giving false testimony in court will prevent one's beloved sister from receiving the death penalty for the presumed murder of her baby. So, broadly, first-level moral enquiry is concerned with specific quandaries. Of course, more general questions such as whether animal experimentation is morally objectionable engages also ordinary people's moral thought. But this does not change the overall picture. Although RE keeps close, initially, to the way ordinary people practise moral enquiry, the former does not correspond completely with the latter. There are important differences. First of all, RE - taken in its narrow sense - is an explicit method, designed to further and facilitate first-order moral enquiry. Secondly, it is distinguished from ordinary people's moral thinking by its critical spirit. The intuitions and principles that are to be brought into equilibrium are not taken at face value; they have to be examined critically. I shall not go into the nature of the intuitions and principles involved, nor shall I be particularly concerned with the way narrow RE is reached among its several elements. What I want to point to, however, is that it is overly narrow in its general conception. As has already been indicated, RE as a method of firstorder moral enquiry is modelled according to the structure of ordinary moral thinking. Seen in this light, morally relevant religious beliefs are conspicuously absent in philosophers' accounts of RE. Their absence is remarkable because religious beliefs have always played a significant role in a great many people's moral thinking. In our Western culture, the religious beliefs concerned bear, of course, predominantly a Christian stamp. In order to bring out the religious element in a simple but intelligent person's moral enquiry, I will relate an episode from Walter Scott's The Heart of Mid-Lothian. 1 Jeanie Deans, the novel's main character, is the daughter of a plain, Calvinist farmer in eighteenth-century Scotland. She is faced with a terrible dilemma. Her sister Effie, whom she adores, has fallen in love with, and got pregnant by, a man whose identity Effie is not allowed nor willing to reveal. Effie hides her pregnancy and loses her baby while lying unconscious in childbed. When the fact of her pregnancy is brought into the open and she is not able to tell what has happened to her child, she is arrested on a murder charge. The only one who can save her is Jeanie. If Jeanie were to declare under oath that Effie had told her about the pregnancy, a death sentence could be Sir Walter Scott (1818/1992), The Heart of Mid-Lothian, London: Everyman's Library.

Background Theories and Religious Beliefs

75

avoided. The problem for Jeanie is precisely that Effie has also kept silent to her. What is she to do? She speaks with her father in guarded terms about her plight. She misunderstands him. She thinks that he sees an opening for her to commit perjury, to go against God's command prohibiting false witness. And then Scott writes: Roaming from thought to thought, she at one time imagined her father understood the ninth commandment literally, as prohibiting false witness against our neighbour, without extending the denunciation against falsehood uttered in favour of the criminal. But her clear and unsophisticated power ofdiscriminating between good and evil, instantly rejected an interpretation so limited, and so unworthy of the Author of the law. She remained in a state of the most agitating terror and uncertainty - afraid to communicate her thoughts freely to her father, lest she should draw forth an opinion with which she could not comply, - wrung with distress on her sister's account, rendered the more acute by reflecting that the means of saving her were in her power, but were such as her conscience prohibited her from using - tossed, in short, like a vessel in an open roadstead, during a storm, and, like that vessel, resting on one only sure cable and anchor, - faith in Providence, and a resolution to discharge her duty. (Scott 1818/1992, 218219)

Jeanie's resolution not to bear false witness is tested up to the final moment. But she cannot bring herself to utter the words of the false statement many present in the courtroom expect her to make. On an earlier occasion she had already declared that it is impossible for her to do what is 'unlawful for a Christian' (170); that it is not man whom she 'fears', but 'the God whose name I must call on to witness the truth of what I say'. When it is objected by Effie's lover who, being an outlaw, is hiding from the police that this God will not be ignorant of the fact that she will not bear false witness for personal profit but in order to save the life of an innocent, her reply is: 'He has given us a law (...) for the lamp of our path; if we stray from it we err against knowledge - I may not do evil that good may come out of it.' 'But you,' she adds, 'you that [knows] all this to be true, which I must take on your word - you that (...) promised [Effie] shelter and protection in her travail, why do not you step forward, and bear leal and soothfast evidence in her behalf, as you may with a clear conscience?' (173) Scott's presentation of the case of Jeanie Deans can be looked upon not only as a wonderful example of an ordinary person's moral thinking, but also as a prototype of the employment of RE as a method for solving first-order moral problems. So Jeanie's 'being tossed like a vessel in an open roadstead' may be understood as her being moved back and forth between her moral beliefs.

76

Ton van den Reid

Note that Jeanie was not solely passive to her beliefs, she put them to critical scrutiny. It was tempting for her to interpret the ninth commandment as only prohibiting false witness against, and not forbidding false witness in favour of her sister. But she rejected the interpretation as below standard, as irrational. Moreover, while knowing of the principle of beneficence, she also appeared to be familiar with the Pauline principle (Romans 3:8) of not doing evil in order that good come out of it. 2 Lying, committing perjury, remains morally objectionable even if much good may result from it, indeed even if a life might be saved. Neither should it escape our attention that Jeanie is aware of a morally .relevant fact: There is a witness available in the person of Effie's lover who could bear witness in favour of Effie without violating any serious moral rule. But above all, the case brings to light the important role religious views play for Jeanie in reaching RE. Thus she is strongly convinced of the status of moral rules as divine commands. The stringency those rules appear to have for her can be accounted for in terms of their divine origin. FurthertTIore, it is essential for a correct interpretation of Jeanie's moral position not to overlook her faith in divine providence. Jeanie apparently believes that both her life and that of her sister are in God's hand. But, more importantly, she also seems to believe that her responsibility for Effie ends where the limits of God's commands are reached. Put in another way, it is her conviction that what she can do out of love and responsibility for her sister is restricted to those actions which are in accordance with the divinely established moral rules. The effects of her refraining from a violation of a moral rule are God's responsibility. This latter belief is implied in Scott's account of the events. Jeanie's moral position would be incomprehensible, or at least liable to misunderstanding, without these religious beliefs. Thus, if we disregard her religious beliefs we could take her to be a kind of rule utilitarian. Indeed, at the time when the events of The Heart ofMid-Lothian took place (1736, to be precise), Francis Hutcheson was professor of philosophy in Glasgow. He taught not only that 'that action is best which procures the greatest happiness for the greatest number', but also that this end can be best pursued by complying with rules and laws without exception, even if 'the violation of the law would be of less evil consequence than obedience to it' (Hutcheson 1738 (4th edn.)/1969, 284). The thing to take to heart here is that we should not overlook the fact that our acting in the world is guided by our thinking. The way we see the situations in which we have to act, the way in which we see ourselves in those situations, as well as the rules and principles which directly influence our deliberations are impregnated by thought, not seldom of a religious nature (cf. Hampshire 1982, 206). For that reason it seems

2

For a modern philosophical defence of the principle, see Geach (1969, 117-129).

Background Theories and Religious Beliefs

77

wrong to abstract from religious beliefs in the structured process of deliberation leading to (narrow) RE.

3

Objections: Foundationalist structure and individualism

It might be objected that the introduction of Jeanie Deans's case is unfit to bear the burden of 'proof' for the incorporation of religious beliefs in the tnethod of RE because the case is misleading as an example of RE. Jeanie's moral thinking seems to reveal a foundationalist structure, whereas RE is coherentist. The first thing to say in reply to this objection is that it gives evidence of a confusion. The procedure of narrow RE should be regarded primarily as a first-order method of moral decision making, not of justification. Justification is an issue of second-order moral enquiry. Foundationalism and coherentism are different approaches to moral epistemology, not to decision making. Moral epistemology is not so much concerned with questions about the rightness or wrongness of particular past or contemplated future actions, but rather with questions of justification and truth, for example with the issue whether we can know our actions to be true and, if not, why.3 Of course, positions taken in epistemology may be reflected in decision making. An epistemological coherentist is likely to be in favour of RE in first-order moral enquiry, but I see no reason why, for example, casuistry or principlism could be no option for him. Pragmatic considerations are allowed to influence the choice of method. On the other hand, foundationalists will presumably be more inclined to support casuistry or principlism. Nevertheless, it seems to me that one may be a (moderate) foundationalist in epistemology and still admit that the only way to track a particular moral truth - What ought I (morally) to do now in this situation? - is to apply the method of RE. Foundationalism in epistemology is not inconsistent with the method of RE in moral decision making. Besides, a foundationalist position does not imply that its adherent is not allowed to move 3

Are not decision making and justification two sides of the same coin, in the sense that someone who takes a moral decision might be convinced of its justifiability? The answer to this question is yes and no. Think again of Jeanie Deans. When she took her decision not to bear false witness, she was convinced of being justified in taking this stand. She surely was able to defend her decision. To this extent, the two-sides-of-the-same-coin view is correct. But Jeanie's capacity to take a firm moral stand and to justify it did not turn her into an ethicist or epistemologist. She was no more capable of using the critical, analytical, secondorder concepts of the ethicist and epistemologist of her time than contemporary ordinary moral thinkers are capable of using those of our time. Nor is an ethicist who is engaged in moral decision making automatically involved in the business of moral epistemology.

78

Ton van den Reid

back and forth between her beliefs and, if need be, give up a basic belief. A foundationalist's system of beliefs has a certain structure; it is built from the bottom up. But the beliefs are not fixed within this structure. A particular belief that was originally non-basic can get at the bottom of the system, and the other way around. 4 So, even if Scott's Jeanie Deans had foundationalist leanings, this would not make it improper for me to present her as an RE-ist in first-order moral enquiry. A second point of criticism may be the strongly individual, personal way of moral enquiry which is suggested by the case of Jeanie Deans. The leading question should not be: What ought I, this particular person, to do? but rather: What ought one, one moral subject in a community of others - what ought we to do in this situation? Surely, equilibrium sought through the method of RE is not an individual cognitive state but a situation of interpersonal agreement? It seems reasonable to incorporate religious beliefs in the method of RE as long as the individual character of RE as a state of the person is emphasized. As soon as the emphasis is changed in the direction of RE as interpersonal agreement, religious beliefs lose their character as natural candidates for being included in the method leading to RE. Religious beliefs are anything but tractable; they are divisive. If RE is primarily a method for reaching interpersonal agreement in moral matters, there is every reason to be wary of the inclusion of religious beliefs in it. In this connection one could point to the way moral deliberation is carried out in ethical advisory committees. Most of the time, religious issues and arguments are avoided as a matter of general policy. My answer to this objection is twofold. Firstly, although there is ground in the literature for the view that RE is a method for a community of moral enquirers (Aronovitch 1996, 401; Daniels 1979a, 257), I cannot see how it is a collective method without being a method of individual enquiry in the first place (cf. DePaul 1993,22-23). It is hard to think of the former method other than as derived from the latter. Secondly, there may be good reasons of a 4

Antony Kenny gives a nice example of such a change. There was a time, he tells us, when he believed that there was a continent called Australia - on the basis of reasons: He was taught so at school; he saw the continent marked on maps; his aunt had told him that she had been there. At that particular time his belief in the existence of Australia was not basic in his noetic system. It was supported by and dependent on reasons. It could heve been shaken by thoughts about the unreliability of his teachers, the poor quality of the school maps and the mendacity of this aunt. But now the belief is basic. And, of course, he could give reasons for his belief in the existence of Australia. But the belief would no longer be dependent on those reasons. If they were to fail, his belief in the existence of Australia would not automatically founder; see Kenny (1983, 18-20, 35). For an example of the other way around, think of someone whose former basic belief in the existence of God has moved to the periphery of his cognitive system.

Background Theories and Religious Beliefs

79

pragmatic kind for members of religiously mixed groups (committees) to mutually refrain from expressing religious convictions when in the process of trying to find interpersonal agreement, i.e. equilibrium, in moral matters. (In a committee meeting, Jeanie Deans and Francis Hutcheson would have agreed upon what a witness in a position like Jeanie's should have done.) One could even go further and, like the later Rawls, argue for the usefulness or even indispensability of the concept of public reason with its related idea of an overlapping consensus among citizens of a democratic polity with different comprehensive (religious) views. That is, one could argue that political argument on fundamental matters such as constitutional essentials and basic issues of justice, typically arising from and being directed at interpersonal agreement, should be as free as possible from the expression of divisive religious beliefs. But this does not alter the fact that, for each of the citizens involved, the beliefs shared in fundamental political agreement, a state of interpersonal equilibrium, are part of a more comprehensive set of beliefs often including religious ones, which are - ideally - in equilibrium (for Rawls, see 1993, 153-158, 168-172, 216-220). Reflective equilibrium thus appears to be again principally a matter of the individual moral (and political) enquirer.

4

Does wide RE add credibility to a moral position?

So far, the question has been focused upon whether beliefs of a religious (or 'comprehensive', weltanschaulich) nature can playa legitimate role in first-order moral enquiry and the corresponding method of narrow RE. In this kind of enquiry one may have reached RE on a particular moral problem, for example, whether or not to tell the truth as a witness for the defendant who happens to be your beloved sister. But then the question may arise whether the solution found through the method of RE is justified. Why precisely, in Jeanie's case, is perjury morally wrong, especially when the law infringed by the defendant seems to be grossly unjust? Why does not here the principle of beneficence override the rule which prohibits giving false evidence in court? And if the answer to the latter question implies a religious belief in divine providence, what then is the warrant of this particular belief? With questions such as these in mind, we should not forget that, in Jeanie's case as well as in other instances of narrow RE, we have to do with honest and often deeply felt beliefs. Moreover, and more importantly, these beliefs have also been considered critically, that is, been pruned and adjusted in order to form a coherent set. But, and this is the thing to emphasize now, all these stipulations and qualifications do not make the particular beliefs warranted or justified. They cannot serve as reliable indicators of the beliefs having positive truth value,

80

Ton van den Reid

or put less strongly, having positive epistemic status. Not even the fact that the beliefs are coherent adds to such a status. Each of them and all of them together may have a high credence level, that is, they may be honestly, sincerely, deeply and coherently believed, but what they still lack is the epistemic value of credibility: Are they worth being believed? Is there good reason to believe them? One can see this as soon as one becomes aware of the fact that two people may have diverse and conflicting sets of beliefs which are held with the same credence levels within the narrow RE of each. Not only is it impossible for both of them to be right, but they may also be both wrong. s The problem's treatment takes us beyond first-order moral enquiry into the area of moral epistemology, part of second-order moral enquiry. Wide RE is supposed to provide a solution to it. In this section, I am not directly concerned with the question whether the method of wide RE in general is able to provide such a solution; my interest is rather in a particular attempt to do so, that of Norman Daniels. More specifically, I will go into Daniels's conception of wide RE as a method for justifying a moral position (reached in narrow RE) with an eye to the place of (systematized) religious beliefs in it. Do those beliefs play a role in wide RE? And if not, why? The crucial step leading from narrow to wide RE is the introduction of so-called background theories or beliefs (between which, by the way, there is no sharp distinction).6 It is common knowledge that Daniels makes this move drawing upon an idea put forward by Rawls. Both philosophers think it possible to bolster up the credibility of the beliefs in narrow RE by linking them to these background theories. The argument seems to be in essentials that a relatively small, coherent set of moral beliefs acquires a higher epistemic status by the incorporation of the beliefs of one or more background theories. The wider the range of beliefs standing in a relation of coherence, the greater the credibility of the members of the set. Of course, some work has to be done in order to achieve this result. There has to be, again, as was the c~se with the method of narrow RE, a process of mutual adjustment, a going back and forth between the original set and the background beliefs. But as soon as the wider equilibrium is reached, there is good reason to attach higher epistemic value to the beliefs in question. The argument is plausible as long as we can assume that the particular background theories have a higher credibility than the original set of beliefs in narrow RE. The assumption is doubtful. How should a coherence 5 6

Fore more about the epistemological notions of warrant, justification, positive epistemic status, credence level, etc., see Plantinga (1993a) and DePaul (1993). Daniels (1979a, 258) explicitly states that 'theory' should be understood in a loose way. Like Daniels, I use the words interchangeably in such a way that, whereas 'theories' should be understood loosely, 'beliefs' should be understood in the sense of 'more or less systematized' .

Background Theories and Religious Beliefs

81

model of epistemic justification allow for a hierarchy between the beliefs in question? This idea seems to have a strong foundationalist flavour. But before reaching a verdict in this matter, let us first try to create some clarity about those background theories. Are they of different kinds? And how are they supposed to function in some more detail? The theories Daniels has in mind are a mixed bunch indeed. In his influential 1979 essay on wide RE, he enumerates such diverse theories as a theory of the person, a theory of moral psychology and development, a general social theory and a theory of the role of morality in society. Besides, he apparently includes in the set alternative moral theories or 'conceptions' (Daniels 1979a, 260-261, 264). The thing to note is that none of the theories on Daniels's list is of a purely descriptive or explanatory nature. They all contain moral or at least normative elements. Otherwise the mutual adaptability of those theories and the considered judgements (intuitions) and principles in narrow RE, required to establish coherence between the several elements of wide RE, would be jeopardized. For, imagine such a theory to be of a purely descriptive and explanatory nature. If it were to contlict with (some element of) narrow REa moral theory in nuce - there would be no question of moving back and forth between the particular background theory and narrow RE. The latter should be abandoned. Unless, of course, the background theory were underdetermined by the facts, or were found to be questionable on factual grounds. (Think, for example, of psychological egoism: if a strong version of it were true, it would support (not: entail) ethical egoism. A going back and forth between it and, say, some form of ethical altruism would be out of the question.) A nice illustration of this concern is the way Daniels deals with the difference between Rawlsian and utilitarian principles of distributive justice. In utilitarian theory, a weaker conception of personal identity is assumed than in Rawls's. Both conceptions are feasible within contemporary theories of the person. There are no hard, decisive facts of the matter to be established through empirical methods. Either conception is influenced by different moral considerations. For this reason it is not possible to appeal to the prior acceptability of a (morally loaded) theory of the person for the justification of one's initial moral theory, accepted in narrow RE. The latter theory is on an equal footing with the theory of the person as far as justification is concerned (Daniels 1979c, esp. 277-281; cf. Daniels 1979a, 263). The several elements, including background beliefs, that make up wide RE are to stand in a horizontal relation of coherence rather than in a vertical foundationalist relation of dependence. If this is so, then nothing has changed substantially compared to our earlier conclusion about the epistemic status of one's beliefs in narrow RE. The fact that a certain number of beliefs is coherent is not at all sufficient to make them credible. Enlargement of the number does not of itself add to their

82

Ton van den Reid

credibility. So the more comprehensive set of beliefs in wide RE being coherent does not make them automatically the more credible. The range of beliefs standing in a relation of coherence has been expanded. This may have some epistemic value. But it does not enhance the prospect that one is tracking the truth in what is being believed. At this point we should remember that the no-credibility objection concerns the initial considered moral judgements (beliefs) in narrow RE. Daniels's aim is to probe and vindicate the general reliability of those judgements. Revealing for his procedure is the statement that the no-credibility criticism 'is at best premature'. He suggests that we must wait and see how wide RE works in overcoming the credibility problem of the initial moral judgements. In the context of a moral theory which could emerge from wide RE, 'we might be able to provide a story', he says - and what he means is a convincing story - 'about the reliability of initial considered judgments' (Daniels 1979a, 271). So, wide RE as a method to defeat the no-credibility objection should be judged by its result. Now, the foregoing criticism was targeted precisely at this result which turned out to be unsatisfactory, to say the least. Therefor~, it comes as no surprise that Daniels ultimately appeals to convergence among moral enquirers in his defence of wide RE as a method to vindicate the general reliability and, thus, the credibility of the initial considered moral judgements (Daniels 1979a, 272282). I shall come back to this point shortly.

5

Religious beliefs in Daniels's conception of wide RE

So far, nothing has been said about the role, if any, religious beliefs, or systems of religious beliefs, play in Daniels's account of wide RE. Indeed, direct reference to, let alone discussion of, the function of religious beliefs in wide RE are hard to find in his earlier works. But there is an important indication in his seminal 1979a article which comes close to the acknowledgement of the legitimate place of religious views in wide RE. In that particular section, Daniels deals with ways in which the move from narrow to wide RE might benefit moral enquiry. Through this move, diverse sources of disagreement between moral agents might be brought to light. Philosophers have often located these disagreements in nonmoral 'facts', but the disagreements might have a more complex structure as may become apparent in the reflective equilibria of different persons (Daniels 1979a, 267). How should we understand this way of deriving benefit from wide RE? Two interpretations are possible. Either one takes the background theories involved as a definite, closed set, consisting of the types of theories mentioned in Daniels's article. Or the set is open and one is allowed to include other types

Background Theories and Religious Beliefs

83

as well. On the first interpretation, quite a few cases of moral disagreement between moral agents will not gain in perspicuity because the sources of the disagreements are to be found in background beliefs which do not belong to the closed set, for example, beliefs of a religious (or weltanschaulich) nature. If, on the other hand, one is not restricted to Daniels's set and it is an open question which background theories are to be accepted, people's actual moral disagreements might give us a clue as to what kind of theories to include. The second way in which moral enquiry might profit by the search for wide RE seems to direct us to the latter interpretation. Here, Daniels thinks of wide RE as facilitating the understanding of 'what factors actually do constrain the considerations people cite as reasons, or treat as "relevant" and "important," in moral reasoning and argument' (l979a, 262). If actual moral reasoning should be used as a benchmark for the construction of wide RE and for an adequate grasp of the character and role of background theories, it is natural to take the set as open and including religious beliefs. The third possible benefit of WRE mentioned, though, runs counter to this interpretation. Daniels appears to hold that disagreements about background theories are more tractable than controversies on moral judgements and principles. With the result that, if moral disagreements could be traced to disagreements about background theory, moral agreement might result (l979a, 262). In case relative tractability should be a characteristic of (systems of) background beliefs, religious beliefs can hardly be reckoned among them, for disagreements about religious matters are anything but tractable. But why should background beliefs be more tractable than the other elements of RE? The reason is Daniels's ambition to show the possibility of interpersonal moral agreement, and of convergence in moral enquiry. He wants an ethical theory to emerge from among the several wide REs of moral enquirers which could command assent from as many of them as possible (Daniels 1979a, 264, 274; note also the title of this essay). Seen in this light, the absence of religious beliefs from the set of background theories becomes intelligible. Now, criticisms of Daniels's attempt to argue for the possibility of interpersonal moral agreement and convergence in ethical theory through wide RE are not hard to come by, even apart from the glaring neglect and disregard of (systems of) religious beliefs.? For my purposes, however, it is more important to point to a recent change in Daniels's views concerning the scope of possible interpersonal

7

See, for example, the conclusion of Michael DePaul, who is relatively sympathetic to the method of wide RE: 'In fact I believe we can say with some 'confidence that the method [of wide RE] has been used by many thinkers for quite some time without yet producing any significant convergence.' (DePaul 1993, 54) For a harsher judgement, see Haslett (1987, 305-311).

84

Ton van den Held

moral agreement through RE, and, especially, the content of wide RE. The change, moreover, brings out into the open his own doubt and retractation concerning his former position. Here I have in mind his essay on 'Reflective equilibrium and justice as political' (Daniels 1996) which is largely an account of his turning away - in Rawls's wake - from the idea of an RE according to A Theory ofJustice to one ala Political Liberalism. The original thought was that the method of wide RE was capable of producing a justification for justice as fairness among all reflective persons. Daniels now considers this to be a philosopher's dream. It is an illusion to think that one philosopher's wide RE can stand in as a proxy for everyone's (Daniels 1996,161). In so thinking, one turns a blind eye to the pervasive and deep fact of serious, reflective people disagreeing about moral and political matters. Instead, as far as political justice (and other normative, community-related issues) is concerned, one has to build on an overlapping consensus between people holding different wide REs. The justification such a consensus might provide is not complete, but only pro tanto, that is, partial. Complete justification is produced by each person's wide RE, and varies among people with different wide REs. The thing to note is that Daniels's acknowledgement of the failure of wide RE as an interpersonal method of justification is accompanied by the admission of religious beliefs to the wide RE of the individual moral enquirer (Daniels 1996, 148-149, 156-157, 168). In retrospect, Daniels calls it 'quite striking' that he paid so little attention 'to the great diversity in beliefs about philosophical, religious and even moral matters that would have to be incorporated in such wide reflective equilibrium' (1996, 146). The changes in Daniels's conception of wide RE, not least the latter one, are to be welcomed. Daniels, however, is reticent about the question why precisely religious beliefs should be admitted in RE. What can their role actually be? To this issue I shall now tum briefly.

6

The role of religious and other fundamental views of life in wide RE

In section 2, I showed not only that religious beliefs are met with in first-order moral enquiry, but also that they have a legitimate place in narrow RE. It would be strange indeed if those beliefs were not to be subjected to closer inspection in second-order moral enquiry. It is, therefore, to be expected that wide RE, now conceived as a fonn of second-order moral enquiry aimed at the credibility of one's moral beliefs within a coherent structure,S will engage religious and 8

Note that a coherent structure need not be coherentist; it might be foundational.

Background Theories and Religious Beliefs

85

other weltanschaulich beliefs. Secondly, additional support for the inclusion of this type of beliefs can be derived from Daniels's account of wide RE. We remember that, according to Daniels, wide RE should help us to understand what factors constrain the considerations people cite as reasons in moral argument. Since religious belief is such a factor, pace the younger Daniels, one better not overlook it in wide RE. Thirdly, it is possible to point to the important role that this type of belief actually plays in the discussion of second-order moral views. I shall give a few examples. The first is concerned directly with the case of Jeanie Deans. She appears implicitly to hold a divine command (or will) theory of moral obligation. That is, very roughly, one has a moral obligation to do action A (or to refrain from doing A), if and only if God commands one to do A (or to refrain from doing A). This theory has been under attack at least since the days of Plato. Recently, Christine Korsgaard has given a both sympathetic and critical assessment of a HobbesianlPufendorflan version thereof. Her analysis is sympathetic in that she agrees that being in a position to create moral requirements implies having legitimate authority to do so. Furthermore, the laws which God requires us to obey are precepts of reason, determined independently of any arbitrary legislative will. Yet, and thirdly, it is not merely their reasonableness that obligates us to obey; the authority behind the requirements is what obligates us. And, finally, she acknowledges that there is nothing inappropriate in the motive of duty. The fact that some action is required, Le. willed by God, is itself, quite apart from any fear of sanction, a reason to do it. On the other hand, Korsgaard is critical of the idea in divine command (voluntaristic) theory that God's authority is based on his power (Korsgaard 1996b, 25-30). Now, this may be true of seventeenth-century voluntaristic thought, but it is not an essential part of the theory. By assuming that the authority of moral requirements as divine commands ensues from divine power, Korsgaard scores a too easy victory for her own Kantian theory of obligation. To indicate why I believe so, think of God's Anselmian great-making properties such as his infinite knowledge, freedom and goodness. Why could not they qualify him as a legitimate moral authority? His authority strikes me anyway as more convincing than that of Korsgaard's construction of a person's own contingent moral identity consisting in her legislative citizenship of the kingdom of ends. The real problem with a divine command theory of moral obligation is not God's lacking legitimate authority, but the possibility of God's existence being doubted or even denied. 9

9

For more about the divine command theory of moral obligation and its strength compared to Korsgaard's theory, see Van den Beld (1997).

86

Ton van den Reid

A second example of how religious or weltanschaulich (systems ot) beliefs function in the discussion of second-order moral views is John Mackie's critique of moral (or value) realism. Some twenty years ago, Mackie famously argued in favour of a kind of moral scepticism by attacking the idea that (moral) values broadly taken - including moral obligations - are objective in an ontological sense. That is, he denied that they are part of the world as observed; instead, they belong to the world of the observer, to his or her desires and interests. It is precisely this combination of denial and positive statement that is definitive of Mackie's second-order brand of moral scepticism. 10 One of his scepticismsupporting arguments is that the idea of values being part of the 'fabric of the world' is 'queer': 'If there would be objective values, then they would be entities (...) of a very strange sort, utterly different from anything else in the universe.' (Mackie 1977, 38) Mackie's presupposition here is a sober evolutionary naturalism: the world is ultimately such as the natural sciences show it to be. He acknowledged, however, that a conception of values as objective in an ontological sense is built into ordinary thought and language. The phenomenology of values seems to underwrite their objectivity, or ontological reality. And not only that, the particular idea is entirely at home in a world interpreted in theistic (or Platonic) tenns (45, 48). Because he thought that theism (or Platonism) could not be defended, he was driven to a so-called error theory of moral scepticism. Moral phenomenology is misleading. Its error must be exposed (which he attempted to do with various arguments, one of which being the argument from queerness mentioned above). Thus Mackie's second-order moral scepticism appears to be based, at least partially, on a fundamental, non-religious world view, Le. naturalism. On the other hand, a religious view of life and the world, such as theism, could be advanced along the lines suggested by Mackie in order to give a coherent account of the phenomenology of (moral) values being objective. Of course, this phenomenology can be disputed with the result that the fit between it and theism disappears. Or one might object to a hard-and-fast ontological understanding of the (supposed) value objectivity. In the latter case, yet another example of the relevance of weltanschaulich beliefs in wide RE can be put forward. According to David Wiggins, protagonist of a 'sensible subjectivism', there are value properties and sensibilities for perceiving them made for each other as equal and reciprocal partners. Blackburn objects to the idea that those sensibilities are 'made for' the properties. He wonders who or what has made them like that. There is no natural story available that can give an explanation. But such a story can actually be told about the reverse relation, 10

Mackie acknowledges that there are other kinds of moral scepticism, see Mackie (1977,16).

Background Theories and Religious Beliefs

87

that is, about how the properties are made for the sensibilities. It is a story about the evolution of mankind. The sensibilities were first to appear, moulded by evolutionary success. Then the properties were projected onto the world. The conclusion is that value realism is untenable (Blackburn 1987, 366-372). It is worth noting that Blackburn recognizes the possibility of a supernatural story with a creator God as the main character about human sensibilities being made for the value properties. He rejects it out of hand. But if it is reasonable to believe in God, such a story could be developed, for example, in the spirit of Plantinga' s epistemology (Plantinga 1993b). Again, the issues of naturalism and theism, or of different kinds of naturalism, appear to playa role in the background of second-order moral controversies. I brought them up, in bare outline, only to show that religious and other fundamental world views should have a place in wide RE.

7

Conclusions

The case of Jeanie Deans was introduced to make clear how important a function religious beliefs can have in first-order moral thinking. There is no good reason to exclude them from narrow RE. The fact that religious beliefs are (often) part of a foundationalist noetic structure does not necessarily prevent the person who has them from using the method of RE, in first-order moral enquiry, to find out what the moral thing to do is in a particular situation. Things are different when wide RE is proposed, in second-order moral enquiry, as a coherentist method of moral justification. Foundationalism and coherentism in (moral) epistemology are incompatible. It is argued that Daniels's brand of coherentism is unsatisfactory, anyway. Background theories being coherent with considered moral beliefs can only add to the epistemic status of those beliefs if the theories concerned have more epistemic weight (or value) than the beliefs. This condition, however, is not met in Daniels's account of wide RE. In wide RE as conceived by the younger Daniels, religious beliefs apparently did not and could not playa role as background theory (nor in any other role). The later Daniels, having changed his conception of wide RE from an interpersonal to a personal method of moral justification, positively admits religious beliefs to it. The new conception should be fit to produce an overlapping consensus between persons with different reflective equilibria on the basis of which fundamental normative, community-related issues could be decided. Finally, some examples have been presented to clarify the role religious and other fundamental views of life and world play in wide RE, which is now

88

Ton van den Reid

conceived as a form of second-order moral enquiry aimed at the credibility of one's moral beliefs within a coherent (not: coherentist) noetic structure.

Ideals and Ideal Theory: The Problem of Methodological Conservatism Wibren van der BurgI

1

Introduction

A well-known problem of reflective equilibrium methods is how to avoid or at least correct for a methodological conservatism or, more positively formulated, how to guarantee adequate critical input. If in the process we use only our own convictions, of whatever kind, how can we hope to do more than systematising our prejudices? In other philosophical theories, the common solution is to suggest that there is some Archimedean point or locus of certainty, but this solution is clearly not available for reflective equilibrium methods. Most strategies to solve this problem in reflective equilibrium methods are based on some form of abstraction or idealisation. We can create, e.g., a distance' from our concrete selves, from our concrete society or from our concrete normative judgements regarding the here and now. The strategy of abstraction from our concrete selves and our concrete society is most prominent in Rawls's theory, in the construction of the persons in the original position, who have to make decisions behind a veil of ignorance. There is yet another way in A Theory oj Justice in which distance from our own society is created which, in my view, is too often uncritically taken for granted. The normative theory that Rawls develops is not meant to be directly applicable to our own society; it is an ideal theory, that is, a theory for a perfectly just society. A process of transformation is still needed to construct a non-ideal theory - a normative theory for the real society we live in. Rawls, however, does not specify how this transformation process is to take place. He only illustrates it with a very limited non-ideal theory about civil disobedience in a nearly just society. Therefore, the general question remains open under what conditions and how precisely an appeal to ideal theory mayor may not be legitimate when constructing a normative theory for the real society we live in or when trying to solve concrete moral problems. My thesis in this article is that the transformation process from a normative theory for an (almost) ideal society to a normative theory for a real society involves so many problems that such an appeal only rarely is valid. I would like to thank Marcel Verweij en Anton Vedder for their helpful criticisms of an earlier draft. 89 ~

van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 89-99. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

90

Wibren van der Burg

Therefore, I suggest a third way of creating distance which has close affinities with the appeal to ideal theory, but is less vulnerable to the objections against it. It is the appeal to ideals as such, to valuable future states of affairs. Ideals, like ideal theory, go beyond the principles and considered judgements about what we should do here and no\v. My argument is that ideals should play a crucial role in reflective equilibrium methods, and that they are especially important as an element which continuously reopens the argumentation processes and the provisional reflective equilibrium to critical perspectives. It can be defended that ideals play an important role in the construction of the original position as well. Vacious authors, e.g. Ronald Dworkin (1978) and T.K. Seung (1993), have argued that the original position is based on a normative deep theory in which the (Kantian) ideals of equality and liberty are central. I will not pursue this line - although I think it a valid one - because I want to focus on ideals as substantive elements in the reflective equilibrium process rather than as the basis for methodological requirements of impartiality.

2

The appeal to ideal theory in normative argument

Both in A Theory ofJustice and in Political Liberalism, Rawls stresses that his primary focus is on developing an ideal theory: a normative theory for a perfectly just society. He argues that we should always start with constructing an ideal theory, which subsequently can be the basis for constructing a non-ideal theory. The concept of ideal theory needs clarification. One possibility, which I will call descriptive ideal theory, is to regard an ideal theory as the elaboration of an ideal. For example, it could be a detailed description of the good society we aspire to, or a description of the ideal doctor. I think this kind of theorising about ideals is most useful in normative argument; I will address its use later. Rawls's two principles for the basic structure of society are part of descriptive ideal theory. However, he also goes beyond this. The second part of his ideal theory might be called normative ideal theory. This is a normative theory not about the basic normative principles on which an ideal society should be built but on the more specific principles that should guide political action in such a society. Description of a future (or imagined) state of affairs shades into prescription about how institutions and people should act in that ideal context. If we want to go from descriptive ideal theory to normative theory for the reality we live in, to moral principles that can serve as action guides, we encounter various problems. There is a problem of transformation (how to translate ideals into principles) and one of indeterminacy (how to balance conflicting principles). It seems, p.evertheless, possible to construe normative principles. The ideal of peace, for example, can be elaborated into a descriptive

Ideals and Ideal Theory

91

ideal theory about a peaceful world in which no nuclear arms exist and no military threats are made between states. This can then be transformed into nonnative principles along two lines of argument;2 one is teleological and results in a general principle 'act according to the most effective strategy towards this peaceful world'; the other is deontological and results in a general principle 'act according to the principles that would be accepted in this ideal, peaceful world'. The first principle can be elaborated into more precise principles about how to reach the ideal of peace, principles which must be based on an analysis of both the real world and the ideal world. For an elaboration of the second principle, however, we only need to analyse and to elaborate the ideal of the peaceful society to know which specific principles would be accepted in that context. In my view, we need both types of principles in a mixed theory. The crucial question is, of course, how to balance the various principles in the case of conflict, but this is a common problem in moral philosophy. The most important point is that, by accepting the ideal of a peaceful society as a point of reference, we can construct a framework in which all relevant aspects can be fitted, including the realistic assessments of world politics. If, however, we start from normative ideal theory in order to develop a normative theory for the real world, we encounter more serious problems. The starting point then is a set of specific principles about how political institutions should be structured and how institutions and individuals should act in a future ideal society. Analogous to the second line of argument I distinguished above, these ideal-theory principles can be transformed into deontological principles for our society. However, there is no analogous possibility to derive teleological principles that can adequately deal with the fact that our society is not so ideal and that can lead us from our realHfe at least in the direction of the ideal. We do not have a mixed normative theory then, but only a theory consisting of deontological principles that have been developed for an ideal situation. Starting from normative ideal theory thus results in a biased and partial normative theory for the real world. The ideal of peace can serve as an illustration again. Our world is still far from peaceful. The principle that we should never threaten other countries with military force is certainly a defensible principle for an ideal society. It is clear, however, that this need not always be the best principle in the world we live in, for example in dealing with very aggressive neighbours or with countries that commit genocide and other crimes against humanity. Threats or even real military force may be necessary and justified against such countries.

2

In section 3, I will show in more detail how we might use ideals to support various types of principles.

92

Wibren van der Burg

So the central problem is clear. Principles that are perfectly justifiable in an ideal society need not be equally justifiable in the non-ideal society we live in. We seem to need at least some teleological principles that show us how to get from our society to the ideal society. Starting from descriptive ideal theory is therefore more fruitful than starting from normative ideal theory. Does this mean that an appeal to nonnative ideal theory is never useful? This would be too hasty a conclusion. Nonnative ideal theory can be the basis for convincing arguments in a special way, which is nicely illustrated by Rawls's own example of normative non-ideal theory: the theory of civil disobedience (Rawls 1971,333-391). Starting from his ideal theory, Rawls argues that even in a nearly just society, civiI disobedience would be justifiable as a form of protest against certain forms of injustice. Even in such an almost ideal society, civil disobedience should therefore sometimes be tolerated as a legitimate form of democratic protest. If this is a valid argument (which I think it is), we can easily infer that, a fortiori, it is sometimes justifiable and should be tolerated in our own, far from nearly just society. Thus ideal theory offers a strong argument against straightforward law-and-order positions that hold that civiI disobedience is never morally justified or politically legitimate. Rawls's theory of civil disobedience is very restrictive and allows only a limited range of protest actions. In an almost perfect society, such a restriction would certainly be justified. However, this is different in our far from perfect society. The appeal to ideal theory is, therefore, much less convincing against those who argue in favour of a much broader scope for justified civil disobedience. Most acts of civil disobedience are directed precisely against the fact that our current society is not an ideal society. Why then should we use a very restrictive theory, valid for a nearly just society, to judge acts that aim to make our unjust society more just? If our society is much less just, the harmful effects of civil disobedience on social order would probably put less weight into the scales. More importantly, the reasons why we have a prima-facie political obligation in an ideal society - e.g. that the law is largely justified - need not necessarily hold (cf. Van der Burg 1989). The interesting conclusion is that sometimes an appeal to nonnative ideal theory can be valid against the arguments of specific opponents, whereas it may not be convincing against the arguments of other opponents in the same debate. In the case of civil disobedience, the appeal is successful against conservative law-and-order positions, but not against more liberal or radical positions that claim a broader justification for civil disobedience. In other cases, the positions may be reversed, and an appeal to ideal theory may be successful against more progressive positions. An example may be income distribution. Rawls's difference principle may be effectively used against radical egalitarians who

Ideals and Ideal Theory

93

claim that everyone should, without exceptions on grounds of merit, earn the same income. It does not help us much, however, once we have accepted that there can be legitimate income differences, in determining how large these differences should be. The difference principle is not effective against the position that there should be no limit to income differences at all because, given the motivations real people have, we must allow them at least the perspective of becoming extremely rich. Here normative ideal theory is of no help. The conclusion so far is that appeals to normative ideal theory in the context of real-life problems are usually of not much use. It is not clear why normative principles that are justified in the context of an ideal society and presuppose this ideal society for their justification, should be an acceptable starting point for thinking about the normative principles for the here and now. We even have reason to believe that sometimes this starting point will lead to a biased approach.

3

The role of ideals in a network model

The negative conclusion in the previous section does not imply that appeals to ideals and to descriptive ideal theory are equally problematic. On the contrary, the example of the ideal of a peaceful world society suggests that there are various ways in which such an ideal can be legitimately translated into more specific principles for real life. Let us therefore make a fresh start. I assume that the goal of a reflecti ve equilibrium process is to develop normative theories for the real society we live in, or to reach justified conclusions concerning concrete normative problems. Robert Heeger and Thea van Willigenburg have developed, in a number of publications, a network model of reflective equilibrium (Heeger 1988 and 1990a; Van Willigenburg and Heeger 1991). According to Van Willigenburg and Heeger (1991), a network model consists of three elements: moral principles, moral intuitions and morally relevant facts. In order to reach justified judgements on concrete moral problems, we should try to establish a reflective equilibrium between these three (groups of) elements. The advantage of their model is that it gives due attention to the morally relevant facts and thus to a realistic assessment of the situation and the larger context we have to deal with. However, in this network model the problem of guaranteeing critical input is particulary urgent. All the normative elements used are directly connected to our own experience. According to Van Willigenburg and Heeger, principles are no more than systematisations of our moral experience, and our moral intuitions are also a reflection of our experience. So how can we go beyond our experience in these models and guarantee critical output?

94

Wibren van der Burg

My suggestion is that introducing a fourth element in the network model, ideals, will be a good way of promoting critical input into the reflective equilibrium process (cf. Van der Burg 1991). Ideals may be defined as follows: 3 Ideals are values that are implicit or latent in the law or the public and moral culture of a society or group that usually cannot be fully realised and that partly transcend contingent, historical formulations and implementations in terms of rules and principles.

This definition combines three elements. First, ideals are values rather than direct action guides like principles or rules. Second, an ideal is both grounded in reality and oriented towards the future. Third, ideals are vague and cannot be completely grasped in a formulation, nor can they be completely realised. Especially the third element is a reason why ideals can fulfil such an important role in warranting critical input. Ideals are much more ambiguous and indeterminate than principles, and can never be fully grasped. Each formulation of a more elaborate conception of an ideal will always describe the ideal only partially. In the light of new situations, we may suddenly see new dimensions of the ideal and reformulate our conception of the ideal. This again may lead to a reformulation of our principles and considered judgements. The fact that ideals transcend every historical formulation thus makes them a continuous source for new critical input into the reflective equilibrium process.. How then do ideals playa role in a network model? The fact that ideals are not direct action guides implies that we need a process of interpretation and transformation in order to connect them to principles or considered judgements. Ideals can be connected to principles along two ways, corresponding to the distinction between deontological and teleological principles. 4 A first step along both ways should be that from concept to conception, 5 e.g. from the abstract formulation of an ideal of non-violence to the more specific description that

3

4

5

I have elaborated this definition in Van der Burg (1997). Taekema (1998) offers an interesting analysis of three ideal-typical models of ideals: cultural, heroical and transcendental. My definition is a combination of the cultural and transcendental models. Although it is impossible to avoid terms like derivation in this presentation, it should be conceived as a two-way process. We can take the reverse steps to construct ideals on the basis of our principles and intuitions. The distinction between concept and conception is central to Dworkin's theory of interpretation (Dworkin 1978) but can also be found in Rawls (1971, 3 ff.). In Rawls (1993,14), he traces the distinction back to Hart (1961,155-159).

Ideals and Ideal Theory

95

no one should ever use any physical force against persons. 6 The second step can either be direct or indirect. A direct derivation is based on the idea that, as much as possible, we should act according to the ideal here and now. The result is a deontological principle, e.g. the principle: 'Do not kill'. An indirect derivation is based on the idea that we should try to realise the ideal in the future, as much as possible, and take those steps that will lead us to that future goal. The result is a teleological principle, e.g. the principle: 'Work towards a nonviolent society' .7 We can now see how ideals can help to counter the built-in methodological conservatism of Van Willigenburg and Heeger's network version of reflective equilibrium. Because ideals are future oriented and transcend every attempt to formulate specific conceptions, they can help us to go beyond our limited moral experience. They constitute a perspective from which we can look at our own society and our moral views with some distance. They are a continuous source of new normative ideas that can be used to criticise our current principles and intuitions. They are thus, usually in a dialectic interaction with new social or technological developments in reality, one of the major factors in the dynamics of our moral views. Ideals are certainly no guarantee for adequate critical input; they are just one element in the process which offers no guarantees for success. The recognition of ideals has to be supplemented by other attempts to counter methodological conservatism. One way is to find some guarantees in the person who makes the judgements, e.g. by focusing on the competent judge (as in Rawls 1951) or, following the suggestions in DePaul (1993), by deliberately seeking to enrich our moral experience and thus developing our moral capacities. Another way is to take the words 'our moral experience' seriously in their intersubjective sense and try to introduce intersubjective criticism and discussion. It is not merely my personal experience in which ideals are grounded and which is summarised in moral principles, but the much richer experience of my society, of my culture. We should nevertheless acknowledge that we will never be fully able to counter methodological conservatism and to take a completely distanced, impartial critical view on our own moral convictions. Despite our attempts, we will always be vulnerable to the criticism that we have not completely gone beyond our own personal cultural biases. But is this really a serious problem 6 7

One could elaborate this conception still further. A more elaborate conception of an ideal would amount to what I have called earlier a descriptive ideal theory. Along similar lines, ideals can also directly and indirectly be connected to our moral intuitions. I have more extensively discussed the connection of ideals, principles and policies, rules and concrete judgements in Van der Burg (forthcoming 1999).

96

Wibren van der Burg

or rather an advantage of reflective equilibrium methods? If our aim is to gain moral knowledge (whatever this might be), it could be a problem, although we could argue against critics that similar epistemological problems are also unavoidable in the sciences. If our aim is, however, to live a morally good life and to act morally responsibly, we should rather regard the connection to our cultural and personal identities as an advantage. It is this concrete historic person who has to act, not some abstract person in an original position or an archangel. If we want our nonnative theories or our moral judgelnents to be effective action guides, they should not be too distanced. A balance should be found between practical effectiveness (people actually living up to their moral views) and moral legitimacy (their moral views being justified). I think a reflective equilibrium which recognises ideals is the best way of striking a balance here: It starts with people's own convictions (including their ideals), and tries to build in as much critical reflection as possible. I do not see how we can do better, being the persons we are rather than some kind of idealised saints.

4

Immanence or transcendence?

So far, I have ignored one major problem of my analysis of ideals. If they are immanent in our culture or experience, how can they also transcend this culture or experience and form a point of critical reflection? An interesting observation to start with is that many authors who recognise ideals (e.g. Dworkin 1986; Selznick 1961) somehow combine the idea that ideals are immanent in reality or experience, with a claim that they transcend reality. I do not think this is accidental - it is in the nature of ideals that they combine both. The question is not whether these two should be combined, but how. Seung (1993) has argued that only if we assume a transcendent existence of our fundamental ideals as Platonic forms, can we find a guarantee for full universality of our nonnative claims. If ideals have no such existence, we can never reach beyond our cultural limitations and attain full universality. The problem with his argument is that, in my opinion, he does not show what such a transcendent existence would be or how we could know it. Nevertheless, his analysis demonstrates an important point: The claim to transcendence is closely linked with the aspiration to universality. This point should be taken seriously, but can be combined with a constructivist interpretation, which to me seems to be more in line with authors like Selznick, Dworkin and Rawls. The starting point should be that ideals are grounded in reality, in our moral experience. They are latent or immanent values embodied in our culture. They have a basis in experience. For example, they may find their origin in so-called negative contrast experiences, experiences

Ideals and Ideal Theory

97

that something is flagrantly unjust or wrong; the ideal can then arise as an idealised negation of this experience. 8 The formulation of an ideal of justice may thus be triggered by concrete experiences of clear injustice. Furthermore, the fact that ideals are grounded in experience implies more than merely that they arise from such concrete negative experiences. They may also be supported by positive experiences, for example, when we repeatedly experience that striving for certain ideals is worthwhile, has positive effects and gives a richer meaning to our lives (cf. Rescher 1987). However, ideals may be grounded in experience, but they cannot be reduced to experience. Their nature is such that, as soon as we recognise them or construct them explicitly, they move us beyond our experiences. Or, perhaps we should formulate this more carefully, in order to avoid the unnecessary suggestion that ideals are independent actors: We move beyond our experiences and beyond our reality as soon as we reflect on the meaning of our ideals. If we formulate a principle of justice, this may be nothing more than the systematisation of our moral experiences, but if we formulate an ideal ofjustice, we reach a more abstract and open level of reflection. At this level, there is always an indefinite surplus of meaning and an openness to yet unforeseen implications. It is not possible to restrict the meaning of the ideal to the original experiences or principles it systematises; there is always a movement towards broader perspectives. Thus, ideals do not have a transcendent existence but they embody a transcending movement towards more general, more universal meanings. The conclusion can be that ideals are both immanent in our culture and embody a transcending tendency. I think this reading of ideals in which transcendence and immanence are combined in a coherent way, is the most defensible one - we do not need to suppose some transcendent existence, which would raise many epistemological and metaphysical problems. I should formulate my point even more precisely, because my interpretation still suggests some form of realism about the existence of ideals in social reality. The suggestion that ideals are latent or immanent in our culture does not mean that they are somewhere out there, merely to be discovered. Our public culture (and even more so the law) is a collective construction, partly grown organically through the ages, partly deliberately created. If I say that a culture embodies certain ideals, this statement presupposes this process of human construction and is itself also the result of my constructive interpretation of that culture. References to the 'immanent existence' of ideals should therefore not be read as reference to a similar form of existence as physical objects have.

8

Cf. Brom (1997, 65); cf. also Van den Brink (1997, 153 ff.).

98

5

Wibren van der Burg

Concluding remarks

In this article, I have argued for the recognition of ideals in reflective equilibrium methods. They can play an important role, especially in the dynamics of reflective equilibrium and in the attempts to get a higher degree of universality and critical reflection on one's own convictions. I have illustrated this thesis with some examples, but I have not given an elaborate demonstration of how this might work in practice; this I have tried to do elsewhere. 9 The focus in this article was on normative moral or political theory; analogous models of reflective equilibrium can be constructed for legal theory, and in such models, ideals fulfil a similar role (Van der Burg 1993). The thesis that ideals are especially important for the dynamics of reflective equilibrium and for the enhancement of critical reflection means that their role may not always be the same. If the field of discussion is one in which the moral convictions are relatively settled and stable, the role of ideals will usually be relatively small. The most relevant dimensions of the ideals will then have been acknowledged and fonnulated in the generally accepted principles. An appeal to ideals to solve concrete problems will then largely be superfluous. If a field is one in which the facts or the moral convictions are in a process of change, e.g. as the result of technological changes, the role of ideals will be much greater. They then provide new perspectives and critical reflection, which may lead to the formulation of new principles and the rethinking of our considered judgements. Similarly, we may expect ideals to be more important in the context of theory construction than in the context of concrete decision making. If we want to develop a normative theory, even a modest one for specific themes such as democracy, ignoring the larger ideals will usually lead to neglecting important dimensions of the theme, if only because many normative p~enomena cannot be fully understood unless we acknowledge that they are ideal oriented (Selznick 1961). For problem solving, especially in relatively stable fields, an appeal to ideals may often be much less necessary as long as the central principles connected to these ideals are reasonably justified and generally accepted. Even then, however, the principles will always be open to revision; reflection on new, so far unrecognised dimensions of ideals may easily distort the provisional

9

In Van cler Burg (1991), e.g., I applied it to democracy and argued that we cannot understand and justify our considered judgements regarding democracy unless we refer to a larger democratic ideal. At Tilburg University, we have recently started a substantial research programme in which the importance of ideals is investigated, both in theory and in practice, for law, morality and politics (cf. Van der Burg and Brom 1998).

Ideals and Ideal Theory

99

equilibrium. to It is the process which is central to the method of reflective equilibrium, in my view, not the result. Ideals help to keep the process of critical reflection alive, and hence they help to keep us as humans alive.

10

The discussion by Marcel Verweij in this volume of how reflection on the ideal of autonomy can lead to the revision of accepted principles is a good example.

Part II The Methodology of Reflective Equilibrium

Rationality, Warrant and Reflective Equilibrium Eberhard Herrmann 1

1

Introduction

The aim of this article is to show the relevance of internal pragmatic realism for the concept of reflective equilibrium regarded as a method for deliberation in moral situations. The main focus will be on the question of how to distinguish between rationality and warrant. To begin with, I will give an account of the debate in moral theorizing. Second, I will focus on some problems caused by reflective equilibrium conceived of as coherentism. Third, I will prepare a solution to these problems by bringing in the perspective of pragmatism. Fourth, I will propose internal pragmatic realism to avoid the risk of relativism threatening the method of reflective equilibrium. Finally, I will relate internal pragmatic realism to Michael DePaul's modification of reflective equilibrium by focusing on his idea of balance and refinement and of experiential warrant. Dealing with these five issues I hope to show that the importance of moral experience for reflective equilibrium lies in the role emotions play in our lives. Furthermore, I hope to show that, although rationality and warrant are different notions, they are both person relative.

2

Reflective equilibrium

On the one hand, moral theorizing as part of our ordinary moral practices is contextual and not so much systematic. On the other band, most moral theorizing done by philosophers aims at constructing theories that are explicit and universal. Philosophers want to identify the principles that govern all theory building and therefore look to epistemology as the study of how we can know things. In epistemology, the two most influential universal approaches to the question of how we can know things are foundationalism and coherentism. Foundationalists claim that some (non-basic) beliefs can be justifiably held on the basis of other (basic) beliefs. It is not difficult to understand why people accept foundationalism. Justifications have to end somewhere. However, we

I want to thank Dirk-Martin Grube, Mark Sluys and Mikael Stenmark for having critically discussed drafts of this article with me. 103 ~

van der Burg and T. van Willigenburg (eels.), Reflective Equilibrium, 103-114. @ 1998 Kluwer Academic Publishers. Printed in the Netherlands.

104

Eberhard Herrmann

may wonder why some beliefs for which justification ends, are nevertheless justified although they do not require justification for themselves. The answer that they are self-evident and therefore self-justifying does not solve the problem. Self-evident beliefs do not seem to be rich enough to allow the deri vation of any interesting moral point of view. So, let us leave foundationalism and look at the coherentist alternative in moral theory, which in its most influential form consists of John Rawls's method of reflective equilibrium. According to Rawls, when we want to account for moral beliefs, we start with referring to general principles. Next, we revise principles and beliefs in the light of each other, until there is equilibrium (Rawls 1971, 19-21 and 48-51). This may sound simple, but there is a fundamental tension that is not always observed. It concerns the meaning of 'equilibrium' . In one sense, it means the feeling of not being puzzled anymore. Most philosophers dealing with the method of reflective equilibrium do not consider this aspect. Rather, they focus on another sense of 'equilibrium' when referring to logical aspects. This applies both to the idea of narrow reflective equilibrium and to the idea of wide reflective equilibrium later introduced by Rawls (Rawls 1974-75, 8). The idea is to also take into consideration alternatives to the moral conception accepted in narrow reflective equilibrium. For that we need to take account of philosophical, scientific and other types of beliefs that might figure in the arguments for and against alternate moral conceptions. As in the original proposal, none of the judgements, principles or theories involved is privileged. How does this idea work when we have to make a moral deliberation? According to Robert Heeger (1992b, 102-110), three elements playa crucial role in moral deliberation: intuitive moral judgements, moral principles and morally relevant facts. Relating these three elements does not take place in a fixed order. Rather, it is the question of a continuous interaction that can result in confirmation or in modification of some elements. Heeger does not state it explicitly, but it seems to be the case that intuitive m~ral judgements in one sense playa crucial role as far as the aim of moral deliberations is concerned. Moral deliberations are acts that involve our life because they concern what is valuable for us. Seen from this perspective, the acceptability of moral intuitive judgements is crucial to moral deliberation because they underlie our actions. Although we would not be able to decide on our actions in concrete cases without intuitive moral judgements, they are not infallible. They can be irrelevant, incomplete, wrong or misleading, since there is always the risk that we see just what we want to see. When we want to scrutinize an intuitive moral judgement, we can ask whether we conceive of the case at issue correctly. We can also ask what moral principles influence our intuitive moral judgements. Compared with intuitive moral judgements, moral principles are different in that they are the result of more reflection. A moral principle does not express

Rationality, Warrant and Reflective Equilibrium

105

a spontaneous reaction to a concrete case. It systematically summarizes intuitive moral judgements about concrete cases in the past. A moral principle reflects the experience people had in earlier cases where it was necessary to make a moral decision. If we want to come up with sufficient reasons to make a moral decision in regard to a concrete case, we have to relate the following questions to each other: Which intuitive moral judgements can persist in the light of moral principles and morally relevant facts? Which moral principles can illuminate the case at issue? Which features of the case at issue are relevant facts? Answering these questions implies a confrontation of intuitive moral judgements, moral principles and morally relevant facts. If the result is a coherent and consistent answer, we can say that the moral decision at issue forms 'a credible judgement' (Heeger 1992b, 110).

3

Coherence

In a specific sense, the method of reflective equilibrium as a method for justification is analogous to the method of coherentism in epistemology. The concept of coherence relates to a system whose elements stand in mutual relations of both consistency and some kind of interdependence. Consistency is, of course, a minimum requirement but it is not enough since, for instance, beliefs about completely unrelated matters fulfil this minimum requirement. There is the need of some kind of dependence. However, it seems difficult to specify it suitably. The closest dependence would be mutual entailment. Yet, it would make one or the other belief redundant, depending on which beliefs we start with. Another way is to say that a belief B coheres with an antecedent set of beliefs if we can infer that belief B from the antecedent set. In addition, one might say that belief B is justified if it survives comparison with competitors, to find acceptance among the beliefs of the antecedent set. However, in this case we have to show why belief B merits acceptance and other equally coherent beliefs do not. Furthermore, we have the question of the antecedent set itself and its justification. If even its justification is a matter of coherence, we have to answer the question of what the antecedent set coheres with in its turn (Grayling 1996, 42 f.). We seem to have a similar difficulty in ethics when it comes to collective deliberations. The coherence of an individual's judgements is not sufficient for a collective decision. Other individuals might appeal to other conflicting intuitive moral judgements, to other moral principles, or might consider other facts to be morally relevant. Thus, different reflective equilibriums are conceivable. To avoid arbitrariness, we need some methodological rules that

Eberhard Herrmann

106

go beyond the logical aspects of coherence. As Heeger points out, the persons who are involved in deliberation should be receptive, awake, open to facts and morally sensitive (Heeger 1992b, 110). Let us imagine the following situation. A man assaults a woman and, when she escapes, asks in which direction she ran. The principle to always tell the truth, confirmed by moral experience, implies answering the question truthfully. However, such a decision needs to be examined by means of another experience, namely that the man's behaviour is strange. To identify the situation correctly presupposes a further kind of moral experience generated by another principle, i.e. that in some situations some people need protection. In our example, this principle may override the principle of always telling the truth because of another moral experience. Seen from this angle, moral experience obviously plays a crucial role in moral deliberations. However, let us explore the concept of rationality first, followed by moral experience regarded as warrant.

4

Rationality

We usually distinguish between theoretical and practical rationality. The fonner applies to beliefs. An irrational belief is one that obviously conflicts with what one should know. According to this characterization, rationality is a person relative concept. What obviously conflicts with what one person should know, need not obviously conflict with what another person should know. That rationality is a person relative concept is also true for practical rationality, a type of rationality that applies to actions. According to a strictly instrumental interpretation, rational action simply means acting in a way that is maximally efficient in achieving one's goals. However, achieving one goal may conflict with achieving another, and no external standpoint from which to evaluate rationally different systems of goals is available. Let us try this approach. Usually, to be rational, a set of beliefs, desires, intentions, decisions or actions must match each other in some way. A person's putative beliefs, for instance, must match the person's desires and decisions, or else they cannot qualify as the individual's beliefs. The same is true the other way around for desires, decisions, etc. (Chemiak 1994, 526). Now, when we evaluate the rationality of a belief, a desire, a decision, an action, a person, we do this by means of the only standards available to us. In a trivial sense these standards are identical with our own current ones (Chemiak 1994, 530). However, the implications are not trivial: it does not make sense to speak of comparing, or coming to agree on, ultimate common standards of rationality. In each case, it is always our own standards to which we must tum in interpreting other standards.

Rationality, Warrant and Reflective Equilibrium

107

Stressing rationality as coherence means to consider the thoughts, beliefs, intentions, decisions and actions of others to be coherent. This looks like an instance of applying the principle of charity. The term, however, is misleading, since there is no alternative if we want to make sense of the attitudes and actions of the agents around us (Davidson 1994, 232 f.). As far as I can see, this does not have to lead to relativism. To see this, let us relate the whole discussion of standards to the idea of problem solving. Although, obviously, there will always be different opinions, we find some solutions to be better than others, and can even argue for them. This is true irrespective of whether in some situations experience means only perceptual experience or includes also moral and existential experience. Communication about the different solutions may in tum lead to the modification of our standards, although the standards are always our own. This type of thinking is neither arbitrary nor merely relative, and it has implications for how one thinks of warrant. I will start with presenting some variations of the received view of the concept of warrant. My aim is to show that, although there is a difference between rationality and warrant, to some extent warrant, too, is person relative.

5

Warrant

Traditionally, the aim of a theory of knowledge is to discern and explain what distinguishes knowledge from mere true belief. According to Alvin Plantinga, the name for this quality or quantity is 'warrant' (Plantinga 1993a, VI and 3). However, what is it precisely that produces warrant? Let us assume a debate in which a questioner does not only ask for evidential support for a claim; the questioner also asks for statements showing the general ways of arguing. Attention is not longer focused on the material grounds presented as the support needed for establishing the initial claim; it has shifted from the question whether there is sufficient evidence for the claim to an inquiry about the general rule or procedure which makes the step from the grounds to the claim a trustworthy step. A statement that has the effect of authorizing the step from the grounds to the claim is a warrant (Toulmin, Rieke and Janik 1979, 43 f.). Obviously, this widens the meaning of Plantinga's use of the word 'warrant'. According to Plantinga, warrant is truth conducive in the sense that it turns mere true belief into knowledge. However, we try to authorize such a step from grounds to claims not only, for instance, in the natural sciences. There, general laws of nature as systematization of perceptual experience fulfil the function of warrant. When it comes to judicial contexts, statutes, precedents and rules do the same. In medicine, diagnostic descriptions do the same

108

Eberhard Herrmann

(Toulmin, Rieke and Janik 1979, 53). Similarly, when we apply the method of reflective equilibrium in situations of deliberation, we give the function of warrant first to moral principles. However, those principles are not important as principles; they are important because they systematically summarize intuitive moral judgements about concrete cases in the past. The crucial point is that it depends on our ability to have a moral experience to what extent we consider the concrete cases to be moral cases. Therefore, we cannot summarize intuitive moral judgements if we do not have a moral experience involving emotions. Well, a critical voice might object, this only shows that warrant conceived as moral experience systematized in principles is person relative. It is at least still an open question whether warrant in the truth conducive sense is person relative. Using some ideas of pragmatism, I want to argue that warrant in the truth conducive sense is person relative as well, while I still accept that there is a difference between rationality and warrant. One main idea in pragmatism is to conceive of truth as warranted assertability. Charles Sanders Peirce sees an analogy between human belief and animal expectation and between human inquiry and animals' exploration of their environment. Inquiry begins with doubt as the result of some belief habit being interrupted by challenging experience. Doubt is thus involuntary and unpleasant. Inquiry is a homoeostatic process by which the organism strives to return to equilibrium, a new habit, a revised belief (Haack 1992, 352). John Dewey's pragmatic idea of warranted assertability presupposes this view. In this view, knowing is a kind of doing, and the criterion of knowledge consists in warranted assertability. Knowledge does not manifest itself as a kind of recording facts, with knowledge consisting in the correspondence of our beliefs to those facts. On the contrary, Dewey views knowing as a constructive conceptual activity that anticipates and guides our adjustment to future experiential interactions with our environment. Concepts are instruments for dealing with our experienced world. They have their function in relation to our experienced world and not to the world as it would be, independent of human experience (Delaney 1995, 199). Dewey does not deny that truth is the correspondence of thought with existence. What he does deny is that it is correspondence between thought and things-in-themselves. Truth is a matter of mutual adjustment of our ideas and of adjustment of them to problematic situations. Not surprisingly, Dewey prefers to work with the concept of warranted assertability (Haack 1996, 651).

Rationality, Warrant and Reflective Equilibrium

6

109

Internal pragmatic realism

In this section, I will try to show the relevance of pragmatism for the method of reflective equilibrium by choosing Hilary Putnam's approach (1983). First, I will present some features of internal pragmatic realism, which are similar to those put forward by Putnam. Second, I will discuss the role our emotions play in conceptualizing values. Finally, the implications of the two steps will enable me to clarify the role of moral experience in the method of reflective equilibrium. Internal pragmatic realism accepts the concept of warranted assertability. This concept presupposes the idea that reality is always conceptualized reality, along with the idea that problem solving plays a crucial role in our lives. Pragmatically speaking, this gives, for instance, well-tried experience in everyday life and science the function of providing us with knowledge. However, there are different conceptualizations according to different kinds of problems in our Ii veSt Thus, scientific conceptualizations are indispensable, but they are not the only ones. We also need conceptualizations of values, a question to which I will return later. Internal pragmatic realism as realism is about what we humans can conceive of as ideal verification. It is in this light that we have to see Putnam's so-called idealization theory of truth: 'To claim that a statement is true is to claim that it could be justified.' If a statement as well as its negation could be justified, 'even if conditions were as ideal as one could hope to make them, there is no sense in thinking of the statement as having a truth value' (Putnam 1983, 85). Instead of 'justified' we could also say 'warranted'. Thus, to warrant a statement or its negation is to consider circumstances to be as ideal as, within our human situation, we could hope to make them. They are the frames within which the process of giving warrant could take place. Obviously, giving warrant to a statement or its negation by considering circumstances to be as ideal as we could hope to make them, does not work everywhere. It works in the sciences and in relation to well-tried experiences in everyday life. There we have developed procedures related to observational evidence by means of which we can or cannot warrant our knowledge claims. From our human perspective of ideal verification, the only criterion for what is a fact is that we have warranted assertability. Truth, rationality and warrant are still separate notions. Nevertheless, they are interdependent notions. Assume that we humans can look into the future and let us assume a person who knows that he or she will be involved in a car accident if driving a car on a certain day next year. Certainly, the person will leave the car at home that day. In a world in which we know about the future, notions such as 'rational', 'warranted' and 'true' will probably have another meaning than they have in

110

Eberhard Herrmann

our actual world. People who can look into the future probably evaluate certain things different from people who do not have this ability, since their ability causes other needs. Probably, for them 'truth' does not relate to reality now but to reality at the end of time. Probably, for them 'rational action' means an action that does not try to hinder what will happen in the future. 2 In other words, if we and our surroundings would be different from what we actually are, probably many things would change. Our needs, value judgements and concept of truth would be different. As it is now, we do not know what will happen. Furthermore, our decisions, based on interests and value acceptance, are among the causes that influence the chain of events. Therefore, we have to ask which conceptualizations, Le. both statements about facts and judgements about values, are acceptable. Both perspectives are necessary since the method of reflective equilibrium is about moral judgements and morally relevant facts. As it is, we are beings who cannot have a view of the world that does not reflect our interests and values. Simultaneously, we are committed to regard some views of the world along with some interests and values as better than others. The question is whether it is possible to discuss which of them are better than others. As far as I can see, there are criteria for such a discussion but only if we accept the role our emotions play in creating and sustaining values. Emotions in the sense of phenomena such as joy, sorrow, hope and fear are always the joy, sorrow, hope and fear of an individual. Certainly, we can have emotions without expressing them in language. However, to be able to identify an expression as the expression of a certain emotion even within oneself presupposes language, and language only exists in relation with other people. Let us consider emotions as part of a process that aims at the conceptualization of values. This process concerns the ideas we humans make of ourselves, our relations to other humans and our environments to cope with the concrete contingencies of life. What I will label 'views of life' are particular expressions about life and systems of such expressions. They are about what it means to be a human being living with love and happiness, suffering and death. These are obvious realities in our lives. However, we can neither completely understand and explain them nor explain them away. In the process of systematization some expressions can condense and, finally, we get views of life in the form of institutionalized world views or ideologies. As far as I can see, the main function of views of life concerns the concrete contingencies of our lives. Views of life provide us with conceptions, pictures and stories by means of which we can experience a fundamental tension. It is the tension between how life could be at its best and how defective it really is. 2

I want to thank Karin Johannesson for making me aware of this point.

Rationality, Warrant and Reflective Equilibrium

111

It is on account of this tension that, by means of provided conceptions, pictures and stories, we can develop a feeling of good and evil, right and wrong. On account of the conceptions, pictures and stories provided we can conceptualize our different emotions and thus create values. Without expressions of the contingencies of life we would not be able to experience the tension between how life could be and real life. We would have no more than pleasant and unpleasant sensations. Without this experienced tension we would not be able to identify certain emotions, such as precisely joy, sorrow, hope or fear. Finally, without those identifiable emotions we would not be able to conceptualize what we conceive of as values. This view has implications for the distinction between facts and values. Certainly, there is a logical distinction between facts and values. However, this distinction does not mean that only facts exist and values do not. Although there is a logical difference between facts and values, there is nevertheless a connection between them. In certain situations the conceptualization of what we conceive of as values will not do. In addition, it is necessary to actually experience the relevant emotions to gain knowledge of the character of certain situations and to identify morally relevant facts. Let me give an example. Suppose that I declare racial discrimination to be among the worst things I know. Such a declaration has specific implications. If I am witness to a white person treating a black person in a way that manifests racism, then other people expect me not to show indifference only. If I remain indifferent, this could mean that my declaration was not honest, or it could mean that I am not able to correctly identify the situation. In the latter case, perhaps, I have read about racism but I have not developed an emotion of disgust as to racism. Without such an emotion I will probably explain the situation differently. Emotions are necessary to decide what actually is the case. As far as I can see, this is one of DePaul's reasons to introduce the following additions to the method of reflective equilibrium. It is not sufficient to fashion the moral and philosophical beliefs into a coherent system only by means of the beliefs one happens to accept. In addition, the inquirer has to consider different moral conceptions and to allow for his or her formative experiences. First, the inquirer has to reflect upon the experiences and activities that are apt to broaden the range of formative experience. Second, the inquirer must actually open him or herself to new experiences by simultaneously avoiding con~pting ones. The inquirer must actually engage in activities that, together with the new experiences, are likely to extend the formative experiences. Without such changes, there would be no refining of the reflective equilibrium. DePaul's proposed method of balance and refinement, when focusing on balance, still defends the coherentist component of reflective equilibrium. However, in

112

Eberhard Herrmann

addition there is also emphasis on refinement, Le. on developing the inquirer's faculty for making moral judgements (DePaul 1993, 173 f.). To clarify what it means to develop this faculty, DePaul sketches an analogy between visual perception and moral experience. We have to develop the faculty of visual perception if, for some purpose, we want to be able to make all the perceptual discriminations that can be made. For that, we need to have perceptual experience. It does not serve merely as additional input yielding a belief we add to the store of our beliefs already accumulated. The experience must somehow alter the nature of the perceptual faculty itself: 'With experience, it functions better, producing beliefs with more warrant or producing warranted beliefs where previously it produced either no beliefs at all or unwarranted beliefs' (DePaul 1993, 204). Something similar is true concerning a person's faculty of moral judgement. This faculty does not come to us in its final form - we must develop it. By thinking about how perception serves to warrant beliefs, we get some guidance in answering the related questions about moral inquiry. Perception works in this way only if we develop it. Moral experience warrants moral judgements only if we develop it. Actually, particularly experience from literature, film, music and art serves to alter the faculty of making moral judgements (DePaul 1993,207). According to DePaul, in moral inquiry the goal is complex, including both rational and warranted belief. Concerning the latter, he gives the following example of a case where we have an especially forceful impression of wrongdoing or evil. He mentions the evening news he recently saw about the aftennath of an Iraqi gas attack on a Kurdish village. He made his judgement in the face of a particularly powerful experience. Part of the experience that prompted him to judge the gas attack to be morally evil involved his emotions, particularly, of outrage and revulsion. However'4it is not just a question of emotions: 'Moreover, it seems to me that this aspect of the moral experience is epistemically relevant. Were I to have seen the films of the village and spontaneously judged the act to be evil, even though my emotions were not engaged in the way I have so poorly described, I would not be able to place the same sort of confidence in my judgment' (DePaul 1993,211). The warrant for the moral judgement is not some diffuse unpleasant sensation. It is a moral experience, i.e. an emotion related to values that in turn are the result of conceptually identified emotions. Therefore, if inquirers want to develop their faculty of making warranted moral judgements, they 'have to pay considerable attention to the nature and character of the moral experiences produced by this faculty. We can only expect that this attention will result in a high level of attunement of moral experience, and hence, moral judgment,

Rationality, Warrant and Reflective Equilibrium

113

to what are believed to be the morally significant features of situations' (DePaul 1993, 212).

7

Conclusions

We could say that the coherentist aspect of reflective equilibrium mirrors Quine's ideas about making some parts of a larger system adjust to other elements. The question of how to understand a particular part can be decided only against a very broad background of considerations. However, by making adjustments in relevant places, it is possible to interpret any given part in different ways. Each of these interpretations is coherent seen from the perspective of the suitably adjusted system in which it is incorporated. This is Quine's idea of ontological relativity (Quine 1969, 26-68). As far as theoretical issues are concerned it does not matter that different interpretations of a given part in a system may be acceptable. However, concerning practical decisions in moral issues different conclusions may exclude each other but there is room for just one way of acting. How to choose in such a situation? One way could be to build on moral realism claiming that values exist independently of us. Then the moral attributes of reality would decide which conclusion is the true one. To me it seems more reasonable to conceive of values as the result of conceptualization in which our emotions and moral experiences play important roles. As to the choice between different reflective equilibriums, therefore, a more promising way is to look for the reasons we can actually share in concrete problem situations. This way can be even more fruitful if more people get training in having moral experiences so that they will have not only rational beliefs but also warranted ones. Certainly, this gives us no guarantee for deciding which reflective equilibrium among several alternatives is the absolutely best one. However, it enables us to solve problems in a way which is acceptable for the people involved. Seen from this perspective, neither the original method of reflective equilibrium nor DePaul's method of balance and refinement is part of ethics as providing a grand-scale theory. Instead, philosophical reflections about reflective equilibrium, its strength and its limitations, are part of ethics as a practice concerning how to make deliberations as rational as possible. If they are to be rational for more than one person, they have to include warranted beliefs. They become warranted through moral experience developed not only by philosophical reasoning but also by way of literature, film, music, art and by expressions in our different views of life. At the start of my article I mentioned the tension between coherentism and foundationalism. Perhaps we should not conceive of them as mutually exclusive

114

Eberhard Herrmann

positions but as aspects pointing to limited functions. When using the term 'balance', DePaul emphasizes the coherentist aspect. However, I hope that my epistemological and ontological considerations have shown why there is a further need of what DePaul calls 'refinement'. Concerning theories about the world we need an opening towards empirical experience and its conceptualizations; concerning moral reflections we need an opening towards moral experience and its conceptualizations. I hope that my epistemological and ontological considerations have shown, furthermore, that there are reasons to prefer a person relative perspective to achieve such openings. Although concerning questions of knowledge as well as concerning questions of value there is a distinction between rationality and warrant, in a certain sense both rationality and warrant are person relative.

The Thinker and the Thinking Process: A FeIninist Perspective on the Moral Faculty Marian Verkerk

1

Introduction

In his book Balance and Refinement, Michael DePaul argues that 'reflective equilibrium seems to represent the attitude that since the experience of any ordinary adult will do for moral inquiry, there is no need to describe in any detail the role of experience in moral inquiry' (DePaul 1993, 139). Reflective equilibrium leaves, so to say, the thinker out of the thinking process: any adult will do. But is this· so? DePaul answers the latter question with a definite 'No' . In the moral reasoning process the character and experiences of the person should not be neglected. While the emphasis DePaul puts on the importance of the thinker in the thinking process can be considered to be unique in the Anglo-Saxon literature, feminist theorists have for several years already stressed the specificity of the knowing subject. Studies of feminist epistemology started with stressing the gender of the thinker in particular. They used terms like 'women's experience', 'women's way of knowing' or 'women's knowledge' . By now, 'women' as a category has lost its credibility. It has become clear that gender hierarchies are not the only ones that influence the production of knowledge; race, class, sexuality, culture and age also play their role in the production of knowledge. In fact, gender identity cannot be adequately understood - or even perceived - except as a component of complex interrelationships with other systems of identification and hierarchy (Alcoff and Potter 1993, 3). These insights have led feminist theorists to the so-called standpoint epistemologies. All knowledge, including moral knowledge, is to be interpreted as situated (moral) knowledge. This means, among many things, that the thinker as well as the thinking process should be positioned in the social. and political context. In this article I want to investigate what feminist epistemology can contribute to the discussion DePaul started around the thinker in the thinking process. In particular I want to relate certain insights gained in feminist moral epistemology to further developments of the method of reflective equilibrium

115

w: van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, ©

1998 Kluwer Academic Publishers. Printed in the Netherlands.

115-126.

Marian Verkerk

116

in moral inquiry.1 I will defend the thesis that the suggestions DePaul gives relating to the importance of the thinker in the thinking process are not sufficient. By emphasising the psychological state of the thinker and thereby abstracting from the thinker's social identity, DePaul's suggestions remain ineffective.

2

Formative experiences

Let us start with considering the following example: Mary is Jack's wife. She is completely devoted to her husband and derives happiness from her deference to him. In fact, Mary defines herself in terms of her subordination to Jack. She is proud to subordinate herself to her husband and derives much of her happiness from the fact that she serves him well. In fact, she believes that her life is in perfect order. She would not choose her life to be different. 2

Would we say that Mary's judgement related to her situation is rational, that she has rational beliefs about her life?3 The most straightforward answer would be: Yes, if Mary considers her beliefs to be in reflective equilibrium. Now, suppose she would say so. We, as outsiders, would still feel some hesitation in affirming that Mary's judgements are rational. Probably we would like to say that she is not justified in subordinating herself and that Mary needs to be more informed about her situation. Mary needs to have a more adequate view of her life. If she were fully informed, she would judge her situation differently. In this case we explain the epistemic deficiency of Mary's beliefs by pointing to the inadequacy of some of her experiences. Experiences can form or alter a person's moral beliefs. One type of experience is called 'reasoning experiences' by DePaul. It involves such things as constructing an argument from propositions one already believes, appreciating a counterexample to a moral principle one accepts or seeing a proposition as obvious upon considering it (DePaul 1993, 141). In the case of Mary we can have doubts about the adequacy of the reasoning experiences she has had so far. Perhaps it shows that Mary is

2 3

So far few attempts have been made to relate feminist theory to more traditional Anglo-Saxon theoretical views. Two anthologies are of particular interest here: Antony and Witt (1993) and Alcoff and Potter (1993). This example is derived from Thomas Hill (1973, 87). A so-called subjective, internalist conception of rationality is presupposed, according to which believing rationally is a matter of believing what, from the believer's own perspective, is likely to be true or would seem to be true upon reflection (DePaul 1993, 102).

The Thinker and the Thinking Process

117

a woman who was raised in surroundings which offered her only some experiences, but not all. She has thought so far that it is a woman's destination to serve her husband well. Her mother and grandmother did, and they seemed to have been quite happy. In fact, Mary knows no better and that is why she thinks her beliefs are in perfect reflective equilibrium. But at the same time it can be shown that Mary lacks important information about other ways in which a woman can live. And by confronting her with new factual information, she may change her beliefs. Let us suppose that Mary has gone through the reasoning experience again. She is informed about other life-styles, perhaps she even has had a course in feminist ethics. But still she is convinced that her judgements about her situation are in perfect reflective equilibrium and that she has no reason to alter her beliefs or to consider them irrational. Is there anything else we can do to show that Mary is not justified in having the beliefs she has? Here DePaul makes a new contribution to the method of moral inquiry. DePaul states that the method of reflective equilibrium needs to be supplemented by another sort of experience in order to account for the epistemic deficiency of one's beliefs. Next to reasoning experience, DePaul introduces the notion of 'formative experiences'. Formative experiences can lead to an alteration of beliefs, but do not fit the mould of reasoning experiences. Certain activities or experiences - for instance actually performing an action someone considers right or wrong, living with a person who has a certain virtue or vice, or reading a novel - can alter people's moral beliefs. Other than the reasoning experience, the formative experience suggests a change in one's moral perspective. It can influence one's faculty of moral judgement, intuition or perception, so that one sees situations in a different way and feels drawn to make different moral judgements or accept different moral principles (DePaul 1993, 146). Now, on the one hand we have seen that Mary's beliefs can be considered to be in reflective equilibrium and, therefore, they may seem rational. On the other hand, we would like to say that Mary's beliefs are epistemically deficient. But the method of reflective equilibrium with its emphasis on reasoning experience cannot show the epistemic deficiency of Mary's beliefs. In fact, they seem to be perfectly rational. Introducing the notion of fonnative experiences, we can account for the epistemic deficiency. Mary's problem is not that she is ignorant of the relevant facts, but that somehow they do not strike her in the right way, and, as a result, that she does not respond to them as she should. DePaul calls Mary's problem a problem of naIvete. The problem has more to do with the person than the circumstances. The naIve person did not have certain formative experiences and, as a result, does not make any judgement about such

118

Marian Verkerk

experiences, nor, more importantly, are the judgements she does make influenced by these experiences (DePaul 1993, 154). The question which remains to be asked is whether Mary's beliefs are not only to be considered epistemically deficient, but also irrational. She would be acting irrationally if it could beshown that she did not do all that she could, as seen form her own point of view, to ensure the truth of her view; that is, subjecting herself to some formative experiences, which might bring to her new judgements. Mary may have been acting irrationally so far if she avoided the formative experiences for no good reason. But suppose Mary were to say to us, 'I just do not want to go through those formative experiences that may lead to an alteration of my judgement, because in doing so, I will risk a divorce. What is more, I think that divorce is a morally bad thing to do, because God forbids it.' In that case, DePaul would say that Mary is under no obligation to set about exposing herself to the formative experiences in question (DePaul 1993, 167). Remembering the subjective conception of rationality, it would even be irrational for Mary to set about acquiring an experience she believes to be corrupting. And so, in DePaul's outlook it seems to be possible that someone can have naIve, but rational beliefs. To sum up the discussion so far: When assessing the rationality of someone's beliefs, it is not enough that moral inquiry is restricted to reasoning experiences as is the case with the method of reflective equilibrium. Moral inquiry should also require that inquirers consider the nature of their formative experiences. In fact, DePaul is very explicit in saying that we need to give up the idea that.we can account for the epistemic status, specifically the warrant, of the beliefs an inquirer is led to by a method of inquiry primarily in terms of features of the method. Instead, the various features of the inquirer play the most significant role in accounting for the epistemic status of the beliefs accepted (DePaul 1993, 19.1). By introducing formative experiences into the process of justification, DePaul criticises the dominance of the so-called intellectualist moral epistemology, Le. our moral beliefs as being warranted in virtue of (i) the transfer of epistemological status by argument and (ii) the augmentation of epistemological status by systematisation. In fact, this intellectualist moral epistemology cannot account for naYvete (DePaul 1993, 197). This is so because the problem with a nai've moral judge is not that there is any information she lacks, or that the system of theories and concepts that she has built up is somehow inconsistent, or even that there is some alternative theory that she has failed to consider. The problem is as DePaul says 'that a significant number of the beliefs she is using to develop her moral theory are the products of a faculty of moral judgements that has not been used enough, or been used on a sufficient number of different types of situations, or objects, or perhaps, been used on the right kinds of situations, or objects, for it to have developed to the

The Thinker and the Thinking Process

119

point where it is competent to make the sorts of discriminations it is being used to make. The result is that the person's moral judgements are not warranted.' (DePaul 1993, 205) Literature, experience and even philosophy can serve to alter the functioning of our faculty of moral judgement for better or for worse (DePaul 1993, 207). And so DePaul concludes: 'We will think of moral inquiry as involving a roughly equal blend of the individual inquirer developing and improving her moral sensibility, and the individual seeking to examine and revise the various judgements her refined sensibility produces and to construct a theoretical account . that systematises these judgements.' (DePaul 1993, 214)

3

Emotional puzzlement

Now let us change Mary's situation somewhat: Mary has become a feminist. She has worked on changing her values and self-conception. She now regards her old values and self-conception as oppressivel She has become a successful professional for whom being a member of her respect-worthy profession is central to her self-conception. Among the standards to which her evaluative self-respect is staked are those she rightly takes to define quality in her profession, and she lives in accord with them, earning the respect and admiration of colleagues whom she, in turn, respects and admires. Mary thus lives congruently with reasonable standards that are central to her normative self-conception and so has good grounds for evaluative self-respect. Yet she cannot feel the worth of what she does and is. She cannot feel proud of herself or take pleasure in her accomplishments or feel satisfied with her life. Mary is harshly critical of herself, dwells incessantly on her failures, feels that her screw-ups give a better picture of her than her so-called successes and fears the inevitable unmasking of her mediocrity. Mary's emotional experience of herself testifies to a lack of evaluative self-respect. At the same time she knows that she deserves to take pride in her accomplishments and that she lives with self-acceptance. She believes she is respectworthy and regards her lack of self-acceptance as ungrounded and disrespectful of herself. 4

It seems something has gone terribly wrong with Mary. In former years, when she was happily married, she was confident of herself and considered herself to be a self-respecting person. Later on, having enriched her experiences, she

4

This example is derived from Robin Dillon (1997, 233).

120

Marian Verkerk

now feels miserable, although she knows she has no reason for it. What has gone wrong? The way Mary feels seems to be completely out of line. In fact, she is irrational: there is no reason for it and Mary would agree herself. She knows she has no reason for it, but still she feels this way. She believes she has all the reasons in the world for feeling self-respect, and yet she does not. She considers her views of her life even objectively justified, but still her emotions say the opposite. How can we make sense of Mary's situation? Are her emotions simply to be regarded as something irrational that should be dismissed? Remember that Mary, in DePaul's terms, has had many formative experiences and so she can count on her moral faculty. Remember also that Mary's beliefs about her situation are now more reliable and justified than the beliefs Mary had when she was still married. She is an enriched or an even more autonomous person than she has ever been. She is not the naive person anymore that she was years ago. It seems that we need some alternative account by which we can make sense of Mary's peculiar situation. In the following it will be shown that DePaul's emphasis on having a certain enriched moral faculty is not enough to make justified moral judgements. A certain self-concept is also needed. In fact, an adequate moral faculty presupposes a self-concept, which itself is socially and politically embedded.

3.1 Experiential understanding Robin Dillon, who in one of her articles discusses situations like the one Mary is in, starts with saying that self-respect is surely one of the morally interesting and personally significant dimensions of human life. Individuals who are blessed with a confident respect for themselves have something that is vital to living a satisfying, meaningful, flourishing life (Dillon, 1997, 226). Self-respect is not a discrete entity but 'is rather a complex of multiply layered and interpenetrating phenomena that compose a certain way of being in the world, a way of being whose core is a deep appreciation of one's morally significant worth. In short, self-respect is an appreciative mode of being toward and with oneself and the world with regard to one's worth.' (Dillon 1997, 228) On the standard account, the emotions of·pride, shame and resentment enact recognition and evaluative self-respect or its loss under certain circumstances. However, this account is not correct as is shown in its inability to explain certain familiar cases of damaged self-respect such as Mary's (Dillon 1997, 232). In fact, Mary has no reason to feel any shame. And so, to make sense of Mary's situation, Dillon introduces another distinction, Le. the distinction between intellectual understanding and experiential understanding. Intellectual understanding involves having beliefs which one has reason to accept as true, then coming by inference to have other beliefs which one takes to be true in virtue of their

The Thinker and the Thinking Process

121

logical relation to warranted beliefs, where the believing, inferring and assessing need not engage emotions. Remember that DePaul would call this kind of understanding the understanding involving the reasoning experience or the reasoning in the intellectualist epistemology. Experiential understanding involves experiencing something directly and feeling the truth of what is experienced. (Dillon 1997,239). Some understanding is explicit (propositional knowledge) and some understanding compromises unarticulated presuppositions implicit in certain ways of being in the world. That is, our ways of being in the world involve multiply layered sets of presuppositions that constitute nonpropositional frameworks of interpreting the world. Now, according to Dillon, and I think she is right here, unqualified self-respect requires appropriate experiential understanding. It seems that Mary lacks precisely this experiential understanding of her situation. In one way she has all the information she needs to have for leading a respectful life. But still she cannot see it the way she is supposed to. She does not feel it this way. To a certain extent this kind of experiential understanding can be compared with DePaul's emphasis on having a certain moral faculty, which in its turn presupposes having certain formative experiences. This can be illustrated by an example DePaul himself gives when explaining the significance of including formative experiences in models of moral inquiry. At a certain point DePaul compares our faculty of moral judgement as analogous to our faculty for making complex perceptual discriminations. Raised in a suburban setting, DePaul had not had many experiences with horses. And so he could not tell the difference between cantering on the left lead or on the right lead. The problem was not that he did not understand the concept of a lead, nor that he lacked any information. In fact, in one sense he could see everything he needed to, but at the same time he was unable to 'see' what he had to in order to apply the concept. After some experiences watching horses, he got to the point where he could immediately see a horse's lead and so he acquired the ability to 'see' something that he had been unable to 'see' before (DePaul, 1993, 203). This 'seeing' can be compared with the experiential understanding Dillon is speaking of. Going back to Mary we could say that, in tenns of DePaul, Mary's moral faculty is still not really altered by the formative experiences she has had. Although she has changed her life and she has had all kinds of experiences, it seems that her judgements are the products of a faculty of moral judgement that has not been developed enough to the point where it is competent to make the sorts of discriminations it is being used to make. She only thinks that she has grounds for having self-respect, but she still does not have the experiential understanding of what it means to apply the concept of self-respect. Her emotional puzzlement is only a case of not yet knowing what it means to be a self-respecting person. In line with DePaul's argumentation, Mary probably

Marian Verkerk

122

needs some more experiences in her new life, she needs to get more accustomed to her new way of living and after a while she will have got rid of these silly emotions. It is just a matter of time and confidence. But is it? Suppose Mary keeps having this emotional puzzlement. Following DePaul's approach, the only thing left to say is that Mary is unfortunately unconfident of herself, or confused. But then again, situations like that of Mary are quite common, especially for women. It seems to me that another analysis of Mary's situation is needed. Although DePaul seems to be right in stressing that moral inquiry should also make room for the development and improvement of moral sensibility, there seems to be something missing. What is missing seems to be a certain concept of the self. Moral inquiry should not only reckon with the fact that persons need enriched fonnative experiences in order to have reliable moral beliefs, but also with the fact that persons should understand themselves in a certain way. The only reason why DePaul attaches importance to having formative experiences or nonpropositional knowledge is that the moral judge is able to change her beliefs to more reliable ones. In this article I want to claim something more. The importance of having formative experiences and thereby having personal transformations lies in the fact that they can change the interests and desires of an individual in ways that we would say are good, or are bringing the person a thicker sense of autonomy and allowing 'her to flourish. In the following section I will elaborate on this.

4

Feminist epistemology

In philosophy, epistemology is traditionally regarded as a theory of knowledge in general, and it rests on the premise that such a general account, justified a priori, is possible. Feminist epistemology has precisely questioned this premise. It is through neglecting the context in which a theory is formed that a general or universal account of the nature and limits of knowledge has always been thought possible. But as soon as social context and the status of inquirers are included in the account, one can have some scepticism about the possibility of such a general and universal theory of knowledge. The politics of knowledge, the impact of the social status as well as the sexed body of the knower upon the production of knowledge are issues on which feminist epistemologists dwell. In short, there is no a priori knowledge, only situated knowledge. It is important to stress here that feminist epistemology does not say that situated knowledge is only gendered knowledge. The project of feminism is more inclusive than a focus on gender alone permits (Alcoff and Potter 1993, 4). It rather concentrates on the situatedness of the knowing subject; a subject whose identity is a multiple-layered one. Although one cannot speak about the feminist

The Thinker and the Thinking Process

123

epistemology, the common denominator seems to be the stress on the particularity and specificity of the thinker in the thinking process. A thinker who is always situated in certain social and political relationships. Feminist theorists claim that in every understanding, be it social, moral or otherwise, an understanding of the thinker herself is presupp~sed. Susan Babbitt makes this point very clear with the following story. When she was living in Italy, she failed to learn to make fresh pasta even though she was shown in detail many times and had a great desire to learn. She knew all the steps but did not understand the procedure. One main reason for this, she presumed, was that she had a concept of people who make pasta from scratch as being unusually competent in culinary affairs and of herself as not being such a person. Had she become integrated enough to be able to see making pasta as part of what one does to get along, she should have understood better. This example very much resembles the one DePaul used in comparing the moral faculty with having a certain perceptual capability. Also, Babbitt stresses the importance of having a certain experiential or nonpropositional understanding in (moral) inquiry. She intellectually knew how to make pasta, but still she failed. Just like DePaul who, on the one hand, knew the difference between cantering a horse on the left lead and on the right lead, but who, on the other hand, could not see the difference. Still, Babbitt gives a different account for her failing to make pasta. It is not that she did not have the capability per se, but that she had a concept of herself that did not enable her to make pasta. Her self-understanding itself prevented her from becoming a good cook. As Babbitt herself says, the self-concept, including understanding of social positioning, is something that people are often unaware of, but which is assumed in the interpretation and application of much information. DePaul probably saw himself a town-boy and not as someone from the province. He simply could not see himself as someone who rides a horse easily. The fact that people can fail to understand or be motivated by information that they have and believe can be explained by pointing to having a particular self-concept; moreover, to a concept of the self whereby this self is always situated in a social and political context. Let us go back to Mary. Mary is emotionally puzzled because she simply cannot feel herself as a serf-respecting person. In DePaul's approach, there are only two options left for Mary, as we have seen. Either she continues working on herself, developing her moral sensibility or we conclude that Mary's psychological state is confused. By stressing the situatedness of the concept of the self, another account of Mary's situation comes up. Mary's situation is not explained by saying that she is still not confident enough of herself, but that she has a concept of herself that is not worthy of self-respect. In fact, Mary's situation is perfectly understandable. On the one hand, she has a correct intellectual understanding of herself. On the other hand, she has an experiential

124

Marian Verkerk

understanding of herself as a degraded and worthless person. As Dillon states correctly, self-respect depends on interpretative self-perception and, what is more, self-respect itself is a mode of normatively interpretative perception of self and worth (Dillon 1997, 243). The source of damage to self-respect is an implicit interpretative framework of self-perception whose organising motif is worthlessness. This framework is not a private phenomenon but is a feature of the historical and socia-political situatedness of individuals. To say that selfrespect is a socia-political construction is to say not only that it develops and plays out against the backdrop of social and political contexts but also, and more importantly, that it is constituted by and reflects prevailing forms of social and political life. In short, socia-political factors have particularly profound effects on self-respect. To quote more of Dillon in this respect: Basic frameworks are constructed in the complex, emotionally charged interplay of self, others and institutions, which begins long before we are capable of conceptualising self, worth, persons, institutions and the relations among them and it shapes and delimits our conceptual schema (Dillon 1997, 244). To acquire reliable moral beliefs is not only a case of having the right psychological state, but also has to do with being socially recognised as a respectable human being.

5

The moral thinker as a self-respecting person

What can we do for Mary? Before answering this question, let me put in a nutshell what the above analysis amounts to. In reflecting on the thinker in the thinking process, we do not only need to pay attention to the moral faculty by which someone judges, but also to the social and political interpretative frameworks the thinker is in. Whether someone has rational or warranted beliefs is not only a question of having the adequate reasoning capacities (as DePaul would say), but also, and more importantly, a matter of her having a sense of her 'self' that would support a full sense of flourishing. One person can have an intellectual understanding of herself as a self-respecting person whereas she still experiences herself as degraded and worthless. For instance, in a society in which 'what is female is worth less' is a basic principle, forces of social subordination are powerful nl0ulders and injurers of basic self-understanding of women. This insight is not only related to situations women are in, but can also account for the degraded self-understanding the chronically ill or ethnic minorities may experience in society. Mary's emotional puzzlement was in fact a sign that her moral understanding of herself was not reliable. In fact, she was not justified in thinking that she was a self-respecting person and her emotional puzzlement told her so. If it is true - as DePaul says - that acquiring adequate moral reasoning depends on bringing about a more appropriate moral faculty,

The Thinker and the Thinking Process

125

the question of what is good for someone cannot be a question of what someone does with. the infonnation. Instead, to the extent that acquiring infonnation is bringing about a more adequate personal situation, the proper fonnulation is what the right kind of information does to the individual (Babbitt 1993, 262). Not only personal, but also social and political transfonnations are necessary to allow ideologically oppressed people to acquire new interests and desires and a new understanding of their positions. For Mary to attain self-respect, the cultural representation of women as worthless should be banished. Now what strategies are left for acquiring this flourishing self? Dillon states correctly in this respect that the simple-to-say solution of transforming society so that all individuals can grow up valuing themselves unconditionally, gives little practical advice for supporting self-respect here and now. But what we do know is that a more adequate interpretative framework should become operative. Dillon gives three suggestions: more loving relationships, a change from a self-obsessive mode to one that centrally involves attentive care might make valuing oneself a real possibility and active political involvement with others to eliminate or transform social forces responsible for deforming selfrespect can bring about situations in which more adequate normative identities and self-understanding are possible (Dillon 1997, 249). In the context of this article I will not elaborate on these findings. But what seems to be important to stress here, is that all three suggestions Dillon made point to some intersubjective understanding of what it means to be a moral subject. The moral reasoner needs to be recognised by others as a being worthwhile. I would like to end with the role literature can play in moral inquiry. In one of the last chapters of his book, DePaul presents some considerations for regarding literature as important in moral inquiry. In the first place, literature is ideally suited for supplying additional cases on which a person can exercise her moral faculties. Secondly, literature supplies cases that are sufficiently distant to minimise the effects of personal bias and the clouding of judgement by excessive emotion while still providing a realistic<enough case to engage our emotions to the degree that seems necessary for the poorer functioning of our capacity for moral judgement. And thirdly, by providing a less complex, partially interpreted scenario, a novel not only provides an example that might help a person to see how to sort out the salient features of a real-life situation for herself, but provides a manageable task on which to exercise a less than completely developed moral faculty (DePaul 1993, 20). All these considerations point to the fact that literature is supposed to be of help in developing a moral capacity or faculty; a moral faculty which itself is not problematised as a social and political phenomenon. How very different is the comment Diana Meyers (1993) has about the role of literature in moral inquiry. 'Literature has the capacity of rupturing familiar cognitive and emotional templates. Especially dissident

126

Marian Verkerk

speech capitalises on this virtue. It displays properties of people that have gone unremarked, and it projects possible constellations of their relationships that have not heretofore been appreciated. Dissident speech baffles routinesed thinking and bestirs improvisational thinking - receptive, generous attention to oneself or others; probing yet playful imagination; accurate yet synthetic interpretation.' (Meyers 1993, 38) In considering the thinker important in the thinking process, one should not only be interested in the psychological state a person is in, but one should account for the political aspects of defining and acquiring an adequate sense of personal integrity. A person's self should not have to be discovered through interaction and involvement - as if it were always there to begin with - but a person's self rather has to be brought about and discovered through her experiences of the effects of actual social and political change and by changes in the external circumstances (Babbitt 1993, 261). One may think that in this article I am pleading for some sort of objectively identified flourishing self as necessary for the rationality of judgements. This is not the case. Like DePaul, I think it is possible that there are naIve, but rational judgers. I also agree with DePaul that moral inquiry can only aid the inquirer in her quest for knowledge, but cannot guarantee truth. The importance of DePaul's contribution lies in his stressing that moral inquiry should also be concerned with refining the ability to make sensitive moral judgements and not only with fitting judgements together into a systematic and coherent whole. My only suggestion in this article is that in refining the moral faculty more is involved than simply refining the psychological state a person is in. In fact, refining the moral faculty means discovering a moral self that is socially recognised and respected. The emotional puzzlement Mary felt, was a sign that something was wrong. It was her emotions that told her so, nobody else did. And so if she had not felt this way, there would have been nothing left for us to say.

Wide Reflective Equilibrium and the Justification of Moral Theory Bo Petersson

1

Introduction

The theory of wide reflective equilibrium (WRE) as it was introduced by John Rawls and further elaborated by Nonnan Daniels in several articles (now brought . together in his Justice and Justification, 1996) is a far-reaching and ambitious project. In one respect, WRE is a method for solving moral problems; in another, WRE is a state of affairs, a result that is achieved by such a method - an equilibrium - that justifies a person's beliefs in certain moral ideas (the personal, subjective level, 'P is justified in believing X') or a certain ethical system or theory (the interpersonal or impersonal, objective level, 'belief X is justified'). In this short paper I will make some comments mainly on Daniels's version of the WRE theory and discuss some difficulties that I believe have consequences for the justification respect of the theory. (All references to Daniels in my text are to his Justice and Justification, 1996.) My first reflection (in section 2 ) deals with the domain of a wide equilibrium. In this aspect, WRE theory seems to be open to different interpretations, a fact which gives us difficulties in discussing justification. My second reflection (in section 3) deals with the concept of coherence. Daniels aims at maximal coherence in WRE to warrant justification. The concept of coherence is, however, complex and composed of different relational concepts. Therefore, it is not at all clear what is meant by 'maximal coherence' or when a set of beliefs has achieved such a state. Daniels's theory seems to deal mainly with the personal level of justification, Le. what a person is justified in believing, but his theory also aims at justification at the objective level (cf., e.g., p. 41, note 1). The problems I outline in sections 2 and 3 are relevant to both levels. Of course, the notion of 'justification' is in itself problematic and would need more elaborate discussion. My interest is in the objective level and in justification of theory. Shall we, e.g., presuppose that justification in such a case implies tluth? I do not think so. But I think we must presuppose that two theories that in some respect exclude one another cannot both be justified. We may have good reasons for both, they may both, e.g., give us a 'best fit' , but if they exclude one another in some respect they cannot both be justified.

127 ~

van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 127-134. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

128

2

80 Petersson

How wide is a wide reflective equilibrium?

A WRE is a state of balance between all beliefs that are relevant to the moral domain. It is, according to Daniels, 'a coherent ordered triple of sets of beliefs' consisting of (a) considered moral judgments, (b) moral principles and (c) background theories (p. 22, 48). This state of balance is achieved, e.g., by a person by playing off his different beliefs against one another, 'working back and forth, making adjustments to his considered judgments, his moral principles and his background theories. In this way he arrives at an equilibrium point that consists of the ordered triple (a), (b) and (c)' (p. 22). The equilibrium point is considered to be a 'best fit' or a 'strongest mutual support' (p. 10) among the components, meaning that a belief (principle, etc.) 'coheres best with the rest of our beliefs' (p. 61). No type of belief has an epistemological priority in WRE. The credibility of a belief is dependent only on how it is supported by or how it supports other beliefs (cf. DePaul 1993, 20). Many passages suggest that WRE is an equilibrium of a radically wide character, e.g. a totality of beliefs of all kinds: ' ... [W]e 'test' various parts of our system of moral beliefs against other beliefs we hold, seeking coherence among the widest set of moral and non-moral beliefs by revising and refining them at all levels' (p. 2). 'This means seeking coherence with all our other moral and nonmoral beliefs - all the other evidence we can bring to bear through empirical and philosophical argument' (p. 6, cf., e.g., p. 10). I will use 'TWRE' for this totality of beliefs. Daniels's way of characterizing background theories in the 'ordered coherent triple' also fits well with interpreting WRE as TWRE. Background theories are composed of both moral and non-moral components, the non-moral being of various kinds, such as 'empirical and philosophical' (p. 6, 8). Their function is to give constraints, reasons, support, explanation, etc., in choosing among and adjusting moral principles and considered moral judgements, trying to make them fit together. But background theori~s are themselves objects of 'adjustment' in this process. Daniels emphasizes 'how diverse the types of beliefs included in wide reflective equilibrium are'. He exemplifies with beliefs' ... about the nature of persons; about the function of ethics in our lives; about the implications of game theory, decision theory, and accounts of rationality for morality; about human psychology, sociology, and political and economic behavior ... ' (p. 6). In my opinion, it is impossible to draw a borderline between beliefs or kinds of beliefs that can or should be brought in and those that can or should not. If we ask why we should believe in a certain theory about the nature of man, we may in part rely upon some pure psychological theory and some 'facts' connected to that theory. But for the judgement that these latter beliefs are reliable, we

Wide Reflective Equilibrium and the Justification of Moral Theory

129

may depend on theories of a medical or physiological character. Where to stop? What will be a relevant non-moral belief seems to depend on how far we are prepared to carry on questioning. Once we have opened up the equilibrium to background theories including 'the broadest range of considerations and arguments that could count as evidence for our moral beliefs' (p. 6), there cannot be a limit that excludes certain kinds of belief. This consequence fits well with the idea that WRE deals with 'coherence among all our moral and nonmoral beliefs' , Le. that WRE is TWRE. However, this picture of WRE as a kind of total outlook on life and the world does not fit all of Daniels's statements about WRE. Some passages tell us that the non-moral beliefs should be 'relevant', which seems to allow us to exclude at least some beliefs when constructing our equilibrium. In explaining the method of working 'back and forth' and making 'adjustments', Daniels maintains that it should be looked upon as analogous to methods in science or scientific reasoning (cf. pp. 22,24, 32, 33, 38, 60, 61). In fact, science and our non-moral observations and 'observational reports' are here mentioned as constituting another reflective equilibrium. Daniels states that 'the credibility we assign to observational reports is itself based on inference from a nonmoral reflective equilibrium' (p. 32). He even speaks of a 'nonmoral wide reflective equilibrium' (p. 32, line 4, my italics). Thus, 'WRE' in these passages cannot mean TWRE, a totality of our moral and non-moral beliefs. It seems as if here we have a non-moral wide equilibrium, NWRE, and a WRE that also includes moral beliefs, MWRE, and that an analogy of method exists between these two equilibria. Evidently, such a division into different wide equilibria presupposes some distinction between what is 'moral' and what is 'non-moral'. Unfortunately Daniels never tells us how he would elaborate such a distinction. I think it also can be argued that no clear-cut distinction can be made between 'moral' and 'non-moral' at a fixed point (cf. Williams 1985, 129-130), which is what we need to separate MWRE from other equilibria. (Nor should the distinction be identified with the distinction between 'nonnative' and 'non-normative'. Daniels would certainly include some normative beliefs in his non-moral equilibrium, e.g. beliefs about what is a better theory and a better explanation.) But suppose we can make a clear distinction and that we can arrange our beliefs in a NWRE and a MWRE. If MWRE and NWRE are disjoint, one could perhaps argue that the beliefs in MWRE get their justification through the coherence of MWRE and that the same is true of NWRE and the beliefs in NWRE. But are MWRE and NWRE really two disjoint sets of beliefs? In the way MWRE is described by Daniels it must be possible to pick up one belief in NWRE - say of a simple empirical character - and use this belief as a part of a background theory in MWRE. (If MWRE and NWRE really were disjoint,

130

80 Petersson

MWRE would be a 'narrow' equilibrium, without 'contact with the world' just the kind of equilibrium Daniels intends to avoid.) Suppose there is a belief B that is justified through the coherence of NWRE, Le. the non-moral wide reflective equilibrium. Suppose B is also a part of a background theory in MWRE. But in MWRE all beliefs are explicitly open to mutual revision and refinement. What gives justification in MWRE - and to MWRE as a whole - is the 'strongest mutual support' or 'best fit' among the components of MWRE. What will happen if 'best fit' is achieved by revising B? Is B then both justified (in NWRE) and not-justified (in MWRE)? To avoid this problem, we can of course state that beliefs in MWRE that also belong to NWRE have a kind of epistemological priority in MWRE. But that would be a new principle in MWRE contradicting the idea that maximal coherence - 'strongest mutual support' - is the only way to get justification in MWRE. B then surely gets its justification from coherence - but from a coherence of another set of beliefs, NWRE. Perhaps then WRE should instead be interpreted as the all-inclusive TWRE, which includes 'all moral and non-moral beliefs'. How shall we handle the idea that, e.g., observational reports and ordinary sensory perceptions get their 'credibility' from 'a nonmoral wide equilibrium' (p. 32)? This idea means that some beliefs in TWRE should get their justification from some selected components, some subset of TWRE. The reason for that cannot be to achieve the 'strongest mutual support' in TWRE, which we have announced to give justification in wide reflective equilibrium. But if we take this latter principle seriously, not only do we have to give up the idea of the 'selected components', but we must also accept that considered moral judgements and moral principles may be reasons for us to change our opinions about and our descriptions of 'facts'. The support, constraints and adjustments must be allowed in all directions (cf. Sayre-McCord 1996, 20 and also note 20; cf. Daniels 1996, 27-29, 49, 61). This version of the TWRE is a radical concept not only because itincludes all possible beliefs but also because it admits moral reasons (considered moral judgements, moral principles) as arguments and constraints for descriptions of facts.

3

What does 'best fit' mean?

Wide reflective equilibrium is said to be found in a 'best fit' among all its different components. We should be looking for a state of maximal coherence. Daniels speaks of what 'best coheres' or what gives the 'strongest mutual support'. The components may also be 'adjusted' or 'revised' in order to achieve this 'best fit'.

Wide Reflective Equilibrium and the Justification of Moral Theory

131

In one way this thought seems both plausible and attractive. It is directly related to the idea that our beliefs about world, man and morality must hang together. There must be some 'convincing' interconnection and coherence among our different beliefs when we judge them to be justified. If they are not coherent, some belief must be unjustified. But what should we mean by 'best fit' and 'maximal coherence'? Already 'coherence' in itself has an unclear meaning. And Daniels does not give us any precise criteria for its applicability. A rough characterization of our 'ordinary use' may be this: A set of beliefs is coherent if and only if it is consistent (noncontradictory) and the elements in the set in some way give support to one another. In Daniels's discussion, expressions like 'be compatible with', 'be reconciled with' and 'constrain' are used for the non-contradictory element of the coherence relation and, e.g., 'support' and 'depend on' are used for the supportive element. However, a 'support' may be understood and also established in many ways. Daniels himself makes the following remarks: 'Coherence involves more than mere logical consistency. As in the sciences, for example, we often rely on inference to the best explanation and arguments about the plausibility and simplicity to support some of our beliefs in light of others' (p. 2). Thus, sometimes we are dealing with very strong relations between what gives the support and what is supported, e.g. an entailment relation. But we are also talking about support when the relation is an implication of a weaker kind or a different kind. The support relation may be of an inductive or a probability character. A belief may also be supported through different kinds of explanatory relations, etc. (cf. the quotation from Daniels above). Geoffrey Sayre-McCord emphazises that we can speak of different versions of coherentism 'as different evidentiaVinferential relations are countenanced as appropriate' (Sayre-McCord 1996, 152). In his own version, coherence is composed of what he calls 'evidential consistency, connectedness ·and comprehensiveness. The first, evidential consistency, sets a necessary and sufficient condition for (minimal) coherence, while the second and third, connectedness and comprehensiveness, serve, when present to increase the relative coherence of a set that is minimally coherent' (Sayre-McCord 1996, 166). If we take coherence to be only consistency, what Sayre-McCord calls 'minimal coherence', we will get the troubles that D.W. Haslett presents in his article 'What is Wrong with Reflective Equilibria?'. Where consistency (coherence) can be achieved by revising A or B, how do I know which one I shall revise and which resulting equilibrium is justified? 'There are, in other words, innumerable different wide reflective equilibria compatible with [Le.

132

80 Petersson

consistent with, my remark] any given set of initial background theories and considered moral judgements, each of these different reflective equilibria being the result of "pruning and adjusting" the initial theories and judgments in different ways' (Haslett 1987, 310). This is, of course, a correct conclusion. But neither Daniels nor SayreMcCord would regard consistency as sufficient for justification. Their concept of coherence also includes other components - the supportive relations. And what they aim at is not only coherence but a 'best fit', a 'strongest mutual support' or a 'maximal coherence' . Obviously, they presuppose that it is possible to speak of degrees of coherence, and they aim at the highest degree. SayreMcCord states: 'The more and the better relations, the greater the degree of coherence enjoyed by the set and the stronger the justification' (Sayre-McCord 1996, 152). How could we apply this to WRE? WRE - and specially in the all-inclusive TWRE version - consists of beliefs of most separate kinds: particular moral beliefs orjudgements, moral principles, background theories - e.g. about persons, man and society - empirical beliefs that are part of or are presupposed in background theories, theories about 'the world' that explain, e.g., some empirical beliefs or some other part of a background theory, etc. Evidently, all kinds of support exemplified above must be relevant - and certainly some not mentioned. We may agree that particular moral judgements must support our moral principles and that moral principles must support the particular judgements. But it seems as if the supporting relation cannot be the same in both directions. The supporting relation from a background theory to moral principles may vary, depending on what kind of background theory it is, what part of it gives the support, and what kind of moral principle we are dealing with. Empirical beliefs may give support to, e.g., a background theory about the nature of man in a way different to the wayan observation supports a particular moral judgement. What is 'maximal coherence' in such a complex set of beliefs? How shall we 'quantify' here? How shall we count 'more relations' and how shall We recognize the 'better relations'? (cf. Sayre-McCord above) 'Best fit' implies that we can compare the degrees of coherence in separate coherent sets of beliefs. Is it the 'strength' of the support relations that counts or is it their quantity? Is the quantity of beliefs included in the coherent set relevant? (At least according to Sayre-McCord it is.) How shall we judge (a low frequency of) mutual support against (a high frequency of) support only in one direction? Shall we prefer a set where some beliefs give support to a very wide domain of other beliefs like when we judge a better explanation to be the one that explains a wider domain of explananda? (Sayre-McCord mentions the problem of how to balance connectedness and comprehensiveness in order to attain an 'over all coherence

Wide Reflective Equilibrium and the Justification of Moral Theory

133

of a set' and admits that there is no algorithm for the solution, cf. n. 57, pp. 186187.) I think there is no easy way of making the discussion of degrees of coherence meaningful. Perhaps it is possible if there are only few elements in a set and we concentrate on one special supportive relation, e.g. entailment. But the wider domain and the more kinds of supportive relation we include, the more problematic the discussion about degrees of coherence becomes. WRE, however we interpret it, is such a wide set of beliefs that it must include several, different kinds of supportive relation. Therefore, 'best fit', 'strongest mutual support', what 'best coheres' etc. do not seem to constitute a useful concept for justification in or of WRE. When we explicitly discuss a certain person's set of beliefs, we could of course relate 'best fit' to some purpose or some feelings of his. The set that fits best could in such a case be the set of beliefs of P that is coherent and that gives P the highest degree of satisfaction, makes him feel happy, gives him a feeling of leading a meaningful life and understanding the world, etc. Of course we can say that such factors are relevant in explaining why P accepts one coherent set of beliefs instead of another. But should we say that P thereby also has good reasons for the one chosen, that he is justified in his choice? I do not think so.

4

Conclusion

I have tried to look critically at the theory of WRE and mainly at some aspects of Daniels's version of it. In section 3, one may even say that my account is 'hypercritical'. Of course, more broadly we can speak of a 'higher degree' or a 'lower degree' of coherence when comparing certain sets of beliefs. In many cases that would perhaps be enough for justification. I certainly believe that WRE theory as it was introduced, is a workable tool to make our nloral judgements and moral theories better, e.g. by excluding incoherencies. But what I insist on is that the concept of 'maximal coherence' or 'best fit' is not applicable when it comes to more systematic and detailed comparisons between coherent sets of beliefs. And that is what is needed if it is to be able to function as a key concept for justification in the theory of WRE. I also share the fundamental idea behind the theory of WRE that our beliefs about the world, man, society and morals must fit together into a continuous and connected whole. The question is how we can develop this into a theory that works. I would prefer this done within an all-inclusive interpretation of the WRE, i.e. a TWRE, in a kind of holistic way.

134

Bo Petersson

Even if Daniels puts some restrictions on what can be reckoned as a considered judgement and also requires a best fit or a maximal coherence for justification, we must conclude that his theory still is a type of coherentism. In his book Value Judgement (1996), James Griffin distinguishes between coherentism and holism. Holism asserts, according to Griffin, that justification is something only connected with the totality of beliefs. In holism we may, e.g., accept that some kinds of belief in a set have more weight than some other without assuming a foundationalism, Le. that some beliefs are certain and indubitable. Griffin states: 'What is impossible to resist is not coherentism, but holism. Holism is the thesis that justification comes only from the whole set of beliefs.' (Griffin 1996, 12) Perhaps Griffin's distinction between holism and coherentism can throw some light on the justification problem of WRE and give us some ideas as to how to modify Daniels's theory. So, what do we and Griffin mean by 'holism'? However, I will not go into this here.

Your Intuition or Mine? Inez de Beaufort

(... ) and there is no real escape from prior moral {intuitions' after all. Our rationalized desires are not a bedrock of morally neutral facts. (Daniels 1996,5)

A. Pregnancies in men? You mean with artificial wombs, big, bigger than male beer bellies, artificial hormones and Caesarians? What about the integrity of the male body? It is disgusting, unnatural and morally irresponsible. Imagine being the child born from a man's body. I mean, Pallas Athena was a God and at least she came from Zeus' head, not his belly. But an ordinary human child? Women should protest: what right do men have to take away this specific female role? Being pregnant is an exclusive female right! Isn't anything sacred anymore? B. Pregnancies in men? Sounds very exciting. What a wonderful opportunity for men. They will finally be able to share the experience of giving life. It will make men more nurturing, less aggressive. What a wonderful opportunity for women. It will enable them to share child-bearing and not only child-rearing. Equality at last. Think also of the possibilities for homosexual couples. A. But aren't you shocked or disgusted by the idea of going against nature? Men are supposed to father children, not to carry them. Male pregnancy is not only unnatural, it is the most unnatural situation I can think of. B. Now, don't exaggerate. I thought you were up to date regarding medical ethics. There are more, equally unnatural situations: think of Dolly, the cloned sheep, or of xenotransplants: a man having a pig's heart, plastic surgery, the operations on transsexuals, and what about ectogenesis or post-menopau·sal motherhood? Do you want more examples? Why should I be disgusted by the idea of going against nature? What makes nature sacrosanct? It seems that the word nature alone can evoke so many sentiments these days. You mention it, and people become ecstatic and ramble on about tropical forests, rare species, unhurried lives, underwater deliveries, whales, listening to oaks, swimming with dolphins, beetroot juice, and so forth. All very nice associations, well but the beetroot - I definitely prefer vodka, also very natural, to beetroot juice but in my view it does not prove in any way that going against nature is always wrong. Sometimes it is, sometimes it isn't. It depends on all kinds of other 135 MZ van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 135-142. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

136

Inez de Beaufort

aspects. The fact in itself that male pregnancies are unnatural doesn't get us anywhere in deciding whether or not to go ahead. It is, for example, quite unnatural to be conceived in a test tube, but that is no good reason not to accept reproductive technologies. Suppose, only for the sake of argument of course, that Robert, now that he has reached the age of sixty, considers dyeing his greying hair. The fact that dyeing one's hair is unnatural is no good reason not to do it. A. The colour of Robert's hair seems to me to be morally totally irrelevant as long as we are not discussing the issue of 'professional' secrecy of hairdressers or spouses. What I find surprising is that your intuitions about this are so very different from mine. B. That is because fundamentally you are a conservative person who is afraid to open his mind for interesting new developments. Your first idea is to say 'No'. A. Well maybe, but if I may return the compliment: you are a naive optimist whose first idea is to say 'Yes' because you are so easily seduced by the newness of developments, and that again has to do with your fear of being held for a conservative. And that obviously has to do with your very conservative father against whom you continue to rebel. B. Why bring up my father? It is like Robert's hair: totally irrelevant. Shall we stop this pseudo-psychological analysis? Do you know what the real problem is? The problem is that your firmly held beliefs on naturalness as a moral criterion are so strong that they will keep you from seriously looking into other arguments. That, I realize, is not a reproach one can hold against the method of reflective equilibrium itself. It is, however, a fact of many moral discussions that the role of beliefs or intuitions or whatever you wapt to call them is so strong that they are almost seen as sort of self-evident, and that it seems as if one is only looking for theories that support the initial beliefs. I think it is an admittedly rather elegant - form of jumping to conclusions. A. But surely one cannot hold the imperfectness of ordinary moral discussions as an argument against such a subtle and sophisticated method as the method of reflective equilibrium. That is unfair. It is like arguing that Escoffier's ideas on cooking are wrong because most people overcook their beetroot. B. Not beetroot again. Well, not altogether. It's more complex than that. The point is that in emphasizing the importance of moral beliefs, and in going back and forth between beliefs and theories, etc., they are given an important role.

Your Intuition or Mine?

137

I, however, fundamentally distrust my intuitive beliefs. Such a belief is at most a starting point for further thinking, something like being intrigued by a problem, or having a hunch, a vague idea of the direction to take. Nothing more. A. Yes, if I had your intuitions, I would be very suspicious of them too! B. What is wrong with my intuitions? Why are they less trustworthy than yours? A. I'm only joking. I take intuitions very seriously indeed, even yours. That is, after all, part of my method. But, seriously, there are probably many issues on which we would have more or less the same intuitions. You cannot base your criticisms on such extreme examples as male pregnancies. B. Yes, we may have the same intuitions on the very, very fundamental moral issues such as not lying, not killing. But the proof of the pudding is in the issues we don't agree on. And we cannot deny that we differ on the issue of male pregnancies, which you think is extreme, and probably on many other issues, particularly those that are at the centre of the medical ethical debate. A. It's like when you meet a person for the first time. You have a feeling: nice person; sexy person; horrible person; boring person. These impressions are the result of your previous knowledge, your life-long experience in meeting people. All those data are stored in your brain. It's not that you don't have to check. You may be mistaken from time to time. You know, like the supposedly exciting rock-musician who turns out to be a dull philosophy professor. But then you may be right in many cases. B. I'm not sure whether you can compare that. Even then I would be very suspicious of my first impressions. Too many instant likes and dislikes. And then there is the danger that it might make it more difficult for me to check my first impressions if a person seemed very unpleasant at first sight. Let's go back to our pregnant men. Explain to me why you think that men should be robbed of the hypothetical possibility of carrying a child. So far, you only came up with your nature intuition. A. Well, for instance, I feel it is better and nicer if there is a difference between men and women. B. There will be plenty of differences left. I mean, I'm sure men will keep their penises. And I don't think they will be able to breastfeed. Besides, pregnancy is not a life-long affair. It usually lasts nine months.

138

Inez de Beaufort

A. I wasn't thinking of that kind of changes; I was thinking of a fundamental barrier between the sexes that should not be crossed. B. Nature revisited! By the way, maybe I know why I react so strongly. In dealing with medical ethical issues I have been confronted with the 'yuk' feeling too often. And the 'yuk' feeling, often attached to the slippery-slope argument, is a very common first intuitive reaction people have to a new development. It is used as a shield in order to avoid considering the issue rationally, as an excuse not to argue about it. Just condemn at first sight. Like me and my unpleasant person. You have to put so much effort into convincing people that they should think about it properly and not confound real thinking about it with just finding the arguments that fit the 'yuk' feeling. A. I see, the trials and tribulations of the medical ethicist... But a first instinctive reaction like the 'yuk' feeling is not the same as a considered judgement or belief. A yuk feeling is far more primitive than an intuition. You have to distinguish between instinctive primary reactions and intuitions. You seem to associate intuition with negative beliefs. I'm perfectly willing to accept that in medical ethics you are often confronted with negative beliefs, but your narrow medical ethical horizon shouldn't cloud your views on intuitions in general. Intuitions are different. B. Tell me more about it. A. Well, they are more fundamental. Like a nagging feeling of 'this is unjust'. This is not to say that an intuition may not be accompanied by a yuk feeling; for example, my intuition about not tampering with nature is accompanied by my yuk feeling on male pregnancies. I'm not saying that they are always accompanied by the yuk feeling, but it happens. It can, however, also be the other way around: my intuition may be accompanied by a feeling of admiration or a sort of happiness: this is good. I haven't yet figured out why, but I have a good feeling about it. B. Alright, so there is a difference between important intuitions in the sense of beliefs and primitive gut reactions. You may say so, but some will say that there is some wisdom in our first reactions and that therefore we should start from there. How do you distinguish between your gut feelings and your beliefs? How do I know that a feeling is primitive and to be disregarded or a sign of old moral wisdom? That an intuition is appropriate or not? How do you distinguish between prejudice and intuition ? You dress them up, your intuitions. I was taught that ethics is about arguments, about good reasons. (By Robert by the way, before

Your Intuition or Mine?

139

the method of reflective equilibrium. I forget the name of that Swedish river he always discussed. Before animal ethics. Those were the days.) And that a good-reasons approach is sufficient. You can discuss them; some are better than others. Plain and simple analysis of arguments. In the course of my thinking I have often changed my mind (that is, my initial beliefs) on the basis of convincing arguments. What was wrong with that? That is, in my view, what makes ethics so exciting. I love to change my mind. I like to be dragged to a conclusion by the magnetic force of logic even if I don't like the conclusion itself. Arguments win. Intuitions lose. A. But if you say you have changed your mind, that means that even you have started with your first ideas: the mind you change? I intuit, you intuit, everybody intuits. B. Sometimes. I don't deny having intuitive beliefs. The problem is the weight they get on the balance of the equilibrium. That worries me. But in the case of male pregnancies I had no first ideas, I had never thought about the subject. Male pregnancies are not possible, probably never will be possible. It is just an interesting hypothetical exercise in arguing. That is why I decided to try to defend it. I didn't want to be stopped by some initial vague feeling of unease. What makes arguments so interesting is that they overrule your initial ideas. That is the whole point. Intuitions are claimed to be the ethical smoke alarms; if you listen to your intuitions, you may still be able to extinguish the fire. But in fact they are just used as stop signs. A. I keep wondering why you are so suspicious of intuitions. B. I suppose because people seem to have widely differing intuitions on the issues that matter. And because I think that intuitions are often subjective prejudices with no other real foundation but that they reflect the prejudices in your education and your time. A hundred years ago people had very different intuitions on, for example, the role of women, on homosexuality, and so forth. You say it is firmly held beliefs, I say it is firmly held prejudice. Also, I'm very worried about the conservative nature of our firmly held beliefs. We have felt this way for ages, therefore we have good reason to think that we are right. We have to show that by running to our theories and back to our intuitions, and so forth. Zigzag, zigzag. I know. You translate your feelings into arguments. Some will tum out to be good arguments. Some will end up in the wastepaper basket. I like to de-intuitize.

140

Inez de Beaufort

A. Yes, hence the reflective equilibrium. That is precisely the point. You are supposed to test and correct intuitions. That is the beauty of the method. Nobody says that you have to follow your intuitions without further analysis. You are right in saying that there is the danger of a certain conservatism. But name any theory without that danger. Not even your good-reasons approach. You can't do without intuitive beliefs. Imagine you were brainwashed into a state where you had no intuitions at all, a kind of moral tabula rasa. You wouldn't know where to begin, you wouldn't know how to check your arguments. They would be dangling in a moral vacuum. B. Well, for a change I do have strong intuitions on the nl0rality of brainwashing. But still I'm not sure. I mean, instead of starting from your intuitions you might as well say: There are usually two possibilities: to go ahead with something, say male pregnancies, or not to go ahead. I would think about the best alternative, considering all the arguments. There is no vacuum. A. How do you get your arguments in the first place? Where do your good reasons come from? I mean, why bother about an argument of autonomy or justice if you don't believe it is relevant? Compare intuitions with the clinical eye of the doctor: having all the relevant medical knowledge is not enough, you have to have this sort of medical intuition to be a good physician, this 'Fingerspitzengefiihl'. A complex sort of extra sense that is definitely not the same as a primitive gut feeling, that enables you to diagnose. Obviously, a doctor also has to check, do tests or whatever, and if the tests don't fit his initial idea, run more tests which mayor may not prove his first idea to be correct, but he will still use this diagnostic intuition. How do you explain the very well-known situation in ethical deliberations of someone who is divided between being convinced by arguments, say those in favour of male pregnancies, and who is, at the same time, unhappy and uneasy about the conclusion? Something keeps him from wholeheartedly agreeing. Something is bothering him. A very important ethical unease. A very different unease than a first reaction. I mean this is after having considered all the relevant arguments. Something is nagging him. Have you ever found yourself in such a situation? You must have, everybody has. B. Of course I have, though not in the case of male pregnancies. Definitely not. If I find myself in such a situation, I would say that either I haven't considered all the arguments deeply enough and my unease is telling me that I have probably missed an argument or something. I would go over the arguments again. Another possibility is that I am in a sense shocked by the conclusion, because, for example, it is new or extravagant or unfamiliar, and therefore my psychological constitution needs some time to adjust. It will come around, if given enough

Your Intuition or Mine?

141

time. In my view, it doesn't prove at all that my feeling of unease is the right feeling. It just makes me reconsider. I have had that experience so often with, for example, music or paintings that I wasn't used to, that were very different from what I grew up with. You don't like it at first, sometimes even experience the yuk feeling but then you will get used to it and even start to appreciate it. People rarely appreciate their first avocado, or their first Pendereski concert (sorry Robert, I don't even like my tenth), or their first Daniel Buren. The solution is patience. A. One final attempt: in medical ethics you are confronted with real-life situations. They will evoke feelings of 'this is wrong' . Those feelings will make you think and write about those situations. They are a reason to bother. They inspire you. They make you really care and not only argue for the sake of arguing. Those feelings come from your firmly held beliefs. Or sometimes, if there is no time to think long, you give an opinion and feel that you are right about it. Your experience, your clinical ethical intuition. B. That is called shooting from the hip, and you are not supposed to do that. And if I do it, I'm supposed to show and argue WHY. If 1just say 'I feel this is wrong', nobody will listen. If I don't come up with convincing arguments, the doctors might as well have decided by throwing dice instead of talking to an ethicist. Or have acted upon their own intuitions, which they didn't trust, otherwise they wouldn't have talked to me. Besides, what is wrong with arguing for the sake of arguing? At most, but I really mean at most, I grant you that intuitions may have a certain heuristic value. But that is no more than a hunchstatus. A modest status. You have to start somewhere. A. We need intuitions and they contain the moral wisdom of ages. All the dangers you have mentioned can be avoided. B. Prejudice or wisdom? Well, I guess I need your gut feelings as a source of criticism. A. And I suppose I need your criticism to keep me from conservatism. Ceterum censeo: male pregnancies are wrong. B. Can't stop, can you? Now you go and run to your principles. And back again. Very healthy all this exercise. You suffer from a severe attack of taking intuitions seriously. You're this embodied levelling rod. I'll go back to male pregnancies.

Inez de Beaufort

142

A. Yuk...

DIE WISSENSCHAFT· So besehlieszen beide denn naeh so manehem Doeh und Wenn, Sich mit ihren Theorien vor die Wissenschaft zu knien. Doeh die Wissensehaft, man weisz es, aehtet nieht des Laienfleiszes. Hier auch schurzt sie nur den Muod, murmult von "Phaotasmen" und Beugt sieh wieder dann auf ihre wiehtigen Spezialpapiere. "Komm", spricht Palmstrom, "Kamerad, alles Feinste bleibt - privat!" Christian Morgenstern

*

Christian Morgenstern (1908), Galgenlieder, Wiesbaden: Emil Vollmer Verlag, p.86.

Part III

The Scope of Reflective Equilibrium: Fundamental Views of Life and Religion

Religious Models and Moral Principles Vincent Briimmer

In the introduction to this volume, a distinction is made between an epistemological and a pragmatic perspective on reflective equilibrium. In the first, reflective equilibrium is seen primarily as an coherentist procedure for . justifying moral principle, and in the second as a pragmatic procedure for achieving moral consensus in a pluralist society. In this paper I would like to reflect on the role which religious beliefs might (or might not) fulfil within each of these perspectives. How we define this role, however, depends primarily on the way in which we understand the relationship between moral commitments and religious beliefs. The main body of my paper will be devoted to an analysis of this relationship. In the final section I will draw some conclusions from this analysis for the role which religious beliefs might fulfil within the procedure of reflective equilibrium. How are moral commitments related to religious beliefs? If religions are primarily ways of Hfe, then they obviously must have implications for the moral commitments of their followers. This applies not only to Christianity but to all religions and also to secular views of life and ideologies such as humanism and Marxism, which for many people today fulfil the role of a religion in their lives. In this way, religious beliefs clearly do entail moral commitments. But do moral commitments necessarily presuppose religious beliefs? Is there in some way an internal relation between morality and religious belief in the sense that morality is logically dependent on religious belief? This claim is clearly more problematical, especially in our contemporary pluralist society, where we need to have common moral commitments, shared by all people in spite of differences in their religious and ideological commitments. Modem society seems to require common moral commitments which are logically independent of the pervasive pluralism of religions or views of life to which various members of society subscribe. The way we are to deal with these issues depends upon the form of discourse we adopt when talking about the moral life. Thus we could express our moral commitments either in terms of moral principles or precepts or in terms of models and metaphors derived from religions or views of life. This distinction needs some elaboration.

145

w: van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, © 1998 Kluwer Academic Publishers. Printed in the Nelher/o.nds.

145-156.

146

1

Vincent Brummer

Moral principles

Morality does not merely deal in general with the things we do or the way we live, but more specifically with the things we ought to do and our policy for living. This policy is often expressed in terms of a code of moral principles. Such principles are rules or precepts in which some specific form of behaviour is described and the circumstances specified in which this form of behaviour is appropriate. Thus, for example, whenever circumstances C obtain (e.g. one is confronted by someone in need), one should always adopt form of behaviour B (e.g. offer assistance). A moral code consists of a coherent system of such principles which together express a policy for living. If we express our policy for living in this way, the relation between religion and morality would consist in the fact that religious believers try to live in accordance with the moral principles entailed by a specific religious tradition. Thus Christians would try to live according to Christian principles, whereas Buddhists would follow Buddhist principles, humanists humanist principles, etc. This way of stating the issue has two important implications. First, this view does not exclude the possibility that the relation between morality and religious belief could become an external one. The fact that religious or ideological beliefs entail moral principles does not exclude the possibility that one might subscribe to the moral principles and at the same time take leave of the religious beliefs. This is well illustrated by a recent controversy among Dutch Christian Democrats on the question whether it is necessary for those who want to become members of the Christian Democratic Party that they should be confessing Christians, or that it is sufficient for them to subscribe to the principles underlying the party programme without also being committed to the Christian faith. The implication is that the principles are not logically dependent on the faith and could also be accepted by people who are not committed Christians. Believers might respond to this point by maintaining that their morality is dependent on their faith, since the latter provides both the reason and the motivation enabling them to live according to their moral principles. Thus Christians subscribe to Christian moral principles because these express the will of God and they are only able to live in accordance with these principles because the Spirit of God inspires them to do so. Valid as this response may be, it does not entail that the believers' moral principles are logically dependent on their faith. Thus contemporary humanists would subscribe to the same moral principles of brotherly love, justice for all, human dignity, etc., without believing in a God who wills that they live according to these principles or inspires them to do so. They would derive their reasons and motivation for this from sources other than the Christian faith. It might of course also be argued that humanist

Religious Models and Moral Principles

147

moral principles are historically dependent on the Christian tradition from which they are ultimately derived. They are merely a secularized version of the corpus christianum. However, such historical dependence does not entail any logical dependence. On the contrary, the example merely goes to show that the moral principles of Christianity are not logically dependent on the Christian faith but can, through a process of secularization, become separated from it. If in this way we define morality in terms of behaving in accordance with general moral principles, we cannot but agree with W.K. Frankena that 'those who think that morality is dependent on religion need not and do not always mean that it is logically dependent on religion. They may mean only that it is causally or historically dependent on religion, or that it is motivationally or psychologically dependent on religion' (Frankena 1981, 15). A second implication is the following. If the relation between religious belief and morality is an external one, it becomes difficult to maintain that the moral principles of believers are distinctive to their religious tradition. In fact the same moral principl~s could be found in more than one tradition. R.B. Braithwaite argues this point as follows. 'I take the fundamental moral teaching of Christianity to be the preaching of an agapeistic way of life. But a Jew or a Buddhist may, with considerable plausibility, maintain that the fundamental moral teaching of his religion is to recommend exactly the same way of life. How then can religious assertions be distinguished into those which ~re Christian, those which are Jewish, those which are Buddhist, by the policies of life which they respectively recommend if, on examination, these policies tum out to be the same?' (Braithwaite 1964, 243-244). According to Braithwaite, the difference between adherents of different faiths (and, we might, add, nonreligious views of life like Marxism, humanism, etc.) is not in the fundamental moral principles which they accept but in their ritual practices and in the stories (or sets of stories) which they associate with their moral commitments. The rituals and stories are useful as psychologically motivating techniques which strengthen believers in their moral commitment, but are nevertheless externally related to the moral principles. We can agree with Braithwaite that religious beliefs and ritual practices provide believers with reasons for accepting moral principles and with a powerful motivation for living according to moral principles. The relation between religions beliefs and ritual practices on the one hand, and such general moral principles on the other, is motivational and not logical. As long as we define morality in terms of such general moral principles, it remains logically coherent for adherents of different religions and views of life to accept the same moral principles even though they do so for different reasons and with different sources of motivation.

Vincent Briimmer

148

However, we also talk about our moral life in terms of models and metaphors. Unlike Braithwaite's 'stories', these are not merely the imaginative means by which we could be motivated to accept moral commitments which are independently expressed. Rather they are themselves the means of expressing moral commitments. What does this imply for the relation between religious belief and morality?

2

Metaphors and models

It could be argued that all our experience involves interpretation. Immanuel Kant is correct in his view that our perception and cognition of the world around us is not merely a passive registration of sensory impressions given to us in perception but an active ordering of such sensory data. Such interpretation is always aimed at our understanding of what we experience, and always involves some form of comparison: I understand x by comparing it to y and noticing that in some way it is like y: I experience x as y (the sound on the telephone as the voice of my friend, the image on TV as a picture of the pope, etc.). Interpretative comparison can take on many different fonns and can be aimed at different kinds of understanding. One way in which we interpret our experience is by means of classification concepts: We note that x is like y because it belongs to the same class or category as y. Thus we always perceive objects as belonging to a kind (tables, chairs, trees, people, etc.) and not as random undefined objects. We classify things intuitively (and often deliberately) in this way in order to be able to cope with them. Thus we can cope with x in the same way as with y because it has the same characteristics as y, can be explained like y, can be treated like y, reacts in the same way as y, should be approached in the same way as y, or because the same attitude or course of action is appropriate in relation to it, etc. In this way classification is a basic function of our human form of life in the world (on this, see Brummer 1981,56 f.). Apart from classification concepts, we also interpret our experience by means of metaphors and conceptual models (see Brummer 1993a, chapter 1). According to Sallie McFague, 'a metaphor is seeing one thing as something else, pretending "this" is "that" because we do not know how to think or talk about "this", so we use "that" as a way of saying something about it. Thinking metaphorically means spotting a thread of similarity between two dissimilar objects, events, or whatever, one of which is better known than the other, and using the better-known one as a way of speaking about the lesser known' (McFague 1982, 15). Such metaphorical thinking is fundamental to all scientific discovery and explanation. I discover how x works by noticing that it is like

Religious Models and Moral Principles

149

y - even though y is something very different. Thus, for example, Newton discovered something about the moon by noticing that it is like a falling apple both being subject to gravity. However, scientists also employ conceptual models, Le. 'sustained and systematic metaphors' (Black 1962, 236), in order to explore and explain the workings of physical phenomena. Thus they explain the behaviour of gases by comparing them systematically with the behaviour of billiard balls, and they explain the behaviour of light rays in terms of waves or in terms of moving particles (see Barbour 1974, 30 and 71 f.). Models and metaphors also playa basic role in religion and theology. Here, however, they are not introduced to help us discover or explain the way physical phenomena work, as in science. Religious models provide an understanding of the meaning (on this concept see Brummer 1981, chapter 9) or significance ofour lives and of the world in which we live and in relation to which we act. In this way they determine our actions and attitudes. Different religions and views of life provide their adherents with different metaphors and models in terms of which life and the world can be understood, and which are definitive for the religions and views of life in question. A similar point is made by Wittgenstein (Wittgenstein 1966, 53 f.), who argues that participating in religious belief requires that we master the technique of using a picture. In learning how to use the appropriate pictures in the appropriate ways, believers must come to see which inferences are to be drawn from the picture and which are not. In discussing this point in Wittgenstein, W.D. Hudson distinguishes two sets of inferences which believers must learn to draw (Hudson 1969, 38 f.). First of all, they must learn to see how the expressions employed in the 'picture' resemble each other, and how they differ from their employment in ordinary non-religious contexts. Thus, for example, if in a theistic religion we use personal models in talking about God and his relation with human persons, we must determine the limits of the models: How is the relation between God and ourselves like and how is it unlike human relations? Here we have to sort out which implications of personal relation concepts, as these are used with reference to human relations, do and which do not carry over to the way we are to understand the relation between God and ourselves. Secondly, believers must come to see what implications the models and metaphors have for their actions and attitudes as they come to interpret their own life and their experience of the world in the light of them. Sorting out these two kinds of implications in a systematic way is one of the important tasks in theological enquiry. Thus Sallie McFague argues that 'the central role of models in theology is to provide grids or screens for interpreting this relationship between the divine and the human. (...) In order to interpret this relationship, conceptual clarity and precision is necessary: the structure implied in the

150

Vincent BrUmmer

relationship must be sorted out and its implications for personal, historical, social, and political life made manifest' (McFague 1982, 125). Religious models determine the actions and attitudes to which believers commit themselves in mainly two ways. Firstly, in understanding themselves and their own Ii ves in terms of the models, believers discover the role they have to play in their lives and actions. As Iris Murdoch puts it, 'man is a creature who makes pictures of himself and then comes to resemble the picture' (Murdoch 1962, 122). Secondly, in interpreting the world in terms of such models, believers come to see which actions and attitudes are appropriate in relation to the world and in the various situations in which they have to act. These two points could be illustrated with reference to the Christian faith. First of all, Christians look on their own lives as lives lived coram Deo in the presence of God. Their role in life is therefore that of 'children of God' , who live and act in fellowship with God. In this way believers are committed to making themselves available as secondary causes through whose agency God (as primary cause) can realize his intentions in the world (see Brummer 1992, chapter 5). Secondly, believers interpret the world in terms of the intentional agency of God. This entails that much of what they experience in the world is for them an object of praise and thanksgiving to the God who brings it about. In this sense J. Neville Ward claims that thanksgiving is 'the essential Christian posture before experience' (Ward 1967,20). However, much else of what they experience is seen as contrary to God's will and therefore to be opposed or changed. This entails a task which believers are called upon to perform in fellowship with God. In these ways the models and metaphors in tenns of which believers interpret their lives and experience of the world, have a commissive force, since, in accepting the interpretation, believers commit themselves to specific attitudes and forms of action. Do these models also have a constative force? Are they also claimed to be factually true? Or are they merely useful fictions in terms of which believers bring order to their lives, without claiming them to be factually true in any sensei] (Similar questions also arise with reference to the use of models in scientific inquiry, and realists differ from instrumentalists as to the answers. See Barbour 1974, chapter 3.) A non-realist and instrumentalist position with reference to the use of religious models is defended by R.W. Braithwaite (Braithwaite 1964. For similar views, see Kant 1793/1968, 868; Miles 1959, 180 f.; and Cupitt 1981). As we pointed out above, Braithwaite holds religious beliefs to be stories in terms of which religious believers express their attitudes and moral commitments, and which also inspire them to adopt these attitudes and these commitments. However, the stories do not have to be believed to be factually true in order to fulfil these functions for the believer. Similarly, W.D. Hudson points out that what Wittgenstein seems 'at times to have come near to suggesting

Religious Models and Moral Principles

151

is that, because religious beliefs have commissive force, that somehow entitles us to by-pass the troublesome problem oftheirconstative force' (Hudson 1969, 44). The trouble with this sort of view is that the actions and attitudes to which believers commit themselves when understanding their lives in terms of their beliefs, become incoherent if they were to reject the factual claims involved in the beliefs. It would be incoherent to live my life as a life in the presence of God if I were to deny that there really is a God in whose presence I live! Since the truth of the belief is a constitutive presupposition of the way of life, the latter would be incoherent if the former is denied (on the role of such presuppositions, see Brummer 1993b). R.W. Hepburn provides a good illustration of this point: 'If I say "the Lord is my strength and shield", and if I am a believer, I may experience feelings of exultation and be confirmed in an attitude of quiet confidence. If, however, I tell myself that the arousal of such feelings and confirming of attitude is the function of the sentence, that despite appearances it does not refer to a state of affairs, then the more I reflect on this the less I shall exult and the less appropriate my attitude will seem. For there was no magic in the sentence by virtue of which it mediated feelings and confirmed attitudes: these were responses to the kind of Being to whom, I trusted, the sentence referred: and response is possible only so long as that exists to which or to whom the response is made' (Hepburn 1957, 148). Truth claims therefore cannot be eliminated from religion. Yet, such claims differ from those in science by being internally related to the religious way of life: Religious truth claims are made with reference to the factual presuppositions which are constitutive for the way of life. Not only does commitment to the way of life presuppose the truth claim, but the truth claim would also lose its significance when it is divorced from the commitment. For this reason such truth claims are 'existential' in a way that the truth claims of science are not. It might make sense to say: 'It is true that the planet Jupiter exists and is the largest planet of our solar system, but I don't really care much about that.' It is, however, absurd to say: 'It is true that God exists and is the personal creator of the universe, but I don't really care much about that.' Here Wittgenstein might comment: 'For a blunder, that's too big'! (Wittgenstein, 1966, 62).

3

Models and principles

We have tried to show that whether or not we could claim that the relation between morality and religious belief is an internal one, depends on the way in which we conceive of and talk about the moral life. If we express our moral commitments in terms of general moral principles, they appear to be externally

152

Vincent Briimmer

related to religious beliefs. They could still be historically or motivationally related to the beliefs but are not logically dependent on them. If, however, we express our moral commitments in terms of the models and metaphors of a specific religion, the relation turns out to be internal. It would be logically incoherent to commit oneself to the way of life expressed in these models, while refusing to believe that they are somehow factually true. I logically cannot live my life coram Deo without believing that there is a God with whom I have a personal fellowship and with whom I am confronted in everything I do. How are these two fonns of moral discourse related? Do we have to choose between them or are they in some way complementary? One possible answer would be to claim that religious believers are committed to a moral life coram Deo which is internally related to their faith, and that they should therefore express their moral commitments in terms of the models of their faith. Nonbelievers, on the other hand, are committed to a morality which is independent of religion and therefore state their moral views in terms of general moral principles. Which form of moral discourse we choose, therefore, depends on whether or not we are believers. Arguing as a believer, Karl Barth seems to hold this sort of view when he asserts that 'ethical theory is not meant to provide man with a programme the implementation of which would be his life's goal. Nor is it meant to present man which principles to be interpreted, applied, and put into practice. (...) Ethics exists to remind man of his confrontation with God, who is the light illuminating all his actions' (Barth 1963, 86). The trouble with this view is that it does not account for the way we in fact talk about morality. It is not true that only believers express their moral commitments in tenns of models and metaphors. Everybody does, even though not everybody would do so in terms of theistic models. Not everyone looks on his own life as a life in fellowship with God. There is an infinite variety of models in terms of which people interpret the meaning of their own lives. Some relate their lives and actions to God, others to Nirvana, the coming revolutionary struggle, the happiness of mankind, the glory of tlie nation, or to some other ultimate concern (see Christian 1964, chapter 9). On the other hand, theists, like everybody else, also refer to moral principles and precepts when discussing moral issues. Clearly we do not in practice choose for one of these two forms of moral discourse to the exclusion of the other. In some way they are complementary. If religious beliefs are not merely to have a motivational relation to our moral commitments but are also to determine the content of these commitments and the nature of the actions to which we are committed, how is expressing these commitments in terms of religious models complementary to expressing them in terms of moral principles? Does the use of models add something to our talk of moral behaviour in terms of principles? Ninian Smart makes the following

Religious Models and Moral Principles

153

suggestion. 'The superimposition of religious upon moral concepts (...) gives the latter a different flavour: (a) because a moral action will have a double significance (not mere kindness, but consecrated kindness; not mere self-control, but a sacrifice, etc.); (b) because the solemnity of moral utterances becomes considerably increased: it is not merely that murder is wrong, but that life is sacred; a bad action is sinful and impious; discrimination against black folk in South Africa is not merely a great injustice, but is it (to quote a churchman's recent pronouncement) blasphemy; marriage is more than a fine institution, it is a sacrament.' (Smart 1966, 24) Thus Smart seems to say that we could all be committed to the same forms of moral behaviour as demanded by our common moral principles, and yet these forms of behaviour could have a quite different flavour depending on the religion (e.g. Christianity) or ideology (e.g. Marxism) or view of life (e.g. humanism) from which we derive the models which we superimpose on our moral principles. This view is faced by two difficulties. First of all, it is not clear what is meant by the 'flavour' which the models give to the forms of moral behaviour described and recommended in our common moral principles. Does this mean merely that adherents of different religions, ideologies, etc., feel differently about the forms of behaviour to which they are commonly committed? Or does it mean that their beliefs increase the solemnity of their moral commitments? I doubt whether many believers would recognize this as an adequate account of the bearing their religious beliefs have on the nature of their moral commitments! A more serious difficulty has to do with the suggestion that adherents of different religions, ideologies, etc., are committed to the same moral actions (even though these have a different 'flavour' for each of them). Even if their observable behaviour is the same, this does not entail that they are doing the same. Action cannot be equated with observable behaviour because any adequate description of an action must include a reference not only to the observable behaviour of the agent but also to his or her intentions in behaving the way he or she does. Stewart Sutherland explains this point with the following example (Sutherland 1982, 160 f.). The observable behaviour of Barry and Brendon is the same and can be described as 'driving lorryloads of food to the refugee camp'. Upon questioning, however, they describe what they do (i.e. their actions) in quite different ways. Brendon says: 'In taking food to the refugee camp, I am meeting the needs of the politically oppressed masses, and in this way preparing the peasants physically for the coming revolutionary struggle, in accordance with the teaching of Mao.' Barry, on the other hand, describes his action differently. He says: 'In taking food to the refugee camp, I am meeting the needs of my fellow creatures, and in this way partially realizing the Kingdom of God on earth in obedience to his will.' Although the observable behaviour

154

Vincent Brummer

of Barry and Brendon is the same, they are nevertheless perfonning very different actions, because the intentions with which they do what they do are quite different. These intentions cannot be adequately stated in terms of their observable behaviour alone. Although they both intend to feed the refugees, this is not a complete description of what they are doing. We must also include the higher level descriptions which they give to their actions, and which express both their higher level intentions and their beliefs about reality which are presupposed by these intentions. Thus, in describing what they do in terms of their respective religious (or ideological) models, they are not merely ascribing a different 'flavour' to the same actions, but are claiming to perform quite different actions, with different higher level intentions which presuppose different and incompatible beliefs about the nature of reality. In the light of this example, it is clear why talking about the moral life in terms of behaviour principles makes the relation between religious belief and morality external, while talking in terms of religious models makes it internal. Behaviour principles describe and recommend fonns of behaviour which could be commonly intended by adherents of different religions or ideologies, and do not, like the religious models, relate what is done to the higher level intentions and beliefs which differentiate the respective religions and ideologies. Should we conclude from this that models and metaphors provide a more adequate fonn of moral discourse than general principles or precepts which could be commonly accepted by adherents of different religions? This conclusion would go too far. True, moral discourse in terms of models is more basic than that in tenns of principles. We do not superimpose our models on our principles, but rather abstract our principles from our models. Nevertheless, we cannot do without this form of abstraction. We need discourse in terms of principles for various reasons, the two most important of which are the following. In the first place, we need general moral principles for the sake of moral instruction. J.R. Lucas explains this point as follows. 'We thus have to reduce the extreme complexity of the Christian life of love to relatively few principles, in much the same way as we have to reduce the complexity of English (...) composition to a few grammatical and syntactical principles when we want to teach someone. We need the moral law in something of the same way as we need grammar books. We fonnulate schematized and condensed sets of rules, because these can be taught to pupils fairly readily, and only when the pupil has mastered these can he develop the finer points of style. And just as we are careful to point out that to write grammatical English is not to have a good English style, so we also insist that to keep the commandments is not to live the Christian life. A man can keep all the rules of grammar, and yet write woodenly, and a man can keep the whole of the moral law and yet live a deadly life.' (Lucas 1976, 100)

Religious Models and Moral Principles

155

Secondly, it is necessary in any society that people co-operate in common action even when they differ in their religious or ideological commitments. In order to achieve consensus on such common action, it is necessary for them to abstract what is common in what they do, and argue about the advisability of that. In this sense we need common moral principles even when our intentions and beliefs (as expressed in our religious or ideological models) differ. Thus Barry and Brendan could only reach agreement about whether they should feed the refugees if they stick to shared moral principles and avoid arguing about whether they should do the will of God or follow the teaching of Mao.

4

Religious beliefs and reflective equilibrium

What does this analysis entail for the role which religious belief could fulfil within the epistemological and pragmatic perspectives on reflective equilibrium? Within the epistemological perspective, it is clear that for religious believers their religious beliefs constitute an essential component within the 'background theories' in tenns of which they justify the moral principles derived from the religious models employed in their understanding of the meaning of their lives and experience. Their moral principles are justified partly in the light of the religious beliefs from which they are derived. Partly but not wholly, since other elements are also relevant within the broad reflective equilibrium required for such justification. Thus, for example, believers will also have to see whether the principles in question are adequate and relevant for the specific factual situations in which they are called upon to act and the concrete moral decisions which they are called upon to make. This entails conversely that moral considerations will also play an important role in the reflective equilibrium in terms of which believers will have to justify their understanding of their own religious beliefs. If their understanding of their beliefs should entail moral principles which are no longer relevant or adequate, this could require them to reinterpret their beliefs in ways which do not have these unacceptable moral implications. This is similar to the way in which developments in our scientific knowledge of the world could give rise to reinterpretations of the faith. Such a process of reinterpretation is characteristic for the dynamics of all religious traditions (see Brummer 1997). Of course, it is only for believers themselves that their religious beliefs fulfil this role in the justification of their moral principles. Those who share their moral principles but not their religious beliefs will have to appeal to other religions or views of life as background beliefs for such justification. However, this merely demonstrates the 'person-relative' nature of all epistemological justification. What counts as sufficient reason or adequate justification for one

156

Vincent Brummer

person does not necessarily count as such for everyone else (see Mavrodes 1970, chapters 1 and 2). From the pragmatic perspective, the procedure which Daniels calls 'political reflective equilibrium' (see the introduction to this volume) has an essential role to play in the task of philosophical ethics to facilitate moral consensus within a pluralist society. Of course the range and content of the consensus which can in fact be achieved remains relati v~ to the specific variety of people with different religious or ideological commitments among whom the consensus is sought. Thus the kind of overlapping consensus which can in fact be achieved will not necessarily be the same in different societies. Nevertheless, moral consensus will always be in terms of shared moral principles and not in tenns of differing religious or ideological models and metaphors in terms of which people understand the meaning of their lives and experience. In this sense Rawls is correct in deliberately excluding comprehensive views of life from the form of reflective equilibrium by which moral consensus is sought (see the introduction to this volume). As I have argued above, the pervasive pluralism of such views of life tends to exclude consensus rather than to facilitate it. Thus Barry and Brendon can co-operate happily in feeding the refugees in spite of their religious or ideological differences. Political reflective equilibrium is an essential procedure in bringing about such co-operation.

Reflective Equilibrium and Cosmological Claims Carl Reinhold Brdkenhielm

One significant development of the theory of reflective equilibrium is the distinction between narrow and wide reflective equilibrium. A narrow reflective equilibrium is the coherence between a set of considered moral judgements and a second set of moral principles. For example, we depart from our convictions that religious intolerance and racial discrimination are unjust. We find that these intuitions withstand criticism and elaborate a certain conception of justice that expresses these judgements in a more general form. But what are the consequences of this theory of justice for the distribution of wealth and for authority? We might find some of the consequences questionable and, subsequently, modify our basic principle. John Rawls writes:

By going back and forth, sometimes altering the conditions of the contractual circumstances, at others withdrawing our judgments and conforming them to principle, I assume that eventually we shall find a description of the initial situation that expresses both reasonable conditions and yields principles which match our considered judgments duly pruned and adjusted. (Rawls 1971,20)

Norman Daniels amended this scheme by introducing the idea of a wide reflective equilibrium (Daniels 1979). He argues - following certain suggestions by Rawls himself - that a third set ofbackground beliefs is important, Le. a set of theories, which may include both moral and non-moral theories. Kai Nielsen suggests that this third set of theories might consist of 'whole moral theories, empirical cum theoretical theories about the function(s) of morality in society, about social structure, the basis of solidarity in society, theories of social stratification, class, and gender, theories about ideology, human nature and the like' (Nielsen 1993, 318). Later in the same article Nielsen also adds 'cosmological claims' and religious beliefs (p. 322). Such background theories might support or weaken the principles arrived at in a narrow reflective equilibrium. For example, Parfit distinguishes between a 'simple' view of personal identity, which he takes to support a deontological moral theory and the 'complex' view, which he takes to support utilitarianism (see Parfit 1973, 148, and Holmgren 1989, 54). Let us, then, depart from the following understanding of a wide reflective equilibrium: 157 ~

van der Burg and T. van Wi/ligenburg (eds.), Reflective Equilibrium, 157-169. @ 1998 Kluwer Academic Publishers. Printed in the Netherlands.

158

CarL Reinhold Brakenhielm

(1) There is at a certain time, T, a certain person, P, such that P after a process of deliberation considering three forms of beliefs (considered moral judgements, J, basic moral principles, M, and background theories, B) at T finds that the relationship between J, M and B satisfies the criterion of consistency.

Needless to say, there are many problems connected with this understanding of a wide reflective equilibrium. Some of them will be discussed later in this article. Let us further assume that cosmological claims can be a part of a wide reflective equilibrium. How does this affect the interpretation of the purpose and justification ofsuch claims? This is the problem upon which I wish to focus in this essay. Firstly, I shall briefly discuss to what question the theory of reflective equilibrium is an answer. Concluding that it may be used as a means to justify considered moral judgements, I will, secondly, discuss some amendments to the theory which I deem necessary for the fulfilment of this purpose. Against this background, I will, thirdly, consider the purpose and justification of cosmological claims when interpreted as background beliefs in a wide reflective equilibrium.

1

The general interpretation of reflective equilibrium

Reflective equilibrium is the answer. But what is the question? If we tum to John Rawls's A Theory ofJustice, the question is formulated in one way - and in a different way a few years later in his article 'The Independence of Moral Theory' (1974-75). Joseph Raz argues that Rawls originally designed his theory of reflective equilibrium as 'a way of collecting evidence about the nature of an independently existing moral reality' (Raz 1982,319). Raz's basic argument for this is that Rawls otherwise would have no justification for his exclusion of unconsidered judgements from the original position. Raz and others claim that Rawls soon after the publication of A Theory ofJustice came to alter the question to which his theory of reflective equilibrium was an answer. Christine Swanton argues that the distinction between value conceptions and values became salient for Rawls. This transformed the epistemological importance of his theory into a theory about the structure of our value conceptions (as opposed to a theory about the structure of value, see Swanton 1991, 577). She cites the following paragraph from Rawls's article 'The Independence of Moral Theory':

Reflective Equilibrium and Cosmological Claims

159

So provisionally we may bracket the problem of moral truth and turn to moral theory: we investigate the substantive moral conceptions that people hold, or would hold, under suitably defined conditions. In order to do this, one tries to find a scheme of principles that match people's considered judgements and general convictions in reflective equilibrium. (Rawls

1974-75,7) , In sum, we may distinguish between a realist and a nonrealist interpretation of the theory of reflective equilibrium. According to the nonrealist interpretation, .the purpose of the theory is not to arrive at a justification of an independently existing moral reality. The purpose is rather to suggest a way in which conflicting moral ideas can coexist in a peaceful way, or to describe a method to increase our knowledge about how our value conceptions develop, or should devel9p, to fulfil certain psychological or social needs. According to the realist interpretation, the purpose of the theory is rather to reveal which values are objectively real. This is an important problem, but I would suggest that the theory of reflective equilibrium can be interpreted in a third way, which leaves the issue of moral realism or nonrealism to be decided through the use of the method. Moral realismlnonrealism is not presupposed by the method. The method of reflective equilibrium is designed to show how a rational argument in moral (and metaethical) questions is to be conducted. Whether or such an argument leads to moral realism is an open question. But does the method satisfy the standards of a rational argument?

2

Reflective equilibrium and the standards of rationality

D.W. Haslett has argued that this method cannot be a method of justification because it tolerates (a) circular reasoning (b) adjustment decisions for which no clear criteria exist (Haslett 1987). Adjustment decisions are the decisions we have to make while we go back and forth between our considered judgements, our moral principles and background theories. We decide what to keep, what to discard and what to modify. Haslett argues that an alternative method is one which requires (1) coherence and (2) consistency with known facts but does

not tolerate circularity and underdetermined adjustment decisions. One way to meet these objections would be to modify the idea of a wide reflective equilibrium to allow for certain background theories, which can and should be exempted from adjustment decisions. Even if there are no infallible 'Archimedean points' of departure in the theory of knowledge, some points are more stable than others. These should - together with the basic logical rules

160

Carl Reinhold Brdkenhielm

- be regarded as the ground upon which these adjustment decisions are made. They should, therefore, be respected in a way other beliefs should not. Which are, then, these provisionally fixed points of departure? Well, firstly, there are certain basic facts of the world and human existence, natural beliefs in the sense of David Hume. For example, we presuppose the outside world and our ability to get some knowledge of this world. We know that the world has existed long before we came to experience it, and that it will continue to exist long after we have passed away. We also know that there is a transcendent reality in the sense that our senses do not reveal the whole of reality. We know for certain that the basic laws of nature will prevail. We know that people have an inner life, have memories, make plans, choose between alternatives and otherwise make decisions. All these things are as certain as things can get. Secondly, there are observation reports, which - as Haslett convincingly argues - have an initial credibility and which resist tinkering and adjustment in the sense that does not attach to the more theoretical constructions. Needless to say, all observations are theory laden, but some observations are more heavily laden than others. In other words, some observations are more resistant to changes in the conceptual framework than others. It might be hard to delimit this category of observations resistant to conceptual change, but it is not hard to make the distinction in practice. Think of the ship that Friday and Robinson Crusoe spot at the horizon. To be sure, Robinson's observation is more theory laden than Friday's. Robinson sees something that Friday does not see. But both see something in the sense that their experiences are experiences of something real. And Robinson's experience being more theory laden than Friday's detracts nothing by itself from the veridicality of his experience. But his observation being more theory laden than Friday's implies that there are more ways in which Robinson's observation could go wrong than Friday's. But if Robinson's observation were to be controlled by others who, hypothetically, would look in the same direction as Robinson, then the reason for questioning Robinson's experience would soon diminish to an insignificant degree. So, natural beliefs and controlled observation reports have a special status that cannot be attached to other background theories. To be sure, these beliefs and reports are not infallible, but it is misleading to treat them as belonging to the same category as other more theoretical background beliefs. They are in a 'privileged position' , meaning that they have a strong initial credibility and should generally be presumed to be true. Following this line of thought, the theory of wide reflective equilibrium should be amended in the following way: (2) There is at a certain time, T, a certain person, P, such that P after a process ofdeliberation considering three forms of beliefs (considered moral

Reflective Equilibrium and Cosmological Claims

161

judgements, J, basic moral principles, M, and background theories, B) at T finds that the relationship between J, M and B satisfies the criteria of consistency and that natural beliefs and controlled observation reports are in a privileged position. Such an amendment would meet Haslett's objection to circularity. But what about his other argument, viz. that the theory of a wide reflective equilibrium (interpreted as a method to justify moral claims) allows for underdetermined adjustment decisions? Haslett claims that the theory does not include any clear guidelines 'for making the many adjustment decisions that would have to be made in achieving a "fit" among the various elements that go to make up a person's equilibrium' (Haslett 1987,310). There might be two equilibria, (A) and (B), where (A) discards a considered moral judgement, J, in favour of a background theory, T, while (B) affirms J and rejects T. How do we decide between these two equilibria? Consistency is insufficient. A rational decision is impossible. There are in fact 'innumerable' different wide reflective equilibria compatible with any given set of considered moral judgements and background theories, 'each of these different reflective equilibria being the result of 'pruning and adjusting' the background theories and the judgements in different ways' (Haslett 1987,310). In sum, reflective equilibria are underdeterrnined by their starting points to an extent that makes a choice between them arbitrary. This argument might lose some of its force considering the introduction of natural beliefs and controlled observation reports. Haslett's objection could then be translated into the following amendment of the theory of wide reflective equilibrium. (3) There is at a certain time, T, a certain person, P, such that P after a process ofdeliberation considering three forms of beliefs (considered moral judgements, J, basic moral principles, M, and background theories, B) at T finds (1) that the relationship between J, M and B satisfies the criteria of consistency in the sense that J logically follows from M and Band (2) that natural beliefs and controlled observation reports are in a privileged position. One argument against (3) could be that it underestimates the difficulties of overcoming the divide between facts and values. There might certainly be some theoretical background beliefs which allow us to connect controlled observations reports with moral principles (in a suitable fit with considered moral judgements). Such a connection can be achieved by a naturalistic value theory or by making moral beliefs applicable through the justification of certain factual beliefs (see further Danto 1976, chapter 1). But the difficulties with naturalistic value theories are well known. And the second solution weakens the relation between

162

Carl Reinhold Brdkenhielm

factual and moral beliefs. In that case, moral beliefs are only negatively controlled by the observation of certain facts. But (3) also faces another difficulty. Consider the relationship between natural beliefs and controlled observation reports. What is the relationship between these two sets of claims? Is it possible to logically infer natural beliefs from observation reports? David Hume's problem of justifying induction are well known, and Roderick Chisholm has argued that there is no deductive nor an inductive relation between certain observation reports and our natural belief in, say, the existence of other minds. Nor is there some form of direct intuition. The relationship must be construed as some form of extralogical inference from certain observation reports to our natural beliefs about the existence of other minds (see Chisholm 1966, chapter 3). There are features of a person's countenance, the sound of her voice, her body language, etc. that indicate certain thoughts and dispositions of mind. Other minds are 'known through' observations and experiences. If this is the relationship between natural beliefs and observation reports, then it would seem misleading to demand a tight logical relationship between these beliefs on the one hand and the considered judgements, moral principles and more theoretical background beliefs existing in a wide reflective equilibrium on the other. We seem to be stuck in a cognitive dilemma: we seek a wide reflective equilibrium that is as little underdetermined as possible, but our cognitive capacity calls for some adjustment of our expectations. A more modest proposal would be to settle for the following version of the theory of a wide reflective equilibrium: (4) There is at a certain time, T, a certain person, P, such that P after a process ofdeliberation considering three forms of beliefs (considered moral judgements, J, basic moral principles, M, and background theories, B) at T finds (1) that the relationship between J, M and B, satisfies the criteria of consistency in the sense that J logically follows from M and B or is justified in some other way and (2) that natural beliefs and controlled observation reports are in a privileged position.

An example of other ways of justification might then be Chisholm's idea of critical cognitivism. This idea would also take care of some of the problems of justifying some of the background theories (such as certain natural beliefs).

Reflective Equilibrium and Cosmological Claims

3

163

The purpose and justification of cosmological claims

Cosmological claims can be characterised as general beliefs about the ultimate framework of human existence. These general beliefs exist in many more or less elaborated forms. In this context it is sufficient to distinguish between five main categories of such cosmological frameworks (Jeffner 1989, 26 f.). The first category is metaphysical materialism. There is no reality apart from matter, which ultimately consists of elementary particles in different combinations. The behaviour and experiences of human beings must be explained in the same way as everything else in the universe - even if these explanations appear to be a little more complicated than explanations of other things. The second type is a modified form of metaphysical materialism, where human beings appear as an inexplicable and unique anomaly in an otherwise material cosmos. This kind of idea is common to different forms of existentialism. The third type is metaphysical idealism. Reality is ultimately made of ideas. Our individual minds may ultimately be part of an all-embracing spiritual unity. Concrete matter is an illusion. In the fourth fonn, the universe appears as the creation of a god or gods. The universe has come about and continues to exist because of a transcendent reality. Human beings hold a significant place in this scheme, but the universe need not have been created solely for that purpose. In the Jewish, Christian and Muslim traditions we find various forms of this idea. The fifth fonn comes closer to some versions of the Eastern religions, where reality is conceived in a circular manner. Human beings exist in a cosmological framework, where their soul moves between different forms of existence or may find rest outside this reality. If any of these cosmological claims are retained in a wide reflective equilibrium, then certain consequences for the interpretation of their purpose and justification seem to follow. Let us, first, consider the purpose of such cosmological claims. Cosmological claims are often developed on the basis of a theoretical interest, Le. to reveal the true nature or structure of reality. But when introduced into a wide reflective equilibrium, their purpose takes on a more practical shape. They are now to assist the quest for moral guidance. We stand in need of orientation on the individual or social level. We look for ethical guidance in the fonn of, say, a principle of justice. Let me take an example from the field of philosophical theology. In his latest book, Ingolf Dalferth suggests that Christian (but also, I would suggest, Jewish and Muslim) belief in creation is made up of three distinctions: (1) between God and the World - and the dependence of the world upon God, (2) between that which is and that which ought to be - which .does not, unfortunately, boil down to the same thing and (3) between the present reality and the god given future perfection of the world. Dalferth argues that such a

164

Carl Reinhold Brakenhielm

belief in the world as God's creation basically has the purpose of enhancing our ability to orient ourselves in the world ('Steigel1lng unserer Orientiel1lngsnihigkeit in der Welt'). Dalferth continues: More precisely, this is the very task of theology: to elaborate in an argumentative way the knowledge how to direct our actions in the world as it is implied in Christian belief and use its cogniti vely and practically effective distinctions as means of orientation for social and individual life. I

So the purpose of these cosmological claims is not only to reveal the ultimate structures of reality, but - more importantly - to assist human activity on the individual and social levels. And this is further emphasised if such a belief is brought into play in the quest for a wide reflective equilibrium. Such an understanding of the purpose of the cosmological quest would clearly fit the purpose of a wide reflective equilibrium as outlined at the beginning of this essay. The theory of reflective equilibrium is designed to guide moral action. If a moral argument is realised according to this method, it is an open question whether it requires certain ultimate moral principles about an independently existing moral reality. I believe that this is the case, but I shall not enter into an argument about that in the present context. If it is the case, then the question about the need of introducing cosmological claims in the wide reflective equilibrium arises. This moves us into the larger issue of justification of cosmological claims.

4

The justification of cosmological claims in the context of a wide reflective equilibrium

How are cosmological claims justified in the context of a wide reflective equilibrium? Well, in one sense, the answer is simple. Cosmological claims are justified as far as they can be retained - more or less modified - in a wide reflective equilibrium. An obvious problem presents itself at the outset of such an exercise. Are not cosmological claims, by their very nature, resistant to modification, let alone falsification? Is it not the very essence of, for example, belief in the world as the creation of God, that it is absolute? This is a large issue which would merit a longer discussion. In the present context, it is sufficient to make a brief comment. It is a basic fact of history that

(Dalferth 1997, 218): 'Denn genau das ist ja die Aufgabe der Theologie: das im christlichen GJaubenswissen implizierte Orientierungswissen und dessen kognitiv und praktisch wirksame Grundunterscheidungen als Orientierungshilfe fUr das gemeinsame und individuelle Leben argumentativ ausluarbeiten.'

Reflective Equilibrium and Cosmological Claims

165

cosmological claims have emerged, developed, been modified and discarded. They have, in sum, been treated and challenged in a way that makes them appear far from absolute. And such changes and developments are found on the historical-collective and the individual-psychological levels. To be sure, Christian theologians such as Karl Barth have claimed that revelation has an absolute character. In his Church Dogmatics, he has argued that 'God's revelation has its reality and truth wholly and in every respect ... within itself' (Barth 1936, 350). Basil Mitchell has argued that Barth's affirmation ignores that it is in the end up to human judgement to detennine whether words or events, which purport to be revelatory are so or not and, if they are, how they are to be interpreted (Mitchell 1981, 144). But even if cosmological claims allow for adjustment and modification, how are they justified in the context of a wide reflective equilibrium? They are, I would argue, clearly, underdetermined by other beliefs. They cannot be interpreted as logical conclusions, nor as more or less probable hypotheses of the universe. There are philosophers and theologians who would not accept this statement. They claim that natural theology is possible, i.e. that there is a rational argument 'from the world to God' . One main argument against their view would be that mere generality of these metaphysical cosmologies precludes the interpretation that they can be deductively or inductively inferred from observation reports or more theoretical background theories about different aspects of reality. This, however, does not imply that cosmological claims are not in some sense controlled by other beliefs in a wide reflective equilibrium. But if it is the case that cosmological claims are underdetermined by other beliefs which have been adjusted to each other in the process of establishing a wide reflective equilibrium, are they not simply dispensable? The answer to this question depends upon their contribution to the achievement of the overall purpose of establishing a wide reflective equilibrium in the first place. Let me return to an earlier line of argument. The theory of reflective equilibrium is designed to show how a moral argument should be designed. It has a pra.ctical purpose. It is designed to help us orient ourselves in challenging moral situations. Let me take an example from medical ethics. Is it proper to give older persons access to transplantation - or should these scarce resources rather be limited to those who are younger? Let us assume that it is reasonable to allow also older persons to benefit from this new medical technology. An argument to this effect could be developed from the idea of human equality, Le. that all humans have equal rights on the basis of the respect that each autonomous individual is entitled to. This idea can be further justified by a theory about an independently existing moral reality giving human beings a special and equal value. Let us assume that a person arrives at such a basic moral principle exists and finds that it is consistent with her considered moral judgements and is suitably adjusted

166

Carl Reinhold Brtikenhielm

to a set of background beliefs. What possible gain could be expected from the integration of these beliefs with certain cosmological claims? The answer to this is, I would suggest,.twofold: cosmological beliefs are of added value to a wide reflective equilibrium if they (1) interpret the moral activity suggested by considered moraljudgemeots (justified by ultimate moral principles in conjunction with background beliefs) in a way that does justice to the emotional importance attached to human morality, and (2) make an overall sense of other beliefs entertained in the equilibrium. To make these contributions more concrete, let us consider two of the most influential cosmological frameworks in the Western heritage. Firstly, there is the modified fonn of materialism and, secondly, the creation theory common to Jewish, Christian and Muslim traditions. Do these cosmologies interpret moral activity in a way so that it does justice to its emotional importance? Robert Heeger once introduced me to an interesting proponent of modified materialism, Vitezslav Gardavsky. He was a Czech philosopher removed when Dubcek fell. He argued that it was atheism, rather than Christian belief, that required justification. In his book God Is not yet Dead (1973), he argued that defeat is clear the very moment we are born. Yet in spite of this, hope continues among humans. Everything in us urges us not to acknowledge death - and here are the roots of theistic ideas. But only if we face reality as it is, can we become authentic human beings. Gardavsky continues: This means that the following statement is valid: if we take death upon us as the future of which each individual is certain, every choice and decision that we make is given a new and radical orientation, i.e. that of going beyond ourselves to our fellows, to the community (Gardavsky 1973,212).

I understand this in the following way. Only modified materialism can match the emotional significance of moral activity. The human being lives alone in a cold, hostile universe. Moral decisions are made without any kind of metaphysical support. It is a uniquely human activity, in which the individual commits herself to a goal which goes beyond her limited existence to a larger vision of human community. Any effort to anchor this in a reality outside the individual decision detracts from the extraordinary character of moral commitment. It is only against the background of the grim realities and futility of human existence that justice is done to the emotional significance of human morality. Compare this to the interpretation of morality from the perspective of the creation theory. Here, moral activity is often taken as one of the ways in which God's purposes and plans are fulfilled. Seeing the duties of ordinary life as part of God's purposes gives them a significance that matches their emotional

Reflective Equilibrium and Cosmological Claims

167

importance. Needless to say, we may lead a moral life without any kind of such religious beliefs. But having such beliefs gives a depth to moral activity it would otherwise lack. So both modified materialism and the creation theory give significance to morality, albeit in radically different ways. How are, secondly, these cosmologies related to other beliefs in described equilibrium? Let me first summarise an answer as to their relationships to scientific background theories. The creation theory is clearly underdetennined by these background theories. Nevertheless, in some modified version it can be shown to be consistent with these theories. Modified materialism excludes the existence of God in a way that does not seem to follow from any known scientific theory. In that sense, it is also underdetennined by any known theory that could appear among the background theories in a reflective equilibrium. (There is, however, at present a significant debate when it comes to the view of human nature that follows from evolutionary theory.) One argument, however, speaks in its favour: it is simpler than a creation theory, which assumes a personal transcendent reality. Simplicity and economy are not usually among the general requirements for a reflective equilibrium, but it might be presumed especially when it comes to the background theories. What about the relationship between cosmological claims and natural beliefs? In what way can cosmological claims make sense of - for example the simple and inescapable conviction that the course of nature will always continue in a uniform way? For David Hume it made no sense to question this natural belief. On the other hand, he found 'all events completely loose and separate'. And he turned to backgammon when he found no rational way of make this observation coherent with his natural belief in the unifonnity of nature. In view of creation theory, there is a solution to this dilemma. Things and events may not have any necessary connections, it is only God, who keeps them together. Take away God - and you get Hume's experience that everything is 'loose and separate' (see furtherWedberg 1959, 115). So on this score, creation theory could make sense in a way modified materialism in a Humean form might not. (There might, of course, be other forms of modified materialism which could have a theory of matter which would make sense of our natural belief in the uniformity of nature.) When it comes to the relationship between cosmological claims and controlled observation reports, the modified materialist finds herself on high ground. There are no observational reports about the existence of a Creator. The proponent of the creation theory might want to widen the concept of observation to cover certain religious experiences. It is questionable whether this is a defensible move, but creation theory faces a larger and more difficult problem, namely the obvious existence of evil. This clearly puts the religious

168

Carl Reinhold Brdkenhielm

believer on the defensive. How is the existence of a good Creator-God consistent with all the evil and suffering there is in the world? In sum, there are clear examples of how cosmologies can (1) contribute to an interpretation of morality and (2) make sense of other beliefs in a reflective equilibrium. But examples to the contrary are equally easy to assemble. It is an open question if any - and, if so, which - of them can be modified or adjusted to be retained in a wide reflective equilibrium. All in all, cosmological relativism seems to be the most reasonable position. Cosmological relativism affions that even the best methods we have for the rational assessment of cosmological claims does not give us any clear answers - or contradictory answers. Cosmological claims are underdetermined by evidence in a more radical sense than other beliefs. Firstly, they cannot be logically inferred from any of the other kinds of beliefs that are found in a reflective equilibrium. But this is a quality they share with - for example - natural beliefs. But secondly, and more radically, we do not even know what would count as a non-logical justification of such cosmological claims. For example: Is the metaphysical materialist allowed to conclude that her position is justified from the fact that she has had no experience of any other reality beyond that which is known through her senses? How do we establish in a credible and reasonable way if this is - or is not - evidence for the position of the metaphysical materialist? It seems that we are at loss to find a conclusive answer to this question. Cosmological relativism would need a longer argument to be sufficiently established. Let us assume that this would result in a reasonable and convincing argument. Then the following question arises: Given cosmological relativism, is it nevertheless rational to affirm a cosmology in the sense of a claim concerning the overall nature or structure of reality? It has been argued that the only rational option is to abstain from such judgements. But there are two arguments against such a position. The first is that we tend to make some such assumptions anyway. It is hard to refrain from constI1lcting some larger picture of the cosmological context of human existence. Therefore, it is better to make an open choice, than to presume neutrality with the risk of slipping into a hidden partisanship. Secondly, many persons find that their cosmological beliefs are part and parcel of a personal experience which opens up reality and unlocks the soul (Tillich 1957, 42). If a person continues to live on the presumption that this experience is an experience of something real, and subsequent experiences and increasing knowledge confirm and sustain this presumption, it seems rational to adhere to this cosmology. And this would be the case even if (1) other persons adhered to other cosmological beliefs under the same conditions, and (2) that their cosmological beliefs remained underdetermined by evidence and no theoretical quarantees were found. If such beliefs were

Reflective Equilibrium and Cosmological Claims

169

retained in a reflective equilibrium, they would at least be rational in a weaker sense, Le. negatively controlled by evidence.

World Views, Value Change and Reflective Equilibrium Anders Jeffner

The problem I want to discuss in the present essay is the following: If we use the process of finding a reflective equilibrium as a decision-making procedure in ethics and introduce what we know about regular value change, what then are the consequences for classical meta-ethical theories? To provide a background to the problem, I must first say something about change in values and views of life (1), and second describe some general theories as to the causes of such changes (2). I can then (3) discuss how the kind of knowledge represented by (1) and (2) affects the outcome of a process of reflective equilibrium and, finally, come to a conclusion in the field of meta-ethics (4). (1) What are the basic values that affect the lives of ordinary people in our societies and what do people in general think about the world and our position as humans in this world? Do beliefs and valuations combine in certain patterns, and are they connected with a basic life-accompanying mood such as optimism or pessimism? These are questions which we have dealt with in Uppsala in connection with more traditional philosophical, ethical and systematic theological studies. A combination of sociological and philosophical methods has turned out to be fruitful in the study of religion and its neighbours. We call the different kinds of research around these questions exemplified studies of ideologies or world views, some results of which I have reported in a Festschrift for Vincent Brummer (Jeffner 1992). The study of world views is closely connected to some of the general value studies which are being much discussed at present. There are some important differences between studies of world views and general value studies but the differences are not particularly important in the present context. 1 It is clear from this kind of research that the process of modernisation in Western societies has been accompanied by a long-term change in valuations and world views. One such change was observed by Weber: the change from a religiously motivated authority in society to a rational legal authority, a process which is accompanied by a diminishing role for religion. The secularisation thesis has been verified in various value researches and research into general For examples of international value studies see Van Deth and Scarbrough (1995) and Inglehart (1997). The development of valuations among young people in Sweden is analysed in Jeffner (1997). 171

w: van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium,

© 1998 Kluwer Academic Publishers. Printed in the Netherlands.

171-174.

172

Anders Jeffner

world views, another change which has been clearly pointed out by Inglehart on the basis of the World Value Surveys (see, e.g., Inglehart 1997). During the last thirty years, Western societies have witnessed a turning from the high valuation of material values, such as economic security, strong military force and the importance of work, towards values that concern freedom of choice, leisure time, toleration of different family patterns, etc. Inglehart calls the former configuration of values 'scarcity values'. He thinks that they emerge when people who were brought up in a situation of material deprivation are becoming aware of the possibility to overcome social and economic insecurity. The new value pattern in the industrialised countries is called 'post-modem values'. Inglehart has shown that the value change can be predicted by looking at the general process of modernisation in these societies. There are many problems connected with both the secularisation thesis and the theory of a tum towards post-modern values. It is not my intention to deal with these problems here. The only thing I want to argue here is that there has been a well-established change in basic valuations and world views in this century and that this change has some form of connection with more concrete social changes such as industrialisation, urbanisation, mass education and other forms of what can be called modernisation. (2) How shall we explain these changing values and world views? There are many, very different theories available and they can be systematised in different ways. I shall first distinguish between general social constructivism, materialistic constructivism, idealism and objectivism and then suggest another grouping of these theories. General social constructivism seems to be the leading theory at present. 2 It states that the changes we have discussed above can be explained by social changes in these societies, world views being social constructions and value change being a function of such factors as industrialisation. Most authors in the field of value studies - Inglehart included - seem to adhere to such a theory. Whether the changes following social development are good or bad is not discussed, and there are no ethical valuations built into the" theory. Societies change, and values and world views change with them. This is all we can say. A slightly different constructivist theory is the materialistic one. It explains the changes in values from specific material changes. At the societal level, one such factor, the means of production and people's relation to it, can be seen as the most decisive one. Marxism is of course the standard example of such a theory. But in Marxism a quite different idea can be found; it is the view that the development based on material change has a certain direction, which is good. 2

For example, the well-known book by Berger and Luckmann (1967).

World Views, Value Change and Reflective Equilibrium

173

There is a goal for the development and therefore there is a kind of ethical criterion built into the theory of change. We get an idealistic type of theory if we say that changes in ideas, including values, are the driving force for the changes in society. Weber's well-known view of the importance of Calvinistic ethics for the development of capitalism is a modest example of a theory of this kind. It can, of course, be totally relativistic, but it can also contain a kind of ethical criterion: the development goes towards a good end, Hegel's grand-scale theory being an example. Idealistic theories, not even in modest variants, do not seem to be very popular among social theorists at present. An objectivist theory sees the changes in a different light. It is supposed that world views can come more or less close to objective reality and that our knowledge of values can be more or less adequate. A change can then mean that we come to know more of objective reality or that we destroy part of our former knowledge. An objectivist theory has an in-built ethical criterion. It is now apparent that we can also divide the theories of change into two categories - theories with an inbuilt ethical criterion and theories without such a criterion, the latter kind being the leading one at present. (3) Given the long-tenn change in basic values and world views described in (1), we can expect a change in the outcome of a reflective equilibrium procedure in a pre-modem, modem and post-modem society even if the relevant empirical facts are the same. If we now add a theory of change without built-in ethical criteria as described in (2), then the result will be a kind of genuine ethical relativism. Ethical decisions change together with changing belief structures in our societies. Actions are not right or wrong but right or wrong given a certain state in the changing situation of values and world views. It is questionable whether we can talk about ethics at all, given these premises. If we keep the first premise - the changing values - and want to save a genuine ethical outcome of a reflective equilibrium we could choose a theory of change with built-in ethical premises. But it is difficult to find good arguments for a Marxist theory or an idealistic theory of a Hegelian type. The objectivist theory remains. However, the fact is that precisely the theory of value change - which is a premise now - is the main reason against an objectivist theory. It seems to be difficult to avoid ethical relativism if we adhere to a procedure of reflective equilibrium as the main decision-making procedure in ethics. But there is a way out for those who feel discomfort with relativism. (4) The first step on an alternative road is to distinguish between values and valuations. The term value studies, which has been well established, conceals this important distinction. Evaluating is what we do when we prefer a certain

174

Anders Jeffner

thing, action or person to another. Health is more valuable to me than wealth. This is a valuation. What we measure in value studies are valuations of different kinds. Valuations change, there is no discussion about that. To speak about values is much more problematic. Saying that there are values as distinct from valuations involves ontological claims. Adherents to values can claim that there is a relation in reality which we can apprehend and which has the unique characteristic of being demanding or repelling. To be honest has a positive value regardless of the general valuation at a certain time. If a valuation corresponds to these values, then the valuation is adequate, according to this line of thought. Can we now admit that there are constant values and changing valuations? To a certain degree, there is nothing strange in this belief. It is like believing in the possibility of reaching something true about reality even if the scientific theories are defective and changing. But we reach a limit if we can show that all valuations are changing. If there is total flux in human valuations, then it is improbable that values exist; but if they do, they must be beyond human interest. Those who argue for emotive or subjectivistic meta-ethical theories often build on such a premise of total change. But there is no empirical evidence at all for this radical diversity of valuations. So a theory of change with an inbuilt ethical criterion is very well compatible with the empirical findings of changing valuations. To talk of changing values, however, is misleading. To use the procedure of reflective equilibrium is then reasonable, even in the face of changing valuations, although there is a price to be paid - a more complicated metaphysics, containing values. To pay such a price is an example of accepting a certain world view. The empirical study of changing world views and the process of finding a reflective equilibrium thus bring us back to the philosophical questions of the choice of world views. Here the criterion of coherence plays a decisive role and for those who take ethics seriously and are neither Marxists nor Hegelians, it points in the direction of some sort of meta-ethical objectivism.

Part IV

The Purposes of Reflective Equilibrium: Modelling Public Debate and Dialogue

Reflective Equilibrium and Public Debate: How to Cast the Public's Web of Beliefs Broadly Enough Jan Vorstenbosch 1

1

Introduction

Public debates are an important instrument of democratic policy making. They enable representatives and governments to take notice of the many aspects, interests, arguments and viewpoints on some policy issue. By incorporating these elements in the procedural fonn and the content of political decisions and of legal regulations, those who are politically responsible can ensure that their decisions are better complied with. There is a principled side to public debates as well. One of the defining characteristics of a democratic policy is that it shows equal respect for all the citizens that are subjected to the policy (Van der Burg 1991). It is not only politically prudent, but also morally obligatory that notice is taken of the interests and views of citizens. In this way, citizens are taken seriously. But besides that, taking into consideration a plurality of viewpoints is relevant to bring politics closer to a justified decision, if not to truth (Mill 1859/1974; Van der Burg 1991). Public debates, in the sense that I take them in this essay, are an important means to attain these ends. These remarks hold especially when it concerns policies that involve moral issues. For moral issues are almost by definition controversial, at least in a pluralistic society, in which people hold divergent views on a wide spectrum of nonns and values regarding public as well as private morality. These opinions, about what is morally acceptable or not, flow from deep-seated views on life, the nature of persons and reality, and what kinds of values are really worthwhile. Although it would be an exaggeration to say that every public debate is of a moral nature - some debates are about economic, aesthetic or political issues in a narrow sense - most of them are. One of the reasons for this is that moral issues have an essentially normative nature, and this means that they cannot be decided by experts - scientific or professional. The opinion of every citizen ought to be taken seriously initially, because in a liberal-democratic society nobody's moral view is a priori privileged. This does not mean that, at the end of the day, all views stand equally well defended. But what it does lead to is some procedural guideline. When moral issues come up in society, everybody

Thanks are due to Anton Vedder, from whose comments this article has profited. 177 W van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 177-190. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

Jan Vorstenbosch

178

is invited to utter his views on them, and there is no obvious way to establish the right view other than by offering views, arguments and opinions for critical debate in the hope that some argued consensus will take shape. My objective in this article is connected with this last observation: the fact that there is no obvious way to establish the right view. Therefore, I want to consider the possibility of using the reflective equilibrium method as a method by means of which arguments, views and opinions can be organized in such a way that there is a greater chance that some view will be justified as the right one. It is not so much my intention to show that the elements that the reflective equilibrium method builds on - judgements, principles and background theories - show up in actual public debates as to evaluate the case for using the method developed in the domain of moral theory (Rawls 1971) or ethical decision making on cases (Van Willigenburg and Heeger 1991) to the much more intractable sphere of public debates. Evidently, this objective presupposes that in actual public debates the three elements are present in some rudimentary form or that they can be introduced into the process without changing the nature of public debates too much. For if this were not the case, the method would be without any 'bite'. What I am after is an analysis of public debates and what I hope that this analysis will show, is the potential of organizing debates - in retrospective, but also while they are taking place - along the lines of the reflective equilibrium method. Of course, organizing should be taken here in a weak prescriptive sense, as a kind of model that might help relevant actors (media people, .ethicists, philosophers, opinion leader) to reach the objectives of a public debate. It will soon become clear that public-debating processes in a democratic society should not be directed, let alone orchestrated, in any strict sense if at all possible.

2

Three problems

Considering the history and theory of reflective equilibrium my objective will meet with a number of difficulties. I can only deal with three problems that, in my opinion, are the most important. By dealing with them I hope at least to prepare the way for a further development of the applicability of the method. The first of these problems concerns the assumption, largely implicit in Rawls (1971), but made explicitly by DePaul (1993) that the reflective equilibrium process has as its substrate and substance the system of moral beliefs of an individual. Public debates, however, are essentially collective. The second problem with modelling public debates on the idea of reflective equilibrium concerns two elements of public debates that distinguish them from debates between ethicists or from deliberations about concrete cases in particular groups. Both elements connect with the fact that public debates have a relation

Reflective Equilibrium and Public Debate

179

with public policy, with the problem of what society or government ought to do concerning some actual problem, while at the same time these debates seem to have an aim and character as well as problems and resources to solve these problems that differ from those of official public policy. The first element is about the strategic nature of interventions that take place in public debates. Because of the potentially political implication of the outcome of public debates, interests are often not very far from the scene. Reflective equilibrium as a method, however, assumes specific conditions of impartiality. The second element is concerned with what Rawls has called the 'burdens of judgment' that often playa significant role in policymaking. Among these are the non-moral questions that can only be answered by procedures of a scientific, technical or administrative nature. It seems that 'the public' in general is not knowledgeable about these things, and so it could be argued that public opinion can better be left out of the decision-making process and that not too much notice be taken of public debates. The third and most important problem is that in modern society there is a plurality of moral opinions on most issues that give rise to debate. Divergence of moral opinion is at the heart of the matter. The method of reflective equilibrium seems to start from elements like considered judgements that have at least initial credibility for those who are prepared to be led by the method. This condition is not fulfilled in the case of public debates, or so it seems. At the very best, the method will help to clarify the different positions of the participants, but it will not lead to a unified, coherent and substanticd result that has any effective, normative impact in politics. Each of these problems will give us cause for specifying and qualifying the use of the method in our case. Before dealing with these problems in detail, I will clarify the notion of public debate. What is to be understood by this notion in the context of this essay? Let me give an example of a specific moral issue that the reader may keep in mind for purposes of understanding: in 1996 a Dutch private clinic claimed to have found a method for sex selection. It offered future parents the opportunity to choose their baby's sex. The news raised a storm of reactions in the Netherlands, and the possibility of closing down the clinic by legal means was discussed in the media for quite some time. Interviews were held in the street to ask lay people for their opinions; principles like the right of people to plan their own families and the duty of medical professionals to offer only reliable methods for help were mentioned, and background theories, for instance on the role of technology in our society and the limits to medical diagnosis, showed up with some frequency in the debate. The debate ended with the decision of the Dutch Minister of Health to ban this technique, although it was unclear whether this was for moral reasons or because the method got the reputation of being unreliable. In this case, a societal reflective equilibrium

Jan Vorstenbosch

180

was reached by taking a closer look at the facts and bringing these new facts to interact with principles like the duty of offering only reliable diagnostic methods. But we can easily imagine the situation to be different (and this may become a reality in the near future). For instance, a better, reliable method of sex selection may force society to reconsider the right of citizens in a democratic society to plan their own families (a principle) or the role of healthcare in society (a theoretical question that often remains in the background of debates).2

3

Public debates

A public debate, as I will take the notion to mean in this article, is an exchange of opinions, arguments and general ideas about a specific issue that is accessible, at least in theory, to everyone (every citizen) and in which, again at least in theory, everyone can participate (Barry 1990). In modem society, the vehicle for this kind of debate is the media - current affairs programmes and talkshows on radio and TV, newspapers and magazines, etc. Especially the second element of my definition - the fact that everyone can participate - means that the object of my exposition is different from two types of debate with which it can easily be confused. First, there is a specific kind of public debate that is a one-time event, organized by someone or some institution in a specific place with, for practical reasons, limited access to the public. Sometimes this sort of debate is organized even more strongly, as in the case of Danish and Dutch experiments with a well-defined debate format, in which carefully selected and well-informed lay people meet experts to discuss the ethical and societal consequences of scientific and technological developments (Feenstra and De Rooij 1994). Much can be said about this kind of debates and their objectives, but this is not my concern in this essay. Second, I exclude political debates in Parliament. These debates are as a matter of course public, but only representatives of the people can participate. There are other reasons to exclude parliamentary debate; for instance, there is a strong strategic, political dimension in these debates and many things that matter in it are of a technical, sometimes scientific, sometimes legal character (for instance, the scope and relevance of an international treaty that the

2

For instance, see Trappenburg (1993) who, following Walzer (1983), makes a big point of demarcating the sphere of health and health care as distinct from that of other societal values and practices.

Reflective Equilibrium and Public Debate

181

government has signed), which makes it less accessible to the public, to say nothing about participation. 3 Of course, there are all kinds of relations between a public debate in the sense I use the term and specific public debates and parliamentary debates. Most of the time, a specific public debate is an element within the broader spectrum of media attention and discussions in newspapers that constitute a public debate. And public debates will influence political discussion in Parliament. Arguments that count in discussions about policy will often tum up in Parliament as well. But the most interesting aspect of public debates is the very idea that they have 'a meaning and logic of their own in a democratic society, with a kind of potential that organized debates and Parliamentary debates do not have. This meaning is foremost that public debates can bridge the gaps that are liable to originate between official political institutions, as managed by the representatives of the people, and the way in which people in society think about policy issues. Furthermore, they can play an important role in the identification and articulation of societal and moral problems and their possible solutions. Following Easton's (1971) conception of politics as the authoritative allocation of values for a society, in Dutch policy theory the process of political decision making on moral issues is often split up in a stage of pointing out the problem, a stage of articulating the problem and a stage of allocating values with regard to the solution of these problems by the authorities. The first two stages prepare the way for the third, the fonnulation of an actual policy. In Easton's words: 'a web of decisions and actions that allocate values' (Easton 1971, 130). Allocation of values here means 'direct successfully how people mayor ought to act regarding a practice' . Now, this is easier said than done. And the fact that public debates are not formally organized, which is an asset for some, does not mean that they are not informally organized in a way that might even be more damaging for democratic decision making. For the purposes of this essay the lack of organization creates a problem of definition as well, because one could wonder whether a public debate is identifiable at all. The fact that it has no defined place or time could raise doubts about its status. A public debate seems to have a vague existence in a space created by the media, often for a limited period. One could raise huge ontological problems about this, but I prefer to take things simpler here and save the ontological problems for some other question of relevance for our theme. The clue to the identification is the occasion or origin of the public debate. Over the last ten years, especially since the ever faster development of science and technology and, more specifically, biotechnology and computer 3

For an extensive argument about the difficulties of conducting a public debate in Parliament, see Jurgens (1995).

182

Jan Vorstenbosch

technology, almost all public debates have started with a news item - the birth of a cloned sheep, the defeat of a chess world champion by a computer, the putative discovery of an aggression gene. These are the kind of things that trigger a public debate, and so at least the starting point of the debate can be determined fairly well. The next stage of the debating process is three-tiered: there is a response from the media - journalists chase scientists, ethicists, representatives of societal organizations or politicians for more information and especially for opinions; there is a response in society - the news is discussed at the offlce~ in homes and bars, obviously because it is easily connected to a broad range of emotions and values; and there is a response from the official political circuit: politicians give their opinions on the matter. These three tiers come together in the media and the media is undoubtedly the most important factor in the origin of a public debate. How long a public debate will go on, how wide it will spread (for instance countrywide or worldwide) and especially when it ends, these are more difficult questions. But, in general, most of the answers will depend on the intensity of the relation of the news with the emotions and the values of the public: how far it reaches into the essentials of the way people see and value life and how they perceive the ways and the risks involved in which something - for instance, a new technology - may affect their own lives. 4 As for the end of a public debate: often it just withers away, sometimes it is ended with the announcement of some sort of official policy, a legal regulation, or some other way of handling societal problems which is available in modern society and which is trusted by the people. 5

4

5

The Theory of Emotions developed by Frijda (1986) offers many possibilities for explaining and interpreting these reactions. Frijda's thesis is that a person's emotions originate in the fact that a person's interests (taken in a broad sense) are perceived by the person to be affected by a development, an action or event. One could argue that this is a much too narrow conception of public debate, because often these specific public debates reflect an ongoing concern with a much larger problem that troubles modern societies, for instance privacy, the welfare state or the moral consequences of new technology. This is certainly true, but by not limiting myself to public debates with a clear time span my analysis would loose greatly in focus. In the end, we may find ourselves back with the use of reflective equilibrium in moral theory that has, as in Rawls (1971), its focus on the articulation of principles that apply to society over a very broad time span or maybe even pretend to be universal. This certainly is and cannot be the objective of the public debates I have in mind.

Reflective Equilibrium and Public Debate

4

183

Reflective equilibrium

Considering the remarks above about the way in which public debates develop, it seems quite difficult and unrealistic or even hopeless, to apply any formal method of ethical reasoning to this process - certainly not the method of reflective equilibrium, which is known as a rather complicated method based on the idea of coherence as a criterion for justified moral judgements. Let me state at the outset that my hopes are not very high. What I certainly do not want to argue is that it would be a good thing to impose the constraints of the method on public debates as they take place, if this would be at all possible. I do not think that the purpose of public debates is to generate justified moral judgements. This would be asking far too much from the fragmented phenomenon which a public debating process is. Nor do I think that public debates can replace the more formalized way in which ethicists and philosophers argue for their theories and positions on ethical issues or, for that matter, the political decision-making process in a liberal-democratic society. These limitations may be sufficient to make clear what I do want to argue for. I think public debates are significant in bringing to light the plurality of views on some new, unusual or even shocking issue which has a great potential for changing things in society. And the method of wide reflective equilibrium can have an important heuristic function in accomplishing this task because it can help to point out relevant judgements, suggest principles and bring in background theories that might otherwise be left out of the debate. It can also have an important evaluative function in these three senses. Moreover, the method could be useful to detect or suggest how judgements, principles and theories cohere to reduce the positions to a manageable number of well-defended viewpoints or perhaps to generate a consensus on some points.

5

Beliefs: individual and collective

One could argue that in these senses the method is already, albeit unconsciously, present in the way in which public debates develop in our society. For the freedom of press and broadcasting, in some ways supported by government through subsidies, and the resulting differentiation in newspapers and broadcasting organizations guarantee to a considerable extent that the plurality of opinions, principles and theories is expressed. It is also at this point, however, that the objections I mentioned earlier on, come to the fore. For reflective equilibrium does not start with a plurality of views, but with three sets of beliefs. To a certain extent, these sets may be in disequilibrium, but on the face of it, they are not marked by the fact that they are contradictory or incompatible, as

184

Jan Vorstenbosch

is the case with some stances taken by individuals or groups in society. The reason for this is simple. These sets of beliefs are all part of a belief system that is attributed to an individual. The assumption is that this system is rational, in the sense that there are no insolvable contradictions. Or, at least, the system can be made rational by the method of reflective equilibrium because this method enables the individual to eliminate elements that are contradictory or that generate tension. So here we run into our first problem. Does it make sense to attribute moral beliefs to entities other than individuals? I think it does and I think that the way in which Rawls introduced the method of reflective equilibrium does not rule out this use of the method. For Rawls made no ontological commitments as regards beliefs. In A Theory of Justice, moral theory is conceived as a propositional system. These propositions can be expressed in statements which individuals can endorse because they believe them to be true. But this is also possible for groups. Groups can endorse propositions as well. Or rather, often it is the other way around: groups are identifiable precisely because the individuals that constitute the group, endorse the same beliefs, especially nloral beliefs. I do not see any epistemological or semantical problem in speaking of a judgement or a principle or a theory (for instance, evolution theory) that is held by a group. The real problem that the objection seems to tum on is that it is difficult to attribute the propositions that are offered for endorsement, to 'the public', or to 'society as such', because there is no such thing. There are only individuals, who can be categorized in groups, and these groups sometimes differ greatly on the most significant matters. But even this position may be too exaggerated. For it seems that the use of a concept like society (or culture) presumes that there are certain propositions or beliefs on the basis of which this society can be identified as a society and on the basis of which the members of the society can be identified as members of that society. I do not want to discuss here the individualism-holism issue in social philosophy, because this would take us too far. But I do not need this idea for my general position on the use of the method of reflective equilibrium in public debates. For this position, it is sufficient that we can ascribe systems of judgements, principles and theories to groups and that the method of reflective equilibrium can help us to identify these systems and can have an important function in finding gaps and shortcomings in the debate as it goes on. I do not want to argue here for the thesis that use of the method leads, or even should lead, to a state of reflective equilibrium with society or its belief system as its substrate. This may hold for some fundamental and uncontroversial domains of political theory on which modem states are built, of a legal or constitutional nature, but it cannot be said to hold for public debates that arise on the occasion of moral controversies.

Reflective Equilibrium and Public Debate

185

I conclude that the use of the reflective equilibrium method in public debates is not meant to lead 'the public' or 'the society' to a state of reflective equilibrium, but to help in bringing to light the variety of views on a particular issue and in making clear how the arguments and considerations for the various positions, are related, both within a position and between positions. There are two important corollaries to this conclusion. First, one could object that this view assumes that, within societal groups, there are stable identities and shared belief systems, and that the method of reflective equilibrium is aimed at disclosing these belief systems. But is this so? Is it not so that in a growing number of moral issues the opinions of people can no longer be classified along the familiar ideological and political lines? My answer to this is that the assumption is not relevant to my thesis. I even think that reflective equilibrium would be most useful in situations in which there is a set of prima facie unrelated intuitions, principles and viewpoints that cannot easily be related to traditional ideologies. I think that in these cases reflective equilibrium might give leads to organizing these elements in coherent positions, which, to some extent, are new and original. This is the more plausible because the problems (environmental problems, genetics, immigration, multiculturalism) are largely new and original. Second, one could ask who are going to apply the method of reflective equilibrium. Is it the task of ethicists to identify these positions and equilibria? Or should it be the opinion leaders or maybe the spokesmen of societal organizations? Let me first refer to a remark made above, that the method of reflective equilibrium is only offered as a method to be used by any participant in the public debate. So there is no logical connection between method and actor. However, considering the elements of the method I would suggest that the method points to a sort of division of labour between three (or four) groups of people. Intuitions are expressed by lay people, in the papers or on radio and television. These intuitions can be made more precise by opinion leaders and spokesmen of societal organizations, and they may be corrected by scientists who criticize the facts on which the intuitions are based. By way of these two processes, intuitions can take the character of considered judgements. Ethicists (of different origin and approach) can try to relate these judgements to existing moral principles or try to amend these principles for new cases, or even suggest completely new principles. Philosophers and scientists with a more general view can suggest background theories of a general scientific (example: evolution theory) or philosophical (example: the nature of a person) character. The upshot of all these contributions, ideally, is a relatively small number of defensible positions on the issue and an uncovering of the real, ethical and ideological, differences, for political choice. I do not think that the dynamics or the results of this process in which intuitions, principles and theories influence one another,

Jan Vorstenbosch

186

could or should be controlled by any particular group of professionals. In this sense, the method is without a 'subject'. But the method offers at least guidance in evaluating the process and the result to all concerned.

6

Morality and strategy

The second problem with putting the method of reflective equilibrium to use in public debates has two elements, both relating to the nature of debates on specific issues with a high potential for technological use and for political implications. The first element concerns the question of whether the political implications of public debates, as they will almost certainly occur, interfere with the ideal of a discussion on merits that public debates should strive for, and in particular the assumption of impartiality that informs the application of the method of reflective equilibrium. I do not think that this is a strong counterargument. Of course, specific participants in the debate, like politicians and representatives of interest groups, will take positions that are to a great extent strategic, or at least interest bound. But the advantage of public debates in this respect is that they display a wider array of opinions and arguments. Besides, the public will almost certainly recognize the strategic nature of arguments and positions on some issue. In the case of moral issues that arise in the wake of new technological findings, - and I concede that these fonn only a subset of public debates - there is an additional factor that makes the strategic nature of positions taken less significant. This is the fact that politicians themselves often are at a loss about what to think, because it is unclear what the relation of the issue is with their general programme. And as for interest groups, for instance business or patient groups, they do not know what their position should be either, because they do not quite see how the developments connect with ~heir own interests and ideals. It is true that in almost all debates there are exceptions to this rule, for instance in the case of science and technology, scientists have a clear stake in a liberal policy and in the case of environmental or animal issues there are strong pressure groups like the Animal Protection Movement and Greenpeace. But this does not detract from the fact that a public debate often is a sort of melting pot of ideas and that this melting pot is not dominated by clear cut interests. (cf. Trappenburg 1993 on issues in medical ethics). Often politicians and interest groups use public debates to see which way the wind is blowing. 6 6

I make an exception for 'public debates' that are orchestrated by the Government in the process of 'selling' a policy or by other institutions in the process of introducing a technology. Examples in the Netherlands are the campaign that accompanied the introduction of a new official policy on organ donation and the

Reflective Equilibrium and Public Debate

187

Incidently, if this analysis of the character of public debates is correct, this is an extra argument for considering the use of the method of reflective equilibrium in the ways I have in mind. If it is clear that a new development creates a kind of disequilibrium, a kind of tension with most of the accomplished belief systems in society, and groups are at a loss about how to respond, then the method of reflective equilibrium may serve to put things in place and help to find an answer in a more orderly way. Of course, in the first instance this applies to the use groups themselves, perhaps helped by philosophers or ethicists or other theorists, could make of the method. It is not an application that is available for my argument. This argument starts with another question, namely that of the significance of public debates for democratic decision making in society and the way in which reflective equilibrium can contribute to this significance. Nevertheless, it seems an essential stage in the decision-making process that the spectrum of coherent, defensible positions and relevant arguments comes to light and it is to the realization of this purpose that reflective equilibrium can contribute.

7

Public debate and science

The second element of this objection concerns the fact that in the case of scientific and technological findings there are often a lot of relevant empirical and technical details that are only available to scientists or insiders. This could easily lead to the argument that lay people, ethicists and others who are no experts should keep their mouth shut until they know what they are talking about, an argument that has a ring of obviousness. Along with this argument, and actually heard more often from the side of the scientists and technologists themselves, is the argument that it is much too premature to start a public debate about these issues because we know so little and, in time, everything will be different from what it looks now. These objections.. could end in a head-on attack on the idea of public debates and even democracy itself, but few people should like to stick their neck out for this conclusion. And rightly so, for it can easily be shown that however complicated the issue, if things are told in a clear and open way and empirical and normative questions are sorted out, it will be the normative, societal and moral aspects that are at the heart of the matter, and

introduction of the chip-card by the banking industry. It is clear that these 'debates' have little to do with my subject. (I thank Anton Vedder for bringing this kind of campaigns to my attention.)

188

Jan Vorstenbosch

these cannot be decided by scientific or empirical means. 7 There is no way to get around the public debate and, as a matter of fact, scientists who know how influential science and technology are for the course society is taking at the end of the twentieth century, should be the first to debate these consequences in public. In the background of this objection is the debate about technocracy versus democracy and, again, it would take us too far to take up that debate here. My position on this question, I think, is clear from what I have said. For argument, I should like to point to an article by Christopher Bertram (1997), which is relevant for the central elements of my argument. It hinges on the idea that 'the ideal of democratic community ... is one where social and political life is governed by principles publicly available to all' . As it is, Bertram directs his criticisms to philosophy and the use of general social and moral theory for justification in politics, but I think that his plea for public transparency could be directed with equal force against the appeal some scientists make to scientific expertise and scientific theory for circumventing extensive public debate on their activities. How does this debate relate to the theme of public debates and reflective equilibrium and especially the problem of the 'burdens of judgment' on the course of this debate? My main argument here is that the very fact that in public debates these aspects are kept in the background or playa minor role is an asset, and that this very fact contributes to the usefulness of the method of reflective equilibrium as a method for organizing normative statements, for public debates. Of course, this does not mean that empirical and technical elements should be ignored, let alone that they should not be understood properly. To ignore them is impossible, to misrepresent them is wrong. And the case I mentioned at the beginning, of the unreliable method for selecting a baby's gender, is a case in point, how important scientific data may be. But what the public debate should be about are, first of all, the possibilities, the cons~quences, the values and the norms, in short the relation of these scientific and technological findings and systems with society and the lives of the individuals that constitute the public.

7

In an article on 'good and bad debates', Propper and Witteveen (1995) discuss four kinds of criteria for good debates: criteria with regard to the quality of the participants, criteria with regard to the subject of the debate, criteria with regard to the debating procedure and criteria with regard to the public function of debates. Concerning the subject of a public debate, they remark that it ought to be a 'big' issue with a political and ideological significance and a structural purport for society, and in any case not a technical subject. Besides, they note that a second criterion is the inclusiveness of the debate: care should be taken that all aspects, viewpoints and arguments that are relevant to the subject are taken into account. This is in line with the gist of the position taken in this article.

Reflective Equilibrium and Public Debate

8

189

Pluralism and reflective equilibrium

The third objection to deal with is linked to this last remark. For if public debates are essentially normative and if, in all probability, the views, arguments and even the concepts are essentially contested, then it is almost certain that in most cases the result of a public debate as described above, will be plurality and conflict, even if it is led and evaluated by means of the reflective equilibrium method. For the most we can hope for if the reflective equilibrium method is used, are an articulated plurality and conflict between views that are coherent on their own terms. So, what is accomplished in a non-descriptive, normative sense by my plea for the use of reflective equilibrium in public debates? The first comment I want to make on this objection is that the critical force of the coherence criterion as it applies to the various positions in public debate should not be underestimated. It may well be possible to eliminate a number of ideas, proposals and arguments on the basis of this criterion itself. Michael DePaul (1993) has suggested that the aim and force of reflective equilibrium lie foremost in its potential to improve the rationality of our beliefs, and not in its capacity to lead us to truth or even warrant. Be that as it may, rationality offers us a normative standard, though not a moral standard. Considering the rational quality of public debates as they take place, it may be worth to try to improve it by taking a method like reflective equilibrium for guidance. Second, the utility of the method resides in the first place in its capacity for structuring and evaluating the process of debating, not in its capacity to assess the end result. It can help to make sure that all relevant elements are brought to the fore, all options are considered and principles and theories are duly noted. There is little doubt that, by doing that, the quality of the end result and of the arguments that it is based on, will be higher than without the method. As stated in the introduction, this deployment of pluralism is an essential and principled element in a democratic society. On this score, the approach taken fits in with the so-called Multiple Advocacy Model of policy making, defended by Majone (1989).8 On the other hand there is no need to canonize the method or to claim that it is the only way to salvation; it just helps. And it would certainly be going too far when it is claimed that the method can give us a guarantee that the end result is satisfying in all respects and for all parties. Third, I suggest that it is possible and sometimes even likely that, at least on a limited set of judgements and principles, and even background theories, there is more agreement than one assumes. The method of reflective equilibrium

8

'Multiple Advocacy is a process of debate and persuasion designed to expose the policymaker systematically to competing arguments made by the advocates themselves.' (Majone 1989, 183)

190

Jan Vorstenbosch

can assist in making this agreement explicit. This purpose is fairly close to the ideas and intentions Rawls had in introducing the method in A Theory of Justice and to the gist of his political philosophy in general, as it comes to the fore in the method of avoidance and his latest stance on the political meaning of overlapping consensus (Rawls 1971, 1993). By excluding all comprehensive philosophical stances and all substantial conceptions of the good that could lead to disagreement, and by limiting himself in the justification for the principles of justice to premises that are uncontroversial, Rawls hoped to create a generally acclaimed basis for his kind of liberalism. Unfortunately, this approach will not do - not for the principles of justice, as time has shown, but certainly not for the introduction of practices and technologies based on scientific inventions, which often run counter to the moral views of substantial groups in society. But what the reflective equilibrium method could do is, more or less on the lines of Rawls's overlapping consensus idea, demarcate a specific set of possibilities, policies and practices that run counter to what is valued by virtually all citizens. This may be a third kind of purpose or result to which the reflective equilibrium method can contribute. When speaking of the 'end result' in this sense, however, we hit upon the complicated relation between public debate and official policy. In any case, it is not so that the ideas that materialize in official policy exhaust the significance of a public debate. To my mind, the level of reflexivity with which society takes its course (Beck 1992) and the liveliness, broadness and quality of the debate itself are equally important. Moreover, the significance wide reflective equilibrium can have for public debates should not be confused with the significance this method can have for the formal political decision-making process or specific stages in this process, like the role of departmental advisory committees (see, in this volume, the contribution by Brom). For this process, wide reflective equilibrium could well be more like a hindrance, especially when it comes to finding compromises and consensus, mechanisms that are essential to the political process. In this context, we may expect more from narrow reflective equilibrium, in which background theories and philosophical considerations supposedly do not playa significant role (Rawls 1974-75). Awareness of the differences between public debates and the political process on the one hand, and between wide and narrow reflective equilibrium (although, for qualifying this difference, see Holmgren 1989) on the other, puts the functions of moral theory and moral philosophy into perspective. But it would take another essay and a different perspective to deal with this problem.

Developing Public Morality: Between Practical Agreement and Intersubjective Reflective Equilibrium Frans W.A. Brom 1

1

Introduction

Analysis of dominant norms and societal values shows that these values often change in the light of societal circumstances and technological developments. In our society, for example, dominant opinions on the treatment of men and women have changed in the last 60 years. And the possibilities brought forward by developments in modem biology and medicine call for a refinement of our opinions on the obligations to embryos and unborn humans. Can the changes in these dominant norms and societal values be considered to be a development of public morality? This question creates two problems. Firstly, what is public morality? It is clear that some dominant nonns and societal values give direction to public life in a pluralist society. In most Western societies, for instance, equal treatment of men and women is a fundamental norm in public morality. The question is, however, under what circumstances such a norm can be considered a part of public morality. Should it be shared by everyone? Does it need to be dominant in the public sphere? Or is acceptance by a majority good enough? Secondly, what kind of change do we call 'development'? It is clear too, that some changes in dominant opinions can be considered to be moral development. Most of us, for instance, consider the equal treatment of men and women a step forward. It is important, however, to note that not every change in dominant opinions should be evaluated, in a positive sense, as growth and development. The question, therefore, is: where do we find criteria to assess development? In this chapter I defend that public morality is a social enterprise in pluralistic societies. It has a discursive character, in which actual and practical agreements - and the argumentation these agreements rest upon - are amended and refined. It develops when people in a modem pluralist society try to elaborate initial social agreement for peaceful cooperation and social life. On the basis of these initial agreements they search for a moral consensus. There is, however, no Archimedean point on which this moral consensus can be built. Therefore, this 'moral consensus' is sought in the form of an intersubjective reflective equilibrium. Thanks are due to Marcel Verweij, from whose comments this article has profited. 191

w: van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

191-202.

192

2

Frans W.A. Brom

Individual morality in a social context

Before we can assess 'public morality' , we need to go into the question of what 'morality' is. Has every individual his or her own morality, or is morality something that is necessarily shared by a group? In an illuminating essay 'On defining "moral"', C.H. Whiteley (1959, 21) concluded that 'the philosophical problem is not that of giving a 'correct' definition, in the sense of one which accords to ordinary usage. For in ordinary usage the words 'moral' and 'morality' have no precise and consistent use' . According to Whiteley there are two nonreducible concepts of morality. The first is the morality of a community: 'The morality of a community consists of those ways of behaviour which each member of the community is taught, bidden and encouraged to adopt by other members' (p. 22) and the second is the morality of an individual: 'My morality consists, not in what other people insist that I should do, but in what I insist that I should do. (...) My morality comprises those actions which I think I ought to do regardless of inclination and regardless of personal advantage' (pp. 22-23). Of course, the 'two concepts of the moral coincide over a wide range: that is, the things that people think they (categorically) ought to do are very largely the things that are enjoined on them by their neighbours. But the coincidence between conscience and convention is logically contingent' (p. 23). I agree with Whiteley that the relation and overlap between individual and social morality are, at first sight, coincidental and logically contingent. But this is only the case as long as we leave the content of our moral obligations out of account. The gap between these two separated concepts of morality can be bridged by the content of moral obligations. When this content is brought into the analysis of the concept of morality, the arguments and reasons behind moral obligations become visible. It is then that we see that moral obligations exist, according to ourselves as well as according to our fellow human beings, on the basis of moral argumentation. Moral arguments bridge the gap between moral obligations according to the community and moral obligations according to ourselves. It is on the basis of moral argumentation that the obligations according to the community can be transformed into individual obligations. The role of argumentation in understanding the concept of morality can be clarified by the distinction between positive an critical morality (derived from Hart 1961, 1977). In a community or social group, a certain morality is actually accepted and shared. This morality is often called the positive, common or social morality. The simple fact that certain dominant patterns of behaviour exist does not, however, make these patterns 'moral'. Only if these patterns are (tacitly) accepted as the right social patterns, they can be regarded as social morality. The (tacitly accepted) idea that social morality consists of social patterns that aim at being the right social patterns, makes it possible to criticise positive

Developing Public Morality

193

morality. Positive morality and other actual social institutions are often criticised in the light of more general moral principles. These moral principles together form a critical morality. The critical morality, however, does not exist. There are several proposals for a critical morality. These proposals are not derived from an Archimedean point, but they come into being when we 'ask whether the things we acknowledge as goods are really valuable, if seen, in the light of all the relevant facts, and as part of the range of values we acknowledge as a whole' (Den Hartogh 1996, 69). Critical morality exists in this way as a 'hermeneutical construct' , meaning that the process of formulating critical moralities never ends; it exists in a repeating process in which a part of morality is criticised in the light of the whole of morality in order to amend this part (and thus morality as a whole). In the process from positive towards critical morality and back, we create our morality. This is, however, no creation ex nihilo, but a creation out of the morality left to us by our predecessors. In this creative process we transform some social conventions into our own morals, and in acting upon our own morals we form the conventions that will be the new points of departure. The dynamics of this process of individual morality making can be seen as a process of trying to reach a reflective equilibrium. By reflecting we aim for a reflective equilibrium of all elements that are (in some way) related to the positive morality from which we started. Morality is not an individual enterprise. It already existed before we came into being, and it will hopefully still be there after we have ceased to be. Human beings maintain an interactive relationship with each other in developing their morality. In theories on the development of individual morality, critical moral thinking is only possible after one has learned the moral rules of the environment in which one grows up (see, for instance, Kohlberg 1981). After I have learned the social morality of my parents and my community, I am able to start - on the basis of this acquired morality - with the development of my own morality. In this respect, morality exists in the same way as our language exists: we learn to speak the words of our ancestors, and with these words we are able to tell our own stories. The development of my morality is a quest for a reflective equilibrium between my experience and the different elements of the social morality I have learned (DePaul 1993).

3

Public morality in a pluralistic context

Morality consists of norms, values and ideals that guide the way in which we act and live. We act, however, in different contexts; one could say that our morality has different parts for different contexts. We have, for instance, a

194

Frans W.A. Brom

morality for the interaction with our parents, partner or children. The norms, values and ideals that guide our behaviour in these family relations are somewhat different from those that guide the way we behave towards our colleagues. We have, in a similar way, a part of morality that consists of ideas on how to interact with non-specific members of the communities we live in. By 'non-specific' I mean that it refers to those people in a way that has no immediate relation to them as a specified person or as specified persons. I call this the morality of the public sphere. I use 'public' here in the way Sir George Cornewall Lewis coined it in 1832: 'Public, as opposed to private, is that which has no immediate relation to any specified person or persons, but may directly concern any merrlber or members of the community without distinction.'2 In this way, everyone has his or her own public morality. Do we have a common public morality? It seems clear to me that in a homogeneous (non-pluralistic) society people will share this public morality. They have a common morality of the public sphere. This public morality provides the basis for societal peace and societal cooperation. In a pluralistic society, however, there are social groups and communities that each have their own social (common) morality. In modem democratic and pluralistic societies, these different social moralities are in a certain way distinct from each other. Some authors even defend, in the line of Alasdair MacIntyre, that these different social moralities arise from radically distinct moral traditions. These moral traditions each have their own moral language and moral rationality. Do these differences exclude moral communication and a common public morality between these distinct moral traditions? According to, e.g., Kees Klop (1997, 14) they do: there may be an overlapping agreement in practical judgements and there may be pragmatic compromises in conflicts, but these agreements and compromises create no common public morality. Klop defends the thesis that modem pluralistic and democratic societies have no common public morality because these societies consist of distinct social groups and communities. This idea of a 'segregation' in public morality presupposes distinct competing traditions, each of which must have sufficient internal coherence so that it can be considered to be one moral tradition, but with enough differences between them so that they can be regarded as distinct moral traditions. I think this 'segregation' of the morality of modem society into distinct moralities is too radical. Firstly, this view of a moral tradition ignores the heretics within each moral tradition. The internal coherency of moral traditions has often, at least in the 2

G. Cornewall Lewis, Remarks on the Use and Abuse of Some Political Terms, quoted in Barry (1965, 190).

Developing Public Morality

195

past, been built upon the exclusion of deviating voices. Historical analysis sometimes gives reasons, as for instance feminist thinking shows, to re-evaluate a tradition's coherence. Heretical voices brought forward points within a certain tradition that are now elaborated in rival traditions. Re-evaluating the exclusion of these heretical voices can start communication with the traditions that arose from these voices. Secondly, this view does not fit with social reality in modem Western societies. In most Western societies, people do not live segregated lives; they participate in various practices and traditions. People can, at the same time, be politically progressive (and members of a left-wing political party), conservative as regards technological developments (and members of an anti-technology ecological movement) and religiously liberal (and members of a liberal church). Or, more specifically, in the political spectrum of, for instance, the Netherlands there are Catholic members of the Labour Party and Hindu members of the Christian Democrat Party. Thirdly, this idea of a radically pluralistic society ignores the fact that there is a common tradition of searching for peaceful cooperation and social life. Although not everybody participates in this common quest, it is dominant in modem societies. This is what Brian Barry (1989b, 164) calls the liberal tradition in modem society. The main elements of this tradition 'were originally adopted for reasons that made sense to people in terms of existing outlooks. Religious toleration seemed better than religious warfare, constitutionalism arose from the desire of various interests to protect themselves, and so on. Later, people like Kant and Mill came along and offered general theories from which, they claimed, the characteristic liberal prescriptions could be derived.' The practice of social life creates practical agreement on common points of departure. It can be doubted whether this agreement is a substantial common consensus. However, it seems to me substantial enough to function as a point of departure for common moral deliberation. In modem Western societies, different moral traditions are not as distinct as they are sometimes considered to be and people are usually not 'locked' within one tradition. Moreover, there is a common tradition of peaceful cooperation. This implies that it is possible to construct a tradition of an 'overlapping discourse' between individuals who do not necessarily share the same ideas and views of life. This tradition of an 'overlapping discourse' differs from Rawls's notion of an 'overlapping consensus' (1993, 144 f.). Rawls's notion of an overlapping consensus is too complicated to discuss here in detail. Instead, I will indicate three differences between public morality as an overlapping discourse and the notion of an overlapping consensus. Firstly, there is a difference in the sources of the overlap. The overlap in overlapping consensus concerns reasonable comprehensive doctrines. These

196

Frans W.A. Brom

doctrines all support - on the basis of their own non-public justifications belonging to the many comprehensive doctrines and acceptable only to those who endorse them - the same fundamental idea of a political conception of justice. The overlap in public morality as an overlapping discourse concerns the moralities for the public sphere different individuals have. They are able to discuss the content of public morality on the basis of some initial agreements. These agreements follow from the common tradition of peaceful cooperation. Secondly, there is a difference in the object of the overlap. The object of an overlapping consensus is a political conception of justice. The content of this conception is not limited to what is agreed upon by the various reasonable comprehensive doctrines. A political conception of justice needs to be freestanding, that is, its content can be presented as independent of the various comprehensive doctrines. Therefore its content is limited to 'fundamentals' such as 'constitutional essentials' and questions of basic justice. The object of public morality as an overlapping discourse is the expansion and refinement of the common tradition of peaceful cooperation. This makes, for instance, that the border between public and private is not theoretically determined but can be considered to be a topic of deliberation in this discourse. And thirdly, there is a difference in the rationality of overlap. The rationality of an overlapping consensus is that of public reason. Public reason abandons arguments that fit into the different comprehensive doctrines. It is freestanding and allows only those arguments which reasonably could be expected to be endorsed by others as consistent with their freedom and equality (Rawls 1993, 218). The rationality of public morality as an overlapping discourse is that of an intersubjective reflective equilibrium. This will be elaborated in the next section. As I will show there, this implies that not only those arguments are allowed that could be expected to be endorsed by others as consistent with their freedom and equality, but also those arguments that could be expected to be endorsed by others as following from the concrete practice of peaceful cooperation. In a pluralist society, common public morality derives from an overlapping discourse. This discourse is characterised as a deliberative interaction between positive and critical social morality. On the one hand, this interaction departs from positive social morality, that is, from defacto agreement; on the other hand, there is critical morality, being the proposals for amendment and refinement of this social morality. Maybe not everybody in modern society is included in this process. Yet, if it is shared by enough people, it can be considered to be a common public morality of a modem society.

Developing Public Morality

4

197

Towards an intersubjective reflective equilibrium

The public morality of our modem Western societies can be seen as the critical scrutiny and elaboration of practical agreement in our society. It is not a 'positive morality' we can apply to concrete moral questions, but it exists in a common practice of critical moral thinking that concerns our positive morality, and also starts from it. In this sense, 'public morality' not only provides the context for our actions, but it also shapes our sense of meaning and values of actions about which we deliberate. In this way, it provides the social context in which we deliberate as well as the norms, values and ideals we deliberate on (Postema 1992, 164). An analysis of this content of public morality cannot be confined to actual (coincidental) moral agreement. It is more than just the actual and practical societal agreements people happen to share. The moral argumentation on which this agreement can be based and which also provides a basis for a critical assessment of this agreement is a necessary part thereof. In our analysis we need to go beyond opinions held by members of our society and look at the more fundamental moral convictions held in our societies. 3 We can only get a clear picture of public morality if we include in our analysis the exchange of arguments on which practical agreements are based and in which these agreements are expanded, amended and refined. Therefore, we need to include the processes of change in our concept of public morality. Its core seems to be in a continual process of interaction in which actual practical agreement is criticised on the basis of proposals for better agreement. In this interaction process, positive morality stands for the norms and values experienced as valid and important in our society and critical morality for analysis, interpretation and rational moral argumentation at the service of democratic procedures and decision making. The problem, however, is that there is not one but that there are several critical moralities from which proposals for better agreement follow. The question, therefore, is how to decide which proposal can or should be endorsed. Central to the idea of WRE is that there is no Archimedean point from which we can derive criteria that may be helpful in answering this question. For the development of individual morality this leads to the idea that every individual has to search for a reflective equilibrium between his or her experience and the different elements of the social morality he or she has acquired and accepted. Is it possible to amend this notion of WRE in a way that it can serve as a model

3

The importance of going beyond 'actual opinions' to 'more fundamental moral convictions' held by people involved in a certain situation is analysed elsewhere in this volume by Johannes van Delden and Ghislaine van Thiel.

198

Frans W.A. Brom

for the dynamics of public morality? I think that a positive answer to this question is possible. In the theory of WRE the development of individual morality is reconstructed as an individual's deliberative process in quest for a RE. From this individual process we need to go on to a process of societal deliberation. A first possibility is that this deliberative process strives for an overlapping agreement between distinct equilibria of distinct individuals. Every individual searches for his or her reflective equilibrium, and there is a common morality in so far as these equilibria overlap. This does not seem helpful to me. Firstly, because in this process there is nothing that helps the individuals to aim at an overlap. And secondly, because such a process leaves the critical scrutiny and the rationality of this social enterprise entirely up to the separate individuals; the social character of the deliberative process is left aside. Everyone strives for his or her own RE, and there is no joint search. Instead of an overlap in distinct REs something else is necessary. A social deliberative process as a joint search of different individuals, should search for something these different individuals share. This shared goal should also be the result of joint critical reflection. In this way, the common reflection directs the process towards overlap. One could, therefore, say that a social deliberative process should search for a common consensus shared by different individuals. The concept of consensus is, however, problematic. It is not necessarily the same as social agreement. The relation between agreement and consensus depends on ideas on moral truth. In the context of this chapter I cannot go into epistemological subtleties (see Van Willigenburg in this volume), but an analysis of Alberto Bondolfi (1994, 124) may be helpful. Bondolfi distinguishes between a substantial conception of the truth and a sceptical one. If there is a substantial and knowable truth, agreement is consensus if and only if it is in accordance with this truth; and if there is no moral truth whatsoever, every agreement whatever its source may be - is consensus. Accordipg to Bondolfi both positions are problematic. In the first, where 'consensus is seen in a close relation to a substantial conception of truth and is perceived as nothing more than consent, an ethical act by a free subject can no longer be achieved'. And in the latter 'where consensus is looked upon purely as a declaration of one's consent, without any argument whatsoever - the ethical is confused with the factual results or with that which is laid down by authority' . Both positions contradict the creative and reflective process in which individuals are trying to reach a reflective equilibrium and thus create their own morality. Neither 'moral truth' nor 'factual agreement' can work as an Archimedean point for this process for individuals and, therefore, neither can work as an Archimedean point for public morality as a social enterprise. A consensus should therefore be characterised as a common RE shared by different

Developing Public Morality

199

individuals. In such a common RE, people do not merely agree; their agreement is the result of a common reflective process in which they have aimed for a common equilibrium. Not only do they share a certain judgement, but they also share - after critical scrutiny - the arguments that support this judgement. Such a consensus can be seen as an 'intersubjective reflective equilibrium' (IRE), a reflective equilibrium shared among different subjects. Against my thesis it can be argued that the idea of an IRE is an unrealistic and unreachable ideal. However, that an IRE is unrealistic is no good reason not to strive for it. Moreover, according to Rawls (1993, 97): 'The struggle for reflective equilibrium continues indefinitely, in this case as in all others.' For him RE is a philosophical ideal as well. In fact, its idealistic character can provide us with a source for the critical assessment of concrete attempts to develop public morality. As Wibren van der Burg defends (in this volu.me), ideals can, just because they transcend factual agreement, help us to criticise factual agreement. To conclude, I reconstructed public morality as an overlapping discourse in which different individuals cooperate in the search for a moral consensus. This moral consensus can best be understood as an intersubjective reflective equilibrium; this is an equilibrium shared among different individuals upon joint critical reflection.

5

Whose deliberative process?

The development of public morality is most wanted in situations in which it is not clear which nonns and values should set public standards. Those situations are considered morally problematic, and in such situations governments often install an ethics committee to solve the problem. One of the ideas behind such a committee is that a small group of persons is better able to reach a rational consensus on such a problem. In a small group direct contact and conversation are possible, and this enhances the exchange of argument. As Kurt Bayertz (1994, 13) states: 'Microinstitutions such as ethics committees and similar bodies are attractive because they present an opportunity for direct communication, an exchange of arguments, and consensus formation.' The problem, however, is that ethics committees seldom reach rational consensus. In practice they reach at best a compromise which is not unacceptable to most members. And even if members of a committee are able to reach agreement on what should be done, the arguments that support their joint decision often differ: an intersubjective equilibrium in a committee is seldom 'reflective', that is, based upon joint critical thinking.

200

Frans W.A. Brom

But even if a committee reaches consensus, what is its value? Behind the creation of an ethics committee lies the idea of 'authority of moral consensus'. According to this idea, a consensus among a group of experts in an ethics committee sets standards for society to cope with a problem. It is necessary to rely on the 'authority' of a consensus because we have - in our pluralist societies - no common basis for moral authority. In the words of Bruce Jennings (1991, 450): 'The renewal of interest in the role of consensus in morality and ethics grows out of the recognition that every other basis for the authority of moral claims - essentially, transcendental reason or revelation - has equal or greater doubts and problems associated with it.' But he adds that it 'is better understood as a necessary precondition for the achievement of moral authority than as a foundation or basis for that authority'. The consensus in a committee, however, is the consensus of others. And there can be a difference between the consensus of others and my own conclusion. If I think that the conclusion I have reached is warranted, its warrant depends - in itself - not on the question whether or not other people agree with it. Their consensus can only function as a moral argument if there are reasons (besides their consensus as such) to trust their judgement. A reason for trusting the consensus judgement of a committee is that the members themselves are trustworthy as moral experts. This can be of help in creating a practical agreement. This is the case when we accept to leave a decision to the deliberation process of a certain group of qualified persons. In those cases we can accept a given consensus of a committee as a practical solution in a concrete case. In the societal search for IRE, however, this is not good enough. In an IRE people share certain judgements as well as the arguments that support these judgements. In order to know which arguments could support a certain judgement, an exchange of arguments is necessary. This implies that the arguments held in common have survived critical scrutiny. Such an IRE can thus only be reached after a free exchange of arguments and opinions. If we want to develop public morality, we need to base it on a free exchange of arguments. Such a free exchange needs to take place in the open; in the public sphere. Hence, the development of public morality should take place in public debates. 4 In these public debates ethics committees can be of help because they provide good arguments. Seen in this way, an ethics committee serves a heuristic goal for public debate. Its function can be to argue for certain proposals for agreement. In these proposals factual information and the different empirical and normative perspectives need to be integrated. It is important that the committee will not search for solutions in vague arguments in which everyone 4

The concept of 'public debate' is elaborated by Jan Vorstenbosch in this volume.

Developing Public Morality

201

can find what he or she wants, but that they have the willingness to show disagreement when such is necessary for the public debate. 5 It is clear that developing public morality is not possible if the argumentation process stays within the committee. Therefore it is necessary that the opinions and arguments are spelled out as fully as possible and are made public. The consensus within a committee, as a consensus of others, can only be authoritative if it functions as an argued proposal. In this way it can be helpful for the development of public morality because it fosters a public debate. The arguments, however, need to be scrutinised in an open discussion. Therefore, in a quest for an IRE it is not only the search for consensus that is important; the expression of argued dissent is as important. Without dissenters arguments are not criticised. I want to stress the importance of dissent because the quest for an IRE seems to presuppose that consensus is good and dissent is bad. This is only partly true. In an IRE the importance of consensus is based upon 'intersubjective reflection', not on the agreement as such.

6

Conclusion

In this chapter I looked for the dynamics and the rationality of the development of public morality. I showed that the theory of WRE can be helpful in reconstructing the development of the morality of an individual in a social context. In a pluralist society a part of the morality of every individual consists of norms, values and ideals for peaceful cooperation with those who do not necessarily share the same ideas or views of life. For peaceful cooperation and social life in a pluralist society it is necessary that there are some practical agreements (overlap) between these moralities for the public sphere. Fortunately, this is the fact in most pluralist societies. In (most) pluralist societies there is an overlapping discourse in which the content of these practical agreements is discussed. I consider this overlapping discourse the· 'public morality' of a pluralist society. In this overlapping discourse actual and practical agreements - as well as the moral argumentation these agreements are based upon - are criticised, amended and refined. Such a discourse is aimed at a moral consensus. There is, however, no Archimedean point on which this moral consensus can be built. We have no criteria for knowing when a certain agreement can be regarded as 'moral consensus'. Therefore, an overlapping discourse can best be considered to be a kind of WRE process. Such a discourse is, however, a social enterprise. Therefore, the WRE process exists in striving towards an intersubjective 5

For a further analysis of 'ethics committees' see Egbert Schroten in this volume.

\202

Frans W.A. Brom

\reflective equilibrium. It is in this process that the public morality of a pluralistic society is developed. The problem is that such a process seems to be too difficult for a society. Therefore, a government creates ethics committees when it thinks that society is in urgent need for amendment and or refinement of common public morality. Such a committee, however, cannot substitute social deliberation and lay down a certain morality for society; the best it can do is to assist society in developing a common public morality. I

II

The Case of the Shared Values: An Example of a Socratic Dialogue l los Kessels

1

Introduction

In his much quoted article 'Outline of a decision procedure for ethics' (Rawls 1951, 177), John Rawls distinguished two types of question concerning justice, only one of which he intended to inquire into: 1. Does there exist a reasonable method to settle disputes about competing interests? 2. How is this method effectively applied in the settling of disputes? The first question is the one Rawls tackled, proposing as an answer his reflective equilibrium method. The second question he left supposedly to practitioners of different disciplines and their theorists, like pedagogues, educationalists, etc. In this article, I will investigate some issues concerning the second question. It is important to note right away that the term 'method' is vague here, and may be misleading. Rawls uses the term to refer to the logical requirements for settling disputes: he defines what a competent moral judge is, what the determining characteristics are of considered moral judgements, what is meant by a satisfactory explication of a judgement, etc. There are, however, different ways or methods to meet these requirements and to produce a thus defined reflective equilibrium. In this sense the term 'method' refers not to logical requirements, but to what Dewey called psychological or chronological requirements, Le. the actual steps, interventions and procedures needed to bring about a reflective equilibrium in a given case. 2 In this article, I will investigate some issues concerning method in this second sense. My aim is to show that, in order to construct an actual reflective equilibrium, one has to overcome several obstacles, concerning both the contents of the inquiry and the interaction between the participants. I will concentrate on a classical method for the investigation of moral and other philosophical questipns, the Socratic dialogue, in our century revived by the work of the Gennan neo-Kantian philosopher Leonard Nelson and his student Gustav Heckmann (Nelson 1970; Heckmann 1993). Why the Socratic dialogue? What is the relevance of the Socratic dialogue for reflective equilibrium? Well,

1 2

An earlier version of this article was published in Kessels (1997). For this distinction cf. Dewey (1916/1966,220); Kessels (1989,38). 203

~

van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 203-215. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

204

los Kessels

in his dialogues Socrates was continuously searching for reflective equilibrium. (Rawls considered moral philosophy explicitly as Socratic; Rawls 1971,49.) And he used specific ways and methods to avoid or overcome the difficulties connected to his search. For instance, he was not interested in all and every inquiry. He was only interested in inquiries that could be performed in an interpersonal, dialectical manner, by question and answer, through dialogue. Socrates talked with people, he never put anything on paper. He could not stand long speeches, nor did he believe in the value of mere texts (cf. Phaedrus 275e, 275a-b). Next, he was only interested in an analysis of ideas if that implied an analysis of the persons proclaiming these ideas. Moral or philosophical analysis had to be a form of self-investigation. Purely abstract, hypothetical discussions were not his cup of tea (cf. Protagoras 331 c, Gorgias 495a, Crito 49c, Politeia 346a, 350e; cf. Seeskin 1987, 1.1.). Besides, his inquiries were based on the assumption that the knowledge he was after, the right answer to a fundamental question, somehow was already there, in the person, in his or her experience. So, personal experience could not be left out of the investigation, at the risk of lapsing into a facile, empty, meaningless exercise. True, knowledge in his view was a prerequisite for real virtue. But this was not so much the theoretical knowledge of the scientist (episteme) as the practical knowledge of the wise person (phronesis) (cf. Kessels and Korthagen 1996). There is no need to go into the metaphysical status of these assumptions here, I would like to concentrate on their heuristic validity, their usefulness for dialectical practice and for bringing about reflective equilibrium in a given case, Le. Rawls's second question. The above-mentioned characteristics of Socrates' approach form the basis of Nelson and Heckmann's neo-Socratic method of 'regressive abstraction'. This implies that a questioner tries to lead a group of answerers to a deeper insight into some fundamental question through the analysis of a single, realistic example, stemming from the experience of one of t~e participants. Starting from this concrete example, the questioner asks 'backward', investigating the assumptions that the experience and the interpretation of the example are based upon (regression). By making these explicit, Le. by analysing what we have to presuppose to be able to make the judgements that we actually make, we may discover the rules and principles on which these judgements are based. This makes it possible to examine them, sharpen or justify them, and hold them to scrutiny. Thus we may develop general insights (abstraction) and produce reflective equilibrium in a group. In this article, I will inquire into how this modem version of the Socratic method can be applied as a tool for achieving reflective equilibrium. I will give an example of how the method was applied in a concrete case, indicate some of the problems that arise in such an investigation and evaluate the results.

The Case of the Shared Values

2

205

Case study

In the case that we will investigate, the Socratic dialogue was held in a business firm, Macmillans, and consisted of three sessions: an introductory meeting, the main dialogue and a follow-up session. The introductory meeting is meant to delimit the topic to be examined as sharply as possible, by formulating it as a single, fundamental, non-empirical question. Let us start by looking at how this went off.

2.1

Introductory meeting: formulating the question

Macrnillans is an international trading company with branches all over the world. The company includes ten divisions and more than sixty branches. Four years ago, Macmillans bought Vasco, a group of four distribution enterprises. The perfonnance of these enterprises was not bad, but their culture was very different from that of Macmillans. This became apparent in all kinds of issues, small ones and bigger ones. For instance, Vasco did not want to cooperate in the company's general policy of creating children's creches. It did not mind that it did not have a well functioning works council. And particularly, it practised a hire-and-fire policy for its personnel: employees were dismissed after a short time (two years). This went strongly against the grain with Macmillans, such a policy being in sharp contrast to the way Macmillans was used to treat its personnel. Until some years ago, it was customary that once people got a job there they stayed there until retirement. Macmillans was a lifetime employment business. Nowadays this cannot be kept up any longer. But Vasco's policy was too much at the other end of the line. This situation raised all sorts of questions for the Board of Directors. In the introductory meeting, with three of the six participants present, these questions were formulated as follows: Is a hire-and-fire policy acceptable? Should we have a common policy on the conditions for employment? Is Vasco a Macmillans company? What is it that unites us? What does the name 'Macmillans' stand for? Next, the group examined which of these questions should be the starting point for the analysis in the main dialogue, which question was central to the problem they wanted to "investigate. This led to still another wording: Which values and norms should all Macmillans companies share?

206

los Kessels

This question, according to the participants, hit the core of the problem with Vasco. Besides, it was also an important question apart from the Vasco example. Everyone agreed to its selection as the starting question. Let us have a closer look at these questions. The first question - Is a hireand-fire policy acceptable? - seems to be on a first-order level: There is a problem and something should be done about it, so we should look for some direction to solve it. The next question goes deeper: Should we have a common policy on the conditions for employment? That is: Should we really issue such a directive? And if so, within what framework and on what grounds? And in the questions that follow, the fundamental nature of the problem shows: Is Vasco a Macmillans company? What is it that unites us? What does the name 'Macmillans' stand for? Which values and norms should all Macmillans companies share? These questions touch the level of principles. They are not 'how' questions, but 'why' questions. They inquire into the underlying reasons for some possible action, and further into the role the company aims to fulfil, the identity it wants to have, the self-image of the group and a valid justification for that. 3

2.2 The main dialogue. The example In the main dialogue, two weeks after the introductory meeting, there were six participants: the three members of the Board of Directors and the Chief Executive Officers for Legal Affairs, Financial Affairs and Human Resources. At the start of the dialogue, each of the participants related an example from personal experience in which the starting question played a central role. After a short deliberation, Johnston's example was selected for further analysis. Although this example was not directly concerned with Vasco, it concerned the question of when precisely a dismissal is acceptable, and thus also the main question which values and norms all Macmillans companies should share. Besides, it was a simple example and well-known to all participants. Johnston (Legal Affairs) explained the case in detail. George Henry, a 55year-old who had worked for Macmillans for more than 25 years, was now being dismissed. Johnston explained his hesitations concerning this dismissal. True, the man could no longer be employed in the near future, because his position had been cancelled. But that could change shortly. He was a man with a great deal of experience and a great loyalty towards the company. Besides, a dismissal entailed costs (a number of months salary). The example was clear to everyone. But not everyone agreed with Johnston's view. Pears (Human Resources) pointed out that although in this case the expectation of a lifetime employment could not be met, it was a 3

Cf. Swieringa and Wierdsma (1992), chI 5.

The Case of the Shared Values

207

'Macmillans-worthy' dismissal, it met the Macmillans norm. He gave the following reasons for his point of view. Acknowledgment of the long-time service was expressed in the way in which Henry had left: In fact, he had received early retirement. He had been given a golden handshake. He got a transition period, a year's time to find another position. Besides, the decision had been taken very carefully. Pears concluded that he, in contrast to Johnston, had no hesitations concerning this dismissal. On the contrary, he fully supported it. These two points of view elicited several reactions, which made the conversation somewhat confusing. Hardwick (President of the Board) presumed that Henry was a victim of the 'Peter Principle' , the principle that many people keep climbing higher until they end up in a position that they are not fit for. Harrison (Board) approved of Henry's dismissal, but not of the norm that Pears explained. In his view, a decision to dismiss should be taken carefully indeed, but without a golden handshake or a transition period. This might have been the standard up to then, but in the future it should be done differently. Melford (Board) wondered whether there would have come a moment for Henry to be employable again, as Johnston supposed. 'And should you keep someone waiting for a train that does not come? Should you hold someone engaged if there is a fair chance that no function will become available?' Foster (Financial Affairs) supported this point of view. 'We are not married to each other. We do not have an engagement till death do us part.' As far as he was concerned, the case was clear; people should simply be judged upon their accountability, that was the norm. Harrison agreed. You should not reward people for the length of their service, only for their performance. But in that case, Melford replied, Henry should have been told at some point that his performance was not satisfactory. And that had never happened; he had always been given the impression that his performance was good enough. 2.2.1 Topical versus strategic dialogue Here a strategic dialogue was inserted. For different points were brought forward here, which made it difficult to keep the central line of the investigation in sight. It seemed as if the participants were giving answers without having posed any questions. Which questions were being answered here? And which of them was the most central one? After some searching, a distinction was put forward between the norm that used to be valid in the past and the norm that should prevail. The central question then was: What precisely had been the standard till then, and what was the norm that should prevail? This often happens in a conversation: Answers are being given before a single question has been stated. As a consequence, participants are mainly trying to answer their own individual questions, resulting in a propounding of individual

208

los Kessels

opinions instead of a joint investigation. In order to keep the subject of conversation shared by all and to make progress in its joint analysis, it is essential that every participant at every moment has a clear view of what precisely is the subject of investigation or the question being examined. In one dialogue, it may be necessary to make this explicit several times. To that purpose, every now and again a strategic dialogue should replace the topical dialogue: The inquiry into the starting question and the exarllple to match have to be interrupted by a secondary inquiry into how best to organize the first, topical inquiry. A strategic dialogue is not only necessary to keep the participants focussed on the subject, but also to keep the subject of investigation clear. For the subject and the starting question are continuously being transformed during the inquiry. For instance, in this dialogue the initial question of which values and norms all Macmillans companies should share, is first transformed into the question of when a dismissal is acceptable, and next into the questions of what used to be the norm in the policy on dismissals and what that norm should be. It will become clear further on in the dialogue that these questions too will have to be made more precise. 2.2.2 The old, present and new norm I will skip the detailed account of the next part of the dialogue and limit myself to a brief summary. For a practitioner, the many winding roads a Socratic dialogue takes are the most interesting, because of the challenge of keeping clear the central line of investigation. But here we will not go into them. 4 The first person to answer the question of what used to be the norm till then and what it should be, was Hardwick. In his view, the nonn that had always been valid in the past was: Do not do stupid things and you will stay till you retire. In other words, loyalty gets rewarded. But Pears disagreed. For if that really had been the norm, Henry would not have been dismissed. He proposed to make a threefold division: the old nonn fronl the past, the present nonn that had been followed in this example and the nonn that should prevail in the future. The norm that Hardwick formulated was the old norm. The present norm, the one that was followed in the example, had been fonnulated by Pears earlier in the discussion: carefulness in taking a decision, a golden handshake and a transition period. But this norm, Harrison repeated, should not be the prevailing one in the future. 'It is no good putting someone in the waiting room like that. You'd better say good-bye to each other, attentively but immediately.' So what should the norm be in the future? Here the windings started. Several formulations of the norm were tried, of which two remained in the end, a harsh one which should prevail when 4

Those interested may read Kessels (1997), ch. 2.

The Case of the Shared Values

209

somebody did not function well (Harrison) and a mild one when someone did function well (Hardwick). A complication in Henry's case was that, although he had not been functioning well, this had never been made clear to him. Still these proposals did not get common assent at first. Foster, for instance, did not agree with considering these two norms to be general norms. He claimed that 'when someone is really good, you do not let him go. So in that case you do not have to arrange an exploration period.' At this point we started investigating a counterexample. But soon time was up. Hardwick inquired finally how Johnston, after this investigation, viewed his initial hesitations as to Henry's dismissal. Johnston replied that he supported the proposal about the two new, complementary norms. There was general satisfaction; the participants had experienced the dialogue as pleasant and productive. 2.2.3 Argumentation analysis, discipline It may be clarifying to consider the development of the dialogue in terms of the figure below, concerning the structure of an argumentation. 5 First of all, the figure shows that in this method a starting question is not answered straightaway, but only after a the detour of connecting it intimately to facts and experience. Thus a large, fundamental question is limited first to a single example and then narrowed down to a specific judgement. Next, the assumptions of this judgement are investigated, on different levels, producing statements of increasing wideness and generality. Hence this progression results in a model having the shape of an hourglass. With the help of this model we can get a clearer view of several aspects of the dialogue, for instance the transformation of the starting question. The original formulation - Which values and norms should all Macmillans companies share? - is first applied to the example, resulting in the transformed question: What is our norm for dismissal? When is a dismissal acceptable? This leads to a number of judgements or standpoints concerning the example: The dismissal causes hesitation. It does not cause hesitation because the decision was carefully arrived at. Though the decision was made carefully, a different kind of decision should be made in the future. This generates the question which of these judgements is to be considered the right one, which requires looking into how a judgement is to be justified: What should be our norm in the policy on dismissals, and why? Thus, narrowing down

5

The figure is based on Leonard Nelson's philosophy, see Kessels (1997). It contains the basic structure of the Aristotelian syllogism, rules and principles being the major premiss, the example (or some core sentences from it) functioning as the minor premiss and the judgement as conclusion. See also Toulmin (1958).

210

los Kessels

the question is achieved here in a number of steps: applying the question to a specific example, formulating possible judgements, justifying these judgements in terms of underlying assumptions or rules and, finally, assessing the justifications of these rules with the help of principles. Next, the interpretation of the facts of the example depends on the norms being used. In terms of the model this means that our representation of an example depends on our judgement on it and on the assumptions we hold to justify this judgement. Consequently, in order to reach a consensus - the joint affirmation of having achieved reflective equilibrium - we have to agree on at least three matters: the facts of an example; a judgement or standpoint concerning the facts; and the justification of that judgement or standpoint (cf. Van Willigenburg and Heeger 1991; Heeger 1990a). At the beginning of a dialogue, these three continuously intermingle, until in the course of the investigation they are disconnected, sharpened and converted from an individual belief into a common view. For instance, Johnston's initial representation of the facts is this: Henry is being dismissed. He has served Macmillans for more than 25 years. At the moment, there is no other position available for him. But there may be a position for him in the near future. Besides, his dismissal entails costs. But others regard this version as erroneous, or at least inaccurate, because an essential element is missing, namely: Henry did not meet the requirements of his function. But he was never informed about this. So, facts are perceived in a different light according to the normative assumptions with which we interpret them, which may lead to different judgements on these facts. When, at the end of the dialogue, agreement seems to be reached on a common interpretation of the facts, this implies that, at the same time, an agreement is reached (here still latent) on a judgement about these facts and on a justification of this judgement. In order to reach such an agreement, however, it is essential to disentangle the different standpoints and arguments into their composite elements: question, facts, judgement and justifying assumptions. An essential point here, which is mostly ignored or overlooked by theorists but vexing any attempt to achieve an actual reflective equilibrium, is that the concepts in the starting question, like 'norm', cannot be defined in advance. It is no use trying to fix the meaning of such a concept in an abstract, purely verbal way at the start. Many conversations are unprofitable and do not lead to inquiry because they remain at an abstract, verbal level, resulting from the

The Case of the Shared Values

211

erroneous belief that we must define our terms first before we can speak about the matter itself. In fact, it is the other way around: first we have to analyse the matter itself, only then is it possible to come to a clear definition of a term. 6 So, it is much better to make the term concrete first, to translate it into events, facts and aspects of the example. For instance, one way to make the concept of 'nonn' concrete is that of Johnston's: 'A man with a great deal of experience and a great loyalty towards the company, like George Henry, is not to be dismissed just like that' . By thus connecting the concept of 'norm' to a concrete example, its multitude of possible meanings is transformed into a single, plain story. Let us conclude the analysis of this case with a brief look at the follow-up session and the final results of the investigation. It shows the most interesting features of the Socratic dialogue.

2.3 The follow-up session: Revaluation of the question Two weeks later the follow-up session took place. All participants had meanwhile read the dialogue report. It appeared not to be difficult to formulate the consensus, which was still implicit in the final part of the main dialogue, explicitly at this point: The 'old norm' was no longer valid. The requirement of carefulness according to the 'present norm' was to be kept up. As to the golden handshake, legal regulations were to be applied (which left some room). If no function is available for an employee, this must be told as soon as possible. If someone does not perform well, this has to be made clear a number of times, together with a limit; if at the limit there is no improvement, then the employee is dismissed. If someone does perform well but the function is cancelled, this has to be told as soon as possible; the period between informing the employee and ending the function should be as short as possible; it should be indicated what is to happen when the function is ended: after dismissal, both the company (for a limited period) and the employee dismissed are to make an effort to find another position. Next, we looked into the way in which this result related to the initial question. True, we did reach a consensus about a norm concerning the policy on dismissals. But was this, then, a norm that all Macmillans companies should share? After 6

The philosophical background of this point is explained in Kessels (1994), note

40.

212

los Kessels

all, that was the question for which we started the investigation. Should this norm then be imposed on the Vasco firms? Harrison presumed that a general problem came to light here. All participants at the dialogue were employed at the main office of Macmillans, and the culture within this office was very different from, for instance, the branches abroad. So we here might consent to a particular norm. 'But that does not mean that this norm can or should be imposed on others outside this building.' This evoked the question: 'Is it after all so serious that the norm fonnulated by us is not shared by others, for instance by Vasco? Should we really try to impose it on them?' Thereupon Harrison drew up a hard, business-like picture of the future Macmillans, as a conglomerate of firms in which the different parts had less and less to do with each other. 'There is only one rule: everything turns on money. That's all.' This once again led to the questions: Do we really have something in common in this conglomerate? Do we share something that unites us? And should we have something like that? Or, in other words: What does the name 'Macmillans' stand for? And should it stand for something particular? Now these were precisely the questions that were formulated as starting questions during the introductory session. So, did we make any progress in this investigation? Here we see once more how the attempt to find a valid answer to the starting question transformed the representation of the facts. For it appeared that this representation required elaboration again: All participants in the dialogue, and George Henry as well, are based at the main office of Macmillans. And this elaboration entailed a transformation of the judgement and its justification too: the consensus that had been achieved a moment ago about the nonn for dismissal was questioned already, for it might not be justified to impose it on others. Is this not a disheartening, even a demoralizing outcome? Conducting a dialogue is not an easy thing to do; it requires effort, patience, discipline and maintaining an attitude of inquiry. In spite of all this, the results of the investigation seem to be purely negative: The starting question - Which values and norms should all Macmillans companies share? - did not obtain any clear answer at the end of the dialogue. Besides, even the one common answer that was found, about a single norm, was questioned immediately. Is it then not an illusion to think that any progress was made in the dialogue? Did it not rather cost a lot of energy without leading anywhere?

The Case of the Shared Values

3

213

Socratic elenchus

This is a classical objection to the Socratic dialogue, one that Socrates himself had to deal with all the time. 7 At the same time, it reveals an essential aspect of the Socratic concept of learning. Socrates believed that before we can learn something we must first unlearn what we think we know but do not really know. We all carry a host of ideas, images, beliefs, preferences, etc., with us. Together these form our mental map, underlying our actions and behaviour. This mental map is not a consistent set; it contains all sorts of 'holes', blind spots, contradictions, deformations or 'idols'. The first thing Socrates tried to achieve in his dialogues was to expose these inconsistencies, to get someone to see that his usual ideas were not sufficient to solve a question on a deep level. This is called the 'elenchus', which means refutation or embarrassment. Only after having been refuted first, after the embarrassment of realizing one's ignorance, after having been brought into a reflective disequilibrium, the person interrogated has a real chance to inquire into a question on a fundamental level. For then he will not any longer be hindered by previous, more superficial ideas. The elenchus is the first of the two main processes of the dialectical method. It is the destructive part, aiming at undermining and eradicating some invalid opinion. The second is the constructive process, by which a better opinion is formed or induced. I will not go into that part here. s The elenchus is meant to expose the participants' ignorance, Of, more embarrassingly, ignorance mistaking itself for knowledge, false conceit of wisdom. Through elenchus Socrates brought his interlocutors to pass judgement upon themselves, and reduced them to a state of doubt or perplexity (aporia).9 Since in Nelson and Heckmann's adaptation of the Socratic method the dialogue is held with a group instead of between two persons only, the elenchus there changes into a group phenomenon. From an intrapersonal contradiction, like in the Platonic dialogues, it is transformed into an interpersonal one. The technique of destruction is thus altered: It is not primarily carried out by the facilitator or interrogator,but, inevitably and effectively, by the participants themselves, as the above dialogue shows. The results, however, are similarly perplexing. Nelson describes the point reached in this dialogue as follows: to 7

8 9 10

Cf. Meno 80a, Theaetetus 150d ff., Thrasymachus in the first book of The Republic, Guthrie (1971), 78 ff. Besides, many of Plato's dialogues end with Socrates' remark that the investigation did not lead to a final conclusion, and that it would be better to continue the conversation 'another time'. Cf., for instance, Theaetetus, Protagoras, Laches, Euthydemus. See Kessels (1997), ch. 3-5. Cf. Sophist 229E-230E. Meno, 7ge. 'Die sokratische Methode', in Nelson (1970) I, 297 (my translation).

los Kessels

214

The famous Socratic confusion has arisen. Everyone is turned desperate. What at first was certain to them has become uncertain. Instead of clarifying their ideas, they feel deprived of the ability to clarify anything at all by thinking.

This is exactly what happened in the dialogue above. The group did not find a definitive answer to its starting question. In a sense, it did not even find an answer at all, it only found another question - a much deeper question, admittedly, but a question all the same: Should we really have any shared values and norms? Should we want to have them? This implies a refutation of their original assumptions: viz., that the company should have a common culture, that the participants in the dialogue are the ones to look after that, that they are actually capable of achieving that. These hidden assumptions were exposed and deprived of their self-evidence. In spite of the friendly atmosphere and the good teamwork in the dialogue, this was an embarrassing experience. It affects the images the participants have of themselves, of their tasks and of the company's identity. At the same time, it is only after such a process of unlearning, after the destruction of some customary, deep-seated but deficient ideas, that the question can be investigated on a deeper, a more fundamental level.))

4

The concept of learning

Often the respondents, baffled and disgusted by the elenchus or destructive process, at this point tend to withdraw from the inquiry. But even then they have, in Socrates' judgement, gained something. For whereas fonnerly, being ignorant, they supposed themselves to have knowledge, now, being ignorant, they are at least conscious of their ignorance and accordingly will b~. more circumspect in the future. Still, the elenchus is not the only result of this dialogue. The joint analysis of a question, some consensus statements, some new, sharpened questions also_ resulted. Besides, there are a number of less conspicuous but equally important outcomes: detecting one's own and other participants' mental models; working towards a shared vision; making tacit knowledge explicit; convergence of concepts and attunement of terminology; team building;

11

Similarly Swieringa and Wierdsma (1992), chapter 8.

The Case of the Shared Values

215

meta-learning. The last point, meta-learning, to my mind is at the same time the most important one and the hardest one to discern at the beginning. On the one hand, the personal experience that it is in fact possible to bring about joint, organizational learning through systematic dialogue, is a necessary condition for acknowledging the significance of dialectics as an educational method. That is why a personal experience with dialectics is so important. On the other hand, to be able to appreciate an experience such as the dialogue above, including the elenchus or confusion, one needs a different concept of learning from our usual one. Normally, we identify learning with teaching, with the transfer of some subject matter, something new that we did not know before, containing new concepts and terms that we never have heard before. In a Socratic dialogue, however, nothing is taught, no subject matter is transferred, no unknown concepts are introduced. In a sense, even what was found out by the participants they had known all along. So, with the traditional concept of learning in mind, it is hard to see that something is being learned at all here. Perhaps one can see that the dialogue is an attempt to substitute poor, defective opinions for better ones by a systematic exchange of question and answer. But detecting the positive value of a, by all appearances, mainly negative result is hard at the start. Nevertheless, this conception of learning, including the elenchus, is one of the essentials of the Socratic method. Without it, reflective equilibrium is thin as paper(s).

Part V Case Studies in Bioethics

The 'Herman Case': The Usefulness of the Wide Reflective Equilibrium Model for Ethics Committees Egbert Schroten 1

1

Introduction

From the introductory chapter to this book it may be clear that there are - or, at least, can be - various types of the wide reflective equilibrium (WRE) model, depending on the purpose for which it is used. It may, for instance, be used as a model for constructing philosophical theories or for decision making; as a framework for structuring moral discussions or for analysing public debate. The question may be put, however, how useful the model is in view of these various purposes. In this contribution I want to examine the usefulness of the WRE model in the context of public morality, more specifically in the context of public policy making in situations which are (taken to be) morally problematic. In a previous chapter in this volume, Frans W.A. Brom already pointed out that, in such situations, governments tend to create committees or working groups with the task to advise as to the policy to be decided upon. This is, at any rate, the case in the Netherlands. Here, I want to focus on a case in animal biotechnology, namely the case of the transgenic bull 'Herman', which attracted much public attention in the Netherlands at the beginning of the nineties. More precisely, I want to concentrate on the work of the Provisional Commission for Ethical Assessment of Genetic Modification of Animals (henceforth: the Commission) concerning this case. As the chainnan of this Commission, which was to advise the Minister of Agriculture, Nature Management and Fisheries (ANMF), I have been intensely involved in the discussions. It is, therefore, interesting to look back, after some time, to what the Commission did. On the basis of an analysis of the advisory work of the Commission concerning the 'Hennan case', then, I want to assess the usefulness of the WRE model in view of the work of a small multidisciplinary task force, which was requested to prepare an advisory paper concerning a specific moral problem.

I want to thank Frans W.A. Brom, the then assistant secretary of the Provisional Commission for Ethical Assessment of Genetic Modification of Animals for having been so kind as to give some useful suggestions and comments on an earlier draft of my contribution. 219 W van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 219-229. @ 1998 Kluwer Academic Publishers. Printed in the Netherlands.

Egbert Schroten

220

Therefore, the first thing to do is to give the necessary information about the Commission and the Herman case.

2

The Herman case

In March 1992, the Commission was installed by the Minister of ANMF. Its task was to give advice to the Minister on the acceptability, from a moral point of view, of research projects carried out under the responsibility of the Minister, in the field of the genetic modification of animals, including projects with animals which had already been made transgenic. The Commission's task was also to give advice to research institutions in the same field, which do not work under the responsibility of the Minister, on their request. The Commission's first assignment was to draw up an advisory report (henceforth: the Advice) concerning a research protocol entitled 'Tissue-specific expression of genes in the mammary glands of genetically modified COWS'.2 It was submitted by Gene Pharming Europe B.V. (nowadays Pharming B.V.), where biotechnological research was done for commercial purposes, and IVO-DLO, the Institute for Cattle Breeding Research of the Service for Agricultural Research, an institute which operates under the responsibility of the Minister. The aim of the research project was to get female offspring from a genetically modified bull, called 'Herman'. The introduction of a gene construct, coding for lactoferrine, a protein with possible anti-infectious effects, had been successful in Herman. As to this aim, the following comment has to be made. The original version of the protocol was not clear in this respect. It gave the impression that, by increasing the concentration of lactoferrine (in combination with another protein, lysosyme) in the mammary glands of cows via genetic modification, the prevention and treatment of mastitis in cattle was intended. However, obtaining a natural antibiotic for mastitis treatment and other biomedical applications was mentioned as well. After consultation of the team which had submitted the protocol, this ambiguity was removed. The Commission concluded that the aim of the research project was twofold: (1) to obtain a sufficient quantity of lactoferrine in order to investigate whether it can serve as a remedy for some bacterial infections; and (2) to obtain more knowledge which is necessary for applying transgenesis in cows in view of the production, via the mammary glands, of proteins for biomedical use. Some background information about the political context of the Commission may be useful. Concerning animal biotechnology, there is, in the Netherlands, 2

Advies inzake het dossier 'Weefselspecifieke expressie van genen in de melkklier van genetisch gemodificeerde runderen' (Voorlopige commissie 1992).

The 'Herman Case'

221

a 'no-unless policy' . In practice this means that genetic modification of animals is forbidden unless it is licensed by the Minister of ANMF. A licence can be obtained if the goal of the research project is important enough, if there are no unacceptable consequences for the health and welfare of the animals involved and if there are no (other) ethical objections. This policy is based on the idea of the 'intrinsic value of animals', which has been established in an official memorandum entitled 'Rijksoverheid en dierenbescherming' (National Government and Animal Protection), published by the Ministry of Culture, Recreation and Social Work in 1981. It is incorporated in the Animal Health and Welfare Act, which was passed by Parliament in 1992. However, at the time that the protocol was submitted, the relevant articles (about animal biotechnology) of this framework act had not been elaborated. Because Gene Pharming Europe carried out the research project in cooperation with a governmental institution (IVO-DLO), the Minister was able to compel this institution to request for his permission. In view of the ethical assessment of the Herman case, the Commission did not have to start from scratch. In 1989, the Minister of ANMF had already installed an Advisory Committee on Ethics and Biotechnology in Animals. I chaired this Committee as well. As it is formulated in its Report, which was published in 1990 (p. 5), developments in biotechnology' ... require a description of the problems of biotechnological activities concerning productive livestock and pets, and a fonnulation of a framework against which these biotechnological acitivities can be evaluated on ethical grounds'. In constructing this framework, the Advisory Committee used Beauchamp and Childress's four principles of biomedical ethics as a model (Beauchamp and Childress, 1983). I shall return to this later on. The point I want to make here is that the Commission not only had a moral framework available but was assigned by the Minister, in his Installation Order, to take into account the Report of the Advisory Committee in which this framework had been worked out. It will, therefore, be necessary to refer to this Report repeatedly.

3

Analysis in the light of WRE

In view of the space available for this contribution, I will not give an extensive summary of the content of the research protocol; this will become clear in the course of my analysis in the light of WRE. The conclusion of the majority of the Commission was that in this case a licence for the genetic modification concerned could be granted because (1) the aim of the protocol was of substantial importance, (2) there were no real alternatives and (3) serious damage to animal health and welfare was not likely to occur. A minority of the Commission

222

Egbert Schroten

pleaded against because (1) the expectations of the protocol were not sufficiently underpinned, (2) the search for alternatives had been insufficient, (3) the financial cost-benefit analysis was not convincing and (4) even a small step in a wrong direction is wrong. Analysing the Advice in the light of WRE would mean to look for (1) moral intuitions and considered moral judgements, (2) moral principles, (3) morally relevant facts, (4) background theories and (5) methodological norms. These elements of the WRE model are, of course, in need of definitions, but for obvious reasons I shall not elaborate on this issue here. In other chapters of this book the reader may find the required critical analysis thereof. In my contribution I shall use my own provisional working definitions, based on my own understanding of what is meant by WRE.

3. J Moral intuitions and considered moral judgements Looking for moral intuitions and basic moral judgements in the Advice, one can point to the hunch that animal biotechnology is morally problematic. Although the Commission did not elaborate on it, it formed the basic idea of its work, and it was, so to speak, constantly present in the discussions. There was no need to elaborate on it because that had already been done by the Advisory Committee on Ethics and Biotechnology in Animals in their Report (1990), and this Report was taken as the starting point for the moral assessment of the protocol. Concerning animal biotechnology, the Advisory Committee made a first attempt to analyse public (moral) concern (1990, 7 ff.). Moreover, the Committee pointed out that there was reason for concern because there were important values at stake, for instance human health and welfare, animal health and welfare and environmental values. So, in the light of the 1990 Report, the idea that animal biotechnology is morally problematic may been seen as a considered moral judgement, in which perhaps several moral intuitions come together. One could even go back to the 1981 Memorandum, where, as I already pointed out above, the 'intrinsic value of the animal' had been formulated as the basic concept of the Dutch 'no-unless policy' . Animals are objects of moral concern. Therefore, impairment of their integrity, in terms of their 'biological intactness' (Heeger 1997,243) can be seen as morally problematic. 3 3.2 Moral principles The Commission made use of the moral framework which had been developed in the 1990 Report (14 ff.). In that framework an attempt is made to 'translate' the four well-known basic principles of biomedical ethics, put forward by 3

Recently, Frans BroID published a more elaborated inventory and analysis of the Dutch public debate on animal biotechnology (Brom t 995, t 997).

The 'Herman Case'

223

Beauchamp and Childress, into principles of animal ethics in the context of animal biotechnology. In other words, the 'Georgetown mantra' was used as a model in animal ethics. The result was as follows. The principle of respect for autonomy was 'translated' into respect for the survival instinct of (individual) animals and their capacities for realizing it. The expression 'intrinsic value of animals' was used, although clarified in tenns of animal health and welfare (p. 14). The principle of beneficence was spelled out in terms of care for animal health and welfare, the principle of non-maleficence in terms of refraining from doing harm to animals or from taking risks to do so. In practice this meant that there is a burden of proof on the basis of good reasons for anyone who would want to use animals in (biotechnological) experiments. The principle of justice was reformulated in terms of (or rather replaced by) the proportionality principle: the end should justify the means. The Advisory Committee added two principles to these four (1990,15). In the first place, the precautionary principle: Always act in such a way that you may be corrected by the consequences of your actions, and secondly, the principle of democratic control (which presupposes openness), in view of the far-reaching consequences of biotechnology for society as a whole. Although the Commission based its evaluation of the protocol on these principles, some of them were interpreted more or less differently, partly in order to cope adequately with the Herman case, partly because the interpretation of some principles in the Report was considered to be too narrow. In the 'autonomy' principle, for instance, three new aspects were introduced: animal integrity, the idea that animals have a value or worth of their own and animal species-specific behaviour. In the Herman case, where tissue-specific expression of genes in the mammary glands was aimed at, the Commission did not think that animal species-specific behaviour would be in danger. However, the principle of animal integrity is violated because of the introduction of a gene construct which is not from the same species, namely a human gene construct. Moreover, animal biotechnology is seen as a next step in the process of instrumentalization of animals for human ends, which means a new threat to animal worth. Another adaptation of the 1990 Report was the Commission's suggestion that the principle of beneficence should not only embrace animal health and welfare but also human health and welfare. As a consequence, the Commission had to balance human and animal interests against each other. Although the principle of justice, 'translated' into the proportionality principle, was taken to be relevant here, and much attention was paid to the question whether there were alternatives, in general human interests were seen as being more important than animal interests. I will return to this below.

222

Egbert Schroten

pleaded against because (1) the expectations of the protocol were not sufficiently underpinned, (2) the search for alternatives had been insufficient, (3) the financial cost-benefit analysis was not convincing and (4) even a small step in a wrong direction is wrong. Analysing the Advice in the light of WRE would mean to look for (1) moral intuitions and considered moral judgements, (2) moral principles, (3) morally relevant facts, (4) background theories and (5) methodological norms. These elements of the WRE model are, of course, in need of definitions, but for obvious reasons I shall not elaborate on this issue here. In other chapters of this book the reader may find the required critical analysis thereof. In my contribution I shall use my own provisional working definitions, based on my own understanding of what is meant by WRE.

3.1 Moral intuitions and considered moral judgements Looking for moral intuitions and basic moral judgements in the Advice, one can point to the hunch that animal biotechnology is morally problematic. Although the Commission did not elaborate on it, it formed the basic idea of its work, and it was, so to speak, constantly present in the discussions. There was no need to elaborate on it because that had already been done by the Advisory Committee on Ethics and Biotechnology in Animals in their Report (1990), and this Report was taken as the starting point for the moral assessment of the protocol. Concerning animal biotechnology, the Advisory Committee made a first attempt to analyse public (moral) concern (1990,7 ff.). Moreover, the Committee pointed out that there was reason for concern because there were important values at stake, for instance human health and welfare, animal health and welfare and environmental values. So, in the light of the 1990 Report, the idea that animal biotechnology is morally problematic may been seen as a considered moral judgement, in which perhaps several moral intuitions come together. One could even go back to the 1981 Memorandum, where, as I already pointed out above, the 'intrinsic value of the animal' had been formulated as the basic concept of the Dutch 'no-unless policy' . Animals are objects of moral concern. Therefore, impairment of their integrity, in terms of their 'biological intactness' (Heeger 1997, 243) can be seen as morally problematic. 3 3.2 Moral principles The Commission made use of the moral framework which had been developed in the 1990 Report (14 ff.). In that framework an attempt is made to 'translate' the four well-known basic principles of biomedical ethics, put forward by 3

Recently, Frans Brom published a more elaborated inventory and analysis of the Dutch public debate on animal biotechnology (Brom 1995, 1997).

The 'Herman Case'

223

Beauchamp and Childress, into principles of animal ethics in the context of animal biotechnology. In other words, the 'Georgetown mantra' was used as a model in animal ethics. The result was as follows. The principle of respect for autonomy was 'translated' into respect for the survival instinct of (individual) animals and their capacities for realizing it. The expression 'intrinsic value of animals' was used, although clarified in tenns of animal health and welfare (p. 14). The principle of beneficence was spelled out in terms of care for animal health and welfare, the principle of non-maleficence in terms of refraining from doing harm to animals or from taking risks to do so. In practice this meant that there is a burden of proof on the basis of good reasons for anyone who would want to use animals in (biotechnological) experiments. The principle of justice was reformulated in terms of (or rather replaced by) the proportionality principle: the end should justify the means. The Advisory Committee added two principles to these four (1990, 15). In the first place, the precautionary principle: Always act in such a way that you may be corrected by the consequences of your actions, and secondly, the principle of democratic control (which presupposes openness), in view of the far-reaching consequences of biotechnology for society as a whole. Although the Commission based its evaluation of the protocol on these principles, some of them were interpreted more or less differently, partly in order to cope adequately with the Herman case, partly because the interpretation of some principles in the Report was considered to be too narrow. In the 'autonomy' principle, for instance, three new aspects were introduced: animal integrity, the idea that animals have a value or worth of their own and animal species-specific behaviour. In the Herman case, where tissue-specific expression of genes in the mammary glands was aimed at, the Commission did not think that animal species-specific behaviour would be in danger. However, the principle of animal integrity is violated because of the introduction ofa gene construct which is not from the same species, namely a human gene construct. Moreover, animal biotechnology is seen as a next step in the process of instrumentalization of animals for human ends, which means a new threat to animal worth. Another adaptation of the 1990 Report was the Commission's suggestion that the principle of beneficence should not only embrace animal health and welfare but also human health and welfare. As a consequence, the Commission had to balance human and animal interests against each other. Although the principle of justice, 'translated' into the proportionality principle, was taken to be relevant here, and much attention was paid to the question whether there were alternatives, in general human interests were seen as being more important than animal interests. I will return to this below.

224

Egbert Schroten

3.3 Morally relevant/acts As I have already indicated, it is not my intention to deal with the (important) question of the definition of 'morally relevant'; that is done elsewhere in this volume by Theo van Willigenburg. In view of biotechnology, which is a new and far-reaching technology, I personally would vote for a moderately particularist view, which combines both moral rules and new moral features. However, the question I want to answer here is: Which factual information influenced the moral discussions of the Commission? At least three aspects of the Herman case should be mentioned explicitly, because they were discussed at length by the Commission. Firstly, the aim(s) of the research protocol. As I already pointed out, originally the protocol claimed to aim at increasing the concentration of lactoferrine (and lysosyme) in the mammary glands of cows in order to diminish the risk of mastitis. This had been rejected by the Commission because it would mean that, eventually, the whole Dutch bovine livestock would have to become genetically modified. Therefore, the aim became the production of a sufficient quantity of lactoferrine in order to investigate whether it could eventually be used as a natural antibiotic for biomedical applications. If this factual claim could be substantiated, this kind of genetic modification of farm animals could serve an important ideal (in the light of the principle of beneficence), namely human and animal health (and welfare). In this perspective, the Commission distinguished another aim, namely obtaining the necessary knowledge concerning the application of genetic modification in animals in view of the production of proteins in the mammary glands for biomedical use. Secondly, in the context of the Dutch no-unless policy however, there was, apart from the question whether this claim could be substantiated, the question whether it might be possible to produce lactoferrine by other means than genetic modification of farm animals. In short, were alternative methods available? In the end, the majority of the Commission decided that there were indeed alternative ways to get lactoferrine by using tissue or fungi, but that these alternatives were, for the time being, in technological respect, too difficult to produce and too risky as well, for instance because cancer-cell lines were used in tissue cultures. Another point that played a role in the discussion about alternatives was the fact that mastitis is a multifactorial disease, which could therefore not be reduced to one (genetic) factor. Control could only be effective by way of a broad approach, directed towards all underlying factors. Therefore one could say that the Commission implicitly rejected the protocol if the aim had merely been the production of a natural antibiotic against mastitis in bovine Ii vestock. It should be underlined, by the way, that the Commission wrestled considerably with the meaning of the concept of 'real alternative'. Is it an

The 'Herman Case'

225

alternative which could theoretically be made available? Or is it something which already is in the process of being developed? Or should it already be on the market? Within the Commission there was no consensus, which was one of the reasons for the majority and minority positions. It was also a reason for the Commission to publish a paper on the issue of alternatives after having given their advice in the Herman case (voorlopige commissie 1993). Thirdly, it was taken into account that, in this experiment, genetic modification was restricted to the mammary glands of a relatively small number of farm animals and was not expected to have serious negative consequences for the environment, nor for the health and welfare of the individual animals involved. However, because this latter point was not certain, the Commission included in its Advice the condition that research on the health and welfare of the animals involved should be part of the experiment. This has indeed been carried out (cf. Van Reenen and Blokhuis 1993).

3.4 Background theories Although it did not play an important role in the argumentation, in the Advisory Committee there was a direct reference to background theories (1990, 8). In the analysis of public debate 'philosophic and historico-philosophic questions' are mentioned, such as the place and task of human beings in Creation or in nature, and the meaning of science and technology There is even a quotation from the Creation stories in the Bible, namely that human beings are put in the 'Garden of Eden' (i.e. the earth) in order 'to work and to take care of it'. It is used to point to the need for 'wise nature management', in particular in connection with biotechnology. In the Commission's Advice however, there is no such reference to background theories. One may, nevertheless, point to the fact that they are present between the lines. In the first place, it is obvious that a certain anthropocentrism is presupposed. Human interests, such as health and welfare, have a predominance over animal interests. Therefore it is taken for granted that it is allowed that animals be sacrificed for 'substantial' human interests (voorlopige commissie 1992, 9). This predominance, however, is not absolute. Animals have a value of their own, an intrinsic value. Therefore it is not allowed to look upon animals as mere commodities or to reduce them to mere human instruments. In using animals for human interests, there should be a proportionality of ends and means. Other terms which are used in this context are 'animal integrity' and 'species-specific functioning' (p. 10). From this it can be derived that, secondly, a certain image of animals is presupposed. In the third place, and against the background of (a moderate form of) anthropocentrism, one may point to the presupposition of human health as a

226

Egbert Schroten

very high value. Needless to say that this is not only the case in the Netherlands but in Western society as a whole. Finally, an increase in knowledge is taken to be important, although it cannot be seen as a knock-down argument in favour of animal biotechnology (p. 6). In the words of the Commission: The mere increase in knowledge is not a value which, taken on its own, is sufficient to justify, from a moral point of view, a certain protocol in the area of animal biotechnology.

3.5 Methodological norms or ideals Like background theories, methodological norms are not explicitly mentioned in the Commission's Report. Here too, however, they appear to be presupposed. The Commission was composed on a broad basis. Moreover, membership was not based on political or religious convictions but on expertise, and the Commission was to work independently. This means that the Commission was supposed to constitute a group where rational discourse, in mutual respect for each other's arguments, should be possible; in other words, where a 'herrschaftsfreie Diskussion' (Habennas) could take place. The fact that, in the end, there has been a vote resulting in a majority and minority position does not belie this basic idea. Another point to be mentioned is the coherence between the various elements of the WRE model. Looking for a reflective equilibrium does not only require an analysis of these elements but also of the interplay between them. In this context Heeger speaks of 'mutual support and adjustment' (Heeger 1992b, 101 f.). I shall not elaborate on this aspect here, for it would require a too detailed analysis of the Report. Suffice it to refer to two issues mentioned above, namely adjustment of the autonomy principle in the light of morally relevant facts of the Herman case and to the discussion on the question of what would be considered to be a real alternative for this case in the light of technological feasibility and risk assessment.

4

The usefulness of the WRE model

In the introduction, I indicated that I want to assess the usefulness of the WRE model for the work of small groups in the context of public morality and public policy. Having analysed, in the light of the WRE model, the work of the Commission concerning the Hennan case, what can be said about the usefulness of this model in such a context? In this final section of my contribution, I will not stick to the Herman case; I wish to make the following, more general, comments.

The 'Herman Case'

227

1. The WRE model is a useful tool to structure, from a moral point of view, an analysis of a commission's report. It offers checkpoints for such an analysis, resulting in a clarification of the kinds of considerations which have been used, or played a role, in building a certain moral position. 2. The other side of this coin is that the WRE model can be useful as an instrument to structure the discussions in a Commission. It can, therefore, be helpful to prepare a position paper by indicating which aspects and kinds of considerations should, at least, be taken into account. 3. As a consequence, the model can be useful as a reminder for the policy makers as to the composition of an advisory committee in the field of public morality. The analysis of moral intuitions, moral principles, relevant facts, background theories and methodology requires a certain (multidisciplinary) expertise to be present in such a committee. 4. In spite of all this, some critical remarks can be made as well. Firstly, a point based on my personal experience as the chairman of some ethics committees in the context,of public policy making. If we want to use the WRE model, which has been developed for individual moral reasoning and theory building, for the work of small groups, like an ethics committee, attention should be paid to the group process. As it is always important to keep an eye on subjectrelated input in moral reasoning, it is the more so when the subject is a group. Aspects such as character, personality, rhetoric capacities of the group members, the quality of the chair and the like are to be taken into account. In this context I would underline the importance of confidence and confidentiality: in particular in a multidisciplinary group, the members are dependent on the reliability of each other's input. In direct relation to this, one could mention the importance of openness. The WRE model presupposes that any relevant information, position or conviction is under discussion. Finally, I would underline the importance of (a respectful attitude towards) religious, philosophical and moral convictions of the group members (and in society). I know, of course, that it is often 'not done' to discuss religious and philosophical convictions in public debate or in ethics committees and that it would be a delicate undertaking to make them the subject of discussion. But they are, as background theories, highly relevant to the moral issues at stake, so that at least we should know something about them and about the role they play in the group discussions. But then, in particular in view of public policy making, a basic attitude of respect should be a conditio sine qua non for joining an ethics committee. Again, in the light of its history, it is obvious that the traditional WRE model does not comprise this dimension of the group process. However, for its usefulness for the work of ethical committees it has to be developed in this direction.

228

Egbert Schroten

5. There is also the question of what can be considered to be a reflective equilibrium in the context of a committee. A consensus would be fine, but, as we all know, this is not always the outcome. Often a committee ends up, after voting, in a majority-minority position. Anyway, it seems to be useful to make a distinction between agreement, consensus and compromise. In an agreement, one agrees on a certain opinion. In a consensus, this is also the case, but one agrees on the arguments as well. A compromise is characterized by the fact that the parties concerned sacrifice, for pragmatic reasons, a part of their positions. In my opinion, in all three cases, agreement, consensus and compromise, one can speak of a reflective equilibrium, depending on a difference in weighing the arguments or in setting priorities. However, it may be important for a committee to point out what sort of equilibrium they have achieved. 6. The WRE model is 'clearly coherentist' (Van der Burg and Van Willigenburg in the Introduction to this volume). But, like all models, it is very rough. Even if, as is done in this volume, its components are analysed and elaborated, it has to be specified in order to make it useful for committee work in the field of applied .ethics. As far as I can see, it is, in its traditional form, like a recipe for making a cake, merely summing up the ingredients but not specifying the method of preparation. It is not my intention to elaborate such a specification here. To give an impression of what I have in mind, I would refer to the so-called 'stappenplan' (step-by-step design), which has been developed at the Centre for Bioethics and Health Law of Utrecht University, in cooperation with 'Zorg Consult Nederland' (Care Consultancy Holland) as a guideline for moral reasoning (Van Willigenburg et al. 1993, 55 ff.),4 on the condition that it is adapted to the WRE model.

4

This step-by-step design runs as follows: Phase I: Explicitness 1. What is the moral question? 2. Which possibilities for action are open at first sight? 3. Which factual information is lacking? Phase II: Analysis 4. Who are involved in this moral issue? 5. Which arguments are relevant for answering the moral question? Phase III: Weighing 6. What is the importance of these arguments for this case? 7. Which possible action is preferable on the basis of weighing the arguments? Phase IV: Action 8. Which concrete steps follow from it?

The 'Herman Case·

229

7. However, specified or not, the WRE model remains 'clearly coherentist' . In my view, this is both its strength and its weakness. To the strong side belongs its openness. It is compatible with and useful for any moral theory, and in a plural society as well, except perhaps in the case of a credo quia absurdum position. In other words, it presupposes some minimal conditions of rationality and common sense. But this openness is also its weakness from a moral point of view. From that point of view, a reflective equilibrium is not necessarily a good equilibrium. From a pragmatic or a prudential point of view it may be acceptable but that does not answer the question of whether it is morally acceptable. Although the WRE model may be indispensable to be able to give an answer to that question, in the end it can only be answered in the light of a moral theory. The WRE model is clearly not an outline of the good life. In a pluralistic society, however, a reflective equilibrium in moralibus is perhaps the best we can get, certainly in view of public policy. In conclusion, then, I want to state that the WRE model, in a more specified form, can be useful in the context of public policy, in situations which are (taken to be) morally problematic. It can certainly also become useful for committee work in the context of public morality if certain conditions are met. In this last section of my contribution I tried to indicate some of them.

The Use of the Reflective EquilibriuIn Method in Normative Veterinary Ethics Bart Rutgers·

1

Introduction

In all branches of veterinary practice - in companion animal, sport animal and fann animal practice - veterinarians are faced with ethical issues regarding the treatment of animals. When veterinarians consider moral questions, they ask, 'What is morally right to do in a particular case?' In seeking answers to these moral questions, the reflective equilibrium method appears to be a suitable approach (Rutgers 1993). This method is used in teaching veterinary and animal ethics to veterinary students at the Faculty of Veterinary Medicine of Utrecht University (the Netherlands). Veterinarians can also use this method when they deliberate about ethical issues in veterinary practice. The proposed method of moral deliberation about ethical issues is based upon a theory of wide reflective equilibrium elaborated by Van Willigenburg and Heeger (1989) in their 'network model' for moral decision making. In this reflective equilibrium method, three elements play an important role: moral intuitions, moral principles and morally relevant facts. A brief summary of the method is given in the next section. Special attention is paid to moral principles that can be used to guide veterinarians in making judgements and decisions about ethical issues regarding animals. In order to show how this reflective equilibrium method works, it will be applied to two concrete moral cases: (1) the declawing of cats and (2) the routine practice of Caesarean section in beef cattle.

2

The network model for moral decision making

Moral reasoning according to the 'network model' of Van Willigenburg and Heeger (1989) implies a continuous interplay between moral intuitions (intuitive moral judgements or convictions), moral principles and morally relevant facts. To reach a justification of a moral judgement in a moral case four activities are required: making explicit the moral intuitions one has with regard to the case;

Thanks are due to Frans W.A. Brom, from whose comments this article has profited. 231

W. van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 231-237. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

Bart Rutgers

232

searching for possibly applicable moral principles and elucidating the case from the perspective of these principles; searching for the morally relevant facts of the case, that is to say, tracing features of the case which are morally relevant in the light of intuitions and principles; balancing intuitions, principles and morally relevant facts until an equilibrium is reached. These activities proceed in a continuous interaction. This means that moral reasoning consists in an interactive process of mutual adjustment and mutual justification to which all four activities contribute in their own way (for a further explanation see the Introduction to this volume).

3

Moral principles

One of the actions we have to take is to search for applicable moral principles and elucidate the ethical issues from the perspective of these principles. Three fundamental moral principles have been elaborated with respect to our relationship with animals (Rutgers 1993; De Boer, Brom and Vorstenbosch 1995; Rutgers, Grommers and Colenbrander 1996; Heeger 1997): - the principle of non-maleficence; - the principle of beneficence; - the principle of respect for animal integrity. The principles of non-maleficence and beneficence bear on both humans and animals, while the integrity principle only refers to animals. According to the principle of non-maleficence one should not inflict harm on living beings (humans and animals). 'Harm' is defined as pain, suffering, discomfort, illness and poor welfare. The principle of non-maleficence is recognized as one of the most basic principles of nonnative medical and veterinary ethics. The concept of non-maleficence is associated with the maxim primum non nocere - 'first of all, do no harm'. This maxim has a prominent position in the ethos of both physicians and veterinarians. Therefore, the duty of non-maleficence is included in most professional codes. According to the principle of beneficence, one should promote the health and well-being of living beings. The duty to benefit humans and animals is an important professional norm in veterinary ethics. With respect to this duty, the professional code for veterinarians of the Dutch Veterinary Medical Association (KNMvD 1992) states that veterinarians are to promote animal health and welfare and to look after the animal owners' interests. Thus, veterinarians have to take into consideration the interests of animals (insofar as their health and welfare

The Use of the Reflective Equilibrium Method in Normative Veterinary Ethics 233

are concerned) as well as the interests of the owners. In practice this may give cause for a conflict of duties. The principle of respect for animal integrity differs from the principles of non-maleficence and beneficence. Applied to actions involving animals, the principles of non-maleficence and beneficence prescribe that one ought not to inflict suffering on animals and that one ought to promote animal health and welfare. The principle of respect for animal integrity is a principle that reflects a sense of respect for animals beyond care for their health and welfare. Animal integrity is defined as the wholeness and completeness of the animal and the species-specific balance of the creature, as well as the animal's capacity to maintain itself independently in an environment suitable to the species (Rutgers 1993; Rutgers, Grommers and Colenbrander 1996; Grommers 1997). Respect for animals (as living beings with a 'good of their own') expresses a moral obligation to consciously refrain from specific interferences in the Ii ves of animals (Heeger 1997). According to the definition of animal integrity above, this implies that man: (1) should not interfere with the wholeness and completeness of the animal, (2) should not disrupt the species-specific balance and (3) should not deprive animals of the ability to maintain themselves independently in an environment suitable to the species (Rutgers, Heeger and Grommers 1998). Infringement or violation of integrity therefore relates to these three criteria.

4 Application of the method to two ethical issues To show how the process of moral reasoning according to the reflective equilibrium method proceeds when it is applied to concrete ethical issues, I shall elaborate two moral cases: (1) the declawing of cats and (2) the routine practice of Caesarean section in beef cattle.

Case 1: Declawing of cats Amputation of the claws of domestic cats (declawing) was introduced in the late sixties as a remedy to prevent cats from furniture scratching (Herron 1967). Since that time there is ethical controversy about the surgery.2 Declawing involves the removal of the distal phalanges and claws, usually from the front feet only. 2

In the Netherlands the ethical debate has resulted in a legal ban on the declawing of cats (Animal Health and Welfare Act 1996). In many other countries it is not forbidden to declaw cats. Declawing is a very common procedure in the USA. The American Veterinary Medical Association has concluded that 'declawing of domestic cats is justifiable when the cat cannot be trained to refrain from using its claws destructively' (AVMA 1991).

234

Bart Rutgers

The case can be described as follows: to prevent a domestic cat from scratching the furniture a veterinarian practitioner is asked to declaw the cat. It is assumed that all other alternatives to prevent the cat from scratching have been unsuccessful and that the owner is not willing to place the cat elsewhere. So, the veterinarian has to make a choice between declawing and refusing the operation. The question is whether it is morally acceptable to declaw a healthy cat for the owner's sake. Many veterinarians intuitively object to performing a mutilating operation on healthy animals. On the other hand, if a veterinarian refuses to declaw the cat, he frustrates the owner's interests. Thus, the moral problem can be regarded as a conflict between the responsibility for the animal and the professional duty to take the owner's interests and wishes into consideration. With respect to the moral responsibility for animals, the principle of nonmaleficence and the principle of respect for animal integrity throw light upon the case. From the perspective of the principle of non-maleficence, the following facts are morally relevant: discomfort associated with surgery and anaesthesia, discomfort because of postoperative complications and the fact that declawed cats meet problems with climbing and defending themselves. Declawing involves an infringement of animal integrity because of the removal of body parts that are essential for the cat's being a cat. Violation of the biological intactness of an animal is incompatible with respect for animal integrity. By having knowledge of the facts in the light of the principles of non-maleficence and respect for animal integrity, the veterinarian can be confirmed in his conviction that it is morally wrong to declaw a healthy cat for non-therapeutic reasons. The principle of beneficence emphasizes the interests of the cat owner. If the veterinarian ought to act in accordance with the interests of the owner (professional code of conduct), refusal of the operation would be contrary to this professional nonn. However, it would be a wrong interpretation of this nonn if even weight were given to all the owner's interests and wishes. A fumiturescratching cat is a nuisance to the owner. Annoyance is the main reason why the owner wants to have his cat declawed. Most people agree that annoyance is not a good reason to hann a healthy animal or to violate its integrity. All things considered, the decision to refrain from declawing a cat can be justified with strong arguments.

Case 2: The routine practice of Caesarean section in beef cattle During recent years, veterinary practitioners are increasingly called upon to perform Caesarean sections in beef cattle. Deliberate breeding of heavy calves has led to an increase in this surgery, particularly in doubled-muscled cattle breeds. A Caesarean section is profitable as long as the market value of the calf

The Use of the Reflective Equilibrium Method in Normative Veterinary Ethics 235

is higher than the costs of the operation. The Caesarean section is a common procedure at many beef cattle farms. Most veterinarians intuitively take the view that the routine practice of Caesarean section is not in accordance with the principles of 'Good Veterinary Practice'. The moral problem is not the choice between performing or not performing the operation. It is clear that a Caesarean section must be performed if it proves to be the only way to save the lives of both cow and unborn calf. In fact, because of the professional duty, a veterinarian cannot refuse to perform a Caesarean section. The routine practice of Caesarean section seems to be problematic because veterinarians are forced to perform an operation that was foreseen and planned and therefore could have been prevented. Thus, the problem is not that the veterinarian performs the operation in a case of emergency. The question is whether the veterinary profession has a responsibility with regard to the prevention of Caesarean sections in beef cattle. Do we have reasons justifying our opinion that the routine practice of Caesarean section is morally wrong? To answer this question we can evaluate the case in the light of the principles of non-maleficence and respect for animal integrity. From the perspective of the principle of non-maleficence, the following facts are morally relevant: the discomfort from the surgery itself, the risk of mortality and the fact that a Caesarean section involves abdominal surgery with the associated risk of peritonitis and wound infection. From these facts it may be concluded that a Caesarean section should be avoided as much as possible. In general, it should be stated that a surgical procedure which is avoidable and which involves animal health and welfare risks is hardly justifiable from a veterinary-ethical point of view. Breeding of heavy calves, which makes Caesarean sections inevitable, can be considered a violation of animal integrity. Integrity is at issue because the criteria regarding the species-specific balance and the animals' ability to maintain themselves independently are not fulfilled (see the definition of animal integrity). The excessively heavy musculature - often in conjunction with a narrow pelvis - upsets the biological balance to such a degree that the cows are no longer able to calve normally and naturally. This deprives the cows of an essential ability in their existence as cattle. Through the deliberate breeding of heavy calves, the cattle are no longer capable of independently maintaining themselves. The infringement of integrity does not take place at the moment of performing the Caesarean section, but at the moment of the choice of the bull, since the cattle breeder knowingly selects a bull that propagates heavy calves. The breeder deliberately runs the risk to have a Caesarean section performed. Paving the way for parturition problems that necessitate a Caesarean section is inconsistent with the principle of respect for animal integrity.

236

Bart Rutgers

Some veterinarians hold that they also have to take the interests of beef cattle fanners into consideration. The fanners have an unmistakable financial interest in breeding beef cattle. But because of the professional responsibility with regard to the prevention of Caesarean sections in beef cattle (which involves animal health and welfare risks) and because of the violation of animal integrity, it cannot be justified that veterinarians give more weight to the economic interests of the farmers.

5 Discussion Moral decision making according to the reflective equilibrium method is a process of balancing moral intuitions, moral principles and morally relevant facts. Reflecting on an ethical issue along the lines of the reflective equilibrium method enables us to give a justification of our moral judgement about that case. The method provides us with a structuring of moral justification that closely resembles our day-to-day reasoning process when trying to reach a moral decision (Van Willigenburg and Heeger 1989). This makes the method attractive to use. On closer inspection, decision making according to the reflective equilibrium method bears a strong resemblance to the way veterinarians try to solve clinical problems. Clinical decision making is a process of balancing clinical intuitions (the 'clinical eye'), pathobiological principles and clinically relevant facts. Because of the methodological resemblance between clinical and moral reasoning it should not be difficult for a veterinarian who is experienced in clinical thinking to make himself familiar with the way of moral thinking along the lines of the reflective equilibrium method. Moral decision making with the help of the reflective equilibrium method can only be fruitful if the following conditions are fulfilled. Firstly, moral deliberation demands openmindedness. The 'ethical practitioner' must demonstrate moral sensibility and responsiveness to moral facts. He knows what it means to judge on an ethical issue from a moral point of view. Secondly, knowledge of ethical background theories is required. It is also important that the ethical practitioner has an understanding of the relevant moral principles. As stated in section 3, the principles of non-maleficence and beneficence deal with animal welfare. We must realize that animal welfare is a moral concept in the context of these principles. Animal welfare researchers use different conceptions of animal welfare to make it amenable to scientific study. However, animal welfare research only gets moral relevance when the relationship with ethics is made explicit (Tannenbaum 1991; Stafleu, Grommers and Vorstenbosch 1996). This is possible by using a 'bridging concept' of animal welfare, which

The Use of the Reflective Equilibrium Method in Normative Veterinary Ethics 237

links scientific research to the ethical concerns about the welfare of animals (Fraser, Weary, Pajor and Milligan 1997; Duncan and Fraser 1997). It must be emphasized that animal welfare concepts differ from the concept of animal integrity. Animal welfare has to do with feeling good and functioning well, whereas integrity relates to biological characteristics of the species in question (see section 3). The criteria used in the assessment of animal welfare are different from those used to judge animal integrity. Another reason for separating the concept of animal welfare from the concept of animal integrity is that the two concepts are linked to different moral principles. Integrity is connected with the principle of respect, whereas welfare relates to the principles of non-maleficence and beneficence. Therefore, it is suggested to distinguish between considerations of integrity and considerations of welfare (Rutgers, Heeger and Grommers 1998).

6 Final remark Veterinarians can individually make use of the reflective equilibrium method when they are faced with ethical issues regarding the treatment of animals. In other words, the method is suitable for 'intrapersonal deliberation', but it can also be used as a model for intersubjective argumentation. A survey of Dutch veterinary practitioners revealed that 71 per cent of the respondents appear to be proponents of taking'collective standpoints' with regard to particular ethical issues (Rutgers and Baarda 1994). When using the reflective equilibrium method as a model of 'interpersonal deliberation', considered judgements which are broadly supported by the veterinary community can be reached.

Infanticide in Reflective Equilibrium? Goran Collste

1

Introduction

In this article, I will applytthe method of reflective equilibrium to a moral issue other than social justice. I will apply it to the question whether infanticide can be morally justified. A specific case, the infant Linda, will be my point of departure, and the case will be discussed in the tenns of the reflective equilibrium method. When discussing this kind of moral problems, i.e. when human life and death are at stake, it becomes necessary to relate the judgements and moral principles to background theories, in this case theories about the nature and value of persons and religious beliefs.

2

Linda's case

One morning, Linda was found on the doorstep of a small hospital. Someone, probably (one of) her parents, must have left her there during the night. There was no message, nothing. The doctor who examined the girl guessed that she was at most one month old. She was mentally retarded, probably caused by brain damage during birth. Linda's life was miserable. She was not able to communicate, she seemed to suffer constant pain and she regularly got cramps that took the breath out of her. In the ward where Linda was placed, particularly two persons had contact with her. One was the head of the ward, an economist who had been placed there because the manager of the hospital wanted a more cost-effective business. The other was the staff nurse who had been ordered to take care of Linda. What should be done with Linda? The two persons having contact with Linda, the head of the ward and the staff nurse, arrived at opposite views. The head of the ward was repelled by Linda. His feelings were a mixture of repulsion and pity. Was her life a life worthwhile to live, he asked himself. Linda regularly got shots for pain relief. Wouldn't it be better to give her a stronger drug that would put an end to her life once and for all, he thought. Besides, the treatment Linda needed was very expensive and the head of the ward was under pressure to reduce costs. 239

M': van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 239-250. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

240

Goran Collste

The staff nurse devoted much time to care for Linda. She felt compassion with her and tried different ways to give her relief from her suffering. But she also tried to find ways to communicate with Linda, and now and then she noticed reactions from her. Linda would smile a little when the staff nurse was laughing and she seemed to enjoy listening to classical music, especially Mozart. It was necessary to arrive at a decision about what should be done with Linda, but how should they proceed? Would it be possible to come to an agreement?

3

Intuitions and arguments

Both the head of the ward and the staff nurse entered into Linda's life. Their views are based on empathy, Le. a combined emotional and rational effort to get some understanding of what Linda's life is like. However, they arrived at different views on what should be done with Linda. Let us now reconstruct their different ways of reasoning. Intuitively, the head of the ward thinks that the right thing to do must be to shorten Linda's life. When reflecting on this intuition, he formulates three arguments for the view that it is morally right to shorten Linda's life. Firstly, her life is void of value to herself. It contains almost no positive feelings or experiences, but instead much pain and suffering. Secondly, her life is of no value to anyone else. She has no friends and no relatives, no one cares for her. Thirdly, the financial resources that it takes to keep Linda alive and to minimise her pain could alternatively be used for other patients, helping them to get a life that is worthwhile. At the same time, after having met Linda, he is hesitant whether he himself would be able to order her killing. Thus, although there are in his view convincing arguments for killing Linda, the head of the ward is still in doubt, and this makes it necessary to engage in a reflective process. Let us now reconstruct the head's reasoning. The first argument can be structured in the following inference:

Argument 1 (1) Linda's life is void of value. (2) One ought l to shorten a human life void of value. (3) It is right to shorten Linda's life.

The concepts 'ought', 'right' and 'should' in these arguments are used in a moral sense. For simplicity, I use them interchangeably.

Infanticide in Reflective Equilibrium?

241

The second premise can be formulated as a moral principle: Principle 1 One ought to shorten a human life void of value.

A reason for this principle is that only those lives that contain more pleasure than pain are worthwhile to live. Thus, the value in human life, i.e. the pleasurable experiences, is the source of the value of a human life. The second argument is structured in the following way: Argument 2 (1) Linda's life has no instrumental value to anyone else. (2) One ought to shorten a human life that is of no instrumental value to anyone else.

(3) It is right to shorten Linda's life. Even here, the second principle can be formulated as a moral principle: Principle 2 One ought to shorten a human life that is of no instrumental value to anyone else.

The same reason as for Principle 1 can be given for this principle. But in this case, it is the pleasure of other persons that counts. Finally, the head's third argument runs as follows: Argument 3 (1) Allocating medical resources to Linda's care, is not allocating them in the most cost-effective way. (2) Medical resources should be allocated in the most cost-effective way.

(3) Medical resources should not be allocated to Linda's care. The second premise is not very controversial. The first, however, is controversial and, in the head's reasoning, dependent on the other arguments, Le. that quality of life, or, more precisely, the pleasure that persons experience in life, is a relevant criterion for priorities given to medical resources. This means that scarce medical resources should be allocated to treatments that may increase pleasure or diminish suffering. The second premise can be formulated as a moral principle:

242

Goran Collste

Principle 3 Medical resources should be allocated in the most cost-effective way. Let us now examine the arguments raised by the staff nurse. For her, the right thing to do is to give Linda as much time and care as possible. Why? Because, firstly, she notices that when she is with Linda, Linda's suffering is limited, that she can even smile now and then. Secondly, on these occasions the staff nurse can feel a kind of relationship or connectedness with Linda. When the head of the ward mentions to the staff nurse that he thinks it would be right to shorten Linda's life, she is astonished and upset. Killing Linda would be the same as offending the inviolability of human life, she thinks. When reflecting on her view, the staff nurse formulates two arguments for taking care of Linda and one argument against the view that it would be right to shorten Linda's life.

Argument 4 (1) Through taking care of Linda, her suffering is limited. (2) It is right to limit human suffering. (3) One should take care of Linda. In this argument, the second premise can also be formulated as a moral principle:

Principle 4 It is right to limit human suffering. The second argument in favour of taking care of Linda is the following:

Argument 5 (1) When taking care of Linda the staff nurse develops a relationship with her. (2) It is valuable to develop a relationship with another person. (3) One should take care of Linda. This argument deals with one of the consequences of caring, Le. that a human relationship evolves. This is especially important when it refers to mentally retarded persons, Le. persons with whom it may otherwise be difficult to establish relationships. Let us formulate the second premise as a moral principle:

Principle 5 It is valuable to establish relationships with other persons.

Infanticide in Reflective Equilibrium?

243

Finally, her third argument is an argument against the standpoint of the head of the ward.

Argument 6 (1) Linda is a human being. (2) Human beings are inviolable and should be protected. (3) It is not right to shorten Linda's life. In this argument, the second premise can be interpreted as a formulation of the Principle of Human Dignity or the Sanctity of Life Principle:

Principle 6 Human beings are inviolable and should be protected. So far, we have rendered the intuitive reactions of two persons who came into contact with the miserable life of a mentally retarded infant. Further, we have formulated their considered judgements, Le. their 'duly reflected convictions' (Aronovitch 1996, 399) and the arguments for their respective convictions. Finally, we have stated the moral principles implicit in their arguments. In this way, we have stated the general moral norms that their argumentation is consistent with. 2 Our next step is to examine the head's and the staff nurse's reasoning. By doing that, we will examine whether their different positions are in a reflective equilibrium.

4

Which position is in equilibrium?

What should be done with Linda? What should be done in a situation like this? So far, by reconstructing the arguments for the opposite views, reflective reasoning has removed some possible obstacles to rationality in ethics, Le. reliance on private intuitions and a lack of consistency between moral intuitions and moral principles. But we are still stuck with two very different positions. One way to come to a decision would, in line with John Rawls's method of reflective equilibrium

2

The judgements and arguments here presented are elaborated in works in applied ethics. See, e.g., Kohl (1978). The arguments of the head of the war9 are found in Harris (1985) and Kuhse and Singer (1985). The arguments of the staff nurse can be found in Donogan (1977) and in Ramsey (1978).

244

Goran Collste

(Rawls, 1951, 181 f.; 1971,48-51), be to ask if any of the positions is defective because it is not in a reflective equilibrium. Let us therefore look at both views from this angle. According to Robert Heeger, this method can be described as ' ... a process in which three elements play an essential role: intuitive moral judgements, moral principles and morally relevant facts' (Heeger 1992b, 95). The aim of the process is to arri ve at - what is altematingly called - an equilibrium, a balance or a coherence of these three elements. We will start with the view held by the head of the ward. His intuition that Linda's life is worthless and, thus, that Linda should be killed is warranted by two arguments. These arguments imply two moral principles: Principle 1: One ought to shorten a human life void of value, and Principle 2: One ought to shorten a human life that is of no instrumental value to anyone else. Further, he argues that the resources needed to care for Linda could be used in a better way. If he were to start his reasoning the other way around and apply principles 1 and 2 to the specific situation, he might reason as follows: Application of both principles will justify Linda's killing. When Linda is killed, her miserable life is ended and no one will grieve for her. The morally relevant facts are in this case, firstly, Linda's suffering and, secondly, the costs of her care. Whether starting from the specific intuition or from the level of ethical principles, the same conclusion will be reached: It is morally right to kill Linda. Thus, this view seems to be in a reflective equilibrium. But perhaps this conclusion is rash? The head of the ward felt, as we noted, some hesitancy concerning this conclusion. The fact that he had come into contact with Linda and that their eyes had met, raised some doubts as to whether killing her really was the right decision. This intuition bothered him and made him dissatisfied with the conclusion that it is right to kill Linda. He still had reasons to look for another solution. Let us now turn to the staff nurse. She looks upon Linda as a person that needs to be taken care of. Her intuitive response is warranted by three nloral principles: Principle 4: It is right to limit human suffering, Principle 5: It is valuable to establish relationships with other persons and Principle 6: Human beings are inviolable and should be protected. For the staff nurse, the morally relevant facts are that Linda is suffering, but also that Linda is a human being. The latter fact is relevant because it is a sufficient condition for being valuable and thus being protected according to the Principle of Human Dignity. When, in this case, we start from the level of principles and apply them to the specific situation, we find that the principles may come into conflict. This is the case if there are no means to limit Linda's suffering. In that case, the staff nurse either has to give up her intuitive reaction that Linda's life should be

Infanticide in Reflective Equilibrium?

245

protected or has to argue that the principle that human beings are inviolable takes precedence over the principle to limit suffering. Only in the latter case, Le. if the staff nurse is able to establish an order of priorities of principles, her view is in a reflective equilibrium. The relation between moral principles, relevant facts and intuitions is illustrated in the following figure: Moral principle One ought to shorten a life void of value.

Relevant facts Linda's suffering

Intuition Linda should be killed.

Human beings are inviolable and should be protected.

Linda is a human being.

Linda should be protected.

When applying the method of reflective equilibrium to the moral question of what should be done with Linda, we have found that neither view as yet fulfils the criteria of a reflective equilibrium. In the case of the head of the ward, there is still a conflict between the principles and his intuition, and in the case of the staff nurse there is a potential inconsistency between principles. At this point, someone might object that neither the head of the ward nor the staff nurse fulfils one of the criteria of a competent judge fonnulated by Rawls and others, i.e. impartiality (Rawls 1951, 181 f.). The decision that Linda should be killed would decrease the financial expenses of the ward, and this would be a gain to the head of the ward personally. On the other hand, the decision to continue Linda's care would be in the interest of the staff nurse, while she has established a personal relationship wit~ Linda. Thus, both the head of the ward and the staff nurse run the risk of arriving at biased decisions. However, the head's involvement and the staff nurse's involvement are not of equal standing. The head's bias is related to his material interest; he is in favour of a decision that will benefit his own career. The staff nurse, on the other hand, will 'gain' emotionally by the decision she defends. Here, the method of reflective equilibrium can be seen as a way to distinguish between reasons and causes. The method focuses on reasons: Which position is reasonable? The interests of the persons involved, should not count; on the contrary, the aim is to arrive at a common conclusion, independent of personal interests. However, the staff nurse's involvement can be seen as an illustration of a potential conflict between two important conditions for a valid ethical

246

Goran Collste

judgement: impartiality and emotional involvement. Emotional involvement is an important condition for morality - Why should we bother about morality if we did not have the ability to engage in other persons' lives? - and it is a means to focus on aspects that may count as reasons in a reflective process. Thus, this engagement should not, as Kantians tend to do, be seen as an obstacle to a rational decision. However, the fact that emotional involvement is important for moral decisions does not exclude the possibility that it may sometimes function as an obstacle to a clear-thought, considered judgement in ethics.

5

The application of background theories

Although both the head of the ward and the staff nurse have been able to structure their views in the process of arriving at a reflective equilibrium, neither has reached the goal. There are still some disturbing doubts on both sides. The head of the ward has argued for a standpoint that he is unwilling to carry out and the staff nurse is facing a problematic conflict of principles. Are we stuck with an unsettled disagreement? Are there ways to continue the process towards an agreement? What will happen if they confront each other with their intuitions and their considered judgements? A possible next step would be to examine and critically assess the theoretical backgrounds of their judgements and principles. In this way, we enter the domain of fundamental views of life and religious beliefs. This step is, especially nowadays, a controversial one to take. It is common to argue that our views are embedded in our traditions and communities and that this puts limits to the possibilities of arguing on the level of world views and religious beliefs. The introduction of the concept 'background theories' and the aim of a wide reflective equilibrium that includes these theories can be interpreted as manifestations of an intention to reach beyond these limits (Daniels 1996). How would such an intention be carried out in the present discussion on the moral question whether infanticide may be justified? Space only allows us to hint at a possible answer. 3 The principles referred to by the head of the ward may be backed by utilitarianism. According to the utilitarian position, the only relevant criterion for a moral judgement is whether an act tends to increase or decrease pleasure or preferences. According to Bentham and J.S. Mill, this view, in turn, is warranted by psychological hedonism. 'Mankind has placed man under the governance of two sovereign masters, pain and pleasure. ' (Bentham 1988, 1; 3

The principles mentioned in this article can be related to a great number of background theories (Collste 1998). Here, I limit myself to some main traditions.

Infanticide in Reflective Equilibrium?

247

Bentham's italics). Thus, to aim at pleasure and avoiding pain is part of man's constitution. A utilitarian ethical theory follows naturally from this view of man. 4 Thus, we have sketched one theory warranting the position of the head of the ward, Le. we have formulated its background theory. Turning now to the staff nurse, her position that it is morally wrong to kill Linda can be warranted by other theories. The principle that human beings are inviolable and should be protected can be seen as a component of the Sanctity of Life Doctrine. According to this interpretation of the doctrine, it implies that human life has intrinsic value. This view may, in its tum, be given both a Kantian and a Christian warrant. For Kant, the starting point of moral reflection is man's rational nature and autonomy. ' ... [A]utonomy is the ground of the dignity of human nature and of every rational nature', Kant writes, and therefore 'man ... exists as an end in himself' (Kant 1785/1983, 35 ff.). When applied to the question whether Linda should be killed or not, it is not clear which position the Kantian principle will support. One can either argue that to kill Linda is to treat her as a means and that thus the act is morally wrong, or one can argue that Linda's life is so miserable and beyond any relief that the only way to treat her as an end is to liberate her from her frustrated life by killing her. How would a Christian warrant be framed? According to a Christian world view, God is the creator of the earth. When man was created, he was created in the 'image of God' (Genesis 1:26), thus man is special in creation. This view is sometimes interpreted as stating an intrinsic value of human life. According to this interpretation, human life is 'sacred'. If human life is sacred, there is a duty to protect human life and it can never be morally defensible to intentionally kill a human being, or, in the words of principle 6: Human beings are inviolable and should be protected. But how can the Christian view of the inviolability of human life be warranted? To answer this question one may, i.a., refer to Augustine's and Aquinas's view of man as a rational being (S1. Augustine 1978; Aquinas 1952, Q 93, art. 8) to Karl Barth's view that each individual is valuable due to the fact that Christ has sacrificed himself for her (Barth 1951, 753) or to Paul Ramsey's view that the value of man is based on a covenant that God has established between himself and man (Ramsey 1970). When we reflect upon or question this kind of background theories, we ask whether they are valid interpretations of our experiences and our knowledge. We can question whether a world view is one-sided, i.e. if it does not take into

4

Here, I will not enter into the discussion whether this defence of utilitarianism commits the n~turalistic fallacy.

248

Goran CoLLste

account the full spectrum of human life. Further, we can find it simplifying or, for that maUer, mystifying the way we experience and interpret life. Let us now apply this way of reasoning to the background theories that are relevant in the discussion on whether Linda should be killed or not. Utilitarianism and Kantianism start from different aspects of human nature. For utilitarianism, the view that man is a sensational being seeking pleasure and avoiding pain, is central. For Kantianism, on the other hand, it is man's rationality and freedom that are the focal points. Man is primarily seen as a rational, deliberating creature. Both aspects of man - man as a sensational being and man as a rational being - seem to be relevant when constructing a view of man. They point, respectively, to important ways of apprehending the nature of human beings. However, separately, they are deficient; in combination, they give a more complete view of man. As a consequence, this may lead us to a 'mixed' ethical theory when we try to integrate both utilitarian and Kantian aspects. Let us now tum to the Christian view. Firstly, one may, in line with many philosophers as well as many Christians, question whether that is not beyond any rational discussion. Is not the Christian view based on a transcendent revelation unattainable for non-Christians? In my view, this objection is invalid. Christianity as well as other religions are interpretations of a reality that is common to all human beings. These interpretations are conceptualised in different religious languages and ritualised in ceremonies and worship. The way in which we experience and interpret the world will be decisive for whether we can experience the reality as 'holy' (Otto 1924) and our lives as dependent on 'a Ground of Being' (Tillich 1948) that goes beyond ordinary life or not. Thus, at least in principle, it is possible to continue the discussion concerning the religion-based moral convictions, and the religious conviction does not provide the believer with any excuse not to try to argue for a theoretical justification of her beliefs. The scope of this article does not allow further examination of the Christian views. I will limit myself to stating, that they may support the view that human beings are conceived as something special in creation, something different from everything else. This conception can be warranted by a phenomenological experience of the other person. She can be experienced as a valuable Thou (Buber 1962) or as 'the Face that one cannot kill' (Levinas 1988). The cognitive content of these phenomenological experiences may be judged in different ways. They may be seen as delusions, as manifestations of an ontological reality or as concrete experiences of moral demands. Our effort to construct a rational basis for our moral views does not stop when we arrive at a coherent reflective equilibrium, or beUer, at reflective equilibria. We may confront these equilibria with each other and carryon the

Infanticide in Reflective Equilibrium?

249

discussion at the level of background theories. The relation between background theory, ethical principles and considered judgements are outlined in the following figure: Background theory Hedonisml utilitarianism

Rationalisml Kantianism

Ethical principle

Considered judgement

One ought to shorten a life void of value. One ought to shorten a life that is not of any value to anyone else.

Linda should be killed.

Human beings should be treated as ends.

Linda should be protected or Linda should be killed. I

Christianity

6

Human beings are inviolable Linda should be and should be protected. , protected.

Conclusion

In this article, I examined the question whether a mentally retarded infant, living with severe pain, should be killed or not. I reconstructed lines of argumentation, from judgements to principles, both in favour of and against the view that the infant should be killed. This kind of disagreement is also related to different background theories. It may be possible to argue even at this level, although the assessment of the arguments will ultimately be done from a personal point of view. While all human beings experience the same world, this assessment will not necessarily imply irrational or plainly subjective choices. The method of reflective equilibrium may help us to open up a position for argumentation and self-critical examination. But where does it lead us in this specific question? Is infanticide justified or not? Will the head of the ward and the staff nurse come to an agreement? So far, we have not enough ground to give a definitive answer. In our examination of background theories, we came across different, relevant theories of the nature and value of human beings. To be able to sketch a more integrated view on the value of human life, we should take into account both the importance of a special dignity of human life and the moral importance of suffering. This may imply a prima-facie duty always to protect human life.

I

250

Goran Collste

It may also, however, imply that one may face situations in which the prognosis of an infant's life is so bad that any person would prefer death to such a life of suffering without any hope of relief. Perhaps there are no final, rational solutions to this kind of moral dilemmas. Perhaps one either has to distance oneself from the specific case and establish some rules for this kind of decisions or, when engaging in the individual case, Ii ve in uncertainty and with an inescapable anguish.

Reflective Equilibrium as a Normative-Empirical Model in Bioethics Johannes J.M. van Delden and Ghislaine J.M. W. van Thiel

1

Introduction

You cannot do ethics until you know the facts. Therefore, the need for empirical research in ethics is very clear. Its task is to describe the morally relevant facts. Thus, facts provide the ethicist with the information she needs. In order to develop an ethically sound protocol about Do-Not-Resuscitate decisions, for instance, one has to know what the chances of success are of cardiopulmonary resuscitation and in what condition the survivor will spend the rest of her life. Likewise, one will want to know what the chances are of getting a certain disease, and what therapeutic possibilities medicine can offer before answering the question whether it is morally acceptable to perfonn a certain predictive genetic test. We take it that this contribution of empirical research to (bio)ethics is undisputed, so we will not go into this any further. Moreover, this kind of research will mostly be performed by doctors, geneticists, etc., not by ethicists. The question can be raised whether data about medical practice are merely the object about which statements are formulated, or whether these data themselves can have an input into the process of formulating arguments. In a deductive mode of reasoning, moral judgements are derived from principles. In this kind of theory usually only the fonner (passive) role is reserved for factual data. Mostly, a certain state of affairs is described, followed by the conclusion that this is desirable or not, given principle X. Period. Often, one of the four principles described by Beauchamp and Childress (1994) functions as X. Apart from this rather passive role of empirical data, we see other, more active ways in which empirical research can contribute to bioethics. In this chapter we will argue that: the object of such research should be practice-internal norms which can be found in the moral intuitions of those working in a practice; the reflective equilibrium model of normative thinking (hereafter: RE model) provides a good method for integrating this kind of empirical research into normative thinking. By implication we will also have to argue that: third-person intuitions can be used in the RE model.

251

W. van der Burg and T. van Willigenburg (eds.), Reflective Equilibrium, 251-259.

© 1998 Kluwer Academic Publishers. Printed in the Netherlands.

252

2

Johannes J.M. van Delden and Ghislaine J.M. W. van Thiel

Ethics and practice

2.1 Ethical reasoning and practice Since the end of the sixties, there has been an increase in public attention for the way medicine deals with patients and illness. One of the consequences has been a growing perception of the need for ethical involvement in medicine. This has changed medical ethics from a professional code of ethics of doctors into a subarea of ethics, taken as a philosophical discipline. Along with this change went a change in actor: while medical ethics used to be an undertaking of doctors themselves, it now became the domain of theologians and philosophers: of ethicists. Often, the work of these ethicists took the form of applied ethics: general ethical principles were applied to clinical dilemmas. One of the (mostly hidden) premises was that medical practice itself contained no special characteristics that should be involved in ethical reasoning. Context, practical wisdom, medical expertise and experience were not perceived as containing morally relevant information. To engage in ethics, one needed to apply general theory to the concrete situation. Many have criticised this deductivism in bioethics. According to Bayles (1984), for instance, principles are by nature formulated in a context-free way, and therefore have to be interpreted in a concrete case before they can do moral 'work' at all. Much of this interpretation remains hidden. Principles thus lead to rigid and fixed distinctions that fail to aid in concrete decision making. There are other points of critique (cf. Toulmin 1981; Van Willigenburg 1991; Beauchamp and Childress 1994) but here we would like to emphasise the point that ethics in its 'application' form fails to take real notice of the field it is judging. 2.2 The relevance ofpractice-internal norms for moral reasoning We think, however, that the practice of medicine does contain its own internal norms that need to be taken into account in moral reasoning. A defence of this statement should, of course, start with explaining what we mean by norms that are internal to a practice. To describe what we mean by a practice, we lean on Alisdair MacIntyre's definition in After Virtue: 'a practice is a coherent and complex form of socially established cooperative human activity' (MacIntyre 1984, 187). Thus understood, it will take little trouble to accept medicine as a practice. But what are nomlS internal to practices? In our view they are not the same as MacIntyre's 'goods internal to practices' (MacIntyre 1984). In his theory these goods are defined by the 'telos' of a practice. By 'norms internal to practices', however, we mean two things. On the one hand, it refers to

Reflective Equilibrium as a Normative-Empirical Model in Bioethics

253

'phronesis': practical wisdom that enables those working in a 'praxis' to act and judge well in concrete cases. In this interpretation internal norms form a subtle, context-driven way of specifying the well-known principles in concrete cases. Practice-internal norms are also action guides, however. They can consist of rules and ideas about how to deal with certain situations in a morally satisfactory way. When trying to describe these internal norms, one may look for the moral intuitions of those working in the practice. Moral intuitions are prereflective judgements about particular cases or situations (Van Willigenburg and Heeger 1989). It is precisely this characteristic, prereflectiveness, that makes intuitions fit for our goal: as soon as people start reflecting on a moral dilemma, the risk increases that unwanted influences colour their judgements. For example, people may echo a dominant theory, like the primacy of the principle of autonomy in health care. Another reason to pay attention to intuitions is that they are often the cause for people to start moral reasoning at all. In the 'application' -form of ethics, intuitions have little or no status. Indeed, DePaul (1993) observed that, oddly enough, the dominant view of analytic philosophy has been that in trying to construct a systematic moral view, we should ignore nearly all of the very moral beliefs that led us to formulate questions about morality in the first place. We, however, think that moral intuitions may provide a good start for normative reasoning.

3

RE thinking as a method of integration

3.1 The RE model If we are right so far, and if it is worthwile from a moral point of view to determine what norms function in practice and how, we also have to describe how that information may be integrated into normative reasoning. But first we should, perhaps, respond to those who are still troubled by the so-called 'is-ought' distinction. We think that we do not have to worry much about that matter here (if at all) because the type of information we aim at is value laden anyway. We do not go from 'is' to 'ought', but try to balance different kinds of 'oughts'. We think RE thinking provides a good way of integrating facts about nonns into moral reasoning. A first reason is the practical orientation of most RE thinking: it is a way of determining what is best to do. And since our main concern here is with bioethics, RE thinking fits well. It does so for another reason: it is a coherentist, non-foundationalist approach that facilitates a real dialogue between theory and practice by not assigning a preferential status to

254

Johannes J.M. van Delden and Ghislaine J.M. W. van Thiel

either of them. Thus, some of the problems of 'application' ethics (see 2.1) will not arise. This dialogue will consist of going back and forth between information stemming from practice (morally relevant facts and moral intuitions of caregivers) and from theory (principles and background theories). We will give an example below.

3.2 Third-person intuitions and the thinker At this point we have to be clear about a special feature of our proposal. Nonnally, RE thinking is thought of as being performed by one person (cf. Rawls 1971, 50), let us call him the thinker. This thinker starts thinking by fomlulating his own considered moral judgements. After this he tries to find moral principles that explicate those judgements. Since either some of the judgements or some of the principles will probably not fit, he will go on to adjust both of them until coherence is reached. We, however, propose to collect information about the moral intuitions of many about a certain moral dilemma. What does this input mean for the equilibrium reached? Is it the interpersonal equilibrium of many or just the intrapersonal equilibrium of the thinker? What happens is the following: the caregivers' intuitions are brought into the RE process in order to enrich the deliberation of the thinker with the norms and practical wisdom of the field. The intuitions of the thinker himself will also, of course, be involved in the RE process. Together with his knowledge of the morally relevant facts, these intuitions form the practical input into the RE process. Hence, the thinker will go on trying to reach a balance. This means that the deliberation process remains the one of the thinker, not of all the workers in the practice. There are several reasons for including information about moral intuitions of those working in practice in moral reasoning. First, it enriches moral reasoning because this kind of information could very well point to relevant aspects of the case or to solutions that one would not have thought of when starting from theory. Second, it enhances moral thinking by taking into account the difficulties and problems that a certain moral dilemma poses in practice. By doing so, and this is the third reason, the chance of implementation of the outcome of moral reasoning is increased. This is so for at least two reasons: doctors and nurses will recognise the attention for their situation and this will add to their willingness to use the moral advice. And also the applicability of the advice itself will have become greater by taking the particularities of medical practice into account. These reasons are all output related: to pay attention to practical norms enhances the strength of the advice and increases the chances of implementation.

Reflective Equilibrium as a Normative-Empirical Model in Bioethics

255

There is yet another reason, however, that is more related to the characteristics of the RE model itself. One of the weaknesses of the RE model is the risk of self-justification: coherence is not a guarantee for moral truth. This chapter is, of course, not the place to deal with this problem in detail (see the chapter by Herrmann in this volume). Nielsen (1982c) already pointed out that the initial credibility of moral judgements increases when they reflect a consensus of many. We would like to add that by including third-person intuitions one reduces the risk just mentioned in two ways. The first is that since the intuitions of many are brought into the RE process, there is a good chance of getting a pluralistic view on the matter at hand. The second is that, although we are primarily interested in the views of the majority, minority positions can be involved in the reasoning.

3.3 The network model We have proposed to integrate facts about moral intuitions into moral reasoning by means of the RE model. Not all versions of this model will do so equally well, however. The appropriateness of a version differs with the goal of the reasoning process. When the purpose is to describe a grand theory, like Rawls's (1971), one will need rather depersonalised considered moral judgements, abstract background theories and general principles. After all, the veil of ignorance was especially designed to be able to ignore morally relevant facts. When one wants to justify a specific course of action in a concrete case, like Van Willigenburg and Heeger (1989), one needs intuitions about the case at hand, information about the morally relevant facts of the case and some specific principles. In this chapter we deal with bioethics, and there, we think, an intermediate level of goals is best suitable: the level of 'modest theories'. Modest theories have a limited range. They cover some theoretical insights, may involve reflective infonnation on fairly concrete problems, but may 3;lso cover some meta-level reflection on the tenability of a specific type of ethical argument (Van Willigenburg 1991). Rules for careful decision making, for instance in a guideline for Do-No-Resuscitate decisions (Van Delden 1993), may serve as an example of such a modest theory. In order to formulate such a theory, we think the network model as described by Van Willigenburg and Heeger (1989) and enhanced by Van der Burg (1991) serves best. This is so because the network model is much more practice oriented than the Rawlsian wide reflective equilibrium and allows morally relevant facts and ideals to playa role in the RE process (see also the Introduction to this volume). The attraction of including ideals, as suggested by Van der Burg, lies in the fact that ideals fonn the nonnative background of principles. This prescriptivity of their own makes ideals particularly fit to correct

256

Johannes J.M. van Delden and Ghislaine J.M. W. van Thiel

for the tendency towards moral conservatism, which is a risk of RE thinking (cf. Van der Burg in this volume). Rather than elaborating theoretical issues any further, we would like to show how the approach we sketched above might work. To this end, we will discuss an example of how empirical information about moral intuitions can be incorporated in a network model in practice.

4

A case study: respecting autonomy in Dutch nursing homes

4.1 Introduction One of the tasks of the Dutch Association of Nursing-Home Care ('Nederlandse Vereniging van Verpleeghuizen' - NVVz) is to help improve the quality of nursing-home care. In order to fulfil this task, several activities have been developed. One of them resulted in criteria for high-quality nursing home care (NVVz 1993). Many of these criteria can be seen as derivations from a leading principle in health care, namely the principle of respect for patient autonomy. In a previous study we asked nurses if they supported the importance that is attributed to respect for patient autonomy and if they behaved as required by the criteria for good nursing home care that are related to respect for autonomy. At that time, we found that 96 per cent of the nurses agreed with the proposition that respect for patient autonomy should be the starting point of care. However, on the practical level, nurses had much trouble in meeting the criteria designed for respect for autonomy by the NVVz. Furthermore, the views of the nurses showed that fundamental problems arose when trying to meet these criteria (Van Thiel and Van Delden 1997). The results of this study gave us two reasons for doing further research on the subject. First and foremost, respect for (patient) autonomy is an important value in health care, and one that cannot be missed if one wants to protect patients against the loss of control over their bodies and their circumstances. Second, we have to acknowledge that the current understanding of the norm of respect for patient autonomy poses problems that need to be addressed, especially in long-term care settings. If we take respect for patient autonomy seriously, we have to develop an understanding of this norm that not only takes into account the limits of autonomous decision making of some patients, but also fits the moral experience of caregivers that give shape to this norm in day-to-day practice. As a first step towards an alternative view on respect for patient autonomy in nursing home care, we started an empirical study in order to elicit these moral intuitions.

Reflective Equilibrium as a Normative-Empirical Model in Bioethics

257

4.2 Methods The dominant understanding of respect for patient autonomy can be characterised as libertarian. This libertarian understanding, with its emphasis on liberty, is one of many possible interpretations of the norm. In the ethical literature, we found many alternative interpretations, which we categorised in three approaches. The first is the Kantian ideal of moral autonomy, in which rationality plays a central role. Second, there is a narrative understanding of autonomy, focusing on personal identity. The third interpretation is derived from an ethics of care perspective, in which interhuman relationships are crucial. We think that most suggestions that are made in the literature can be seen as ultimately based on one of these four (including the libertarian) interpretations of respect for patient autonomy. In line with these interpretations, we developed four views on patient autonomy in nursing home care. Each of these consisted of the same elements, namely: a concept of persons (containing a description of what is considered important in life according to the view concerned), a characterisation of good care and a view on dealing with patients who are not (fully) competent. In a questionnaire these views were presented. We asked 100 nurses and 50 nursinghome physicians (response rates 94 per cent and 62 per cent respectively) to indicate which of the views they would prefer if they could choose one for their own nursing home. The response to this question gives information about the more general views on autonomy of caregivers. But which approach do nurses and physicians have when confronted with (descriptions ot) concrete situations? Do they still think that the approach they preferred regardless of specific circumstances, is also the best one when they know more about some relevant facts of a case? In order to answer these questions we designed ten vignettes. Vignettes are concise case descriptions in which a limited number of variables are varied. The respondents were asked to choose one out of four options (each of them based on one of the four interpretations of autonomy) for each vignette. Each option was a comment on the case and suggested an approach to the issue at stake. The respondents were instructed not to choose the option that best described every-day practice, but to pick the one that best reflected their normative ideas. 4.3 Results The libertarian interpretation of respect for autonomy proved to be the most popular: 39 per cent chose the libertarian view. Of the respondents, 33 per cent preferred the narrative understanding, 18 per cent the Kantian and 10 per cent the ethics of care perspective. Those who chose the narrative approach were significantly more often employed in psychogeriatric wards. The most

258

Johannes J.M. van Delden and Ghislaine J.M. W. van Thiel

straightforward explanation for this is probably that insight into someone' s lifestory may help when decisions have to be made for incompetent patients. As we mentioned above, the view that is based on an ethics of care was chosen by only 10 per cent of the respondents. We had expected this percentage to be higher because in a nursing home setting elements of life such as vulnerability and dependency are more important than in general. The ethics of care perspective is the view that is - more than the other views - sensitive to these elements. When we look at the interpretation of respect for autonomy that is preferred in more concrete case descriptions, even more questions arise. The results show that the choice for a view on good nursing home care does not correlate with the choice of approach when confronted with case descriptions. In more concrete cases, the influence of relevant circumstances is significant. We found strong influences of most (combinations ot) variables on the chosen approach in the vignettes. We conclude that caregivers make radically different choices when confronted with concrete case descriptions instead of a more general view on good care. This means that caregivers generally agree with a certain understanding of respect for autonomy in care, but still think that it should be interpreted differently when circumstances change. Ideas on how patients' autonomy should be respected are heavily dependent on the circumstances of the case. In general, the results showed that caregivers tend to shape their respect for autonomy according to the specific needs or abilities of patients. They do not think that patients should be treated equally under all circumstances.

4.4 From empirical data to reflective equilibrium After having obtained the information about the moral intuitions of our respondents, we had to look at a different level of moral information: the principles. As stated above, our goal was to describe a view on respecting patient autonomy in a nursing home setting. Therefore, we had to confront the data with the four interpretations of the principle of respect for patient autonomy. Obviously, the data did not point to one view as the most favoured one under all circumstances. In order to reach a coherent view, we had to elaborate the most important ideas lying behind the four views on autonomy: liberty, rationality, personal identity and interhuman relationships. By doing so, we were able to describe why one view was preferable in one situation while another one was under other circumstances. This led to what one may call a narrow reflective equilibrium: a coherent set of statements based on intuitions and different interpretations of the principle of autonomy. Clearly, more independent input was necessary in order to reach

Reflective Equilibrium as a Normative-Empirical Model in Bioethics

259

a wide equilibrium. This is, of course, a common step in the RE process, but we had a special, additional reason. In this project some information about the different views on autonomy was already given in the questionnaire. Thus, intuitions and principles were not completely independent. (This is not to say that we ended up only with information we provided ourselves: the many surprises in the data refute this idea.) The independent input we found, as stated in 3.3, in ideals. For this purpose we analysed views on the relationship between patient and caregiver, ideas about being a good caregiver, and the ideal of autonomy. The result is a subtle, balanced view on how to respect patient autonomy in a setting where autonomy cannot be taken for granted for many reasons. Respect for autonomy is, for example, understood differently when dealing with fully competent patients and when caring for incompetent ones. Communication on the basis of reason is considered an important feature of respect for autonomy of the former patients. In dealing with incompetent patients, however, a more relational interpretation is suggested.

5

Conclusion

We have argued for including the intuitions of those working in a practice into nonnative thinking, and we have stated why the network version ofRE thinking provides a good method for integrating such information. The view behind both these claims is that only by doing so one takes real notice of the normativity embedded in the practice itself. We think this is a prerequisite for a real dialogue between theory and practice. As was stated in the introduction: to do ethics one has to know the facts. These are facts about the consequences and circumstances (for instance), but also facts about the nonns that function in practice and how. Only then can good facts make good ethics.

Bibliography

Advisory Committee Ethics and Biotechnology in Animals (1990), 'Ethics and Biotechnology in Animals', Wageningen: Netherlands Council for Agricultural Research. Alcoff, L. and E. Potter (eds.) (1993), Feminist Epistemologies, London: Routledge. Antony, L.M. and ChI Witt (1993), A Mind of One's Own: Feminist Essays on Reason and Objectivity, Boulder, Colo.: Westview Press. Applbaum, A.I. (1998), 'Are Violations of Rights Ever Right?', Ethics 108, 340-366. Aquinas, Thomas (1952), The Summa Theologica, I, Chicago: Encyclopaedia Britannica. Archard, D. (1995), 'Moral Partiality', Midwest Studies in Philosophy XX, 129141. Aronovitch, H. (1996), 'Reflective Equilibrium or Evolving Tradition?', Inquiry 39,399-419. Audi, R. (1993), 'Ethical Reflectionism', The Monist 76,295-315. Audi, R. (1996), 'Intuitionism, Pluralism, and the Foundations of Ethics', in W. Sinnott-Armstrong and M. Timmons (eds.), Moral Knowledge? New Readings in Moral Epistemology, New York/Oxford: Oxford University Press, pp. 101-136. Austin, J.L. (1979), 'A plea for excuses', in J.L. Austin, Philosophical Papers (ed. by J.O. Urmson and G.J. Warnock), New York/Oxford: Oxford University Press, pp. 175-204. AVMA (1991), Animal Welfare Position Statements and Background Information, American Veterinary Medical Association. Babbitt, S.E. (1993), 'Feminism and Objective Interests: The Role of Transfonnation Experiences in Rational Deliberation', in L. Alcoff and E. Potter (eds.) (1993), Feminist Epistemologies, London: Routledge, pp. 245-263. Baier, A. (1985), Postures of the Mind, Minneapolis: University of Minnesota Press. Barbour, lG. (1974), Myths, Models and Paradigms, London: S.C.M. Press. Barry, B. (1989a), Theories ofJustice, Berkeley: University of California Press. Barry, B. (1989b), 'The Light that Failed? A Review of Alasdair MacIntyre, Whose Justice? Which Rationality?' , Ethics 100, 160-168. Barry, B. (1990), Political Argument, Berkeley: University of California Press.

262

Bibliography

Barry, B. (1995), 'John Rawls and the Search for Stability', Ethics 105, 874-915. Barth, K. (1936), Church Dogmatics, 1/1 (trans!. by G.T. Thomson et a1.), Edinburgh: Clark. Barth, K. (1951), Kirchliche Dogmatik, 11114, Munchen: Chr. Kaiser Verlag. Barth, K. (1963), The Humanity of God, (transl. by Th. Wieser), Richmond, Va.: John Knox Press. Bayertz, K. (1994), 'Introduction: Moral Consensus as a Social and Philosophical Problem', in K. Bayertz (ed.), The Concept ofMoral Consensus: The Case of Technological Interventions into Human Reproduction, Dordrecht: Kluwer Academic Publishers, pp. 1-15. Bayles, M.D. (1984), 'Moral Theory and Application', Social Theory and Practice 10, 97-120. Beauchamp, T.L. and J.F. Childress (1983), Principles ofBiomedical Ethics, 2nd edn., New York/Oxford: Oxford University Press. Beauchamp, T.L. and J.F. Childress (1994), Principles ofBiomedical Ethics, 4th edn., New York/Oxford: Oxford University Press. Beck, U. (1992), Risk Society: Towards a New Modernity, London: Sage. Benn, S.I. (1988), A Theory of Freedom, Cambridge: Cambridge University Press. Bentham, J. (1988), Principles ofMorals and Legislation, Amherst: Prometheus Books. Berger, P. and Th. Luckmann (1967), The Social Construction of Reality: A Treatise in the Sociology of Knowledge, New York: Anchor. Bertram, Chr. (1997), 'Political Justification, Theoretical Complexity, and Democratic Community', Ethics 107, 563-583. Black, M. (1962), Models and Metaphors, Ithaca N.Y.: Cornell University Press. Blackburn, S. (1987), 'How to Be an Ethical Antirealist', in P.A. French, Th.E. Uehling, Jr. and H.K. Wettstein (eds.), Midwest Studies in Philosophy, Vol. XII: Realism and Antirealism, Notre Dame: University of Notre Dame Press, pp. 361-375. Blackburn, S. (1993), 'Can Philosophy Exist?', in J. Couture and K. Nielsen (eds.), Meta-Philosophie: Reconstructing Philosophy? New Essays on Metaphilosophy, Calgary: University of Calgary Press, pp. 163-187. Bondolfi, A. (1994), 'Coming to Consensus: An Ethical Problem in Law and Politics', in K. Bayertz (ed.), The Concept ofMoraI Consensus: The Case of Technological Interventions into Human Reproduction, Dordrecht: Kluwer Academic Publishers: pp. 123-128. Braithwaite, R.B. (1964), 'An Empiricist's View on the Nature of Religious Belief', in J. Hick (ed.), The Existence of God, New York: Macmillan. Brandt, R.B. (1959), Ethical Theory: The Problems ofNormative and Critical Ethics, Englewood Cliffs, N.J.: Prentice-Hall.

Bibliography

263

Brandt, R.B. (1979), A Theory of the Good and the Right, Oxford: Clarendon Press. Brandt, R.B. (1989), 'Morality and Its Critics', American Philosophical Quarterly 26,2,89-100. Brandt, R.B. (1996), 'Science as a Basis for Moral Theory', in W. SinnottArmstrong and M. Timmons (eds.), Moral Knowledge? New Readings in Moral Epistemology, New YorlclOxford: Oxford University Press, pp. 200-214. Brink, D.O. (1987), 'Rawlsian Constructivism in Moral Theory', Canadian Journal of Philosophy 17, 1, 71-90. Brom, F.W.A. (1995), Waarom is biotechnologie bij dieren moreel problematisch? Een inventarisatie en analyse van argumenten uit het publieke debat, Utrecht: Centrum voor Bio-ethiek en Gezondheidsrecht. Brom, F.W.A. (1997), Onherstelbaar verbeterd: Biotechnologie bij dieren als een moreel probleem (doct. thesis Utrecht University), Assen: Van Gorcum. Brummer, V. (1981), Theology and Philosophical Inquiry, London: Macmillan. Brummer, V. (1992), Speaking of a Personal God, Cambridge: Cambridge University Press. Brummer, V. (1993a), The Model of Love, Cambridge: Cambridge University Press. Brummer, V. (1993b), 'Wittgenstein and the Irrationality of Rational Theology' , in 1.M. Byrne (ed.), The Christian Understanding of God Today, Dublin: Columba Press. Brummer, V. (1997), 'The Identity of the Christian Tradition' (valedictory lecture Utrecht University), Utrecht: Faculty of Theology. Buber, M. (1962), Jag och Du, Stockholm: Bonniers. Carens, J.H. (1995), 'Aliens and Citizens: The Case for Open Borders', in W. Kymlicka (ed.), The Rights of Minority Cultures, Oxford: Oxford University Press, pp. 331-349. Carr, S. (1975), 'Rawls, Contractarianism and Our Moral Intuitions', The Personalist 56, 83-95. Chemiak, C. (1994), 'Rationality', in S. Guttenplan (ed.), A Companion to the Philosophy of Mind, Oxford: Blackwell, pp. 526-531. Chisholm, R. (1966), Theory of Knowledge, Englewood Cliffs, N.J.: PrenticeHall. Christian, W.A. (1964), Meaning and Truth in Religion, Princeton: Princeton University Press. Cladis, M. (1994), 'Wittgenstein, Rawls and Conservatism', Philosophy and Social Criticism 20, 29-41. Clarke, S.G. (1987), 'Anti-Theory in Ethics' ,American Philosophical Quarterly 24,3,237-244.

264

Bibliography

Cohen, J. (1994), 'A More Democratic Liberalism', Michigan Law Review 92, 1506-1543. Collste, G. (1998), Is Human Life Special? Philosophical Perspectives, Linkoping: Centre for Applied Ethics. Cupitt, D. (1981), Taking Leave o/God, London: S.C.M. Press. Dalferth, I.U. (1997), Gedeutete Gegenwart. Zur Wahrnehmung Gottes in den Erfahrungen der Zeit, Tiibingen: Mohr/Siebeck. Dancy, J. (1981), 'On Moral Properties', Mind 90,367-385. Dancy, J. (1984), 'On Coherence Theories of Justification: Can an Empiricist be a Coherentist?', American Philosophical Quarterly 21, 4, 359-365. Dancy, J. (1985), Introduction to Contemporary Epistemology, Oxford: Blackwell. Dancy, J. (1993), Moral Reasons, Oxford: Blackwell. Daniels, N. (ed.) (1975), Reading Rawls, Oxford: Blackwell. Daniels, N. (1979a), 'Wide Reflective Equilibrium and Theory Acceptance in Ethics', Journal of Philosophy 76, 5, 256-282. Daniels, N. (1979b), 'Rights to Health Care and Distributive Justice', Journal ofMedicine and Philosophy 4, 174-191. Daniels, N. (1979c), 'Moral Theory and the Plasticity of Persons', The Monist 62, 3, 267-287. Daniels, N. (1980a), 'On Some Methods of Ethics and Linguistics', Philosophical Studies 37,21-36. Daniels, N. (1980b), 'Reflective Equilibrium and Archimedean Points', Canadian Journal of Philosophy 10, 1,83-103. Daniels, N. (1980c), 'Reflective Equilibrium and Theory Acceptance in Ethics', Canadian Journal of Philosophy 10, 1, 83-103. Daniels, N. (1988), Am I My Parents' Keeper? An Essay on Justice between the Young and the Old, New York: Oxford University Press. Daniels, N. (1996), Justice and Justification: Reflective Equilibrium in Theory and Practice, Cambridge: Cambridge University Press. Danto, A.C. (1976), Mysticism and Morality: Oriental Thought and Moral Philosophy, Harmondsworth: Penguin. Davidson, D. (1994), 'Davidson, Donald', in S. Guttenplan (ed.), A Companion to the Philosophy of Mind, Oxford [etc.]: Blackwell, pp. 231-236. Davis, R.B. (1995), 'The Principlism Debate: A Critical Overview' , Journal ofMedicine and Philosophy 20, 85-105. De Boer, I.J.M., F.W.A. Brom and I.M.G. Vorstenbosch (1995), 'An Ethical Evaluation of Animal Biotechnology: The Case of Using Clones in Dairy Cattle Breeding', Animal Science 61, 453-463.

Bibliography

265

DeGrazia, D. (1992), 'Moving Forward in Bioethical Theory: Theories, Cases, and Specified Principlism' ,Journal ofMedicine and Philosophy 17, 511-539. Delaney, C.F. (1977), 'Rawls on Method', Canadian Journal of Philosophy, suppl. 111,153-161. Delaney, C.F. (1995), 'John Dewey', in R. Audi (ed.), The Cambridge Dictionary of Philosophy, Cambridge: Cambridge University Press, pp. 198-200. Den Hartogh, G. (1996), 'Soziale nnd kritische Moral', in K. Golser and R. Heeger (eds.) Moralerziehung im neuen Europa, Brixen: Verlag A. Weger. DePaul, M.R. (1986), 'Reflective Equilibrium and Foundationalism' ,American Philosophical Quarterly 23, 59-69. DePaul, M.R. (1993), Balance and Refinement: Beyond Coherence Methods of Moral Inquiry, London/New York: Routledge. Dewey, J. (1916/1966), Democracy and Education, New York: Free Press. Dillon, R.S. (1997), 'Self-Respect: Moral, Emotional, Political' , Ethics 107, 2,226-249. Donogan, A. (1977), The Theory of Morality, Chicago [etc.]: The University of Chicago Press. Duncan, I.J.H. and D. Fraser (1997), 'Understanding Animal Welfare' , in M.C. Appleby and B.O. Hughes (eds.), Animal Welfare, Wallingford: CAB International, pp. 19-31. Dworkin, R. (1978), Taking Rights Seriously, Carrlbridge, Mass.: Harvard University Press. Dworkin, R. (1986), Law's Empire, London: Fontana. Easton, D. (1971), The Political System: An Inquiry into the State of Political Science, Chicago: Chicago University Press. Ebertz, R. (1993), 'Is Reflective Equilibrium a Coherentist Model?', Canadian Journal of Philosophy 23, 193-214. English, J. (1978), 'Ethics and Science', Proceedings o/the XVI World Congress of Philosophy, pp. 227-233. Feenstra, M. and J. de Rooij (eds.) (1994), Het publiek debat: Praktijk, ethiek en methodiek, Den Haag: Nederlandse Organisatie voor Technologisch Aspectenonderzoek. Feinberg, J. (1986), Harm to Self, OxfordlNew York: Oxford University Press. Frankena, W.K. (1981), 'Is Morality Logically Dependent on Religion?' , in P. Helm (ed.), Divine Commands and Morality, Oxford: Oxford University Press. Fraser, D., D.M.Weary, E.A. Pajor and B.N. Milligan (1997), 'A Scientific Conception of Animal Welfare that Reflects Ethical Concerns', Animal Welfare 6, 187-205. Frijda, N. (1986), The Emotions, London: Cambridge University Press.

266

Bibliography

GaBie, W.B. (1955-56), 'Essentially Contested Concepts' , Proceedings of the Aristotelian Society 56, 167-198. Gardavsky, V. (1973), God Is not yet Dead, (transl. by V. Menkes), Harmondsworth: Penguin. Geach, P. (1969), God and the Soul, London: Routledge & Kegan Paul. Gibbard, A. (1995), 'Why Theorize How to Live With Each Other', Philosophy and Phenomenological Research 55, 2, 323-342. Glover, J. (1977), Causing Death and Saving Lives, Harmondsworth: Penguin. Goodman, N. (1955), Fact, Fiction and Forecast, Cambridge, Mass.: Harvard University Press. Graham, K. (1975), 'Moral Notions and Moral Misconceptions' , Analysis 35, 3,65-78. Grayling, A.C. (1996), 'Epistemology', in N. Bunnin and E.P. Tsui-James (eds.), The Blackwell Companion to Philosophy, Oxford: Blackwell, pp. 38-63. Greenawalt, K. (1994), 'On Public Reason', Chicago-Kent Law Review 69, 669-690. Grice, G.R. (1978), 'Moral Theories and Received Opinion', Proceedings of the Aristotelian Society, suppl. 52, 1-12. Griffin, J. (1996), Value Judgements: Improving Our Ethical Beliefs, Oxford [etc.]: Clarendon Press. Grommers, F.J. (1997), 'Consciousness, Science and Conscience', in M. Dol, S. Kasanmoentalib, S. Lijrrlbach, E. Rivas and R. van den Bos (eds.), Animal Consciousness and Animal Ethics, Assen: Van Gorcum, pp. 198207. Guthrie, W.K.C. (1971), Socrates, Cambridge: Cambridge University Press. Gutting, G. (1982), 'Can Philosophical Beliefs Be Rationally Justified?', American Philosophical Quarterly 4, 315-330. Haack, S. (1992), 'Pragmatism', in J. Dancy and E. Sosa (eds.), A Companion to Epistemology, Oxford: Blackwell, pp. 351-357. Haack, S. (1996), 'Pragmatism', in N. Bunnin and E.P. Tsui-James (eds.), The Blackwell Companion to Philosophy, Oxford: Blackwell, pp. 643-661. Hampshire, S. (1983), Morality and Conflict, Cambridge, Mass.: Harvard University Press. Hampshire, S. (1982), Thought and Action, London: Chatto and Windus. Hampton, J. (1989), 'Should Political Philosophy Be Done Without Metaphysics?', Ethics 99, 4,791-814. Hanen, M. (1983), 'Justification as Coherence', in M.A. Stewart (ed.), Law, Morality and Rights, Dordrecht [etc.]: Reidel, pp. 67-92. Hare, R.M. (1971), 'The Argument from Received Opinion', in R.M. Hare, Essays on Philosophical Method, London: Macmillan, pp. 122-136.

Bibliography

267

Hare, R.M. (1981), Moral Thinking: Its Levels, Method, and Point, Oxford: Oxford University Press. Hare, R.M. (1996), 'Foundationalism and Coherentism in Ethics', in W. SinnottArmstrong and M. Timmons (eds.), Moral Knowledge? New Readings in Moral Epistemology, New Yorlc/Oxford: Oxford University Press, pp. 190-199. Harman, G. (1978), 'Reasons', in J. Raz (ed.), Practical Reasoning, Oxford: Oxford University Press, pp. 110-117. Harris, J. (1985), The Value ofLife: An Introduction to Medical Ethics, London: Routledge & Kegan Paul. 11art, H.L.A. (1961), The Concept of Law, Oxford: Clarendon Press. Hart, H.L.A. (1977), 'Immorality and Treason', in R. Dworkin (ed.), The Philosophy of Law, Oxford: Oxford University Press: pp. 83-88. Haslett, D.W. (1987), 'What Is Wrong with Reflective Equilibria?', Philosophical Quarterly 37, 305-311. Heckmann, G. (1993), Das sokratische Gespriich: Erfahrungen in philosophischen Hochschulseminaren, Bonn: Philosophisch-Politische Akademie. Heeger, F.R. (1988), 'Mensen, dieren en principes: Over de ethische toetsing van proeven met dieren', Filosofie en Praktijk, 9, 4,169-180. Heeger, F.R. (1990a), 'Ethische Knotenpunkte in. der Biotechnologie', in Zentrumfiir Niederlande-Studien, Jahrbuch 1990, Munster: Regensberg, pp. 113-118. . Heeger, F.R. (1990b), 'The Virtue of Intellectual Honesty' , in R. Veldhuis, A.F. Sanders and H. Siebrand (eds.), Belief in God and Intellectual Honesty, Assen: Van Gorcum, pp. 11-25. Heeger, F.R. (1992a), 'Eigenwert uDd Verantwortung: Zur normativen Argumentation in der Tierethik', in W. Harle, M. Marquardt and W. Nethofel (eds.), Vnsere Welt: Gottes Schopfung (Marburger Theologische Studien 32), Marburg: Elwert, pp. 251-267. Heeger, F.R. (1992b), 'Views of Life in New Situations: Moral Thinking about Animal Experiments and Biotechnology' , in G. van den Brink, L. van den Brom and M. Sarot (eds.), Christian Faith and Philosophical Theology, Kampen: Kok Pharos, pp. 92-102. Heeger, F.R. (1992c), 'Ansvarig Deliberation vid Etisk Radgivning' , in C.H. Grenholm and G. Lantz (eds.), Etik, Religion och Samhiille, Nora: Nua Doxa, pp. 137-151. Heeger, F.R. (1993), 'What is Meant by "The Tum to Applied Ethics"?', in F.R. Heeger and T. van Willigenburg (eds.), The Turn to Applied Ethics: Practical Consequences for Research, Education, and the Role ofEthicists in Public Debate, Kampen: Kok Pharos, pp. 9-16.

268

Bibliography

Heeger, F.R. (1997), 'Respect for Animal Integrity?', in A. Nordgren (ed.), Science, Ethics, Sustainability: The Responsibility ofScience in Attaining Sustainable Development (Studies in Bioethics and Research Ethics 2), Uppsala: Acta Universitatis Upsaliensis, pp. 243-252. Hepburn, R.W. (1957), 'Poetry and Religious Belief', in A. MacIntyre (ed.), Metaphysical Beliefs, London: S.C.M. Press. Herman, B. (1996), The Practice of Moral Judgement, Cambridge, Mass.: Harvard University Press. Herron, M.R. (1967), 'Declawing the Cat', Modern Veterinary Practice 48, 40-43. Hill, Th. (1973), 'Servility and Self-Respect', The Monist 57,87-104. Holmgren, M. (1989), 'The Wide and Narrow of Reflective Equilibrium', Canadian Journal of Philosophy 19, 43-60. Hudson, W.D. (1969), 'Some Remarks on Wittgenstein' s Account of Religious Belief', in G.N.A. Vesey (ed.), Talk ofGod (Royal Institute of Philosophy Lectures, Vol. II), London: Macmillan. Hutcheson, F. (1738 (4th edn.)/1969), 'An Inquiry concerning Moral Good and Evil', in D.D. Raphael (ed.), British Moralists 1650-1800, Vol. I, Oxford: Clarendon Press, pp. 261-299. Inglehart, R. (1997), Modernization and Postmodernization, Princeton, N.J.: Princeton University Press. Jamieson, D. (1991), 'Method and Moral Theory', in P. Singer (ed.), A Companion to Ethics, Oxford: Blackwell, pp. 476-487. Jeffner, A. (1989), Att Studera Manniskosyn: En oversiktlig Problemanalys, Linkoping: Universitet i Linkoping. Jeffner, A. (1992), 'A New View of the World Emerging among Ordinary People', in G. van den Brink, L. van den Brom and M. Sarot (eds.), Christian Faith and Philosophical Theology, Kampen: Kok Pharos, pp. 137-145. Jeffner, A. (1997), 'Vart tar livsfagorna vagen', in Bildningsgang, Natur och Kultur 75 ar, Stockholm: Natur och KuItur. Jennings, B. (1991), 'Possibilities of Consensus: Toward Democratic Moral Discourse', Journal of Medicine and Philosophy 16,4,447-463. Jonsen, A.R. and S. Toulmin (1988), The Abuse of Casuistry, Berkeley: University of California Press. Jurgens, E.C.M. (1995), 'Het publieke debat in het parlement: Een fictie', in K. van Kersbergen and I. Propper (eds.), Publiek debat en democratie, Den Haag: Sdu, pp. 99-116. Kamm, F.M. (1992), 'Non-Consequentialism, the Person as an End-in-Itself, and the Significance of Status', Philosophy & Public Affairs 21, 354-389.

Bibliography

269

Kant, I. (1785/1983),· 'Grounding for the Metaphysics of Morals', in I. Kant, Ethical Philosophy, Indianapolis: Hackett Publishing Company. Kant, I. (1793/1968), Die Religion innerhalb der Grenzen der blossen Vernunft, Darmstadt: Wissenschaftliche Buchgesellschaft. Kekes, J. (1986), 'Moral Intuition', American Philosophical Quarterly 23,1, 83-93. Kenny, A. (1983), Faith and Reason, New York: Columbia University Press. Kessels, J.P.A.M. (1989), 'Kennis van kennis: Een ontwikkelingsonderzoek in didactiek van filosofie' (doct. thesis Utrecht University). Kessels, J.P.A.M. (1994), 'Inleiding: De socratische methode bij Plato en Nelson', in L. Nelson (1994), De socratische methode, (ed. by J.P.A.M. Kessels), Amsterdam: Boom Meppel, pp. 7-44. Kessels, J.P.A.M. (1997), Socrates op de markt: Filosofie in bedrijf, Amsterdam: Boom Meppel. Kessels, J.P.A.M. and F.A. Korthagen .(1996), 'The Relationship Between Theory and Practice: Back to the Classics' , Educational Researcher 25, 3,17-22. Klop, C.J. (1997), Bestaat er publieke moraal? (inaugural lecture) Nijmegen: University of NijmegenlFaculty of Philosophy. Klosko, G. (1993), 'Rawls's "Political" Philosophy and American Democracy' , American Political Science Review 87, 348-359. KNMvD (1992), Code voor de dierenarts, Utrecht: Koninklijke Nederlandse Maatschappij voor Diergeneeskunde. Kohl, M. (ed.) (1978), Infanticide and the Value ofLife, New York: Prometheus Books. Kohlberg, L. (1981), Essays on Moral Development, Vol. I, The Philosophy ofMoral Development: Moral Stages and the Idea ofJustice, San Francisco [etc.]: Harper & Row. Korsgaard, Chr. (1996a), Creating the Kingdom ofEnds, Cambridge: Cambridge University·Press. Korsgaard, Chr. (1996b), The Sources ofNormativity, Cambridge: Cambridge University Press. Kovesi, J. (1967), Moral Notions, LondonlNew York: Routledge & Kegan PaullHumanities Press. Kuhse, H. and P. Singer (1985), Should the Baby Live? The Problem of Handicapped Infants,Oxford: Oxford University Press. Kymlicka, W. (1993), 'Moral Philosophy and Public Policy: The Case ofNRTs', Bioethics 7, 1-26. Levinas, E. (1988), Etik och oiindlighet, Stockholm: Symposion. Little, D. (1984), 'Reflective Equilibrium and Justification', Southern Journal of Philosophy 22, 373-387.

270

Bibliography

Lucas, J.R. (1976), Freedom and Grace, London: S.P.C.K. Lyons, D. (1975), 'Nature and Soundness of the Contract and Coherence Arguments', in N. Daniels (ed.), Reading Rawls, Oxford: Blackwell, pp. 141-167. MacCallum, G. (1967), 'Negative and Positive Freedom', The Philosophical Review 76,312-34. MacIntyre, A. (1984), After Vil1ue, Notre Dame: University of Notre Dame Press. Mackie, J. (1977), Ethics: Inventing Right and Wrong, Hannondsworth: Penguin Books. Majone, G. (1989), Evidence, Arguments and Persuasion in the Policy Process, New Haven: Yale University Press. Mavrodes, G. (1970), Beliefin God, New York: Random House. McDowell, J. (1981), 'Non-Cognitivism and Rule-Following', in S.H. Holtzman and C.M. Leich (eds.), Wittgenstein to Follow a Rule, London: Routledge & Kegan Paul, pp. 141-162. McDowell, J. (1985), 'Values and Secondary Qualities', in T. Honderich (ed.), Morality and Objectivity, London: Routledge & Kegan Paul, pp. 110-129. McFague, S. (1982), Metaphorical Theology, London: S.C.M. Press. McNaughton, D. (1996), 'An Unconnected Heap of Duties?', Philosophical Quarterly 46,433-447. Meyers, D. (1993), 'Moral Reflection: Beyond Impartial Reasons', Hypatia 8,3,21-47. Miles, T.R. (1959), Religion and the Scientific Outlook, London: Allen & Unwin. Mill, J.S. (1859/1974), On Liberty, Harmondsworth: Penguin. Mill, J.S. (1863/1969), 'Utilitarianism', in J.M. Smith and E. Sosa (eds.), Mill's Utilitarianism: Text and Criticism, Belmont: Wadsworth, pp. 31-88. Miller, D. (1992), 'Distributive Justice: What the People Think', Ethics 102, 3,555-593. Miller, D. (1994), 'Book Review of K.R. Scherer (ed:), Justice: Interdisciplinary Perspectives', Social Justice Research 7, 2, 167-188. Mitchell, B. (1981), The Justification of Religious Belief, New York: Oxford University Press. Murdoch, I. (1962), 'Metaphysics and Ethics', in D.F. Pears (ed.), The Nature of Metaphysics, London: Macmillan, pp. 99-123. Nelson, L. (1970), Gesammelte Schriften in Neun Biinden (ed. by P. Bernays and W. Eichler et al.), Hamburg: Felix Meiner. Nelson, L. (1994), De socratische methode, (ed. by J.P.A.M. Kessels), Amsterdam: Boom Meppel. Nelson, M.T. (1990), 'Intuitionism and Conservatism', Metaphilosophy 21, 3,282-293. Nielsen, K. (1977), 'Unsere wohlerwogene Urteile' , Ratio 19, 40-48.

Biblio graphy

271

Nielsen, K. (1982a), 'Grounding Rights and a Method of Reflective Equilibrium', Inquiry 25, 277-306. Nielsen, K. (1982b), 'On Needing aMoral Theory', Metaphilosophy 13,97-111. Nielsen, K, (1982c), 'Considered Judgements Again', Human Studies 5, 117129. Nielsen, K. (1985), Equality and Liberty, Totowa, N.J.: Rowman and Allenfield. Nielsen, K. (1988), 'In Defence of Wide Reflective Equilibrium', in D. Odegard (ed.), Ethics and Justification, Edmonton, Alb.: Academic Printing and Publishing, pp. 19-37. Nielsen, K. (1991), 'Rawls and the Socratic Ideal', Analyse & Kritik 13,67-93. Nielsen, K. (1993), 'Relativism and Wide Reflective Equilibrium', The Monist 76, 316-332. Nielsen, K. (1994a), 'Methods of Ethics: Wide Reflective Equilibrium and a Kind of Consequentialism', Journal of Philosophy 25, 2, 57-72. Nielsen, K. (1994b), 'How to Proceed in Social Philosophy: Contex~a1ist Justice and Wide Reflective Equilibrium', Queen's Law Journal 20, 89-137. Noble, C.N. (1979), 'Normative Ethical Theories', The Monist 62, 496-509. Norman, W. (1998), '''Inevitable and Unacceptable": Methodological Rawlsianism in Anglo-American Political Philosophy' , Political Studies 46, 2, 276-294. Nussbaum, M. (1986), The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy, Cambridge [etc.]: Cambridge University Press. NVVz (1993), Kwaliteitskenmerken verpleeghuiszorg, Utrecht: Nederlandse Vereniging van Verpleeghuizen. Otto, R. (1924), Det heliga, Stockholm: SKD bokforlag. Parfit, D. (1973), 'Later Selves and Moral Principles', in A. Montefiore (ed.), Philosophy and Personal Relations, London: Routledge & Kegan Paul, pp. 137-169. Plantinga, A. (1993a), Warrant: The Current Debate, New York: Oxford University Press. Plantinga, A. (1993b), Warrant and Proper Function, Oxford: Oxford University Press. Postema, G. (1992), 'Public Faces - Private Places: Liberalism and the Enforcement of Morality', in A.W. Musschenga et al. (eds.), Morality, Worldview and Law: The Idea of a Universal Morality and Its Critics, Assen: Van Gorcum, pp. 153-175. Propper, I. and W.J. Witteveen (1995), 'Goede en slechte debatten', in K. van Kersbergen and I. Propper (eds.), Publiek debat en democratie, Den Haag: Sdu, pp. 13-29. Putnam, H. (1983), 'Realism and Reason', in H. Putnam, Philosophical Papers, Vol. 3, Cambridge: Cambridge University Press.

272

Bibliography

Quine, W.V.O. (1960), Word and Object, Cambridge, Mass.: MIT Press. Quine, W.V.O. (1969), Ontological Relativity and Other Essays, New York: Columbia University Press. Raikka, J. (1996-97), 'Are There Alternative Methods in Ethics?', Grazer Philosophische Studien 52,173-189. Ramsey, P. (1970), The Patient as Person, New Haven/London: Yale University Press. Ramsey, P. (1978), Ethics at the Edges ofLife: Medical and Legal Intersections, New Haven: Yale University Press. Rawls, J. (1951), 'Outline of a Decision Procedure for Ethics', Philosophical Review 60,177-197. Rawls, J. (1971), A Theory of Justice, London: Oxford University Press. Rawls, J. (1974-75), 'The Independence of Moral Theory', in Proceedings and Addresses of the American Philosophical Association, 48,5-22. Rawls, J. (1980), 'Kantian Constructivism in Moral Theory', Journal of Philosophy 77, 515-572. Rawls, J. (1985), 'Justice as Fairness: Political not Metaphysical', Philosophy & Public Affairs 14,223-251. Rawls, J. (1987), 'The Idea of an Overlapping Consensus', Oxford Journal of Legal Studies 7, 1, 1-25. Rawls, J. (1989), 'The Domain of the Political and Overlapping Consensus', New York University Law Review 64, 2, 233-255. Rawls, J. (1993), Political Liberalism, New York: Columbia University Press. Raz, J. (1978), 'Reasons for Action, Decisions and Norms', in J. Raz (ed.), Practical Reasoning, Oxford: Oxford University Press, pp. 128-143. Raz, J. (1982), 'The Claims of Reflective Equilibrium', Inquiry 25,307-330. Raz, J. (1990), 'Facing Diversity: The Case of Epistemic Abstinence', Philosophy & Public Affairs 19, 3-46. Reader, s. (1997), 'Principle Ethics, Particularism and Another Possibility', Philosophy 72, 269-292. Rescher, N. (1987), Ethical Idealism: An Inquiry into the Nature and Function of Ideals, Berkeley: University of California Press. Richardson, H.S. (1990), 'Specifying Norms as a Way to Resolve Concrete Ethical Problems', Philosophy & Public Affairs 19, 279-310. Ross, W.D. (1930), The Right and the Good, Oxford: Clarendon Press. Rutgers, L.J.E. (1993), 'Het weI en wee der dieren: Ethiek en diergeneeskundig handelen' (doct. thesis Utrecht University). Rutgers, L.J.E. and D.B. Baarda (1994), 'Normatieve vragen in de diergeneeskundige beroepsuitoefening: Een verkenning' , Tijdschrift voor Diergeneeskunde 119, 525-535.

Bibliography

273

Rutgers, L.J.E., F.J. Grommers and B. Colenbrander (1996), 'Ethical Aspects of Invasive Reproduction Techniques in Farm Animals', Reproduction in Domestic Animals 31,651-655. Rutgers, L.J.E., F.R. Heeger and FJ. Grommers (1998), 'The Concept of Animal Integrity in Animal Ethics' (unpublished document). 'Rijksoverheid en dierenbeschenning' (1981), Den Haag: Ministerie van Cultuur, Recreatie en Maatschappelijk Werk (Tweede Kamer, 16966-2). Sandel, M.J. (1982), Liberalism and the Limits ofJustice, Cambridge: Cambridge University Press. Sayre-McCord, G. (1985), 'Coherence and Models for Moral Theorizing', Pacific Philosophical Quarterly 66, 170-190. Sayre-McCord, G. (1996), 'Coherentist Epistemology and Moral Theory', in W. Sinnott-Armstrong and M. Timmons (eds.), Moral Knowledge? New Readings in Moral Epistemology, New York/Oxford: Oxford University Press, pp. 137-189. Seeskin, K. (1987), Dialogue and Discovery: A Study in Socratic Method, Albany: State University of New York Press. Selznick, P. (1961), 'Sociology and Natural Law', Natural Law Forum 6, 84108. Sencerz, S. (1986), 'Moral Institutions and Justification in Ethics', Philosophical Studies 50, 77-95. Seung, T.K. (1993), Intuition and Construction, New Haven/London: Yale University Press. Shafer-Landau, R. (1997), 'Moral Rules', Ethics 107,584-611. Shiner, R.A. (1992), Norm and Nature: The Movements of Legal Thought, Oxford: Clarendon Press. Shiner,R.A. andJ.E. Bickenbach (1976), 'Misconceptions about Moral Notions', Analysis 36, 55-67. Singer, M.G. (1986), 'Ethics, Science and Moral Philosophy', in J.P. DeMarco and R.M. Fox (eds.), New Directions in Ethics: The Challenge ofApplied Ethics, New York: Routledge & Kegan Paul, pp. 282-298. Singer, P. (1974), 'Sidgwick and Reflective Equilibrium', The Monist 58, 490517. Smart, N. (1966), 'God, Bliss and Morality', in LT. Ramsey (ed.), Christian Ethics and Contemporary Philosophy, London: S.C.M. Press. St Augustine (1978), 'On the Holy Trinity', in P. Schaff (ed.), A Select Library of the Nicene and Post-Nicene Fathers o/the Christian Church, Grand Rapids, Mich.: Eerdmans. Staffeu, F.R, F.J. Grommers and J. Vorstenbosch (1996), 'Animal Welfare: Evolution and Erosion of a Moral Concept' , Animal Welfare 5, 225-234.

274

Bibliography

Stein, E. (1994), 'Rationality and Reflective Equilibrium', Synthese 99,2,137172. Sunstein, C.R. (1995), 'On Legal Theory and Legal Practice', in I. Shapiro and J. Wagner DeCew (eds.), Theory and Practice (NOMOS XXXVII), New YorkILondon: New York University Press, pp. 267-287. Sutherland, S.R. (1982), 'Religion, Ethics and Action', in B. Hebblethwaite and S.R. Sutherland (eds.), The Philosophical Frontiers of Christian Theology, Cambridge: Cambridge University Press, pp. 153-167. Swanton, Chr. (1991), 'The Role Played by the Method of Wide Reflective Equilibrium in Moral Epistemology', Dialogue 30, 575-589. Swanton, Chr. (1992), Freedom.' A Coherence Theory, Indianapolis/Cambridge: Hackett Publishing Company. Swieringa, J. and A. Wierdsma (1992), Becoming a Learning Organization.' Beyond the Learning Curve, Wokingham: Addison-Wesley. Taekema, S. (1998), 'Vormen van het ideaalbegrip', in W. van der Burg and F.W.A. Brom (eds.), Over idealen.· Het belang van idealen in recht, moraal en politiek, Deventer: W.E.J. Tjeenk Willink (in press). Tannenbaum, J. (1991), 'Ethics and Animal Welfare: The Inextricable Connection', Journal o/the American Veterinary Medical Association 198,1360-1376. Tillich, P. (1948), The Shaking of Foundations, New York: Charles Scribner's Sons. Tillich, P. (1957), Dynamics of Faith, New York: Harper & Row. Timmons, M. (1987), 'Foundationalism and the Structure of Ethical Justification', Ethics 97, 595-609. Toulmin, S.E. (1958), The Uses ofArgument, Cambridge: Cambridge University Press. Toulmin, S.E. (1981), 'The Tyranny of Principles' ,Hastings Center Report, 31-39. Toulmin, S., R. Rieke and A. Janik (1979), An Introduction to Reasoning, New York: Macmillan. Trappenburg, M.J. (1993), Soorten van gelijk.· Medisch-ethische discussies in Nederland, Zwolle: W.E.J. Tjeenk Willink. Van Delden, I.J.M. (1993), Beslissen om niet te reanimeren.· Een medisch en ethisch vraagstuk (doct. thesis Utrecht University), Assen: Van Gorcum. Van den Beld, A. (1997), 'Wat heeft God met de moraal te maken? De goddelijke gebodstheorie van morele verplichting' , Bijdragen, tijdschrift voor filosofie en theologie 58, 362-380. Van den Brink, B. (1997), The Tragedy of Liberalism (doct. thesis Utrecht University), Utrecht: Department of Philosophy.

Bibliography

275

Van der Burg, W. (1989), 'The Myth of Civil Disobedience', Praxis International, 287-304. Van der Burg, W. (1991), Het democratisch perspectief: Een verkenning van de normatieve grondslagen der democratie (doct. thesis Utrecht University), Arnhem: Gouda Quint. Vander Burg, W. (1993), 'De gele oormerken, of: de betekenis van incoherenties in ethiek en recht', Rechtsfilosofie en Rechtstheorie 22, 37-50. Vander Burg, W. (1997), 'The Importance of Ideals' , Journal ofValue Inquiry 31,23-37. Van der Burg, W. (1999), 'The Morality of Aspiration: A Neglected Dimension of Law and Morality', in W.J. Witteveen and W. van der Burg (eds.), Rediscovering Fuller, Amsterdam: Amsterdam University Press (forthcoming). Van der Burg, W. and F.W.A. Brom (eds.) (1998), Over idealen: Het belang van idealen in recht, moraal en politiek, Deventer: W.E.J. Tjeenk Willink. Van Deth, J.W. andE. Scarbrough (1995), The Impact ofValues, Oxford: Oxford University Press. Van Reenen, C.G. and H.J. Blokhuis (1993), 'Investigating Welfare of Dairy Calves Involved in Genetic Modification: Problems and Perspectives', Livestock Production Science 36, 81-90. Van Thiel, G.J.M.W. and J.J.M. van Delden (1997), 'Dealing with Patient Autonomy in Dutch Nursing Homes' , Health Care in lAter Life 3, 177-186. Van Willigenburg, T. (1991), Inside the Ethical Expert: Problem Solving in Applied Ethics (doct. thesis Utrecht University), Kampen: Kok Pharos. Van Willigenburg, T. (1998a), Moreel chauvinisme, (inaugural lecture), Rotterdam: Erasmus University. Van Willigenburg, T. (1998b), 'New Casuistry: What's New?', Philosophical Explorations 1, 152-164. Van Willigenburg, T. and F.R. Heeger (1989), 'Justification of Moral Judgements: A Network Model', in Societas Ethica Jahresbericht 1989, Hannover: Societas Ethica, pp. 53-61. Van Willigenburg, T. and F.R. Heeger (1991), 'Rechtfertigung moralischer Urteile: Ein Netzmodell', Zeitschriftfiir Evangelische Ethik, 35,88-95. Van Willigenburg, T., A. van den Beld, F.R. Heeger and M.P. Verweij (1993), Ethiek in praktijk, Assen: Van Gorcum. Vedder, A.H. (1995), 'The Values of Freedom' (doct. thesis Utrecht University). Vedder, A.H. (1997), 'Privatization, Information Technology and Privacy: Reconsidering the Social Responsibilities of Private Organizations', in G. Moore (ed.), Business Ethics: Principles and Practice, Sunderland: Business Education Publishers, pp. 215-226.

276

Bibliography

Verweij, M.F. (1994), 'De ethicus als amateur-jurist? Over de rol van morele principes in een morele redenering', in W. van der Burg and P.C. Ippel (eds.), De Siamese tweeZing: Recht en moraal in de biomedische praktijk, Assen: Van Gorcum, pp. 71-87. Voorlopige commissie ethische toetsing genetische modificatie van dieren (1992), 'Advies inzake het dossier "Weefselspecifieke expressie van genen in de melkklier van genetisch gemodificeerde mnderen"', Den Haag: Ministerie van Landbouw, Natuurbeheer en Visserij (92/dwtp/f308). Voorlopige commissie ethische toetsing genetische modificatie van dieren (1993), 'Discussienota "Altematieven": De beoordeling van mogelijke alternatieven in het kader van de ethische evaluatie van transgenese bij dieren', Den Haag: Ministerie van Landbouw, Natuurbeheer en Visserij (93/dwtp/f241 ). Walker, B. (1995), 'John Rawls, Mikhail Bakhtin, and the Praxis of Toleration', Political Theory 23, 1, 101-127. Walzer, M. (1983), Spheres ofJustice: A Defence of Pluralism and Equality, New York: Basic Books. Ward, J. N. (1967), The Use of Praying, London: S.C.M. Press. Wedberg, A. (1959), Filosofins historia: Nyara tiden till romantiken, Stockholm: Bonniers. Wellman, C. (1971), Challenge and Response: Justification in Ethics, Carbondale: Southern Illinois University Press. Whiteley, C.H. (1959-60), 'On Defining "Moral"', Analysis 20, 141-144 (reprinted in G. Wallace and A.D.M. Walker (eds.) (1970), The Definition of Morality, London: Methuen, pp. 21-25). Williams, B. (1985), Ethics and the Limits of Philosophy, London: Fontana. Wittgenstein, L. (1966), Lectures and Conversations on Aesthetics, Psychology and Religious Belief (ed. by C. Barrett), Oxford: Blackwell. Wittgenstein, L. (1967), Philosophical Investigatipns (transl. by G.E.M. Anscombe), Oxford: Blackwell. Wolf, S. (1992), 'Morality and Partiality', in J.E. Tomberlin (ed.), Philosophical Perspectives, Vol. 6, Ethics, Atascadero, Calif.: Ridgeview, pp. 243-259.

Name Index

Alcoff, L. 115, 116, 122 Antony, L.M. 116 Applbaum, A.1. 31 Aquinas, Thomas 247 Archard, D. 49 Aristotle 57, 61, 209 Aronovitch, H. 9, 16, 56, 78, 243 Audi, R. 7, 53 Austin, I.L. 59, 63 Baarda, D.B. 237 Babbitt, S.E. 123, 125, 126 Baier, A. 11 Barbour, I.G. 149, 150 Barry,B. 7,9,12,13,180,194,195 Barth, K. 152, 165,247 Bayertz, K. 199 Bayles, M.D. 252 Beauchamp, T.L. 1, 13,29,32,37,38, 221,223,251,252 Beck, U. 190 Benn,S.I.64 Bentham, 1.246,247 Berger, P. 172 Berlin, I, 59 Bertram, Chr. 188 Bickenbach, J.E. 61 Black, M. 149 Blackburn,S. 3, 8, 86, 87 Blokhuis, H.J. 225 Bondolfi, A. 198 Braithwaite, R.B. 147, 148, 150 Brakenhielm, C.R. 22 Brandt, R.B. 3, 8, 9 Brink, D.O. 1 Brom, F.W.A. 15, 16, 24, 97, 98, 190, 219,222,232 Brummer, V. 21,22,148-151,155,171 Buber, M. 248 Carens,I.H. 1 Carr, S. 7

Cherniak, C. 106 Childress, J.F. 1, 13, 29, 32, 37, 38, 221,223,251,252 Chisholm, R. 162 Chomsky, N. 7 Christian, W.A. 152 Cladis, M. 9, 11 Clarke, S.G. 11 Cohen, J. 12 Colenbrander, B. 232, 233 Collste, G. 25, 246 Cupitt, D. 150 Dalferth,I.U. 163 Dancy,J.8,49,51,52 Daniels, N. 1-3,7-11,13, 19,31,55, 56, 73, 78, 80-85, 87, 127-134, 135, 156, 157,246 Danto, A.C. 161 Davidson, D. 107 Davis, R.B. 30 De Beaufort, I. 21 De Boer, I.J.M. 232 DeGrazia, D. 36, 39 Delaney, C.F. 7, 108 Den Hartogh, G. 193 DePaul, M.R. 1,9,10,31,33,78,80, 83,95,103,111-114,115-126, 128, 178, 189, 193, 253 De Rooij, J. 180 Dewey, 1. 108, 203 Dillon, R.S. 119-121, 124, 125 Donogan, A. 243 Duncan, 1.1.H. 237 Dworkin, R. 2,6, 11, 17,32,90,94, 96 Easton, D. 181 Ebertz, R. 1, 7, 15 English,J.6 Feenstra, M. 180 Feinberg, 1. 64

282

Name index

Frankena, W.K. 147 Fraser, D. 237 Frijda, N. 182 Gallie, W.B. 72 Gardavsky, V. 166 Geach, P. 76 Gibbard, A. 3, 11 Glover, J. 13, 17 Goodman, N. 7 Graham, K. 61 Orayling, A.C. 105 Greenawalt, K. 12 Grice, G.R. 2,7 Griffin, J. 134 Grommers, F.l. 232, 233, 236, 237 Guthrie, W.K.C. 213 Gutting, G. 6 Haack, S. 108 Habermas, J. 226 Hampshire, S. 4, 76 Hampton, J. 9 Hanen, M. 1,2,17 Hare, R.M. 3, 29, 30, 35, 45 Harman, G. 43 Harris, l. 243 Hart, H.L.A. 94, 192 Haslett, D.W. 3,8, 17,83, 131, 132, 159-161 Heckmann, G. 203,204,213 Heeger, F.R. 1, 14, 15,30,34,41,53, 55,73,93,95,104-106,166,178, 210,222,226,231-233,236,237, 244,253,255 Hegel, G.W.F. 173, 174 Hepburn, R.W. 151 Herman, B. 46, 47, 52 Herrmann, E. 20, 255 Herron, M.R. 233 Hill, Th. 116 Hobbes, Th. 85 Holmgren, M. 4,8, 157,190 Hudson, W.D. 149-151 Hume, D. 160, 162, 167 Hutcheson, F. 76, 79 Inglehart, R. 171, 172

Jamieson, D. 2 Janik, A. 107, 108 leffner, A. 23, 163, 171 Jennings, B. 200 Jonsen, A.R. 4 Jurgens, E.C.M. 181 Kamm, F.M. 49 Kant, I. 4, 46, 56, 85, 90, 148, 150, 195,246-249,257 Kekes, J. 7 Kenny, A. 78 Kessels, l.P.A.M. 24, 203, 204, 208, 209,211,213 Klop,C.J.194 Klosko, G. 4, 12, 15 Kohl, M. 243 Kohlberg, L. 193 Korsgaard, Chr. 4, 85 Korthagen, F.A. 204 Kovesi, J. 58, 61-63, 65-67, 70 Kuhse, H. 243 Kymlicka, W. 1 Levinas, E. 248 194 Lewis, Little, D. 3 Lucas, J.R. 154 Luckmann, Th. 172 Lyons, D. 1, 3 MacCallum, G. 59 MacIntyre, A. 194, 252 Mackie, J. 86 Majone, G. 189 Marx, K. 172-174 Mavrodes, G. 156 McDowell, J. 52 McFague,S.148-150 McNaughton, D. 51 Meyers, D. 125, 126 Miles, T.R. 150 Mill, J.S. 29,177,195,246 Miller, D. 4, 15 Milligan, B.N. 237 Mitchell, B. 165 Morgenstern, C. 142 Murdoch, I. 150

a.c.

Name index Nelson, L. 203, 204, 209, 213 Nelson, M.T. 7 Newton, I. 149 Nielsen, K. 1, 3, 6, 7, 9, 12, 55-57, 157,255 Noble, C.N. 7 Norman, W. 2-4, 14 Nussbaum, M.C. 57 Otto, R. 248 Pajor, E.A. 237 Parfit, D. 157 Peirce, C.S. 108 Petersson, B. 21 Plantinga, A. 80, 87, 107 Plato 85, 86, 96, 213 Postema, G. 197 Potter, E. 115, 116, 122 Propper, I. 188 Pufendorf, S. 85 Putnam, H. 20, 109 Quine, W.v.o. 7,45,113 Riiikka, J. 2, 9 Ramsey,P.243,247 Rawls,J.l,2,4-13, 16, 19,31-33,55, 56,79-81,84, 89,90,92,94-96, 104,127,156,157-159,178,179, 182,184,190,195,196,199,203, 204,243-245,254,255 Raz, J. 1,9, 11,43,44, 158 Reader, S. 52 Rescher, N. 97 Richardson, H.S. 36, 37 Rieke, R. 107, 108 Ross, W.O. 14,51 Rutgers, LJ.E. 14,25,231-233,237 Sandel, MJ. 9 Sayre-McCord, G. 1,130-1.32 Scarbrough, E. 171 Schroten, E. 24, 201 Scott, W. 74, 75, 78 Seeskin, K. 204 Selznick, P. 96, 98 Sencerz, S. 3,7,8 Seung, T.K. 90, 96 Shafer-Landau, R. 52

283

Shiner, R.A. 17,61 Singer, M.G. 7 Singer, P. 8, 9, 243 Smart, N. 152, 153 Socrates 23, 24, 73, 203-205, 208, 213215 St. Augustine 247 Stafleu, F.R. 236 Stein, E. 3 Sunstein, C.R. 17 Sutherland, S.R. 153 Swanton, Chr. 3, 55,57,59,63,64, 158 Swieringa, J. 206, 214 Taekema, S. 94 Tannenbaum,J.236 Tillich, P. 168, 248 Timmons, M. 3 Toulmin,S.E.4, 107, 108,209,252 Trappenburg, M.J. 180, 186 Van Delden, JJ.M. 15,25,197,255, 256 Van den Beld, A. 19,22,85 Van den Brink, B. 97 Van der Burg, W. 15, 17, 19,29,32, 92,94,95,98,177,199,228,255, 256 Van Deth, J.W. 171 Van Reenen, C.G. 225 Van Thiel, GJ.M.W. 25, 197,256 Van Willigenburg, T. 4,14, 18, 19,29, 30,34,35,41,49,53,93,95,178, 198,210,224,228,231,236,252, 253,255 Vedder, A.H. 18, 19,64,66,70 Verkerk, M. 20 Verweij, M.F. 17,99 Vorstenbosch, J.M.G. 23, 200, 232, 236 Walker, B. 9, 12 Walzer, M. 180 Ward, J.N. 150 Weary, D.M. 237 Weber, M. 171, 173 Wedberg, A. 167

284

Name index

Wellman, C. 2 Whiteley, C.H. 192 Wierdsma, A. 206, 214 Wiggins, D. 86 Williams, B. 11, 12, 129 Witt, Ch. 116 Witteveen, W.J. 188 Wittgenstein, L. 42, 58-61, 63, 72, 149151 Wolf, S. 49

Tabula Gratulatorum

Dr Thomas Achen, Norrkoping - Sweden Dr Keimpe A. Algra, Utrecht - The Netherlands Prof. dr Brenda Almond, Hull - United Kingdom Prof. dr Eberhard Amelung, Icking - Germany Prof. dr Svend Andersen, Arhus - Denmark .Prof. dr Antonio Autiero, Munster - Germany Drs C.M. Bekker, Utrecht - The Netherlands Drs G.C. Bergshoeff, Vianen - The Netherlands Prof. dr J.A. Bergstra, Utrecht - The Netherlands Prof. Goran Bexell, Lund - Sweden Prof. dr Laszl6 Boda, Budapest - Hungary Dr Th.A. Boer, Utrecht - The Netherlands Bishop Dr Gusztav Bolcskei, Debrecen - Hungary Dr Alberto Bondolfi, Zurich - Switzerland Dr h.c. P.H. Bool, Rhoon - The Netherlands Dr J.A.M. Bransen, Utrecht - The Netherlands Prof. Ruth F. Chadwick, Preston - United Kingdom Prof. dr Enrico Chiavacci, Firenze - Italy Commissie Ethiek van de Koninklijke Nederlandse Maatschappij voor Diergeneeskunde, Utrecht - The Netherlands Ass. Prof. Graciela Damilano, Buenos Aires - Argentina Prof. dr Tj. de Cock Buning, Amsterdam - The Netherlands Drs P. de Greeve, Rijswijk - The Netherlands Babette de Kanter-Lover, LL.M., Utrecht - The Netherlands Dr Eef Dekker, Gouda - The Netherlands Prof. dr G.A. den Hartogh, Haarlem - The Netherlands Prof. dr hab. Aniela Dylus, Warszawa - Poland Pfarrer Prof. dr Ulrich Eibach, Bonn - Germany Dr Nienke Endenburg, Hoenderlo - The Netherlands Ersta Institute for Health Care Ethics, Stockholm - Sweden Dr A.M.E. Fermin, Utrecht - The Netherlands Dr Leo Fretz, Amersfoort - The Netherlands Prof. Chaim Gans, Tel Aviv - Israel Prof. dr Volnei Garrafa, Brasilia - Brazil Prof. dr Chr. Gestrich, Berlin - Germany Prof. dr med. Wolfgang Goetze-Claren, Dusseldorf - Germany