Keeping Balance: On Desert and Propriety 3110327422, 9783110327427

Citation preview

Keeping Balance On Desert and Propriety

PRACTICAL PHILOSOPHY Herausgegeben von / Edited by Herlinde Pauer-Studer • Neil Roughley Peter Schaber • Ralf Stoecker Band 10 / Volume 10

The aim of the series is to publish high-quality work that deals with questions in practical philosophy from a broadly analytic perspective. These include questions in meta-ethics, normative ethics and "applied" ethics, as well as in political philosophy, philosophy of law and the philosophy of action. Through the publication of work in both German and English the series aims to facilitate discussion between English- and Germanspeaking practical philosophers.

Diana Abad

Keeping Balance On Desert and Propriety

ontos verlag Frankfurt I Paris I Ebikon I Lancaster I New Brunswick

Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliographie; detailed bibliographic data is available in the Internet at http://dnb.ddb.de

North and South America by Transaction Books Rutgers University Piscataway, NJ 08854-8042 [email protected]

United Kingdom, Ire, Iceland, Turkey, Malta, Portugal by Gazelle Books Services Limited White Cross Mills Hightown LANCASTER, LA1 4XS [email protected]

Livraison pour la France et la Belgique: Librairie Philosophique J.Vrin 6, place de la Sorbonne ; F-75005 PARIS Tel. +33 (0)1 43 54 03 47 ; Fax +33 (0)1 43 54 48 18 www.vrin.fr

2007 ontos verlag P.O. Box 15 41, D-63133 Heusenstamm nr. Frankfurt www.ontosverlag.com

ISBN 13: 978-3-938793-18-3 2007 No part of this book may be reproduced, stored in retrieval systems or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use of the purchaser of the work Printed on acid-free paper ISO-Norm 970-6 This hardcover binding meets the International Library standard Printed in Germany by buch bücher dd ag

Acknowledgments I thank Rüdiger Bittner, Martina Herrmann, Dorothea Kraus, and Ralf Stoecker for their invaluable help and support, Day Petersen for revising the English manuscript, the members of the Kooperative Praktische Philosophie at the Department of Philosophy at the University of Göttingen, where I stayed during the winter semester 2005 / 06, for their instructive comments on the chapter on William Wollaston, as well as the Department of Philosophy at the University of Bielefeld for the opportunity of writing this book there.

Contents Introduction

7

Part 1: Desert I. Feinberg's analysis 1. Basics: "x deserves y in virtue of z" 2. Polar and nonpolar desert objects 3. The preinstitutionality of desert 4. Propriety II. Rawls's ostensible challenge 1. Rawls's argument as most people understand it 2. Rawls's argument as Sandel understands it 3. Rawls's text as Pogge understands it III. Propriety: a preliminary account 1. Fittingness 2. Propriety 3. Desert judgments 4. Conclusion

11 13 13 14 16 19 21 22 24 26 31 33 35 40 44

Part 2: Propriety I. Roderick Milton Chisholm 1. Requirement 2. Wider states of affairs 3. Overriding 4. What we ought to do 5. Supererogation 6. Conclusion II. Samuel Clarke 1. The propriety of things and actions 2. The obligation to act appropriately 3. The happiness of the virtuous 4. Conclusion

45 47 47 49 52 54 55 60 61 63 65 69 71

III. Richard Price 1. Moral rightness a) Rectitude b) Absolute and practical virtue c) Critique of practical virtue d) "Supererogatory" actions 2. The beautiful 3. Desert a) Particular circumstances b) The appropriate 4. Conclusion IV. William Wollaston 1. Kinds of meaningful action 2. The meaning of actions 3. Wollaston's correspondence theory of moral action 4. Lying actors 5. The appropriate treatment of humans by humans 6. The appropriate treatment of humans by God 7. Critique of Wollaston's theory in the literature a) Arguments from circularity b) Joynton's rescue attempt c) Propriety and moral realism 8. Conclusion V. Immanuel Kant 1. Morals a) Morals in the Grundlegung b) Desert in the "Tugendlehre" c) Critique of the "Tugendlehre"-account 2. From morality to happiness 3. The worthiness to be happy and the concept of the highest good a) Worthiness b) The highest good c) Heteronomy and the natural pursuit of happiness 4. Conclusion

75 78 78 80 82 84 85 87 88 90 92 95 96 101 104 108 109 110 112 113 114 116 119 121 122 122 123 126 128 130 130 131 133 135

VI. Propriety and desert 1. The concept of propriety 2. The concept of desert a) Price's account b) Desert bases as criteria of distinction c) Fittingness as desert base d) Desert judgments and propriety judgments e) Conclusion

137 137 139 140 141 143 144 147

Part 3: Applications I. Saving retributivism 1. Hampton's retributivism a) Moral injury b) Retribution 2. Inherent difficulties 3. Retributivism, rightly conceived a) The problem of tautology b) The concept of desert in Hampton's theory c) Conclusion II. Explaining moral residue 1. The persisting Ought a) Feelings b) Giving up the agglomeration principle c) Conclusion 2. Claims a) Hohfeld on absolute claims b) Thomson on non-absolute claims c) Trolleys and transplants d) Claims and moral residue e) Conclusion 3. Desert

149 151 152 154 157 159 161 161 162 165 167 168 169 170 173 175 176 178 180 181 182 184

Bibliography Index of names Index of subjects

189 197 199

Introduction Everybody ought to get what they deserve. For a long time this has been the definition of justice, and in everyday life, the concept of desert is still pervasive. People say, for instance, that a football team deserve to win, that a painting deserves admiration, that a criminal deserves punishment, that a good pupil deserves to get good marks, that every person deserves respect. In ordinary language, there is nobody and nothing that could not deserve something, and nothing that could not be deserved. Talk of desert is not only ubiquitous, though. There is manifestly also a strong normative force connected to it, which is undoubtedly the reason why it was the central concept of justice from ancient times until the 20th century and the emergence of the concept of fairness. Few things move us to such heights of indignation as people not receiving their deserts. The concept of desert seems to entail that something goes singularly wrong if people are denied what they deserve, the whole world seems to be thrown off balance, and we will not rest content until things are put right again and balance is regained. In view of this, it is rather surprising that the concept of desert has never really been analysed, neither as the definiens of justice nor in ordinary language. This is my aim in this book: I want to know what it is to deserve something. I want to know what it takes to keep the world in balance. I will start in Part 1 by discussing the most seminal paper there is on the subject, written by Joel Feinberg. Feinberg does not give us a substantive account of the concept of desert, but he explains the logic of desert, and most importantly, he does give us a clue: he tells us that to deserve something means that it is appropriate to get it. Thus, desert is grounded on propriety, so the key to understanding the one is to understand the other. Yet, the concept of propriety is not any clearer than the concept of desert and we will have to deal with that first. To start with, I will differentiate between propriety and fittingness and try to illustrate these terms by way of metaphors. Two shards of a broken bowl are fitting to each other, but it is not required that they be put together again. By contrast, playing seven

8

notes of a major scale and the eighth note of that scale are not only fitting to each other, but playing the seven notes also requires that the eighth follow, something is wrong if it does not. Hence, propriety is a form of fittingness, but differs from it by an additional normative element. On this basis, I will propose a preliminary account that I think grasps the gist of propriety: propriety involves the requirement that incomplete states of affairs be completed. Obviously, this needs a lot more explaining, since the next question immediately suggests itself: what is that supposed to mean? Part 2 is dedicated to elucidating this notion. Following Feinberg's lead, I will come first to Roderick M. Chisholm, who examines the logic of propriety. From him we learn that one ought to do what is appropriate if that is not overridden. This establishes the important connection of propriety with what ought to be done, but for all that, it does not offer a substantive account of propriety, either. Still, Chisholm gives us a new lead to follow: he points us to 18th-century rationalism, to Samuel Clarke, Richard Price, and William Wollaston, for whom the concept of propriety was the central notion of their moral theories, and eventually to Immanuel Kant. Clarke and Price see propriety as some kind of aptitude or suitability, as in a means being apt to achieve a certain end. This account is not without merit, but it is certainly too strong. Manifestly, not in every case where we talk of propriety do we talk about means and ends. Still, Price's theory in particular is of the utmost importance, if not exactly because of his concept of propriety, as he is the first one ever to give an account of desert that is explicitly grounded on propriety. According to him, desert is that propriety that is based on another propriety, or in short: desert is second-order propriety. I will then turn to Wollaston, who does, at last, develop a substantive and viable account of propriety. For Wollaston, acting appropriately means treating persons and things as they really are. Kant's concept of the worthiness to be happy deals with a special case of desert and propriety, namely the case of deserving happiness in virtue of being moral. It is most illuminating to consider his account, especially since his concept of the highest good offers an answer to the question of why it is even important to us that propriety obtain. In the end, his answer is based on an anthropological fact: we human beings just work that way

9

that we treat things as they really are and that we need to be able to hope for things to be treated that way. In Chapter VI of the second part, I will put Wollaston's concept of propriety and Price's account of desert to the test and ultimately develop my own account of what it is to deserve something. I will argue that we can basically accept Wollaston's account of propriety as it stands, but Price's account of desert needs to be modified, since it does not cover all the cases we want to count as cases of desert. Desert cannot be secondorder propriety, as, although desert itself is always propriety, it is not always based on another propriety. The problem is that propriety is too strong to serve as a desert base in every case, because it is essentially normative. By contrast, there is no such difficulty with fittingness, as I introduce the term in Part 1, and thus it can serve as a desert base instead. Hence, I will offer this account of desert: to deserve something means that it is appropriate that one get it because one is or has done something fitting. Yet, a difficulty immediately arises with this account: if the desert base is being or having done something fitting, there needs to be something that what one is or has done is fitting to. In some cases, the only candidates for that are, for want of a better term, some kind of Platonic Forms. So the account of desert I will offer comes at a high metaphysical price. It presupposes incomplete states of affairs that require something, and it presupposes something like Platonic Forms. I will not endeavour to argue independently for the existence of such entities. Still, if we want to continue talking about desert, and talking about desert presupposes such entities, this is a reason for accepting them. I will give a further reason for accepting the concept of desert and everything it presupposes in Part 3 by discussing two fields of application. The concept of desert can help us solve some problems in moral philosophy where other concepts fail us. By means of the concept of desert, I will first repudiate a fundamental objection against retributive theories of punishment, and show that retributivism works in principle. Thus, the concept of desert is a tenable means, and maybe even an indispensable means, to justify punishment. In the second chapter of Part 3, I will turn to the debate on moral dilemmas and argue that no explanation offered so far succeeds in

10

explaining why there is moral residue. Only by means of the concept of desert can we give a reasonable account of why we ought to apologize or pay compensation, say, in conflict situations. Hence, as expensive as the metaphysical underpinnings of the concept of desert may seem to be, if we take into account the pervasiveness of the concept in ordinary language and the theoretical work it can do which cannot be covered by other normative concepts, simply abandoning it might prove equally expensive. There just might be something, after all, to our feeling that, when people are denied what they deserve, the world is unbalanced.

Part 1: Desert As we will see in greater detail in the first section of this part, desert is a three-place relation, "x deserves y in virtue of z", and the discussion concentrates mainly on the question concerning which individuals may replace the variables. For instance, some say that only persons can deserve something, and only in virtue of actions they are responsible for.1 I shall not pursue this question in this book. Rather, I want to know what it means to deserve something. In other words, the literature is concerned with the "x", "y", and "z" in the relation; I will be concerned with the "deserves". The second question is independent of the first. The question what it means to deserve something is not answered by stating what individuals can replace "x", "y", and "z". In fact, it may very well be that this can be stated only if one understands the meaning of the concept of desert first. For on the face of it, all sorts of individuals can replace the variables, and all sorts of individuals do replace them in actual desert judgments. In Ancient Greece, for instance, people believed that a man could deserve something in virtue of his noble birth. Maybe the above-mentioned champions of desert in virtue of responsible action are right, maybe one cannot deserve anything in virtue of one's noble birth. My point is that the Ancient Greeks may have said something false at worst, but in no event have they said something meaningless or nonsensical. Since their desert judgments must in any case have been meaningful, even though they might have been false, it follows that if they are false indeed, they are so not because the concept of desert does not allow this kind of judgment. It does. If they are false it must be due to other reasons than conceptual ones.2

1

2

Cf. Sadurski, Giving desert its due, p. 117; Rachels, What people deserve, p. 157; Rachels, The elements of moral philosophy, p. 143; Sher, Desert, pp. 37 ff.; for an opposing view, cf. Cupit, Desert and responsibility, pp. 92 ff.; Feldman, Desert: Reconsideration of some received wisdom, pp. 186 f. Lamont follows a similar line of argument, The concept of desert in distributive justice.

12

Thus, I do not think there is anything to be gained by considering the question of which individuals are fit to replace the variables, so instead I will concentrate solely on the different question of what it means to deserve something. Moreover, in contrast to "x", "y", and "z", "deserves" is no variable in the desert relation. Hence, I will assume that all desert judgments, as varied as they may appear due to the replacement of the variables with all sorts of individuals, are based on one and the same concept of desert. There are some who tend to fragment desert, depending on context, into two or more concepts which have little in common with each other, but if there is an analysis of desert as a single, unified concept which covers all the cases, this would be, on grounds of simplicity, a superior analysis and therefore preferable. I shall endeavour to give such an analysis.

I. Feinberg's analysis It is impossible to inquire into the meaning of desert and not take notice of the most important contribution on the subject, Joel Feinberg's "Justice and personal desert" (1963). Feinberg was the first one ever to try to systematically analyse the concept of desert, and to the best of my knowledge, nobody since has participated in the debate who has not referred to this seminal paper.3

1. Basics: "x deserves y in virtue of z" The first thing to be said in answer to the question of what it means that somebody deserves something is that she ought to get it, although not unconditionally, but just pro tanto, as Feinberg puts it. In other words: the fact that somebody deserves something is always a reason for giving it to her, but not always a conclusive reason – something more important could count against it.4 Maybe Sawyer deserves the gold medal in heptathlon, but if she does not actually win she ought not to get it. Probably Feinberg's most important insight is that in order to deserve something one has to fulfil certain conditions he calls desert bases. One cannot deserve something for no reason at all, but only in virtue of something, a property or an action. Moreover, that property or action must be the person's who deserves something in virtue of it – let us call her the desert subject. Sawyer cannot deserve the gold medal in heptathlon for no reason at all, but only because, say, she is the best athlete competing; and she cannot deserve it because of her fans' outstanding support, because that is not a property or action of hers.5 It is important to realize that desert always needs a base, since not all desert judgments explicitly state one. Feinberg makes clear that 3

4 5

Kleinig, The concept of desert, and Sterba, Justice and the concept of desert, have offered two accounts after Feinberg, but they do not really surpass him. Feinberg, Justice and personal desert, p. 60. Cf. ibid., pp. 58 ff.

14

propositions like "You deserved that!" are always elliptical and imply a desert base so that they always have to be supplemented – like this, for instance: "You deserved that because you were naughty!". Thus, Feinberg shows that all desert judgments are judgments of the form "x deserves y in virtue of z" and that they mean that the desert subject x pro tanto ought to get the desert object y in virtue of the desert base z. In our example, Sawyer is the desert subject, the gold medal is the desert object, and Sawyer's being the best athlete competing is the desert base.6 At this point, two questions arise, one about desert objects and one about how desert is related to institutions.

2. Polar and nonpolar desert objects The desert objects Feinberg is concerned with are treatments by persons. Incidentally, the only desert subjects he is concerned with are persons, as well. This is because he wants to examine the connection between desert and justice. Still, this restriction does not mean that things other than persons and treatments cannot be desert subjects and objects. Feinberg does not mean to say that a painting cannot deserve admiration nor a villain to be struck by lightning. He only means to say that the questions whether a painting deserves admiration and whether a villain deserves to be struck by lightning are not questions pertaining to justice.7 The subject matter of my inquiry, however, is not as restricted as Feinberg's. I will assume that anything can be desert subject and object, but that aside, I will stay with Feinberg for a moment and see what else there is to learn from him. The treatments one can deserve, according to Feinberg, are varied, but they have something in common, as well, and that is their being affective: typically, the desert subject experiences them as pleasant or unpleasant. Moreover, Feinberg orders the various desert objects into five classes: first,

6

7

On this point, cf. McLeod, Contemporary interpretations of desert. Introduction, pp. 61 f. Cf. Feinberg, Justice and personal desert, p. 55.

15

awards of prizes; second, assignments of grades; third, rewards and punishments; fourth, praise and blame; and fifth, compensations. It is not important to Feinberg whether or not this list is exhaustive. What is important to him is that the typical desert objects he is concerned with fall into these classes, and that they are not reducible to one another.8 These classes have what Feinberg calls different symmetries: the first two are classes of nonpolar, the other three are classes of polar desert. What exactly does that mean? In respect to polar desert, one can be said to deserve good or to deserve ill – reward or punishment, praise or blame, and so on. [...] Nonpolar desert, on the other hand, has a different sort of symmetry. When it is a prize, an honorable office, or a grade that is in question, we divide persons not into those who deserve good and those who deserve ill, but rather into those who deserve and those who do not.9

Feinberg's point here is that in some cases we speak of somebody deserving good or ill, whereas in other cases this does not make sense and we rather speak of somebody deserving something or not. "Reward and punishment" is a pair of polar notions, of antonyms; the one is good desert, the other ill desert. By contrast, a grade one deserves for a piece of homework does not have an antonym, and that is why it does not make sense to speak of good or ill desert with regards to grades. The only thing that can be said about a grade in this respect is that you either deserve it or you do not.10 This is an instructive illustration of our ways of speaking about desert objects in different cases, but one has to be careful about what exactly is at issue here. Feinberg's differentiation of desert objects into polar and nonpolar is first and foremost a linguistic differentiation, not a substantial one. "Reward" has an antonym, "grade" has not, therefore we can talk differently about each of these desert objects. Nonetheless, any desert object is always either deserved or not. Someone who does not deserve a reward does not for that reason alone 8 9 10

Cf. ibid., pp. 62, 56. Ibid., p. 62. Cf. ibid., pp. 66 f.

16

deserve punishment. Whether or not punishment is deserved is a different question and independent of no reward being deserved.11

3. The preinstitutionality of desert A second question Feinberg treats concerns the connection between desert and institutions. According to him, desert is "logically prior to and independent of public institutions and their rules", or "preinstitutional" for short.12 This marks the fundamental difference between desert and claims, and Feinberg rightly shows that desert cannot be reduced to claims.13 Yet, this is controversial. N. Scott Arnold and David Cummiskey have tried to show that desert can be rightly understood only from within institutional contexts. According to them, desert is dependent on institutions, as both the desert bases as well as the desert objects are determined by the aims of the institutions.14 For instance, the aim of the Olympic Games is to distinguish the best athletes. This aim determines what the desert base is, namely to be the best athlete, and what the desert object is, namely a token of distinction like a medal. Only within the institution is the practice of distinguishing best athletes at all comprehensible; without it, it is unclear why being the best athlete is something so wonderful that one ought to get something for it. In other words: for them, desert is "institutional". This is not altogether far-fetched. It is not self-evident why that person, of all people, who runs a distance of 110 metres, of all distances, and is the fastest to cross ten hurdles along the way, of all obstacles, ought to get a gold medal, of all things. All of this is arbitrary, after all, and therefore it seems plausible to say: this is just the way we set the rules for racing 11 12

13

14

Cf. Sverdlik, The logic of desert, p. 319. Feinberg, Justice and personal desert, p. 87, cf. p. 56; also cf. Galston, Justice and the human good, p. 170; Lucas, On justice, p. 167; Miller, Social justice, p. 91. The term "preinstitutional" is McLeod's, Contemporary interpretations of desert. Introduction, p. 68. Feinberg, Justice and personal desert, pp. 85 ff.; on this point, also cf. Feinberg, The nature and value of rights, and Feinberg, Duties, rights, and claims. Arnold, Why profits are deserved, pp. 390 f.; cf. Cummiskey, Desert and entitlement, p. 18.

17

hurdles at the Olympic Games, and that is the only reason why the fastest runner at hurdles ought to get a gold medal. However, this institutional account of desert faces two problems.15 The first is that it makes desert judgments relative to institutions. This relativity is problematic, because we can easily imagine institutions that give people things they do not deserve, and indeed, institutions are sometimes criticised for not distributing goods according to desert. This criticism would be groundless if desert were institutional. We can imagine a system of "criminal justice" the aim of which is the furtherance of humiliation and in which, appropriate to this aim, rapists are treated to three weeks' vacation in the Caribbean for their deeds. Arnold sees this problem and tries to solve it like this: The objection proceeds too quickly; from the fact that someone deserves something, it does not follow that, all things considered, he ought to get it. [...] desert claims are best thought of as having prima facie significance. An analogy with promise keeping is illuminating: from the fact that A promises B to do x, one cannot conclude that A ought to do x. It is, however, one reason in favor of A's doing x, though that reason may be overridden by other moral considerations. Similarly, desert claims within evil institutions do, on the interpretation offered here, have some moral significance, but such claims are easily trumped by a consideration of the wickedness of the institution in question.16

So, Arnold's point is that within "evil institutions" people do deserve what they are afforded by them; in that particular system of criminal justice, rapists do deserve their vacations. Still, from the fact that they deserve it, we cannot infer that they really ought to be treated to their vacations, because that is outweighed by other considerations and the desert judgment is overridden. If Arnold's opponents' "objection proceeds too quickly", his own justification sets in far too late. It is utterly unconvincing to concede even so much that rapists can deserve vacations at all, and then try to escape the unwelcome consequences of such an account by pointing out that that is 15

16

For the following line of argument, cf. Sher, Desert, p. 15; McLeod, Desert and institutions, pp. 186 f. Arnold, Why profits are deserved, p. 394.

18

not the last word in the matter. Rapists not only ought not to get vacations, they do not even deserve them. Arnold's rather feeble attempt only serves to highlight the fact that institutionalists cannot defend their account against this problem. An institutional account of desert leads to untenable relativism. The second problem with this account is, quite simply, that there are too many cases of desert without an institutional context. Murder is criminal, independent of whether there are institutions that say so, and therefore a murderer deserves punishment, independently of whether there are institutions to mete it out. The respect which some people say human beings deserve qua human beings is, if it is deserved at all, deserved independently of institutions. Just so, the break someone deserves after toiling all day long in the garden is independent of any institution. That shows that desert is preinstitutional, otherwise we could not understand these cases of desert. In fact, we cannot completely understand even those desert judgments which are set in institutional contexts if we consider the insitutional plane only. An expansion of our example of the Olympic Games will show this: in this case, it really is an institution that determines who ought to get what for what. Nevertheless, the Olympic Games themselves did not simply arise out of the blue. They were instituted, and that would hardly have been the case if there was nothing admirable about athletes independent of them, prior to them, so that they ought to get something for it. Why are there institutions like the world championships in cherry stone spitting? Manifestly because some people think it is a splendid achievement to spit cherry stones as far as possible, and they hold excellent spitters in such a high regard that they want to reward them. Arnold and Cummiskey are right insofar as that for some desert judgments institutions are important. Even Feinberg admits that much.17 The point is not to say that institutions are unimportant, the point is that we have to take care to see in what way institutions are important for some desert judgments. In many cases, as in the Olympic Games, the desert objects and desert bases really are determined only by the institutions in the context of which they are set. Nevertheless, the fact that sometimes 17

Cf. e.g. Feinberg, Justice and personal desert, p. 71.

19

desert objects and desert bases are determined by institutions does not imply that the notion of desert itself is dependent on institutions.18

4. Propriety So we learn much from Feinberg about the concept of desert, first and foremost the logic of desert judgments, that is, that desert judgments have the form "x deserves y in virtue of z", and that that means that x ought to get y in virtue of z. Nevertheless, that alone does not suffice for a complete account of what it is to deserve something. The problem is not that we do not know much about the variables "x", "y", and "z", yet. The problem is that, even if we knew which individuals to replace the variables with, we still would not know in what kind of relation exactly they stand. This problem can be illustrated by way of an example: Amber deserves to be punished, because she has murdered Bailey. What does it mean that Amber deserves punishment in virtue of the murder? In what kind of relation exactly do Amber, the murder, and the punishment stand? As we have learned from Feinberg, in the following: in virtue of the murder, Amber ought to be punished. However, we need to go a step further: in what sense "ought" she to be punished? Ought she, because she has some kind of claim to it? Ought she, because it is right that she be punished? Ought she, because she is worthy of it? And if one of these, or something else is applicable, how are we to understand that in its turn? And what exactly does the murder have to do with it? Does it make it true that she ought to be punished? Does it make her worthy of it? How does it do that? Unfortunately, Feinberg's analysis of desert judgments as three-place relations has tempted all other participants in the debate to regard the question pertaining to the replacement of the variables with particular individuals as the only one left open.19 This, however, does not help us at

18 19

Cf. Kleinig, The concept of desert, pp. 84 f., 88. McLeod gives a good survey of the debate, Contemporary interpretations of desert. Introduction.

20

all if we want to know in what kind of relation these stand to each other, that is, if we want to know the very meaning of desert judgments. So, what is the meaning of desert judgments? Feinberg gives us a lead: To say that a person deserves something is to say that there is a certain sort of propriety in his having it.20

David Miller puts the point similarly: In its widest, loosest sense, 'A deserves X' means 'It is fitting for A to have X' or 'X is due to A'. 21

Thus, to deserve something means that it is appropriate or fitting to get it. But what exactly does that mean, it is appropriate? Miller does not answer this question, but Feinberg tries to: To deserve something, one must be qualified in still a third sense: one must satisfy certain conditions of worthiness which are written down in no legal or official regulation.22

Hence, the propriety which lies in desert is a kind of worthiness. However, one concept is as dark as the other one and we would like to know more. Feinberg does give us more: the conditions of worthiness a person has to satisfy in order to deserve something are – the desert bases.23 That is not what we wanted to know. We did not want to know how to tell whether somebody deserves something. We wanted to know what it means to deserve something. Feinberg and Miller see desert as some kind of propriety, but they do not say what they mean by it. I want to say: this is precisely the right lead to follow. The key to understanding desert is to understand the propriety which constitutes it. Pursuing this lead is the purpose of this inquiry. 20 21 22 23

Feinberg, Justice and personal desert, p. 56. Miller, Social justice, p. 83. Feinberg, Justice and personal desert, p. 57. Ibid., p. 58.

II. Rawls's ostensible challenge Feinberg's paper on desert was published in 1963. That raises the question of why, in the more than 40 years since, nobody has ventured to follow the lead I have pointed out in the last section, that desert and propriety are related, and tried to solve this puzzle. The answer to this question is connected to the fact that eight years after Feinberg's paper, John Rawls's A theory of justice (1971) was published. This book occasioned a shift in the debate on justice away from desert and towards fairness. Moreover, many readers understand some of Rawls's arguments to discredit the notion of desert so that since then desert has played almost no role in political and moral philosophy, and presumably, little need was felt to elucidate it. The passages from A theory of justice which primarily gave rise to this understanding were these: It seems to be one of the fixed points of our considered judgments that no one deserves his place in the distribution of native endowments, any more than one deserves one's initial starting place in society. The assertion that a man deserves the superior character that enables him to make the effort to cultivate his abilities is equally problematic; for his character depends in large part upon fortunate family and social circumstances for which he can claim no credit.24 The distributive shares that result do not correlate with moral worth, since the initial endowment of natural assets and the contingencies of their growth and nurture in early life are arbitrary from a moral point of view. The precept which seems intuitively to come closest to rewarding moral desert is that of distribution according to effort, or perhaps better, conscientious effort. Once again, however, it seems clear that the effort a person is willing to make is influenced by his natural abilities and skills and the alternatives open to him. The better endowed are more likely, other things equal, to strive conscientiously, and there seems to be no way to discount for their greater good fortune. The idea of rewarding desert is impracticable.25

24 25

Rawls, A theory of justice, p. 104. Ibid., pp. 311 f.

22

In the first of these two quotations, Rawls says that nobody deserves their native endowments. In the second, he says that our native endowments influence much if not all that we do. He says more than this in the second quotation, but what exactly that is, and what exactly his ostensible argument against desert is, is controversial. Interestingly, this argument is primarily discussed by Rawls's opponents who end up by saying that whatever it is, it does not work. There are three variant readings of this supposed argument.

1. Rawls's argument as most people understand it Most often, Rawls's argument is read like this: we can only deserve something in virtue of a desert base, if we also deserve that deserve base; everything we do depends on our native endowments – ultimately, they are our desert bases; we do not deserve our native endowments; therefore, we cannot ever deserve anything.26 Read like this, the argument is typically rejected as absurd at once. Nozick points out that the requirement that desert bases themselves be deserved is highly implausible; we only have to have them and use them to ground desert.27 Zaitchik spells this out in detail and notes that Rawls's argument as it is usually read is subject to an infinite regress: Thesis 1 [X deserves Y in virtue of having Z only if X deserves to have Z.], together with the truism that all deserving is deserving in virtue of some ground or other, sets up the following argument. Suppose that Jones claims to deserve something in virtue of ground Z. By 1 he must deserve to have Z. He can deserve this only in virtue of some new ground for deserving Z'. But, again by 1, Jones must now deserve to have Z'. This will require yet another ground Z" and so on ad infinitum. Obviously, these conditions on deserving are never satisfied by anyone. Since the argument is perfectly general, we can state its conclusion as 'no one ever deserves anything'.28 26

27 28

Cf. Zaitchik, On deserving to deserve, p. 372; Nozick, Anarchy, state, and utopia, p. 172; Sher, Desert, p. 24. Cf. Nozick, Anarchy, state, and utopia, p. 225. Zaitchik, On deserving to deserve, p. 373.

23

Thus, according to Zaitchik, Rawls can only come to the conclusion that no one ever deserves anything by stating conditions for deserving something that cannot ever be fulfilled; and they cannot ever be fulfilled, because they open an infinite regress, so that in this reading of Rawls's argument there cannot ever be any fundamental desert bases that actually ground desert. First of all, it is worth noting that in this reading of Rawls's argument as well as in Nozick and Zaitchik's replies the term "desert base" is used in a peculiar fashion. This will become clear when we consider again our example of Amber's deserving punishment for having murdered Bailey. The first premise of Rawls's argument in this reading is: we can only deserve something in virtue of a desert base if we also deserve that desert base. Accordingly, Amber would deserve punishment for murdering Bailey only if she had deserved murdering Bailey. I do not think that Rawls meant to say anything like this. However, we can circumvent this difficulty and clear up this reading. The idea behind it is that we cannot deserve anything for something we have no control over or that we cannot ascribe to ourselves; and if everything we do depends on our native endowments, which Rawls tells us we cannot ascribe to ourselves, we do not ever deserve anything. Thus, the cleared-up version of our example would go something like this: Amber deserves to be punished in virtue of having murdered Bailey only if the murder can be ascribed to her. Amber's ability to murder Bailey depends on her native endowments – a certain amount of physical strength to accomplish the deed, say – which she cannot ascribe to herself. Since these cannot be ascribed to her, nothing she does that depends on them can be ascribed to her. Therefore, the murder cannot be ascribed to her. As the murder cannot be ascribed to Amber, she does not deserve to be punished. The trouble is that we can clear up Nozick and Zaitchik's replies in the very same way. This argument is absurd: it is wholly implausible that we cannot ascribe to ourselves anything we do that depends on our native endowments. It is enough, as Nozick rightly says, to have them and to use them.

24

As I have already noted, though, this reading of Rawls's texts is put forward primarily by his opponents who then go on to point out its absurdity. Hence, we may suspect that this is not the most charitable reading possible.

2. Rawls's argument as Sandel understands it The absurdity of Rawls's argument as it is usually read has led Michael Sandel, another of Rawls's opponents, to offer an alternative reading. It is so very self-evident that desert has not to be grounded on a desert base that needs to be itself deserved that Rawls simply cannot mean that.29 He must be out for something different: To say, as Rawls does, that I do not deserve the superior character that led me to realize my abilities is no longer enough. To deny my desert, he must show that I do not have the requisite character, or alternatively, that I have it, but not in the requisite sense. But this is precisely the argument Rawls' theory of the person allows him to make. For given his sharp distinction between the self, taken as the pure subject of possession, and the aims and attributes it possesses, the self is left bare of any substantive feature or characteristic that could qualify as a desert base. [...] On Rawls' theory of the person, the self, strictly speaking, has nothing, nothing at least in the strong, constitutive sense necessary to desert. [...] Rawls can accept that some undeserved desert base is necessary to desert, only to claim that, on an adequate understanding of the person, this condition could never in principle be met! [...] This is what makes [the characteristics] attributes rather than constituents of my person; they are mine rather than me, things I have rather than am.30 On Rawls' conception, no one can properly be said to deserve anything because no one can properly be said to possess anything, at least not in the strong, constitutive sense of possession necessary to the notion of desert.31

29 30 31

Cf. Sandel, Liberalism and the limits of justice, p. 85. Ibid. Ibid.

25

Sandel's point here may best be understood by recalling the cleared-up version of Nozick's reply to the usual reading of Rawls's argument. Nozick's reply was that the requirement that one's native endowments be themselves deserved is implausible; it is enough to have and use them to ground desert. Although Sandel speaks of character instead of native endowments, he means the same thing and his argument starts where Nozick's leaves off: since Nozick is manifestly right, Rawls has to deny that we even have our native endowments at all to get his argument off the ground. At first sight, this seems a very odd idea. If I do not have my native endowments, who does? Sandel's point, though, is that there are two senses in which Rawls can speak of someone "having native endowments", an attributive and a constitutive sense. In the constitutive sense of "have", a person's native endowments comprise that person; they are what we can say that person is. In the attributive sense of "have", things are different. In this sense there is the person on the one hand and the native endowments on the other hand which cling to her as my clothing clings to me. My clothing does not make me up, I just wear it; I am something different from my clothing. Thus, as Sandel understands it, Rawls's argument is that to ground desert people would have to have their native endowments in a constitutive sense; and that this is false, since they have them in an attributive sense only which cannot ground desert. Still, this differentiation between the constitutive and attributive senses of "having native endowments" is precarious. It is not at all clear what is left of a person if her native endowments do not comprise her, but only cling to her as her clothing does; that is, it is not clear what sort of thing a person even is. This is exactly where Sandel's criticism starts. According to his reading, Rawls does have an argument against desert that is not patently absurd, but only at the price of having to presuppose an oddly thin account of personhood which nobody can make anything of and which, therefore, is implausible.

26

Even worse, not only is this account of personhood implausible, it does not sit well with the rest of Rawls's theory either, as both Nozick and Sandel point out: This line of argument can succeed in blocking the introduction of a person's autonomous choices and actions (and their results) only by attributing everything noteworthy about the person completely to certain sorts of 'external' factors. So denigrating a person's autonomy and prime responsibility for his actions is a risky line to take for a theory that otherwise wishes to buttress the dignity and self-respect of autonomous beings; especially for a theory that founds so much (including a theory of the good) upon persons' choices. One doubts that the unexalted picture of human beings Rawls' theory presupposes and rests upon can be made to fit together with the view of human dignity it is designed to lead to and embody. 32

Sandel's reading of Rawls's argument is a little more charitable than the first one, but really only a little. Sandel is perfectly right, of course, the account of personhood presupposed by his reading of Rawls's argument is wholly implausible; but then, it was he who foisted it on Rawls in the first place. Read like this, Rawls's argument is absurd as well, just not as obviously so as in the first reading of it. Therefore, we still do not know what exactly Rawls's argument against desert is.

3. Rawls's text as Pogge understands it Thomas Pogge reconstructs Rawls's text in a way which does not fail because of absurdity and which therefore is the best way.33 According to Pogge, Rawls's critics who offer the discussed readings set in much too late: Rawls is not at all concerned with the question of whether, within a cooperative system, people ought to get what they deserve in virtue of their native endowments according to the rules of that cooperative system – of course they ought to. Rather, he is concerned with the prior question of whether people have a right to the cooperative system 32

33

Nozick, Anarchy, state, and utopia, p. 214; cf. Sandel, Liberalism and the limits of justice, p. 95. Pogge, Realizing Rawls, pp. 63 ff.

27

being instituted in such a way that they will be rewarded for their specific native endowments in it.34 So the question is not whether people can deserve something for being, say, beautiful within the scope of the cooperative system they live in – they can if there are rules pertaining to that –, but rather whether beautiful people can rightfully demand that a cooperative system be established such that there will be rules in it affording them advantages for their beauty. According to Pogge Rawls's answer to this question is: no, they do not have a right to that, because the differences in persons' native endowments have no implications for their moral worth; they are morally arbitrary. Since native endowments are morally arbitrary, it is not appropriate to demand a cooperative system rewarding them.35 Let us leave aside the question of whether or not Rawls is right on this issue. The point is: if Rawls is not even concerned with the question his opponents attribute to him, that is, the question of whether anybody can ever deserve anything, but rather with the question Pogge points out, then the quoted passages of Rawls's cannot be construed as an argument against desert at all. Then, all Rawls is saying is that desert subjects cannot infer further rights from their desert, although they do deserve something. Moreover, Pogge's reading explains how exactly the concept of desert is connected with the concepts of ascription and of a person's moral worth in Rawls, which in the passages quoted at the beginning of this section gave rise to the assumption of an argument of Rawls's against desert. The connection is that a person's moral worth that is not influenced by her native endowments may be influenced in other ways, namely by that person's properties or actions which she can ascribe to herself. Only when there is such a change in a person's moral worth is Rawls inclined to speak of desert: Desert, for Rawls, is moral deservingness, a reflection of one's moral worth in virtue of which alone one can Deserve anything. This may not be the ordinary sense of the term. Rawls suggests as much when he occasionally speaks more fully of moral desert and once, when he has the broader sense in mind, talks of people being 'deserving in the ordinary sense' (TJ 74). Still, there can be no doubt 34 35

Cf. ibid., pp. 63 f., 75. Cf. ibid., pp. 73, 78 f.

28

that the stronger, narrower notion is the one he is using. It follows that many things deserved in the ordinary sense are not Deserved in Rawls's. You may have worked all your life on your autobiography and thus deserve the National Book Award, but you would still not Deserve it in Rawls's sense, unless your efforts somehow testified to your superior moral worth.36 Something is due you in the narrow sense only if you are entitled to it in virtue of your moral Deservingness, that is, in virtue of a moral quality of yourself or a moral quality of something you have done.37

Hence, Rawls himself has a concept of desert, only one that is much narrower than in ordinary language. He just limits speaking of desert to cases with a certain sort of desert base, that is, moral properties and actions of persons that influence those persons' moral worth. To mark that difference between Rawls's use of the term "desert" and our usual one, Pogge writes "Deserve" with a capital D when he means Rawls's use of the verb, and with a lower case d otherwise. Even though Pogge speaks here of different senses of the term "desert" in Rawls, it is clear that there is only the aforementioned difference in acceptable desert bases at work. In some cases, desert objects ought to be got only in virtue of morally relevant desert bases, in others not, and Rawls only speaks of the first sort of case. As far as the concept of desert is concerned, though, Rawls's concept of desert is no different from our usual one. This peculiarity in Rawls's manner of speaking also explains the mistake of those of his followers who think that the concept of desert presupposes that the desert subject be responsible for the desert base.38 Only in some cases, namely exactly in those Rawls limits himself to, are the desert bases such that the desert subjects are responsible for them, in others, they are not like that. However, the notion of desert is the same in all these cases.39 36 37 38 39

Ibid., p. 77. Ibid., p. 78. Cf. Fn. 3 above. Cf. Cupit, Desert and responsibility, pp. 92 ff.; Feldman, Desert: Reconsideration of some received wisdom, pp. 186 f.

29

So the challenge many have seen in Rawls's text really is no challenge after all. He has not offered an argument against desert, he has only been read that way. Rawls does limit his use of the term "desert" to cases with morally relevant desert bases. He does not speak at all about the other cases. That is an oddity in Rawls that is not in accord with our ordinary way of speaking about desert, and it certainly has contributed to Rawls's being misunderstood by so many in this respect. But this oddity is not detrimental in any way to the concept of desert in general. Since at the same time, this serves as an explanation for why the debate on the concept of desert has been virtually non-existent in the past 30 years, the path is free to pick up Feinberg's lead again and to examine the connection between desert and propriety in order to find out what it is to deserve something.

III. Propriety: a preliminary account To deserve something means that it is appropriate to get it. What does that mean? As a preliminary account of the concept of desert I suggest the following: with the desert base the desert subject creates an incomplete state of affairs that requires completion. The completion then consists in the desert subject getting the desert object. Alternatively, we could say that the desert subject creates a state of affairs with the desert base which cannot be left standing, but which calls for a response. But what does that in turn mean, that a state of affairs is incomplete and requires completion? The completion consists of two components, one of which is a relation I shall call fittingness, and the second of which is a normative element that, taken together with fittingness, make up a relation I shall call propriety. Both components can be illustrated by metaphors. The desert object must be fitting to the desert base in the sense of pieces fitting together. A model for this metaphor can be seen in the practice of Ancient Greeks of swapping σύµβολα as tokens of contract. Symbols in this original sense of the word are two halves of an object, like a bowl or a coin, that is broken by the two parties to a contract, where each party keeps one as proof of the contract and to identify the other party.40 An identification is necessary if, say, one party sends a proxy and the other party wants to make sure that he is not an impostor. The identification works like this: by breaking the bowl one gets two shards that fit each other. No shard of another bowl will fit exactly to the original shard so that one can be sure that the owner of the other shard is rightfully a party to the contract. In a way, puzzle pieces may be thought a more catchy example of things that fit each other. The complication with them, however, is that somebody made them such as to fit each other and that is not necessarily so 40

Cf. e. g. Herodot, Historien, Buch VI, 86 β 1, p. 820; How / Wells, A commentary on Herodotus II, p. 99; Liddell / Scott, A Greek English lexicon.

32

with desert. So I will stick to the shards of the bowl as my primary example of fittingness, the exact line of fracture of which is not elaborate in that way. This is a most simple and straightforward concept of fittingness that also lies at the heart of desert. There is this difference, though, that in cases of desert, sometimes there are more than one shard that will fit to the first one. Thus, it is possible that for one desert base there are several fitting desert objects. Yet, this notion alone does not suffice to make desert completely clear. There is a normative element missing. Two shards of a bowl may be rejoined again or they just as well may not. Shards do not require being put together as if that were the state they belonged in. With desert, things are different. It is not the case that one may just as well assign the desert object to the desert subject in virtue of the desert base as not. There is not only fittingness between the desert base and the desert object, but also propriety, that is, something goes wrong if the desert subject does not get the desert object. In contrast, nothing goes wrong if the two shards are not rejoined. Propriety can also be illustrated by a metaphor. Playing seven notes of a major scale is incomplete. That is so because a major scale's seventh note is the leading note that just leads to something, namely the eighth note. So when the eighth note does not follow the leading note leads to nothing, and that is not right. A major scale of which only seven notes are played cannot be left standing like that. This is evidenced by the fact that most listeners can hardly bear to hear it so that they add the eighth note either mentally or by singing it. However, this unbearableness for listeners is just the ratio cognoscendi; the impropriety of the eighth note missing explains the listeners' reaction, not vice versa. The contrast between fittingness and propriety becomes clear by considering the case of playing seven notes of a pure minor scale which works differently. The seventh note of a pure minor scale is not a leading note and that is why one has to take care to notice whether or not one has completed an octave. This is not made apparent by the situation itself, because the situation does not require it in this case. When I practise A minor scales on the piano absent-mindedly – practising scales is not very challenging – I regularly stop not at the note A, but at C. So I turn A minor

33

into C major, because in playing inattentively I automatically take my cue from the leading note that tells me when the scale is completed – it is only that with A minor this is a mistake. Thus, playing seven notes of a pure minor scale is not incomplete nor does it require completion as playing seven notes of a major scale does. It is fitting to play a minor scale's eighth note, but it is not inappropriate not to do so. In short: propriety is fittingness where the fitting counterpart is required. Hence, all propriety is fittingness, but not all fittingness is propriety as well. Incomplete states of affairs do not require just anything, they require something fitting. Nonetheless, it does not follow from something's being fitting alone that it is also required. This normative element of requirement is what distinguishes propriety from fittingness. Accordingly, there are some cases, as the example of the symbols shows, where there is fittingness, but no propriety. At this point, we see why Feinberg's insight that desert necessarily needs a desert base is so important: it makes clear which two things are related by relation of propriety, namely the desert base and the desert object. More examples will elucidate both these relations, fittingness and propriety, still further.

1. Fittingness A different manner of speaking about fittingness is to say that the fitting is a "response" to that state of affairs it is fitting to. One example of this manner of speaking and at the same time of a case where there is fittingness without propriety is Rüdiger Bittner's theory of action. Bittner does not talk of fittingness in general, but of the relation between reasons and actions done for those reasons. As he puts it, this relation is such that the action is a response to the reason obtaining.41 We could also say: this relation is such that the action is fitting to the reason.

41

Bittner, Doing things for reasons, § 118.

34

For the purposes of this inquiry it is not important whether or not Bittner's theory is a good theory of action. Whatever else it may be, it is a splendid illustration of fittingness. Bittner explains what he means when he says that something is a response to a state of affairs by way of examples: Examples abound in games: to your sharp serve I respond with a long return, you draw spades and in response I trump with a jack, and when you leave me, in dominoes, with a three, my response is to get rid of my double three in turn. Examples occur outside games. You insult me, and I respond in kind, or else I respond by walking away silently. You sell me a lemon, and I cut up your tires. The lights turn red, and I step on the brakes. You do me a favor, and I do you a favor in turn. You publish a book in philosophy with big claims and little argument, and I write a scathing review. Simply, you ask me what time it is, and I tell you.42

These are wonderful examples of actions that are fitting to states of affairs or other actions. As Bittner is only concerned with theory of action, he limits himself to this kind of example, but it should be clear that fittingness in general is not limited to action. Its being a fine day is fitting to its being somebody's birthday as well. Bittner examines, too, how these relations of fittingness differ from other relations. In his main example of one particular game of chess the question is: why is moving the pawn a response to the bishop's threat and not a response to the castle's position? The answer to this question is, according to Bittner, that we often describe situations by telling stories about them, stories in which the states of affairs are linked to each other in such a way that some states of affairs are responses to some others, and to some others again they are not. Yet, being a response is not somehow inherent in particular states of affairs, just in the particular situation they are thus that they are one. In a different situation it could be different.43 According to Bittner, it is with the help of these stories we find out what is fitting to what: 42 43

Ibid., § 119. Cf. ibid., § 121.

35

It is an empirical matter. We can only gather from cases which things make up a history, and in particular a history consisting of a state or event and an action that is a response to it. Admittedly, this is to shift the burden of explaining what a response is, hence what a reason is, to the relevant stories. [...] Thus, if you really do not understand the difference between the two relations, that of my pawn move and your bishop's threat, and that of my pawn move and your castle's sitting on a1, then you need some more chess training. [...] Since I shall not run out of relevant stories that might help you to grasp the point, I need not ever be remiss in my explanatory duties, even though I shall not offer you a general criterion of what it is for an action to be a response to something.44

This last phrase is surprising, because we would have thought that there is a general criterion after all: namely, being a response, or fittingness. Of course, this criterion does not by itself explain what state of affairs in particular is a response to what other; this work has to be done, as Bittner shows, by the stories we tell about the states of affairs. Nevertheless, as varied as the states of affairs may be that are linked by all sorts of stories, their theme is always the same: fittingness.

2. Propriety For the purposes of this inquiry, the important form of fittingness is, as I have said, propriety, that is, fittingness where the fitting counterpart is required. The above-mentioned example of playing seven notes of a major scale is a simple but illuminating example of propriety. There is another, much more substantial, example from the sphere of music. Leonard Bernstein once gave a lecture on the first movement of Beethoven's Fifth Symphony, "How a great symphony was written". By means of Beethoven's sketch books, Bernstein examines the development of particular passages, and why Beethoven chose the final version and rejected others. Bernstein's conclusion is that the whole first movement is one direct development from the initial motif, in fact, that it is the correct and the only possible development:

44

Ibid., § 123.

36

[O]ut of [these three G's and an E flat] has grown the first movement of a great symphony, a movement so economical and consistent that almost every bar of it is a direct development of these opening four notes. [...] [The composer] must have the ability to know what the next note has to be to convey a sense of rightness, a sense that whatever note succeeds the last is the only possible note that can happen at that precise instant.45

According to Bernstein, Beethoven rejected the sketches of the first movement he did not work into the final version, because for various reasons they just were not right. We could also say that they were not what the preceding required as coming next, or that they were not appropriate to the whole. For example, the flute was cancelled from the instrumentation, because in comparison with the other instruments it did not sound "masculine" enough and therefore stood out disagreeably; static passages were substituted for, because they hindered the dynamic flow of the music and the development up to the climax; and passages that, taken alone, were "stimulating" had to go, because, compared to the final version, they were less "bold"; and so forth. Bernstein's main point is that it is not our expectation that makes us reject these passages as not as good as the final version just because they are not what we are used to hear. No, it is the nature of music itself.46

Again, for the purposes of this inquiry, it is not important whether Bernstein is right in this respect. In any event, his lecture wonderfully illustrates what propriety is. Roderick Chisholm gives further examples of propriety, and not only from the sphere of music: Requirement may be illustrated by the following: promise-making requires – or calls for – promise-keeping; being virtuous, according to Kant, requires being 45 46

Bernstein, How a great symphony was written. Ibid. This quotation is from the German version of the lecture; my translation.

37

rewarded; the dominant seventh requires the chord of the tonica; one color in the lower left calls for a complementary color in the upper right. [...] 'A curve may demand a certain completion and seem to reject all others; this completion appears as appropriate to the nature of the curve.' [Mandelbaum] [...] Many words ending in 'worthy' or 'able' suggest other types of requirement: e.g., 'praiseworthy', 'blameworthy', 'valuable', 'despicable', 'dubitable'. 'Requires' could be replaced by 'fitting'.47

Chisholm himself does not differentiate between propriety and fittingness, and thus it is important to point out that he also does not take note of the difference in his examples. While most of his cases are certainly cases of propriety, it is at least doubtful whether, say, the Mandelbaum example is one as well. To me, for one, it seems that a certain completion is fitting to a curve rather than required by it. The most impressive examples of propriety, though, or rather of impropriety, can be found in literature, in stories of lovers who, because of extraneous circumstances, most particularly because of malicious scheming, cannot be together. There are countless variations of this motif. One of the best known surely is William Shakespeare's Romeo and Juliet (1597), in which the happiness of the main characters is ruined by their families' animosity that makes it impossible for them to publicly announce their marriage. Juliet is bullied by her family into marrying somebody else so that she sees feigning her own death by means of a narcotic as her only way of escape. Romeo, who coincidentally does not receive Juliet's message revealing her plan, kills himself with poison at his beloved's "tomb". When Juliet awakens only minutes later, she despairs and takes her own life with Romeo's dagger. The main characters in Friedrich von Schiller's Kabale und Liebe (1784), Luise and Ferdinand, fare similarly, as their love is dashed due to their belonging to different social classes. Ferdinand's father plots against them to thwart their relationship. By arresting Luise's parents and threatening to kill them he forces Luise to write a love letter to someone else. He then gives this letter to Ferdinand who is driven to despair by jealousy. For Luise, there is no way out of the situation: if she tells 47

Chisholm, Ethics, p. 147.

38

Ferdinand of his father's scheme, her parents will be killed; if she does not tell him, their love will be frustrated. In the end, both take fatal poison together and dying Luise tells Ferdinand the truth. A relatively new story is especially interesting in this context, Murder most foul (2002) by Lois Cloarec Hart and Carol Paradee, the basic plot of which is similar to Kabale und Liebe. In this story, set in the USA in the 1960s, Patricia and Delia fall in love with each other. In that country at that time, homosexuality is rejected as perverted and regarded as a mental disorder to be "treated" by serious psychiatric operations like lobotomies. Patricia's mother deprecates the girls' relationship and schemes against it with the help of a man she has designated to be Patricia's husband. Just days before Patricia and Delia can elope, Patricia is forced into marriage and into aborting all contact with Delia. Both their lives seem irredeemably ruined. Over forty years later, with Patricia's mother and husband now dead, Patricia and Delia resume contact and fall in love with each other again. Patricia's son, though, fears that his career as a high-ranking politician might suffer repercussions from his mother's lesbianism. Therefore, he kills Delia and so prevents a renewed relationship between them. In a scuffle preceding his arrest he inadvertently kills Patricia as well. This scheming obstruction of Patricia and Delia's plans and hopes, this malicious devastation of their happiness because of others' malignity and interests is inappropriate to their love. Not only is it unfitting that Patricia and Delia have been robbed of the opportunity to live their lives together, it must never have happened. The world Hart and Paradee create in their story, and in which that happens, is not right. The indignation one feels about this course of the story is an indication that something has gone horribly wrong here. The same can be said, of course, of Romeo and Juliet's and of Luise and Ferdinand's lots. The parallels between Murder most foul and Kabale und Liebe are obvious; so why is Murder most foul especially interesting for my purposes? The special thing about Murder most foul is that one of the two authors, Lois Hart, herself had the feeling that the story could not be left as it was. In private correspondence she told me that it had been painful for her to

39

write the story that way, but since the story had developed its own dynamics she was loath to change it within its own scope. For that reason, she wrote a short story, "Lost and found" (2004), which provides an alternative ending to Murder most foul in which Patricia and Delia come together in their old age after all, and which she prefaces as follows: I felt the need to write an alternative ending to Delia and Patricia's story, so, with Carol's permission, I decided to 'play God' and rework their fate.48

In Murder most foul the world is unbalanced, and "Lost and found" is the author's attempt to set it right again, because a world unbalanced is just plain wrong. In his phantastic novel The Eyre Affair (2001), Jasper Fforde similarly plays with two variant endings to one story, one of which is inappropriate and one of which is not. Fforde describes an alternative world which, though very similar to ours in most respects, is very different in others. In an alternative England Charlotte Brontë wrote an alternative Jane Eyre. In this version, Jane comes to work for Edward, they fall in love with each other, and they want to marry. However, the marriage is obstructed by Edward's brother-in-law who discloses that Edward is already married. Jane flees and coincidentally comes to live with as yet unknown relatives. So far, everything is the same as in the book as we know it, but the ending is different: Jane marries her cousin St John and follows him to Calcutta. Literary critics in Fforde's novel agree that, while Jane Eyre is the greatest English literary work, this ending is not in accord with the rest of the book and thus very dissatisfying. Due to many very complicated entanglements in this alternative world, Fforde's heroine uses an abstruse machine, the "Prose Portal", to get into the original manuscript of Jane Eyre where she befriends Edward and somehow manages, after his first wife's death, to bring Jane to go back to him so they can marry after all – just like in the original story. Although Fforde describes some traditionalists who reject this altercation because in the alternative world it is not from Brontë herself, most characters acclaim the "new" ending as finally an appropriate end to a great piece of work. 48

Hart, Lost and found, Disclaimer.

40

In this novel, as in the aforementioned works, a love story does not end appropriately so that the world is thrown off balance, and as in "Lost and found" the author restores the balance, because it is just wrong if Jane and Edward do not come together. It is this that judgments of propriety state in general: they are judgments that the world is unbalanced if an incomplete state of affairs is not completed, and that that is wrong.

3. Desert judgments I have already pointed out how desert judgments are to be understood: with the desert base the desert subject creates an incomplete state of affairs that requires completion which consists in the desert subject's getting the desert object. A detailed examination of a few examples of desert judgments (DJ) will further this understanding. DJ1: Arminia Bielefeld deserve to win. Arminia Bielefeld played well in the last football match. They tackled the opposing team early, passed the ball back and forth well, fought till their last breath, had some very good chances to score, but unfortunately, they did not. The much inferior opposing team, on the other hand, needed only the one counterattack they managed to score. As German sports journalists are wont to say in cases like this one: the opposing team turned events upside down. They also say that the Arminia deserved to win. Grounding this desert judgment on the concept of the fittingness of symbols, this means that, first, there is the fact of the Arminia's good play. This fact is like one shard of a broken bowl. The course of the world will provide a counterpart to that shard, a win, a loss, or a tie. The first counterpart is fitting to the Arminia's shard, the other two counterparts are not. Additionally, a win is not only fitting to the incomplete state of affairs the Arminia created with their good play, it is also required, as the eighth note is required by a major scale's first seven notes. If the Arminia lose, the course of the world is not only unfitting, but something goes wrong as

41

well, like when, after the first seven notes, instead of the next note of the scale the next semitone follows. So, Arminia Bielefeld deserve to win means that the course of the world did not deliver an appropriate counterpart to the Arminia's good play, although that was required. DJ2: The murderer deserves to be punished. Amber murdered Bailey. This is true, there are no reasonable doubts about it, and there is no excuse. Amber deserves to be punished. By murdering Bailey she has created an incomplete state of affairs that calls for completion. In this case, it is not the course of the world that delivers this completion, but typically the state. Moreover, there is not exactly one completion that is fitting to Amber's crime. There are all sorts of completions the state could deliver and more than one could be fitting. The state could ignore Amber's crime and do nothing about it at all; it could confiscate her driver's license; it could react with a suspended sentence, a prison sentence, or capital punishment. It could react in totally different ways as well, like torturing Amber's children, or deporting Amber to a different country, or appointing her as new minister of agriculture. Some of these reactions are clearly unfitting, like appointing Amber as minister or torturing her children; the fittingness of others is dubious, like capital punishment or deportation; and there are several which are fitting, like prison sentences of 14 and 15 years. Thus, Amber deserves to be punished means that by murdering Bailey she has created an incomplete state of affairs that requires one of several possible completions to be delivered by the state. DJ3: The most qualified applicant deserves to get the job. Ryan is the country's best carpenter, but unfortunately, she is currently unemployed. What does she deserve? At first sight, there does not seem to be much of a difference between the most qualified applicant for a job and a murderer, as far as desert judgments are concerned. Both deserve what they deserve in virtue of something they are or have done. Nonetheless, Ryan's case is a bit more complicated than that. The murderer creates an incomplete state of affairs

42

that calls for completion, punishment, that is allotted to her by the state. But what exactly does the most qualified and unemployed carpenter deserve? Let us suppose a job for a carpenter is advertised. Then there is no problem, the most qualified applicant deserves to get this job. What, though, if there is no such advertisement? Then Ryan still deserves a job. However, she does not deserve that somebody in particular create a job for her. In this latter case, this desert judgment is similar to DJ1 above: the most qualified but unemployed carpenter deserves a job, but if there is none available, the course of the world simply did not deliver what is required. DJ4: The most beautiful participant deserves the first place in the beauty contest. Robin has entered a beauty contest. Of all participants, she is the most beautiful. Therefore, Robin deserves the first place in this contest. There is only one difference that is of interest to us here between the most beautiful participant in a beauty contest and a murderer and a carpenter: unlike the latter two, the most beautiful participant creates an incomplete state of affairs by a property she is not responsible for. By being the most beautiful participant Robin creates an incomplete state of affairs that is completed by her winning the contest. DJ5: Whoever has toiled all day in the garden deserves to rest in the evening. Erin spends her free day gardening. All day long she mows the lawn, pulls out weeds, plants a new bed of flowers, trims the hedge, and chaffs. After all this work she is exhausted in the evening. Yet again, by working so hard Erin creates an incomplete state of affairs that calls for completion. Yet again, there is not only one completion that is a fitting counterpart to Erin's work. All sorts of things may be viable possibilities, stretching out on the sofa, praise, an extra nice dinner – it depends on the circumstances and maybe on what Erin wants. It may not be immediately apparent how Erin's work requires a rest after work. After all, we would not necessarily say that Erin does herself

43

wrong or acts contrary to what the situation requires if she does not stretch out on the sofa, but goes to a party instead. But that is only because there are so many possible appropriate completions to Erin's work. If Erin chose an inappropriate completion, like cleaning the whole house after eight hours of hard, uninterrupted yard work, we would tell Erin that something is going wrong here and that she ought to take a break. Thus, Erin's work does require something. It just may not seem like it initially, since it is rather undetermined what exactly it is that Erin's work requires. DJ6: The landscape deserves to be preserved. On the western slope of the Sierra Nevada in California giant sequoias grow, the tallest and some of the oldest living trees on Earth. In 1890 the US government decreed that these woods be preserved and set up Sequoia National Park for this purpose. One can say that these woods deserve their preservation. Obviously, we cannot say that the trees created an incomplete state of affairs calling for completion by doing something, but we can say that they require completion by what they are. We can say that, but just like with Erin's work in DJ5 it may not be immediately evident why anything should be required by a property of trees, presumably their uniqueness. That they do, nonetheless, require something can, as in Erin's case, be best seen when we consider an inappropriate completion. Let us say woodcutters approach intent on cutting down the sequoias. In this case, too, we want to say that this is inappropriate, that this is, as it were, the wrong note after the scale's first seven. So, the sequoias' uniqueness at least requires that people do not destroy them wantonly, and therefore, that they be protected against such destruction and preserved. On the face of it, DJs1 to 6 are very different kinds of desert judgments. They differ from each other in several respects. Sometimes the desert subjects are single persons, sometimes groups of persons, sometimes things. Sometimes they deserve something in virtue of actions, sometimes in virtue of properties, for which sometimes they are responsible, and sometimes not. Sometimes the desert object is assigned by other persons,

44

sometimes whether or not the desert object appears is left to the course of the world. Sometimes the desert object is definitely determined, sometimes we do not know exactly what it is. Sometimes this whole process is institutional, sometimes it is not. As diverse as these cases are with respect to desert subjects, desert objects, and desert bases, this analysis shows that at the root of them all there is always one and the same concept of desert. I will return to these examples at the end of Part 2 when I have fully developed the concept of desert in order to analyse them more accurately still.

4. Conclusion So now we know that desert is propriety. Hence, the key to understanding what it means to deserve something is to understand what it means that it is appropriate to get it. Moreover, we know that the set of propriety relations is a proper subset of the set of fittingness relations. There is a like relation between desert and propriety. Clearly, although deserving something means that it is appropriate that one get it, not in every case in which it is appropriate to get something it is also deserved. Desert is a proper subset of the appropriate. The first step toward understanding the concept of desert will therefore be to understand what propriety is. Once we have managed that, we will next need a criterion of distinction that marks the boundary between the subset of the desert relations and the set of all the propriety relations. I have outlined a preliminary account of propriety in this part: propriety is the required completion of an incomplete state of affairs. Obviously, this is only a provisional attempt. The tasks I shall endeavour to undertake in the next part are an analysis of this concept that will lead to an analysis of the concept of desert, as well as finding a suitable criterion of distinction between propriety and desert.

Part 2: Propriety The important clue we found in the first part was that in order to understand desert we need to understand propriety. In this second part, we will follow this lead, and it points us first to Roderick Chisholm. Chisholm's papers on propriety, or requirement, as he calls it, have a lot in common with Feinberg's. Like him, Chisholm does not give an account of the meaning of propriety, but rather of its logic which helps us understand how the concept of propriety is connected to what we ought to do. Although Chisholm does not get us much closer to the meaning of propriety, he gives us another clue, again just like Feinberg. Following this lead in turn, we will come to Samuel Clarke and thus to 18th-century British Rationalism. For Clarke, as for the other two proponents of this school of thought, Richard Price and William Wollaston, the concept of propriety is the central concept of their moral theories. It is important to note here that all of them speak of fittingness rather than of propriety. They do not differentiate between these two terms as I do. However, as will become clear in the discussion of their works, what they call "fittingness" really is what I call "propriety". For the purposes of the differentiation I propose, it is more expedient to reserve the term "fittingness" for a relation where the fitting counterpart is not required, and otherwise to speak of propriety. The British Rationalists, like Immanuel Kant after them, are concerned with a special case of propriety or rather desert, namely the case of deserving happiness in virtue of being virtuous. With the exception of Price, these authors are solely concerned with the happiness God assigns us in the afterlife. Clarke tries to give an account of the meaning of propriety; he considers it to be a kind of suitability. According to him, to do something appropriate is to choose an apt means to an end. In the special case he deals with, being virtuous is suitable to achieve happiness. Although this gets us a little closer to an understanding of propriety, it still does not quite get us there, as this is certainly too strong. We do not

46

always think of suitability in this sense when we speak of something's being appropriate. Accordingly, questions still remain open in Clarke's theory. The first, and for the purposes of this inquiry more important, question is just how we are to understand propriety if the concept of propriety as suitability is too strong. The second question arises when we consider the special case of desert Clarke is concerned with. Why ought we to get something at all for being virtuous, and even if so, why happiness, of all things? In effect, this comes down to the question of why it is important to us that propriety be observed. Corresponding to these two questions, two new trails start from Clarke that we have to follow. The first of these trails leads from Clarke to Price. Price sees more clearly than Clarke that although propriety has something to do with suitability, it is not identical with it. However, we do not learn more than that from Price as far as that question is concerned. All the same, his contribution to this inquiry is of the utmost importance, as he is the first one ever to explicitly explain desert in terms of propriety, and it is his account of desert that will serve as a basis for my own. Pursuing this lead further, we move from Price to Wollaston for whom propriety is the requirement to treat things as they really are. His contribution is as important as Price's, and I will adopt his concept of propriety. The second trail starting from the discussion of Clarke's theory leads to Kant. His concept of the highest good gives us an answer to the question of why it is important to us that things happen appropriately. Ultimately, it is an anthropological matter: it is just the way we human beings work that we need reasonable hope for propriety being observed. I will expound on these considerations in the following chapters, and on this basis, I will finally give an analysis of the concept of desert in the last chapter of this part: desert is propriety that has fittingness as a desert base.

I. Roderick Milton Chisholm What Feinberg did for the concept of desert in "Justice and personal desert", Roderick Milton Chisholm did for the concept of propriety in two papers, "The ethics of requirement" (1964) and "Practical reason and the logic of requirement" (1974), the first of which was published soon after Feinberg's paper. Chisholm's term is "requirement", but he makes clear that he intends it to be taken as "fitness" or "fittingness" in Clarke's sense of the word, and thus as identical with what I call "propriety".49 Where Feinberg primarily examines basic aspects of the meaning and logic of the concept of desert, Chisholm does exactly the same thing for the concept of propriety. He is chiefly concerned with how the logic of propriety works and how that concept is related to other concepts, most importantly to what we ought to do. Unlike Feinberg, though, Chisholm proceeds in a very formalistic fashion. His account of the concept of propriety is predominantly made up of a series of definitions he proposes and which he explains by means of axioms and theorems.

1. Requirement Chisholm's starting point is the formal sentence "'pRq' – 'p when it obtains requires q'".50 Going on from there, he defines an actual requirement thusly: D1 p requires q = Df p obtains and pRq. In other words, p does require q, or p in fact requires q, when p is such that (i) it obtains and (ii) if it were to obtain it would require q. Thus the defined 'p requires q', unlike the undefined 'pRq', implies that p obtains.51 49 50 51

Chisholm, Ethics, p. 147. Chisholm, Logic, p. 4; Chisholm, Ethics, p. 147. Chisholm, Logic, p. 4.

48

In "Practical reason and the logic of requirement", in contrast to "The ethics of requirement", Chisholm defines each of his terms twice, once as a possibility ("p when it obtains ...") and once as an actuality ("p obtains and ..."). For the purposes of this inquiry the definitions of the first sort are not important. Hence, I will ignore them and only consider those of the second sort. Chisholm puts forth a number of axioms in order to explicate in more detail what he means when he says that p requires q. First of all, there is (A1): p and q are states of affairs. The second axiom (A2) states that if a requirement relation holds, it holds necessarily. These two axioms set the ground rules, as it were. More important for Chisholm's purposes are (A3) which states that if one state of affairs requires another state of affairs, these two states of affairs are logically compatible with each other, and especially (A4): Our fourth principle tells us, on the other hand, that two states of affairs that are logically compatible with each other may yet have requirements that are not logically compatible with each other: A4 (Ex)(Ey)(Ez) {¬N[(x&y) does not obtain] & xRz & yR¬z} There is, therefore, the possibility of a conflict of actual requirements. That is to say, there may occur one state of affairs which requires a certain thing q and another state of affairs which requires not-q.52

I have promised Heather to go to the movies with her tonight. This requires me to go to the movies tonight. On the other hand, I have promised Holly not to go to the movies tonight. This requires me not to go to the movies tonight. Since the state of affairs of my having promised Heather is compatible with the state of affairs of my going to the movies tonight (A3); and the state of affairs of my having promised Holly is compatible with the state of affairs of my not going to the movies tonight (A3); and, additionally, the state of affairs of my having promised Heather is compatible with the state

52

Chisholm, Logic, pp. 5 f.; cf. Chisholm, Ethics, pp. 147 f.

49

of affairs of my having promised Holly; I am under two conflicting requirements (A4). Now what?

2. Wider states of affairs On the basis of his axioms, Chisholm states five theorems, two of which, (T3) and (T4), are especially important for solving the problem of conflicting requirements: T3 pRq ⊃ ¬(pR¬q) T4 (pRq & sR¬q) ⊃ {¬[(p&s)Rq] v ¬[(p&s)R¬q]} [...] No state of affairs would impose incompatible requirements (T3). Thus if, in accordance with the possibility countenanced by (A4),53 two states of affairs that are compatible with each other happen to impose requirements that are incompatible with each other, then at least one of those requirements is not imposed by the conjunction of those two states of affairs (T4). [...] Principle (T4), then, allows for the possibility that, although p is such that if it were to obtain it would require q, there occurs a wider state of affairs that entails p and that does not require q. [...] We can say, then, that there may be requirements that conflict with each other (A4). But if there occurs something p which requires q, then this same thing p does not also require not-q (T3).54

Chisholm here speaks of a "wider state of affairs" which is a "conjunction of two states of affairs". The wider state of affairs in my example is my having promised to go to the movies tonight and my having promised not to go to the movies tonight. The problem is that it is not clear what I am required to do in this situation, to go to the movies or not to go to the movies. (T4) now states that it is possible "that, although p is such that if it were to obtain it would require q, there occurs a wider state of affairs that entails p and that does not require q." J. W. N. Watkins has seen a problem here: 53

54

In the original text it says "(A3)", but this is manifestly a misprint of which there are several in the paper. Chisholm, Logic, pp. 7 f.

50

Professor Chisholm says that his requirement-relation 'is like the relations of logic' (p 5). But which logical relation is it like? Not the relation of logical implication; for that obeys the following principle: if a conclusion q is logically implied by premiss p, then it is logically implied by an augmented premiss p&x, whatever x may be (and even if x is inconsistent with p). I will follow Popper in referring to this as 'the principle of the augmentation of premisses'. It is a conspicuous feature of Chisholm's requirement-relation that it does not obey anything like the principle of the augmentation of premisses: it can easily happen that p requires q but p&r does not require q. [...] Now in Chisholm's 'logic of requirement', so far as I understand it, the principle of the augmentation of premisses is rejected[.]55

Taken out of context, this could be taken simply as a question as to what exactly the requirement relation is like and as a request to Chisholm to explicate this. Yet, the further course of Watkins's criticism shows that he really thinks Chisholm's theory wholly inept so that his line of argument is best understood like this: Chisholm says that the requirement relation is like the relations of logic. The only relation of logic that could even possibly be considered to bear any similarity to requirement is the relation of logical implication. However, requirement cannot be anything like implication, since implication obeys the principle of the augmentation of premises, whereas requirement does not. Therefore, Chisholm can only say that requirement is like the relations of logic on the basis of rejecting the principle of the augmentation of premises, and only on this basis can he formulate (T4) as well. But obviously, it is completely absurd to reject this principle. So much for Watkins. Chisholm answers this challenge directly: [Watkins] then notes correctly that in the sense in which I have understood the concept of requirement, 'it can easily happen that p requires q but p&r does not require q.' And then he draws the conclusion that the requirement relation 'does not obey anything like the principle of the augmentation of premisses'. This conclusion is, of course, correct if we take it to mean that the result of replacing 'logically implied' by 'required' in the above statement of the principle of the augmentation of premisses would be false. But Watkins' conclusion is incorrect if

55

Watkins, Comment: What does Chisholm require of us?, pp. 36 f.

51

one takes it to mean, as evidently he does, that 'in Chisholm's logic of requirement ... the principle of the augmentation of premisses is rejected'. For I am as certain as Watkins is, that the principle of the augmentation of premisses is true.56

Chisholm's reply is typically cryptic, and thus a little difficult to understand. First of all, Chisholm makes quite plain that he does not reject the principle of the augmentation of premises. If he does not, though, what of Watkins's criticism that the requirement relation does not obey this principle? Well, you simply cannot substitute "implication" for "requirement", these are two different relations, the first of which obeys said principle, the second of which does not. It has not been shown that it is a problem if the requirement relation does not obey this principle. But then, Watkins would presumably answer, how can Chisholm go on and say the requirement relation is like the relations of logic? At this point I have to say that Watkins simply has misunderstood Chisholm. This is what Chisholm says about the requirement relation in its full context: Our second general principle tells us that the relation of requirement is like the relations of logic: if it holds between any two states of affairs, then it holds necessarily between those states of affairs.57

The question arising from this statement of Chisholm's that the relation of requirement is like the relations of logic, hence, is not as Watkins would have it "which relation of logic [it is] like",58 but rather in what respect it is like them. Chisholm only meant to say that both kinds of relations have something in common: if they hold, they both hold necessarily. Thus, there is not anything to be said against wider states of affairs as Chisholm construes them.

56 57 58

Chisholm, Reply, p. 48. Chisholm, Logic, p. 5. Watkins, Comment: What does Chisholm require of us, p. 36.

52

3. Overriding Back to the problem: am I required both to go to the movies and not to go to the movies tonight? In such a case (T4) promises help, stating that a wider state of affairs entails at most one of the requirements entailed by the "singular" states of affairs. But how is this decrease in requirement accomplished? By overriding one of the requirements: "if two states of affairs are such that they would impose incompatible requirements, then one or the other of those incompatible requirements would be overridden by the conjunction of the two states of affairs."59 The relevant definition is (D3) – (D2) is cast only for the possible case: D3 The requirement for q that is imposed by p is overridden by s = Df (i) [pRq, (ii) ¬[(p&s)Rq], and (iii) p&s is logically compatible with q, and (iv)] p&s obtains.60

So I am not required both to go to the movies and not to go to the movies, resulting in a dilemma, but rather my promise to Heather and my promise to Holly taken together at most require one or the other – say, to go to the movies. To be precise, the fact that the wider state of affairs of the promise to Heather together with the promise to Holly requires me to go to the movies does not affect the fact that the singular state of affairs of my promise to Holly alone still, even after being overridden, requires me not to go to the movies. This requirement does not vanish in the face of the requirement to go to the movies, it just loses its force. There are two more important things about overriding to take into account: first, a requirement's being overridden may not be the last word in the matter, since "an overriding may itself be overridden". Second, "a requirement for something q may be overridden without being replaced by a requirement for not-q".61 Chisholm illustrates this second point by way of an example: 59 60 61

Chisholm, Logic, p. 8. Chisholm, Logic, p. 9; cf. Chisholm, Ethics, p. 148. Chisholm, Logic, p. 9; cf. Chisholm, Ethics, p. 148.

53

If I am required to go to Boston in order to meet a friend, and then learn that he is dead, the requirement may be overridden without my being required not to go to Boston.62

Joseph Raz complains that Chisholm does not sufficiently differentiate here. Chisholm's example may be acceptable, but there are other examples sufficiently similar to his in which nonetheless we would either not typically speak of a requirement being overridden or in which Chisholm cannot say that even though he should: I promised John to make him a gift of my dog on the first of July. Therefore, I have a reason to do so. On the first of June John released me from my promise and I no longer have a reason to give him my dog. This is not a case of conflict nor does Professor Chisholm claim that it is. Yet it is a case of overriding in his logic. There is, of course, nothing wrong in stretching the concept of overriding to cover cases where no conflict exists. It may, however, be thought somewhat unfortunate that such cases should be drastically separated from others which seem to resemble them. Imagine, e.g., that John did not release me from my promise but that my dog died on the first of June. p&r does not require q, but this is no longer a case of overriding since p&r is incompatible with q. Professor Chisholm's requirement of compatibility in the definition of 'would be overridden' does not seem adequate to its task.63

Although Raz speaks here of "reasons to do something", that is only his translation of "requirement to do something". Chisholm disagrees with this translation, as, while he does think that whenever something is required there are reasons for doing it, there is not always a requirement whenever there are reasons for doing something,64 but for the present purposes that is beside the point. In any event, Chisholm admits Raz's point, and introduces a new definition in order to cope with this difficulty: Let us say that, in such a case, the requirement for q is nullified. This concept might best be defined, in the following way, by appealing to temporalized physical possibility: 62 63 64

Chisholm, Logic, p. 9. Raz, Comment: Reasons, requirements and practical conflicts, p. 31. Chisholm, Reply, p. 44.

54

D10 The requirement for q, that is imposed by p, is nullified at t by s = Df (i) pRq, (ii) s occurs at t, and (iii) it is physically impossible that, if s occur at t, then q occur at t or after t. (We might remind ourselves that 'if John's sitting occurs at t then John's not sitting occurs after t' does not describe a state of affairs that is impossible.)65

So requirements may be overridden, overridings of requirements may themselves be overridden, and requirements can be nullified.

4. What we ought to do Since requirements may be overridden or nullified, the existence of a requirement alone does not imply that one ought to do what is required. Before defining "ought to do" in (D9), Chisholm takes several intermediate steps which, with the exception of the first one, (D4), are not important for the present purposes. With (D4) now, Chisholm defines on the basis of (D3) what it means that something ought to be: D4 It ought to be that q obtains = Df There is an x such that x requires q and it is false that the requirement of q imposed by x is overridden.66

In this definition, Chisholm uses other variable names than in the others, but we should not let that irritate us. In (D4) x requires something, therefore x, like p in the other definitions, stands for a state of affairs (A1). Hence, (D4) states that a state of affairs ought to be if it is required and this requirement is not overridden. On the basis of what ought to be, Chisholm later defines what we ought to do in (D9): We may now define 'the ought to do' by reference to 'the ought to be':

65 66

Ibid., p. 46. Chisholm, Logic, p. 11; cf. Chisholm, Ethics, p. 149.

55

D9 S ought to perform A = Df It ought to be that S's performing A obtains. Having defined the 'ought to be' in terms of requirement, and having characterized requirement in terms of relations holding necessarily among propositions or states of affairs, we may say, with Samuel Clarke, that our duties are a function of the eternal relations of fitness that hold among things.67

The "ought to do" thus is a special case of the "ought to be", since evidently not everything that ought to be ought to be done by somebody. It is possible that there ought to be fine weather – who ought to do that? Taking (D9) and (D4) together, Chisholm can now say that a person S ought to perform an action A if it ought to be that S perform A; or in other words, if S is required to do A and this requirement is not overridden.

5. Supererogation Initially, Chisholm's detour via (D4) may seem unnecessarily complicated without yielding much gain in knowledge. As moral philosophers, are we not much more interested in what we ought to do than in what ought to be? So the one is a special case of the other, what do we learn from that? The worth of this differentiation shows when we consider a special problem. Chisholm's account of requirement raises the question of what place it leaves in his theory for supererogatory actions, that is, actions which are good without being required. Raz, for one, reproves: I agree with Professor Chisholm that whenever one has a duty to do A it is required that one shall do A, or as I prefer to put it, one has a reason for doing A. But Professor Chisholm maintains that the contrary is equally true: whenever one has a reason to do A (it is required that he should do A) he has a duty to do A. This makes it impossible for him to deal with the problem of supererogation.68

Chisholm, however, is able to solve this problem precisely by means of his differentiation between what ought to be (D4) and what one ought to do (D9). Let us say that a state of affairs that is brought about by a given 67 68

Chisholm, Logic, p. 13; cf. Chisholm, Ethics, p. 150. Raz, Comment: Reasons, requirements and practical conflicts, p. 34.

56

action ought to be. That alone does not imply that a particular person is required to bring about this state of affairs. Can't we say, then, that S's bringing it about that p obtains is supererogatory if and only if (i) it ought to be that p obtains but (ii) it is not the case that S ought to bring it about that p obtains? (Perhaps we should add 'and it is not the case that S ought not to bring it about that p obtains'.) Thus if it ought to be that the poor are fed and if I'm not obligated to feed them (and not obligated not to feed them), then if I do feed them, my act is supererogatory.69

So if the state of affairs brought about by my action is required, but it is not required of me, specifically, to perform that action, then my action is supererogatory. Anscombe raises a more complex problem in this context. She considers mercy as an instance of a supererogatory act: We say a punishment may 'fit' a crime. Now mercy can be noble. What facts will make it so? I don't say there aren't any – but mercy at least is not an obligation in the particular case. It is always supererogatory; otherwise it would be justice, not mercy. Professor Chisholm's account leaves no room to this. Mercy would have to be required in an overriding way, to have the right to override the fittingness of punishment.70

This case is more complex than Raz's because, unlike Chisholm's example of feeding the poor, this is a case of overriding. Thus, the question is not only whether Chisholm can deal with supererogatory actions within the scope of his theory, but additionally, how he can explain that sometimes supererogatory actions can override required actions. Chisholm replies directly to this challenge: [T]he example is difficult because it is not entirely clear what the moral principles are that it presupposes. 'Mercy at least is not an obligation in the particular case.' Just how are we to take the situation? We may assume that someone A has performed a sinful act and that A's performance of that sinful act requires that A be punished. We may also assume 69 70

Chisholm, Reply, p. 40; cf. Chisholm, Ethics, p. 153. Anscombe, Comment, p. 21.

57

that there is some other person B who is such that his act of mercy in withholding punishment from A is supererogatory. I would suppose that, if B's act of mercy in withholding punishment is supererogatory, then it is not the case that B ought to punish A and it is also not the case that B ought not to punish A. If my account of supererogation is correct, there is a certain good which B's act brings about but which is such that 'in the particular case' it is not B's obligation to bring it about. What would this good be? One possibility is this: that someone, someone or other, ought to show mercy toward A. Another is this: that now and then B ought to show some mercy, or he ought to show mercy toward some person or other. In any of these cases, we could say that B's act of mercy toward A is supererogatory. What happens, then, to the requirement that A be punished? Perhaps we could say that B's act of mercy overrides it.71

Since, as I have already said, two things are at issue in this case, namely first, Chisholm's account of supererogation, and second, his account of overriding, his answer is rather intricate and needs to be worked out in detail. The problem of supererogation first. B's act is supererogatory if it brings about a state of affairs that ought to be without its being the case that it is required of B specifically to bring it about. Chisholm mentions three candidates for states of affairs that ought to be in this case: first, that someone ought to show mercy toward A; second, that B ought to show mercy now and then; and third, that B ought to show mercy toward some person or other. First of all, I suggest deleting the "ought" in all three possibilities. It is awkward at best to say that it is a state of affairs that B ought to show mercy toward somebody. If at all, the state of affairs in question is that B shows mercy toward somebody. All the same, there is a problem with the last two of these three candidates. How can the state of affairs that B shows mercy now and then or toward somebody or other ought to be, without its being the case that it is required of B specifically to show mercy now and then or toward somebody or other? Chisholm cannot say that what he rather meant was that the state of affairs that B shows mercy now and then or toward somebody or other ought to be, without its being the case that it is required of B specifically to show mercy toward A specifically. He cannot say that, 71

Chisholm, Reply, p. 41.

58

because his definition of a supererogatory action was this: an action of a person bringing about a state of affairs p that ought to be, without its being the case that that person specifically is required to bring that state of affairs p about. Thus, we have to replace both instances of the variable p with the same state of affairs, and not once with the state of affairs that B shows mercy now and then or toward somebody or other, and once with the state of affairs that B shows mercy toward A. This will not do at all, so only the first candidate for a state of affairs brought about by a supererogatory actions remains: it ought to be that somebody shows mercy toward A; it is not required of B specifically to show mercy toward A; therefore it is supererogatory of B if she shows mercy toward A. Consequently, Chisholm can reply to the first of Anscombe's two charges. He can say that mercy is supererogatory according to his account of supererogatory actions, too. However, we have already seen that this case is more complex, and Anscombe brought forth a second charge. There was not only the question of whether mercy was supererogatory according to Chisholm's account, but also the question of how a supererogatory action can take precedence over required actions. Here now, overriding comes into play: "B's act of mercy overrides [the requirement that A be punished]".72 There is yet another problem for Chisholm here, though. Here is a reminder: D3 The requirement for q that is imposed by p is overridden by s = Df (i) [pRq, (ii) ¬[(p&s)Rq], and (iii) p&s is logically compatible with q, and (iv)] p&s obtains.73

So, spelt out in full, what has to be the case for B's act of mercy toward A (s) to override the requirement that A be punished (q) imposed by A's sinful act (p)? First, pRq, A's sinful act requires that A be punished. Let us assume that. Second, ¬[(p&s)Rq], it is not the the case that A's sinful act and B's act of mercy toward A taken together require that A be punished. 72 73

Ibid. Chisholm, Logic, p. 9; cf. Chisholm, Ethics, p. 148.

59

That sounds rather odd already. The third condition, though, settles it at the latest: A's sinful act and B's act of mercy toward A are not logically compatible with A's punishment, since showing mercy toward A just entails not punishing A. In other words: on Chisholm's account of overriding, B's act of mercy toward A cannot override the requirement that A be punished. Does that mean that Anscombe is right in accusing Chisholm that there is no place for supererogation in his theory, and moreover that this is a dilemmatic situation with two conflicting requirements that Chisholm wanted to avoid in the first place? No, because in answering Anscombe's challenge, Chisholm simply chose the wrong "s". We cannot replace that variable with "B's act of mercy toward A", but we can replace it with something else. Maybe A is pregnant, or on her deathbed, or 102 years old, and that is why B ought to show mercy toward her. Then we can say that although A's sinful act requires that A be punished, it is not the case that A's sinful act and A's pregnancy taken together require A's punishment, albeit A's sinful act and A's pregnancy are logically compatible with A's being punished. Indeed, it is not B's act of mercy toward A that overrides the requirement that A be punished, but rather some state of affairs or other concerning A. It is remarkable that Chisholm, who is usually a real stickler as far as his formalisms are concerned, made such a mistake. This mistake of Chisholm's has unnecessarily obscured his response to Anscombe's challenge. If we stick to Anscombe's much more precise wording, we can see that Chisholm can answer her, after all. Her charge was: "[Mercy] is always supererogatory" and "mercy would have to be required in an overriding way".74 Anscombe quite rightly saw that mercy cannot override, but must be itself required, and whatever it is that requires mercy overrides the requirement of punishment. Chisholm can reply to both of Anscombe's charges: mercy understood as incorporating the first candidate for a state of affairs that ought to be is supererogatory; and a state of affairs concerning A can require that A be shown some mercy and this requirement can override the requirement that A be punished. 74

Anscombe, Comment, p. 21.

60

6. Conclusion Chisholm's account is invaluable for understanding the concept of requirement or propriety. The logic he develops shows how we are to speak of requirement, how requirements can conflict, how these conflicts are to be resolved, and how requirement is connected with what we ought to do. I have already pointed out the parallels between Chisholm's achievements regarding the concept of requirement and Feinberg's regarding the concept of desert. There are further parallels. Feinberg explicates the logic of the desert relation, yet he does not give a substantive account of it, and thus does not tell us what it means to deserve something. Chisholm wonderfully elucidates the logic of the requirement relation, but he as well leaves us with the substantial question of what it means that something is appropriate or required. I have promised Heather to go to the movies with her tonight, hence it is appropriate for me to go to the movies with her tonight if that is not overridden, therefore I ought to go to the movies with her tonight. As in Feinberg's desert relation, the question remains open how exactly we are to understand this "ought". What is the connection between the state of affairs of my having promised and the state of affairs of my going to the movies tonight? It is not an oversight on Chisholm's part that this question remains unanswered. He is concerned with logical relations between terms and he does not claim to give substantive definitions of these terms. Nevertheless, within the scope of this inquiry, that is just the question we need to consider next. Thanks to Chisholm, we now know how to talk about propriety correctly. Now we want to know what propriety is. There is yet another parallel between Chisholm and Feinberg: like Feinberg, Chisholm gives us a lead to follow. After his definition of the "ought to do" (D9), he refers to Samuel Clarke. Clarke, and the school of British Rationalists after him, do try to offer a substantive definition of propriety.

II. Samuel Clarke Samuel Clarke (1675-1729) was a high-ranking clergyman who was excellently versed in the natural sciences. Among other things, he was chaplain to Queen Anne and designated as Archbishop of Canterbury. Yet, he was not appointed to that post, since he was accused of arianism after publishing an essay on trinity – that is, the heterodox though widespread belief that Jesus, albeit divine, was created. Moreover, Clarke belonged to the circle around Isaac Newton; he was his neighbour and translator of Newton's "Optics", and so convinced of Newton's theory that he did not simply translate a treatise of the Cartesian physicist Jacques Rohault's, but rebutted it in footnotes on Newtonian grounds. He is especially known for his correspondence with Gottfried Wilhelm Leibniz that was published in 1717.75 Clarke was also Boyle Lecturer twice, in 1704 and 1705. The Boyle Lectures were a series of annual sermons initiated by the physicist Robert Boyle in 1691 with the aim of defending Christian doctrine against unbelievers by grounding it on Newtonian foundations. Since Clarke was an advocate of Christianity as well as of a Newtonian conception of natural science, he was eminently suitable to deliver lectures in this series. The "unbelievers" he opposed in particular were not Jews nor Muslims, but rather voluntarists and, most importantly, Thomas Hobbes and his followers.76 Both the voluntarists and Thomas Hobbes believe that whether something is good or evil is dependent on a will – hence the name: voluntarists hold that things are neutral in themselves, and that it is only the will of God that makes them good or evil, and for Hobbes this power rests with the will of the sovereign.

75

76

Cf. Vailati, A demonstration of the being and attributes of God. Introduction, pp. xi f.; Schneewind, Moral philosophy I, pp. 293 f. Cf. Vailati, A demonstration of the being and attributes of God. Introduction, p. x; Jacob, The Newtonians and the English revolution 1689 - 1720, pp. 32 f., 144, 146, 159; Hunter, Science and society in Restoration England, pp. 183 ff.

62

Clarke, by contrast, is a non-voluntarist, and in his second Boyle Lecture A discourse concerning the unalterable obligations of natural religion, and the truth and certainty of the Christian revelation (1705), he emphatically argues against voluntarism. For him, things are good or evil by nature. If that were not so, it would be pointless to ascribe moral properties to God, and no law would be better than any other, and Clarke rejects both of these views as absurd,77 so absurd, in fact, that he rather rudely berates his opponents: These things are so notoriously plain and self-evident, that nothing but the extremest stupidity of mind, corruption of manners, or perverseness of spirit, can possibly make any man entertain the least doubt concerning them.78

Thus, for Clarke the non-voluntaristic grounds of his sermon are manifest and he goes into them only because he thinks his opponents either too dumb or too depraved to realize them. Accordingly, he presents his own views on the good and evil of things in a very impatient and erratic way without heeding details and connections. For all that, Clarke's main focus is the concept of propriety. This concept yields an instrument for him he can use against his opponents, and he builds his whole moral theory on the basis of this central notion. I have already pointed out that I use terms differently from Clarke and the other rationalists in order to tidy up terminology. What they call "fitness" or "fittingness", by which they mean a requirement in Chisholm's sense, I call "propriety"; whereas I reserve "fittingness" for a similar relation which entails no requirement, though, a relation which the rationalists do not name at all.

77 78

Clarke, Discourse, pp. 609 ff., 627. Ibid., p. 609.

63

1. The propriety of things and actions That there are differences of things; and different relations, respects or proportions, of some things towards others; is as evident and undeniable, as that one magnitude or number, is greater, equal to, or smaller than another. That from these different relations of different things, there necessarily arises an agreement or disagreement of some things with others, or a fitness or unfitness of the application of different things or different relations one to another; is likewise as plain, as that there is any such thing as proportion or disproportion in geometry and arithmetic, or uniformity or difformity in comparing together the respective figures of bodies. Further, that there is a fitness or suitableness of certain circumstances to certain persons, and an unsuitableness of others; founded in the nature of things and the qualifications of persons, antecedent to all positive appointment whatsoever; also that from the different relations of different persons one to another, there necessarily arises a fitness or unfitness of certain manners of behaviour of some persons towards others: is as manifest, as that the properties which flow from the essences of different mathematical figures, have different congruities or incongruities between themselves; or that, in mechanics, certain weights or powers have very different forces, and different effects one upon another, according to their different distances, or different positions and situations in respect of each other.79

So, for Clarke, there are differences between things and between persons, and that means – contra the voluntarists and Hobbes – that things and persons have natures, and that the differences and relations between them are grounded on their different natures. In short, Clarke is a realist. From these differences and relations "fitnesses" arise, and from the way Clarke talks about them here he seems to mean a kind of suitability by that. A "fitness [...] of the application of different things [...] one to another" very much sounds like Clarke means to say that you had better take a knife instead of a cup to cut a loaf of bread, and if there are several knives to choose from, take the sharpest. This applies to persons and their differences as well: that person had best lead the safari who knows about the area and animals, not the tourist who cannot tell a giraffe from a hippo nor a forest from veld. 79

Ibid., p. 608.

64

As far as the different kinds of behaviour that are due to different persons are concerned, though, this reading is tricky. It would be easy to think that propriety in the treatment of people simply means good manners: say, that it is seemly to bow before a king, but not before a menial. Clarke does not explicate this in detail so we do not know his exact meaning here, but this is not the only possible reading. In contrast to good manners, we could think of propriety in treating people like this: we had better give medicine to the ill person rather than the healthy one. This reading is closer to an account of propriety as suitability than the first one, insofar as the medicine operates as it is supposed to in the ill person and not in the healthy one, just like a knife operates as it is supposed to when cutting bread and a cup does not. In a different context, however, Clarke seems to mean something wholly different when he speaks of propriety: [B]ut those things only are truly good in their own nature, which either tend to the universal benefit and welfare of all men, or at least are not destructive of it. The true state therefore of this case, is plainly this. Some things are in their own nature good and reasonable and fit to be done; such as keeping faith, and performing equitable compacts, and the like; [...] Other things are in their own nature absolutely evil; [...] and these cannot by any law or authority whatsoever, be made fit and reasonable, or excusable to be practised. Lastly, other things are in their own nature indifferent; that is [...] such things, whose tendency to the public benefit or disadvantage, is either so small or so remote, or so obscure and involved, that the generality of people are not able of themselves to discern on which side they ought to act: and these things are made obligatory by the authority of laws[.]80

In this passage, Clarke wants to call those things appropriate which are good by nature, whereas before he spoke of things as appropriate that were apt means to certain ends. Still, we need not be concerned about this equivocation of "propriety" as "good" on the one hand and "propriety" as "apt" on the other, since in this passage too he has a kind of suitability in mind: those things are appropriate and good which are suitable to promote the welfare of all men. 80

Ibid., pp. 610 f.

65

In lining up "good and reasonable and fit to be done", Clarke reveals himself as a rationalist. We can see what is appropriate and good, since God has given us reason for that purpose. If we need instruction and divine revelation in morality, this is not because reason was not necessary to this end, but rather because it is fallible.81 So even though Clarke does not make his meaning absolutely clear in all these passages, we may reasonably take him to mean propriety to be a kind of suitability or aptness of means to bring about certain ends, and he additionally marks the special suitability to bring about general welfare as "good". Moreover, this shows exactly how the voluntarists and Hobbes have got it all wrong, according to Clarke, and why it cannot depend on the will of God or the sovereign what is appropriate: whether or not something promotes the general welfare, or whether or not something is an apt means to an end, is in the nature of things and cannot be determined on a whim.

2. The obligation to act appropriately In the course of his sermon, Clarke focuses on those appropriate actions which are also good actions, that is, as we have seen in the last section, on actions that promote the general welfare. This is his focal point, because after all, he chiefly opposes the voluntarists and Hobbes, and he thinks these people have not rightly understood the difference between "good" and "evil". He probably would not go so far as to think them too dumb to see which means are apt for which ends. As far as good actions are concerned, now, Clarke's main point is that we are obligated to perform them: [T]hese eternal and necessary differences of things, make it fit and reasonable for creatures so to act; they cause it to be their duty, or lay an obligation upon them, so to do[.]82

81 82

Ibid., pp. 612, 652 ff. Ibid., p. 608.

66

Clarke's rationalism leads him to this assumption. Since reason discerns what is appropriate, Clarke draws a parallel between morality and theoretical cognition. Given that reason discerns correctly, it is as necessary to act according to the discerned nature of the thing, that is, appropriately, as it is necessary to agree to a true proposition after discerning the demonstration to be correct. For originally and in reality, it is as natural and (morally speaking) necessary, that the will should be determined in every action by the reason of the thing, and the right of the case; as it is natural and (absolutely speaking) necessary, that the understanding should submit to a demonstrated truth. And it is as absurd and blame-worthy, to mistake negligently plain right and wrong, that is, to understand the proportions of things in morality to be what they are not; or wilfully to act contrary to known justice and equity, that is, to will things to be what they are not and cannot be; as it would be absurd and ridiculous for a man in arithmetical matters, ignorantly to believe that twice two is not equal to four; or wilfully and obstinately to contend, against his own clear knowledge, that the whole is not equal to all its parts. The only difference is, that assent to a plain speculative truth, is not in a man's power to withhold; but to act according to the plain right and reason of things, this he may, by the natural liberty of his will, forbear. But the one he ought to do; and it is as much his plain and indispensable duty, as the other he cannot but do, and it is the necessity of his nature to do it.83

Thus, the necessity of theoretical cognition consists in our being unable to disagree with what our reason discerns to be true. By contrast, in morality we are able to act differently from what our reason discerns to be appropriate, and therefore, Clarke takes moral necessity to be only an obligation to act appropriately which manifestly is not a necessity at all. Nevertheless, as Clarke sees this difference, the question arises what exactly he takes the parallel between morality and theoretical cognition to be, if we assume that he means more than simply that in both cases we tend to use the word "necessary". A parallel seems to imply that just as it is necessary to assent to a demonstrated truth, it is necessary to act in a way that is discerned to be good, but now Clarke says that moral necessity is something other than necessity in theoretical cognition. 83

Ibid., p. 613.

67

Schneewind attempts to explain the alleged parallel like this: Moral necessity, then, is best understood as absolute or mathematical necessity in cases where the reasons for conclusions are reasons justifying action, not reasons grounding theoretical truth. [...] To be a rational agent is not only to be able to be moved by reasons; it is to be unable to escape being moved by reasons, at least to some extent. We pursue and must pursue what we think to be good; and if we justify our acts to ourselves by thinking of them as good, then we fall under the further demands of the eternal fitnesses to which, as rational agents, we are able to respond.84

Hence, for Schneewind, it works something like this: as rational agents, we pursue what we think to be good; in doing so we are not only receptive to reasons, we cannot get around them – this is what makes us rational in the first place. Propriety is one of the sources of these reasons: if we deem something to be appropriate, it is a reason for us to do it. Since, according to Schneewind, our rationality as agents consists in our being moved to act by reasons, and propriety gives us reasons to act, we cannot help but be moved to act by propriety, just as we cannot help but be moved to assent to a demonstrated truth. This is a splendid explanation of what Clarke must have in mind. Nonetheless, the problem remains that there does not seem to be a real parallel there between morality and theoretical cognition. The problem is that it just is not the case that we always do what we see we ought to do, and Schneewind himself admits as much. For Schneewind, it is weakness of will that is at the root of this problem.85 We could say that it simply takes so much more to act well than to assent to a true proposition. In this latter case, we only have to see that it is true and already the job is done. Regarding action, though, after we have discerned what is right we still have work to do, like putting our body in motion, and all sorts of things can go wrong here, like passions interfering. Then maybe we could say that Clarke's parallel is not a real parallel, as we do not always get everything right in morality, but if we did, it would be a parallel. 84 85

Schneewind, The invention of autonomy, pp. 317 f. Cf. ibid., pp. 318 f.

68

This cannot be all there is to it, however. Schneewind thinks that weakness of will is the problem, because he subscribes to Clarke's premise that we always pursue what we think to be good. This is not so. There can be a person who correctly discerns what is good and what is evil, and who still, in the face of this knowledge, deliberately chooses evil.86 Thus, the parallel Clarke makes out only works for people for whom "pursue the good" is for action what "aim for the truth" is for theoretical cognition. What Clarke may be out for here is more of a psychological description of people who pursue the good, correctly recognize it, and who are not interfered with in any way. Such people, once they see what is good, do not stop to ask themselves what they ought to do. They see it and that is all that it takes for them to act accordingly. This course of action is not necessary in the theoretical sense. It probably does not even occur very frequently. It is a course of action, though, which we can say works just like theoretical cognition: they see and accept. Since it is neither necessary nor even usual we cannot say that, in general, moral cognition works like theoretical cognition. Yet, we can say that this case is something of a precedent or a guideline of how to do things correctly in morality and which other cases are assessed by, like showing somebody who made a mistake in mathematics how to do it correctly and urging her to proceed accordingly. So Clarke has to say that things may work like this in morality, that this is a parallel to how things work in theoretical cognition, and that this is how things should work in morality. The problem with this, of course, is whence an obligation to do what is good comes. Since such people pursue the good as a matter of course, it makes as little sense to tell them that they ought to do so as it does in the theoretical case to tell people that they ought to assent to a demonstrated truth. They just do. Thus, Clarke would need to work in the demand on a prior level: we ought to be like such people, that is, we ought to pursue the good. The fact that he so naturally makes such people his model case for how things work correctly in morality already shows that he does indeed make that demand. This fact shows that he probably did not even think of the possibility that someone could deliberately choose what he knows to be 86

On this point, cf. Stocker, Desiring the bad, pp. 747 ff.

69

evil, just as Schneewind did not. In fact, it is a rather common idea that all humans pursue the good in any case, and then weakness of will remains the only explanation left for failures in that respect. An indication for this is that professed wicked characters in fairy tales or animated movies always tend to seem slightly ridiculous, like the witch who says that she "delights" in somebody's being ill or that, for her, ugly things are "beautiful". Her values are diametrically opposed to the "real" values, and we see how outlandish an outlook that is just in virtue of the fact that she does not even have a proper vocabulary to articulate her own values. She is simply wrong, in every sense of the word. Clarke and Schneewind evidently share this common idea that nobody could really be like such a witch. It does not even occur to them that "pursue the good" is a demand that has to be explicitly made, and therefore, they treat it as if it referred to something that people do of their own accord anyway, just like people do when aiming for the truth in theoretical cognition. This is probably not true, but in this way we can understand why, for Clarke, there is a parallel between acting morally and theoretical cognition: he thinks it as natural for people to pursue the good as it is for them to aim at the truth. What always happens in theoretical cognition, namely that people assent to demonstrated truths, happens in morality only in the model case in which that happens what is supposed to happen, namely that people do what they discern to be good. This model case serves as a guideline for all the other cases so that we are able to say that all those other cases should work like the model case, and that is why there is no necessity here but only obligation.87

3. The happiness of the virtuous So the nature of things alone directly puts us under a moral obligation to act appropriately. This involves that we are obligated regardless of whether or not the action in question will make us happy.88 87

88

In a different context, Velleman makes a similar point as I did in this section, if with a slightly different argument, The guise of the good, pp. 8, 15 ff. Clarke, Discourse, pp. 608, 641.

70

At this point, two problems arise for Clarke both of which he tries to solve in one go. First, there is the non-voluntarists' notorious difficulty that in their attack against voluntarism they seem to render God superfluous for morality – we are morally obligated by the nature of things alone, after all – so in order to avoid that unpalatable consequence they have to find some important role for him to play in morality.89 Second, although we ought to do what is appropriate regardless of whether that makes us happy, we are such that we need the hope of happiness in order to be able to do what is appropriate in the first place:90 [T]he practice of virtue is often threatened with great calamities, losses, and sometimes even with death itself. And this alters the question, and destroys the practice of that which appears so reasonable in the whole speculation, and introduces a necessity of rewards and punishments. For though virtue is unquestionably worthy to be chosen for its own sake, even without any expectation of reward; yet it does not follow that it is therefore entirely selfsufficient, and able to support a man under all kinds of sufferings, and even death itself, for its sake; without any prospect of future recompense.91

In Clarke, we already find the problem that human beings need the hope that virtue and happiness are compatible with each other that Kant was concerned with later as well.92 Clarke's solution to both problems at once is that God does have an important part to play in morality, since only he can guarantee that the virtuous really do become happy which we must hope for. Interestingly, Clarke does not believe that God was always necessary for that part. Originally, virtue and happiness coincided naturally. Someone eating and drinking in moderation, for instance, was healthier, and thus happier, than somebody who constantly stuffed himself and got drunk.93 Here and now, though, things are quite different due to "some great and general corruption and depravation", including detrimental physiological accidents: 89 90 91 92 93

Cf. Schneewind, Moral philosophy I, p. 294. Cf. Schneewind, The invention of autonomy, p. 322. Clarke, Discourse, p. 629. Cf. section V.3.b) below. Clarke, Discourse, p. 644.

71

[T]he condition of men in the present state is plainly such, that this natural order of things [that the virtuous are happy] in the world is manifestly perverted[.]94

As has already been pointed out, however, we need a connection between being virtuous and being happy, and here God comes into play. Although what is appropriate does not depend on God's will, and although we are obligated to act appropriately even without God's command, nevertheless, because he himself is both rational and moral, he does command us to act appropriately.95 As God commands us to act appropriately, we either heed his command or we do not when acting in a morally relevant manner, even though his command is unnecessary. Thus, there is not only a difference between people who act appropriately and those who do not, but correspondingly a difference between people who obey God and those who disobey him. In turn, God has to react to these people, and for Clarke, that means that he has to treat them according to their obedience or disobedience toward him, as otherwise there would be no proof that he cared about that at all, and this again would be inconsistent with the fact that God is moral which Clarke established in another treatise.96 Therefore, since this life does not work out as it is supposed to, God takes care that in the afterlife the virtuous are granted happiness according to their degree of virtue and the vicious receive their due punishment. Thus, Clarke manages to secure both an important place for God in morality and the requisite hope of happiness for the virtuous.

4. Conclusion The trail we followed from Chisholm to Clarke has led us to a first inkling of what propriety is. Clarke describes the relation between two things which are appropriate to each other as a suitability of the one for the other, as the aptness of a means to achieve an end. This applies to the propriety of persons as well. 94 95 96

Ibid. Ibid., p. 637. Ibid., pp. 642 ff.

72

In the course of his sermon, Clarke distinguishes a special kind of propriety, namely the aptness of actions to promote human welfare. In this special case, he identifies "the appropriate" with "the good". On the basis of this special case, Clarke construes his moral theory: we are rationally and by the nature of things obligated to perform good actions, and if we do God grants us happiness in the afterlife. This is a first approach to an account of the concept of propriety, and it gives us an idea of what kind of relation things are in that are appropriate to each other. However, as many first approaches do, it leaves a lot of questions open. First, describing the propriety relation as a means-end-relation is too strong. It is at least odd to say that the eighth note is "apt" to complete the major scale of which seven notes have been played. On the other hand, Clarke's suggestion is not exactly absurd either. We simply want to know more about it than Clarke tells us. Second, Clarke's inference of a connection between virtue and happiness seems rather ad hoc; his moral theory seems more like a listing of facts: it is the case that certain actions are good; additionally, it is the case that we are obligated to perform good actions; and furthermore, it is the case that God will make us happy according to our virtue. It is not altogether implausible to link these facts to each other, but here, too, Clarke does not make clear how he comes from one to the other. Here, too, we want to know more than Clarke tells us. Corresponding to these two questions left open in Clarke's theory, there are two new leads. One of these leads follows a problem secondary to the purposes of this book. Yet, it will also lead us to an understanding of a basic question that arises within the scope of this inquiry, namely an understanding of why we ought to get anything at all for being virtuous, and why, if we ought to get anything, it should be happiness, of all things. Putting this more generally, the second lead pursues the question of why it is important to us that propriety obtain, and I will return to it in Chapter V on Kant. The more important lead as far as this inquiry is concerned continues to follow the problem of what a substantive account of propriety might be and it leads on to Richard Price and eventually to William Wollaston. Although

73

Price was a follower of Clarke's, he does not restrict himself to simply developing Clarke's ideas. Rather, he surpasses him by far and construes a most refined and original rationalistic moral theory.97

97

Cf. Raphael, Review, p. xv; Thomas, The honest mind, pp. 19 f.

III. Richard Price Richard Price (1723-1791) was, as I have already said, influenced by Samuel Clarke – as well as by Ralph Cudworth and Plato –,whom he had a lot in common with. Like Clarke, he was a clergyman, a non-voluntarist, he suffered reprisals for his beliefs, and he was educated in the sciences, most importantly in mathematics. Unlike Clarke, Price was politically active. Price was Welsh and a Dissenter. Dissenters rejected the orthodox doctrine of trininty and the Anglican rites of sacrament. For these reasons, they were banned from attending university at Oxford, for instance, and even though they could study at Cambridge, they were not allowed to graduate. They were also barred from holding public offices, something which the City of London used to extort money from them. The City appointed Dissenters as candidates for public offices, thus putting them in a dilemma: they must not accept the election, since that would have involved accepting the Anglican rites of sacrament, which they rejected; they must not not accept the election, since it was forbidden not to take an office one was elected to; or, they had to go against their consciences, accept the rites of sacrament, and thus renounce their own beliefs. Most of them rejected the offices and were forced to pay a fine to the City.98 It was this kind of persecution that motivated Price to become involved in politics and to publish numerous pamphlets demanding freedom of conscience and worship. With his demands that humans be allowed to follow their own consciences freely and be responsible for themselves, he openly supported the American War of Independence and the French Revolution, as well. This earned him the respect of the French and even an invitation from the Americans to become a citizen of the United States, but it also earned him the enmity of Edmund Burke, who emphatically opposed Price in these matters.99

98 99

Thomas, The honest mind, pp. 13 f.; Thomas, Political Writings. Introduction, p. ix. Raphael, The moral sense, pp. 99 f.; cf. Schneewind, Moral philosophy II, p. 586.

76

Additionally, Price was very well-educated in fiscal matters and mathematics. He concerned himself with the British national debt and with probability calculus, particularly with the work of Thomas Bayes. Furthermore, the invention of life insurance is attributed to him.100 In this chapter, though, we will be concerned with the book Price has written on moral philosophy, A review of the principal questions in morals (1758, 2nd edition 1769, 3rd edition 1787). Like Clarke's lectures, it is relatively unknown, but those who do know it cannot praise it highly enough. Schneewind and Raphael, for instance, write about it: Price's Review of the Principal Questions in Morals [...] is one of the most remarkable of British eighteenth-century treatises on ethics, bringing to some of the major controversies of the time a new degree of sophistication and insight.101 Price's contribution to moral philosophy consists of a single, comparatively small book, but this one volume is probably worth more than any of the more weighty tomes produced by most of his contemporaries; of them perhaps none but Hume and Butler challenge serious comparison with Price as a thinker on moral subjects. Rashdall says of him: 'His Review of the principal Questions and Difficulties in Morals I regard as the best work published on ethics till quite recent times. It contains the gist of the Kantian doctrine without Kant's confusions.'102

Price's primary aim in his Review is to construe a moral theory in which morality is not arbitrary, but fixed for all rational beings at all times. Hence, he opposes both voluntarists and sentimentalists, particularly Francis Hutcheson. He thinks it absurd that it should be impossible to apply moral standards to God – which would be the case if what is morally good and evil was only dependent on his will, like the voluntarists assume, and from which would follow that we could not say of God that he was good –, or that we might just as well consider other things to be morally good and evil than we do if we just felt differently than we do – which 100

Thomas, Political Writings. Introduction, pp. xiii ff.; Raphael, The moral sense, p. 99. 101 Schneewind, Moral philosophy II, p. 586. 102 Raphael, The moral sense, p. 101. Rashdall refers to Price's book by the title of the first two editions; in the third "and difficulties" is deleted.

77

would be the case if what is morally good and evil was only dependent on our psychological constitution, like the sentimentalists assume.103 This concern leads Price to have a rationalistic and realistic view of morality. Actions really have moral properties. And since actions have their moral properties necessarily, experience cannot show us what is morally right. Rather, we discern this by reason, that faculty of discernment empiricists have simply overlooked as a source of new simple ideas. For Price, there are three different kinds of perception regarding moral actions: In considering the actions of moral agents, we shall find in ourselves three different perceptions concerning them, which are necessary to be carefully distinguished: The first, is our perception of right and wrong. The second, is our perception of beauty and deformity. The third we express, when we say, that actions are of good or ill desert.104

It may seem surprising that Price speaks of "perceptions" in this context, as this typically belongs to the empiricists' vocabulary, and Price is not an empiricist but a rationalist. What does he mean by saying that reason perceives? Price's terminology here and elsewhere is irritating and not always adequate to his own position. Later critics fault him, for instance, for simply taking over Locke's differentiation between simple and complex ideas, and for considering his idea of the morally right as a simple idea, whereas that does not come up to what he really has in mind.105 This is correct, but Price just uses the vocabulary of his time, the only vocabulary there is for him, and that is empiricistically coined. Thus, I suggest not attaching too much importance to it. What he means by "perceptions" is just that, when we consider moral actions, several different things strike us, namely whether they are right, beautiful, and deserving. These three perceptions are not independent of 103

Cf. Thomas, The honest mind, p. 20. Price, Review, p. 13. 105 Cf. Raphael, The moral sense, pp. 127 f.; Thomas, The honest mind, pp. 46 ff.; Hudson, Reason and right, p. 19. 104

78

each other: whenever an action is morally right, it is also beautiful and deserving – these three properties always come together. Still, they are not reducible to one another, either, but rather, they are different aspects of the action: reason discerns whether it is right; feelings tell us whether it is beautiful; and that it is deserving means that it stands in a certain relation to other actions and maybe states of affairs, which is also discerned by reason. This indicates that we indeed have different faculties of discernment for the different aspects of an action, that is, reason for rightness and deservingness, and feeling for beauty. Nonetheless, Price is a rationalist, since an action's beauty, as we shall see, is only a secondary aspect for him, while its rightness and deservingness are central.

1. Moral rightness So, what does it mean that an action is morally right? This is not at all easy to say, since, according to Price, moral rightness is a simple idea which cannot itself be defined, but rather for which we can only give synonyms.106 The synonym Price uses most often for moral rightness is propriety. I shall return to that later.107 Nonetheless, there are still several things to say about morally right actions.

a) Rectitude For Price, that an action is morally right first of all means that one is obligated to perform it. Obligation to action, and rightness of action, are plainly coincident and identical; so far so, that we cannot form a notion of the one, without taking in the latter.108

106

Price, Review, p. 41. Cf. section 3.b) below. 108 Price, Review, p. 105. 107

79

Hence, if reason discerns which action is the right one, it thus also discerns what one is obligated to do. When reason discerns this correctly, Price calls this rectitude, and rectitude is both a law and a rule for us: From the account given of obligation, it follows that rectitude is a law, as well as a rule to us; that it not only directs, but binds all, as far as it is perceived.109

At this point, Price does not make the difference is between rule and law sufficiently clear. Rectitude serves as a rule of action insofar as it gives us information of what would be right to do and thus points us that way. It would be easy to think, reading the quotation, that it is the rule's task to point us toward the right action, and the law's task to obligate us to that action. However, when explicating this further in the next paragraph, Price surprises us by saying that it is the rule of action that obligates us, as well: "it is such a direction as implies authority"; "Reason is the guide, the natural and authoritative guide of a rational being"; "here he is tied in the most strict and absolute manner".110 Maybe we should not view rule and law as disparate things, but rather as something like two sides of a medal, something which belongs together and thus should not be considered each by itself. Yet, there is another paragraph concerning rectitude as a law of action, and here Price characterizes the law quite differently: That is properly a law to us, which we always and unavoidably feel and own ourselves obliged to obey; and which, as we obey or disobey it, is attended with the immediate sanctions of inward triumph and self-applause, or of inward shame and self-reproach, together with the secret apprehensions of the favour or displeasure of a superior righteous power, and the anticipations of future rewards, and punishments.111

Here, the law is connected to consequences, with rewards and punishments in the future – that is, in the afterlife. Price's point here, though, cannot be 109

Ibid., p. 108. Ibid., p. 109. 111 Ibid. 110

80

that these consequences are a reason to follow the law. It is true that, according to Price, certain actions ought to have certain consequences – doing what is morally right is deserving, after all, as we will see –, but an action that is performed in expectation of these consequences is therefore precisely not morally right any longer.112 I shall return to this problem in the next section. So, these consequences only accompany morally right actions, or ought to accompany them, but they do not have anything more than that to do with them. Thus, I suggest not separating law and rule of action strictly as the two mentioned paragraphs and their different characterizations indicate, but rather treat them as one and the same thing. So, rectitude discerns which action is the morally right one, points us that way, obligates us to it, and what is more, we are by this discernment alone directly motivated to do it.113

b) Absolute and practical virtue Price's account of rectitude faces serious problems, though, even if we put aside the problem of direct motivation by discerning what is morally right. Most importantly, as Price himself sees, we are not always right in moral matters. He makes out several causes for this, but for him, the main problem is that in order to be perfectly moral we would have to have perfect knowledge and perfect reason, and we simply do not.114 In view of this problem, he differentiates between two kinds of virtue, absolute virtue and practical virtue. Absolute virtue is determined by the actions; certain actions are "in themselves", "materially", "objectively" right, and performing these is absolutely virtuous.115 The problem here is that we cannot know which actions these are. We probably manage to be absolutely virtuous from time to time, but since that is not in consequence of correct discernment, it only 112

Cf. Thomas, The honest mind, p. 108. Price, Review, pp. 185 ff. 114 Ibid., pp. 166 ff. 115 Ibid., p. 177. 113

81

amounts to lucky strikes. Practical virtue, on the other hand, is relative to the agent; here, it depends on an agent's beliefs whether his actions are "formally" right.116 To put it differently, an agent is practically virtuous if she follows her conscience.117 Manifestly, this cannot mean that the agent ought to do simply what she is inclined to do. She has a duty to inform herself and assess the situation as well as she can. Thus, in contrast to absolute virtue, practical virtue is agent-relative. It still is "real" virtue, though, according to Price. To achieve absolute virtue is simply impossible for agents, and therefore we cannot demand from them any more than their being as practically virtuous as possible.118 Interestingly, whether an agent is to be called virtuous or not depends solely on her practical virtue, and not at all on her absolute virtue.119 This is not because we can only be absolutely virtuous by coincidence at best, but rather, because absolute virtue is compatible with bad character. In other words, it is possible to perfom an objectively right action from morally indifferent or evil motives, and therefore, it cannot depend solely on an action's absolute rightness whether or not the agent is to be called virtuous. Additionally, she must act "from a consciousness of rectitude", that is, she must be practically virtuous.120 This also indicates why an agent must not act in expectation of future consequences as the passage on the law of action suggested: she must do the right thing, not because it will secure her a place in heaven, but rather, because it is the right thing. Moreover, Price sees practical virtue as a requirement of personal integrity.121 The idea here is that something is wrong with someone who firmly believes that a certain action is morally right, and that it is within her power to perform it, and yet does not perform it, even when performing that action would have been objectively wrong.122

116

Ibid. Ibid., p. 179. 118 Ibid., p. 180. 119 Ibid., p. 179. 120 Ibid., p. 184. 121 Cf. ibid., p. 180. 122 Cf. Broome, Normative requirements, for a modern treatise of this issue. 117

82

c) Critique of practical virtue In view of the reprisals Dissenters had to suffer in Great Britain mentioned at the beginning of this chapter, and of Price's resulting commitment to freedom of conscience, it is hardly surprising that he gives the individual's conscience, in the form of practical virtue, a prominent role to play in his moral theory. However, there is a serious problem with this, in that differentiating as Price does between absolute and practical virtue, and making an agent's virtuousness solely dependent on his practical virtue, undermines his own moral realism and rationalism. Because the only important thing the agent needs to do in order to act rightly is to inform herself about the situation and assess it carefully, neither the true moral properties of the situation and the action nor reason, the task of which is discerning these properties correctly, are all that crucial anymore. If truth is not important, we can get by with experience, since experience enables us to assess situations very carefully just as well. As I said, it is not at all surprising that Price differentiates between absolute and practical virtue and that he focusses on the latter. Within the scope of his moral theory, though, this differentiation is unnecessary, and in fact, it is better off without. Of course, Price is quite correct in saying that we would need perfect reason in order to discern infallibly what is morally right in each and every situation. Nonetheless, although we are not infallible in discerning what the right thing is, it does not follow that we ought not to aim for it. Since infallibility in discerning what is right is admittedly impossible, Price hits upon the idea of demanding rather little from agents: they ought only to inform themselves well and act accordingly. Behind this is a skepticism about morality that does not fit his rationalism and, furthermore, that is not justified. Even though we are not infallible in discerning the right, we are quite good enough to aim for it with fairly good chances of success, and thus we ought to aim for it, too. Thus, the first argument Price offers for focussing on practical virtue does not work: of course, we can always doubt whether we really succeeded in getting things right, and we must not be paralyzed by such

83

doubts, since otherwise we would be unable to act at all. On the other hand, such doubts must not lead to morality's becoming so very undemanding that we are virtually faultless as long as we only follow our well-informed consciences. From the possibility of doubt, it does not follow that we ought not to aim for what there is doubt about: for the right. The second argument Price offers for focussing on practical virtue is that absolute virtue is compatible with bad character or evil motives. Although this is quite true, it is no reason to attach importance to practical virtue only, all the more so as it is not at all clear how focussing on practical virtue can produce a remedy for this problem. Let us recall that practical virtue involves informing oneself about the situation and then following one's conscience. Yet, this account of practical virtue does not rule out either bad character or evil motives. Such an account of practical virtue allows for practically virtuous murderers. Because Price focuses solely on practical virtue, he cannot even call in absolute virtue as a corrective and say that it is absolutely wrong to murder somebody, after all. An individual's conscience neither reliably leads to right actions nor is it indicative of good character. So, practical virtue neither fits into Price's rationalistic and realistic background, nor can it serve, as Price hopes, as an adequate standard for moral action. All the same, Price does have a point when he says that, because absolute virtue is compatible with evil motives, it cannot be all there is to an agent's virtuousness. How can we save this point without referring to practical virtue? In his discussion of practical virtue, Price points out the motive from which we ought to act, namely "from a conscience of rectitude". For Price, acting from this motive just is practical virtue. It should be clear, however, that "acting from a conscience of virtue" is not at all the same as "following one's conscience". If Price confused these two things, he did at least make out at one point what really is important, and that is the right motive. Price correctly sees that absolute virtue alone is not enough for morally right action, and he sees as well what motive has to be added to it for an agent to be called virtuous. For both these things he does not need practical

84

virtue. Thus, in what follows, I will treat Price's account of moral rightness as including only the demand that an agent be absolutely virtuous from the right motive, or to put it differently: that an agent do what is right, because it is right.

d) "Supererogatory" actions Equating rightness of and obligation to an action raises another problem for Price: there seem to be actions that are right one is not obligated to perform, and moreover, performing such actions seems to be more laudable than just doing one's duty.123 As we have seen, Chisholm, too, faces the difficulty of whether there could be supererogatory actions if one is obligated to right actions. So does Clarke, really, although he does not deal with it. Now, how does Price solve the problem of the possibility of supererogatory actions? The first part of the problem is whether Price can even admit that there are right actions we are not obligated to perform after equating rightness of and obligation to an action. He admits that there are such actions in this sense: some duties are general duties only, like benevolence. Generally speaking, benevolence is right, and we are obligated to be benevolent, and on this general level, both these things are equated. This is compatible with the fact that we are not obligated to be benevolent in every single instance, and thus have latitude in particular cases whether or not to be benevolent here and now. In these single cases we are not strictly obligated to be benevolent, and here we can say that benevolence is right without our being obligated to it.124 So, Price cannot really say that there are actions which are truly supererogatory, since these single cases are cases of fulfilling a general duty all the same. He can say, though, that there are right actions we are not strictly obligated to.125 Here, Price is thinking of something like the imperfect duties of natural law theory. 123

Ibid., pp. 119 ff. Ibid., pp. 120, 122 f. 125 Cf. Thomas, The honest mind, pp. 69 f. 124

85

The second part of the problem is that these "particular cases" seem to be more laudable than actions which are just dutiful. Price explains this by having recourse to the above-mentioned motive of acting "from a conscience of rectitude". Someone who acts rightly even though she is not strictly obligated, shows a higher degree of conscience of rectitude than someone who only does what is right when she is strictly obligated. Surely, it is more laudable to do the right thing even in situations when we do not necessarily have to than just when we are strictly obligated.126 Presumably, Price does not mean this to say that someone who helps out voluntarily with the Salvation Army, but who does not keep her contracts punctually and correctly, is better than someone who keeps her contracts without ever getting involved in anything else. Rather, of two people, both of whom act equally right in situations of strict obligation, but only one of which does what is right even when she is not strictly obligated, this one is more commendable than the other. Thus, in a manner of speaking, like Chisholm, Price finds a place for "supererogatory" actions in his moral theory; yet, we have to put that in inverted commas, since they really are fulfilling of a duty all the same for Price.

2. The beautiful A second thing we perceive in moral actions, according to Price, is whether they are beautiful or deformed. Although he speaks here of beauty, it really is a moral perception, not an aesthetic one: he speaks of actions as being beautiful in the sense of the ancient καλóν or pulchrum,127 not in the sense that they are pretty to look at. What does it mean that an action is beautiful in this sense? It means that we like it, that we react to it with a feeling of pleasure: We are plainly conscious of more than the bare discernment of right and wrong, or the cool judgment of reason concerning the natures of actions. We often say of 126 127

Price, Review, p. 123. Ibid., p. 62.

86

some actions, not only that they are right, but that they are amiable; and of others, not only that they are wrong, but odious and shocking. [...] [T]hese epithets [...] signify not any real qualities or characters of actions, but the effects in us, or the particular pleasure and pain, attending the consideration of them.128

Thus: The truth seems to be that, 'in contemplating the actions of moral agents, we have both a perception of the understanding, and a feeling of the heart'[.]129

At the beginning of this chapter, I have said that Price opposes sentimentalism, especially Francis Hutcheson's. Now we see that Price does not mean to deny that we do react with feelings to actions. The question is what the significance of these feelings is. The sentimentalists claim that a moral sense in us determines whether or not an action is morally right. For Price, it is exactly the other way round: right actions are apt to cause a feeling of pleasure in us. There are objects which have a natural aptitude to please or displease our minds. [...] One account, therefore, of the sentiments we are examining, is; 'that such are the natures of certain actions, that, when perceived, there must result certain emotions and affections.'130

So, it is not like we observe an action, feel something about it, and this is what determines whether the action is right; but rather, exactly because it is right we feel as we do about it. The action's rightness is just what causes our feeling of pleasure.131 An action's beauty, that is, the pleasure it causes in us, therefore is more like an epiphaenomenon for Price: the moral properties reside in the actions themselves, reason does the real moral work in us, and furthermore, really rather on the side, the action's properties cause feelings in us.

128

Ibid., p. 57. Ibid., p. 62. 130 Ibid., p. 58. 131 Ibid., p. 63. 129

87

3. Desert The third, and for the purposes of this inquiry most important, perception we have according to Price regarding moral actions is whether they are of good or ill desert. Desert is, for Price, a "species" of the right. It is needless to say any thing to shew that the ideas of good and ill desert necessarily arise in us upon considering certain actions and characters; or, that we conceive virtue as always worthy, and vice as the contrary. These ideas are plainly a species of the ideas of right and wrong.132

This is misleading. "Species" sounds like "subset" and seems to imply that some, but not all right actions are of good desert. That is not so: all right actions are of good desert. Rather, the idea of desert is an idea of a certain relation between two right actions, so on the one hand it is indeed an idea of what is right, thus a "species", yet on the other hand, and beyond that, it is also an idea of another, relative element of right actions. Price characterizes this relation as follows: I apprehend no great difficulty in explaining these ideas. They suppose virtue practised, or neglected; and regard the treatment due to beings in consequence of this. They signify the propriety which there is in making virtuous agents happy, and in discountenancing the vicious. When we say, a man deserves well, we mean, that his character is such, that we approve of shewing him favour; or that it is right he should be happier than if he had been of a contrary character.133

And further: The moral worth or MERIT of an agent, then, is, 'his virtue considered as implying the fitness, that good should be communicated to him preferably to others; and as disposing all observers to esteem, and love him, and study his happiness.' – Virtue naturally, and of itself, recommends to favour and happiness, qualifies for

132 133

Ibid., p. 79. Ibid.

88

them, and renders the being possessed of it the proper object of encouragement and reward.134

Several things come together here which we need to be careful to separate and examine. To begin with, we can sum up the basic idea of desert like this: if an agent performs a right action, then an action that will make that agent happy or is otherwise good for her, is itself right. After what has been said in the last section about the beauty of actions, we can easily understand how that is connected with our approval of doing good to someone who deserves it, or the desert subject for short: since an action that makes the desert subject happy is itself right, it is an action that causes pleasure in us.

a) Particular circumstances But how is it that an action that makes the desert subject happy is itself right? This is not because we have a general duty to make desert subjects happy. We cannot subsume this under the virtue of benevolence, either, since for that purpose it does not matter whether or not somebody is a desert subject. As far as benevolence is concerned, we ought to do well by others in general. Rather, for Price, this is because given circumstances play a part in determining an action's character, so that usually morally indifferent actions can become right or wrong under certain circumstances. This applies to right actions in general, as well as actions giving a desert subject something good in return for her own right action in particular. We could call this a form of particularism, and we find Price's clearest statement of it in his characterization of the virtue of gratitude: The consideration that we have received benefits, lays us under peculiar obligations to the persons who have conferred them; and renders that behaviour, which to others may be innocent, to them criminal. [...] With respect to this part of virtue, it is proper to observe, that it is but one out of a great variety of 134

Ibid., p. 81.

89

instances, wherein particular facts and circumstances constitute a fitness of a different behaviour to different persons independently of its consequences.135

Hence, it is appropriate to treat different people differently if one stands in different relations to them. The fact that I have received a benefit from someone, and not from another, may put me under an obligation towards the first one and not towards the other one. So if, for instance, Leigh has helped me move and Mason has not, it is possibly a breach of duty, or an impropriety or a meanness, not to invite Leigh to my housewarming party, whereas it is quite alright not to invite Mason. The last quotation also makes clear that it is not a specific feature of gratitude that particular circumstances determine what is right, but rather that this happens often, if not always. Thus, this idea can be most clearly understood when considering the virtue of gratitude, but it also lies at the heart of Price's concept of desert. This can be seen, too, from the fact that Price chooses even the same words to characterize gratitude and desert; in both cases the circumstances "render" a certain action right: "The consideration that we have received benefits, [...] renders that behaviour [...] to them criminal",136 and "Virtue [...] renders the being possessed of it the proper object of encouragement and reward".137 So, the fact that somebody has done what is morally right changes the circumstances such that now it is right to do her good in turn. Let us assume that I have an apple, and that it does not matter whether I give it to Skyler or to Trinity, it is morally indifferent. However, if Skyler now does something right in contrast to Trinity, the circumstances change, and now it would be right in turn if I gave the apple to Skyler and not to Trinity, or to put it differently: under these new circumstances, Skyler deserves the apple.

135

Ibid., p. 152. Ibid. 137 Ibid., p. 81. 136

90

b) The appropriate Price's basic idea of desert thus should be clear: the circumstance that someone does something right, makes it right to do her something good in turn. Price gives something of a definition of desert, as well, that recalls this basic idea, but which moreover explicitly links desert to propriety, or fitness in his terms: The moral worth or MERIT of an agent, then, is, 'his virtue considered as implying the fitness, that good should be communicated to him preferably to others[.'] 138

Here now is the connection between desert and propriety we have been looking for: Price explains desert by propriety. So, what about propriety then? I have already mentioned in the section on moral rightness that "the appropriate" is the synonym Price uses most often for "the right" and that, what both these terms denote, is a simple idea which cannot be defined further. Yet, Price does go on to say something further about rightness which makes it easier to understand what he means by that, namely that we are obligated to do what is right. He does say something further about propriety, as well: Fitness and unfitness most frequently denote the congruity or incongruity, aptitude or inaptitude of any means to accomplish an end. But when applied to actions, they generally signify the same with right and wrong; [...] It is worth observing, that fitness, in the former sense, is equally undefinable with fitness in the latter; or that it is as impossible to express, in any other than synonymous words, what we mean [by both].139

Thus, Price tells us that there are two different senses of the terms propriety, or fitness, both of which are undefinable, namely first, that there are appropriate means to ends, which means that these means are suitable to achieve ends, and second, the other sense of propriety, which is the 138 139

Ibid. Ibid., p. 104.

91

important one for our purposes here. We have already got to know the account of propriety as aptitude in Clarke's theory, and we have seen that it can be no more than a first approach to an understanding of this concept, and that we cannot leave it at that. At this point, Price cuts in and distinguishes more carefully between propriety as suitability and propriety as rightness. Still, the question is whether the two senses of propriety Price differentiates have nothing at all in common. After all, they both are senses of "appropriate", and that at least suggests that they are related, such that we can better understand what the one means by understanding the other. In the first sense of propriety, a connection is made between two actions or an action and a state of affairs, namely a means and an end. This connection is such that the means is apt to achieve the end. In order to get out of the room, we had better go through the door instead of running against the wall. "Going through the door" is appropriate in this sense to "getting out of the room", while "running against the wall" is not appropriate in this sense. It would now appear that we can understand the second sense of propriety in a way parallel to the first one, only in the second sense, the relation between two actions or states of affairs is not a means-endrelation. Price does not say anything explicit about this second kind of relation. Still, we have a parallel in the first sense of the relation, at least, we can understand fairly well, and Price does give some examples for propriety relations in the second sense: To treat a party of rebels, after they had surrendered themselves upon certain terms stipulated with them, in the same manner as if they had been reduced by force, would be generally disapproved [...] A general would be universally condemned, who, by means of any treacherous contrivance should engage his enemies to trust themselves in his power, and then destroy them. How different are our ideas of such conduct from those we have of the same end gained by open and fair conquest? Would it be indifferent whether a person, supposed to be just returned from some unknown country or new world, gave a true or false account of what he had seen? Is there a man in the world who, in such a case, would not think it better to tell truth than needlessly and wantonly to deceive? [...] Can we, when we

92

consider these things, avoid pronouncing, that there is intrinsick rectitude in keeping faith and sincerity, and intrinsick evil in the contrary?140

These examples are obviously not examples of means-end-relations. Here, the second sense of propriety is important. In the light of these examples, the parallel between the two senses of propriety can be demonstrated: shooting rebels who have surrendered, or lying when telling about a journey, is no more appropriate than running against a wall in order to get out of a room. Moreover, as the first example shows, that shooting rebels who have surrendered is inappropriate, also means that it is inappropriate to treat rebels who have surrendered the same as rebels one has defeated in battle. Here, too, particular circumstances come into play. Treating both sorts of rebels alike does not take into account the differences between them, it does not do justice to them, it is inappropriate.

4. Conclusion Hence, Price shows, first, that propriety is not suitability, but only something similar, and second, he explains desert by propriety. Now, how exactly are we to understand this relation between desert and propriety as Price sees it? He says that desert is a species of propriety. Primarily, this means that desert is propriety, but a special case of propriety: desert is that propriety that is only itself propriety on account of another propriety. This is a rather complicated way of saying something quite simple: let us recall Skyler, Trinity, the apple, and the fact that particular circumstances make a difference. Initially, it is morally indifferent whether I give the apple to Trinity or to Skyler. But then Skyler does something morally right, and that means: something appropriate. This is a particular circumstance that makes the decisive difference. The propriety of what Skyler does grounds the propriety of giving the apple to her and not to Trinity. Because what she has done is appropriate, it is appropriate that she get the apple, or in short: 140

Ibid., p. 133.

93

Skyler deserves that I give her the apple, because she has done something appropriate. A somewhat simpler way of putting Price's account of desert thus is: desert is second-order propriety. This also shows that Price does not only give an answer to the question of what the connection is between desert and propriety, but also an answer to the question of what the difference is between them. Let us recall that desert is a proper subset of propriety, so that all desert is propriety, but not all propriety is desert. This raises the question of where exactly the boundaries of this subset are. Price gives us an answer to this very question. Now, the question remains whether Price's answer is correct, and since his answer centres on propriety, this depends chiefly on how exactly we are to understand this concept. The problem is that, although Price does have a more refined concept of propriety than Clarke, it is still insufficient. Understanding propriety as "something similar to suitability" is no sufficient basis for examining whether the account of desert as secondorder propriety is a good one. There is yet another proponent of the school of British Rationalism, though, who can help out at exactly this point: William Wollaston.

IV. William Wollaston There is not much to say about William Wollaston's life (1660-1724). He was a contemporary of Clarke's and, like him, English. Initially, he was a teacher at a secondary school in Birmingham until inheritance and marriage made him well-to-do in his late twenties so that he retired.141 His major work in philosophy is his book The religion of nature delineated (privately released in 1722, publicly published in 1724). Although Wollaston quite clearly stands in the British Rationalist tradition, he develops a peculiarly original account of propriety as compared to Clarke's and Price's, and this is precisely why his work is of the utmost importance to the present inquiry. As we have seen, for Price, propriety is "something similar to suitability", and that leaves us at a loss as to how to understand this concept. This is exactly where Wollaston's theory comes in to help us out. Wollaston only very rarely speaks explicitly of "propriety" or "fitness",142 but the striking similarities in his language to Clarke's especially indicate that he, too, is seeking an explanation of propriety, as we will see more clearly in this chapter. For this purpose, Wollaston develops a kind of correspondence theory of morality in his The religion of nature delineated: he starts with a correspondence theory of truth and then tries to apply it to moral actions. According to him, the propriety of a moral action is determined by its correspondence or non-correspondence with certain states of affairs or other actions. Wollaston's correspondence theory of truth tallies with the one we still know today: II. Those propositions are true, which express things as they are: or, truth is the conformity of those words or signs, by which things are expressed, to the things themselves, Defin.143

141

Brown, William Wollaston. Cf. e.g. Wollaston, The religion of nature delineated, p. 45. 143 Ibid., p. 8. 142

96

This means nothing but that a truth-bearer is true, if and only if it expresses a state of affairs and this state of affairs obtains, as we would say today. As far as propositions as truth-bearers are concerned, this is not hard to understand. If propositions correspond to the states of affairs they are about, they are true. Of course, reason is the faculty that discerns truth,144 whence it follows that Wollaston is a rationalist as well. How can we apply this to actions, though? Wollaston tries to do this in the following way: III. A true proposition may be denied, or things may be denied to be what they are, by deeds, as well as by express words or another proposition. It is certain there is a meaning in many acts and gestures.145

Hence, according to Wollaston, not only propositions have meaning, but so do many actions. The questions now are: which actions have meaning; how we are to understand that actions have meaning; how Wollaston's correspondence theory actually works; and finally, on the basis of all this, what exactly propriety is.

1. Kinds of meaningful action For Wollaston, there are three kinds of meaningful action only one of which he acknowledges as important for his own purposes, though. The first kind of meaningful action are gestures. Every body understands weeping, laughing, shrugs, frowns, etc; these are a sort of universal language. [...] But these instances do not come up to my meaning.146

The second kind of meaningful action are actions that are indifferent by nature and obtain their meaning by convention only and which, therefore, might just as well have had a different meaning than they actually do.

144

Cf. ibid., p. 45. Ibid., p. 8. 146 Ibid. 145

97

It may not be improper by the way to observe, that the significancy here attributed to men's acts, proceeds not always from nature, but sometimes from custom and agreement among people, as that of words and sounds mostly doth. Acts of the latter kind may in different times and places have different, or even contrary significations. The generality of Christians, when they pray, take off their hats: the Jews, when they pray or say any of their berakoth [benedictions], put them on. The same thing which among Christians denotes reverence, imports irreverence among the Jews. The reason is, because covering the head with a hat [...] is in itself an indifferent thing, and people by usage or consent may make it interpretable either way.147

The third and last kind of meaningful action, and finally the interesting one, according to Wollaston, are those actions which have their meanings by nature and that are not gestures. He contrasts this kind with the first kind of meaningful action thusly: There are many acts of other kinds, such as constitute the character of a man's conduct in life, which have in nature, and would be taken by any indifferent judge to have a signification, and to imply some proposition, as plainly to be understood as if it was declared in words: and therefore if what such acts declare to be, is not, they must contradict truth, as much as any false proposition or assertion can.148

And he goes on to contrast the third kind of meaningful action with the second kind as follows: But acts of the former kind, such as I chiefly here intend, have an unalterable signification, and can by no agreement or force ever be made to express the contrary to it. Aegon's treating the flock, and disposing of it as if it was his, can by no torture be brought to signify, that it was not his. From whence it appears, that facts express more strongly, even than words themselves; or to contradict any proposition by facts is a fuller and more effectual contradiction than can possibly be made by words only. Words are but arbitrary signs of our ideas, or indications of our thoughts [...]: but facts may be taken as the effects of them, or rather as the thoughts themselves produced into act; as the very conceptions of the mind brought forth and grown to maturity; and therefore as the most natural and express representations of them. And, beside this, they bear certain respects 147 148

Ibid., p. 12. Ibid., p. 8.

98

to things, which are not arbitrary, but as determinate and immutable as any ratios are in mathematics. For the facts and the things they respect are just what they are, as much as any two given quantities are; and therefore the respects interceding between those must be as fixed, as the ratio is which one of these bears to the other: that is, they must remain the same, and always speak the same language, till things cease to be what they are. I lay this down then as a fundamental maxim, that whoever acts as if things were so, or not so, doth by his acts declare, that they are so, or not so; as plainly as he could by words, and with more reality. And if things are otherwise, his acts contradict those propositions, which assert them to be as they are.149

The last but one quotation in particular shows that, according to Wollaston, meaningful actions imply propositions. This needs some explaining, and I will come back to this point in the next section. First, though, we need to turn our attention to this three-fold distinction Wollaston makes between kinds of meaningful action: gestures, actions with conventional meaning, and actions with natural meaning. Wollaston explicitly says that only the last kind is important for his purposes and that raises the question, why? After all, all of these kinds of action are meaningful so why does he limit his theory to actions with natural meaning only? To examine actions with conventional meaning first, the key here probably is that such actions only arbitrarily imply the proposition that they do; they might just as well have implied a different one. Wollaston does not explicitly say why this is a problem, but he seems to think that actions with arbitrary meaning are incapable of truth-value: if taking one's hat off can just as well imply reverence for God as putting one's hat on, it is evidently not in the nature of these actions to imply reverence for God, and if something is not in the nature of a thing, then there is no truth in the matter, since according to the correspondence theory, truth consists in the correspondence of proposition and thing. At least, that is the best reconstruction we can get here of a problem with conventionally meaningful actions. Yet, this will obviously not do at all. First, it is cheating, since "a thing's nature" and "a thing" are not identical. Second, being capable of truth-value cannot depend on whether 149

Ibid., pp. 12 f.

99

or not a meaning is arbitrary for Wollaston, as in the last quotation he also mentions that words have only arbitrary meaning. If whatever has only arbitrary meaning is incapable of truth-value, we could not lie with words, and that is an unacceptable consequence, even for Wollaston. So there is no reason for Wollaston to leave actions with conventional meaning out of consideration. Evidently, gestures do not have the same fault as conventionally meaningful actions for Wollaston, as, since they are a "universal language" everybody understands, they presumably have their meaning by nature. Nevertheless, he does not think that gestures are a relevant kind of action within the scope of his theory either, and again, he does not tell us why. The one thing he does say regarding this issue is that the kind of action he has in mind is "such as constitute the character of a man's conduct in life".150 The main examples of this kind of action Wollaston discusses are consorting conjugally,151 shooting soldiers,152 and promising.153 These examples indicate that "character of conduct in life" surely does not mean a classification by social rank, that is, a classification of actions by whether they are characteristic of members of a certain class, like for instance the different codes of conduct of Norman knights and Anglo-Saxon swineherds Walter Scott describes in Ivanhoe. This would be too narrow an understanding, since obviously swineherds can consort with their partners, kill people, or promise something just as well as knights can. On the other hand, it would be too broad to see the kind of action Wollaston has in mind simply as intentional action,154 first, because evidently, we can gesticulate intentionally, too, and second, because it is not clear why going to the toilet intentionally has more to do with the conduct of our life than shrugging intentionally. We simply do not know which actions are "such as constitute the character of a man's conduct in life". The only reasonably clear criterion Wollaston gives for the kind of action he has in mind is this:

150

Ibid., p. 8. Ibid., p. 11. 152 Ibid., pp. 8 f. 153 Ibid., p. 10. 154 Feinberg, Wollaston and his critics, p. 312. 151

100

I lay this down then as a fundamental maxim, that whoever acts as if things were so, or not so, doth by his acts declare, that they are so, or not so; as plainly as he could by words, and with more reality. And if things are otherwise, his acts contradict those propositions, which assert them to be as they are.155

Taking this criterion as a basis, though, there is no reason to leave gestures out of consideration either, because obviously we can act as if things were different than they are by gesticulating, be it intentionally or even involuntarily as well. Hence, Wollaston distinguishes three kinds of meaningful action only one of which he thinks relevant to his purposes, namely actions with natural meaning. We have now seen, however, that he does not sufficiently justify the exclusion of the other two kinds, so that actions with conventional meaning and gestures both also really fall within the scope of his theory. Maybe that is why the different kinds of meaningful action have hardly been dealt with in the literature. Only Mackie and Joynton mention the difference between naturally and conventionally meaningful actions.156 Still, an interesting point by the side is that Wollaston's theory, if we take it to include naturally meaningful actions only as he explicitly does, is just the opposite of Austin's speech act theory.157 For Austin, certain propositions are not assertions, but rather, in virtue of their illocutionary role, actions: if somebody, for instance, says "I promise" in the right kind of context, then this is not a proposition about his promising something, but rather the promise itself, and the whole thing works only within a conventional framework.158 Wollaston's theory is just the other way round. Some actions are assertions, and they are so by nature. Now, the important thing about Wollaston's criterion for meaningful actions for the purposes of this inquiry is that from it we can derive the concept of propriety: treating things as they really are is acting appropriately. This concept of propriety will need to be fleshed out a little 155

Wollaston, The religion of nature delineated, p. 13. Mackie, Hume's moral theory, p. 20; Joynton, Wollaston's theory of declarative actions, p. 440. 157 Cf. Mackie, Hume's moral theory, p. 20; Joynton, Wollaston's theory of declarative actions, p. 439. 158 Austin, Other minds; How to do things with words. 156

101

more, of course, and Wollaston's correspondence theory of morals is one way to do that. I will return to it in the section after the next.

2. The meaning of actions We have seen in the last section that, for Wollaston, some actions have meaning, as they "would be taken by any indifferent judge to have a signification, and to imply some proposition".159 How exactly are we to understand this? There are two points that need explaining. First, we need to assess what Wollaston is really out for here. He simply cannot mean that actions actually imply propositions in the modern sense of "imply", that propositions are logically entailed by actions such that: I take off my hat → "I revere God". Rather, writing in 1722 and thus naturally ignorant of today's intricacies of the word, Wollaston must use "imply" in a much more innocent sense that has nothing at all to do with logical entailment. Probably, the best way to understand Wollaston in this point is by way of a Gricean account of the meaning of actions. By Grice's account, not only words and sentences are meaningful, but so is behaviour, or at least, it can be. The meaning of behaviour arises, for Grice, from the fact that an audience can take it as evidence for something's being the case. Thus, for instance, an onlooker could take someone's groaning to be evidence for his being in pain.160 One rich example of how this concept can be fruitfully put to work is Jean Hampton's theory of retributive punishment in which she illustrates how a criminal's harming a victim can be taken as evidence of the victim's diminished value, how this is something we can "read off", as she says, of the criminal's actions.161 Consider two examples of Wollaston's which are exactly in this vein: If a body of soldiers, seeing another body approach, should fire upon them, would not this action declare that they were enemies; and if they were not 159

Wollaston, The religion of nature delineated, p. 8. Grice, Meaning revisited, pp. 292, 295. 161 Hampton, Correcting harms versus righting wrongs; cf. Part 3. Chapter 1 below for a detailed account of Hampton's theory, and Section 1.a) of that chapter for further similarities between Wollaston and Hampton. 160

102

enemies, would not this military language declare what was false? No, perhaps it may be said; this can only be called a mistake [...]. Suppose then, instead of this firing, some officer to have said they were enemies, when indeed they were friends: would not that sentence affirming them to be enemies be false, notwithstanding he who spoke it was mistaken? The truth or falsehood of this affirmation doth not depend upon the affirmer's knowledge or ignorance: because there is a certain sense affixed to the words, which must either agree or disagree to that, concerning which the affirmation is made. The thing is the very same still, if into the place of words be substituted actions. The salute here was in nature the salute of an enemy, but should have been the salute of a friend: therefore it implied a falsity. Any spectator would have understood this action as I do; for a declaration that the others were enemies. Now what is to be understood, has a meaning: and what has a meaning, may be either true or false: which is as much as can be said of any verbal sentence.162

Firing upon a body of soldiers can be taken as evidence for its being the case that these soldiers are enemies: "Any spectator would have understood this action as I do; for a declaration that the others were enemies." In the second example, this point is even more manifest: In the Jewish history we read, that when Abimelek saw Isaac sporting with Rebekah, and taking conjugal liberties, he presently knew her to be Isaac's wife; [...] it is sufficient for my purpose, and to make acts capable of contradicting truth, if they may be allowed to express things as plainly and determinately as words can. Certainly Abimelek gave greater credit to that information which passed through his eye, than to that which he received by the ear; and to what Isaac did, than to what he said. For Isaac had told him, that she was not his wife, but his sister.163

For Abimelek, Isaac's behaviour was more conclusive than his words. In Wollaston's terms, this is a case of "facts express[ing something] more strongly, even than words themselves",164 or of "acts declar[ing something ...] as plainly as [...] by words, and with more reality."165 Clearly, we can 162

Wollaston, The religion of nature delineated, pp. 8 f. Ibid., p. 11. 164 Ibid., p. 12. 165 Ibid., p. 13. 163

103

rephrase this adequately by saying that in contrast to Isaac's assertion that Rebekah was his sister, the conjugal liberties he took with her were evidence of their real relationship. His actions meant that he and Rebekah were husband and wife, and Abimelek took them as such. Hence, we might reconstruct Wollaston's point that actions imply propositions as simply stating that actions have meaning, as just explained, and this meaning may be put in the form of a proposition. It is obvious that this proposition stating the action's meaning is either true or false. It is certainly misleading that Wollaston goes on to say that since this proposition is true or false, the action itself is true or false accordingly, in just the same sense that the proposition is true or false. An action is not a truth-bearer as a proposition is. Still, this is a relatively minor point and we should not attach too much importance to it. Rather, we should understand it simply as a short-cut way of saying that the proposition stating the action's meaning is true or false. For brevity's sake, I will continue to speak of action's implying propositions. Now, in order to understand what this means exactly we have to understand how Wollaston's correspondence theory of moral action works, and we will come back to that presently. Second, we have to clear up some misunderstandings regarding Wollaston's point of action's implying propositions which has been grossly distorted by critics. Thus, the completely false claim that all actions imply propositions has repeatedly been imputed to Wollaston.166 This is so manifestly wrong that I cannot see how this idea could even occur to anyone, since Wollaston himself does not once speak of "all" actions, but always only of "many".167 So, for Wollaston, it does not go for all actions that they imply propositions, but it probably goes for all three kinds of meaningful actions: laughing could imply that something is funny; taking off one's hat could imply that one is honouring God; shooting somebody could imply that the person shot at is an enemy. 166

E. g. Tweyman, Reason and conduct in Hume and his predecessors, p. 90; Tweyman, Truth, happiness, and obligation, p. 38; Feinberg, Wollaston and his critics, pp. 306, 311, 313. 167 Joynton meticulously repudiates this allegation, Wollaston's theory of declarative actions, pp. 439, 442 ff.

104

Manifestly, all of this only "could" be implied, and this is the point R. F. Holland challenges. He thinks that Wollaston is really being rather optimistic in his judgment that one action implies exactly one proposition. He points out that an action may indeed imply several propositions and it may not always be clear which ones.168 Granted, it is not impossible that an action's meaning may be ambiguous, even if we adopt the Gricean interpretation of Wollaston's point, and Wollaston need not deny it. The question is what this challenge really amounts to. Yes, we have to be careful lest we err, and sometimes it may be difficult or even impossible to find out what an action's meaning is, but we have to be cautious, and the same troubles may arise, when we consider the meaning of words and sentences as well. This does not make much of a difficulty either, so Holland's is hardly a persuasive argument against Wollaston. The possibility of ambiguity does not at all preclude actions from being meaningful nor does it imply that we are not fairly good at discerning the meaning of a given action. Certainly, Grice's account of the meaning of actions is not altogether uncontroversial,169 so that reading Wollaston along Gricean lines will not clear his theory of all problems whatsoever. Still, all in all, it is quite a respectable account, and Wollaston is really best off with this interpretation.

3. Wollaston's correspondence theory of moral action Now that we know which actions are meaningful, according to Wollaston, and also how we are to understand that these actions are meaningful, it is time to see how Wollaston's correspondence theory of moral actions proper works in order to substantiate his concept of propriety. The example of firing upon soldiers mentioned in the last section may serve as illustration. There are friendly soldiers approaching. They are shot. Shooting friendly soldiers does not correspond to their being friendly. How exactly are we to take this? 168 169

Holland, Thought, word and deed, pp. 129 f. Cf. e. g. Schiffer, Meaning, and the discussion in Grandy/Warner, Philosophical grounds of rationality.

105

For Holland, this is quite an odd idea: In the case of propositions, the making of a comparison with an existing state of affairs in order to see whether or not the proposition is true is something we reckon we know how to manage. [...] The Gedanken-experiment of doing the same thing with an action however is in danger of being stifled at its inception by our sense of the utterly queer. For since actions consist in the changing of states of affairs, how are they to be prised apart from states of affairs in order to be compared with them?170

Well, if we understand Wollaston's theory of the meaning of actions in Gricean terms, a correspondence theory of moral action could work like this: there is the state of affairs that those soldiers approaching are friendly. It would really be a misrepresentation to say that the action of shooting them "consists in the changing of this particular state of affairs". Shooting a soldier alone does not make him an enemy – although he might conceivably change sides after the incident. Rather, the shooting may be taken as evidence that the approaching soldiers are enemies. The meaning of that action may be put in the form of the proposition: "These soldiers approaching are enemies". So here we have two propositions, one describing an existing state of affairs – "These soldiers are friendly" (S1) – and one stating the meaning of the action of shooting them – "These soldiers are enemies" (S2): Proposition S1 ↑ State of affairs

Contradiction

Proposition S2 ↑ Action

In Wollaston's terms, there is no correspondence between these two propositions. He does not make clear what exactly correspondence between two propositions involves, but at the very least they must not contradict each other. Evidently, both the propositions cannot be true at the same time, and since the first one correctly describes an existing state of affairs, it is true. That much we know from our correspondence theory of truth. It fits the state of affairs, and since the second one contradicts the 170

Holland, Thought, word and deed, p. 131.

106

first one, it does not fit the state of affairs, and thus it is false. Wollaston makes this particularly clear in another example: If A should enter into a compact with B, by which he promises and engages never to do some certain thing, and after this he does that thing: in this case it must be granted, that his act interferes with his promise, and is contrary to it. Now it cannot interfere with his promise, but it must also interfere with the truth of that proposition, which says there was such a promise made, or that there is such a compact subsisting. If this proposition be true, A made such a certain agreement with B, it would be denied by this, A never made any agreement with B. Why? Because the truth of this latter is inconsistent with the agreement asserted in the former. The formality of the denial, or that, which makes it to be a denial, is this inconsistence. If then the behaviour of A be inconsistent with the agreement mentioned in the former proposition, that proposition is as much denied by A's behaviour, as it can be by the latter, or any other proposition. Or thus, if one proposition imports or contains that which is contrary to what is contained in another, it is said to contradict this other, and denies the existence of what is contained in it. Just so if one act imports that which is contrary to the import of another, it contradicts this other, and denies its existence.171

Wollaston's point now is that the action of shooting the soldiers does not fit the state of affairs that they are friendly in exactly the same way as the proposition implied by that action contradicts the first proposition. Clearly, all of this depends on there being a definite direction of fit. In the case of shooting soldiers, there is no difficulty. Either they are or they are not enemies, so shooting them either fits that state of affairs or it does not. A problem arises, however, from the fact that there are cases where there is no such definite direction of fit. The example of A's promise to B just quoted is a case in point. Here there are two actions, A's promising not to do p and A's doing p, both of which imply contradicting propositions. Wollaston tells us that since A did promise B not to do p, doing p does not correspond to the promise and, therefore, doing p is false. While there is clearly no correspondence between promising not to do p and doing p, this case is not as clear-cut as the case of shooting the soldiers. That is so, because, in contrast to that example, it is not clear which proposition has 171

Wollaston, The religion of nature delineated, p. 10.

107

priority, as it were, and thus determines the truth-value of the other. The chronological order of the actions alone cannot be the crucial factor in the matter. Let us assume that I hit my neighbour, thus implying that I disdain him. The next day I regret striking him and by way of apology I give him a flower, implying that I actually esteem him. Cannot my regret be real, cannot I honestly mean that I do esteem him just because I hit him the day before? I do not think so. Hence, since there is manifestly no correspondence between the two actions, the best Wollaston can get in this case, at least from his correspondence theory of action alone, is that it was either wrong to promise not to do p or wrong to do p, which is obviously true, but we would have liked to get more than that; we would have liked a theory to tell us that doing p was wrong, and why. This kind of case does present a difficulty for Wollaston, although not so much of a difficulty that he is forced to say something completely counterintuitive. He simply needs to supplement his theory in a way that would enable him to say that by his promise A changed the world somehow such that his breaking the promise does not fit this new state of affairs, whereas my having hit my neighbour did not change the world in the requisite manner. In any event, by now it should be clear how his correspondence theory of moral action could be made to work in principle, and that it does work in cases like shooting the soldiers, so that it is by no means as absurd as Holland and others make it out to be. To return to the concept of propriety derived in the penultimate section, we can now say that in order to act appropriately, that is, in order to treat things as they really are, we need to act in such a way as not to contradict how they really are as explained in this section. There must be a correspondence between how a thing really is and how it is treated. Thus, even in the problematic case it is not inconceivable to say that A, in doing p after having promised B not to do p, does not treat his promise to B as what it is, and therefore, that A acts inappropriately.

108

4. Lying actors Before we come to see how Wollaston applies his theory to morality proper, and thus, before we see his concept of propriety at work, there is still one question to be answered that is raised by his criterion for meaningful actions and hence by his concept of propriety. How can Wollaston deal with actors who professionally act as if things were otherwise, or in short, who lie by their actions when, for instance, they act like somebody is being executed while in fact nobody is being harmed? We want to say that this is harmless, but the question is if and how Wollaston can say so, too. It would be implausible to claim that actors are harmless because their actions have no meaning, and that, therefore, they do not lie. After all, we can understand what actors do when they act, and Wollaston would be hard-pressed to explain how we can understand something that has no meaning. Maybe Wollaston could say instead that actors do not seriously assert things to be other than they are so that they do not lie. This, however, would raise the problem that an action's meaning is no longer the important thing, but rather what the actor intends to say, and so far that was not at all important to Wollaston. The only plausible thing for Wollaston to say in answer to this problem is that actors do in fact lie, and then make the context the deciding factor in so far as harmlessness is concerned. As long as one is a spectator in a theatre and knows what is happening on stage such lies are indeed guileless and therefore do no harm. In contrast, we can imagine a group of actors improvising a scene in the street and a passer-by who, maybe because she arrives later on the scene, does not know that these people are actors and that this is not a real execution. Such a passer-by might rightly feel hoaxed. The best answer for Wollaston to this question thus seems to be that actors do lie, and the harmlessness of their lies depends on context. Presumably, he has to deal with mistaken people in the same way. Somebody who is mistaken lies in any event, even though she may not be

109

responsible for her mistake. This is indicated by his example of the mistaken officer.172

5. The appropriate treatment of humans by humans So, for Wollaston, acting appropriately is treating things as they really are. He is mainly concerned with moral actions, though, and that raises the question of how his theory, as I have outlined it, can be applied to humans. How are we to treat other humans in order to treat them as they really are? First of all, according to Wollaston, we have to keep in mind that all human beings' highest good is their happiness. Each human naturally desires their happiness. From this fact, Wollaston draws the rather dubious conclusion that happiness is desirable, and even that each human has a duty to make themselves happy.173 Therefore, this is the major aspect in terms of which we have to consider our fellow human beings: as beings who pursue their happiness and who are susceptible to pleasure and pain. Treating them as they really are, and thus, treating them appropriately, consequently is to do justice to the fact that their happiness is their highest good.174 Obviously, Wollaston has to complement this with further assumptions. As things stand, Wollaston could not condemn a person for intentionally kicking somebody else. He could not say that the kicker is not treating the kicked as a being susceptible to pleasure and pain, as, since she kicked him intentionally, we may assume that she did it precisely because he is susceptible to pain. Only on the assumption that the other is susceptible to pain is it even intelligible to kick him intentionally – nobody kicks a rock in order to hurt it. Hence, the kicker treats the kicked exactly as he really is. Some remarks in another context suggest that Wollaston backs up his theory with the golden rule,175 so that he could tell the kicker that she is a human being pursuing her happiness, and that includes not to be kicked. It 172

Cf. the example of the soldiers shot upon, Section 2. above. Wollaston, The religion of nature delineated, pp. 35, 38. 174 Cf. ibid., p. 38. 175 Cf. ibid., p. 129. 173

110

is unreasonable not to assume this of her fellow human beings as well so that in fact she has not treated the kicked as he really is.

6. The appropriate treatment of humans by God So humans ought to treat other humans as they really are. According to Wollaston, this also goes for God: he, too, ought to treat humans as they really are. Interestingly, although both humans and God ought to treat humans as they are, there is a different standard for each. As far as the treatment by other humans is concerned, the important thing is that we pursue our happiness; from God's point of view, it is our virtue that sets the relevant standard. [I]t is as certain, that there is a particular providence, as that God is a Being of perfect reason. For if men are treated according to reason, they must be treated according to what they are: the virtuous, the just, the compassionate, etc. as such, and the vitious, unjust, cruel, etc. according to what they are: and their several cases must be taken and considered as they are: which cannot be done without such a providence. Against all this it has been, as one might well expect, objected of old, that things do not seem to be dealt according to reason, virtuous and good men very oft laboring under adversity, pains, persecutions, whilst vitious, wicked, cruel men prevail and flourish.176

Wollaston does not go into this difference of standards; maybe we can help him out by saying that there are different aspects to us, and depending on which aspect we are considering, different treatments are appropriate. It is not necessary that it is God, of all "people", who treats us according to the standard of our virtue – Price, after all, speaks of the propriety of our showing favour to the virtuous –, but then it is not surprising either when we bear in mind that he is traditionally in the role of passing judgment of our virtue; Clarke links our virtue to the happiness God assigns us as well, and as we will see in the next chapter, so does Kant. 176

Ibid., p. 110.

111

Since Wollaston was influenced by Clarke, it is not altogether astonishing that he ascribes God the same place as Clarke does. However, he does go beyond Clarke's views in this respect. As we have seen, for Clarke, our virtue and the happiness God assigns us are just two states of affairs side by side, as it were, which raises the question, at least as far as Clarke's theory is concerned, what exactly they have to do with each other that could explain why we ought to get the one on the basis of the other. Wollaston offers an answer to this question that is chiefly supported by his non-voluntarism: rationally speaking, the same moral laws apply both to us and to God, and that means that he has to treat humans appropriately just like we do, and our virtue is an important aspect of us that needs to be done justice to. Consequently, because God is rational, he treats humans as they really are and gives them happiness in accordance with their virtue. The obvious objection Wollaston himself immediately sees is, of course, that humans are not as happy as they are virtuous. Wollaston's answer to this objection is that this just seems to be so as long as we consider this life only: [I]f the virtuous man has undergone more in this life, than it would be reasonable he should suffer, if there was no other; yet those sufferings may not be unreasonable, if there is another. For they may be made up to him by such injoyments, as it would be reasonable for him to prefer, even with those previous mortifications, before the pleasures of this life with the loss of them [i.e. the mortifications]. [...] On the other side, if the vitious and wicked men do prosper and make a figure; yet it is possible their sufferings hereafter may be such, as that the excess of them above their past injoyments may be equal to the just mulct of their villainies and wickedness. [...] If what is objected be in many instances true, this only inferrs the necessity of a future state: that is, if good and bad men are not respectively treated according to reason in this life, they may yet be so treated, if this and another to follow be taken together into the account.177

Therefore, for Wollaston, the imbalance between virtue and happiness in this life is in fact a proof for the existence of an afterlife in which God levels out this imbalance such that everything is as it rationally should be. 177

Ibid., pp. 113 f.

112

7. Critique of Wollaston's theory in the literature Most commentators ridicule rather than criticize Wollaston. This began with a footnote in David Hume's A Treatise of human nature in which he parodies Wollaston's theory – although he does point out some noteworthy difficulties, too –178 and this attitude is still common today.179 At the same time, these critics give such a distorted presentation of Wollaston's theory that it rather seems as though they have not read Wollaston proper, but only Hume's parody. This impression is shared by Feinberg,180 and he emphatically decries this abuse: I am no ardent admirer of Wollaston. His conclusions are not especially congenial to me, and his argumentation is undistinguished. But after discovering in various commentaries the amount of scorn and ridicule that have been heaped on his now forgotten theory, and the shocking misrepresentations and sophistries used in 'refuting' it, I am moved (out of a simple sense of justice!) to set the record straight.181

This is quite admirable of Feinberg. Unfortunately, though, his sentiment is undermined by the fact that Feinberg himself is one of those critics who impute absurd claims to Wollaston, like, for instance, the claim that all actions imply propositions.182 Yet, apart from all this, there are some problems with Wollaston's theory discussed in the literature which do have to be taken seriously. There are three lines of attack against Wollaston each of which can be subsumed under the general heading of arguments from circularity. I will illustrate these variants in the next section, then discuss Joynton's rescue attempt, and conclude by exonerating Wollaston.

178

Hume, A treatise of human nature, p. 461 n. Cf. e. g. MacIntyre, A short history of ethics, pp. 170 f. 180 Feinberg, Wollaston and his critics, pp. 308 f. 181 Ibid., p. 308. 182 Cf. Section 1 above and 7.a) below. 179

113

a) Arguments from circularity The main objection that has been put forward against Wollaston from Hume and Price to present-day critics is that his argument is circular. They allege that, for Wollaston, moral wickedness depends on the falsehood of actions. However, the objection goes, Wollaston already has to presuppose moral principles in order to mark some actions as false in the first place. In Hume's oft-cited footnote it says: Besides, we may easily observe, that in all those arguments there is an evident reasoning in a circle. A person who takes possession of another's good, and uses them as his own, in a manner declares them to be his own; and this falshood is the source of the immorality or injustice. But is property, or right, or obligation, intelligible, without an antecedent morality? A man that is ungrateful to his benefactor, in a manner affirms, that he never received any favors from him. But in what manner? Is it because 'tis his duty to be grateful? But this supposes that there is some antecedent rule of duty and morals.183

So the objection is: it is only because we already have a concept of property, for instance, that we are even in a position to say that someone acts contrary to truth when she treats something that belongs to somebody else as if it were hers. Likewise, only on the presupposition that we have the duty to be grateful can we say of an ungrateful person that she acts contrary to truth. The point here must be this: let us assume that Cory does Drew a favour, and Drew does not invite Cory to her birthday party. According to the critics, this alone does not mean anything at all. Only when we presuppose moral concepts or duties can we characterize Drew's behaviour as ungrateful. In other words: sometimes it makes sense to say that an action is contrary to truth only on the presupposition of moral principles. Therefore, the moral evaluation of an action cannot depend on its truthvalue, since this in turn is already dependent on the action's moral evaluation. 183

Hume, A treatise of human nature, p. 462 n.; also cf. Price, Review, p. 125; Feinberg, Wollaston and his critics, p. 311; Mackie, Hume's moral theory, p. 22.

114

There is a variant of the problem of circularity: so there are true and untrue actions according to Wollaston – so what? Why ought we always to act in accordance with truth? Wollaston can only say this if there is a moral duty to be truthful. From this it follows that the moral evaluation of an action depends on its truth-value, while its truth-value itself is subject to moral evaluation.184 Feinberg takes this problem too lightly when he says that this objection is "the lamest of all of Hume's 'refutations'", because it is "a question which neither has nor needs an answer".185 It is a reasonable question to ask, and it has a reasonable answer, too. Yet another objection is closely connected to the problem of circularity as well, and that is that in Wollaston's theory there can be no grades of wickedness. All untrue actions are equally contrary to truth and thus equally wicked, which is absurd: Unhappily, however, Wollaston cannot explain on the single ground of fidelity to truth why treating a post as if it were a man is not precisely as evil as treating a man as if he were a post. Both modes of conduct are absurd, declarative of falsehood, and out of accord with nature, and equally so.186

This objection is closely connected to the problem of circularity, because it is carried on along the lines that a further criterion is needed in order to properly grade wickedness, and that this criterion is just one of those moral considerations Wollaston has to presuppose.

b) Joynton's rescue attempt Olin Joynton has ventured to defend Wollaston's theory against these challenges by shifting the focus away from the truth-values of actions to their conduciveness to making people happy: Whether rightly or wrongly, [Wollaston] regards the whole process of finding assertions in actions and then testing their truth as a more reliable procedure for 184

Hume, A treatise of human nature, p. 462n.; Mackie, Hume's moral theory, pp. 21 f. Feinberg, Wollaston and his critics, p. 312. 186 Ibid., p. 308; cf. Mackie, Hume's moral theory, p. 56. 185

115

identifying the moral qualities of actions than that of predicting their expected bearing on human happiness. This does not entail that interference with truth is what makes wrong actions wrong: what makes them wrong is that they threaten human happiness. Suppose, by analogy, that someone has graded a set of exams on a point system and that any exam scoring below 70 points is to receive a failing mark. The point system coincides with the quality of work and makes it easy to flip through the set and pick out the failing exams, but it is poor work that makes a student fail, not the number of points received. [...] By equivocating 'because' we might be troubled momentarily by this specious case of circularity: Smith's exam is a failure because it has a score of 65, and it has a score of 65 because it is a failure. In simplest terms this is the same kind of mistake made by those who charge Wollaston with circularity.187

Thus, Joynton thinks an action's truth-value to be merely something like a ratio cognoscendi, whereas an action's conduciveness to happiness is the ratio essendi which of course is the truly important thing. This manoeuvre gets Wollaston's argument out of the vicious circle all right, because now an action's truth-value is no longer important in evaluating its moral wickedness. An action is wicked if it is detrimental to happiness, and that is that. Moreover, this solves the additional problem mentioned, since here is a perfect criterion for grading: the more detrimental to happiness an action is, the more wicked it is, too.188 Of course, now we are at pains to explain why Wollaston takes the "detour" via determining an action's truth-value at all. Joynton dismisses this problem rather curtly by saying that Wollaston simply has more trust in our ability to discern the truth than in our ability to see what is conducive to happiness and what is not.189 This is really quite a daring move, as discerning the truth in moral matters can hardly count as child's play. Joynton is undoubtedly right in saying that happiness plays a large part in Wollaston's theory; it is our highest good, after all. Nevertheless, by no means does it play such a major part that everything depended on it. Joynton's manoeuvre makes a utilitarian out of the rationalist Wollaston 187

Joynton, The problem of circularity in Wollaston's moral philosophy, p. 441. Ibid., p. 442. 189 Ibid., pp. 442 f. 188

116

really is. A rescue attempt that requires Wollaston not to be Wollaston is unacceptable, and therefore I suggest not following Joynton's lead in this matter.

c) Propriety and moral realism Wollaston's critics focus primarily on the problem that the truth of actions cannot be a criterion of moral evaluation for Wollaston, since it is itself subject to moral evaluation. This affects Wollaston's concept of propriety in particular, and indeed, it seems that it is this that the critics are really aiming at. To make this clearer, let me rephrase each variant of the problem of circularity as applied to the concept of propriety. First, the description of how things really are may sometimes be a description of moral properties; in such cases, moral evaluation determines what is appropriate treatment; therefore, appropriate treatment cannot be the criterion of moral evaluation. Second, Wollaston has to presuppose a moral duty to treat things appropriately in order to be able to say that we ought to treat things as they really are. Third, the concept of propriety presupposes moral considerations, because otherwise there is no criterion for grading. However, Wollaston can be defended against these objections up to a point. At the very least, we can say that from these problems it does not follow that Wollaston's concept of propriety is as untenable as most critics make out. We have to realize that Wollaston's concept is a very strong concept, and we have to take seriously his claim that moral evaluation really does not ground propriety, but rather depends on it. For this, we have to bear in mind that Wollaston, like Clarke and Price, is a moral realist. He does not say he is as explicitly as Price does, but his expression "as things really are" strongly suggests that his view of morality is a realistic one. Presupposing moral realism, though, we can indeed say that it is only dependent on the real properties of the situation whether Drew is ungrateful towards Cory, and that there is no need to back this up with moral principles like, say, a duty to be grateful. Similarly, moral concepts such as property, right, and duty do not simply appear out of the blue, but rather they have to be justified. The very point of a realistic account of morality, now, is precisely that this justification need not rest on moral

117

concepts or principles itself. Then, we can say that whatever justifies rights, for instance, in a non-moral way is what makes it inappropriate to violate rights, and because it is inappropriate, it is morally wrong to violate rights. I do not mean to say that moral realism does not face any difficulties whatsoever, and therefore, that the mentioned objections do not really touch Wollaston's theory. I do want to say, though, that moral realism is not an absurd position by any means. To be sure, Wollaston does not defend his realism, and to do it for him would go far beyond the scope of this book. But we can at least say that Wollaston's concept of propriety does work and can respond to the objection of circularity within a respectable and relatively widespread metaethical framework. The second variant of the problem of circularity does not pose a problem specific to Wollaston's concept of propriety either. Propriety is a normative concept, and of course, that raises the question of why we ought to act appropriately. This is a problem quite parallel to other views of morality: morality in general is normative, and that, too, raises the question of why we ought to be moral. Typically, there are two sorts of answer to this question. Either we state non-normative considerations in order to explain or justify this normativity, and that raises the further question of whence these non-normative considerations derive their force; or we state other normative considerations, and then we beg the question. Wollaston's realistic account of morality suggests that he would lean towards an answer along the first of these lines, though he is not committed to it, whereas the objection of circularity is based on the second kind of answer. Again, my point is not that these issues do not pose a real problem for Wollaston, especially since his realism does not pin him down to the first kind of answer and both kinds of answers are fraught with problems in any case. I only want to say that his concept of propriety is, in this regard, not worse off than morality in general, and that therefore it is not manifestly more absurd than morality in general either. Of course, we can accuse Wollaston of ignoring this problem, but in this as well he is not the only one. In a similar way we may answer Feinberg's challenge that Wollaston has to presuppose moral considerations, because otherwise there would be

118

no criterion for grading the gravity of impropriety. Feinberg's point is that, since treating a person as if she were a post is inappropriate, and treating a post as if it were a person is inappropriate as well, both these treatments are equally inappropriate. In answer to this version of the problem of circularity, like in answer to the last one, we can say again that Wollaston's concept of propriety is no worse off than morality in general, for we might just as well say, quite parallel to the objection, that, since both lying and murdering somebody are morally wrong, both of these things are equally wrong. This is patently absurd, and since both cases are parallel, possibly the challenge against propriety is as well. Maybe Feinberg could reply that we must not construe these cases as parallel to each other, because in the case of morality it is not a problem to include further moral considerations in order to explain gradual differences, whereas this is exactly the problem for the concept of propriety, as here we just must not presuppose moral considerations. We can answer this problem, nevertheless, like the first variant of the problem of circularity, namely that in view of Wollaston's realism, it simply is not settled yet that it is really moral considerations he has to presuppose in order to get a grading criterion. Clearly, these considerations do not refute any of the objections brought forth against Wollaston's theory, and they were not meant to. They were meant to show that these objections do not really aim specifically at Wollaston's concept of propriety, but rather at the moral realism it presupposes. Certainly, a lot of questions about moral realism remain unanswered, it is prone to this sort of objection, and the champions of realism have yet to offer a satisfactory defence. Still, the judgment of whether or not realism fails is still pending, especially since no version of moral anti-realism is better off as things are. Consequently, it is not absurd to be a moral realist, and it is acceptable for Wollaston, who has written a treatise on normative ethics after all, to disregard problems with the metaethical framework he presupposes in this treatise. Given this metaethical framework, Wollaston's concept of propriety can be exonerated from the charge of circularity.

119

8. Conclusion So, according to Wollaston, we ought to act appropriately, and that means we ought to treat things as they really are. Thus, we finally have the means for examining in Chapter VI whether Price's account of desert passes. Let us recall that, for Price, desert is propriety founded on another propriety, or second-order propriety. Consequently, the key to understanding what desert is, is to understand what propriety is. Price has not offered a sufficiently clear concept of propriety, and this is where Wollaston can help out. Moreover, Wollaston's concept of propriety can help explain why Clarke and Price understood propriety as they did, namely as "suitability" and "something similar to suitability", respectively. At first sight, "treating things as they really are" does not seem to have much to do with "choosing apt means to an end", and that raises the question of whether Wollaston's concept is close enough to Clarke's and Price's, vague though they may have been, to serve as a basis for them. However, this first impression is deceptive. Although Clarke and Price do not offer substantive accounts of propriety, Clarke does concentrate on the paradigm case of propriety, and Price sees that this paradigm case is not the whole truth. Aptness is the paradigm case of propriety, since choosing a means unsuitable to a certain end – like trying to peel an egg with a pneumatic hammer – is one of the most evident cases of just not treating things as they really are. Here, finally, the lead we followed from Clarke comes to its end, as Wollaston's concept of propriety can indeed serve as a basis for Price's concept of desert. Before I can turn to that, though, there is still another point at issue. In discussing the objection of circularity to Wollaston's concept of propriety, another important aspect was hinted at. A version of this argument was the problem of why we ought to act appropriately. I have replied in Section 7.c) that this objection really cannot be aimed against Wollaston's concept of propriety, but rather against the metaethical framework he presupposes, so that, given this framework, his concept of propriety can pass. Nevertheless, shining through this objection is a need for understanding

120

why propriety is important to us, and we must not simply sweep aside this need. The last great representative of the rationalist tradition, Immanuel Kant, can help answer this question. His concept of the worthiness to be happy is closely related to Price's concept of desert, and Kant's concept of the highest good answers the very question of why propriety is so important to us.

V. Immanuel Kant With Immanuel Kant (1724-1804), we return to the second of Clarke's leads that we have left to one side for the time being in favour of the first one concerning the problem that we still had no substantive account of the concept of propriety. This led us first to Price and eventually to Wollaston's solution to this problem. Kant, too, is concerned with this problem when he discusses his concept of the worthiness to be happy. His importance for the purposes of this inquiry, though, does not lie there, but rather in his contribution to solving the second problem we have seen in Clarke. This second problem was that, apart from the question of how exactly we are to understand propriety, there is also the question of what virtue and happiness have to do with each other such that we ought to get the one for the other. Why ought we to get anything at all for being virtuous, and then, why happiness, of all things? In the end, this comes down to the question of why it is important for us to receive happiness for our virtue. This question, of course, refers primarily to the special case of desert Clarke considers, namely desert of happiness in virtue of being virtuous. At this point, we have come full circle. In the discussion of Wollaston's general concept of propriety, as opposed to this special case of desert, we have seen the need for understanding why propriety and thus desert are important to us. This need now can be satisfied by showing why it is important to us to get what is deserved in the particular case, and applying this to the general case. Clarke pointed out the right way to follow in this matter as well: he saw that human beings need to be able to hope for virtue and happiness to be compatible in order to be virtuous in the first place. This is why we ought to get happiness for being virtuous. This line of thought is picked up by Kant. In his concept of the highest good he answers this very question that has been hinted at by Clarke and that we have come across again in Wollaston: why is it important for us to get what we deserve? Like Clarke and Wollaston, Kant is concerned with the special case of deserving happiness in virtue of being virtuous, or moral, as Kant calls it,

122

even though he does not speak of "desert", but rather of "worthiness". Still, he means the same thing, and if Kant considers the same case as two of his predecessors, his treatment of it is much more extensive.

1. Morals The central passage in Kant's works for his characterization of morals is the first section of his Grundlegung zur Metaphysik der Sitten (1785). However, if we want to know what account Kant gives of desert, the "Tugendlehre" in Metaphysik der Sitten (1797) where he explicitly defines deservingness, or meritum, seems to be the most relevant text. In view of the fact that for the purposes of this inquiry Kant's concept of morals has to be seen exactly from the angle that it is a desert base for happiness, the question arises as to what extent these two passages are compatible with each other, or if they are not, which one takes precedence.

a) Morals in the Grundlegung In the Grundlegung Kant gives an answer to the question of what we have to do in order to become worthy of happiness: we have to be moral. This takes two things: first, we have to perform the right action, and second, we have to have the right maxim in performing the action. The first point is quite plain. In order to be moral, we have to do what is right, that much is clear. Right actions are, according to Kant, those that we are bound by duty to do. Like Price, though, Kant is impressed by the possibility that doing the right thing is compatible with being immoral. Price is led by this problem to an untenable differentiation between absolute and practical virtue. By contrast, Kant chooses a more promising path which, nonetheless, Price already hinted at. Thus, according to Kant, performing the right action is not enough. Something else is required as well, and that is the right maxim. We can perform right actions according to wrong maxims, for instance, if we do

123

what is our duty simply because we are so inclined, or because of some consequences. To act from inclination or for the sake of consequences is to act from the wrong maxim, because, although this may lead to externally dutiful actions, it will do so only by accident, as here it is not the rightness of the action that is the deciding factor in its performance, but rather inclination or the action's consequences. If that is the case, however, these wrong maxims can just as well lead to wrong actions as to right ones. In order to make sure, therefore, that we do what is right, we have to act in accordance with a maxim that makes it impossible to act wrongly. Performing the right action in accordance with the right maxim thus means: acting as duty demands and exactly because it is our duty, or in short: acting from duty. Only if both these conditions are met are we virtuous, and only then are we worthy to be happy.190 As I have said, this is already hinted at in Price, namely in the motive we have to have in performing the action. Kant's "acting from duty" and Price's "acting from a conscience of rectitude" have the same basic structure.191

b) Desert in the "Tugendlehre" Interestingly, in the introduction to Metaphysik der Sitten Kant gives a quite different answer to the question of what we have to do in order to deserve happiness than in the Grundlegung: If someone does more in the way of duty than he can be constrained by law to do, what he does is meritorious (meritum); if what he does is just exactly what the law requires, he does what is owed (debitum); finally, if what he does is less than the law requires, it is morally culpable (demeritum).192

190

Kant, GMS, Section I. Cf. Chapter III.1.c) above. 192 Kant, MS, p. 227 AA / p. 382 in the translation. All references to Kant's works in this chapter are to the Akademie Ausgabe (AA), except the quotations where I will give the relevant pages both of the AA as well as of the particular translation used. 191

124

This takes up a rather different intuition we have about desert than the one that is worked with in the Grundlegung. In the Grundlegung, we are told that we have to do a certain thing in order to deserve happiness, namely acting from duty. By contrast, in the Metaphysik der Sitten, the important thing is not doing a particular thing, but rather performing an action that stands in a certain relation to other actions we could perform instead in the situation. There are actions which we can be "constrained by law" to perform, and there are actions that go beyond that, with which we do more than with enforceable actions, and it is only in virtue of such actions that go beyond the enforceable that we deserve happiness. When you pay taxes you do exactly what you can be constrained by law to do, you do your duty. If, in addition, you donate for the renovation of the public indoor swimming pool you do more than what you can be constrained to do, and hence you deserve happiness in virtue of this. The familiar intuition behind this is that we need not only do something particular in order to deserve happiness, but rather that we deserve happiness only when we act better than we have to. Acting as one has to and acting better than one has to each corresponds to the fulfilment of a certain kind of duty Kant differentiates between in the "Tugendlehre", namely the duties of right and the duties of virtue.193 Many commentators think that this differentiation between duties of right and duties of virtue corresponds to the differentiation between perfect and imperfect duties. On the other hand, several deny this on the grounds that although all duties of right are perfect duties, duties of virtue can be perfect as well as imperfect.194 Yet, whether or not these differentiations correspond to each other is not relevant for my purposes, nor are other differentiations Kant makes between duties, like between positive and negative duties, or between duties towards oneself and duties towards others.195 Therefore, I will not address the issue of these differentiations and their correspondences here. 193

For an account of the difference between this differentiation of duties and the one in the GMS, cf. e.g. Schaller, From the Groundwork to the Metaphysics of Morals. 194 Cf. e.g. Schaller, From the Groundwork to the Metaphysics of Morals, p. 333; Hill, Kant on imperfect duty and supererogation, p. 357. 195 For a survey of the differentiations, cf. Hill, Kant on imperfect duty and supererogation.

125

Duties of right now are such duties as are concerned with actions, that is, if one has a duty of right, one is obligated to do something. By contrast, duties of virtue are duties that deal with maxims to actions, that is, if one has a duty of virtue, one is obligated to have a certain maxim.196 The duty to have a maxim entails the duty to at least sometimes act in accordance with this maxim, too.197 Nonetheless, since there is no concrete duty in this case to perform a particular action on a particular occasion, the focus of the duty is not on the action but on the maxim. So if someone has the maxim of, say, helping people in need, she must at least sometimes help people in need, even though she need not help every person in every predicament. Otherwise it is not the case that she really has this maxim – except maybe if she is never in a position to be able to help people in need. These maxims that are the object of duties of virtue are determined, according to Kant, by ends which we have a duty to have. There are two sorts of this kind of end: one's own perfection, and the happiness of others.198 From the differentiation between duties of right and duties of virtue according to their objects, that is, actions on the one hand and maxims on the other, another difference between these kinds of duties follows. We can be constrained to fulfil our duties of right, we cannot be constrained to fulfil our duties of virtue. We can be constrained to do something, but we cannot be constrained to do it in accordance with a particular maxim.199 I will concentrate only on the point that we can be constrained to do our duties of right, and not on the point that we can be constrained by law to do them, as Kant says. Kant's expression suggests that by duties of right he means such duties that arise from the laws of a state,200 but that is controversial,201 and for the present purposes it is not important. So, taking all of this together, we can say that someone fulfilling her duty of right acts as she has to, because she can be constrained to do so, while someone fulfilling her duty of virtue acts better than she has to, because she cannot be constrained to do so – she "does more than she can 196

Kant, MST, p. 388. Cf. Gregor, Laws of freedom, p. 103. 198 Kant, MST, pp. 382 f., 385. 199 Cf. ibid., p. 383. 200 Cf. e.g. Gregor, Laws of freedom, p. 26. 201 E.g. Hill, Kant on imperfect duty and supererogation, p. 356. 197

126

be constrained to do". Consequently, we deserve happiness in virtue of fulfilling our duties of virtue, whereas we do not deserve anything in virtue of fulfilling of our duties of right.202 As Kant says: "Fulfilment of these [i.e., the duties of virtue] is merit."203

c) Critique of the "Tugendlehre"-account There are two problems with Kant's account in the "Tugendlehre" in which he considers the desert base for happiness to be the fulfilment of duties of virtue. The first problem is that it is simply not the case that in fulfilling duties of virtue one acts better than in fulfilling duties of right, or that one does more than what one can be constrained to do. The line of argument in the "Tugendlehre" is as follows: (1) One only can deserve something in virtue of doing more than what one can be constrained to do. (2) One cannot be constrained to have particular maxims. (3) Duties of virtue have maxims as their objects. Thus: In fulfilling duties of virtue, one does more than what one can be constrained to do, and therefore one deserves something in virtue of this. Put this way, we can easily see what is wrong with this argument: although it is quite correct that we cannot be constrained to fulfil our duties of virtue, this alone does not mean that in fulfilling them we have done more than what we can be constrained to do. Just because we do something we cannot be constrained to, it does not follow that it is better than what we can be constrained to do, too. The only difference is that we can be constrained to do the one, but not the other. That entails no judgment that the latter is better than the first. So Kant does not show that in fulfilling

202 203

Johnson, Kant's conception of merit, p. 313. Kant, MST, p. 390 AA / p. 521 in the translation.

127

duties of virtue we do "more" than what we can be constrained to do, but only that we do something "other" than what we can be constrained to do. The second problem lies in the difference of evaluation between duties of virtue and duties of right that follows from the account in the "Tugendlehre" which Kant cannot allow within the scope of his own moral theory. Fulfilling our duty is the central requirement in Kant's moral philosophy.204 In the "Tugendlehre" now, a differentiation is made between kinds of duties from which a different evaluation of the fulfilments of each kind of duty follows. Kant says that fulfilling one's duties of virtue is "merit (meritum) = +a; but failure to fulfil them is not in itself culpability (demeritum) = -a, but rather mere deficiency in moral worth = 0".205 Since he is explicitly concerned here with the fulfilment of duties of virtue only, it is not clear how we are to evaluate duties of right, the debitum, within the scope of this scheme. Still, since there are only these three possibilities mentioned, merit, culpability, and deficiency in moral worth, one of these three values has to be assignable to the fulfilment of duties of right. The passage from the Metaphysik der Sitten quoted at the beginning of the last section makes it plain that Kant expressly intended to differentiate between the debitum and the meritum, so that fulfilling our duties of right just cannot be deserving. It cannot be culpability either, of course, because in doing the debitum we do exactly what is owed. Therefore, fulfilling one's duties of right can only be evaluated as "deficient in moral worth", or as morally indifferent, according to the "Tugendlehre". Hence, fulfilling our duties of right has the same moral status as, say, wiping spectacles, and it is, morally speaking, quite indifferent whether we wipe our glasses or whether we fulfil our duties of right. Yet, in view of the fact that, within the scope of Kant's moral philosophy, it simply is not the case that the fulfilment of duties is indifferent, but rather the crucial point, this is just not an adequate evaluation of the fulfilment of duties of right. Consequently, in contrast to what Kant says in the "Tugendlehre", there can be no difference in evaluating the fulfilments of different kinds of duties. 204 205

Especially clear in Kant, KpV, p. 86. Kant, MST, p. 390 AA / p. 521 in the translation.

128

Therefore, it cannot be that we deserve happiness in virtue of fulfilling our duties of virtue, while deserving nothing for fulfilling our duties of right. Since duties of right cannot be in the same class as morally indifferent actions, and since all of the above can be applied, mutatis mutandis, to duties of virtue, it follows that one deserves happiness in virtue of fulfilling any of these two kinds of duties. With this, though, Kant's account in the "Tugendlehre" collapses, and here too, what remains is what Kant already told us in the Grundlegung in reply to the question of what we have to do in order to deserve happiness, and that is: we have to act from duty.

2. From morality to happiness So now we know what we have to do, according to Kant, in order to deserve happiness: we have to be moral. The question now is what exactly happens when we are moral. According to Kant, the following happens: if we are moral, we get to be happy. At first sight, this is manifestly false: commonly, human beings are moral and happy to different degrees, there is no proportionality between these two things. Whether we are happy is dependent on all sorts of accidental things; our morality has little, if anything, to do with it. Kant admits that the course of this world does not conform to our ways of thinking in this matter, but rather conforms to laws of nature, and that therefore our happiness in this world does not, in fact, depend on our morality. However, the accidental happiness in this world is not the kind of happiness Kant means, anyway. He has in mind our happiness in the afterlife. This happiness, by contrast, is indeed caused by our morality. Thus, according to Kant, we set a cause in this world, here we have to be moral, and in another world, in our afterlife, an effect follows, there we are going to be happy in proportion to our morality. For a cause in this world to be able to have an effect in another, though, some kind of link is needed. For Kant, this link is God. God makes us happy in the next life in virtue of and in proportion to our morality in this life.206 206

Kant, KrV, pp. 525 f.; KpV, pp. 113 ff.

129

So far, Kant adopts the outlines of the story Clarke and Wollaston have already told before him. Yet, from here he goes on to develop the story far beyond these outlines. For the proportional assignment of happiness in virtue of morality, we need two further presuppositions: first, the immortality of the soul without which there is no afterlife, and second, the existence of God. Both these points are outside of the realm of our knowledge and cannot be proven theoretically. Nevertheless, they can be postulated. Postulates are necessary conditions for something that "indubitably ought to exist", and have, for Kant, objective reality despite their unprovability. According to Kant, the proportional assignment of happiness in virtue of morality indubitably ought to exist; for this, God's existence and the soul's immortality are necessary conditions; therefore, God's existence and the soul's immortality are postulated; and consequently, they are objectively real.207 The fact that the existence of God is only postulated has an effect on his makeup as the link between morality and happiness. Postulating the existence of God means nothing less than making our own concept of God.208 Still, we do not simply make our own concept of God arbitrarily, but in accordance with moral principles:209 we make a God, because it ought to be that happiness is allocated in proportion to morality, and we make him exactly so that he allocates it just so. Because we make him to allocate happiness just so, that is what he necessarily does. We do not have any other concept of God – if he did not do that, he would be a different God than the one postulated, and therefore, he would not be objectively real. Thus, "our" God necessarily assigns happiness in virtue of and in proportion to morality.210 As I have already said, this is essentially the story Clarke and Wollaston tell as well, but Kant has a much harder time dealing with God than they. Postulating God's existence so that he necessarily has the properties he needs to do what he was postulated for in the first place, really rather seems like prestidigitation. By contrast, the British 207

Kant, KrV, pp. 421, 425 f.; KpV, pp. 125 f., 134. Cf. Kant, Vornehmer Ton, p. 401 n. 209 Cf. Kant, KrV, pp. 530 f. 210 Cf. Kant, Religion, p. 99. 208

130

Rationalists spent next to no time on a proof of God's existence – maybe because Clarke, at least, was a clergyman –, and simply assumed it as a matter of course. On the other hand, as non-voluntarists they had the mentioned difficulties of finding an important place for God in morality at all.

3. The worthiness to be happy and the concept of the highest good As Kant basically retells more or less the same story as Clarke so far, he also faces the same problems Clarke does at this point. The first problem in Clarke's theory was that it is not clear which kind of link there is between morality and happiness. The second problem in Clarke's theory, that recurred in the objections against Wollaston, is that we want to know why it is important for us to get happiness in virtue of our morality. In contrast to Clarke, though, Kant does go on at this point and introduces his concept of the worthiness to be happy and his concept of the highest good in order to deal with these problems, although his primary contribution to this inquiry is in his dealing with the second problem.

a) Worthiness The first problem Clarke faces is that it is not apparent from the story he tells why virtue ought to result in happiness; why ought God to assign us happiness for our morality? What are the grounds for this Ought? At this point, Kant introduces his concept of the worthiness to be happy in order to establish the connection. Accordingly, we are worthy to be happy in virtue of and in proportion to our morality; and we get to be happy in virtue of and in proportion to our worthiness. Kant does not explicate further what exactly he means by worthiness, but again and again, he quite emphatically expresses the intuition that, for him, is behind this concept:

131

[A]n impartial rational spectator can take no delight in seeing the uninterrupted prosperity of a being graced with no feature of a pure and good will[.]211 For, to need happiness, to be also worthy of it, and yet not to participate in it cannot be consistent with the perfect volition of a rational being that would at the same time have all the power, even if we think of such a being only for the sake of experiment.212 [O]ne inevitable judgement must have forced itself upon [mankind reflecting upon right and wrong]. It could never be that the issue is all alike, whether a man has acted fairly or falsely, with equity or with violence, albeit to his life's end, as far at least as human eye can see, his virtues have brought him no reward, his transgressions no punishment. It seems as though they perceived a voice within them say that it must make a difference.213

We can rephrase this intuition of Kant's in terms of its being inappropriate that the virtuous are unhappy and the vicious happy. With our virtue we create an incomplete state of affairs that requires completion that consists in our getting to be happy in virtue of and in proportion to our morality. To put it in Wollaston's terms, the virtuous are not treated as they really are if they do not get to be happy in virtue of and in proportion to their morality. Hence, it appears that we can quite nicely understand Kant's concept of worthiness, and thus the connection between morality and happiness, as desert: we deserve happiness in virtue of and in proportion to our morality, and we get happiness in virtue of and in proportion to our deservingness. How exactly we are to understand desert here we will have to see with the help of Price and Wollaston in the next chapter.

b) The highest good So with his concept of the worthiness to be happy Kant means to say that we deserve happiness in virtue of our morality.

211

Kant, GMS, p. 393 AA / p. 49 in the translation, cf. KU, p. 443. Kant, KpV, p. 110 AA / pp. 228 f. in the translation. 213 Kant, KU, p. 458 AA / pp. 128 f. in the translation. 212

132

This leaves the second problem we have already seen in Clarke. It is not at all evident why we deserve something in virtue of our morality at all, and if we do, then why happiness, of all things. As I have already said, this comes down to the question of why it is important for us to get happiness for our virtue. For the purposes of this inquiry, the importance of Kant's contribution lies in answering this question by means of the concept of the highest good. Kant defines what he thinks to be the highest good as follows: Thus happiness in exact proportion with the morality of rational beings, through which they are worthy of it, alone constitutes the highest good of [the] world[.]214

Hence, the highest good consists of two components for Kant, namely, our morality and the happiness we get in virtue of and in proportion to our morality. The main question now is why the highest good consists of two components at all, when it is morality that is the focal point of Kant's moral philosophy. Why does happiness come into play in addition to that, even though it is not happiness in general, but only the happiness we are worthy of? The problem is that, as far as acting is concerned, morality taken by itself does not get us very far. The moral law to which our morality needs to conform may give the agent's will the right form and is its motive. In other words, if an agent's will passes the test of the categorical imperative, and if, moreover, it is this fact that is her reason why she wills the way she does, this agent is moral. The problem is that that alone does not tell the agent what to do. Her will does not need the right form and a motive only, but an object as well.215 She cannot simply will the right way, she must will something the right way.216

214

Kant, KrV, p. 528 AA / pp. 681 f. in the translation; cf. KpV, pp. 110 f., 119. Cf. Kant, KpV, pp. 108 ff.; cf. also Düsing, Das Problem des höchsten Gutes in Kants praktischer Philosophie, p. 30; Wood, Kant's moral religion, p. 62; Engstrom, The concept of the highest good in Kant's moral theory, pp. 751 f. 216 Kleingeld puts the point in a similar way, Moral und Verwirklichung, p. 428. 215

133

To put it differently, the formula of morality in the Grundlegung demands that an agent act from duty, but it does not tell the agent positively what to do, since the formula does not substantiate that. Thus, the agent needs an object for her will, and that just is her happiness, because she pursues that naturally. In turn, the problem with happiness, according to Kant, is that happiness in general is not worthwhile, but only that happiness we are worthy of.217 So happiness is something the agent wills, but she ought to will it in the right way, and that is what she needs morality for. Consequently, morality determines agency formally, and happiness determines it materially. Each component by itself is insufficient and that is why Kant unites them both in the concept of the highest good.

c) Heteronomy and the natural pursuit of happiness Yet, Kant's concept of the highest good has been severely criticized. A particularly powerful objection has been raised by Lewis White Beck: Kant simply cannot have it both ways. He cannot say that the highest good is a motive for the pure will, and then say that it is so only under the human limitation that man must have an object which is not exclusively moral [...]. The theory of the Analytic requires him to deny that the concept of the highest good provides an autonomous motive. While the hope for the highest good may in fact be a necessary incentive to do that to which the concept of duty would not move man, it is clear that to admit the latter human – all-too-human – fact into the determination of conduct in accord with moral norms is to surrender autonomy.218

Beck's criticism comes down to saying that, with his concept of the highest good, Kant undermines his own moral theory. As I have already pointed out, in Kant's moral philosophy, morality, and along with it autonomy, is the focal point. If Kant now assigns an equally important part in his moral 217 218

Cf. Kant, KrV, pp. 527 f. Beck, A commentary of Kant's Critique of practical reason, p. 244; cf. also Murphy, The highest good as content for Kant's ethical formalism, p. 105; Auxter, The unimportance of Kant's highest good, p. 130.

134

philosophy to happiness, this means that along with it heteronomy is admitted, because human beings are not rational in their pursuit of happiness, but rather externally determined, and thus heteronomous. Hence, as Beck goes on, Kant's concept of the highest good can only lead to a "semblance of morality" at best.219 Maybe Beck is right, and Kant's concept of the highest good does not agree with everything he says elsewhere. I cannot examine this in detail, that would be far beyond the scope of this book. The point is that Kant cannot do without the concept of the highest good. The key to understanding this lies just in the answer to the question of why we deserve happiness in virtue of our morality; or, to put it differently, why happiness is the object of our will. This is so, because we are human. As human beings we simply are not rational beings only, but natural, finite, sensual beings, and as such we just naturally pursue our happiness. We cannot say that these are different things, that morality need not care about our sensual nature; we cannot say that morality is valid, no matter if it is our undoing. After all, our rationality and our sensuality are often in conflict with each other; what we ought to do from a moral point of view can conflict with our happiness, and what we ought to do in order to become happy may be immoral. Beck suggests solving this problem in such a way as to let only virtue count from a moral point of view and to disregard happiness completely. However, in doing so he does not do justice to our humanity. If we have to deny such an important part of ourselves for the sake of morality, we may quite reasonably ask why morality should concern us at all, and why under these conditions we should not "regard the moral laws as empty figments of the brain".220 As Steven G. Smith rightly states: The merely formal moral law does not hang in the air, enjoining universality of practical judgment in the abstract; it appears to us as a condition of our alreadyongoing quest for happiness. Human morality is structured a priori by the fact that we are human beings and not angels. Therefore, if we cannot be moral 219 220

Beck, A commentary of Kant's Critique of practical reason, p. 244. Kant, KrV, p. 526 / p. 680 in the translation.

135

human beings – if the highest good is demonstrably unattainable because its component parts are irreconcilable – then we cannot be moral at all.221

Consequently, the solution to the problem of conflict between morality and the pursuit of happiness must be to combine both these things, just like Kant does in his concept of the highest good. This way, there is hope that morality and happiness may be compatible, which does justice to our rationality as well as to our sensuality. We have already seen this point shine through in Clarke who says that without hope of happiness we cannot be virtuous. His problem was, as I have said, that although he established a connection between these two things, our need for happiness and our capability of being virtuous, he did not tell us what kind of connection that is. Hence, Clarke could not reply to a criticism like Beck's: he could not give an answer to the objection that morality need not care about our needs in order to be valid. Kant does have an answer. He shows that morality cannot come round anthropological facts. If morality wants to have anything to say to human beings, it has to take into account that it is, in fact, human beings it is speaking to. Kant's concept of the highest good shows that it is not true that morality does not have to care about our needs, because a morality that does not take into account how human beings work is invalid for us.

4. Conclusion Kant's primary contribution within the scope of this inquiry thus is his answering the question that was raised in connection with both Clarke and Wollaston, namely why it is important to us that propriety obtain. Kant answers this question on the basis of the special case of desert of happiness in virtue of morality. Nonetheless, we can apply his conclusions to propriety in general on which desert is grounded, after all. 221

Smith, Worthiness to be happy and Kant's concept of the highest good, p. 172; cf. Düsing, Das Problem des höchsten Gutes in Kants praktischer Philosophie, pp. 33 f.; Kleingeld, Moral und Verwirklichung, p. 430; for a similar line of argument, though in a different context, cf. O'Neill, Kant on reason and religion, pp. 272 ff.

136

With his concept of the highest good, Kant shows us what morality and happiness have to do with each other, so that it becomes plain why we ought to get something in virtue of our morality, and why we ought to get happiness for it, of all things. Morality gives an agent's will the right form and is her motive, and happiness is the object of her will, because the agent as a finite sensual being naturally pursues it. This is an anthropological fact: we human beings just work this way that we pursue our happiness and need the hope for its being attainable, too. Since this is the way it is, it is important for us to obtain happiness for our virtue. So how can this be applied not only to desert, but to propriety in general? Well, since desert of happiness in virtue of morality just is a special case of propriety, since the concept of desert is grounded on the concept of propriety, and since propriety is treating things as they really are, it follows that assigning happiness in virtue of morality is treating the moral person as she really is. Now, it would appear that the anthropological fact Kant cites can be put generally as follows: human beings just work like this, that they tend to treat things as they really are and that, moreover, they need the hope for things being treated like this. Obviously, this is not so in every single instance, but most of the time people do choose means apt for their ends, tell the truth, use things as they are supposed to be used, make differences in their treatment of different persons according to whether they know each other, whether they are friends or relatives, whether they like each other; and they expect others, and sometimes from the course of the world, to act in like manner. In short: since this is so, it is important for us that propriety obtain, or that incomplete states of affairs be completed. Evidently, this is not a normative justification of propriety. In this chapter, neither Kant nor I have said anything at all as to why propriety ought to obtain. He only has said something about why it is important to us that it does.

VI. Propriety and desert All the pieces have fallen into place, and the two most crucial findings we have gained so far in this second part are Price's concept of desert and Wollaston's concept of propriety. Finally, we are in a position to assess the former by means of the latter, and to go on from there and develop an understanding of what it is to deserve something. Before coming to desert, let us first consider propriety again.

1. The concept of propriety For Wollaston, propriety is treating things as they really are. He uses his correspondence theory of moral action to substantiate this, as I have pointed out in Chapter IV, and although this theory is much more sound than it is commonly supposed to be, it should be noted that it is not the only conceivable way of substantiating it. A different way is suggested by Bittner's theory of action as discussed in Chapter III.1 of Part 1, and there may be still other ways. Thus, not adopting Wollaston's theory does not make the concept of propriety void. So, apart from such concrete theoretical underpinnings, what exactly does that mean, "as things really are"? Especially within the scope of British Rationalism, it is plausible to take that to mean that we ought to treat things according to their nature. In that context, "the nature of things" sounds very Aristotelian like the telos of things, their function, what they are there for, what they are good for, what their essence is. Clarke in particular leans towards this understanding when he tells us that things and persons have natures from which a propriety of certain modes of conduct arises. If that were so, we would have to abandon talk of propriety. A concept of propriety that had to presuppose such a nature of things would be untenable. There is no such nature of things. It is plausible to understand "as things really are" like that, but it does not do justice to the way in which even the British Rationalists use the

138

concept of propriety. "As things really are" refers to that relatum that makes something else appropriate; as Chisholm would put it, it is that which requires something else, it is the p in the relation "p requires q". This p, let us call it the propriety base, is a state of affairs, and that can be just anything. The state of affairs that Cory has done Drew a favour makes it appropriate that Drew invite Cory to her birthday party. The state of affairs that Cory has done Drew a favour is "as things really are". In no sense of the word does Cory's having done Drew a favour have anything to do with Cory's nature. Rather, it is just a fact, how things are, what is true of Cory. In other words, it is nothing special. Since Cory has done Drew a favour, Drew owes gratitude to Cory, and if she is ungrateful towards Cory, she does not treat Cory as if Cory had done her a favour. Accordingly, not to invite Cory to her birthday party, taken as a sign of ingratitude, is inappropriate of Drew. An anecdote might illustrate this point further. When the Italian football referee Pierluigi Collina led his first international match, Germany v. The Netherlands, the German Football Association gave him a hair dryer to mark the occasion. This is an unusual gift for such an event under the best of circumstances; but the point is that Collina suffers from a metabolic disorder which causes him to be bald. This disorder is not in Collina's nature. He simply has no hair, that is all. In view of this disorder, giving him a hair dryer is not treating him as he really is, and therefore inappropriate. To be sure, the point is not that such a gift might be offensive, but rather that Collina cannot know what to do with it. Hence, contrary to first appearances, the concept of propriety does not presuppose a nature of things, and therefore, it does not need an unacceptable metaphysical foundation. The preliminary account of the concept of propriety I have offered in Part 1 can thus be taken like this: propriety is the required completion of an incomplete state of affairs, and that means that there is something about things and persons that requires they be treated in a certain manner.

139

2. The concept of desert Understanding propriety is the key to understanding what it means to deserve something. Now we know that, that the desert subject x deserves the desert object y in virtue of the desert base z, means that it is appropriate for x to get y in virtue of z, and that that in turn means that the state of affairs that x is, or has done, z constitutes how x really is, and that this requires something, namely that x get y. In other words, if the desert subject does not get the desert object, it is not treated as it really is, namely as a subject that has set the desert base. Thus, we can see how the two-place propriety relation can be the foundation of the three-place desert relation. The desert subject x and the desert base z are not strictly distinct. Propriety encompasses a requirement to treat things as they really are: x is the thing, and z is how it really is. Accordingly, the treatment due to that thing so constituted is getting the desert object y. In short: in Chisholm's propriety relation "p requires q", p and q can be replaced with x, y, and z of the desert relation as follows: "(x and z) requires y", which is just another way of putting "x deserves y in virtue of z". Hence, we have grounded the concept of desert on the concept of propriety, but now a new problem arises, the problem of distinction. When playing seven notes of a major scale, it is appropriate to play the eighth note as well, it is required, the major scale is not treated as it really is, if the eighth note is not played; but it is not the case that the playing of seven notes of a major scale deserves that the eighth note follows. All desert is propriety, but not all propriety is desert. Therefore, having grounded desert on propriety is not enough, we also need a criterion of distinction. To illustrate the problem with an analogy: all squares are rectangles, but not all rectangles are squares, too. So far, I have concentrated on showing that squares are rectangles and on saying what exactly rectangles are. This is important in order to understand what squares are, but it does not suffice. In addition, we have to learn that squares are rectangles with equal sides.

140

a) Price's account Price has been the first and, to the best of my knowledge, the only one so far who not only grounded desert on propriety, but who also gave a criterion of distinction. His point is that desert is a species of propriety, namely it is that propriety that itself is appropriate in virtue of some other propriety. Recall the example of the apple: Skyler does something appropriate. This in turn makes it appropriate that I give her an apple. Consequently, Skyler deserves the apple, because she has done something appropriate. In other words, the desert relation is a propriety relation that in turn has some other propriety relation as its propriety base, or rather as its desert base, as the case stands. First of all, this criterion of distinction would be suitable, because, obviously, not every propriety is based on another propriety. Let us say that the appropriate thing Skyler did was helping me when I was in need. She treated me as I really was, as somebody in need; however, that I was somebody in need, was not because I had treated anything else as it really was. Thus, when Skyler helped me, it was appropriate, but I did not deserve it. By contrast, if I give her the apple, I treat Skyler as she really is, that is, as my benefactress; and she is my benefactress, because she has treated me as I really was, that is, as somebody in need. Therefore, it is not only appropriate to give Skyler the apple, but moreover, she deserves it. As it stands, this criterion is too narrow; we have to broaden it for cases like punishment. Evidently, punishment is not deserved for appropriate, but rather for inappropriate behaviour. This needs only a minor adjustment, though, as we can say that the desert base is not a propriety relation, but, put more generally and a bit awkwardly, propriety-affecting behaviour. The main problem with this account, however, is that it does not work for a lot of cases we want to count as cases of desert. Arminia Bielefeld deserve to win because of their good play. If the desert base itself has to be a propriety relation, we have to say that the Arminia's good play has been appropriate. The question is who or what has the Arminia treated as it

141

really is? The only possibility in this case is: the game of football itself, or Platonically speaking: the Form of Football. The problem with taking desert as second-order propriety, or more accurately, the problem with it for Price's account, is that, even though we may say something like this, that the Arminia has treated football as it really is, that they have played football as it should be played, or at least, that they have done better at it than their adversaries, it is not the case that the Form of Football requires football to be played in a certain manner. This way of putting it indicates that it is not quite correct to speak of Football as a Platonic Form in the strict sense of the term. Yet, for want of a better term, as well as for brevity's sake, I will refer to relata such as the present one as Forms. The present problem is even better illustrated in the case of the beauty queen, and moreover, in this case yet another problem arises. Robin deserves the first place in the beauty contest, because she is the most beautiful contestant. First, the additional problem is that the desert base in question is a property and it is odd to say that one can treat things as they really are with a property. It is odd, but maybe we can get around that somehow. In any event, this case illustrates particularly well the main problem: even if we can say that Robin treats the Form of Beauty as it really is with her being beautiful, it is not the case that the Form of Beauty requires of her – or of anybody – to be beautiful. Consequently, although Price's account of desert as second-order propriety seems to be promising, because he manages both to ground desert on propriety and to state a criterion of distinction between desert and propriety, he still cannot account for all the cases we want to count as cases of desert, and that is a point against his theory. Hence, the question remains how to understand desert.

b) Desert bases as criteria of distinction At this point, it is tempting to have recourse to the debate on desert in the literature and try to find a criterion of distinction there, after all.

142

It is tempting to say that desert relations are those propriety relations which permit of only specific things as desert bases. That would give us a splendid criterion and, moreover, it would include the debate in the literature thus far. Most importantly, it would help escape the most disagreeable problem of having to presuppose Platonic Forms. The problem with this attempt is that it just does not work. To illustrate this, let us say that only actions the desert subject is responsible for are viable desert bases. Desert, then, would be propriety with a responsible action as the propriety base. There should not be too much importance attached to this particular example. The arguments against it can be brought forth, mutatis mutandis, against any similar attempt, as well as against any attempt at reducing the acceptable classes of desert subjects or objects. There are two problems with this attempt which I have already outlined at the beginning of Part 1. First, the heterogeneity of desert judgments attests to the fact that anything can be a desert base. We speak as a matter of course of the beauty queen deserving the first place, of a painting deserving admiration, of a landscape deserving preservation. Therefore, this attempt does not cover all common desert judgments any more than Price's does. Indeed, for any attempt at reducing the acceptable class of desert bases, we can easily find more than enough examples of desert judgments with different desert bases. Given this, it would be pure dogmatism to say that only desert judgments grounded on certain desert bases would be "real" ones, taking all others to be metaphoric at best. Second, this attempt blurs the distinction between the truth and the meaning of desert judgments. To take up the example from Part 1 again: the Ancient Greeks believed that a man could deserve something in virtue of his noble birth. Maybe this desert judgment is wrong, I do not know. Nonetheless, if one were to say that that is not the right kind of desert base, then this judgment is not wrong, but rather meaningless. This is an unacceptable consequence, especially in view of the fact that for any given desert base there are countless desert judgments that cannot be subsumed under it. An analysis of a concept from which it follows that the concept is meaningless in countless cases is untenable.

143

As I have said, it is possible that one cannot deserve anything for certain things, that is, that some things cannot ground desert, and consequently, that some desert judgments are false. Still, there have to be independent reasons for that – that can neither be derived from the concept of desert nor can the concept of desert be grounded on that.

c) Fittingness as desert base So if neither Price's account of desert as second-order propriety nor the account of desert as propriety grounded on a specific kind of desert base holds, what else is left? Since Price's account fails in some, but not all, cases of desert, the key is modifying his theory so as to cover all cases. The main problem with Price's account is that not every desert base is a propriety relation, because it is not true in every case that that thing of which we can say that it is treated as it really is with the desert base requires that very treatment. To recall the aforementioned examples: the Form of Football does not require football being played at all, let alone in a certain manner; the Form of Beauty does not require anyone to be beautiful. The normative element of requiring something now is the very thing that distinguishes propriety from fittingness. Fittingness is the nonnormative relation that underlies propriety as I have explained in Part 1. I have illustrated the difference by way of the examples of the shards of a broken bowl and of the major scale. The shards of the bowl are fitting to each other, but they do not require to be rejoined. By contrast, the eighth note is fitting to the first seven of the major scale already played, and furthermore, it is required as well. The shards are fitting to each other; the eighth note is not only fitting, but appropriate. The point now is that while propriety is too strong to be taken as desert base in each case of desert, exactly because of the normativity inherent in it, fittingness is eminently suited for that task, since fittingness is essentially non-normative. In every other respect, the relation of fittingness to propriety is sufficiently close so that everything that was plausible in Price's account regarding propriety is true of fittingness as well.

144

Consequently, desert is that propriety that has at least fittingness as its propriety, or desert, base. The "at least" is added, because, clearly, there are still many cases of desert, as Price himself shows, that have propriety as desert bases, or rather propriety-affecting behaviour which includes impropriety as well, as, for instance, the cases of deserved rewards and punishments. In those cases, however, in which this is not true, the desert base is a fittingness relation. This account preserves the advantage of Price's of giving us a good criterion of distinction between desert and other propriety relations, since evidently, as I have already said, not only do not all propriety relations ground on other propriety relations, but neither do all fittingness relations. On the other hand, this account remedies the disadvantage of Price's in which not all cases of desert were covered. With at least fittingness as desert base, all cases of desert are adequately explained, as I will show in the next section.

d) Desert judgments and propriety judgments So desert is that propriety that is appropriate in virtue of a fittingness. In order to show that this concept of desert indeed covers all cases, I now return to the examples of desert judgments I have considered in Part 1 and elucidate their exact meaning. DJ1: Arminia Bielefeld deserve to win. The Arminia's good play is fitting to the Form of Football, or at least, it is more fitting than their adversaries' play. For that reason, it is appropriate for the Arminia to win. Thus, the Arminia's play was fitting, and therefore, they deserve to win. DJ2: The murderer deserves to be punished. Killing Bailey was not only unfitting, it was inappropriate: Amber has not treated Bailey as she really is, say, as a sentient being striving to survive. This case shows why the desert base is "at least" a fittingness, as there are cases like this one in which the desert base is not only a

145

fittingness relation, but a propriety relation as well. Killing Bailey was inappropriate, and therefore, Amber deserves to be punished. DJ3: The most qualified applicant deserves to get the job. Ryan's qualification is fitting to the profession of a carpenter. Therefore, it is appropriate for her to get a job as a carpenter. Because Ryan's qualification is fitting, she deserves the job. DJ4: The most beautiful participant deserves the first place in the beauty contest. Robin's property of being beautiful is fitting to the Form of Beauty. Since it is a contest we are speaking of, the important thing is not only to be beautiful, but to be the most beautiful participant. Contests are a comparative matter. Thus, it is fitting to be the most beautiful participant, and hence, Robin deserves to win. DJ5: Whoever has toiled all day in the garden deserves to rest in the evening. Erin's gardening is fitting to the state the garden has been in; maybe presupposing the garden to have been in a bad state, but I am not sure about that. Maybe there is even propriety here: maybe Erin has decreed her garden to be of a specific kind, and maybe with her work she treats it as it really is according to her definition. At least, however, her work is fitting, and therefore, Erin deserves to rest in the evening. DJ6: The landscape deserves to be preserved. The sequoias' uniqueness may be fitting to many things, let us say it fits Earth's great variety of landscapes. Because the trees' uniqueness is fitting, they deserve to be preserved. To be honest, I am not quite sure about DJ6; I am not sure whether the trees' uniqueness really is fitting to anything. In that case, we could not say that the landscape deserves to be preserved, but only that the preservation is appropriate. Practically speaking, there is not much difference, since, because the desert relation is a propriety relation, the normative force is the

146

same in both relations. Both judgments state that the trees ought to be preserved. The difference is that maybe we cannot say that they "deserve" that. I do not want to decide this case here, it is unclear. What is clear, though, is that whoever says that the trees deserve to be preserved in virtue of their uniqueness, says that their uniqueness is fitting to something, like for instance Earth's variety of landscapes. That leaves the cross-check. After all, limiting the desert bases to fittingness relations is supposed to serve as a criterion of distinction between desert and propriety. Now we have seen that the various desert judgments can be adequately explained by that. That leaves showing that propriety judgments (PJ) do not ground on fittingness. PJ1: The eighth note is appropriate to the first seven notes of the major scale played. PJ2: It is appropriate to court the prince.222 PJ3: Living happily ever after is appropriate to Patricia and Delia's love for each other. To be honest again, I cannot think of anything to which either playing the first seven notes of a major scale or the fact of being a prince is fitting. In each case where there is nothing at all to which either of these is fitting, there really is "only" a propriety judgment to be made. Of course, no one can rule out the possibility of construing cases in which either of these things is fitting to something else. In those cases, we would really have to say that the eighth note and the courting are deserved. I have already said that anything can be a desert base provided that it is fitting to something, and that we cannot reduce the class of desert bases to certain kinds. Hence, it would be very odd to say that, of all things, playing seven notes of a major scale and being a prince are totally unsuitable for being desert bases. From the fact that they can be fitting to something else, however, it does not follow that they are always fitting to something else. Whenever there is something to which either is fitting, desert follows; if there is nothing like that, propriety follows.

222

I thank Ralf Stoecker for this example.

147

With PJ3 I fare like with DJ6: I am not sure into which class this judgment belongs. I suppose one could say that, to judge from the story, Patricia and Delia belong to each other, and that one could consequently say that it is fitting that they love each other. If one says that, one has to say that their living happily ever after is not only appropriate, but also deserved. Yet, maybe we can say no such thing, and in that case, their living happily ever after is not deserved, but just appropriate. As far as that is concerned, I can only state here that it is not always possible to determine unequivocally whether particular judgments are propriety or desert judgments. There is a grey area between propriety and desert, and with judgments within that grey area, like DJ6 and PJ3, it is difficult to say which side they belong to. Still, even though the line between desert and propriety is surrounded by a grey area, it is clear where it is to be drawn: desert is that subset of propriety that is appropriate in virtue of fittingness. It is also clear that the question of whether a particular judgment is a desert judgment or a propriety judgment cannot be settled by analytic, but only by substantial reasons.

e) Conclusion This concept of desert as propriety with fittingness as propriety base is easy to understand and covers all the cases we usually count as cases of desert. Going through the particular examples in the last section, however, has shown that one problem remains which Price already had with his account: since it is fittingness as desert base that constitutes desert, there still always has to be something there which the desert base is fitting to. With this solution, as with Price's, there is no better candidate available for that than Platonic Forms, as for instance the Form of Football in the Arminia example, or the Form of Beauty in Robin's case, although I have already pointed out that this is an incorrect term for the things I have in mind, and that I use it only for want of a better one. To be sure, the things I have in mind dwell in the metaphysical vicinity of Platonic Forms, and the term will do as a placeholder at a pinch.

148

In any event, this concept of desert has a high metaphysical price: first, it presupposes propriety, that is, it presupposes incomplete states of affairs that require completion; and second, it presupposes Platonic Forms. I have not endeavoured to argue for the existence of such entities independently of the present inquiry – how could anybody in present times? Still, I have provided what can be considered as an indirect argument: if we want to talk about desert, and talking about desert presupposes such entities, this is a reason for accepting them. Of course, given the cost, we could just as well opt for simply abandoning the concept of desert altogether. Yet, I shall try to give further reasons for keeping the concept in the next part, namely that this concept can help us in solving problems of moral philosophy where other concepts fail. If these reasons do not turn out to be convincing, I shall not insist on preserving the concept of desert. I shall insist, though, on having analysed it correctly.

Part 3: Applications Now we have an account of the concept of desert, and we also see that it comes at a price: in order to speak of desert we have to presuppose some rather substantial metaphysical underpinnings. As I know that some will be tempted to abandon the concept of desert altogether in view of this price, I shall show in this part that abandoning the concept will be expensive as well. The concept of desert is not only pervasive in ordinary language, that is, not only do all sorts of people speak of desert in all sorts of everyday situations, but furthermore, there is a place for this concept in moral philosophy that cannot be easily covered by other normative concepts. This is, of course, especially manifest in theories that are explicitly grounded on the concept of desert, as, for instance, retributive theories of punishment. Retributivism is an attempt to substantiate the basic propriety, or rather impropriety relations that ground certain desert relations, namely those that have punishment as a desert object. The problem here is that this is an attempt to apply a concept nobody has clearly understood as yet. Only with the concept of desert developed in this book can we make plain what exactly retributivism is even about. More importantly, this concept can be used for more than merely clarifying what the basic retributivist idea is. We need a clear concept of desert in order to be able to defy fundamental objections against retributivism which themselves are based on a misunderstanding of the concept, and thus to be able to allow retributivism the status of a theory of punishment to be reckoned with that its followers claim for it. In short, we need a concept of desert in order to justify punishment. Moreover, the worth of the concept of desert shows not only in helping to clarify theories, but also in the fact that with it we have a further normative instrument which we can solve moral problems with where other normative concepts fail us. The residue of moral dilemmas, for instance, can be far better explained with the concept of desert than with any other concept offered in explanations so far.

150

These two points indicate that the concept of desert is not only important for understanding what people mean when they talk about desert as they do every day, but that it can do real explanatory work that cannot be provided by other normative concepts.

I. Saving retributivism Put very roughly, there are two kinds of justification of punishment. The first kind is consequentialist and points out the beneficial consequences of punishing wrongdoers. The other kind is retributivist and is guided by the idea that wrongdoers deserve punishment. This is only a rough differentiation, since there are countless variants of both kinds of justification that are all quite different from each other. In addition, there are several hybrid forms that try to unite the best of both kinds. Each version of retributivism explicitly presupposes the concept of desert and thus is one of the most prominent fields of application for this concept in moral and political philosophy. This has not led retributivists to give a general analysis of the concept, but each tries to give their own account of what it means that a wrongdoer deserves punishment. Each theory of punishment, regardless of which kind, has inherent difficulties. However, the retributivists' falling back on the concept of desert has prompted their consequentialist opponents to a fundamental criticism of the retributive enterprise itself: [W]hat pass for retributivist justifications of punishment in general, can be shown to be either denials of the need to justify it, or mere reiterations of the principle to be justified, or disguised utilitarianism. Assertions of the type 'it is fitting (or justice requires) that the guilty suffer' only reiterate the principle to be justified – for 'it is fitting' means only that it ought to be the case, which is precisely the point at issue. Similarly, since justification must be in terms of something other than the thing in question, to say that punishment is a good in itself is to deny the need for justification.223

223

Benn, An approach to the problems of punishment, p. 327; cf. Honderich, Punishment, p. 15; Duff, Trials and punishments, pp. 3 ff., 195 ff.; Walker, Why punish?, pp. 83, 119, 136; Walker, Just desert or just desertion?, p. 162; Braithwaite / Pettit, Not just deserts, pp. 11, 165; Matravers, Justice and punishment, p. 5; Slattery, The myth of retributive justice, pp. 31 ff.; Thornton, Against retributivism, pp. 83 f.

152

Put more succinctly, the objection against retributivism is that, when they say that it is justified to punish wrongdoers because they deserve it, they say no more than that "it's right to punish criminals because doing so is right".224 This is a mere tautology, though, and hence cannot possibly justify punishment. There are fundamental objections against consequentialist theories of punishment as well, basically the same sorts of objections that are put forth against consequentialism in general – most importantly, the objection that consequentialism entails some "repugnant conclusions". Yet, here I want to concentrate on the fundamental objection against retributivism and show by means of the concept of desert I have developed in this book that this objection is misguided. It is chiefly the lack of an accurate account of desert, together with the inherent problems each retributive theory has and that have rightly been pointed out by their opponents, that lead consequentialists to impute to retributivists that their basic principle is a mere tautology and that any retributive justification of punishment is doomed to failure simply due to analytic reasons. As soon as we know what it means to deserve punishment, though, it becomes quite plain that this is not so. It may be that no version of retributivism actually works, but that has to be due to other difficulties, not because retributive justifications of punishment are analytically unsound.

1. Hampton's retributivism Opponents of retributivism mainly attack particular retributive theories in order to show that retributivism in general fails as well, because its basic idea of desert is just "an unargued and unarguable intuition".225 Therefore, I myself will consider a particular retributive theory, Jean Hampton's, and exemplary show with the help of the concept of desert that

224

Dolinko, Retributivism, consequentialism, and the intrinsic goodness of punishment, p. 507. 225 Duff, Trials and punishments, p. 3.

153

despite many inherent problems with this theory, the fundamental objection of the unjustifiability of retributivism does not pass. Hampton's theory is but one of many retributivist attempts to justify punishment. Still, I am inclined to think that it is the best attempt yet, since for Hampton, in contrast to other proponents, the crucial point is the victim of the crime, and thus, she is best able to grasp what exactly the wrong is the wrongdoer commits, namely hurting the victim. Other, more prominent, theories make out the wrong to be the liberty the wrongdoer takes and that law-abiding citizens deny themselves,226 and that seems to me to be a gross misrepresentation of what is wrong with, say, rape. Still, it is of little consequence that I prefer Hampton's theory to others. Everything I have to say regarding her theory can, mutatis mutandis, be applied to any other retributive justification of punishment. Hampton first introduced her theory in "The retributive idea" (1988) and refined it in a series of further papers227 leading up to "Correcting harms versus righting wrongs: The goal of retribution" (1992). In this last paper, she explicitly takes on the challenge against retributivism in general: [C]ritics of [the retributive] approach to punishment are, in my view, right to charge that its supporters appeal to little more than intuition when they defend the idea that all and only wrongdoers 'deserve' to suffer. In this paper, I will develop what I call the 'expressive' theory of retribution in order to explain and defend the idea of retributive desert.228

The foundation of the idea of retributive desert is the connection between punishment and "that which makes the wrongful action wrong".229 So first of all we need an account of wrongful action that can serve as the desert base for punishment.

226

Cf. e.g. Morris, Persons and punishment; Murphy, Marxism and retribution; Murphy, Three mistakes about retributivism; Finnis, The restoration of retribution. 227 Cf. e.g. Hampton, A new theory of retribution; Hampton, An expressive theory of retribution. 228 Hampton, Correcting harms versus righting wrongs, p. 1659. 229 Ibid., p. 1661. Italics deleted.

154

a) Moral injury Hampton defines wrongful actions in general as actions that violate a moral standard valid in the situation. Manifestly, not all wrongful actions in this sense are worthy of punishment. A famous counterexample Hampton discusses is the case of the ship owner who kept his ship tied to a dock despite the fact that the dock owner had forbidden it, because a storm approached that doubtlessly would have wrecked the ship. During the storm, the ship caused considerable damage to the dock. Keeping the ship tied to the dock was wrongful of the ship owner, he infringed on the dock owner's rights in doing so, but general opinion is that he ought not to be punished but only compensate the dock owner for the damage.230 So which are the wrongful actions that ought to be punished? According to Hampton, only "moral injuries" are worthy of punishment, that is, those wrongful actions that are "an affront to the victim's value or dignity".231 Here, Hampton presupposes a Kantian-egalitarian concept of the value of human beings: all human beings are autonomous rational endsin-themselves, from which it follows that they all have the same value.232 A moral injury now Hampton defines as damage to the realization of a victim's value, or damage to the acknowledgement of the victim's value, accomplished through behavior whose meaning is such that the victim is diminished in value.233

Hence, in order to understand what a moral injury is we have to understand four things: first, what does Hampton understand by meaning of behaviour? Second, what is diminishment? Third, what is realization of a victim's value? And fourth, what is acknowledgment of a victim's value? Hampton's understanding of the meaning of behaviour goes back to Grice. Accordingly, a wrongdoer intends by the moral injury of his victim to give evidence of the victim's lesser and his greater value.234 We can 230

Cf. ibid., p. 1666, 1664. Ibid., p. 1666. Italics deleted. 232 Cf. ibid., pp. 1667 f. 233 Ibid., p. 1679. 234 Cf. ibid., pp. 1674 ff.; Grice, Meaning. 231

155

"read off" this message concerning values of the wrongdoer's action, this is what he is "saying".235 As I have already pointed out in Chapter IV.2 of Part 2, Hampton's manner of speaking about the meaning of actions is surprisingly close to Wollaston’s. Her characterization of what the wrongdoer does is strongly suggestive of his characterization of what makes an action wrong. This is even more pronounced in Hampton's first paper on retributivism, "The retributive idea", where she says that the wrongdoer makes "a false moral claim" and hence "denies moral reality".236 In comparison, let us consider again Wollaston's account of the meaning of actions and the possible contradiction of reality: I lay this down then as a fundamental maxim, that whoever acts as if things were so, or not so, doth by his acts declare, that they are so, or not so; as plainly as he could by words, and with more reality. And if things are otherwise, his acts contradict those propositions, which assert them to be as they are.237

Here, we can see quite clearly that, for both Hampton and Wollaston, the major moral problem of wrongful actions is that with them something is said that does not conform to reality. In other words: for both of them, wrongful actions are wrongful, because they are lies. This was just by way of mentioning an interesting parallel, though. The second question that is raised by Hampton's definition of a moral injury is what she understands by diminishment. Hampton assumes that all humans have equal value and she makes quite plain that nothing can touch this value. Then, what could it mean that the wrongdoer still diminishes the victim's value? According to Hampton, the victim's value generates an entitlement of the victim to be treated in a certain manner. The wrongdoer violates this entitlement in treating the victim differently from what the entitlement requires. Here, Grice's account of meaning comes into play. The wrongdoer does not really diminish the victim's value, since that is

235

Hampton, Correcting harms versus righting wrongs, p. 1674. Hampton, The retributive idea, p. 125; cf. also Hampton, An expressive theory of retribution, p. 12. 237 Wollaston, The religion of nature delineated, p. 13. 236

156

impossible, but he seems to do so. His action counts as evidence that his victim is worth less than he is.238 This inappropriate treatment, as Hampton calls it as well, is exactly what constitutes "damage to the realization of a victim's value", the third point in her definition. The realization of a person's value is accomplished by fulfilling the entitlement to a certain treatment generated by the value. Thus, it is not simply the fact that the wrongdoer lied regarding the victim's value that makes his behaviour worthy of punishment, it is that, by lying in this manner, he did not treat her in accordance with her entitlements. The fourth question that is raised by Hampton's definition of a moral injury is what she means by acknowledgment of the victim's value. This point concerns the problem that a wrongdoer does not treat the victim inappropriately with every wrongful action, like, say, with an unsuccessful attempt at murder that fails so utterly that the victim does not even realize that she is a victim. In such a case, there is possibly no real damage to the realization of the victim's value, yet we do want to say that failed attempts at murder still are moral injuries and thus worthy of punishment. How can we? Here, the crucial thing is the wrong message. Failed attempts at murder may not "treat" the victims inappropriately in each case, but they always "say" that the victim's value is less than it actually is. This is an abnegation of the victim's value and that is sufficient for her to be morally injured.239 Consequently, whenever a wrongdoer conveys the message with his wrongful action that the victim is less valuable than she really is, which most often happens by treating the victim inappropriately and thus violating her entitlements generated by her value, this wrongful action is a moral injury.

238 239

Hampton, Correcting harms versus righting wrongs, p. 1674. Cf. ibid., pp. 1678, 1681.

157

b) Retribution Hampton wants to establish a link between punishment and that which is wrong in the wrongdoer's action. This has led us first to the concept of moral injury. Not all wrongful actions ought to be punished, but only moral injuries. Now that we know what a moral injury is, we have to consider two questions next: first, since a moral injury does not really touch a victim's value, but only seems to, what is so terrible about it that we should even care? Second, given that we should care, why should we link moral injuries to punishment, of all things? Why is punishment, of all things, the "appropriate response"240 to a moral injury? Hampton's reply to the first question goes as follows: The answer to the 'ought' question is straightforward: to be valuable is to be something about which human beings should care, and something for which they should care. To witness the value of someone being misrepresented is to witness behavior that ought to violate one's sense of what is appropriate for this person, and prompt a reaction that attempts to stop, and repudiate, that misrepresentation. [...] Nor does a Nietzschean attitude on the part of the victim suffice to answer the value-denial, because regardless of whether an immoral action shakes a victim's confidence in her worth, if it has damaged the realization or the acknowledgement of her value, it is morally offensive, and the representation of her diminishment becomes unacceptable. I cannot explain that offensiveness by appealing to something else, because it is the foundation of our objection to wrongdoing; it is part of what it means to say that something is valuable that we ought to care about preserving and acknowledging that value.241

Consequently, for Hampton, it is analytically true that, if something is valuable to us, the representation of this value concerns us, and that, accordingly, even the appearance of diminished value is inappropriate, unacceptable, objectionable. This, taken together with the admission that there is no further explanation than this, is the point in Hampton's theory where the fundamental criticism against retributivism itself sets in that retributivism simply begs the question. I shall return to this point later.

240 241

Ibid., p. 1685. Ibid., pp. 1683 f.

158

Assuming that moral injuries really are offensive, though, there is still the question of why we should react to it with punishment, of all things, or as far as criminals are concerned, why the state should react in such a way. First of all, the issue here is what exactly an "appropriate response" to a moral injury is. As the wrongdoer objectionably diminishes the victim's value, the appropriate response consists in the "vindication" of the victim's value by an "event" that denies the message of the wrongdoer's superior value and the victim's inferior value in a manner that asserts both of them to be equally valuable. It is this vindication that Hampton calls "retribution",242 whence the name "retributivism" for this kind of theory derives. Clearly, this event by which the victim's value is vindicated need not necessarily be punishment by the state. Hampton herself considers the most common and successful retributive response to a moral injury to be an apology of the wrongdoer to the victim.243 Exactly what kind of event is needed in order to achieve vindication is determined by the gravity of the moral injury. In cases of most serious moral injuries, like rape, an apology is woefully insufficient, and it is not enough either to simply assert that victim and wrongdoer are equally valuable, because this is "to accomplish virtually nothing". When we face actions that not merely express the message that a person is degraded relative to the wrongdoer but also try to establish that degradation, we are morally required to respond by trying to remake the world in a way that denies what the wrongdoer's events have attempted to establish, thereby lowering the wrongdoer, elevating the victim, and annulling the act of diminishment.244

Thus, if the wrongdoer's action can count as evidence of the diminished value of the victim, the response consists in denying this, and for this a more conclusive proof of the equal value of victim and wrongdoer is needed,245 and in cases of serious moral injuries this just is punishment by

242

Cf. ibid., p. 1686. Cf. ibid., pp. 1694, 1697 f. 244 Ibid., pp. 1986 f. 245 This is especially clear in Hampton, An expressive theory of retribution, p. 12. 243

159

the state. At this point as well, Hampton discloses some striking similarities to Wollaston. Thereby, the state as punishing instance does not only act as the victim's deputy who often is not able to deny the wrongdoer's evidence by herself. Maybe she sometimes is, but the important thing is not only to lower the wrongdoer, as Hampton calls it, but also to do it in a manner that establishes the equal value of wrongdoer and victim. The state's impartiality guarantees, according to Hampton, that the wrongdoer's value is not represented as lower than it really is either.246 Hence, the justification of punishment, according to Hampton, is that the wrongdoer morally injures the victim, that is, he misrepresents the victim's value as being lower than it really is which is offensive and requires a response that denies this misrepresentation and establishes the equal value of victim and wrongdoer which is often accomplished by punishing the wrongdoer.

2. Inherent difficulties Hampton's attempt to justify punishment by saying that the wrongdoer deserves to be punished in virtue of the moral injury he caused the victim has some inherent difficulties which I will only mention here without really going into them. For my purposes, it is not important whether or not Hampton can defend her theory against these objections. The first problem concerns moral injuries. According to Hampton, those wrongful actions are worthy of punishment which convey a certain message about the victim's value. Wrongful actions that do not convey any such message are still wrongful, and maybe the wrongdoer ought to compensate damage caused by them, or do something else, but he ought not to be punished for them. The problem is that there are wrongful actions that do not convey such a message, but that nevertheless the wrongdoer ought to be punished for, like, for instance, vandalism, burglary, or theft. Such things ought to be punished, but the wrongdoer hardly asserts anything about the victim's value with them. Thus, it seems as if Hampton 246

Cf. Hampton, Correcting harms versus righting wrongs, pp. 1692 ff.

160

does not adequately grasp what is wrong in wrongful actions and makes them worthy of punishment.247 Yet, even assuming that Hampton's account of moral injuries is adequate, and further assuming that the victim's value really ought to be vindicated, there is still the second problem that it is not clear that this ought to be accomplished by punishment. This, however, is exactly what is at issue: the justification of punishment. A justification is necessary, because punishing the wrongdoer means causing him such harm as, under different circumstances, would infringe on his rights. We need an account of why the circumstances are different here, and therefore, why it is alright to punish a wrongdoer. Hampton only seems to show that some kind of vindicating reaction to the wrongdoer's action is required, but not the "harsh treatment" of punishment, especially since, for Hampton, state punishments are not the exclusive, and maybe not even the major, cases of retribution.248 Thus, the criticism comes down to saying that Hampton falls short of her own aim, namely "linking retributive punishment to that which makes the wrongful action wrong".249 She can neither explain what makes wrongful actions wrong nor can she establish the link between this and punishment. This is certainly not all there is to criticize in Hampton's theory, but I want to leave it at these two major objections. Even if her theory should fail because of these problems, it does not concern my point here. My point here is that from these inherent difficulties, even if they lead to the failure of Hampton's and any other retributive theory, it does not follow that the idea underlying these theories is misguided.

247

Cf. Dolinko, Some thoughts about retributivism, p. 554; Braithwaite / Pettit, Not just deserts, p. 160; Duff, Penal communications, pp. 37 f.; Matravers, Justice and punishment, pp. 76 f.; Marshall, Harm and punishment in the community, p. 78; for Hampton's repsonse, cf. Correcting harms versus righting wrongs, p. 1680. 248 Cf. Duff, Penal communications, pp. 38 f.; Matravers, Justice and punishment, p. 78; Harward, The bitter pill of punishment: retribution, p. 200. 249 Hampton, Correcting harms versus righting wrongs, p. 1661.

161

3. Retributivism, rightly conceived The objection I am discussing in this chapter and that I have already presented at the beginning of it is aimed against the retributive enterprise itself. It is the objection that retributivists base their theories only on an intuition when they say that punishment is justified because the wrongdoer deserves it, an intuition they not only do not justify, but that they simply cannot justify. According to them, that the wrongdoer deserves punishment means nothing more than that it is right to punish him, which just begs the question. "It is right to punish the wrongdoer, because it is right" is a mere tautology and cannot serve as a justification of punishment.

a) The problem of tautology I have already indicated at which point Hampton's opponents set in with this fundamental criticism of her theory: Hampton assumes that it is inherent to the concept of a human being's value that a misrepresentation of that value is offensive, and she thinks this offensiveness to be so basic that it cannot be explained by means of anything else. It is just the misrepresentation's offensiveness that requires punishing the wrongdoer. R. A. Duff thinks that this line of argument is circular: Hampton claims that a retributive response is needed to moral injuries that damage the realization or acknowledgment of the victim's value; by understanding the concept of a 'moral injury', we are meant to understand why they require a retributive response. But she cannot then without circularity argue that what makes an action morally injurious is the fact that it would damage the acknowledgment of the victim's value if there was no retributive response to it: that would be to define moral injuries, in part, by reference to the need for a retributive response – whereas the notion of a moral injury was itself supposed to show why a retributive response is needed.250

Nevertheless, Duff's presentation of the problem is based on a misunderstanding. It is no part of the definition of a moral injury that it 250

Duff, Penal communications, p. 38; cf. Matravers, Justice and punishment, p. 78.

162

requires a retributive response; a moral injury only involves the damage to the realization or to the acknowledgment of the victim's value. Certainly, if the wrongdoer damages the victim's value, a retributive response is required, but that is not so because it is entailed in the concept of moral injury, and Hampton does not try to justify it by saying so, but rather by saying that this response is "appropriate". Even though Duff's presentation misses the point, that does not mean that the fundamental objection against retributivism in general cannot be raised against Hampton's theory in particular; it just needs to be put correctly. The objection then goes as follows: Hampton herself admits that she cannot explain why misrepresenting the victim's value is offensive by something else so she simply has to presuppose it. Because she simply presupposes this offensiveness, she can, accordingly, presuppose the desert of punishment that arises from this offensiveness. This, however, cannot just be presupposed, it needs to be justified. At this point now, we need an account of what exactly desert is in order to see that this objection, taken as a fundamental criticism against retributivism, does not go through. For that purpose, I will first "translate" Hampton's theory so as to make plain that her theory is, indeed, best understood as incorporating the very concept of desert I have expounded in this book.

b) The concept of desert in Hampton's theory This translation is fairly easy, since Hampton's vocabulary is rather similar to the one I have used in this book, as can be seen in several passages. Most importantly, Hampton does not only explicitly say that her theory is based on the idea of desert251 – that alone is not saying much –, but she goes on to present her programme like this: To understand retribution, we must link the point of the retributive response to the wrongfulness of the action.252 251 252

Cf. Hampton, Correcting harms versus righting wrongs, p. 1659. Ibid., p. 1660.

163

Here, we can understand Hampton as saying that the wrongfulness of the action is the desert base, the point of the retributive response is the desert object, while obviously, the wrongdoer is the desert-subject, and, by the fact that retribution is a "response" to the wrongfulness, as characterizing what kind of link there is between them, and that is, as characterizing the desert relation. This raises the question of whether the action's wrongfulness is a desert base in the sense of the concept of desert developed in this book. For that, the wrongdoer would have to have treated the victim unfittingly or inappropriately. According to Hampton, the wrongfulness of the action consists in the wrongdoer's lowering the victim, and she tries to illustrate what exactly she means by that: To understand what I mean, consider what we might say about the marring of some valuable artistic object. Let us suppose that someone spray painted over the pages of the Book of Kells. We would be furious and deeply angry about the damage this person had done, and that is because we take the book's value to preclude many kinds of treatment with respect to this object, including anything that would mar the artwork. So our view of the value of this object implies that only certain kinds of treatment are appropriate for it. Let us say therefore that its value generates certain entitlements. For as long as the Book of Kells has that value, it has these entitlements, which include being preserved, treated with care, and so forth.253

This is a wonderful description of what Hampton means by diminishment. It is only a bit awkward to speak of the book's value generating entitlements. I, for one, find it odd to speak of a book's being entitled to a certain treatment. Still, it is easy to understand what Hampton is aiming at, and she even uses the word herself in this passage: certain kinds of treatment of the Book of Kells are not "appropriate". This book requires to be treated in a certain manner, it is not being treated as it really is if it is not treated in this certain manner, but spray painted over. What the vandal does to the book in this example, the wrongdoer does to the victim. He treats her inappropriately. This is what explains the offensiveness of the victim's lowering, too, that Hampton explains "without 253

Ibid., p. 1674.

164

appealing to something else". The victim has a certain value, and if the wrongdoer misrepresents this value, he just does not treat the victim as it really is, that is, as a person of higher value. We can understand this perfectly well, there is no need to appeal to anything else that could explain this further. Thus, Hampton does make out a suitable desert base, the wrongdoer's inappropriate treatment of the victim. The desert object is, for Hampton, "the point of the retributive response", and that is, as we have seen, retribution as evidence of the equal value of victim and wrongdoer. The wrongdoer needs to deserve punishment, and that means that his inappropriate treatment of the victim itself makes a certain treatment of him appropriate in turn. At this point, the translation is especially easy, since none is needed at all: If we can understand the nature of this second category, then we will be in a position to isolate the element in the first category of wrongful conduct that makes a retributive response to such actions appropriate.254 The preceding analysis has isolated the ingredient in wrongful conduct that makes it the sort of conduct for which a retributive response is appropriate.255

So Hampton herself speaks of retribution as being an "appropriate response", and thus uses exactly the same terms as I have in this book. It is easily understood how retribution can be a response to the lowering of the victim: the wrongdoer has said something false and that is put right again. Moreover, it is clear that by this response's being appropriate, Hampton means that it is required. Consequently, Hampton's theory is based on the very concept of desert I have expounded in this book. "The wrongdoer deserves punishment in virtue of his lowering the victim" accordingly means "In lowering the victim, the wrongdoer has treated her inappropriately which in turn makes it appropriate that he be punished."

254 255

Ibid., p. 1664. Ibid., p. 1685.

165

c) Conclusion This exemplary "translation" of Hampton's theory shows that the fundamental objection against retributivism is unfounded. The objection that the basic retributive intuition is a mere tautology and thus cannot serve as a justification of punishment rests on a misunderstanding of the concept of desert retributivism draws on. To say that the wrongdoer's inappropriate treatment of the victim makes punishing the offender appropriate, clearly is not merely reiterating that punishing the wrongdoer is right. Certainly, the concept of propriety is a normative concept, but normativity has to come into play at some point, after all, otherwise we could never say that we ought to punish. This alone cannot be enough to object that retributivism begs the question. Analysing Hampton's theory as I have done in the last section evidently is not adding anything substantial to it. The translation is meant only to clarify what she is saying. Substantiating what exactly is meant by the wrongdoer's inappropriate treatment of the victim, and what the corresponding propriety of punishment consists in, is completely Hampton's business, not mine. Maybe she does not do justice to this task, maybe she does not substantiate these points adequately. If that is so, these are problems inherent to her own particular theory, not problems concerning the retributive enterprise itself. So maybe Hampton's theory fails because of its inherent problems mentioned above, and maybe any current retributive theory of punishment fails because of similar problems. The point of this chapter was not to defend a particular retributive theory, but rather to show with the help of the concept of desert that retributivism at least does not fail due to analytic reasons, but can work in principle, and in fact, that the concept of desert can serve to justify punishment.

II. Explaining moral residue In the discussion on moral dilemmas, one argument that is put forth for the existence of such dilemmas is that there is moral residue in some cases and the only explanation of this moral residue is simply the existence of moral dilemmas.256 An example of Judith Jarvis Thomson's illustrates this:257 I promise Jamie to give her a banana today. Later, I promise Kody to give her a banana today, too. To my dismay, though, I find that I only have one banana and that, therefore, I cannot keep both promises. What ought I to do now? Let us say that the fact that I promised the banana to Jamie first determines that I ought to give the one banana I have to her. In this situation there are two moral requirements I am subject to, but that are such that I cannot fulfil both. That is why this kind of situation is called a moral dilemma, even when, as in the present case, it is clearly determined which of the conflicting requirements I ought to fulfil.258 Now, what is moral residue? Although in our situation it is clear which promise I ought to keep, namely the promise to Jamie, there is still Kody there, and I have given her a promise too, after all. The point is that giving the banana to Jamie cannot be all I ought to do in this situation. I cannot give Kody a banana, but I still ought to do something, like asking Kody beforehand to release me from my promise, or explaining to her afterwards why I broke it, or apologizing to her, or compensating her, or feeling sorry, or regretting it, or feeling guilty. Different authors favour different things from this list as moral residue. It is not important to me in this context what exactly moral residue is, nor, for instance, whether or not it can consist of feelings, or whether some feelings like regret can and others like feelings of guilt cannot constitute moral residue.

256

Cf. Williams, Ethical consistency; Marcus, Moral dilemmas and consistency; Sinnott-Armstrong, Moral dilemmas, p. 44. 257 Cf. Thomson, The realm of rights, p. 82. 258 Cf. Foot, Moral realism and moral dilemma, p. 38.

168

What is important to me here is the explanation of this moral residue: whatever I ought to do regarding Kody, why ought I to do it? I shall concentrate on this question alone, and in the process, I may not do justice to the discussion on moral dilemmas as it is led. Thus, I do not care whether the above-mentioned situation really is a moral dilemma at all, and consequently, in what follows, I will speak only of "conflict situation". I may turn some lines of argument upside down as well: some authors want to take the existence of moral residue in some cases as an explanation of other things, like for instance, that there even are moral dilemmas, whereas I want to know what explains moral residue itself. Hence, I will not mind whether the explanation I offer for moral residue is suitable for the purposes of the champions of moral dilemmas. Finally, I will simply assume that there is something I ought to do regarding Kody in the conflict situation, although that manifestly does not go without saying. In short, I shall concentrate on showing that, given that there is moral residue in conflict situations, the explanations of it offered so far are not adequate, and that the best explanation is that I ought to do something regarding Kody in the conflict situation, because she deserves it.

1. The persisting Ought Bernard Williams in his paper "Ethical consistency" (1965) and after him Ruth Barcan Marcus in "Moral dilemmas and consistency" (1980) were two of the first authors ever to consider the problem of moral residue. The starting point of their theory is phenomenological: even if we assume that Agamemnon made the right decision when he had to choose between saving his campaign and saving his daughter's life, that is, even if it had been the case that he ought to have preferred the campaign, it was not irrational of him to mourn for Iphigeneia. Quite on the contrary, we want to say that it was decent of him to regret that he had to sacrifice Iphigeneia, this is what admirable moral agents tend to do.259

259

Cf. Williams, Ethical consistency, pp. 172 f.

169

Therefore, finding out what we ought to do in a conflict situation and doing it cannot be the end of the story; this does not fit the moral facts. Rather, in this kind of situation there is still moral residue, like Agamemnon's regret.260

a) Feelings Both Williams and Marcus primarily consider feelings as moral residue, namely regret and feelings of guilt, respectively. This does not apply only to tragic cases such as Agamemnon's, of course, but to me in our example as well: I ought to regret that I broke my promise to Kody or I ought to feel guilty about it. At this point, however, there is protest against Williams and Marcus's account, and it comes, interestingly, from other proponents of moral residue in conflict situations, like Philippa Foot and Judith Jarvis Thomson. The problem they rightly see is that, although in every conflict situation there is moral residue, feelings of regret or guilt are not called for in every conflict situation. Maybe I do no damage to Kody when I break my promise to her, because, say, she got a banana today from somebody else, or because she has a banana plantation in her backyard anyway and does not care for one banana more or less, or because, when she goes to buy a banana in the supermarket, it so happens that she is the 100.000th customer and wins 100.000 euros.261 The point is not that in such circumstances there is no moral residue at all so that I do not have to do anything further regarding Kody. The point is rather that in such circumstances I really ought not to feel regret or guilt about breaking my promise; on the contrary, maybe I ought to rejoice instead that in virtue of my broken promise Kody got so lucky and won so much money. In other words, we need a concept of moral residue such that it obtains in every conflict situation. To consider only particular feelings as 260

Cf. Marcus, Moral dilemmas and consistency, pp. 131 f., 137; Williams, Ethical consistency, p. 175. 261 Cf. Foot, Moral dilemmas revisited, pp. 176 f., 183; Thomson, The realm of rights, p. 96.

170

moral residue does not do justice to this fact, because it is not true that we ought to have these feelings in every conflict situation. This is a real problem for Williams and Marcus, although I only mention it and put it aside immediately. As I have said, it is not important to my purposes here which kinds of moral residue there are, but only what the explanation of moral residue in general is, whatever it may consist of.262 There is nothing to be said against applying the explanation Williams and Marcus offer for the feelings of regret and guilt they think we ought to have in conflict situations to moral residue in general. I think it cannot be disputed that one kind of moral residue in our example is that I ought to apologize to Kody for breaking my promise, and in what follows I will consider this as the moral residue in our situation.

b) Giving up the agglomeration principle So, according to Williams and Marcus, why ought I to apologize to Kody for breaking my promise to her? My promise to Jamie implies that I ought to keep that promise. Likewise, my promise to Kody implies that I ought to keep that promise. In fact, however, I can only keep one of these promises, and as our story goes, I ought to keep my promise to Jamie. Williams and Marcus's point now is that it is nonetheless still true that I ought to have kept my promise to Kody so that this Ought persists, as it were, and this fact of the persisting Ought is what explains why I ought to apologize to Kody.263 This Ought not acted upon is, in Williams's metaphor, a "moral impulse" that needs to find a new object,264 like, in our case, the apology to Kody.

262

In reaction to this difficulty in Williams and Marcus, Foot, Moral realism and moral dilemma, tries to develop an account of moral residue that is based on reasons for actions; her explanation of residue, nonetheless, is roughly the same as Williams and Marcus's – cf. p. 42 –, thus I will not go into it here. 263 Cf. Williams, Ethical consistency, p. 179; Marcus, Moral dilemmas and consistency, pp. 131 f. 264 Williams, Ethical consistency, p. 172.

171

Yet, how can Williams and Marcus say that this Ought toward Kody persists? How can I ought to do something I cannot do? Thomson has put this problem very pointedly: It is an odd idea. I will certainly feel you have been unhelpful if when I tell you about my predicament, and ask what I ought to do, you tell me 'Well, as a matter of fact, you ought to give C a banana and you ought to give D a banana.' I just told you I have only one banana.265

Williams and Marcus's point is that, although I cannot keep both promises taken together, and therefore, I ought not to keep both promises taken together, it is still true, at least before I give away my banana, that I can keep each promise taken by itself, and therefore, that I ought to keep each promise taken by itself. Hence, both deny what Williams calls the agglomeration principle. This principle states that from the fact that I ought to do α, and from the fact that I ought to do β, it follows that I ought to do both α and β. From this principle, together with the principle that Ought implies Can, in turn it follows that if I cannot do both α and β, it is not the case that I ought to do both α and β, from which again it follows that I ought not to do α or β.266 We can easily see why this is an unwelcome conclusion for Williams and Marcus. If both the agglomeration principle and the principle that Ought implies Can were valid, the case in our example would stand like this: I cannot keep both promises; therefore I ought not to keep both promises; since I ought to keep my promise to Jamie, it follows that I ought not to keep my promise to Kody. Yet, if I ought not to keep my promise to Kody, then there is nothing left that explains why I ought to apologize to her. But I ought to. Consequently, Williams and Marcus deny the agglomeration principle. Nonetheless, they cannot leave it at that, because as soon as I have given the banana to Jamie it is no longer true of the promise to Kody, even taken by itself, that I can keep it. How can they now still say that I ought to keep it? 265 266

Thomson, The realm of rights, p. 83. Cf. Williams, Ethical consistency, pp. 180 ff.; Marcus, Moral dilemmas and consistency, p. 139.

172

As I have tried to argue throughout, it is surely falsifying of moral thought to represent its logic as demanding that in a conflict situation one of the conflicting ought's must be totally rejected. One must, certainly, be rejected in the sense that not both can be acted upon [...]. But this does not mean they do not both (actually) apply to the situation; or that I was in some way mistaken in thinking that these were both things that I ought to do. I may continue to think this retrospectively, and hence have regrets; and I may even do this when I have found some moral reason for acting on one in preference to the other. For while there are some cases in which finding a moral reason for preference does cancel one of the ought's this is not always so. I may use some emergency provision [...] which deals with the conflict of choice, and gives me a way of 'acting for the best'; but this is not the same as to revise or reconsider the reasons for the original ought's, nor does it provide me with the reflexion 'If I had thought of that in the first place, there need have been no conflict'. It seems to me impossible, then, to rest content with a logical picture which makes it a necessary consequence of conflict that one ought must be totally rejected in the sense that one becomes convinced that it did not actually apply.267

So the point is this: now that I have given the banana to Jamie it may no longer be the case that I ought to give the banana to Kody, because I can no longer do so. However, as long as I could have done so I ought to have done so, even though it was already true at that time as well that I ought to have given the banana to Jamie. The statement "I ought to give the banana to Kody" was true in the conflict situation, and now that the conflict is resolved the adjusted statement "I ought to have given the banana to Jamie" is still true, and in this sense the Ought toward Kody persists, even though I cannot act upon it any longer. Since the Ought toward Kody persists, but I cannot act upon it any longer, I ought to do something else instead, namely apologize to Kody. This apology is the new object of the moral impulse I still have after the conflict situation has been resolved.

267

Williams, Ethical consistency, pp. 183 f.

173

c) Conclusion One objection that is raised against Williams and Marcus is that solely because people, even admirable people, have certain feelings in certain situations, it does not follow that they ought to have these feelings.268 This objection is really rather astonishing, since it is so evidently misguided. It is certainly true that Williams in particular bases his theory on the intuition that in particular situations people react in a particular manner, but he does not leave it at that by any means. His giving up the agglomeration principle is exactly meant to justify these feelings: decent people do not simply have certain feelings in conflict situations, but they are justified in this, because in these situations the Ought that is not acted upon persists. Obviously, this does not touch the objection that these particular feelings are not necessarily justified in every conflict situation, but as I have already said, there is nothing against applying Williams and Marcus's explanation to moral residue in general that obtains in every conflict situation. Still, giving up the agglomeration principle is a rather serious deviation from commonly accepted deontic logic. Many things in deontic logic are controversial, but the agglomeration principle is not one of them, so that giving it up has to be justified itself. Marcus is rather cavalier about it. She simply gives it up and notes: Such a claim is of course a departure from familiar systems of deontic logic.269

By contrast, Williams acknowledges that this really is a problem for him he has to face, and he replies by taking the offensive: Now the mere existence of such cases is obviously not enough to persuade anyone to give up agglomeration of ought [...]. I do not want to claim, however, that I have some knock-down disproof of the agglomeration principle; I want to claim only that it is not a self-evident datum of the logic of ought, and that if a 268

Cf. Foot, Moral realism and moral dilemma, pp. 40 f.; Sinnott-Armstrong, Moral dilemmas, p. 46. 269 Marcus, Moral dilemmas and consistency, p. 139.

174

more realistic picture of moral thought emerges from abandoning it, we should have no qualms in abandoning it. We can in fact see the problem the other way round: the very fact that there can be two things, each of which I ought to do and each of which I can do, but of which I cannot do both, shows the weakness of the agglomeration principle.270

This would be an acceptable course to take indeed. If we assume that there really is moral residue in conflict situations, and if abandoning the agglomeration principle in fact is the only explanation we have of this moral residue, then this is a reason for abandoning it. As I have said, I am assuming here that there is, in fact, moral residue in conflict situations. Nevertheless, a persisting Ought is not the only possible explanation for the fact that I ought to apologize to Kody. There are other explanations, and if these can do without denying the agglomeration principle, that will be something that counts in their favour against Williams and Marcus's account, exactly because the agglomeration principle is otherwise uncontroversial. There is yet another point against Williams and Marcus's explanation of moral residue that is indicated by Thomson: Why should [moral residue] not be explained by appeal to the fact – it remains a fact – that I broke the second promise? Why should we think we can pass from the fact of the broken promise to the fact of the later moral residue only by way of an intermediary fact to the effect that I ought to have kept the promise? You would think this true only if you thought that the broken promise could have an impact on my later thoughts and feelings only if I ought to have kept it. But why should we think that so?271

Thomson's criticism comes down to saying that an explanation is the better the less it has to presuppose, and Williams and Marcus's explanation presupposes an "intermediary fact" which we do not yet see that we should need. If there was an explanation of moral residue that manages without presupposing any intermediary facts, that explanation would be better than Williams and Marcus's for it. 270 271

Williams, Ethical consistency, p. 182. Thomson, The realm of rights, p. 85.

175

So we are looking for an explanation of moral residue in conflict situations that conforms to the agglomeration principle and needs no intermediary facts. Since it is Thomson who raises this last objection against Williams and Marcus it seems reasonable to look to her for such an explanation.

2. Claims In contrast to Williams and Marcus, Thomson does not think that the Ought not acted upon in conflict situations persists. Since she does not believe that, her explanation of moral residue leaves the agglomeration principle untouched so that it meets the first condition of being a better explanation than Williams and Marcus's. As Thomson herself initially demands to get directly from the fact of my broken promise to the fact of the moral residue, it would appear that her own explanation meets this demand, but this is not so. In the paragraph immediately following the one quoted in the last section in which she criticizes Williams and Marcus she writes: I think we do think we can pass from the fact of the broken promise to the fact of the later moral residue only by way of some intermediary fact. 'You broke a promise. So what? How does that fact yield the moral residue if it is not the case that you ought to have kept it?' One answer, which I think very plausible indeed, is this: in making a promise one gives a claim, and breaking a promise is therefore failing to accord a claim, and that fact explains the moral residue – for a claim is equivalent to a constraint on the claim-giver's behavior that includes such things as that the claim-giver may have to make amends later if he or she does not accord the claim.272

So, according to Thomson, the key to the explanation of my apology to Kody lies in the fact that Kody has a claim against me in virtue of my promise to her that I did not accord. Because I did not accord Kody's claim against me, I ought to apologize to her.

272

Ibid., p. 85.

176

The first question here is what Thomson understands by a claim. Thomson bases her theory essentially on Wesley Newcomb Hohfeld's concept of a claim. However, she does not agree with him on a few points and modifies his concept accordingly. Hence, Thomson does not exactly adopt Hohfeld's concept of a claim, but his influence on Thomson's theory is enormous, and we will not understand her theory without knowing about Hohfeld's concept, according to which, in Thomson's terminology, claims are absolute.

a) Hohfeld on absolute claims In his paper "Some fundamental legal conceptions as applied in judicial reasoning" (1913), Hohfeld does not strictly define legal relations in general nor claims in particular, but rather he tries to show by means of a kind of "map" what each legal relation is.273 Legal relations are relations individuals stand to each other in by law, as, for instance, when one person has a claim against another. This map is such that it represents the legal relations as "correlatives" and "opposites". Since each correlative and each opposite to each legal relation is a legal relation itself, and thus each legal relation is illustrated only by means of showing its connections to other relations that are illustrated in turn by their connections to the first relations, among others, this procedure is circular and does not yield real definitions, as Hohfeld readily admits. Nevertheless, even showing the connections between relations is illuminating insofar as it is a major gain in systematic knowledge, and thus Hohfeld's map is instructive. Now, what are correlatives and opposites? Hohfeld explicitly states that correlatives are equivalent relations,274 that is, if and only if a person X stands in a legal relation to another person Y, Y stands in the correlation to X, so that the legal relation of Y to X is the correlative of the legal relation of X to Y. It stands to reason that opposites are complementary relations, 273

Hohfeld, Some fundamental legal conceptions as applied in judicial reasoning I, p. 36. 274 Ibid., p. 38, cf. also p. 39.

177

that is, if and only if X stands in a legal relation to Y, then X does not stand in the opposite relation to Y. The correlative to the claim relation is a duty. A claim's opposite has no real name, so Hohfeld simply calls it "no-right".275 The trouble with the no-right is that it is not a real legal relation at all, but simply there in order to tidy up Hohfeld's map. As such, it does serve an important purpose, to be sure, since Hohfeld can explain a few things on his map better with it than without it. As far as the explanation of a claim is concerned, though, we may safely disregard it, since a no-right is nothing more than a negated claim, and saying "If X has a claim against Y it is not the case that X has no claim against Y" does not really help. That leaves us duty, the claim's correlative, to answer the question what a claim is with. It would appear that referring to the correlative does not help either, because it will just lead us in a circle: the answer to the question "What is a claim?", we learn from Hohfeld's map, is "The correlative to a duty"; but the answer to the question "What is a duty?" simply is "The correlative to a claim", which obviously explains nothing at all. This is what Hohfeld means when he says that his paper yields only a map and no real definitions of legal relations. Yet, on this point exactly Hohfeld does offer something more when he quotes a legal case and evidently agrees with it: A duty or a legal obligation is that which one ought or ought not to do.276

At first sight, this is a curious way of putting this point, since it appears as if a duty has nothing at all to do with what one ought to do: one ought to do one's duty, or not. Hohfeld means, of course, that a duty is that which we ought to do, and this "do" is meant in a sufficiently wide sense that it comprises omissions as well. So how exactly are we to understand the "is" in "A duty is that which we ought to do"? This "is" cannot mean that "to have a duty" and "ought to do something" are equivalent relations, because not always when somebody ought to do something does somebody have a claim to its being 275 276

Ibid., pp. 36, 39. Ibid., p. 38.

178

done, and hence, because of the equivalence of claim and duty, one does not always have a duty to do what one ought to do. Therefore, the existence of a duty is not equivalent to, but only implies, that one ought to do what one has a duty to do. This addition is helpful, because it helps us avoid the circle in the explanation of both terms in question. Now we can say: if X has a claim to something against Y, this means, according to Hohfeld, that Y has a duty toward X to accord it, and this in turn means that Y ought to accord it to X. Or in short: if X has a claim to something against Y, then Y ought to do it. This is why Thomson calls Hohfeld's concept of claims "absolute".

b) Thomson on non-absolute claims As I have already said, Thomson essentially adopts Hohfeld's concept of a claim, yet disagrees with him on two points. That is because, for Thomson, Hohfeld's concept is too narrow in one respect, and too wide in another, and she tries to improve on it in these respects. According to Thomson, Hohfeld's concept is too narrow in that he limits it to legal claims, that is, to claims that are grounded in law. Thomson shows, though, that there is no reason not to apply Hohfeld's ideas to claims grounded in morality as well.277 On the other hand, Hohfeld's concept is too wide for Thomson in that, according to it, a claim's obtaining against Y that Y do p implies that Y ought to do p. Or in Thomson's terminology: the problem with Hohfeld's concept is that, according to it, claims are absolute.278 Thomson disagrees, and she does so, precisely because there is moral residue in conflict situations. The banana example is hers, after all, and it is intended to show that I ought not to do something regarding Kody even though Kody has a claim against me, and therefore, it is intended to show that claims are not absolute. I ought not to give the banana to Kody, because I cannot give the banana to Kody, since I gave it to Jamie, as I ought to have done. Nonetheless, according to Thomson, Kody still has a 277 278

Thomson, The realm of rights, pp. 73 - 77. Cf. ibid., pp. 79 f.

179

claim against me that I ought not to accord, and this is in direct opposition to Hohfeld: for him, there is no such thing as a claim that ought not to be accorded, because a claim just implies that one ought to do what the claim is to. Thomson disagrees with Hohfeld on this point, because the claim against me that I ought not to accord still serves an important purpose on her account: the fact that it still persists and that I do not accord it explains why I ought to apologize to Kody. This point leads Thomson to understand the concept of a claim to be less strict than Hohfeld does. The existence of moral residue shows, according to Thomson, that a claim cannot imply that the "claimee" ought to accord the claim. Still, that does not mean that Thomson is so far opposed to Hohfeld that she thinks that a claim's obtaining and that which the claimee ought to do do not ever have anything to do with each other. Often we ought to do what there is a claim to, and exactly for the reason that there is a claim to it,279 just not always. Sometimes there is a reason against doing what somebody has a claim to. Thus, Thomson says that to have a claim against somebody to something means that this somebody ought to accord the claim unless there is a reason for him not to.280 Evidently, she cannot leave it at that, as without specifying this proviso this just means that we sometimes ought to accord claims and sometimes we ought not to, which really comes down to saying that a claim's obtaining does not tell us anything at all about whether or not we ought to accord it. Thomson defines the circumstances which can count as a reason against according a claim by means of her "Tradeoff Idea": The Tradeoff Idea: It is permissible to infringe a claim if and only if infringing it would be sufficiently much better for those for whom infringing it would be good than not infringing it would be for the claim holder.281

In other words: according the claim is to the claimant's advantage; infringing the claim is to other people's advantage; it is permissible to infringe a claim if the latter advantage is sufficiently greater than the first 279

Ibid., pp. 97 f. Ibid., p. 149. 281 Ibid., p. 153. 280

180

advantage. How much greater does the latter advantage have to be in order to be sufficiently greater? Thomson does not answer that question.282 She does give us something like a rule of thumb, though. The more stringent the claim is, the larger the "increment of good" has to be for infringing the claim to be permissible. How stringent a claim is is determined by the "Aggravation Principle". The Aggravation Principle: If X has a claim against Y that Y do alpha, then the worse Y makes things for X if Y fails to do alpha the more stringent X's claim against Y that Y do alpha[.]283

Therefore, according to Thomson, there is the following connection between a claim's obtaining and what we ought to do: the claimee ought to accord the claim, unless the advantage yielded to other people by infringing the claim would be sufficiently greater than the advantage yielded to the claimant by according the claim; and how much greater the advantage has to be depends on how stringent the claim is which in turn depends on how bad it would be for the claimant if the claim is not accorded.

c) Trolleys and transplants The most prominent problem with Thomson's account is that her consequentialist "Tradeoff Idea" cannot do justice to all cases. Since this problem is so prominent, I will only mention it here without going into it in any detail, as it has been sufficiently discussed elsewhere,284 and, more importantly, it really does not matter one way or the other for the purposes of this chapter. 282

Ibid., p. 153. Ibid., p. 154. 284 Cf. Gert, Transplants and trolleys; Mack, Of transplants and trolleys; Fisher / Ravizza, Thomson and the trolley; Gorr, Thomson and the trolley problem; Norcross, Rights violations and distributive constraints; for an opposing view, cf. Thomson, Reply to critics; Thomson, Reply to commentators; Rosenberg, Contractarianism and the "Trolley" problem. 283

181

The problem concerns two cases Thomson discusses and which Thomson's opponents regard as sufficiently similar to each other. Nevertheless, Thomson treats both cases in a different manner which, according to her opponents, in view of their similarities, she cannot justify within the scope of her own theory. These two cases are the Trolley Problem and the Transplant Problem. In the Trolley Problem, a trolley is out of control; the only thing that can be done is to throw a switch. If the switch is not thrown, the trolley will hit and kill five persons; if it is thrown, it will kill one. Thomson thinks it justified to throw the switch such that only one person is killed.285 The Transplant Problem is about one healthy person and five persons that each need a different organ in order to survive. If the necessary organs are removed from the healthy person and transplanted to the five patients, the healthy person will die and the five patients will survive. If this is not done, the healthy person will live and the five patients will die. In this case, Thomson thinks it is not justified to save the patients at the expense of the healthy person's life.286 Thomson's opponents share her intuitions about how we have to assess these two cases, but they do not see how Thomson can say what she does about them within the scope of her own theory. With this, I will put aside this problem I only wanted to mention, since it is not important in this context.

d) Claims and moral residue Now that we know how Thomson understands claims we can return to our starting question: why ought I to apologize to Kody for breaking my promise to her? The story Thomson tells us about it is not so very different in its outlines from Williams and Marcus's story, except for the manifest and crucial difference that it is not the fact that I ought to have given the banana to Kody, but rather the fact that Kody has a claim against me to a 285 286

Thomson, The realm of rights, pp. 176 ff. Ibid., pp. 135 ff.

182

banana that explains the course of the story. Thomson's story thus goes like this: My promising a banana to Jamie generates a claim of Jamie's against me that I give her a banana today. Likewise, my promising a banana to Kody generates a claim of Kody's against me that I give her a banana today. If we assume for a moment, with Hohfeld, that claims are absolute, this story would continue as follows: that there is a claim against me implies that I ought to do what the claim is to. Together with the principle that Ought implies Can, this means that if it is the case that I ought to give the one banana that I have to Jamie, and that, therefore, I cannot give a banana to Kody, then it is the case that I ought not to give the banana to Kody. From this it follows that Kody has no claim against me to a banana. But if Kody has no claim against me to a banana, this would have to be the end of the story. I do what I ought to when I give the banana to Jamie because she has a claim against me, and I would not have to do anything more, if it were true that claims are absolute. Yet, this cannot be the end of the story, as evidently I still ought to do something: I ought to apologize to Kody. There is nothing left to explain why I ought to apologize if Kody simply has no claim against me, according to Thomson. For her, a better explanation is to say that Kody does, in fact, have a claim against me to a banana, even though I ought not to give her the banana, and the fact that I do not accord a claim explains why I ought to do something else, namely to apologize.287

e) Conclusion The first of two problems with Thomson's explanation of moral residue is that it does not work. The fact that I ought not to do something cannot explain why I ought to do something entirely different instead that has nothing at all to do with what I ought not to do.

287

Ibid., pp. 91 f., 93, 100.

183

According to Thomson, Kody has a claim against me to a banana that I ought not to accord. Still, she only has a simple claim to a banana, not a complex claim to a banana or, if she does not get a banana from me, an apology. Thus, it is not at all clear how this simple claim of Kody's against me to a banana could possibly explain why I ought to apologize to her. A claim of Kody's against me to a car might just as well explain why I ought to give her a lollipop. Now, Thomson does not actually say that it is Kody's claim against me to a banana that explains why I ought to apologize, but rather the fact that I do not accord it. However, that does not help much, since the nonaccordance of a claim to a banana can no more explain why I ought to apologize than the claim itself, particularly since, according to Thomson, I ought not to accord the claim in the first place. Maybe Thomson could reply that my not according the claim to a banana itself generates a new claim, namely a claim of Kody's against me to an apology. Then, Kody would have, first, a claim against me to a banana I ought not to accord, and if I do not accord it then this generates, second, a claim of Kody's against me to an apology, and if Kody has a claim against me to an apology, I ought to apologize – unless I ought not to, of course, which would presumably lead to a third claim's being generated. I, for one, do not see why not according a claim that I ought not to accord should generate a new claim to an apology, but even if this manoeuvre worked for Thomson and she could offer us a working explanation of moral residue, at this point at the latest the second problem with her account arises. We were looking for an explanation that did not need to presuppose any "intermediary facts" like Williams and Marcus's, and as far as this point is concerned, Thomson's account is even worse off than theirs: they had to presuppose one intermediary Ought, Thomson has to presuppose at least two intermediary claims. In view of this, the question still is whether there is not a better explanation of moral residue available, and I suggest that the best explanation of moral residue is the fact that Kody deserves an apology from me.

184

By the bye, if Thomson's explanation of moral residue does not pass, as I have argued, then her argument against Hohfeld's claim that claims are absolute falls as well. As I have already suggested, her argument is based on her assumption that moral residue simply cannot be explained in any other way than by presupposing the existence of claims that ought not to be accorded. Yet, if moral residue can, in fact, be explained in another way, as I shall show in the next section, then at least Thomson's reason for assuming claims to be non-absolute must be dropped.

3. Desert Why ought I to apologize to Kody for breaking my promise to her to give her a banana today? Well, because she deserves that I apologize to her for that. First, then, for Kody to deserve an apology from me there has to be a suitable desert base, that is, at least a relation of fittingness, or unfittingness, as it were. I have given a promise to Kody. Then I broke this promise. These two facts do not fit each other. Breaking a promise is not a response to having given a promise. Maybe there is even impropriety in this case. Maybe we can say that I do not treat Kody as she really is, namely as somebody I have promised something to. It does not really matter, though, whether we can reasonably say this. For desert, the only thing that matters is that the desert base is something fitting or unfitting, not that, beyond that, it is something appropriate or inappropriate, and this is given here. The fact that I have done something unfitting in breaking my promise to Kody in turn makes certain behaviour toward her appropriate. It is now appropriate that I apologize to her. Maybe other forms of behaviour are appropriate as well, maybe I ought to compensate her, because she could not profitably resell the banana as she had planned, or maybe I ought to feel guilty, because now she has to starve. What exactly the moral residue is depends entirely on the particular circumstances of the case, yet, whatever I ought to do regarding Kody, I ought to do it, because she deserves it.

185

This is a simple and straightforward explanation of why I ought to apologize to Kody, and it avoids both the problems faced by the other two explanations discussed. Here, no Ought persists, and therefore the agglomeration principle is left untouched. Maybe this could be thought controversial insofar as it could be said that breaking my promise to Kody is not only unfitting, but inappropriate, after all, and that this implies that I ought not to have treated her like this. If that were so, does my explanation not come down to something like Williams and Marcus's? No, it does not, as for desert it is not important that breaking the promise was inappropriate; the only thing that matters is that it is unfitting. Unfittingness does not imply that I ought to have kept the promise, just as little as the unfittingness of socks to sandals implies that one ought not to wear both socks and sandals. Desert just is not second-order propriety as Price would have it, then this objection would be justified, but rather it is propriety that is based on fittingness, and hence, this objection does not go through. Regarding the second problem that an explanation is the better the less it has to presuppose, my account fares better than the other two as well, because it does not need to presuppose any "intermediary facts". The intermediary fact in Williams and Marcus's explanation was that the Ought not acted upon in the conflict situation persists as a moral impulse. The fact that I have given Kody a promise accordingly leads to the existence of an impulse which in turn explains why I ought to apologize. In Thomson's account, there were intermediary claims generated in conflict situations: my promise generates a claim of Kody's against me I ought not to accord, and this claim explains why I ought to apologize. By contrast, no intermediary fact is needed if we just say that I ought to apologize to Kody, because she deserves it. Here, there are only two facts, first, the fact that I broke my promise, and second, the fact of my apology. These facts have certain properties, namely breaking the promise is unfitting to having given the promise, and the apology is appropriate to this unfittingness.

186

These properties are certainly an independent element in the explanation. Nonetheless, in contrast to the persisting Ought and the generated claims, they are not "intermediary facts". We cannot reasonably paraphrase this explanation, saying that my promise to Kody leads to something which in turn explains why I ought to apologize to her, and the properties of unfittingness and propriety are somehow built into this "something". Rather, breaking my promise is itself unfitting, and that leads directly to the fact that I ought to apologize to Kody, because that itself is appropriate. Maybe it could be objected that it depends on how this "something" is described whether or not this explanation really needs less facts than the other two. Maybe we cannot make an intermediary step in the explanation out of propriety, but then that might just be a linguistic peculiarity of that concept. Things could be different if we replace propriety with desert. When we speak of desert rather than of propriety it does seem as though there were an intermediary fact, after all: my breaking the promise leads to Kody's deserving an apology from me, and this desert of Kody's in turn explains why I ought to apologize to Kody. This explanation thus seems to run just parallel to the other two accounts, and it is not quite clear why it should be better in this regard. It may look as if this explanation runs parallel to the other two accounts, but in fact, it does not, because to deserve something means nothing other than that one ought to get it because that is appropriate. Consequently, there is no linguistic peculiary in propriety that makes it seem like we can make do with two facts when indeed we need three, but rather it is in desert that makes it seem like we need three facts where there are only two. We cannot meaningfully replace the definiens of desert in the offered alternative explanation; we cannot meaningfully say: my breaking the promise leads to the fact that I ought to apologize to Kody, because that is appropriate, which in turn explains why I ought to apologize to Kody. Therefore, in explaining moral residue by desert it only seems as if an intermediary fact was needed, but that is not so. There are, of course, metaphysically expensive assumptions we have to make for this explanation, as I have shown in Chapter VI of Part 2, so I do

187

not pretend that there is nothing at all that could be said against this explanation of moral residue. In view of these high metaphysical costs, some people might be inclined to regard the concept of desert in a quasiAnscombely way as a relic from a time when everybody still believed in such metaphysics, a relic that nowadays, however, has become untenable and consequently must be rejected.288 Nevertheless, the explanation of moral residue I have offered in this section is the best there is, in view of the fact that Williams and Marcus have to deny the agglomeration principle, and Thomson has to postulate claims of which it is quite doubtful that they can really serve as an explanation. Thus, like Williams did for his, I will take the offensive for my explanation and say: to the question of why I ought to apologize to Kody in our example of a conflict situation the answer "because she deserves it" is the best answer there is. The fact that this is the best answer there is is a reason to accept it along with everything it presupposes. Furthermore, the fact that this answer is the best there is indicates that conceptions of morality that try to make do with the normative concepts of obligation, claims, and duties alone are incomplete, since they cannot do justice to cases like this conflict situation, and it consequently shows that a place in morality is due to the concept of desert. Moreover, we need to take into account that the concept of desert is still predominant and pervasive, even though it seemed discredited in philosophical discussion for decades after Rawls, and in view of this fact, simply abandoning the concept might be just as expensive. An explanation for the persistence of the concept of desert against all odds we find in Kant: we human beings simply work that way that propriety is important to us, and that we must hope for the world to be in balance.

288

Cf. Anscombe, Modern moral philosophy.

189

Bibliography Anscombe, G. E. M.: "Comment", in: Stephan Körner (ed.), Practical Reason, Oxford: Blackwell, 1974, pp. 17 - 21. ––- : "Modern moral philosophy", first published in: Philosophy 33, 1958, pp. 1 - 19; quoted from: G. E. M. Anscombe, The collected philosophical papers of G. E. M. Anscombe. Volume III: Ethics, religion and politics, Oxford: Blackwell, pp. 26 – 42. Arnold, N. Scott: "Why profits are deserved", in: Ethics 97, 1987, pp. 387 - 402. Austin, John L.: How to do things with words, Oxford: Clarendon Press, 1962. ––– : "Other minds, Part II", in: Proceedings of the Aristotelian Society, Supp. 20, 1946, pp. 148 - 187. Auxter, Thomas: "The unimportance of Kant's highest good", in: Journal of the History of Philosophy 17, 1979, pp. 121 - 134. Beck, Lewis White: A commentary on Kant's Critique of practical reason, Chicago: University of Chicago Press, 1966. Benn, S. I.: "An approach to the problems of punishment", in: Philosophy 33, 1958, pp. 325 - 341. Bernstein, Leonard: "How a great symphony was written. Leonard Bernstein talks about the first movement of Symphony No. 5", on: Ludwig van Beethoven / Leonard Bernstein / New York Philharmonic: Symphony No. 5 in C minor, Op. 67, The royal edition, Sony Classical GmbH, 1992. Bittner, Rüdiger: Doing things for reasons, Oxford: Oxford University Press, 2001. Braithwaite, John / Pettit, Philip: Not just deserts. A republican theory of criminal justice, Oxford: Clarendon Press, 1992. Broome, John: "Normative requirements", in: Jonathan Dancy (ed.), Normativity, Oxford: Blackwell, 2000, pp. 78 - 99. Brown, Charlotte R.: "William Wollaston", in: Edward Craig (ed.), Routledge encyclopedia of philosophy CD-ROM. Version 1.0, London: Routledge, 1998. Chisholm, Roderick Milton: "The ethics of requirement", in: American Philosophical Quarterly 1, 1964, pp. 147 - 153. (quoted as: Chisholm, Ethics) ––– : "Practical reason and the logic of requirement", in: Stephan Körner (ed.), Practical Reason, Oxford: Blackwell, 1974, pp. 1- 17. (quoted as: Chisholm, Logic) ––– : "Reply to comments", in: Stephan Körner (ed.), Practical Reason, Oxford: Blackwell, 1974, pp. 40 - 53.

190

Clarke, Samuel: A discourse concerning the unchangeable obligations of natural religion, and the truth and certainty of the Christian revelation, in: Samuel Clarke, The Works. Vol. II, 1738, reprint New York: Garland, 1978. (quoted as: Clarke, Discourse) Cummiskey, David: "Desert and entitlement", in: Analysis 47, 1987, pp. 15 - 19. Cupit, Geoffrey: "Desert and responsibility", in: Canadian Journal of Philosophy 26, 1996, pp. 83 - 99. Dolinko, David: "Retributivism, consequentialism, and the intrinsic goodness of punishment", in: Law and Philosophy 16, 1997, pp. 507 - 528. ––– : "Some thoughts about retributivism", in: Ethics 101, 1991, pp. 537 – 559. Duff, R. A.: "Penal communications: Recent work in the philosophy of punishment", in: Crime and Justice: A Review of Research 20, 1996, pp. 1 - 97. ––– : Trials and Punishments, Cambridge: Cambridge University Press, 1986. Düsing, Klaus: "Das Problem des höchsten Gutes in Kants praktischer Philosophie", in: Kant-Studien 62, 1971, pp. 5 - 42. Engstrom, Stephen: "The concept of the highest good in Kant's moral theory", in: Philosophy and Phenomenological Research 52, 1992, pp. 747 - 780. Feinberg, Joel: "Duties, rights, and claims", first published in: American Philosophical Quarterly 3, 1966, pp. 1 - 8; quoted from: Joel Feinberg, Rights, justice, and the bounds of liberty. Essays in social philosophy, Princeton: Princeton University Press, 1980, pp. 130 - 142. ––– : "Justice and personal desert", first published in: Carl J. Friedrich / John W. Chapman (eds.), Nomos VI: Justice, New York: Atherton Press, 1963, pp. 69 97; quoted from: Joel Feinberg, Doing and deserving. Essays in the theory of responsibility, Princeton: Princeton University Press, 1970, pp. 55 - 94. ––– : "The nature and value of rights", first published in: Journal of Value Inquiry 4, 1970, pp. 243 - 257; quoted from: Joel Feinberg, Rights, justice, and the bounds of liberty. Essays in social philosophy, Princeton: Princeton University Press, 1980, pp. 143 - 155. ––– : "Wollaston and his critics", first published in: Journal of the History of Ideas 38, 1977, pp. 345 - 352; quoted from: Joel Feinberg, Rights, justice, and the bounds of liberty. Essays in social philosophy, Princeton: Princeton University Press, 1980, pp. 307 - 314. Feldman, Fred: "Desert: Reconsideration of some received wisdom", first published in: Mind 104, 1995, pp. 63 - 77; quoted from: Fred Feldman, Utilitarianism, hedonism, and desert. Essays in moral philosophy, Cambridge: Cambridge University Press, 1997, pp. 177 - 192. Fforde, Jasper: The Eyre affair, London: Hodder & Stoughton, 2001. Finnis, John: "The restoration of retribution", in: Analysis 32, 1972, pp. 131 - 135. Fisher, John / Ravizza, Martin: "Thomson and the trolley", in: Journal of Social Philosophy 23, 1992, pp. 64 - 87.

191

Foot, Philippa: "Moral dilemmas revisited", first published in: W. Sinnott-Armstrong / D. Raffman / N. Asher (eds.), Modality, morality, and belief, Cambridge: Cambridge University Press, 1994, pp. 117 - 128; quoted from: Philippa Foot, Moral dilemmas and other topics in moral philosophy, Oxford: Clarendon Press, 2002, pp. 175 - 188. ––– : "Moral realism and moral dilemma", first published in: Journal of Philosophy 80, 1983, pp. 370 - 398; quoted from: Philippa Foot, Moral dilemmas and other topics in moral philosophy, Oxford: Clarendon Press, 2002, pp. 37 - 58. Galston, William A.: Justice and the human good, Chicago: University of Chicago Press, 1980. Gert, Bernard: "Transplants and trolleys", in: Philosophy and Phenomenological Research 53, 1993, pp. 173 - 178. Gorr, Michael: "Thomson and the trolley problem", in: Philosophical Studies 59, 1990, pp. 91 - 100. Grandy, Richard E. / Warner, Richard (eds.): Philosophical grounds of rationality. Intentions, categories, ends, Oxford: Clarendon Press, 1986. Gregor, Mary J.: Laws of freedom, New York: Barnes and Noble, 1963. Grice, H. P.: "Meaning", in: Philosophical Review 66, 1957, pp. 377 – 388. ––– : "Meaning revisted", first published in: N. V. Smith (ed.), Mutual knowledge, New York: Academic Press, 1982, pp. 223 - 243; quoted from: H. P. Grice, Studies in the way of words, Cambridge, Mass.: Harvard University Press, 1989, pp. 283 - 304. Hampton, Jean: "A new theory of retribution", in: R. G. Frey / Christopher W. Morris (eds.), Liability and responsibility. Essays in law and morals, New York: Cambridge University Press, 1991, pp. 377 - 414. ––– : "An expressive theory of retribution", in: Wesley Cragg (ed.), Retributivism and its critics. Archiv für Rechts- und Sozialphilosophie, Beiheft 47, 1992, pp. 1 25. ––– : "Correcting harms versus righting wrongs: The goal of retribution", in: UCLA Law Review 39, 1992, pp. 1659 - 1702. ––– : "The retributive idea", in: Jeffrie G. Murphy / Jean Hampton, Forgiveness and Mercy, Cambridge: Cambridge University Press, 1988, pp. 111 - 161. Hart, Lois Cloarec: "Lost and found", published in: Lois C. Hart, Assorted Flavours. A collection of lesbian short stories, Clayton: P. D. Publishing, 2005, pp. 163 194; quoted from: http://www.storycorner.org/ljchart/lostandfound.htm, 2004. Hart, Lois Cloarec / Paradee, Carol: Murder most foul, http://www.academyofbards.org/fanfic/l/loischart_cparadee_murder.html, 2002. Harward, Donald W.: "The bitter pill of punishment: Retribution", in: Journal of Value Inquiry 10, 1976, pp. 199 - 204. Herodot: Historien. Zweiter Band: Bücher VI - IX, 5. Jh. v. Chr., Josef Feix (ed.), Darmstadt: Wissenschaftliche Buchgesellschaft, 1995.

192

Hill, Thomas E., Jr.: "Kant on imperfect duty and supererogation", first published in: Kant-Studien 62, 1971, pp. 55 - 76; quoted from: Ruth F. Chadwick (ed.), Immanuel Kant. Critical Assessments. Volume III: Kant's moral and political philosophy, London: Routledge, 1992, pp. 346 - 369. Hohfeld, Wesley Newcomb: "Some fundamental legal conceptions as applied in judicial reasoning I", 1913, in: Wesley Newcomb Hohfeld / Walter Wheeler Cook (ed.): Fundamental legal conceptions as applied in judicial reasoning, New Haven: Yale University Press, 1966, pp. 23 - 64. Holland, R. F.: "Thought, word and deed, Part II", in: Proceedings of the Aristotelian Society, Supp. 54, 1980, pp. 119 - 132. Honderich, Ted: Punishment: The supposed justifications, New York: Harcourt, Brace & World, 1969. How, Walter Wybergh / Wells, Joseph: A commentary on Herodotus. Volume II: Books V - IX, Oxford: Clarendon Press, 1968. Hudson, W. D.: Reason and right. A critical examination of Richard Price's moral philosophy, London: Macmillan, 1970. Hume, David: A treatise of human nature, 1739-40, 2nd edition, L. A. Selby-Bigge / P. H. Nidditch (eds.), Oxford: Clarendon Press, 1978. Hunter, Michael: Science and society in Restoration England, Cambridge: Cambridge University Press, 1981. Jacob, Margaret C.: The Newtonians and the English Revolution 1689 – 1720, Hassocks, Sussex: Harvester Press, 1976. Johnson, Robert N.: "Kant's conception of merit", in: Pacific Philosophical Quarterly 77, 1996, pp. 310 - 334. Joynton, Olin: "The problem of circularity in Wollaston's moral philosophy", in: Journal of the History of Philosophy 22, 1984, pp. 435 - 444. ––– : "Wollaston's theory of declarative action", in: Southern Journal of Philosophy 19, 1981, pp. 439 - 450. Kant, Immanuel: Grundlegung zur Metaphysik der Sitten, 1785, in: Königlich Preußische Akademie der Wissenschaften (ed.), Kants gesammelte Schriften. Band IV, Berlin: de Gruyter, 1968. Translation: Mary J. Gregor, Groundwork of the metaphysics of morals, in: The Cambridge edition of the works of Immanuel Kant. Practical Philosophy, Cambridge: Cambridge University Press, 1996, pp. 37 - 108. (quoted as: Kant, GMS) ––– : Kritik der praktischen Vernunft, 1788, in: Königlich Preußische Akademie der Wissenschaften (ed.), Kants gesammelte Schriften. Band V, Berlin: de Gruyter, 1968. Translation: Mary J. Gregor, Critique of practical reason, in: The Cambridge edition of the works of Immanuel Kant. Practical Philosophy, Cambridge: Cambridge University Press, 1996, pp. 133 - 272. (quoted as: Kant, KpV)

193

––– : Kritik der reinen Vernunft, 2nd edition, 1787, in: Königlich Preußische Akademie der Wissenschaften (ed.), Kants gesammelte Schriften. Band III, Berlin: de Gruyter, 1968. Translation: Paul Guyer / Allen W. Wood, The Cambridge edition of the works of Immanuel Kant. Critique of pure reason, Cambridge: Cambridge University Press, 1997. (quoted as: Kant, KrV) ––– : Kritik der Urtheilskraft, 1790, in: Königlich Preußische Akademie der Wissenschaften (ed.), Kants gesammelte Schriften. Band V, Berlin: de Gruyter, 1968. Translation: James Creed Meredith, The critique of judgement, Oxford: Clarendon Press, 1989 (quoted as: Kant, KU) ––– : Die Metaphysik der Sitten, 1797, in: Königlich Preußische Akademie der Wissenschaften (ed.), Kants gesammelte Schriften. Band VI, Berlin: de Gruyter, 1968. Translation: Mary J. Gregor, The metaphysics of morals, in: The Cambridge edition of the works of Immanuel Kant. Practical Philosophy, Cambridge: Cambridge University Press, 1996, pp. 353 - 604. (quoted as: Kant, MS and MST for the "Tugendlehre") ––– : Die Religion innerhalb der Grenzen der bloßen Vernunft, 1793, in: Königlich Preußische Akademie der Wissenschaften (ed.), Kants gesammelte Schriften. Band VI, Berlin: de Gruyter, 1968. (quoted as: Kant, Religion) ––– : Von einem neuerdings erhobenen vornehmen Ton in der Philosophie, 1796, in: Königlich Preußische Akademie der Wissenschaften (ed.), Kants gesammelte Schriften. Band VIII, Berlin: de Gruyter, 1968 (quoted as: Kant, Vornehmer Ton) Kleingeld, Paulien: "Moral und Verwirklichung", in: Zeitschrift für philosophische Forschung 44, 1990, pp. 425 - 441. Kleinig, John: "The concept of desert", first published in: American Philosophical Quarterly 8, 1971, pp. 71 - 78; quoted from: Louis P. Pojman / Owen McLeod (eds.), What do we deserve? Oxford: Oxford University Press, 1999, pp. 84 92. Lamont, Julian: "The concept of desert in distributive justice", in: The Philosophical Quarterly 44, 1994, pp. 45 - 64. Liddell, Henry George / Scott, Robert: A Greek English lexicon, 9th edition, Oxford: Clarendon Press, 1996. Lucas, J. R.: On justice, Oxford: Clarendon Press, 1980. MacIntyre, Alasdair: A short history of ethics, London: Routledge & Kegan Paul, 1966. Mack, Eric: "Of transplants and trolleys", in: Philosophy and Phenomenological Research 53, 1991, pp. 163 - 167. Mackie, J. L.: Hume's moral theory, London: Routledge & Kegan Paul, 1980.

194

Marcus, Ruth Barcan: "Moral dilemmas and consistency", first published in: Journal of Philosophy 77, 1980, pp. 121 - 136; quoted from: Ruth Barcan Marcus, Modalities. Philosophical essays, New York: Oxford University Press, 1993, pp. 125 - 141. Marshall, S. E.: "Harm and punishment in the community", in: Wesley Cragg (ed.), Retributivism and its critics. Archiv für Rechts- und Sozialphilosophie, Beiheft 47, 1992, pp. 75 - 82. Matravers, Matt: Justice and punishment. The rationale of coercion, Oxford: Oxford University Press, 2000. McLeod, Owen: "Contemporary interpretations of desert. Introduction", in: Louis P. Pojman / Owen McLeod (eds.), What do we deserve? Oxford: Oxford University Press, 1999, pp. 61 - 69. ––– : "Desert and institutions", in: Louis P. Pojman / Owen McLeod (eds.), What do we deserve? Oxford: Oxford University Press, 1999, pp. 185 - 195. Miller, David: Social justice, Oxford: Clarendon Press, 1976. Morris, Herbert: "Persons and punishment", in: The Monist 52, 1968, pp. 475 - 501. Murphy, Jeffrie G.: "Marxism and retribution", first published in: Philosophy and public affairs 2, 1973, pp. 217 - 243; quoted from: Jeffrie G. Murphy, Retribution, justice, and therapy, Dordrecht: Reidel, 1979, pp. 93 - 115. ––– : "The highest good as content for Kant's ethical formalism. Beck versus Silber", in: Kant-Studien 56, 1965, pp. 102 - 110. ––– : "Three mistakes about retributivism", first published in: Analysis 31, 1971, pp. 166 - 169; quoted from: Jeffrie G. Murphy, Retribution, justice, and therapy, Dordrecht: Reidel, 1979, pp. 77 - 81. Norcross, Alastair: "Rights violations and distributive constraints: Three scenarios", in: Pacific Philosophical Quarterly 76, 1995, pp. 159 – 167. Nozick, Robert: Anarchy, state, and utopia, New York: Basic Books, 1974. O'Neill, Onora: "Kant on reason and religion", in: Grethe B. Peterson (ed.), The Tanner lectures on human values 18, Salt Lake City: University of Utah Press, 1997, pp. 267 - 308. Pogge, Thomas W.: Realizing Rawls, Ithaca: Cornell University Press, 1989. Price, Richard: A review of the principal questions in morals, 1787, D. D. Raphael (ed.), Oxford: Clarendon Press, 1974. (quoted as: Price, Review) Rachels, James: The elements of moral philosophy, New York: Random House, 1986. ––– : "What people deserve", in: John Arthur / William H. Shaw (eds.), Justice and economic distribution, Englewood Cliffs: Prentice-Hall, 1978, pp. 167 - 196. Raphael, D. D.: "A review of the principal questions in morals. Editor's introduction", in: Richard Price, A review of the principal questions in morals. ––– : The moral sense, Oxford: Oxford University Press, 1947.

195

Rawls, John: A theory of justice, Cambridge: Belknap Press of Harvard University Press, 1971. Raz, Joseph: "Comment: Reasons, requirements and practical conflicts", in: Stephan Körner (ed.), Practical Reason, Oxford: Blackwell, 1974, pp. 22 - 35. Rosenberg, Alexander: "Contractarianism and the 'trolley' problem", in: Journal of Social Philosophy 23, 1992, pp. 88 - 104. Sadurski, Wojciech: Giving desert its due: Social justice and legal theory, Dordrecht: Reidel, 1985. Sandel, Michael J.: Liberalism and the limits of justice, Cambridge: Cambridge University Press, 1982. Schaller, Walter E.: "From the Groundwork to the Metaphysics of Morals. What happened to morality in Kant's theory of justice?" in: History of Philosophy Quarterly 12, 1995, pp. 333 - 345. Schiffer, Stephen R.: Meaning, Oxford: Clarendon Press, 1972. Schiller, Friedrich von: Kabale und Liebe. Ein bürgerliches Trauerspiel, Frankfurt / Main: Suhrkamp, 1999. Schneewind, J. B. (ed.): Moral philosophy from Montaigne to Kant. Vol. I, Cambridge: Cambridge University Press, 1990. (quoted as: Schneewind, Moral philosophy I) ––– (ed.): Moral philosophy from Montaigne to Kant. Vol. II, Cambridge: Cambridge University Press, 1990. (quoted as: Schneewind, Moral philosophy II) ––– : The invention of autonomy. A history of modern moral philosophy, Cambridge: Cambridge University Press, 1998. Scott, Sir Walter: Ivanhoe, Ian Duncan (ed.), Oxford: Oxford University Press, 1998. Shakespeare, William: Romeo and Juliet, Harold Jenkins (ed.), London: Methuen, 1982. Sher, George: Desert, Princeton: Princeton University Press, 1987. Sinnott-Armstrong, Walter: Moral dilemmas, Oxford: Blackwell, 1988. Slattery, Brian: "The myth of retributive justice", in: Wesley Cragg (ed.), Retributivism and its critics. Archiv für Rechts- und Sozialphilosophie, Beiheft 47, 1992, pp. 27 - 34. Smith, Steven G.: "Worthiness to be happy and Kant's concept of the highest good", in: Kant- Studien 75, 1984, pp. 168 - 190. Sterba, James P.: "Justice and the concept of desert", in: The Personalist 7, 1976, pp. 188 - 197. Stocker, Michael: "Desiring the bad. An essay in moral psychology", in: Journal of Philosophy 76, 1979, pp. 738 - 753. Sverdlik, Steven: "The logic of desert", in: Journal of Value Inquiry 17, 1983, pp. 312 - 324.

196

Thomas, D. O.: "Political writings. Introduction", in: Richard Price / D. O. Thomas (ed.), Political writings, Cambridge: Cambridge University Press, 1991, pp. vii - xxii. ––– : The honest mind. The thought and work of Richard Price, Oxford: Clarendon Press, 1977. Thomson, Judith Jarvis: "Reply to commentators", in: Philosophy and Phenomenological Research 53, 1993, pp. 187 - 194. ––– : "Reply to critics", in: Philosophy and Phenomenological Research 58, 1998, pp. 215 - 222. ––– : The realm of rights, Cambridge: Harvard University Press, 1990. Thornton, Mark: "Against retributivism", in: Wesley Cragg (ed.), Retributivism and its critics. Archiv für Rechts- und Sozialphilosophie, Beiheft 47, 1992, pp. 83 - 92. Tweyman, Stanley: Reason and conduct in Hume and his predecessors, Den Haag: Martinus Nijhoff, 1974. ––– : "Truth, happiness, and obligation. The moral philosophy of William Wollaston", in: Philosophy 51, 1976, pp. 35 - 46. Vailati, Ezio: "Introduction", in: Samuel Clarke / Ezio Vailati (ed.), A demonstration of the being and attributes of God, Ezio Vailati (ed.), Cambridge: Cambridge University Press, 1998, pp. ix - xxxi. Velleman, J. David: "The guise of the good", in: Nous 26, 1992, pp. 3 – 26. Walker, Nigel: "Just deserts or just desertion? An attack on modern retributivism", in: Richard Mowery Andrews (ed.), Perspectives on punishment. An interdisciplinary exploration, New York: Peter Lang, 1993, pp. 161 - 170. ––– : Why punish? Oxford: Oxford University Press, 1990. Watkins, J. W. N.: "Comment: What does Chisholm require of us?" in: Stephan Körner (ed.), Practical Reason, Oxford: Blackwell, 1974, pp. 36 - 40. Williams, Bernard: "Ethical consistency", first published in: Proceedings of the Aristotelian Society, Supp. 39, 1965, pp. 103 - 124; quoted from: Bernard Williams, Problems of the self, Cambridge: Cambridge University Press, 1973, pp. 166 - 186. Wollaston, William: The religion of nature delineated, 1724, reprint New York: Garland, 1978. Wood, Allen W.: Kant's moral religion, Ithaca: Cornell University Press, 1970. Zaitchik, Alan: "On deserving to deserve", in: Philosophy and Public Affairs 6, 1977, pp. 370 - 388.

Index of Names Anscombe, G. E. M., 56-59, 187 Arnold, N. Scott, 16-18 Austin, John L., 100 Auxter, Thomas, 133n.218 Bayes, Thomas, 76 Beck, Lewis White, 133-135 Beethoven, Ludwig van, 35f. Benn, S. I., 151n.223 Bernstein, Leonard, 35f. Bittner, Rüdiger, 33-35, 137 Boyle, Robert, 61 Braithwaite, John, 151n.223, 160n.247 British Rationalists, 8, 45, 60, 62, 93, 95, 129f., 137 Broome, John, 81n.122 Brown, Charlotte R., 95n.141 Burke, Edmund, 75 Butler, Bishop Joseph, 76 Chisholm, Roderick M., 8, 36, 45, 4760, 71, 84f., 138, 139 Clarke, Samuel, 8, 45f., 55, 60, 61-73, 75, 76, 84, 91, 93, 95, 110f., 116, 119, 121, 129f., 132, 135, 137 Cudworth, Ralph, 75 Cummiskey, David, 16-18 Cupit, Geoffrey, 11n.1, 28n.39 Dolinko, David, 152n.224, 160n.247 Duff, R. A., 151n.223, 152n.225, 160n.247f., 161f. Düsing, Klaus, 132n.215, 135n.221 Engstrom, Stephen, 132n.215 Feinberg, Joel, 7, 8, 13-20, 21, 29, 33, 45, 47, 60, 99n.154, 103n.166, 112114, 117f. Feldman, Fred, 11n.1, 28n.39 Fforde, Jasper, 39f.

Finnis, John, 153n.226 Fisher, John, 180n.284 Foot, Philippa, 167n.258, 169, 173n.268 Galston, William A., 16n.12 Gert, Bernard, 180n.284 Gorr, Michael, 180n.284 Grandy, Richard E., 104n.169 Gregor, Mary J., 125n.197, 125n.200 Grice, H. P., 101, 104f., 154f. Hampton, Jean, 101, 152-165 Hart, Lois Cloarec, 38f. Harward, Donald W., 160n.248 Herodot, 21n.40 Hill, Thomas E., Jr., 124n.195, 125n.201 Hobbes, Thomas, 61, 63, 65 Hohfeld, Wesley Newcomb, 176-179, 182, 184 Holland, R. F., 104f., 107 Honderich, Ted, 151n.223 How, Walter Wybergh, 31n.40 Hudson, W. D., 77n.105 Hume, David, 76, 112-114 Hunter, Michael, 61n.76 Hutcheson, Francis, 76, 86 Jacob, Margaret C., 61n.76 Johnson, Robert N., 126n.202 Joynton, Olin, 100, 103n.167, 112, 114116 Kant, Immanuel, 8, 36, 45f., 70, 72, 76, 110, 120, 121-126, 187 Kleingeld, Paulien, 132n.216, 135n.221 Kleinig, John, 13n.3, 19n.18 Lamont, Julian, 11n.1 Leibniz, Gottfried Wilhelm, 61 Liddell, Henry Georg, 31n.40

198

Locke, John, 77 Lucas, J. R., 16n.12 MacIntyre, Alasdair, 112n.179 Mack, Eric, 180n.284 Mackie, J. L., 100, 113n.183, 114n.184 Marcus, Ruth Barcan, 167n.256, 168175, 181, 183, 185, 187 Marshall, S. E., 160n.247 Matravers, Matt, 151n.223, 160n.247f., 161n.250 McLeod, Owen, 14n.6, 16n.12, 17n.15, 19n.19 Miller, David, 16n.12, 20 Morris, Herbert, 153n.226 Murphy, Jeffrie G., 133n.218, 153n.226 Newton, Isaac, 61 Norcross, Alastair, 180n.284 Nozick, Robert, 22f., 25f. O'Neill, Onora, 135n.221 Paradee, Carol, 38 Pettit, Philip, 151n.223, 160n.247 Plato, 75 Pogge, Thomas, 26-28 Price, Richard, 8f., 45f., 72f., 75-93, 95, 110, 113, 116, 119f., 121, 122f., 131, 137, 140-144, 147, 185 Queen Anne, 61 Rachels, James, 11n.1 Raphael, D. D., 73n.97, 75n.99, 76, 77n.105 Rashdall, Hastings, 76 Ravizza, Martin, 180n.284 Rawls, John, 21-29, 187 Raz, Joseph, 53, 55f. Rohault, Jacques, 61 Rosenberg, Alexander, 180n.284 Sadurski, Wojchiech, 11n.1

Sandel, Michael J., 24-26 Schaller, Walter E., 124n.193f. Schiffer, Stephen R., 104n.169 Schiller, Friedrich von, 37f. Schneewind, J. B., 61n.75, 67-69, 70n.89f., 75n.99, 76 Scott, Robert, 31n.40 Scott, Sir Walter, 99 Shakespeare, William, 37 Sher, George, 11n.1, 17n.15, 22n.26 Sinnott-Armstrong, Walter, 167n.256, 173n.268 Slattery, Brian, 151n.223 Smith, Steven G., 134f. Sterba, James P., 13n.3 Stocker, Michael, 68n.86 Stoecker, Ralf, 146n.222 Sverdlik, Steven, 16n.11 Thomas, D. O., 73n.97, 75n.98, 76n.100, 77n.103, 77n.105, 80n.112, 84n.125 Thomson, Judith Jarvis, 167, 169, 171, 174f., 175-184, 187 Thornton, Mark, 151n.223 Tweyman, Stanley, 103n.166 Vailati, Ezio, 61n.75f. Velleman, J. David, 69n.87 Walker, Nigel, 151n.223 Warner, Richard, 104n.169 Watkins, J. W. N., 49-51 Wells, Joseph, 31n.40 Williams, Bernard, 167n.256, 168-175, 181, 183, 185, 187 Wollaston, William, 8f., 45f., 72, 93, 95-120, 121, 129f., 131, 135, 137, 155, 159 Wood, Allen W., 132n.215 Zaitchick, Alan, 22f.

Index of subjects N.B.: Subjects need not occur as terms on the pages specified, but may rather just indicate the subject matter discussed there. agglomeration principle, 170, 173-175, 185, 187 aggravation principle, 180 apology, 107, 158, 167, 170-172, 175, 179, 181-182, 184-187 aptitude. See means-end-relation balance, 7, 10, 39f., 111, 187 burglary, 159 circularity, 112-114, 115, 116-118, 119, 161, 176-178 claims, 175-184, 187 absolute, 176-178, 179, 182, 184 according, 178, 179f. not according. See claims, infringement of as correlatives of duties, 177 generation of, 182f., 185f. infringement of, 154, 160, 175, 179f., 182-184, 185 legal, 178 moral, 178 non-absolute, 178-180, 184 normativity of, 179f. stringency of, 180 See also moral residue, claims as compensation, 15, 154, 159, 167, 175, 184 conflict of requirements. See moral dilemmas correspondence, 25f., 98, 104-107 criterion of distinction. 44, 139-141, 144, 146. See also desert and propriety deontic logic, 173

desert and claims, 16, 19, 26f. bases, 13f., 18f., 20, 22f., 24, 31, 33, 43f., 139, 140, 142f., 144, 146 bases, responsible actions as, 11, 21, 27f., 42, 43f., 142 as completion, 31-33, 40-44. See also propriety as completion concept of, 7-9, 13, 20, 31, 139-144, 147f. concept of, metaphysical price of, 9f., 148, 149, 186f. concept of, pervasiveness of, 7, 10, 149, 187 concept of, price of abandoning, 10, 149, 187 concept of, unity of, 12, 44 definition of, 9, 31, 46, 139, 144 examples of, 7, 11, 13f., 16-18, 19, 23, 40-44, 89, 92f., 140f., 144-146, 184 of happiness, 8, 45f., 69-71, 72, 87f., 110f., 121, 122-124, 126, 128-132, 134, 132 and institutions, 16-19, 26f., 44 judgments, analysis of, 40-44, 144146, 184 judgments, truth of v. meaning of, 11, 142 Kant's account of, 122-132 logic of, 7, 11f., 13-16, 19, 32, 33, 43f., 47, 60, 139 normativity of, 7, 13, 19, 145f., 157159, 161f., 163 objects, 18f., 31, 33, 44, 139 objects, polar & nonpolar, 14f. objects, affective character of, 14

200

desert (continued) in ordinary language, 7, 10, 27-29, 149f. preliminary account of. See desert as completion Price's account of, 8f., 87-89, 90, 92f., 119, 140f. and propriety, 7, 20, 21, 29, 44, 45, 90, 92f., 139, 140f., 144, 146f. of punishment, 162-165. See also retributivism as second-order propriety, 88, 93. See also desert, Price's account of subjects, 13, 43, 139 subjects, persons as, 11 duty acting from, 81, 83f., 85, 123, 124, 128, 133 as correlative of claim, 177 normativity of, 177f. of right, 124-128 of virtue, 124-128 empiricism, 77, 82 entitlement,155f., 163 evidence, 101-103, 105, 154, 156, 158f., 164. See also meaning of actions fairness, 7, 21 feelings of guilt, 167, 169f., 184 fittingness, 7f., 31f. examples of, 34f., 185. See also symbols non-normativity of, 8, 9, 31-33, 143, 185 as response, 33-35 See also propriety and fittingness happiness, 70, 109f., 114-116, 132f., 133-135, 136 highest good, 109 Kant's concept of, 8, 46, 120, 131135 hope, 9, 46, 70f., 121, 135, 136, 187

integrity, 81 intermediary facts, 174f., 183, 185 intuitions underlying desert, 124, 130f. underlying moral residue, 168, 173 underlying retributivism, 152f., 161, 165 justice, 7, 14, 21, 56, 66, 151 legal relations, 176f. major scale, 32f., 35, 40f., 43, 72, 139, 143, 146 meaning of actions, 96, 98, 100, 101104, 108, 154-156 means-end-relation, 8, 45f., 63-65, 71f., 90-92, 119, 136 mercy, 56-59 moral dilemmas, 9f., 48f., 52-54, 59, 149, 167-170, 172-175, 178, 185, 187 "the end of the story" in, 167, 169, 171, 182 moral impulse, 170, 172, 185 moral injury, 154-162 appropriate response to, 157, 158f., 161-163 definition of, 154, 156, 161f. moral motivation, 70, 72, 80, 121, 132f., 136, 187 moral realism, 63, 77, 82f., 116-118 moral reality, 155, 169, 174 moral residue, 10, 149, 167-187 claims as, 181-183 feelings as, 167, 169f., 173 moral worth. See value murder, 18, 19, 23, 41f., 144f., 156 nature of things, 63-65, 66, 69f., 72, 96101, 102, 137f. non-voluntarism, 62, 70, 75, 111, 130. See also voluntarism no-right, 177 nullification, 53f.

201

omission, 177 Ought implies Can-principle, 171, 178, 182 overriding, 8, 13, 17, 52-54, 55, 56f. 58f., 60 particularism, 88f., 92 persisting Ought, the, 170-175, 185f. Platonic Forms, 9, 140f., 142, 143, 144, 147f. principle of the augmentation of premises, 50f. promise, 17, 36, 48f., 99, 100, 106f., 167, 169-171, 174f., 181f., 184186 propriety Clarke's account of, 8, 45, 63-65, 71, 119. See also means-end-relation as completion, 8, 40, 44, 138, 148. See also desert as completion concept of, 7f., 32, 45, 60, 137f. definition of, 8, 100, 107, 109, 119 examples of, 35-40, 48f., 52, 56-59, 60, 63f., 72, 91f., 101f., 104f., 106f., 109f., 110f., 113, 116, 119, 138, 146f. See also major scale and fittingness, 7f., 31-33, 35, 37, 4044, 45, 62, 143 importance to us, 8, 46, 72, 120, 121, 132, 135f., 187 logic of, 8, 45, 47-55, 60 normativity of, 7f., 9, 31-33, 40f., 43, 45, 54f., 60, 65-69, 69f., 72, 78-81, 90, 117, 143, 145f., 165 preliminary account of. See propriety as completion Price's account of, 8, 80, 81, 83, 9092, 119. See also means-endrelation Wollaston's account of, 8f., 46, 95, 100, 104-107, 116-118 See also desert and propriety punishment, 7, 9, 15, 16, 18, 19, 23, 41f., 56-59, 70f., 79, 140, 144f., 162-165

punishment (continued) consequentialist justification of, 151f. justification of, 9, 149, 159, 160, 162, 165 retributivist justification of. See retributivism by the state, 41, 158f., 160 worthy of, 154, 156, 159f. rape, 17f., 153, 158 rationalism, 65, 66-69, 73, 77f., 82f., 96, 115 regret, 107, 167, 168-170, 172 relativism, 17f. requirement. See propriety retribution, 162-164 as vindication, 158, 160 retributivism, 9, 101, 149, 151-165 fundamental objection against, 151f., 153, 157, 161f., 165 inherent problems of, 152, 153, 159f., 165 different theories of, 153 rights, 117, 154 sentimentalism, 76f., 86 suitability. See means-end-relation supererogation, 55-59, 84f. symbols, 31f., 33, 40, 143 theft, 159 tradeoff idea, 179, 180 transplant problem, 180f. trolley problem, 180f. value acknowledgment of, 154, 156, 157, 161f. diminishment of, 101, 154, 155f., 157, 158, 163 of human beings, 21, 27f., 101, 154156, 157-159, 160, 161f., 163f. realization of, 154, 156, 157, 161f. representation of, 154f., 156, 157159, 161, 164. See also evidence

202

vandalism, 159, 163 voluntarism, 61f., 63, 65, 70, 76. See also non-voluntarism weakness of will, 67

worthiness, 19f., 70, 87, 122, 132f. Kant's concept of, 8, 120, 121, 123, 130f. wrongful actions. See moral injury