Political Legitimization without Morality?
J¨org K¨uhnelt Editor
Political Legitimization without Morality?
123
Editor Dr. J¨org K¨uhnelt Thoner Weg 26 90425 N¨urnberg Germany
[email protected]
ISBN: 978-1-4020-8575-8
e-ISBN: 978-1-4020-8576-5
Library of Congress Control Number: 2008927738 2008 Springer Science+Business Media B.V. No part of this work may be reproduced, stored in a retrieval system, 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 by the purchaser of the work.
c
Cover painting: ‘Collage’, Karl Heinz K¨uhler, 2007. Printed on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com
Preface
The initial idea for this anthology arose during my work at the interdisciplinary Collaborative Research Centre (SFB) 485 Norm and Symbol at the University of Konstanz. My research project on the potential of Hobbesian contract theory was influenced by the focus of the SFB on social phenomena such as pluralism and cultural change. In this context, I realized that the Hobbesian idea to refer only to instrumental rationality and basic egoistic interests to legitimize a state has, on one hand some advantages for pluralistic societies: All individuals are supposed to share these premises independent of the personal values they might hold. On the other hand, a rational legitimization must cope with the fundamental problem of explaining and legitimizing those tasks of legal states that go beyond the idea of a minimal state. Although my research was focused on the idea of solving this problem with a modification of the Hobbesian argument, I became interested in the more general question of which role morality could or should play in legitimizing a state. Within the current discussion, not only rational but also political accounts of legitimacy can be attractive as soon as they try to avoid contentious normative premises. To analyse some of the core ideas within the current discussion, I organized an interdisciplinary workshop at the University of Konstanz in December 2004 in which different perspectives from sociology, politics and philosophy were compared and analysed. The quality of the proposed ideas supported the aim to edit a systematic volume that provides an overview not only of some of the core problems, but also of the potential of rational and of political accounts of legitimacy within the contemporary discussion. Completing the present volume was only possible with the help of many people who supported me from the initial idea of the volume until its publication. First of all, I would like to thank my PhD supervisor Gottfried Seebaß for his trust and support. Moreover, I need to thank him and all participants of his research colloquium for their helpful suggestions during the discussion of my initial ideas on the theme of the conference. The participants at that time were Barbara Guckes, Monika Oertner, Bernhard Th¨ole, Neil Roughley, Julius Sch¨alike, Stephan Schlothfeldt and Michael Schmitz. I am especially grateful to Stephan Schlothfeldt and Michael K¨uhler, who always found time to answer my questions and discuss my ideas, and whose comments significantly improved this project. Furthermore, I would like to thank the SFB for its generous funding of the workshop. I also need to thank Luc Bovens v
vi
Preface
and Stephan Hartman, heads of the research group philosophy probability and modelling, for their co-sponsoring and for hosting the website. Next, I am very grateful to Sebastian Schleidgen, my assistant, for his helpful comments in many personal conversations concerning the workshop and for being an extremely reliable aid for its organisation. For elaborating the systematic focus of the anthology, I am again very grateful to Michael K¨uhler and Stephan Schlothfeldt. Many thanks also go to Christoph Schmidt-Petri and Franz Huber for their conceptual suggestions. Furthermore, I would like to thank the publishing house Springer, especially Floor Oosting and Ingrid van Laarhoven, for their professional support of this project. I would also like to thank Indumathi Srinivasan and her team for their professional copyediting of the Manuscript. Moreover, I am very pleased and thankful that Karl Heinz K¨uhler allowed me to use one of his pictures for the book cover. At last, I would like to thank Sebastian Schleidgen once more for his help to edit the final manuscript, as well as Eric Besselmann, Julia Day and Michelle Wilson for their proofreading. University of Konstanz, Germany Shanghai, China J¨org K¨uhnelt
Contents
State Legitimacy and the Role of Morality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J¨org K¨uhnelt
1
Part I How Necessary is Morality to Legitimize a State? State Legitimacy and Social Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Christopher Morris The Claims of States and the Claims of Morality . . . . . . . . . . . . . . . . . . . . . . . 33 Neil Roughley Legitimacy and Justice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Wilfried Hinsch Political Legitimacy and Its Need for Public Justification . . . . . . . . . . . . . . . . 53 Michael K¨uhler Consent, Obligation, and Legitimacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Frank Dietrich
Part II Which Role can Rationality Play for State Legitimacy? On the Rationality and Stability of a Minimal Consensus . . . . . . . . . . . . . . . 73 Reinhard Zintl A Commentary on Zintl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Rafaela Hillerbrand vii
viii
Contents
Rational Egoism, Morality and Human Rights . . . . . . . . . . . . . . . . . . . . . . . . . 99 Matthias Kaufmann A Commentary on Kaufmann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Stephan Schlothfeldt Political Contractarianism and Equally Distributed Basic Rights . . . . . . . . 119 J¨org K¨uhnelt Part III How to Ensure the Stability of a Legitimate State? Value-Mistaken and Virtue-Mistaken Norms . . . . . . . . . . . . . . . . . . . . . . . . . . 139 Philip Pettit A Commentary on Pettit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Gottfried Seebaß Political Norms, Markets and Social Capital . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Michael Baurmann Do Multinationals Create Social Capital Just Like That? . . . . . . . . . . . . . . . 181 Christoph Schmidt-Petri Cultural Diversity and Liberalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Russell Hardin Redistributing Liberty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Julius Sch¨alike The Authors of the Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
State Legitimacy and the Role of Morality An Introduction ¨ J¨org Kuhnelt
Within their territories, states claim to be the only authority to determine the law and to use coercive power to punish those who act illegally. Both claims are considered to be acceptable if a state acts in accordance with the framework of a legitimate constitution. A legitimate constitution usually allows a state to establish a stable framework for the peaceful coexistence of its citizens. However, depending on the exact content of a constitution, legal states can be obliged to establish different frameworks and therefore influence the lives of their citizens differently. The aim of a libertarian state, for instance, is restricted to securing public order and property rights. In contrast, an egalitarian state also seeks to guarantee that all citizens have more or less equal opportunities to fulfill their life plans. The question of which constitution or state is supposed to be legitimate depends on the underlying justification. As different justifications are available, the crucial question of how to find a convincing justification for a legitimate state arises. The papers in the present volume propose different answers to the questions of how to justify legitimacy and how to maintain a legitimate state. All authors recognize that conflicts can occur if a state is legitimized with reference to substantial normative premises. Still, they offer distinct solutions. For a better comparison of their arguments, it is helpful to outline the background of the current discussion in political philosophy. By analysing the role of morality, or to be more precise, the reference to normative premises, I think, in principle, three accounts of legitimacy can be distinguished. Normative premises play a dominant role in a moral justification, whereas their role is limited in a political justification. A rational justification has aim to avoid contentious normative premises completely. However, all three accounts face the initial problem that their justification can be criticised. Furthermore, each type of justification has to cope with diverse problems providing arguments to explain the stability of the legitimized state.
1 Three Approaches of State Legitimacy Within the first approach theorists argue that substantial normative premises are necessary to legitimize a state. A moral account of legitimacy can – widely defined – not only be based on substantial moral principles or values but can also be founded J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
1
2
J. K¨uhnelt
on religious values. Consequently, the primary aim of a state is not to establish public order, but to ensure that citizens live in accordance to a certain moral or religious standard. These extensive duties of a state are supposed to be legitimized by a convincing justification that shows the “truth” or appropriateness of the underlying normative premises. A moral justification not only ascribes an obligation to obey the law to all citizens, but also legitimizes sanctions to guarantee that everybody normally has enough incentive to act in accordance with it. However, it is a problem within moral accounts of legitimacy that different and incompatible conceptions of state legitimacy can be proposed, because theorists can rely on different sets of normative premises. This problem can be illustrated nicely, if one takes a closer look at contemporary societies. As long as all individuals share the underlying values in a homogenous society, a moral justification may be attractive. However, such societies are located on one side of a continuum and it is unlikely that many contemporary societies are homogenous. Pluralistic societies are located on the other side of this continuum and many individuals within such societies hold different and conflicting moral or religious views. Furthermore, some individuals living in pluralistic societies can be described as rational egoists, as they only try to fulfill their personal interests and consider religion, morality or moral obligations to be irrelevant. At least to a certain extent, current western societies are pluralistic. If, however, individuals have different and probably incompatible opinions on the properties of a legitimate state, the question of why a legitimate state should be justified with reference to a certain set of values arises. Nevertheless, this question is only relevant if agreement or compliance with a legitimate constitution is considered to be important for its justification. Hence, it is helpful to analyse the connection between justification and agreement in more detail. In my opinion it is not possible to fully distinguish between the justification of a legitimate state and the agreement of its citizens. On one hand, it is not necessary that everybody hypothetically or actually agrees to a legitimate constitution. On the other hand, the justification of a state would be more than questionable if only the leaders or a minority of its citizens actually agree to it. For the further analysis, I assume a justification to be convincing if at least most citizens could agree to it. In this context, it is obviously necessary to specify the properties of hypothetical agreement. Although hypothetical agreement is certainly less demanding than actual agreement, it is not easy to define. However, analyzing a moral justification, one could possibly argue that rational egoists or individuals holding conflicting values are excluded from the justification; hence it is not apparent that they could agree to it. Nevertheless, problems arise not only in defining hypothetical agreement; it is also necessary to asses the importance of the potential to convince citizens in relation to other aspects of justification. One can argue, for instance, that it is more important to base a constitution on a normative conception of social justice, than to convince as many citizens as possible. Finally, even if individuals have reason to obey the law because they could agree to the constitution, they still may not be sufficiently motivated to actually do so. A detailed analysis of these aspects is not necessary to show the following consequence: Within a non-homogenous society it is unlikely that a state, which dictates a
State Legitimacy and the Role of Morality
3
certain moral or religious standard, at the same time relies on a constitution to which most individuals could agree. In other words, it is doubtful that most individuals in a pluralistic society could agree to a state that is based on a certain set of moral values as long as a significant number of individuals are rational egoists or hold different moral or religious views and therefore prefer different ways to live their lives. Of course, one can argue that the proposed moral justification is “true” or “reasonable” and is therefore supposed to override other moral claims or conflicting individual interests. However, different sets of normative premises lead to different and possibly conflicting conceptions of legitimate constitutions and, on principle, every normative theory can claim overridingness. Still, sanctions can be “justified” to ensure the stability of a moral state, even if a significant number of individuals claim that they are excluded from justification. Nevertheless, these individuals have a good reason not agree to the constitution and to complain that the conflicts between different normative perspectives are not solved by justification, but by coercion. Furthermore, if theorists assume that hypothetical or actual agreement is more or less irrelevant for a convincing justification, even moral or religious fundamentalist states could be “justified”. Theorists within the second approach take these problems seriously and propose a political account of legitimacy. They admit to a reasonable disagreement on moral values but assume that no such disagreement exists as far as political values are concerned. In relation to moral values, political values have a limited scope because they are focused primarily on peaceful coexistence and fair cooperation within a state.1 Hence, states are supposed to be liberal, as they are not allowed to dictate moral standards that citizens are required to maintain. Still, it has to be analyzed whether or not this restriction is capable of solving the problems discussed above. The first of two distinct types of political justifications refers directly to political values. For instance, a liberal theory can be based on a set of political rights which are assigned to everybody in the society. However, this is only one possibility of a direct reference to values. Some theorists understand the liberal idea differently and claim, for instance, that each individual requires at least similar chances to fulfill his personal life plan. The former conception is, for instance, used by Libertarians who specify a state in which property rights are dominant and based on natural law.2 Given this libertarian understanding of property rights, any kind of substantial redistribution program of the state, for example to ensure social justice, would amount to theft and hence be illegitimate. Egalitarian philosophers employ the latter idea and ascribe legitimacy to a state only if it is built upon a certain account of
1 This distinction between moral and political values is rather technical, as they share the property to specify a normative fundament of legitimacy. In this terminology, a justification that refers to religious values would also be considered as a moral justification. However, because of their limited scope, political values can be distinguished from moral values. Nevertheless, some scholars use morality or moral assumptions synonymous with normative assumptions. 2
Nozick (1974): Anarchy, State, and Utopia.
4
J. K¨uhnelt
equality.3 A significant conflict arises, as Egalitarians surely would consider a state to be illegitimate if a feasible governmental redistribution program to guarantee sufficient equality is not developed or launched. The comparison of Libertarians and Egalitarians illustrates that conflicting conceptions of legitimacy can be proposed even within political accounts of legitimacy. Of course, one has to admit that more individuals can agree to a political justification than to a moral justification and liberal theories usually legitimize similar core tasks of a state, for instance, the prohibition of murder. Still, substantial conflicts on the necessity of further governmental tasks remain, for example, on the role of social justice. Even if theorists dedicate much effort into designing a convincing normative theory, more or less the same problems that arise for moral accounts also arise for political accounts. Different sets of political values can be proposed to legitimize different constitutions and individuals often have no sufficient reason to change their normative perspective: It is more than questionable whether Libertarians could agree to an Egalitarian constitution and vice versa. Therefore, rational egoists and a sufficient number of individuals holding conflicting values will claim to be excluded even from a political justification and consider the state to be illegitimate at least from their individual perspective. Although these individuals will not feel obliged to obey the law, they will still act in accordance with it, at least as long as they fear governmental sanctions. Nevertheless, if justification is taken seriously, it cannot be reduced to a potential problem of stability. The second type of a political justification is meant to solve these problems with an indirect reference to political values: A legitimate constitution is supposed to be specified with reference to a fair consensus between all or most citizens. According to the contractualist conception, individuals are supposed to agree to such a consensus within a hypothetical thought experiment. Because of this agreement, individuals not only accept the legitimacy of the state, but also have a reason to obey the law. In his later works, John Rawls developed a prominent argument for an overlapping consensus on a legitimate constitution within a pluralistic society.4 A fair and stable consensus between different moral groups specifying basic and further governmental tasks is realistic within his theory, as Rawls relies on the normative assumption that all individuals are supposed to support only reasonable comprehensive doctrines and thus share the aim of a fair cooperation. Individuals holding reasonable moral or religious doctrines are supposed to recognize when their personal values endanger this fair consensus. In this case, they will make concessions and withdraw conflicting normative claims. Nevertheless, rational egoists and individuals supporting unreasonable normative doctrines are excluded from negotiations, as they do not share the proposed aim of a fair cooperation. These consequences show that contractualist theories generally, at least implicitly, rely on political values which are used indirectly to enable and frame a certain idea of a fair consensus, which specifies the tasks of a legitimate state.
3
Dworkin (2002): Sovereign Virtue.
4
Rawls (1993): Political Liberalism.
State Legitimacy and the Role of Morality
5
The indirect reference to political values can be criticized in two ways: Firstly, conflicts can arise even between individuals sharing the aim of a fair consensus, as different conceptions of fairness can be proposed. Secondly, not everybody will share the normative assumption that a fair consensus is more important than personal interests or conflicting normative values. It is questionable, for instance, whether individuals who support Libertarian or Egalitarian conceptions would accept the normative framework necessary for a fair consensus, as they would have to withdraw substantial elements of their positions to avoid the conflict described above. Furthermore, rational egoists will not be convinced of any idea of fairness. These problems illustrate why some scholars support an account of legitimacy that tries to avoid controversial normative assumptions completely. This third approach is based on a rational account of legitimacy and relies on the idea that states are legitimate because they are a necessary condition for citizens to improve the satisfaction of their interests. Thomas Hobbes’ Leviathan is a prominent example for a contractarian theory.5 The idea is to avoid disagreement on the proposed justification by avoiding any reference to controversial moral or political values. Contractarians replace controversial normative assumptions with the help of empirical premises. They refer primarily to the rationality of individuals and to certain basic interests that all individuals are supposed to share. It is assumed, in a hypothetical thought experiment, that rational individuals would not interact efficiently if neither law nor moral norms were established in a society in a so-called state of nature. Contractarians argue that rational individuals have problems of interaction that can be solved only by a legal state that establishes a certain framework enabling basic security as well as efficient cooperation and coordination. This legal state is considered to be legitimate because it enables gains of cooperation which are advantageous for everybody. Hence everybody could rationally agree to this state. However, to prove whether or not this argument is convincing, it is necessary to analyse its structure in detail. To avoid normative premises within their justification, contractarians neither assume that all individuals support certain moral values, nor that they share particular altruistic interests. Hence, they analyse how egoistic individuals interact in a state of nature if law and moral norms do not exist. Although, all individuals share the interest not to be killed or robbed, they are supposed to kill and steal, given this is an efficient means of satisfying their interests in material goods or security. Even if it would be more beneficial for individuals to live in an environment without theft and murder, as long as no legal state guarantees public order, it is supposed to be rational for every individual to use all possible means to satisfy his interests. This behaviour is explained in relation to the framework of interaction in the state of nature: It is irrational for individuals to constrain their behaviour as long as they cannot be sure
5
Hobbes (1991 [1651]): Leviathan. The term contractarianism is often used for Hobbesian contract theories and contractualism for Kantian contract theories. For instance, Robert Sugden gives a good overview on the distinction of Hobbesian and Kantian contract theory. Sugden (1993): “The contractarian enterprise”.
6
J. K¨uhnelt
that everybody else does the same. They would just become more exploitable and reduce their options to act without any chance to establish a public order, because others have no incentive to constrain their behaviour as well. Thus, individuals in the state of nature live in permanent insecurity regarding their lives and goods, leading to a very low production and efficiency. A state is considered to be the only solution for these problems because only a state is powerful enough to establish sanctions and hence provide an incentive, even for egoistic individuals, to obey a general prohibition of theft and murder. This sketch of the argument shows why contractarian theories are supposed to avoid controversial normative assumptions, and therefore do not face the problems arising from pluralistic societies. Not only rational egoists but also the remaining individuals will be able to fulfill their aims better in a legal state than in the state of nature. Consequently, they are supposed to recognise that the advantages of a legal state provide good reason to accept its legitimacy. Clearly contractarian theories also face several problems: Firstly, individuals supporting substantial normative values could complain of being unconvinced of this argument because they evaluate their normative point of view to be more important than their egoistic interests. In this case they can argue that they are systematically excluded from the justification, as the contractarian argument refers primarily to egoistic interests. Secondly, it is doubtful if a rational justification can really avoid all normative premises. Contractarians claim that all individuals could be able to agree to a legitimate state because of its advantages in relation to the state of nature. Difficulties arise, for instance, if some individuals in the status quo have enough power to maintain certain privileges that are not advantageous for everybody in relation to the state of nature. One has to keep in mind that contractarians do not refer to more than rational egoism. Rational egoists, however, are only concerned about others as long as cooperation is a necessary means to fulfill their personal aims. Therefore, why should powerful rational egoists in the described status quo give up their privileges as long as they do not accept the normative premise that a state is only legitimate if it is advantageous for all in relation to the state of nature? Thirdly, even if this normative premise is generally accepted. In relation to a strong inequality of power in the state of nature even a slavery state could be beneficial for everybody.6 Finally, a rational justification has to cope with a quite prominent problem: In any case, for any state that may be rationally justified, the so-called freerider or compliance problem arises. Generally, rational egoistic individuals will agree that it is beneficial to create a state that secures, for instance, individual rights. However, as no obligations bind egoistic individuals to keep their promises or agreements, it can be rational – at least under certain circumstances – for an egoist behave opportunistically and to steal or kill: This would be the case if one is in an occasion in which theft or murder is very beneficial to him, has no negative consequence for the institution of individual rights, and the danger of getting caught and sanctioned is
6 James Buchanan discusses this problem in more detail. Buchanan (1999 [1975]): The Limits of Liberty.
State Legitimacy and the Role of Morality
7
sufficiently low.7 However, if a certain number of individuals break the law, negative consequences for its stability arise. Within this context, two rival starting points are available to legitimize a state: Firstly, the problems of rational accounts of legitimacy can motivate scholars to remain within the framework of political accounts. Still, these scholars usually take the problems of political accounts seriously and accept the need to discuss the question of how necessary morality is to legitimize a state. Consequently, they try to avoid controversial normative premises and seek new arguments within a modified political justification. Secondly, if scholars believe normative premises of political accounts to be too controversial, they will argue within the framework of rational accounts of legitimacy. However, in this case, it appears to be necessary to specify the empirical assumptions more precisely and analyse the exact role of rationality. Therefore, scholars search for a convincing description of individual behaviour or a stable substitution for the normative obligation to obey the law. Regardless of which starting point a scholar takes, both strategies appear to be in-between the spectrum of classical political and rational accounts of legitimacy. Therefore, stability problems may not only arise within rational accounts of legitimacy. For instance, if political accounts refer to abstract premises to achieve an uncontroversial justification, they will probably face the problem that such arguments fall short in motivating the citizens to obey the law. In consequence, scholars of both accounts of legitimacy are confronted with the difficulty of finding an appropriate substitute or backup for governmental sanctions, as sanctions alone are problematic and often not sufficient to ensure the stability of a legal state.
2 General Overview The description above is meant to map out different arguments within the discussion in political philosophy in order to clarify the background of the papers and problems addressed in this volume. As already mentioned, the contributors of this anthology mainly focus on the questions of how to justify legitimacy or maintain a legitimate state. They share the opinion that a reference to substantial normative assumptions cannot solve all problems of state legitimacy and agree that different means must be found to ensure the legitimacy and stability of a legal state. The papers cover central topics of the discussion as they either offer modifications for rational accounts of legitimacy or propose revised political accounts of legitimacy. Of course, within this volume only a small spectrum of the current discussion within political theory can be presented. Furthermore, the authors focus on state legitimacy, even though some of the suggested ideas and conceptions would also be interesting to justify the legitimacy of international institutions. Nevertheless, the
7 For instance, Russell Hardin (2003) gives a good general introduction in the Stanford Encyclopedia of Philosophy.
8
J. K¨uhnelt
papers in the present volume are supposed to provide the reader with an overview on general aspects of state legitimacy. The idea was to compile inderdisziplinary perspectives to discuss some of the central problems of state legitimacy. Scholars from different academic backgrounds investigate the problems of normative and empirical assumptions and present new ideas within the general discussion to cope with these problems. The sections of the volume cover three areas of analysis. The first section focuses on the necessity of morality to legitimize a state. Within the second section, scholars discuss which role rationality and empirical assumptions can play within state legitimacy. The third section offers some practical suggestions to ensure the stability of legitimate states. A number of commentaries provide the reader with a different perspective on the problems and advantages of the presented ideas.
2.1 How Necessary is Morality to Legitimize a State? The papers in the first section investigate, which role morality should have to develop a convincing normative conception of legitimacy. The authors take the mentioned conflicts between different normative assumptions and positions seriously and analyse alternatives for current political accounts of state legitimacy. They propose and discuss different conceptions of legitimacy that are supposed to avoid contentious moral or political values. As these approaches cannot fully rely on a normative obligation to obey the law, some papers also discuss the necessity of possible substitutes for this obligation. All papers share the aim of providing a convincing conception of legitimacy, but work with miscellaneous normative and empirical assumptions. As a result, the contributions show different possibilities to synthesize mainstream ideas to provide new conceptions of legitimacy. Christopher Morris suggests a conception of basic legitimacy that requires a less demanding justification, which can be fulfilled with either normative or rational arguments. Morris argues that legitimate states have a right to exist and a right to rule. He distinguishes between full legitimacy, which he believes to be the traditional conception, and his suggestion of weak or basic legitimacy. Whereas full legitimacy imposes the obligation for everybody to obey every valid law, basic legitimacy only imposes an obligation not to undermine the state and to support it in certain ways. Additionally, he analyses which possible substitutes of morality exist for this reduced obligation to obey the law. In his commentary, Neil Roughley discusses the source of the obligations of basic legitimacy and suggests deriving them from moral premises. Wilfried Hinsch analyses the normative limits of political justice within the contractualist framework. He distinguishes between conceptions of justice and conceptions of legitimacy. Hinsch notes that the contractualist idea of public justification cannot be extended to cover each and every political decision because it has to cope with the problem of reasonable disagreement. Hence, a conception of justice can only specify proto-rights that contain the idea of political equality or the right of
State Legitimacy and the Role of Morality
9
political participation. To clearly define these rights within a legitimate state, Hinsch suggests a conception of legitimacy which has to rely on non-argumentative procedures, such as majority voting. This conception of legitimacy specifies a just state, because it is based on just proto-rights. As this state fulfils the demands of justice, everybody has an obligation to obey its law. Michael Kuehler questions whether or not it is possible for non-argumentative procedures to specify legitimate norms autonomously if they are still bound to the demands of justice. Frank Dietrich discusses the justification and necessity of an obligation to obey the law. Within contract theories, he argues, the obligation to support and comply with the political institutions of one’s country is usually based on the principle of consent. He investigates the central idea of contract theory that uses either tacit or hypothetical consent. As a first result, he notices that substantial problems exist within Kantian and Hobbesian variants of contract theories to justify this obligation. Therefore, he examines Allen Buchanan’s suggestion of political legitimacy in which the obligation to obey the law is regarded as obsolete. In opposition to Buchanan, he argues for the necessity of such an obligation and suggests that Hobbesian contract theories have to be investigated further to provide a suitable justification.
2.2 Which Role Can Rationality Play for Legitimacy? The papers in the second section analyse the potential role of rationality or other empirical assumptions in legitimizing a state. If theorists try to avoid controversial moral or political values, the question arises of whether or not to substitute these normative assumptions within a rational justification of a legitimate state. A central problem is the opportunistic behaviour of egoistic individuals, which can lead to stability problems within a state. Therefore, some authors investigate the possibility to connect normative and rational arguments. Furthermore, some authors analyse how different models of individual behaviour and different descriptions of the state of nature influence the results of contractarian arguments. Reinhard Zintl argues while contractarianism can convincingly show the advantages of a state in comparison to anarchy, it has difficulties determining a specific state. To clarify this problem, he investigates on an abstract level the different possibilities for a minimal consensus on the properties of a legitimate state. Instead of starting with a more or less explicit set of specifications – as it is common in political theory – he analyses the outcomes of the interaction of different actors with different preferences and dispositions. Actors close to the model of homo oeconomicus identify their common interests and find a rational and efficient constitution. Nevertheless, as they are opportunists, they have to cope with serious stability problems. Less opportunistic “complex” actors behave virtuously but hold different ethical or religious convictions which endanger the possibility of a minimal consensus. In this context, Zintl suggests that individuals should invest in virtue and information processing capabilities to ensure a stable and efficient society. In her commentary,
10
J. K¨uhnelt
Rafaela Hillerbrand questions whether or not an investment in virtue is really possible without increasing conflicts between different ethical convictions. Matthias Kaufmann proposes that state legitimacy needs a reference to morality and rationality. On one hand, he argues that even rational individuals have good reasons to live and participate in a moral society. On the other hand, he admits that virtuous citizens are important to secure the stability of a state. Kaufmann focuses on the problem that rigorous moralists could justify arbitrary restrictions on the behaviour of others with reference to their idea of virtue. He argues that a reference to long-term rational egoism could provide a protection for individuals against this tendency, as rational egoists would usually select humanitarian values without oppressive elements. From a moral point of view, Kaufmann proposes that the idea of freedom as non-domination also supports the rational demand for a morality without the danger of fanaticism. In his commentary, Stephan Schlothfeldt questions whether rational egoists would always vote for a humanistic morality or act in accordance to it. J¨org K¨uhnelt believes rational accounts of legitimacy to be especially promising in relation to pluralistic societies. Nevertheless, he acknowledges that contractarian theories are still based on problematic normative assumptions, cannot convincingly exclude slavery states from the list of possible results, and struggle with the freerider problem as they assume egoistic benefit maximizing behaviour of individuals. In this context, James Buchanan’s modified contractarian theory becomes interesting, as he avoids certain problematic premises. Nevertheless, K¨uhnelt criticises that Buchanan relates the distribution of rights to the distribution of power and interests and thus cannot guarantee everybody stable basic rights within a minimal state. In relation to this result, K¨uhnelt argues for relativistic contractarian theory that could be based on a more realistic model of individual behaviour, a more specific description of the distribution of interests in the state of nature, and a more explicit reference to certain general economic arguments.
2.3 How to Ensure the Stability of a Legitimate State? The papers in the third section analyse the practical problem of how to ensure the stability of a legitimate state. This problem is generally discussed in relation to rational accounts of legitimacy, as their focus on egoistic interests leads to the aforementioned freerider problem. However, not only in pluralistic societies, individuals can have different reasons to be insufficiently motivated to obey the law. As mentioned above, even the justification of a political account of legitimacy is not necessarily able to motivate a sufficient number of citizens to obey the law. Of course, the use of governmental sanctions is one possibility to ensure the stability of a state. Nevertheless, sanctions are a harsh and problematic means to seek stability and sanctions are surely the only available option. The following papers propose not only alternatives to governmental sanctions, but also analyse certain restrictions to governmental tasks to ensure the stability of legal states.
State Legitimacy and the Role of Morality
11
Philip Pettit discusses the possible role of social esteem to support socially attractive norms. Although Pettit admits that internalization can support stability, he argues that social norms may often be supported or reinforced by the fact that deviant behaviour is usually punished by the withdrawal of approval, or the appearance of disapproval. Furthermore, the positive approval of others might be a stronger reason to comply with a norm than the acceptance of the underlying value. He believes that many norms at least stabilize and fixate because of an error on peoples’ opinion of the attitudes of others. Pettit describes the case of value-mistaken norms, where individuals comply with norms because they wrongly imagine that others support the underlying values. He refers to the example that states fear to inform the public that fewer people comply with tax laws than generally believed. Pettit suggests to use these effects for instance to improve institutional design in matters of the compliance problem. In his commentary, Gottfried Seebaß questions whether value mistaken norms can really be considered as stable norms. Michael Baurmann focuses on the role of social capital to support civic virtues and to ensure a stable civil society. Social capital theory shows that certain personal attitudes and dispositions, for instance trust and reciprocity, are usually established in small groups, but also have positive effects for the individual behaviour within a society. However, Baurmann distinguishes two forms of social capital and shows that one has destructive consequences while the other has the desired supportive effects for a democratic order. The preferred form of social capital is supposed to create bridging social networks, establish norms of generalized reciprocity and generalized trust, and promote the commitment to public goods. In this context, Baurmann analyses the possible role of markets and multinational companies for the creation and support of social capital. Finally, he suggests the need for a virtuous equilibrium between markets, politics, and civil society. Nevertheless, Christoph Schmidt-Petri questions whether multinational companies really support the kind of social capital that is needed for a civil society. Russell Hardin analyses how liberalism should cope with problems caused by cultural diversity. He investigates why liberal theory should rely on universalistic principles and thus exclude versions of group liberalism which support illiberal practices that are likely to be destructive to the survival of the social order. He distinguishes between political, economic and social liberalism. Political liberalism is supposed to liberate from despotic or oligarchic political control, whilst economic liberalism tries to liberate from political decisions concerning who has to work, to produce or to trade. Finally, social liberalism has the aim to liberate from social conventions. Hardin argues that even if new versions of welfare liberalism were to be enacted, they usually share the property of being self-enforcing, once maintained. Their stability is based on the fact that they are mutually advantageous for most political important groups in liberal societies. In relation to this result, Hardin argues that group based liberalism is fundamentally different. Conflicts arise especially in relation to social liberalism, which is supposed to liberate from social norms and thus might undercut the ties of certain communities. He discusses certain examples of illiberal groups and refers, for instance, to the problem of honour killings. In his
12
J. K¨uhnelt
comment, Julius Schaelike suggests that these issues have to be addressed by an underlying theory of distributive justice. Acknowledgments I am very grateful to Michael K¨uhler, Christoph Schmidt-Petri, Stephan Schlothfeldt and an anonymous referee for critical commentaries on earlier versions of this introduction. Furthermore, I would like to thank Michelle Wilson for her suggestions on style and language.
References Buchanan, James (1999 [1975]): The Limits of Liberty – Between Anarchy and Leviathan, Indianapolis: Liberty Fund. Dworkin, Ronald (2002): Sovereign Virtue, Cambridge (MA): Harvard University Press. Hardin, Russell (2003): The Freerider Problem, in: Zalta, Edward (ed.): The Stanford Encyclopedia of Philosophy,
. Hobbes, Thomas (1991 [1651]): Leviathan, Tuck, Richard (ed.), Cambridge: Cambridge University Press. Nozick, Robert (1974): Anarchy, State, and Utopia, New York: Basic Books. Rawls, John (1993): Political Liberalism, New York: Columbia University Press. Sugden, Robert (1993): The Contractarian Enterprise, in: Gauthier, David and Sugden, Robert (eds.): Rationality, Justice and the Social Contract. Themes from Morals by Agreement, Ann Arbor: University of Michigan Press.
Part I
How Necessary is Morality to Legitimize a State?
State Legitimacy and Social Order Christopher Morris
Our world is one of states. Virtually all of the land mass of the globe today is the territory of a (single) state, and virtually all humans alive are subject to the laws of a state. Some countries or societies are degenerate cases of a state (e.g., Somalia, Congo), and others are relics of pre-modern Europe (e.g., Liechtenstein, San Marino, the Vatican). But even these aspire to be states or are often thought of as states. “International relations” are principally relations that obtain between states. The concept of legitimacy seems to be essential to making sense of this world of states. States always claim to be legitimate, and they are often recognized as such. Legitimacy is widely held to be very important, and the lack of it debilitating. With some frequency it is said that we face a crisis of legitimacy, and often a particular state is said to lack legitimacy. Without legitimacy, it is sometimes thought, there is not much that states, even very powerful ones, can do. There is, however, massive confusion about legitimacy. Contemporary discussions betray a notable lack of clarity and understanding about legitimacy and associated concepts. Sometimes it seems that parties to a dispute have different conceptions in mind – possibly different concepts – and may not, in effect, be talking about the same thing. Other times it is hard to know what they are thinking of. Disorder in the world may not be unrelated to disorder in our thoughts. What follows is an opinionated overview of the topic of state legitimacy and of its connections to social order. Much that I shall offer will be analysis, with part of the needed evidence or argument left out for reasons of space. But I think that clarity alone is immensely helpful with respect to the questions at hand.
1 The Importance of States In the conditions of modern times states are very important for the maintenance of social order and prosperity. Some people – I am thinking especially of classical liberals or libertarians, and anarchists – have expressed scepticism about this claim, and their case is often not without merit. However, states seem to play a significant role in preserving many of the conditions of social order, especially in situations of conflict and instability when that order is threatened. That is not to say that states J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
15
16
C. Morris
are the sole or even principal source of social order; the various and numerous kinds of social relations of “civil society” clearly are important. But states and their legal systems are important. Sceptics may – again often rightly – point out that many of the major threats to social order in fact originate with the state system; for instance, the principal parties to the major conflicts of the last century were states, and the agents of the great massacres of our time have typically been states. The point is well taken, and the balance sheet will surely need to mention the Nazi and Soviet states, among others. States are with us for now, however, and in our present conditions it is hard to see how their abolition would improve things. The rule of law, basic security of person and possessions, the often intricate rules and standards necessary for commerce, even civil society, all depend to a considerable extent on states. Even if many political societies need to be organized as states mainly because they are surrounded by hostile states, there is as of yet no feasible alternative to the state system.1 There is much to be said against states and for alternative forms of political organization.2 Globalization and political experiments such as the European Union may transform the state system considerably. And many of the powers associated with states have been eroded in recent decades. But states will be with us for some time, and I presume that their contribution to the maintenance of social order will not significantly decrease for some time. The question of their legitimacy is thus of some importance.
2 The Importance of Legitimacy Legitimacy seems quite important to a state’s contribution to the preservation of social order. Just as the effectiveness of governments or regimes is limited by their loss of legitimacy, so states may be hampered by failing to be legitimate. Legitimation crises once were thought to threaten the internal life of Western states. Now they are said to be especially worrisome in international affairs.3 One of the principal reasons why legitimacy is important is simply that without it states cannot efficiently perform the tasks we assign to them. States not infrequently impose sanctions or use force to secure order. But there are limits to what they can force people to do. Without other incentives, states are ineffective.4 The thought here is that legitimacy is needed to allow states to do what they need to do. A legitimate state is one that is 1 The remark about being surrounded by hostile states refers the process of “secondary state formation” whereby one state or centralized political society is formed by reaction to the threatening presence of another. See my discussions in: Morris (1998): An Essay on the Modern State, p. 80. 2
And I have said some of it. Morris (1998), An Essay on the Modern State, esp. chs. 1–3 and 10.
3
Robert Kagan (2004): “America’s Crisis of Legitimacy”, in: Foreign Affairs 83, 2, pp. 65–83.
4
I think the state’s dependence on force or coercion greatly exaggerated by philosophers and theorists, who invariably invoke a simplified Weberian “definition” of the state. The state is rightly seen as requiring considerable political power in order to carry out its assigned tasks, but authority is a species of power and the fixation on coercion makes us lose sight of the state’s authority. I discuss these matters below. See also my “Is the State Necessarily Coercive?” (manuscript).
State Legitimacy and Social Order
17
more likely than an illegitimate one to be supported by its subjects and assisted, or at least not hindered, by others. A legitimate state, or one that is believed to be legitimate? It is not always clear; in fact, the distinction is sometimes not clearly drawn.5 Very often the notion of legitimacy in these contexts seems to be a de facto one. The force of the argument itself suggests that it is the belief in a state’s legitimacy that is crucial – a state that was legitimate but not believed to be so would be no more effective than an illegitimate one thought to be legitimate (it might even be less effective). Some social scientists and political commentators do not draw the distinction and think of legitimacy as consisting in the widespread belief in legitimacy. As stated, this is an absurd or at least circular conception of legitimacy and has often been recognized as such. (What would it then be to believe something to be legitimate?) A de facto notion of legitimacy seems to presuppose the intelligibility of a de jure or normative notion. The basic idea behind these kinds of conceptions is that legitimacy lies in the beliefs or other attitudes of peoples, especially those subject to the state’s laws. Social stability and thus social order most certainly depend directly on people’s beliefs. But it is hard to believe that legitimacy itself is a property of people’s mental beliefs or mental attitudes. For the concept of legitimacy is in the first instance a normative one. Someone who believes that a particular state is legitimate believes it has a certain status. That status is important and may contribute significantly to social order. The first question is in what consists this status?
3 What is Legitimacy? The legitimacy that interests us here is a property of states. But we speak of legitimacy in many contexts. Moving a pawn sideways is not a legitimate move in chess; that is, it is not allowed by the rules of the game. Were a head of state or a legislature to act in ways prohibited by the state’s constitution, it would be acting illegitimately. In games, as in politics, we often say “You can’t do that!” in response to a player’s move. The notion of legitimacy here is that of lawfulness or legality: something is legitimate insofar as it is in accordance with the rules of the game, in this case, the law. This notion of legitimacy is extremely useful in political life. We often wish to know whether actors are conforming to the rules of the game. It can be very important to know that a government has assumed power legitimately, just as in
5
“Legitimacy arises from the conviction that state action proceeds within the ambit of law, in two senses: first, that action issues from rightful authority, that is, from the political institution authorized to take it; and second, that it does not violate a legal or moral norm. Ultimately, however, legitimacy is rooted in opinion, and thus actions that are unlawful in either of these senses may, in principle, still be deemed legitimate. That is why it is an elusive quality. Despite these vagaries, there can be no doubt that legitimacy is a vital thing to have, and illegitimacy a condition devoutly to be avoided.” Robert W. Tucker and David C. Hendrickson (2004): “The Sources of American Legitimacy”, in: Foreign Affairs 83, 6, p. 18 (emphases added).
18
C. Morris
olden days much depended on ascertaining the legitimate heir to the throne. But this understanding of legitimacy does not always help us when we wish to know whether the constitutional order itself, the state, is legitimate.6 Some have argued that legality exhausts the notion of legitimacy and have denied that one can speak of the legitimacy of a system of rules or constitutional order. This position seems even less plausible today than in centuries past. Few now seem genuinely to accept it.7 In international contexts, states are sometimes said to be legitimate to the extent that they are recognized as such by other states. We are familiar with the manner in which states may withhold recognition from other states (e.g., Israel, Taiwan). The thought here may be that if a state is recognized as legitimate by most (legitimate) states, it is legitimate. The first problem is the familiar one of determining whether recognition here is supposed to be evidential or constitutive. It seems bizarre to think that recognition could be constitutive. If it were, the notion would lose much of its interest or importance. Legitimacy would then be analogous to membership in a club and would amount to little more. The club in question – the set of states – does not have a distinguished history, not to mention a discerning membership committee. There may, of course, be uses for a club of this kind – “the family of nations” – but it would be best not to think of legitimacy in this way; most of us presumably wish to deny (genuine) legitimacy to many member states of the current United Nations. Recognition at best would be an evidential relation. More precisely, a case could be made for understanding people’s belief in a state’s legitimacy (or their support for a state) as evidence for its legitimacy. When one understands the factors that often lead one state to recognize another as legitimate, it is often hard to be particularly impressed. However, if many reasonable and well-informed people believe a particular state to be legitimate, that might be reason to agree – not conclusive, of course, but grounds nevertheless. Some take widespread belief in legitimacy to be constitutive of legitimacy, but again this is implausible; at best the relation seems evidential. There are some complicated relations to be disentangled here – we often need to distinguish people’s belief in an institution’s legitimacy, their support for the institution, their endorsement of it, and their consent to it. Some of the distinctions will be made below. Legitimacy is a status, and we do not seem to make much progress understanding what it is by examining people’s beliefs, legality, or recognition. A more promising approach would be to try to determine the nature and terms of the status accorded a legitimate state. Not that long ago people still spoke of a child’s illegitimacy; to call someone a “bastard” was then to make a quite specific accusation. An illegitimate child is one born out of wedlock, and this was once a shameful way to come
6
I qualify the remark (“not always”), as some legal thinkers – for instance, Lon Fuller or Ronald Dworkin – argue that genuine legality has some legitimating force. 7 The scepticism underlying this view may be more interestingly expressed by other accounts of legitimacy – for instance, that a state is legitimate insofar as it has a legal system and satisfies the other conditions for being a genuine state.
State Legitimacy and Social Order
19
into the world. For our purposes the noteworthy feature of this condition was the child’s status, its lack of a certain kind of standing. It could not inherit and was not due other benefits, and in the social hierarchies that were once all determining, its relative place was low. Hierarchies like these are not part of our political world; the official discourse of the modern “family of nations” is egalitarian – all states, like all citizens, are equal. Not to have legitimacy is to be excluded from the family or club and to be denied the normal benefits of membership. What are the normal benefits of the status bestowed on legitimate states? There are in the first instance a bundle of rights and powers. A legitimate state may do certain sorts of things, and others may not prevent them from so acting. A legitimate state, to start with, has a right to exist; its very existence is permissible. Consider, for instance, some of the hostility to Israel. It does not merely reflect a denial of particular powers to the Israeli state; rather, it expresses the conviction that Israel has no right to exist.8 Similarly, the position of the People’s Republic of China regarding Taiwan, we may suppose, is that it has no right to exist as an independent state.9 National liberation or independence movements seek a state for their group or people, and they suppose that this state would have a right to exist. I shall think of the right to exist as the first of two major rights possessed by legitimate states. It may be interpreted as a mere liberty or Hohfeldian privilege – the absence of a duty forbidding the thing in question. But I assume that if a state is legitimate and therefore has a right to exist, other parties have a duty not to destroy it and that this duty is correlative to the right to exist. So I shall interpret this right as a claim-right and not a mere liberty. The right to exist attaches in the first instance to a territory. All modern states have territories, for the most part compact and contiguous, and outsiders are forbidden access to these territories without permission. Others, including states, have a duty of non-interference, and a state may defend itself against trespassers and invaders. As one may threaten a state without encroaching on its territory, blockades and sieges may also violate its right to exist. Legitimate states not only have a right to exist, but they also possess a right to act or to do certain things. Most important of these rights is that to rule. States are forms of political organization, and many of the powers they claim are related to governance. A legitimate state possesses the powers necessary to govern. The law is one of the ways, if not the principal way, in which states act. The right to rule includes the right to make, adjudicate, and enforce laws. In relations to other states and to those not subject to its laws, a legitimate state would posses the right to form alliances, to enter into agreements, to make treaties. The right to rule consists first of all of these rights to make, adjudicate, and enforce law and of these powers regarding agreements with other states.
8 Some might say “exist there”, in a particular place. But modern states are necessarily territorial, and a particular state would cease to be if it lost its territory. 9 Some think that merely admitting that Taiwan is an independent state (a descriptive claim) would be for China to concede too much – evidence to my mind of the confusion about legitimacy.
20
C. Morris
The rights to exist and to rule are not exhausted by the rights and powers I have enumerated. In particular there is an associated claim to exclusivity that attaches itself to these rights. A state’s right to exist rules out alternative arrangements of its population and territory; its right to rule excludes alien sources of law that it does not recognize. The latter claim is hard to express exactly, at least succinctly, as it raises a number of controversies about international law and the nature of legal systems. I do not wish to broach these here and shall try to indicate the kind of exclusivity I am thinking of as simply as I can. States establish the law and apply it in their territories, and they claim that only they can do so. In classical terms, states claim sovereignty, to be the ultimate source of political authority in a realm.10 A state determines what is the law in the sense that, even if it does not create it, it ascertains that it is what it is. The state is the ultimate determiner of what is the law of its realm. The sole virtue, in my view, of the influential Weberian “definition” of a state as “a human community that (successfully) claims the monopoly of the legitimate use of physical force within a given territory”11 is the recognition of the claim to exclusivity. States seek to monopolize their powers to govern. This exclusivity is one of the main sources of the controversies surrounding the powers claimed by states. A legitimate state, then, has a right to exist and a right to rule.12 What duties or obligations are correlative with these rights, especially the second? A state’s right to exist presumably would entail a number of obligations of subjects not to threaten its existence. The right to rule would also ascribe to subjects an obligation not to undermine the state in various ways and in some situations to support it actively. To violate the (genuine) laws of a state would typically be one way of undermining its powers to rule and subjects would presumably have an obligation to obey the law. Traditionally, the obligation to obey the law – that is, the obligation to obey each and every valid law that applies to one – has been associated with legitimate states.13
10
See: Morris (1998): An Essay on the Modern State, ch. 8; Morris (2000): “The Very Idea of Popular Sovereignty: ‘We the People’ Reconsidered”, in: Social Philosophy & Policy 17, 1, pp. 1–26; Morris (2002): “Sovereignty”, in: Encyclopedia of Democratic Thought, edited by Paul Barry Clarke and Joe Foweraker, pp. 673–676.
11
Weber goes on to note that “the right to use physical force is ascribed to other institutions or to individuals only to the extent to which the state permits it. The state is considered the sole source of the ‘right’ to use violence.” Max Weber (1946 [1919]): “Politics as a Vocation”, in: H. Gerth and C. Wright Mills (tr. & eds.): From Max Weber: Essays in Sociology, p. 78. Weber elsewhere gives a much more complete and better one. See Weber (1947): The Theory of Social and Economic Organization (Part I of Wirtschaft und Gesellschaft), A.M. Henderson and T. Parsons (tr.), p. 156.
12
To forestall misunderstanding: this statement is meant neither to be exhaustive of the state’s legitimate powers nor as a definition of legitimacy. The ‘has’ does not express a biconditional.
13
See, for instance, A. John Simmons (1979): Moral Principles and Political Obligations; Leslie Green (1988): The Authority of the State; and David Copp (1999): “The Idea of a Legitimate State”, in: Philosophy & Public Affairs 28, 1, pp. 3–45. A good collection of contemporary articles may be found in William A. Edmundson (ed.), (1999): The Duty to Obey the Law Selected Philosophical Readings. See also Simmons (2001b): “Legitimacy”, in: Encyclopedia of Ethics, Lawrence C. & Charlotte B. Becker, (eds.), 2nd ed., pp. 960–963.
State Legitimacy and Social Order
21
In recent years scepticism about this obligation has been growing, to the point that “philosophical anarchism” – the position that existing states are illegitimate – is no longer a minority view. If legitimacy entails an obligation to obey, then a denial of the latter implies a rejection of legitimacy. The debates here are complex. I am not sure that the usual ways of formulating the questions and positions are best, and I shall frame things somewhat differently. The matter of greatest controversy is whether we, as subjects14 of a particular state, have an obligation to obey the law and to support the state in various ways. The question is not entirely clear at the outset, and the nature of the obligation in question must first be made clear before we can address it. Philosophers immediately distinguish between legal and moral obligations and typically assume that the question must involve the latter. The law trivially imposes a “legal obligation”; the interesting question, most philosophers think, is whether there is a moral one. I do not like this way of proceeding for a number of reasons. For one, I am not always able to locate the boundary between the moral and the non-moral and am suspicious of the ease in which many are able to do so. I also think the grammar of predicating ‘moral’ and ‘legal’ of ‘obligation’ is misleading. Moral and legal obligations are not species of obligations. They are all obligations simpliciter – that is, normative demands or requirements of a certain kind. It is just that they have different sources and that this is what is indicated by the predicate. A moral obligation is simply an obligation whose source is morality or that part of morality which issues requirements of this kind (e.g., justice), and a legal obligation is one whose source is the law. Were etiquette or custom to be a source of obligation – which I think they can be – then there would be other sources of obligation. Now it may be that obligations which have their source in justice always or virtually always override those given to us by the law or by custom. American philosophers commonly assume this must be the case, and they may be right. But to formulate the matter in way which implies there are different species of obligation and that one species necessarily is the determining one is not the best way to frame these difficult questions. For it can lead us not to see what is most radical about the modern state: its claim to supremacy. A political authority is supreme in this sense if it has the power to regulate all other sources of authority (e.g., conscience, church, kin, corporate bodies). The state can be understood as an alternative system of social control to morality, and we do not notice this extraordinary claim if we assume that obligations of justice always override those of the law. So I shall talk about obligation simpliciter. To have an obligation is to be subject to a kind of normative requirement. The state claims to impose obligations on us, and it understands these as requirements created or ratified by the law. The law says that one must respect the property of others, pay one’s taxes, obey the speed limit,
14
Someone is a subject of a set of laws insofar as he or she is directly obligated by them (or is liable to some change in his or her normative situation). A subject of a legal system or state is someone who falls under the latter’s jurisdiction. For the most part we shall be thinking of citizens, but they are a proper subset of the class of subjects.
22
C. Morris
and obtain a license if one wishes to practice dentistry or to be a plumber. It does not say one has a legal or moral obligation to do these things.15 The predicate does not indicate a species of obligation. Obligations may have different sources without being different kinds of requirements. It is especially important to recognize this in the context of a discussion of the powers of the modern state. This particular form of political organization arose in Europe in the context of deep disputes and conflicts over faith and conscience. For some theorists and statesmen, the main purpose of the state was to offer an effective way of settling or adjudicating these disputes, or at least enabling people to live with them. The forms of political organization prevalent in pre-modern Europe proved not to be effective with these and other forms of conflict. The state proposed to make governance territorial and to establish itself as the ultimate source of authority within this territorial realm. What the state determines to be lawful is meant to settle the question of what to do for all persons who find themselves in its territory, no matter what their allegiances might otherwise dispose them. The state’s word, so to speak, is to overrule or pre-empt that of pope or conscience. The idea is that the state can determine what people are to do and that this determination is genuinely to govern them – that is, to determine what they do. The state does not deny that individuals may have obligations of conscience or faith (“moral obligations”). It merely tells them they are to act in the ways that the law instructs them to act. This understanding of the modern state sees it not as something that can create “moral obligations” but as a form of political organization which can create requirements for us whatever our other obligations – or our disagreements about our obligations. The state thus understood, we may say, is a “conflict resolution system” which is a substitute for or alternative to morality or justice.16 This is not to say that states are independent of justice or that unjust states have the powers they claim. Rather, it is to recognize that the state purports to be a system of governance that can substitute for others, especially morals, when they fail to preserve social order. To assume our obligation to obey the law must be moral is to have us overlook this pretension of states. Now we may not accept the state’s claim to be independent of morality and of faith. Theists of more than one kind may plausibly reject it, and moralists may not accord the state the powers it claims. But I do not think we can pose the questions we need to ask without first recognizing the claims that states make. I think
15
You appear in front of a magistrate and ask, “This obligation, your Honour, is it a moral one or merely a legal obligation?”
16
This is one way of reading Hobbes. The Limits of Liberty by James Buchanan (1975) may also be read this way. Edward McClennen also wants to understand political and legal norms as substitutes for morality. See McClennen (1990): “Foundational Explorations for a Normative Theory of Political Economy”, in: Constitutional Political Economy, 1, 1, pp. 67–99; McClennen (1993): “Rationality, Constitutions, and the Ethics of Rules”, in: Constitutional Political Economy, 4, pp. 173–210; and McClennen (2003): “Prudence and Constitutional Rights”, in: University of Pennsylvania Law Review, 151, 3, pp. 917–961.
State Legitimacy and Social Order
23
it will be helpful to talk simply about obligations, understood univocally, without assuming at the outset that obligations from one source always override those from another.17 We could even try to do without the language of obligation, as the aim of the law – the parts of law concerned with normative requirements – is principally to affect our behavior. What the state wants is to have us act in certain ways. Its capacity to impose normative requirements is supposed to be a means of affecting our behavior. Theorists of a certain disposition will see this power as residing in the state’s capacity to sanction illegal behavior, and there is no doubt that sanctions, as well as force or even violence, can be effective ways of influencing people to act as one wishes. Many other theorists, myself included, will understand the law’s attempt to affect our behavior as involving in the first instance an appeal to authority. The fact that the law requires someone to do something is meant to be a reason for him or her so to act. Hobbes clearly explains: “Law in generall, is not Counsell, but Command. . . addressed to one formerly obliged to obey him [who commands]”, where command is “where a man saith, Doe this, or Doe not this, without expecting other reason than the Will of him that sayeth it.”18 The claim about law may be false – and it presumably would at best be credible in the case of a legitimate state – but again it seems necessary to recognize that this is the claim that is made. The reasons offered by authorities purport to be, in Hart’s term, “content-independent”.19 It is hard to understand the law or the state without admitting this. What I am proposing is that we understand the state as a form of political organization that claims for itself the power to impose requirements which are reasons for action for all those subject to its rule. A legitimate state is one that has a right to exist and to rule; that is, if a state is legitimate, then it has these rights. Its subjects have an obligation not to undermine its rule and to support it in a number of ways. We need to say a bit more, however, about the way in which obligations are meant to affect our behavior. Deliberation normally involves weighing reasons against one another and opting for a course of action that is supported by the balance of reasons. The familiar utilitymaximizing conceptions of rationality, dominant in some of the social sciences, are special cases of this more general “balance of reasons” account. Norms generally and obligations imposed by the law specifically can be reasons for action. They can, much like sanctions and similar considerations, affect the balance of reasons. The fact that a particular act is forbidden by the law may, on this understanding, be a 17
Note that my reluctance to understand the obligations imposed by (legitimate) states as moral does not imply that the grounds for the legitimacy of states must also be non-moral. The question of the nature of political or legal obligations is distinct from the nature of the justification of the institutions that create them. I discuss the grounds of legitimacy in Section 5 below.
18 19
Hobbes (1991 [1651]): Leviathan, Richard Tuck ed., ch. xxvi, p. 183, and ch. xxv, p. 176.
“A reason is content-independent if there is no direct connection between the reason and the action for which it is a reason.” Joseph Raz (1986): The Morality of Freedom, p. 35. See also Hart (1982): “Commands and Authoritative Legal Reasons”, in: Essays on Bentham: Jurisprudence and Political Theory, pp. 243–268.
24
C. Morris
reason against it, one which is simply added to the other reasons against. Obligations thus understood would be no different from other kinds of reasons. This is a mistaken understanding of obligations and of norms in general; obligations are not merely a kind of especially weighty reasons. Reasons of this kind are meant to be pre-emptive, and sometimes they seem to be just that. How to understand this feature of norms is, of course, a matter of considerable controversy. I favor Raz’s account of pre-emptive reasons as involving second-order exclusionary reasons requiring that one not act for certain reasons (the ones excluded), as well as the accounts of Gauthier and McClennen, which are specifically adapted to formal maximizing conceptions of practical rationality.20 The obligations imposed by the law, then, I shall think of as requirements which are pre-emptive. They are reasons that displace or exclude other reasons for acting differently. The laws of legitimate states are meant to be of this kind. The law is supposed to affect our behavior by affecting our reasons for action. Much of the time it does not matter whether it is successful in so doing; many acts are overdetermined, and the reasons provided by the law are merely supplementary. Most of us, for instance, refrain from killing our neighbors or kidnapping children independently of the laws forbidding such acts. But in many instances the existence of a law may be instrumental in securing social order and other goods. The aim of the state is practical, to secure social order and other goods by controlling our behaviour. It does not matter whether we obey the law out of respect for it, only that our acts conform with the law, for whatever reason. So sanctions can be useful as supplementary reasons for action for those occasions when we fail to do what we have reason to do.21 What is important is our behaviour, specifically its conformity with the law. Given the practical nature of the law and of the state generally, there is an additional reason to emphasize reasons rather than moral obligations.22 Let us return to the question of the obligations entailed by the rights to exist and to rule possessed by legitimate states. Minimally there would be an obligation not to undermine the state and to support it in certain ways. I am thinking of certain acts which would contribute to undermining a state’s rule and even its existence – for instance, publicly showing disrespect for the authorities in certain circumstances, encouraging sedition, supporting revolution or anarchy in various ways.23 In many instances public violations of the law would undermine a state’s rule. In certain 20 Raz’s account is first laid out in: Raz (1999 [1975]): Practical Reason and Norms. The most easily accessible presentation of David Gauthier’s account is found in: Gauthier (1986): Morals By Agreement, Ch. VI. But his later essays are in many ways better; see “Commitment and Choice: An Essay on the Rationality of Plans”, in: Francesco Farina, Frank Hahn, and Stefano Vannucci (eds.), (1996): Ethics, Rationality, and Economic Behaviour, pp. 217–243. 21
I am thinking only of legitimate states here. Tyrannical ones presumably need to rely on these supplementary measures to a much greater extent. See my “Is the State Necessarily Coercive?”
22 23
See also the remarks below about “internalism” about obligations.
These examples are meant as illustrations. Lest they be misunderstood, I should note that it is not the case that all obligations should be enforced. Clearly, liberal states would want to permit
State Legitimacy and Social Order
25
circumstances subjects of a legitimate state would also have a number of obligations to do things to support its rule – for instance, paying taxes, serving in the military. These obligations, we should note, are perfectly compatible with the absence of an obligation to obey each and every law that applies to one. The obligation to obey the law is more demanding than that not to undermine the state and to support it; it is to conform to each and every law that applies to one. Consider the examples of infractions of the law that have no adverse effects that are now standard in the literature – for instance, failure to come to a complete halt at stop signs on deserted roads in the countryside. Or think of violations of more misguided laws – for example, US laws forbidding young people from drinking alcohol before the age of 21 (even, in some states, in the company of their parents). Failure to conform with the law in such cases would not undermine the state, even if it would violate an obligation. Consider also laws which forbid people, no matter how knowledgeable, from providing certain services (e.g., providing a medicine, rewiring a house) for others without a licence. Were an unlicensed (knowledgeable, competent) physician to provide his wife with a useful drug or were someone very knowledgeable about electrical wiring to do repairs on her cousin’s house for a fee, the law typically would be violated. A legitimate state possesses a right to exist and a right to rule. But these rights can be interpreted in a number of different ways. Legitimacy may come in different forms or strengths depending on the specific obligations correlative to these rights. Too often it is simply assumed that legitimacy is a univocal notion. Given the disagreements about how to understand the concept, we should not expect it to be univocal, much less simple. I shall distinguish between two forms of legitimacy. There are undoubtedly more, and my characterization of the notion to now has been very simple. For our purposes we need to distinguish between a weaker and a stronger notion. Weak or basic legitimacy imposes an obligation on subjects not to undermine the state and to support in certain ways. And strong or full legitimacy imposes, in addition, an obligation on subjects to obey the law – that is, an obligation to obey every valid law that applies to them. In this essay, in order not to complicate matters overly, I shall assume that if someone has an obligation in these contexts, he or she has a pre-emptive reason to act as required. This is an internalist assumption about obligations.24 Given the practical aim of the law, the state claims that its directives are reasons for action – specifically, content-independent, pre-emptive reasons – assuming internalism here makes this claim evident. (Alternatively, we could formulate things differently, without the internalist assumption.) To summarize, if a state is legitimate, it has a right to exist and a right to rule. If a state possesses weak or basic legitimacy then subjects have an obligation (and a pre-emptive reason) not to undermine the state and to support it in various ways.
citizens to do many things (e.g., speak in favor of revolution) even if they have obligations not to do them. 24
The internalism here, it is worth noting, is about reasons, not motives.
26
C. Morris
If it possesses strong or full legitimacy, subjects also have an obligation (and a pre-emptive reason) to obey the law. We should noted that non-subjects will in both cases have obligations not to undermine legitimate states and, in certain circumstances, to support it. Interestingly, the obligations of subjects and non-subjects toward a weakly legitimate state may be quite similar.25 What I have characterized as strong or full legitimacy is, I believe, the traditional conception.26 On this (strong) conception, if it turns out that subjects lack the obligation to obey a state, it is illegitimate. Indeed, on this view, legitimacy must be denied to states where subjects occasionally lack the obligation to obey particular laws (e.g., traffic or drug and alcohol laws). That is, even if virtually all subject are obligated to obey virtually all laws most of the time, and have pre-emptive reasons to do so, the fact that some lack these obligations and reasons on occasion suffices to rob the state of legitimacy. This is seems to be a feature of the traditional conception, which is part of the reason I wish to distinguish it from another, which I have dubbed weak or basic legitimacy.
4 What is the Value of Legitimacy? Legitimacy presumably has considerable instrumental value. A state which is legitimate and widely thought to be so may elicit more voluntary support or at least acquiescence from subjects than would otherwise be available. Legitimacy might also secure a state wider support from non-subjects. States that obtain such support will typically be better able to achieve their ends with less resistance; they will need to rely less on alternative means of achieving their ends, such as coercion or payoffs. I am supposing that legitimacy is a necessary condition of genuine authority. States claim authority. If something is a genuine authority, its directives will be pre-emptive reasons for action. Authority is a power to get people to act in ways requested. So to the extent that a state possesses genuine authority, it will be more effective in achieving its ends. Legitimacy confers on states a right to exist and a right to rule. A legitimate state is one which may do what it is authorized to do; that is, its lawful acts are all permissible. Certain kinds of challenges and criticisms of a state will be rebutted by establishing its legitimacy. Being able to addresses such challenges and to answer such criticisms is also valuable, instrumentally and perhaps intrinsically. There are a number of advantages to be had by the possession of legitimacy.
25
This means that an account of weak legitimacy will not address the “particularity problem”, that of showing that the obligations to obey are particular to citizens of a state. See Simmons (1979): Moral Principles and Political Obligations, and Green (1988): The Authority of the State.
26
Again, see the works by Simmons, Green, and Copp, cited above.
State Legitimacy and Social Order
27
5 The Grounds of Legitimacy What can establish or confer legitimacy? There is and has been considerable controversy as to the grounds of legitimacy. There are a number of well-known accounts, and I shall say little to resolve the debates between their partisans. Rather I wish to explore the implications of some of these accounts for the legitimacy of states. Suppose a state to be reasonably just and efficient. It adheres to most of the normal constraints of justice regarding life, liberty, and possessions, and most of its activities it carries out with reasonable efficiency. Both assumptions may be difficult for some critics of states to accept. For instance, if we have certain kinds of rights to liberty and to property, then it will be hard to see how states could fail to violate these rights and be just. I shall not linger on this point here and simply urge sceptics to suspend belief for a moment.27 Suppose, in addition, that this state satisfies one or more of the following conditions: (1) it does many useful and worthy things – for instance, it provides security for its subjects, a number of important public or collective goods, relief of poverty and illness, (2) its subjects would consent to it under certain appropriate conditions, (3) its subjects consent to it. Readers will recognize a number of familiar accounts of the grounds of state legitimacy. Satisfaction of the first condition would show that the state in question was beneficial to or served the interests of the governed. My formulation of this conditions does not distinguish between different members of a broadly consequentialist family of accounts, some stressing mutual advantage, others adopting non-Paretian accounts of the general or overall good; all are lumped together in one class. I trust the differences between the second and third conditions will be clear: hypothetical vs. actual consent. People may genuinely consent to something they would not consent to under certain (hypothetical) conditions. More importantly, the fact that people would consent to something – hypothetical consent – does not mean they have in fact consented to it. ‘Hypothetical consent’ is a misnomer, as consent normally must be an act in order to be genuine; to consent to something must involve an act of will, a dated event. I am leaving out much detail, of course. I do not wish to claim that the three families of accounts are jointly exhaustive. My purpose is to say something about the general implications of the main kinds of accounts of the grounds of the state’s legitimacy. I shall limit myself to making two principal claims. The first is that a state’s satisfaction of any of the three accounts may be sufficient to secure weak or basic legitimacy. We can see easily enough how a state’s satisfying one of these
27 In my An Essay on the Modern State I require that states be relatively just and efficient in order to be legitimate. See also: Morris (2005) “Natural Justice and Political Legitimacy”, in: Social Philosophy & Policy 22, 1, pp. 314–329, which addresses the question of whether states necessarily are unjust.
28
C. Morris
conditions might plausibly give it a right to exist and to rule, with subjects having an obligation not to undermine it and to support it in various ways. Now it is likely that the satisfaction of different conditions may affect the particular obligations we have. For instance, consent might secure more demanding obligations than mutual benefit. The second claim is that only the third condition – genuine consent – is likely to ground a state’s strong or full legitimacy. This claim is more controversial, and I cannot adequately defend it here.28 Strong legitimacy entails that subjects have an obligation to obey the law, that is, to comply with each and every law that applies to them. I have assumed that such an obligation entails a pre-emptive reason to comply. It is hard to see how anything short of (genuine) consent could secure such a conclusion. This may be the emerging consensus in the literature. To claim strong or full legitimacy is to make a very strong claim. It is very hard to see how it could be secured by either of the first or the second kind of account. The obligation to obey the law entailed by the stronger kind of legitimacy is difficult to establish for reasons that have been widely explored in the literature. The benefits typically offered by just and efficient states, as great as they may be, are not of the sort that can establish that all valid laws are content-independent, pre-emptive reason for all subjects. The nature and scope of the quantifiers are crucial elements of the argument, even if the obligation to obey here is particularized to subjects of the law. Not all persons or humans are subjects, but all subjects have obligations to obey each and every law that applies to them. Were each subject to have consented (genuinely) to a state, then it may be that they would have the obligation to obey. But, as is well-known, no state has ever been able to obtain the genuine consent of more than a small proportion of its subject population. Suppose just and efficient states possess at most weak or basic legitimacy. Subjects, then, will have obligations not to undermine their state and to support it in various ways. We may suppose that the content and strength of these obligations will vary. For instance, citizens will have more obligations than mere subjects. But, as I noted earlier, everyone, independently of being a subject of a state, will have some obligations not to undermine and possibly even to support any state that possesses basic legitimacy. The mere existence of obligations of this kind does not distinguish subjects and outsiders. This is an interesting feature of the kind of account of (weak) legitimacy that I am offering here.
28
The original modern defense of consent in these contexts is, of course, John Locke’s political theory. Simmons’ view is “Lockean”. See his books on Locke: Simmons (1992): The Lockean Theory of Rights and: Simmons (1993): On the Edge of Anarchy.
State Legitimacy and Social Order
29
6 Legitimacy and Social Order While we may reasonably think of most of our constitutional democracies as states with weak or basic legitimacy, it is implausible to think that any of them is strongly or fully legitimate. Should this conclusion be cause for concern or despair? For the most part, I shall argue, it is not. There is, however, something paradoxical about this conclusion, and I shall make some concluding remarks about this. What effects might the fact that our states are not fully legitimate have on social order, assuming this fact to be common knowledge? I shall argue that the effects are not to be lamented. Consider our attitudes towards compliance with the law. There usually are many reasons why we might be moved to comply with a valid law that applies to us. We might think this law a good one, either intrinsically or because of its likely consequences. The law in question may apply a norm or principle we hold dear or are committed to, the violation of which would be shameful. We might worry about the possible consequences of non-compliance on our society, one of which might be encouraging others to take the law in question, or laws in general, less seriously. We might fear detection and the application of sanctions, as well as effects on our reputation. We may have a general policy of obeying the law unless there is some special circumstance clearly warranting disobedience. Some laws may be conventions in the special sense that the fact that others conform to them is a reason to follow suit. We may also comply with most laws ignorantly; we are not aware that we are conforming to them. This presumably is commonplace; most of us are unaware of most of the laws that apply to us and that we are not tempted to violate. It may also be that while we disapprove of the law in question, we think that our system of government to be a good one and that we should rarely disobey a valid law. Or we may identify as nationals of our state; that is, we think of our state as the embodiment of our nation, our nation-state – a kind of Verfassungspatriotismus. Our motives are often multiple or mixed. Any of these reasons may suffice to secure our conformity with the law on most occasions. And conformity is often, if not usually, overdetermined. But none of these reasons suffice to establish that all subjects of the law of a state have a content-independent, pre-emptive reason to obey each and every law that applies to them. The mere fact that a particular norm is a valid law of a weakly or basically legitimate state and applies to one need not be a pre-emptive reason for action. Consider the proverbial stop sign in the American desert. Traffic laws of this kind are important and worthwhile; we need to be concerned about cultivating habits of disregard for such laws. But suppose it is the middle of the day, no one is in sight, we are quite certain that the stop sign is not a trap. We slow down and drive through the stop sign without coming to a complete stop. We knowingly disobey the law. It is hard to believe that such behavior is action against a pre-emptive reason. Many laws are rules (as opposed to standards), and many apply in situations where there occasionally is little point in complying with them as long as one’s non-conformity does not have bad effects. (Other examples involve stupid laws or ones which apply equally to experts and the ignorant.)
30
C. Morris
Unlike some philosophers, I do not wish to deny that sometimes – perhaps quite often – a valid law is a content-independent, pre-emptive reason for action for most people. I am not sceptical or political authority per se; I merely think that it is less extensive than it is claimed to be. Consider three general possibilities regarding a law. A valid law is (1) always a content-independent, pre-emptive reasons to obey, (2) never a content-independent, pre-emptive reasons to obey, (3) sometimes a content-independent, pre-emptive reasons to obey. In all three cases other reasons for conformity still obtain. Were it the case that there never are any content-independent, pre-emptive reasons to obey, that would not affect the other reasons we may have to obey, for instance, because of possible effects on others or because of loyalty to our country. I assume that possibility 2 does not obtain with regard to states that are weakly legitimate. So the two possibilities to consider are 1 and 3. What are the likely effects to be of the realization that 3 and not 1 obtains? If the valid laws of weakly legitimate states are sometimes but rarely content-independent, pre-emptive reasons, then that will make a difference. But if they are often or even usually reasons of this kind, then the differences between 1 and 3 will be minimal, of little practical importance. One might think that widespread recognition of the fact that the valid laws of a weakly legitimate state are not always content-independent, pre-emptive reasons would encourage non-compliance. But, I repeat, the other reasons favoring compliance would still obtain. In addition, the fact that an act of non-compliance would encourage disobedience on the part of others might in some circumstances, when significant non-compliance would be bad, itself be a supplemental reason to comply. States will rarely, if ever, possess strong legitimacy, and I have suggested we are not worse off for this fact. This may be true, but there is still something odd or possibly paradoxical about this fact. States claim full legitimacy. The idea seems to be part of our modern conception of the state. Allen Buchanan has recently expressed puzzlement at philosophers’ expectations of states. He distinguishes political legitimacy and political authority, the former consisting the moral justification for exercising political power, the latter including in addition a right to be obeyed.29 The distinction Buchanan draws is similar to my distinction between weak and strong legitimacy.30 He argues that
29
The exercise of political power for Buchanan is “the (credible) attempt to achieve supremacy in the making, application, and enforcement of laws within a jurisdiction.” Allen Buchanan (2004): Justice, Legitimacy, and Self-Determination: Moral Foundations for International Law, ch. 5, p. 233. Note that this characterization will not fit political powers, e.g., in medieval Europe, which do not claim exclusive authority or ultimate power. Presumably it is mainly modern states that fit Buchanan’s characterization.
30 My current distinction is only implicit in the account offered in my An Essay on the Modern State. I argue in that book that even if a state is legitimate – in what I now call the weak sense – it will not possess the full set of powers it claims, including sovereignty and the power to create content-independent, presumptive reasons for all of its subjects. In recent work I have explicitly
State Legitimacy and Social Order
31
once we recognize how demanding the notion of political authority is, and how unconnected it appears to be with the important questions concerning the morality of political power. . . it is puzzling that some recent political philosophers seem to have assumed that an account of political authority must be a centerpiece of a viable political theory.31
There is nothing puzzling here. It is not merely “some recent philosophers” who have focused on what Buchanan calls political authority. It is the major political thinkers of modern times, starting with Hobbes.32 We could, I think, dub what I have called strong or full legitimacy classical legitimacy. More importantly, this conception of legitimacy seems the natural one to start with. The state after all does not say that we may obey some laws and not others if we decide this is best (without its authorization). It reserves for itself the exclusive right to determine or ascertain what our obligations are, and it designates these as laws. Buchanan’s comments suggests he does not take seriously the state’s claim to (what I call) authority. He may be right, but that does not alter the presumptions of states. It is possible, of course, to take the state’s claims seriously and to deny that they are all true. To say that a state’s legitimacy is merely weak or basic is to deny it the strong or full legitimacy it claims. I think that states lack strong or full legitimacy. So allowing for terminological and conceptual differences, I do not disagree with Buchanan’s conclusion – states lack what he calls political authority. But I think there is something odd about the fact that states are not strongly or fully legitimate. They could be, of course, but it is hard to see how they ever would be. And this is peculiar. For what it means is that our fundamental form of political organization is one that cannot measure up to its self-image. The way in which the state presents itself is, it would seem, mistaken. It claims powers it never fully possesses. We can get by and even do well without strong or full legitimacy, and there is nothing peculiar in our realization of this fact. But states cannot get by claiming less. That fact is peculiar. Acknowledgments I am very grateful to David Gauthier, J¨org K¨uhnelt, Neil Roughley and Bruno Verbeek for critical comments on an earlier draft.
drawn a distinction between weak and strong kinds of legitimacy in response to some comments John Simmons made in his review of my book: Simmons (2000): “Review of Morris’ Essay on the Modern State”, in: The Philosophical Review 109, 2, pp. 271–273. See also Simmons’ seminal essay: Simmons (2001a) “Justification and Legitimacy”, reprinted in his collection, Justification and Legitimacy: Essays on Rights and Obligations, pp. 122–157. 31
Allen Buchanan (2004): Justice, Legitimacy, and Self-Determination: Moral Foundations for International Law, p. 241.
32
For Hobbes, law is command, not counsel. “C OMMAND is, where a man saith, Doe this, or Doe not this, without expecting other reason than the Will of him that sayes it.” Leviathan, Ch. 25, para. 2.
32
C. Morris
References Buchanan, Allen (2004): Justice, Legitimacy, and Self-Determination: Moral Foundations for International Law, Oxford: Oxford University Press. Buchanan, James (1975): The Limits of Liberty, Chicago: University of Chicago Press. Copp, David (1999): “The Idea of a Legitimate State”, in: Philosophy & Public Affairs 28, 1, pp. 3–45. Edmundson, William A. (ed.), (1999): The Duty to Obey the Law Selected Philosophical Readings, Lanham (MD): Rowman & Littlefield. Gauthier, David (1986): Morals By Agreement, Oxford: Clarendon Press. Gauthier, David (1996): “Commitment and Choice: An Essay on the Rationality of Plans“, in: Farina, Francesco, Hahn, Frank, and Vannucci, Stefano (eds.): Ethics, Rationality, and Economic Behaviour, Oxford: Clarendon Press, pp. 217–243. Green, Leslie (1988): The Authority of the State, Oxford: Clarendon Press. Hart, H.L.A.: “Commands and Authoritative Legal Reasons”, in: Essays on Bentham: Jurisprudence and Political Theory, Oxford: Clarendon Press, 1982, pp. 243–268. Hobbes, Thomas (1991 [1651]): Leviathan, edited by Tuck, Richard, Cambridge: Cambridge University Press. Kagan, Robert (2004): “America’s Crisis of Legitimacy”, in: Foreign Affairs 83, 2, pp. 65–83. McClennen, Edward F. (1990): “Foundational Explorations for a Normative Theory of Political Economy”, in: Constitutional Political Economy 1, 1, pp. 67–99. McClennen, Edward F. (1993): “Rationality, Constitutions, and the Ethics of Rules”, in: Constitutional Political Economy 4, pp. 173–210. McClennen, Edward F. (2003): “Prudence and Constitutional Rights”, in: University of Pennsylvania Law Review 151, 3, pp. 917–961. Morris, Christopher W.: “Is the State Necessarily Coercive?” (manuscript). Morris, Christopher W. (1998): An Essay on the Modern State, Cambridge: Cambridge University Press. Morris, Christopher W. (2000): “The Very Idea of Popular Sovereignty: ‘We the People’ Reconsidered”, in: Social Philosophy & Policy 17, 1, pp. 1–26; Morris, Christopher W. (2002): “Sovereignty”, in: Clarke, Paul B. and Foweraker, Joe (eds.): Encyclopedia of Democratic Thought, London & New York: Routledge, pp. 673–676. Morris, Christopher W. (2005): “Natural Justice and Political Legitimacy”, in: Social Philosophy & Policy 22, 1, 314–329. Raz, Joseph (1986): The Morality of Freedom, Oxford: Clarendon Press. Raz, Joseph (1999 [1975]): Practical Reason and Norms, Oxford: Clarendon Press. Simmons, A. John (1979): Moral Principles and Political Obligations, Princeton (NJ): Princeton University Press. Simmons, A. John (1992): The Lockean Theory of Rights, Princeton (NJ): Princeton University Press. Simmons, A. John (1993): On the Edge of Anarchy, Princeton (NJ): Princeton University Press. Simmons, A. John (2000): “Review of Morris’ Essay on the Modern State”, in: The Philosophical Review 109, 2, pp. 271–273. Simmons, A. John (2001a): “Justification and Legitimacy”, reprinted in his collection, Justification and Legitimacy: Essays on Rights and Obligations, Cambridge: Cambridge University Press, pp. 122–157. Simmons, A. John (2001b): “Legitimacy”, Encyclopedia of Ethics, edited by Becker, Lawrence C. and Charlotte B., 2nd edn., New York and London: Routledge, pp. 960–963. Tucker, Robert W. and Hendrickson, David C. (2004): “The Sources of American Legitimacy”, in: Foreign Affairs 83, 6, pp. 18–32. Weber, Max (1946 [1919]): “Politics as a Vocation”, in: Gerth, Hans and Mills, C. Wright (tr. and eds.): From Max Weber: Essays in Sociology, New York: Oxford University Press, pp. 77–128. Weber, Max (1947): The Theory of Social and Economic Organization (Part I of Wirtschaft und Gesellschaft), Henderson, A.M. and Parsons, T. (tr.), New York: Oxford University Press.
The Claims of States and the Claims of Morality A Commentary on Morris Neil Roughley
In his insightful paper Chris Morris argues that modern states purport to possess legitimacy in both a weak and a strong sense, where legitimacy is understood to be a normative status, a feature that is not amenable to explication in empirical terms. Rather, it can only be explicated in the normative terms of the rights and powers it entails and the obligations with which these are correlated. The legitimacy claimed by states entails the rights to exist and to rule, rights to which correspond obligations on the part of the state’s subjects to refrain from undermining it and to obey its laws. The distinction between weak and strong legitimacy is a matter of the strictness of the obligation of a state’s subjects to obey its laws. In the strong case, subjects are obligated to obey each and every law; in the weak case, this is not so. However, because strong legitimacy could only result from genuine consent of all a state’s subjects and because no such universal consent is feasible, modern states suffer from a normative deficit by their own lights: they don’t possess the strong legitimacy they claim for themselves. This, Morris says, is peculiar, but unworrying for the question of social order, as there are plenty of other reasons for people to obey the laws of weakly legitimate states. As far as I can see, Morris and I are in broad normative agreement: I also take it that the subjects of modern democratic states can be obligated to obey the law or to desist from undermining the state. I also see no grounds to assume that the subjects of any historically existing states are obligated to obey each and every law. What I want to discuss in the following is the contention that states nevertheless claim that their subjects are thus obligated or, as I shall put it, that the obligations of their subjects to obedience are “exceptionless”. I shall raise questions that focus primarily on the nature of the claim thus ascribed. I shall also remark briefly on the conditions of its justifiability. One of the main motivations for the idea that modern states claim legitimacy is the refusal to see the state purely as a construction that amasses and concentrates causal powers to constrain the inhabitants of a certain territory.1 Thus viewed, states would have the same status as organised criminal organisations, except that their structures would be more public and their workings presumably more effective. We
1
Morris: this volume, p. 16.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
33
34
N. Roughley
can assume that this is not the self-understanding of the majority of the people whose job it is to represent the state. Now, the view for which Morris argues is not that the representatives of our states tend to see the structures within which they work as having a normative status that goes beyond that of a monopoly of force. Rather, he asserts that such a claim is made by the states themselves. My first query is quite simply what it means to say that a state is the author of such a claim. According to Morris, the claim is made as part of “the way in which the state presents itself” or its “self-image”.2 What this might mean is that, if we survey the legal proclamations, particularly the constitutions, of modern states, we will always find passages containing claims of this ilk. According to a somewhat vaguer version, Morris’s assertion could be verified by seeing the way in which the official talk of all sorts of representatives of modern states tends to imply some such thing, even if it doesn’t involve any such explicit statement. It would be part of the rhetoric generally employed by such people. Or perhaps the idea of “claiming” at work here is not to be taken literally as a form of linguistic behaviour, but rather as in some way “performatively” implied or implicated by the kind of institutions that modern states set up. According to these variants, the “claim” would be a generally present empirical feature of the workings of the modern state. Another way of interpreting Morris’s assertion would be to understand it as conceptual in nature, an understanding perhaps suggested by the statement that “the idea seems to be part of our modern conception of the state”.3 This might appear somewhat strange, as it would seem to entail that any modern political organisation that didn’t claim legitimacy would fail to be a state. Or would it perhaps just be a pre-modern state? Certainly, pre-modern forms of social organisation tended strongly to be accompanied by explicit claims on the part of their rulers to be invested with authority by divine agency. Is there a sense in which modern states’ claims to legitimacy – as opposed to the grounds for those claims – are of an essentially different kind to those of, say feudal lords? It would be helpful, then, if we had somewhat more criterial specificity as to the idea of state authorship of claims. It will emerge why I think that this is important. The main issue I want to focus on, however, concerns the kind of claim states make, in particular the construal of the central notions of rights and obligations as “statuses”. To begin with, the example Morris offers in order to clarify what is meant by a normative status to my mind raises more questions than it answers: the use of the term “bastard” a couple of decades ago involved assigning a specific normative status, above all the denial of certain rights, on the basis of the assignee’s failure to satisfy certain empirical conditions.4 The structures that both made the relevant rights available and set the conditions for their conferral or denial are perfectly clear: they are a matter of legal structures that had been institutionally established at a specific historical juncture and which have in the meantime been more or less 2
Ibid., p. 31.
3
Ibid., p. 30, my emphasis.
4
Ibid., pp. 18–19.
The Claims of States and the Claims of Morality
35
dismantled. In general, within the purview of a state, subjects have rights that can be made sense of in terms of the correlative obligations towards them on the part of other subjects, where the relevant statuses are all conferred by the institution. Where, however, we are dealing with rights of the state itself, there seems to be little chance of such institutional conferral. Rather, it seems that, for a state to be able to claim legitimacy in Morris’s sense, it would somehow have to be established in advance that fulfilling certain empirical criteria would confer the relevant bundle of rights and obligations. But what might it be that establishes such status-conferring mechanisms? The most obvious candidate here would appear to be morality. It might be argued that the rights to exist and to rule claimed by the modern state, along with the correlative obligations of its subjects toward the state, are moral statuses. According to this conceptualisation, morality would both make the relevant concepts of right and obligation available and stipulate the criteria the satisfaction of which confers those statuses. What the modern state would then be claiming would be that it satisfies those pre-political criteria. Call this “the moral statuses construal”. Morris explicitly rejects such a moral construal of the relevant rights and obligations. He offers three reasons for this,5 none of which I find convincing. Firstly, he says that the distinction between the moral and the non-moral is not easily located. I agree. But vague boundaries don’t make for an invalid distinction. Secondly, Morris argues that the predicates “moral” and “legal” as applied to obligations (and rights) should not be seen as designating species, but rather sources of such statuses. The obligations thus generated should therefore be understood to be simply obligations without qualification. However, I fail to see how either talk of obligations simpliciter or replacing the idea of species with that of source helps to illuminate what is at issue. What these moves might be aiming to establish is the idea that all obligations are in some sense “on a par”, independently of the sort of considerations that generate them. Perhaps, then, we should think of the relationship between source and product in this case as analogous to that between the cause, say, of a fire and the fire itself, where the rational way to deal with the fire may be independent of how it was started. But this would be extremely implausible relative to obligations. If certain obligations have their “source” in morality, whereas others don’t, then surely this is important for the way in which a practical reasoner should treat them. How a rational subject should deal with the claims on her raised by different systems for the orientation of her action, particularly when they conflict, will depend on how much weight these systems are to be accorded. That in turn will depend on the kind of considerations – moral or other – that give rise to, and are given normative shape in these obligations. A third reason Morris adduces for rejecting the moral statuses construal is supposed to derive from the state’s claim to “supremacy”, that is, from its claim to provide reasons that override all others, including moral reasons. We miss this “extraordinary” point, Morris claims, if we assume that reasons of justice (moral
5
Ibid., pp. 21–22.
36
N. Roughley
reasons) always override legal reasons. I would have thought that the opposite is the case: wouldn’t a theorist who makes precisely that assumption be particularly likely to notice that states deny it? If, that is, they do: were the moral statuses construal to be correct, then the claim to authoritative exclusivity raised by states would itself be a moral claim. One reason why it may seem that this cannot be right is that the reasons frequently thought to speak against accepting the state’s authority – whether in general or in particular instances – are themselves moral reasons, access to which may be thought to be reserved for the individual agent’s conscience. But the state’s claim to authoritative exclusivity tolerates no competing grounds, which, so it may seem, reasons of conscience would be if both were to be situated within the moral domain. However, it is unclear why this should be thought to be an argument against the moral statuses construal. To begin with, the conflict between different sorts of claims is obviously not a sufficient reason for seeing them as belonging to different normative spheres, or as having different normative “sources”. Claims grounding in the demands of benevolence and fairness can conflict, but that doesn’t prevent them both being moral. More to the point, it is not internal to the demands of fairness that they are only pro tanto considerations that can be relativised by other moral reasons. Rather, the possibility of their relativisation by other considerations depends on the adoption of a broader moral perspective within which they are given a certain amount of weight. That is, within moral reasoning, we have to step outside the bounds of fairness in order to decide whether, in some particular case, reasons of fairness should be the exclusive arbitrators of what is to be done. But that doesn’t change the fact that, as long as we are attempting to work out what would be fair, only relevant considerations such as equality of opportunity or distribution are to be taken into account. Most importantly, the moral statuses construal would insist on a distinction between the moral point of view and the point of view of an institution that purports to satisfy the conditions which, according to morality, confer certain rights on that institution and certain obligations on its subjects. According to such a conception, the state should be seen as making two distinct claims. The first is that it indeed satisfies the empirical conditions that, by the lights of morality, confer legitimacy. The second is a claim concerning the content of the statuses morality confers where those conditions are satisfied. If Morris is right that states claim that their subjects are obligated to obey every law that applies to them, then all the moral statuses construer needs to do is to deny that this is the content of the obligations conferred by satisfying the relevant empirical conditions. The modern state, so the argument would run, claims certain moral statuses that morality wouldn’t confer whatever empirical criteria were to be fulfilled. The state’s claim to “supremacy” would not fall outside the moral sphere; it would be a moral claim that is morally unacceptable. If this were indeed to be the structure of the phenomenon under investigation, there would appear to be nothing particularly “peculiar” about it. We are perfectly familiar with cases in which institutions or persons legitimately invested with a moral function overestimate the scope or extent of the authority bestowed by that function. For instance, this seems like a fair description of the way many parents and
The Claims of States and the Claims of Morality
37
educators have historically exercised their functions. So is Morris wrong that there is something “peculiar” about the way in which modern states “present themselves”? There is one way in which the moral statuses construer could find himself confronted with an uncomfortably strong variant of the sense of paradox Morris wants to communicate. He would do so if, returning to my introductory query about the kind of claim modern states make, he interpreted it as conceptually necessary. If the modern state necessarily claims that its citizens are obligated to obey every law in every situation in which it applies to them, and if this claim involves the state’s overestimation of the authority invested in it by morality, morality would appear to be internally incoherent. That result would be so peculiar that it would count strongly against the moral statuses construal. So anyone with such a view would have to deny that there is anything conceptually necessary about the state’s claim to impose exceptionless obligations to obedience. But there are good independent grounds for doing so. I suggest that the apparent necessity with which the modern state makes such a claim is best explained in pragmatic terms: it is simply very hard to imagine that the directives issued by a central institution which aims at coordinating the behaviour of all those subject to it could be effective if those directives were given explicit pro tanto or conditional formulations. Such directives would, for empirical reasons, inevitably give rise to confusion and abuse, at least where we are dealing with organisations on the scale of modern states. Perhaps, then, the public claim to command obedience to all laws is best understood as a pragmatic device employed in order to achieve some level of empirical approximation to the function states ought morally to fulfil. This would tie in with the fact that liberal states tend to open up some amount of space within which the stringency of the enforcement of certain laws is, more or less officially, left up to the judgement of individual law enforcers. People who cross a street in spite of the prohibitive little red man on the side of a traffic light may find themselves treated in very different ways by passing police, just as considerable leniency may be shown towards people who, according to the letter of the law, ought to be prosecuted for minor drug offences.6 Where such forms of leeway are semi-officially established in the practices of the keepers of law, might we not be justified in interpreting the claim made by the state as demanding less than obedience to every law that applies to one’s behaviour? In other words, would not a more performative interpretation of the idea of a state’s authorship of claims enable us to avoid saddling the state with a claim that cannot conceivably be acceptable? For obvious reasons of space, Morris develops no extended argument for his claim as to the grounds for legitimacy. Nevertheless, I would like to make a concluding remark on what he does say about the relationship between universal consent and the obligation to obey every law. He sensibly recommends that we keep track of the nature and scope of the universal quantifiers at work here.7 However, in attempting to do so, I find myself unclear as to why we should think that the
6
Compare Morris’s own examples: this volume, p. 25 and p. 29.
7
Morris: this volume, p. 28.
38
N. Roughley
universality of the subjects of consent transfers to the universality of the objects of obedience. Assuming here that universal consent is indeed the sort of criterion necessary to establish legitimacy, why should one think that even it, should it ever be forthcoming, would place all consenters under obligations to obey every single law? If it is correct that consent generates obligations, then it might seem to follow that the consent of all generates obligations for all. But why should that be thought to entail that the content of those obligations specify obedience to every single law? If what is consented to were to be a particular coherent set of clearly identifiable norms, then it might seem clear that one has subjected oneself to governance by those norms in every situation they cover. However, Morris leaves no doubts that this is not what he means by the phrase “consent to a state”. Following Hart and Raz, he insists that the norms to which exceptionless obedience would be demanded under full legitimacy are “content-independent”, that is, norms whose authority results from the pedigree of their authors, not from the acceptability of what they prescribe. But what sort of features might a state possess consent to which would substitute for consent in individual cases? The best candidates here are procedural features characteristic of democratic states: the separation of powers, the possibility to vote out lawmakers whose products appear unacceptable and so on. But why should universal consenters to these and other features be thought thereby to abrogate their moral rights to refuse compliance if a system with such procedural virtues nevertheless produces laws that turn out to be stupid or even just over-ambitious in scope? Even the best possible procedure – one that everyone should consent to as a system – is liable to have an output elements of which are unacceptable. As I read Morris, he and I agree on this. The point of disagreement I want to conclude by noting is that, where Morris takes it to be implausible that the empirical conditions for “full legitimacy” could ever be met, I see no reason for assuming that the satisfaction of any empirical conditions could invest a state with the statuses Morris picks out with the notion of full legitimacy.
Legitimacy and Justice A Conceptual and Functional Clarification Wilfried Hinsch
1 Two Concepts of Legitimacy It is helpful to distinguish between concepts of legitimacy and conceptions of legitimacy.1 A concept of legitimacy gives us the basic meaning of the term “legitimate” and tells us what we say about, e.g., a social rule or a political decision when we call it legitimate. It also specifies the role of “legitimacy” in social theory and practical political deliberation. Conceptions of legitimacy, on the other hand, specify the criteria that have to be met by a rule or decision in order to actually be legitimate in the relevant sense. It is important to distinguish clearly between two different concepts of legitimacy: the empirical concept of the social sciences and the normative concept of political philosophy. It seems that the empirical and the normative concept of legitimacy are easily confused. Political commentators are prone to vaccilate between them. There are two reasons for this: First, the empirical and the normative concept have in common that both take “legitimacy”, “legitimate”, and their cognates to be terms that indicate normative status. In one way or other, legitimate rules and decisions are characterized by the fact that they have (or that they are believed to have) normative authority.2 They are conceived of as binding by those who ascribe legitimacy to them, i.e., as imposing duties and obligations.3 Second, in line with both concepts “legitimacy” applies primarily in the domain of the political and relates to the exercise of coercive state power. Assuming that legitimate rules and decisions have normative authority and impose corresponding obligations, we think it appropriate that they be enforced by state action should that prove necessary. In any case, legitimacy is taken to be a necessary condition of any justifiable use of state power. We ascribe legitimacy to monarchs, national and international regimes, laws, or parliamentary decisions. I shall confine myself mainly to discussing the legitimacy 1
The explicit distinction between “concepts” and “conceptions” has been introduced by Rawls (1993), p. 14n; Rawls (1999), p. 5; who gives credit for it to Hart (1994), p. 160. 2
Cf. Flathman (1993), p. 527.
3
For a general account of normative authority, see Raz (1979) and (1986), part one.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
39
40
W. Hinsch
of norms. This helps to keep things manageable. Moreover, norms seem to have a certain priority. Whatever we may wish to call legitimate in the domain of the political is either itself a norm – or something made up of norms like institutional arrangements – or it is legitimate in virtue of being in conformity with certain norms. Hence, to understand the legitimacy of norms is basic to understanding political legitimacy in general. The empirical understanding of legitimacy goes back to Max Weber whose understanding of legitimacy is still looming large in the social sciences and enjoys also some prominence among political analysts and commentators. There are, however, alternative empirical conceptions, for instance Niklas Luhmann’s account of procedural legitimacy, which is explicitly critical of Weber but shares the latter’s concern with empirical meaning and explanation.4 For the purpose of conceptual clarification, a basic understanding of Weber’s conception of legitimacy should suffice.5 Roughly speaking, Weber’s view is that a norm or an institutional arrangement is legitimate if, as a matter of fact, it finds the approval of those who are supposed to live in this group. “Approval” in this context is to be understood as “sincere recognition” of a norm as providing an appropriate standard of conduct that must be complied with because it is the correct or right standard and not only because of the threat of social sanctions. Sincere approval of norms constitutive for a social order by the participants turns an order into what Weber called “geltende Ordnung”, i.e., into an order that has normative authority for those involved. To call an institutional arrangement legitimate thus implies that there are people who sincerely believe that these arrangements should be regulative for their conduct irrespective of external sanctions. Legitimate arrangements are, therefore, arrangements that receive voluntary support. Since no political regime or social order could persist without a fairly high level of unenforced compliance with its rules, the empirical understanding of legitimacy naturally occupies a central place in explanatory theories of social order. Notwithstanding an air of higher moral ambition that may be associated with the term “legitimacy,” the empirical concept does not imply any evaluative or normative commitment on the side of those who, as external observers, use it to describe a social order. One may maintain that a particular arrangement is legitimate in the empirical sense – i.e., that it finds the approval of those who have to abide by it – and still believe that it is highly unjust or unacceptable for other reasons. “X is legitimate” taken in the empirical sense implies that there are people who approve of x, but the person making the statement (say a social scientist) need not be one of them. Given the empirical understanding, to say that an institutional arrangement is a legitimate one is to make a factual claim about the subjective state of mind of particular individuals that belong to one political society. This is how Weber uses
4 5
Cf. Luhmann (1969), chap. I.2.
For the basics, see Weber (1922), ch. 1, 5–7. A detailed discussion of Weber’s conception would go beyond the scope of this paper. For Weber and for a more comprehensive account of legitimacy in general see Beetham (1991), part one.
Legitimacy and Justice
41
the term “Legitimit¨ats-Glaube,” to refer to the subjective basis which makes the crucial difference between mere social regularity and expediency, on the one side, and social order based on recognized rules, on the other.6 And Robert Dahl makes the same point when he distinguishes between an ethical and a psychological sense of “legitimacy” and explains that the latter simply is “a belief in the rightness” of a decision or process of decision making.7 Let us now turn to what I have called the normative concept of legitimacy. This is central for the understanding of legitimacy, e.g., in the work of J¨urgen Habermas (1988) and John Rawls (1993). The normative concept differs in at least two ways from the empirical one. First, it involves “objective” criteria of legitimacy that are alien to the empirical concept. To say that an institutional arrangement is legitimate in the normative sense is not simply to say something about the subjective state of mind of particular individuals. It is to say that the arrangement meets certain substantive requirements – say standards of justice and rationality expressed in a normative conception of legitimacy – irrespective of whether people believe that they are met or not. Mere factual approval of social rules and institutions, therefore, does not suffice to lend normative legitimacy to them, though in case the approval is sufficiently widespread and stable, the empirical conception of legitimacy may apply. Objectivity here is to be understood in a weak sense. I do not mean to say that the normative concept of legitimacy involves uniquely correct or true criteria of legitimacy. On the contrary, I assume that there are various incompatible, normative conceptions of legitimacy specifying different criteria for legitimate norms. In order to employ a normative concept of legitimacy, we are not tied to any particular conception of legitimacy. We do not even need to assume that all but one of these conceptions are wrong. What we do have to presuppose, however, is that legitimacy is not simply a matter of subjective belief but one of actually meeting substantive requirements and, in particular, requirements of justice and rationality. Second, the normative concept of legitimacy involves moral approval on the side of those who apply it to a rule or a system of rules. To say that an institutional arrangement is legitimate in the normative sense is to publicly recognize that it has moral standing, and not simply to report that there are people who believe that the arrangement has moral standing. The difference between normative and empirical legitimacy is that between (1) expressing a normative belief that one actually holds and thus committing oneself to the normative and inferential consequences of holding that belief and (2) stating that there are people who hold that belief without committing oneself to the truth (or validity) of this belief and its inferential consequences, on the other. Normatively speaking, legitimate arrangements are not simply schemes of social rules we (or others) approve of, but arrangements that have normative authority in actually imposing obligations on participants. Any particular normative conception of legitimacy, therefore, has to expound its substantive criteria of legitimacy in a way that explains why meeting these criteria actually confers
6
Cf. Weber (1922), 6.
7
Cf. Dahl (1956), p. 47n.
42
W. Hinsch
normative authority on norms, institutions, or persons. Thus, a normative conception of democratic legitimacy, for instance, has to be laid out in a way that makes intelligible the normative capacity of democratic procedures to generate morally (as opposed to merely legally) binding rules.
2 Contractualism I shall now turn to a particular normative conception of legitimacy, the contractualist conception, and then discuss the relationship between the concepts of justice and legitimacy in the next section. The contractualist ideal of a just and legitimate social order is well expressed by Rawls’ conception of a well-ordered society. A well-ordered society is a fair system of social and political cooperation which is effectively regulated by a public conception of justice. More specifically, the following conditions must be fulfilled: (1) All political and social institutions conform to principles of justice which are, after due consideration and for good reasons, recognized by all citizens. (2) Everyone acts in conformity with the public conception of justice. (3) Everyone knows that conditions (1) and (2) are satisfied, and this is itself common knowledge. (4) A consensus exists as to the justification of the first principles of justice.8 Two features of a well-ordered society are worthy of special attention. First, all citizens willingly comply with its laws and with fair settlements in general. In a well-ordered society, what game theorists call “the compliance problem” has been solved. It is a stable and hence highly efficient system of social cooperation. Second, the institutions and rules governing such a society are recognized by all citizens as being just and legitimate, not only de facto but for “good reasons.” An easily recognizable causal relationship obtains between the two features since the rational beliefs held by the citizens of a well-ordered society as to the justice and legitimacy of its institutions provide the empirical basis for its stability. Unlike stability-oriented considerations, however, the “good reasons,” which are presumed to justify the citizens’ beliefs, raise a genuine philosophical question: the problem of justifying political principles and institutions. Contractualist conceptions of justice and legitimacy are based on a notion of public justification.9 Presupposing the idea of free and equal democratic citizenship, it is assumed that all members of a society have an equal claim to live in a society the norms and institutions of which they can freely endorse in the light of their wellconsidered beliefs and interests. To expect a person to comply with rules that seem arbitrary from her point of view is normally inconsistent with recognizing her as a free and equal person with a firm intention to cooperate fairly. Each citizen has the same claim to live in accordance with her well-considered opinions and interests. It is unreasonable to expect a person to recognize (as opposed to merely comply with) 8
Rawls (1993), p. 35ff., p. 66f
9
See Rawls (1993), lect. I & VI; Rawls (1999), p 115; Hinsch (2002), Kap. 5.
Legitimacy and Justice
43
rules that have been established against her will and which must appear arbitrary and partial from her point of view. The rules of fair social cooperation are ideally rules that can be rationally accepted by all citizens who are supposed to comply with them. The requirement of publicity incorporated in the idea of a well-ordered society may seem important for several reasons, one of which is stability. Stable political and social cooperation presupposes that the parties involved not only recognize and obey certain rules but that they also be aware of the fact that each of them is doing so. This is guaranteed by the condition of publicity. Moreover, in recognizing fair principles of cooperation, citizens make manifest to one another that they respect each other as partners with equal rights, which again contributes to stability. More importantly, publicity and public justification matter because of the difficulties in realizing the value of autonomy in the face of the “fact of pluralism.” From a philosophical point of view, the primary issue is not that only publicly recognized norms are capable of establishing a stable scheme of cooperation. Rather, it is that citizens as autonomous moral agents have a claim that the norms regulating their mutual affairs be norms they can reasonably accept in the light of their well-considered beliefs and interests. Given the fact of pluralism, the requirement of public justification means that basic norms of a well-ordered society must secure the consent of citizens whose moral, philosophical, and religious views are at least partially incompatible. It is thus not inconceivable that no publicly justified political principles can be found. This, of course, would be fatal for the contractualist approach. If the contractualist ideal of a just and legitimate political order is to amount to more than an elusive utopia, two things must be shown: First, that principles of political justice can be formulated upon which citizens who hold incompatible moral and religious views might reasonably be expected to agree, and, second, that the content of these principles can be specified through the processes of democratic decision making in such a way that the resulting laws and institutions may also expected to find reasonable approval. The first problem concerns the most basic normative foundations of a well-ordered society. It is resolved when a conception of political justice is found which can be publicly justified. The second problem concerns the legitimization of particular political decisions regarding the concrete institutional setup of a well-ordered society. From a contractualist point of view, the latter problem is solved once a conception of political decision making is found that is capable of lending normative authority to specific norms that go well beyond basic moral principles and address concrete questions of social organization.
3 Justice vs. Legitimacy “Justice” and “legitimacy,” when applied to political institutions, are often treated as interchangeable concepts. This seems reasonable insofar as legitimate political decisions are first and foremost decisions that meet the most basic requirements
44
W. Hinsch
of political justice. We should be aware of the fact, however, that we are dealing with two normatively distinct, if related, concepts. Principles of political justice are identified by means of rational argumentation which, in accordance with the contractualist approach, takes the form of a public justification of norms. If, after due consideration of all the relevant facts, values, and principles, a reasonable disagreement concerning first principles of justice for political institutions persisted, then the notion of political justice would, for precisely this reason, remain indeterminate and empty. In that case, we would either have to forgo the attempt to establish institutions that are binding for all or continue on the basis of a mere modus vivendi. Hence the contractualist conception of justice presupposes that there are at least some very basic moral principles and norms that are beyond reasonable disagreement. But even if one were to succeed in identifying publicly recognized principles of justice, it would be mere fancy to expect that the reasonable consensus obtained could be extended to cover each and every particular political decision. Even under the admittedly idealized condition that the process of reaching decisions be fully rational and reasonable, we must face the fact that in many cases justified differences of opinion will continue to exist as to how a political problem may be solved as expediently and fairly as possible. In other words, we must take into account the possibility of reasonable disagreement. Let us say that, as a matter of ideal theory, we have a situation of reasonable disagreement if the following asymmetry obtains: All parties involved have sufficiently strong rationally and morally sustainable reasons to support their respective claims and positions, but no party is capable of putting forward generally acknowledged reasons strong enough to eliminate all competing claims and positions of the other parties. Given the fact of reasonable pluralism, any reasonable contractualist political and moral theory that proceeds from the idea of publicly justified norms must, therefore, accommodate this fundamental possibility and incorporate non-argumentative forms of collective decision making in order to meet the requirements of public justification. Ex hypothesi in a situation of reasonable disagreement no further argument could yield a reasoned consensus on a contested question. Whenever an agreement not to agree would not do, because there is no way to avoid a collectively binding regulation, the argumentative deadlock in cases of reasonable disagreement is a reason to employ non-argumentative forms of collective decision making (e.g., setting up a lottery or having a vote). In view of the contractualist requirement of public justification, this seems perfectly reasonable as long as the chosen non-argumentative procedure is sufficiently efficient and fair and finds the approval of those involved. Given the fact of reasonable disagreement and given the unavoidability of a collectively binding decision, all parties involved have a reason to accept fair and effective forms of non-argumentative decision making and to recognize the resulting decisions as publicly justified and binding. The contractualist idea of public justification thus includes argumentative and non-argumentative parts, the former being concerned with questions of political justice, the latter with questions of legitimacy.
Legitimacy and Justice
45
If all moral and political disagreement could, at least in principle, be traced back to factual errors or logical inconsistencies, a non-argumentative procedure like voting could, with regard to the moral justification of rules, never be more than a pragmatic way of dealing with ignorance, irrationality, or limited resources of other kinds. Sure enough, given the constraints of real life, voting even then would provide us with a morally attractive way of dealing with disagreements and conflict. Normally, voting is morally preferable to having a fight, and most of the time the attractions of secession are limited too. Taking a pragmatic perspective, however, the mere fact that a vote with a certain result was would, by itself, not make the result, say a particular regulation, a morally appropriate or correct result. If we proceed from the idea of reasonable disagreement, however, voting becomes a necessary part of the public justification of social arrangements and does indeed endow these arrangements with moral standing and normative authority. The notion of what is politically legitimate, as distinct from what is politically just is hence only partially determined by rational argumentation. As far as constitutional democracies are concerned, its content is fixed to a much greater extent by voting, typically by majority decision. As a result, political legislation may in many cases be put legitimately into effect, against which reasonable objections can be raised, at least on the part of some citizens. Such controversial decisions cannot be fully justified on the basis of substantive argument alone but only by appealing to the fact that they are the result of a fair decision-making process, recourse to which appears publicly justified under the given conditions. A conception of normative legitimacy, therefore, presupposes an account of the conditions under which collectively binding decisions can be reached by the factual working of non-argumentative procedures like voting. To say, then, that a norm is legitimate (normatively speaking) is to say not only that the norm meets certain requirements of basic justice but also that it has a certain factual pedigree. The factual or pedigree element is crucial for a concept of normative legitimacy and hence shared by all conceptions of normative legitimacy, be they as far apart as the dynastic conception of legitimacy of, say, a Talleyrand and the radical democratic conception of a Rousseau. Still, the procedural, non-argumentative part of legitimacy and its argumentative, non-procedural part that relates to the basic requirements of political justice do not have equal standing. Given the contractualist idea of public justification, rational argumentation is prior to all non-argumentative forms of decision making. Whether a particular political procedure (say majority vote) is a fair procedure and meets the requirements of political justice is not itself to be settled by voting. It is to be settled by arguments that involve substantive notions of political justice. The public justification of social norms and institutional arrangements by means of nonargumentative procedures thus presupposes a reasoned consensus on the principles of justice that specify ideas of political equality, determine the rights of political participation, and also define the bounds within which collective decisions may intervene with individual liberty.
46
W. Hinsch
4 Proto-Rights Having discussed the concepts of justice and legitimacy in the last section, we are now in a position to apply the distinction between the two concepts to a notorious difficulty found in liberal theories of constitutional democracy. It is the function of the constitution of a democratic society to set up an institutional framework for collective decision making and to enunciate the conditions under which voting as a non-argumentative procedure may issue in legitimate legislation, i.e., in political decisions that have normative authority and impose corresponding obligations. The contractualist ideal of political legitimacy would then be realized only to the extent that concrete political decisions are arrived at in conformity with a constitution that can be publicly justified. See, e.g., Rawls’ liberal principle of legitimacy: our exercise of political power is fully proper only when it is exercised in accordance with a constitution the essentials of which all citizens as free and equal may be expected to endorse in the light of principles and ideals acceptable to their common human reason.10
A serious flaw of this formulation of liberal legitimacy is that it already presupposes a constitution as a given framework for the legitimate exercise of political power. The principle thus fails to account for what gives legitimacy to the political constitution of a society itself. Moreover, J¨urgen Habermas has pointed out that there is an unresolved tension here between the principle of popular sovereignty and the idea (found in the work of Rawls and others) that – as a matter of political justice – there are certain constitutional liberties and rights that have a normative authority irrespective of popular sovereignty.11 Indeed, if we assume – In line with the mainstream of contemporary political philosophy – that for the most part the requirements of political justice are given by a set of basic civil and political liberties, we have to conclude that being part of the content of political justice, the content of these basic liberties cannot be fixed by vote or any other form of non-argumentative decision making. Questions of justice are to be settled by argument and not by vote. This, however, would mean that the most important elements of modern liberal constitutions – the basic political and liberal rights incorporated in these constitutions – remain excluded from the processes of democratic decision making and hence out of reach of popular sovereignty. The philosopher kings, it seems, have taken over the pouvoir constituante, leaving the democratic sovereign nothing but to work out the legal details of political organization. It seems possible, however, to address the misgivings of Habermas by denying that the rights and liberties guaranteed by Rawls’ first principle are actually constitutional rights.12 Pursuing this line of argument, I propose to distinguish between proto-rights and constitutional rights. Proto-rights are part of the content of
10
Rawls (1993), p. 137.
11
Cf. Habermas (1995).
12
Keep in mind, though, that Rawls takes these basic rights and liberties to be constitutional rights: Rawls (1993), p. 228f.
Legitimacy and Justice
47
principles of justice and can be established within a theory of political justice solely by substantive philosophical argument. Constitutional rights, on the other hand, are those (rights) familiar from the political constitutions of liberal democracies. They have been established by the procedures of democratic legitimization and hence may be seen as a result of the exercise of popular sovereignty. Proto-rights and constitutional rights differ in various regards. First, proto-rights are moral and not legal rights. Individuals have them independently of whether these rights are publicly and institutionally recognized in their society or not. Second, proto-rights are established solely by means of substantive philosophical argument and not by any kind of a legislative procedure that involves non-argumentative steps of collective decision making. Thirdly, proto-rights are noninstitutional rights in the sense that individuals can have them even though these rights are not backed up by institutional structures that guarantee their appropriate administration and enforcement. Finally, proto-rights are less specific than constitutional rights in the sense that different constitutional rights (say of freedom of speech) may guarantee the same proto-right. As opposed to the proto-rights incorporated in first principles of political justice, constitutional rights are (1) legal rights that are (2) based not only on arguments but also on procedures of constitutional law making that (in a democracy) involves voting as a non-argumentative way of collective decision making. Furthermore (3), constitutional rights are institutional rights backed up by a framework of legal structures that guarantee their appropriate administration and enforcement. And (4) they are more specific than proto-rights in the sense that different constitutional rights may protect the same proto-rights. The last point deserves our special attention because it allows to resolve the alleged conflict that radical democrats like J¨urgen Habermas see between the claim of political philosophy that the content of political justice – insofar as the basic liberties are concerned – is to be fixed solely by substantive moral argument, and the claim of popular sovereignty that it is the people who determines (say in a constitutional convention) the specific scheme of basic political and civil liberties that should effectively regulate the mutual political relations between free and equal citizens. If we assume that schemes of constitutional rights are always more specific than the proto-rights of political justice, then these schemes cannot be fully specified by means of (those) arguments providing the basis of the proto-rights. Furthermore, if all the constitutional schemes that incorporate the proto-rights of political justice are supposed to be equally just schemes, (then there may be) reasonable disagreement may arise as to which specific scheme should be established in a particular society. In order to give normative authority to any such scheme, we therefore need a non-argumentative way of collective decision making which, if we believe that there are equal proto-rights of political participation in a society of free and equal moral persons, will be a procedure of democratic decision making. I shall now illustrate the legal and philosophical relevance of the distinction between proto-rights and constitutional rights by discussing briefly the constitutional rights of freedom of speech in the U.S. and in Germany.
48
W. Hinsch
5 Freedom of Speech Freedom of speech and freedom of the press is guaranteed in plain words and without restriction by the U.S. constitution’s First Amendment of 1791: “Congress shall make no law [. . .] abridging the freedom of speech, or of the press.” The formulations in Article 5 of Germany’s Grundgesetz of 1949 are more differentiated. They reflect the technological advances that have changed the forms of public exchange since the 18th century. Article 5, paragraph 1, guarantees the right to free expression and dissemination of opinions “in speech, writing and pictures,” freedom of press and freedom of “reporting by means of broadcasts and films” as well as freedom to “inform oneself without hindrance from generally accessible sources.” The German Grundgesetz, however, does not give a guarantee without qualifications. Article 5, paragraph 2, specifies as “limits” to these freedoms the general laws, the protection of young persons, and the right to personal honor. Nevertheless, the obvious differences in the wording of the two constitutional texts do not in themselves justify the conclusion that the constitutional rights guaranteed therein are not the same. In comparing the constitutional rights of different countries, we must take into account not only the relevant constitutional documents but also established institutional practices of law interpretation and law enforcement. If we do this, it appears there are, indeed, important commonalties between the U.S. and the German rights to freedom of speech, but it also becomes obvious that significant differences exist.13 Even though the First Amendment does not mention broadcasts and films, the U.S. Supreme Court did interpret the term “speech” in the 20th century consistently in a way that included the expression of opinion in films, radio, and also in television.14 Moreover, freedom of speech does not enjoy absolute protection in the U.S. constitution, either. In Virginia vs. Black et al.,15 the Supreme Court rules, for example, that speech aiming at harassment or intimidation of fellow citizens does not enjoy protection by the U.S. constitution. Three congruencies of U.S. and German constitutional freedom of speech rights seem to be of particular importance in this context: First, in both constitutions, freedom of speech occupies a prominent position, and an extensive protection of public speech is regarded as an indispensable prerequisite of a free and democratic society. Second, in both constitutional orders freedom of speech is treated not only as a necessary condition of democratic self-government but also as an indispensable prerequisite for the development of the human personality, while, in the reasons stated in court decisions, it is the importance of freedom of speech for democracy
13
My still rather rhapsodic comparison between the constitutional law of free speech in the U.S.A. and Germany has greatly benefited from Tribe (1988), p. 185ff., Kommers (1989), p. 366ff., Grimm (1995), and Eberle (2002).
14
Cf. Brugger (2001), p. 157f.
15
538 U.S. (2003), p. 1ff.
Legitimacy and Justice
49
that is often privileged.16 Third and last, it holds for both orders that restrictions on freedom of speech in concrete cases may not be justified directly by the content of speech but require that hazards for other constitutional values – e.g., the public order or (in Germany) human dignity – be proven to be immediately involved. In particular, speech may not be restricted only because it expresses controversial political opinions or is critical of government. Without going into greater detail here, these commonalties support the idea that, both in the U.S. and in Germany, the same proto-right of freedom of speech is protected by constitutional regulations. These important commonalties go along, however, with a number of equally important discrepancies. Freedom of speech together with freedom of conscience and religion introduce the American Bill of Rights and are clearly brought to bear in the standing jurisdiction of the Supreme Court as a paramount constitutional value. In the ordering of the German Grundgesetz, however, freedom of speech is subordinate to human dignity as protected in Article 1. Indeed, the Federal Constitutional Court treats protection of freedom of opinion (like the protection of all other basic freedoms as well) as an aspect of the right to human dignity. But freedom of speech is thereby just one aspect among many, and it is often restricted on the basis of other human dignity considerations (especially of protection of personal honor and personality rights) in a way that would be unconstitutional in the U.S.17 In 1987, the Federal Constitutional Court ruled, for example, that it was an undue violation of Franz Josef Strauß’ (a German-Bavarian politician) personality rights to portray him as a pig that copulates with other pigs wearing judicial robes. Around the same time, however, the Supreme Court, (in 1988) in Hustler Magazine vs. Falwell (485 U.S. 46, 1988), ruled that a caricature of the reverend Jerry Falwell (in those days a political leader of the so-called “moral majority”), in which he, being drunk, has sex with his mother in a garden shed, is an exercise of the freedom of speech protected by the Bill of Rights.18 In the U.S. – contrary to Germany – substantially defamatory statements regarding political opinions are protected even if they are false on factual counts, as long as they are not made in bad faith. Thus – as the Supreme Court states in New York Times vs. Sullivan19 – the critical faculty of the press which is essential for a democratic public shall be ensured even in marginal cases.20 Whereas something like the defamation of groups is unknown in U.S. constitutional law, freedom of opinion has been repeatedly restricted by the jurisdiction of the German Constitutional Court in order to protect the personality rights of
16
For Germany see the discussion of these two aspects of freedom of speech in Glaeser (1972), esp. p. 83f. and Grimm (1995), p. 1698.
17
Cf. Nolte (1990) and Grimm (1995), p. 1703.
18
See the comparison of the Strauß decision of the German constitutional court with Hustler Magazin vs. Falwell. in Nolte (1988) and in Eberle (2002), p. 207f.
19
376 U.S. 254, 1964.
20
Cf. Tribe (1988), p. 862ff.
50
W. Hinsch
members of minorities against libel and slander that do not aim directly at individuals and their rights but at groups. The threat of punishment – which is compliant with the German constitution – for the corpus delicti “incitement” (Volksverhetzung) in 130, German penal code, belongs also to this context. A further important difference concerns the expression and dissemination of radical political beliefs. The German Grundgesetz embodies the concept of a “militant” or “well-fortified” democracy (streitbare Demokratie). After the experience of the breakdown of the Weimar Republic and the atrocities of the national socialists, individual liberties shall, in case of emergency, be subordinated to the exigencies of defending a free and democratic basic order. In the American/U.S. system, even seditious libel and propaganda of revolutionary and subversive doctrines fall under the protection of freedom of speech. A statutory or judicial ban of national socialist propaganda – as it has been practiced in Germany since the end of World War II – could not be sustained before the U.S. Supreme Court.21 What these comparisons rhapsodically show is that the constitutional right to freedom of speech protected by positive law in the U.S. – notwithstanding the mentioned commonalties – can obviously not be the same constitutional right as the one which – under the same title – is protected by the German Constitutional Court. For the set of actions, or rather of action types, that are protected in the two law systems, respectively, are not coextensive – even though they are overlapping in central areas. The comparisons also show to what degree the constitutional design of proto-right notions of freedom of speech in the two countries depend on specific national experiences and circumstances. What this suggests is that the equal right to freedom of speech that, as part of Rawls’ first principle, is a requirement of political justice cannot be identical with the rights of freedom of speech familiar from liberal democratic constitutions. All we can say at the moment (at least if all the involved constitutions are just) is that citizens who enjoy the U.S. or the German constitutional right of freedom of speech also enjoy the right that is part of the content of Rawls first principle of justice. If we conceive of constitutional rights as bundles of more specific rights, we may say that the U.S. and the German “freedom of speech bundle” somehow overlap and that all specific freedom of speech rights that are necessary prerequisites of political justice lie in the overlapping section.
6 Conclusions Political justice and democratic legitimacy are two equally necessary and mutually dependent components of a well-ordered society that is effectively regulated by publicly justified principles and norms. Given the fact of reasonable pluralism, the proto-rights of political justice, which are identified by means of moral argument, are an insufficient basis for the public justification of constitutional rights and other specific social norms. By themselves, they do not provide a basis for endowing 21
Cf. Eberle (2002), p. 202.
Legitimacy and Justice
51
constitutional rights and other norms with legitimate normative authority. The process of the public justification of norms has to be a two level procedure that incorporates argumentative and non-argumentative ways of collective decision making. There can be no legitimate constitution with normative authority that does not contain the proto-rights of publicly justified principles of political justice. And there cannot be a legitimate constitution that has not been subject to an adequately framed democratic process which involves voting procedures as a fair and rational way of resolving reasonable disagreements in matters of political justice and expediency. It is outside the scope of this article to discuss in any detail the extent to which the U.S. and German constitution actually meet the demands of political justice with regard to their freedom of speech clauses. It seems clear to me, however, that, in principle, as far as freedom of speech is concerned, both constitutions can be considered just constitutions, notwithstanding the fact that they do not contain the same constitutional rights of freedom of speech. Given the distinction between proto-rights and constitutional rights, both constitutions are just constitutions if, and only if, they effectively protect the same proto-rights that partly define political justice whose content is not subject to the procedures of political decision making. Constitutional rights, on the other hand – as the freedom of speech rights we find in the U.S. and German constitutions – may be seen as institutionalized proto-rights in need of democratic legitimization. Their constitutional specification is highly dependent on the historical, cultural, and institutional context of a particular political society and hence goes beyond what could be justified solely in terms of arguments that do not allow for reasonable disagreement. To respond to the misgivings Habermas expressed about Rawls’ understanding of constitutional democracy, we are now in a position to say that constitutional rights – along with all other legal rights – obtain normative authority only by means of a process of political legitimization, but only on the assumption that this process meets the demands of the proto-rights which are part of the content of political justice, which, in turn, is not subject to the outcomes of political procedures.
References Beetham, David (1991): The Legitimation of Power, London. Brugger, Winfried (2001): Einf¨uhrung in das o¨ ffentliche Recht der USA, M¨unchen, 2nd edn. Dahl, Robert A. (1956): A Preface to Democratic Theory, Chicago/London. Eberle, Edward J. (2002): Dignity and Liberty. Constitutional Visions in Germany and the United States, Westport/London. Flathman, Richard E. (1993): “Legitimacy”, in: R. E. Goodin, P. Pettit (eds.): A Companion to Contemporary Political Philosophy, pp. 527–533. Glaeser, Walter Schmitt (1972): “Die Meinungsfreiheit in der Rechtssprechung des Bundesverfassungsgerichts”, in: Archiv des o¨ ffentlichen Rechts, 97: pp. 60–123 (1. part) und pp. 276–298 (2. part). Grimm, Dieter (1995): “Die Meinungsfreiheit in der Rechtssprechung des Bundesverfassungsgerichts”, in: Neue Juristische Wochenschrift, 27: pp. 1697–1776. Habermas, J¨urgen (1988): “Law and Morality”, in: The Tanner Lectures on Human Values VIII, Salt Lake City: pp. 217–279.
52
W. Hinsch
Habermas, J¨urgen (1995): “Reconciliation through the Public Use of Reason. Remarks on John Rawls’s Political Liberalism”, in: The Journal of Philosophy, XCII, pp. 109–131. Hart, Herbert L. A. (1994): The Concept of Law, Oxford, 2nd edn. Hinsch, Wilfried (2002): Gerechtfertigte Ungleichheiten. Grunds¨atze sozialer Gerechtigkeit, Berlin/New York. Kommers, Donald P. (1989): The Constitutional Jurisprudence of the Federal Republic of Germany, Durham/London. Luhmann, Niklas (1969): Legitimation durch Verfahren, Neuwied. Nolte, Georg (1988): “Falwell vs. Strauß: Die rechtlichen Grenzen politischer Satire in den USA und der Bundesrepublik”, in: Europ¨aische Grundrechtszeitschrift, pp. 253–259. Nolte, Georg (1990): Beleidigungsschutz in der freiheitlichen Demokratie, Berlin/Heidelberg/ New York. Rawls, John (1993): Political Liberalism, New York. Rawls, John (1999): A Theory of Justice, Cambridge (MA), 2nd edn. Raz, Joseph (1979): The Authority of the Law. Essays on Law and Morality, Oxford. Raz, Joseph (1986): The Morality of Freedom, Oxford. Spencer, Martin E. (1970): “Weber on Legitimate Norms and Authority” in: British Journal of Sociology, 21, pp. 123–134. Tribe, Laurence H. (1988): American Constitutional Law, Mineola (NY), 2nd edn. Weber, Max (1922): Wirtschaft und Gesellschaft. Grundriss der verstehenden Soziologie, T¨ubingen.
Political Legitimacy and Its Need for Public Justification A Commentary on Hinsch ¨ Michael Kuhler
Taken for granted that political disagreements cannot always be solved by means of an argumentative consensus, a level of non-argumentative procedures that nonetheless are able to produce mutually acceptable norms seems to be an appealing option. Such a concept of legitimacy, in contrast to the underlying concept of justice, rests upon the idea that the mutual acceptability of its normative results can be assured because the non-argumentative procedures are themselves publicly justified and, hence, just. Furthermore, this option is accompanied by a number of individual (moral) rights that have to be recognized regardless of the results of any non-argumentative procedure. They may be called proto-rights. Can such a concept of legitimacy successfully be based upon the concept of justice, as Wilfried Hinsch suggests?1 What kind of connection has to be presupposed? And how plausible is it? In order to answer these questions I will begin with some remarks on the argumentative limits of public justification. Then I will mention some problems en route to legitimacy because of its need for public justification. Finally, the considerations so far will allow me to make some short and concluding remarks on the role and normative authority of proto-rights.
1 On the Argumentative Limits of Public Justification The idea of public justification can be considered as the heart of every Rawlsian contractualism, like e.g. Hinsch’s,2 and, hence, of the notion of justice. It rests upon the idea that a norm can be justified to every member of society by pointing out reasons which every person can accept from her own well-considered point of view. So far, this might look like each person considers a norm to be justified if it serves her personal interests. Where then lies the difference to the contractarianist point of view of rational egoism? The crucial difference lies in the notion of generality, put forward in the claim of justifying a norm to every person who shall obey it. Thus,
1
See Hinsch: this volume, p. 43f.
2
Cf. ibid., p. 42f.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
53
54
M. K¨uhler
a political norm is publicly justified only if every relevant individual point of view has been equally taken into account.3 However, such a notion of generality leaves room for two different interpretations. The first one sees the generality involved only as a matter of parallel, individual considerations.4 The second, and contractualist, interpretation conceives the idea of generality also in terms of fairness and puts it into the individual considerations right from the start. It implies that all citizens respect one another and their beliefs and interests as prima facie equal. Hence, the persons are to be understood not simply as rational egoists, as in contractarianist theories, but as “moral persons”.5 That means that every person considers the reasons for a norm not only in the light of her own interests but in the light of the interests of all other persons as well. Therefore, a norm cannot be justified to a moral person if it could be reasonably rejected by another person.6 Furthermore, moral persons are willing to act according to publicly justified norms because of their public justification, even if so acting would contradict their actual interests. A society consisting of moral persons can then be described in the Rawlsian term of a “well-ordered society”.7 Given this conception of public justification, the question arises within which limits can it contribute to the well-ordered society. This question may be discussed starting in terms of the compliance problem. A well-ordered society based on this conception of public justification seems to have the advantageous feature of solving it because the reasons to justify a norm publicly are at the same time the personal reasons of each moral person to recognize that norm. And as this is generally known, no one has to fear uncooperative behaviour by others, so everybody will comply with a publicly justified norm. However, one has to admit that this advantageous feature is not the benefit of public justification alone. It rather seems to be the consequence of presupposing moral persons. They are simply defined in a way that they can only accept and act on personal reasons that are identical with the publicly justifying reasons. Thus, the compliance problem is ruled out from the beginning and cannot be solved by public justification on its own. However, if not all persons are in fact moral persons and, consequently, publicly justifying reasons do not necessarily motivate persons
3 The importance of the term “equally” connects justice with equality, at least at some basic level of respect. Ernst Tugendhat calls this the proposition of symmetry (“Symmetriesatz”). Cf. Tugendhat (1997), p. 70. “Contractualism”, then, represents the Kantian tradition, while “contractarianism” stands for the Hobbesian approach. 4 This view is even open to a contractarianist account of mutual justification between rational egoists. However, the success of such an account is either dependent upon sheer luck – maybe the parallel, individual considerations all lead to the same norm or maybe they don’t – or it is based on power – which in turn raises the question if power can serve as a justifying reason. 5
Cf. Hinsch (2002), p. 130, and Rawls (1993), p. 81ff.
6
For the distinction between the “reasonable” and the “rational”, see Rawls (1993), p. 48ff. For the idea of “reasonable rejection”, see Scanlon (1998), p. 191ff. 7
See Rawls (1993), p. 35ff., also cf. Hinsch: this volume, p. 42.
Political Legitimacy and Its Need for Public Justification
55
to act upon them, does that mean the conception of public justification described by Hinsch8 is bound to collapse as a whole? To answer that question, it might be helpful to distinguish more strongly between motivation and justification. The compliance problem, in my view, is primarily a problem of motivation and has to be seen independently. If so, the strength of public justification still lies in its potential to produce mutually justifiable norms.9 That means no one could reasonably reject a publicly justified norm on the grounds of just his personal reasons. However, the question if that also means that he would be motivated sufficiently has to be left open. Leaving that question open now creates another problem. The presupposition of moral persons then seems to lose its ground and would turn into a rather implausible free-standing concept. Therefore, one should no longer presuppose moral persons. The idea of public justification rather obligates every person to adjust her interests in a way that contains the notion of contractualist generality and fairness. Hence, the concept of moral persons has itself to be understood as a normative concept. Citizens in a well-ordered society ought to see themselves and all others as moral persons and act accordingly. However, contractualists still have to present personal reasons why one should see himself as a moral person and not, for example, as a rational egoist.10
2 Some Problems en route to Legitimate Norms Given the fact of “reasonable pluralism”,11 no longer can it be expected that all political disagreements can be settled by means of public justification. Therefore, other, non-argumentative, ways of finally reaching mutually acceptable norms have
8
Cf. Hinsch: this volume, p. 42f.
9
Therefore, I think it would be a wise move to concentrate on this strength of the conception of public justification, for it also points out a weak spot of contractarianism. How can contractarianism hope to establish mutually justified norms, if not by sheer luck or force of power? The only possible way, it seems, leads exactly to a Rawlsian conception of public justification, save the aspect of motivation. The main reason for this is that even a contractarian norm that claims to be justified to a number of rational egoists has to take all these egoistic points of view into account. As neither sheer luck nor power are satisfactory starting points when aiming at a mutually justified norm, the only possible solution left seems to be a prima facie equal respect of all persons and their beliefs and interests. This would exactly be the contractualist notion of generality and fairness and thus be the contractualist conception of public justification. See K¨uhler (2006), ch. 3. 10
Such personal reasons may be identical with publicly justifying reasons. However, given that this cannot be generally assumed, public justification in itself, therefore, doesn’t seem to create sufficient personal reasons for all members of society to see themselves as moral persons and not as rational egoists. So why not be a rational egoist and play one’s own cards of power and cheating? I think there are indeed good personal reasons for being a moral person, especially when asking that question in the context of how to live a good life. Of course, that discussion would lead too far away from the focus of this paper, so I have to leave it open here. See ibid., ch. 4.
11
See Rawls (1993), p. 36.
56
M. K¨uhler
to be established. With the distinguished concept of legitimacy, Hinsch has proposed a likely solution for the problem of reasonable pluralism. The main idea is that the members of a society should at least reach an “overlapping consensus”12 concerning the features of fair non-argumentative procedures. Such procedures, e.g. majority decisions, can then give a society the means to establish norms whose content has failed to be mutually justifiable but that are nonetheless generally binding. To ensure that the non-argumentative procedures are fair, they in turn are the subject of public justification. One might say that legitimate norms are mutually acceptable because they are indirectly publicly justified. However, by admitting reasonable pluralism concerning the content of norms, the question arises how fair non-argumentative procedures can be established if their features are to be publicly justified as well. How can it be ensured that there is no reasonable disagreement about them, too? For example, why should minorities agree to a procedure like majority decision if it is foreseeable for them that it will produce only a legitimate norm whose content doesn’t reflect their well-considered interests? One has to bear in mind here that these interests are indeed reasonable interests which by definition cannot be rejected within public justification. Furthermore, there are different, and even incompatible, conceptions of legitimacy imaginable because the notion of objectivity involved has to be understood in a weak sense.13 So why be willing to agree to non-argumentative procedures knowing that this could mean giving up one’s own reasonable interests? Or, to put this question in theoretical terms: How may the level of legitimacy be of help if it also hinges on public justification and inherits its problem of reasonable pluralism? The answer to this question is still that establishing a level of legitimacy lies in the well-considered interests of all persons concerned. Even granted this answer, one has to ask again if the individual considerations involved are reasonable or rather rational ones, especially when considering the question left open concerning the motivating personal reasons. If, on the one hand, the individual considerations are just rational ones, then some kind of motivational contractarianism seems to be the root of contractualism. On the other hand, if the individual considerations shall be reasonable ones, then further personal reasons to see oneself as a moral person have to be presented. The underlying question is the following: Why not be a rational egoist within the limits of reasonable pluralism, i.e. why not insist on one’s own reasonable interests and combine them with one’s power and ability to cheat in order to enforce establishment of the favoured, then legitimate, norms? The contractualist answer, I think,14 would mainly lead back to the conception of public justification. A limited rational egoism would contradict the normative claim of respecting all others as moral persons. Whereas such mutual respect would pass the test of public justification, rational egoism would fail because of its aspects of power and cheating. Hence, there seems to be no other morally and politically 12
See ibid., p. 133ff.
13
Cf. Hinsch: this volume, p. 41.
14
See also Hinsch (2002), p. 161ff.
Political Legitimacy and Its Need for Public Justification
57
acceptable option of establishing legitimate norms than by reaching an overlapping argumentative consensus concerning the features of fair non-argumentative procedures. This means, on the one hand, that maybe some persons, on the grounds of reasonable pluralism, cannot be motivated to agree to any non-argumentative procedure that will likely produce legitimate norms which are not in their interests. On the other hand, it can be questioned whether such an egoistically biased refusal to find common argumentative ground could be publicly justified to all other persons.
3 Some Short Notes on the Role and Normative Authority of Proto-Rights The considerations so far, stressing the need of legitimacy to rest upon public justification, allow for some short and concluding remarks on the role and normative authority of the proto-rights described by Hinsch. Proto-rights, in opposition to constitutional rights, are to be conceived as moral rights, and people have them independently of their actual institutional recognition. They are solely the result of substantive philosophical argument. Therefore, they also have normative authority over constitutional rights, which is simply the consequence of a contractualism based on public justification. To ease possible concerns about the normative authority of proto-rights and their tension with the principle of public sovereignty, Hinsch states that proto-rights are more broadly and vaguely defined than constitutional rights.15 The latter are able to specify the former in various different ways. However, if proto-rights are the result of the conception of public justification and if legitimate constitutional rights are in turn specifications of proto-rights, then the question remains which influence the conception of public justification has on the details of the specific content of constitutional rights. Concerning freedom of speech, one might ask how much influence proto-rights would have on the detailed scope and content of this protected freedom. May, for example, all kinds of speech that directly attack the basic institutions of a wellordered society, like Nazi propaganda, legitimately be forbidden? Do proto-rights include the possibility to rightfully claim that certain detailed contents should be institutionalized as constitutional rights? If so, one might object that this seems to leave no more room for cultural differences concerning the details of the constitutional rights. Philosophers would then indeed seem to have regained their assumedly relinquished throne, at the cost of public sovereignty. On the other hand, one might ask if the normative authority of proto-rights does not simply ensure that the content of any norm can always be questioned as to whether it is mutually justifiable or not, i.e. whether it would pass the test of public justification, either directly or indirectly, via the conception of legitimacy.
15
Cf. Hinsch: this volume, p. 47.
58
M. K¨uhler
References Hinsch, Wilfried (2002): Gerechtfertigte Ungleichheiten. Grunds¨atze sozialer Gerechtigkeit, Berlin: Walter de Gruyter. K¨uhler, Michael (2006): Moral und Ethik – Rechtfertigung und Motivation, Paderborn: Mentis. Rawls, John (1993): Political Liberalism, New York: Columbia University Press. Scanlon, Thomas M. (1998): What We Owe to Each Other, Cambridge (MA): Harvard University Press. Tugendhat, Ernst (1997): Dialog in Leticia, Frankfurt am Main: Suhrkamp.
Consent, Obligation, and Legitimacy Frank Dietrich
Contract theory in its various classic and modern forms starts with two basic moral assumptions. The state of nature is described as a state of freedom and as a state of equality.1 Originally, the individuals are not subordinated to any political power; restrictions of their freedom have to meet with their approval. They are equal in so far as nobody is authorized in virtue of his religious insight, virtuous character, or intellectual capabilities to rule over others. According to these assumptions the state is only entitled to wield political power, if each individual consents to its sovereignty. The contract that is signed in the state of nature specifies the conditions under which the agents of the state can legitimately use coercive force. However, the act of consent not only warrants the state’s use of power but also creates political obligations for its subjects. By entering into the contract, the individuals voluntarily undertake obligations that were absent from the state of nature. In this paper I shall be concerned exclusively with the explanation of political obligations that contract theory offers. Political obligations are commonly understood as moral requirements to support and comply with the institutions of one’s country of residence.2 They are special obligations in the sense that they bind the individual to a particular political entity. For instance, it is generally assumed that military service or payment of taxes can only be required by the state one lives in. At first glance a consent-based justification of political obligations appears to be attractive. The requirement of consent protects the individuals from becoming bound to a state against their will. Since the obligations are self-imposed, they can be expected to be in the individuals’ best interests. But the contract-based justification of political obligations faces, as is well-known, considerable problems. The idea of a contract that is concluded in the state of nature is purely fictitious; there is no 1 The moral foundations of the argument that contract theory puts forward are clearly expressed in John Locke’s “Two Treatises of Government”: “To understand political power right, and derive it from its original, we must consider what state all men are naturally in, and that is, a state of perfect freedom to order their actions (. . .). A state also of equality, wherein all the power and jurisdiction is reciprocal, no one having more than another: there being nothing more evident, than that creatures of the same species and rank (. . .) should also be equal one amongst another without subordination or subjection (. . .)”. Locke (1993), p. 116. 2
Simmons (1979), p. 29.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
59
60
F. Dietrich
historical evidence that any state came into being through a voluntary agreement of its subjects. Even if some states were founded in the dim and distant past by contract, nothing would follow for their contemporary citizens. Since, according to contract theory, consent is required of each individual personally, the decisions of some remote forefathers cannot have any binding force. But the citizens of contemporary states have not signed a contract, nor have they expressed in any other explicit form their willingness to comply with the existing legal and political institutions. In what follows I will examine two strategies to tackle the aforementioned problems. Firstly, I will consider the possibility of replacing explicit consent by a different form of consent. Scholars who work in the framework of contract theory usually resort either to tacit or to hypothetical consent. Secondly, I will discuss the attempt to establish a concept of political legitimacy that dispenses with the principle of consent altogether. In this context I will focus primarily on the theoretical account, which Allen Buchanan has developed in his recent work.
1 Consent-Based Obligations The contract-based justification of political obligations that I sketched out in my introductory remarks encounters serious difficulties. Scholars who wish to stick to the central idea of contract theory base their arguments either on tacit or on hypothetical consent. I shall consider both alternatives in turn.
1.1 Tacit Consent The term tacit consent denotes an indirect way to express one’s agreement. The consent is not given explicitly, but has to be inferred from the person’s behavior. The most famous instance of a contract theory that relies on tacit consent is to be found in John Locke’s political philosophy. In Locke’s opinion mere residence on the state’s territory indicates that the individuals approve of the political authority.3 Provided that the individuals are not forcefully prevented from leaving the country, they have a choice between different political and possibly even anarchist communities. If they voluntarily decide to remain in the state, they thereby tacitly consent to the obligations that are connected with residence. Locke’s argument is thought to be questionable inter alia for the following two reasons. Firstly, it is commonly assumed that acts of consent have to be performed knowingly. The individuals must be aware that their behavior is interpreted as a 3 In his “Two Treatises of Government” Locke states: “(. . .) Every man, that hath any possession, or enjoyment, of any part of the dominions of any government, does thereby give his tacit consent, and is as far forth obliged to obedience to the laws of that government, during such enjoyment as anyone under it; whether this his possession be of land, to him and his heirs for ever, or a lodging only for a week; or whether it be barely traveling freely on the highway; and in effect, it reaches as far as the very being of anyone within the territories of that government”. Locke (1993), p. 176.
Consent, Obligation, and Legitimacy
61
sign of approval. An act or omission cannot count as tacit consent, if its meaning is not properly understood. Evidently, the individuals usually do not consider their residence as the result of a morally significant choice. They are not aware that they undertake political obligations, if they forego emigration. Consequently, living on the state’s territory cannot be seen as a conscious act of consent.4 Secondly, in order to give his consent to X, either explicitly or tacitly, the individual must have at least one feasible alternative. We do not correctly speak of an act of consent, if the individual is not able to refuse X and to choose Y. Since emigration usually implies very high costs, it seems to be for many people not a realistic option. In his essay “Of the Original Contract” David Hume made this point very clearly: “Can we seriously say, that a poor peasant or artisan has a free choice to leave his country, when he knows no foreign language or manners, and lives, from day to day, by the small wages which he acquires. We may as well assert that a man, by remaining in a vessel, freely consents to the dominion of the master; though he was carried on board while asleep, and must leap into the ocean, and perish, the moment he leaves her”.5 Other versions of the tacit consent argument appear to be no less problematic. For instance, it has been argued that participation in political procedures, such as elections, can be interpreted as an indication of consent.6 But if voting is voluntary, some citizens will stay away from the ballot. Consequently, the criterion of participation cannot ensure the consent of the whole population. Moreover, elections are concerned with the choice of governments; they are not about the formation or maintenance of states. The individuals who take part in an election do not necessarily approve of the existing political institutions. Even anarchists, who firmly oppose the state, can have strategic reasons to cast their votes. They may, for example, wish to support the party that seems most likely to strengthen individual liberties. However, one can think of conditions under which the objections to Locke’s account would be less persuasive. Firstly, the state could officially inform its inhabitants that residence is regarded as a sign of tacit consent. The citizens would then be aware that they undertake political obligations, if they decide to remain in the country. Secondly, and more precariously, the state could expand the options of its citizens. If they were allowed to emigrate and to secede, they would have the widest possible freedom of association and dissociation. Only the free choices of other persons could hinder them in achieving their political goals. Thus, the attempt to found an independent state will fail, if not enough persons are prepared to take
4
Simmons (1993), pp. 225–232.
5
Hume (1993), p. 283.
6
In his postscript to the second edition of “Consent, Freedom and Political Obligation” John Plamenatz writes: “The citizen who votes at an election (. . .) has voluntarily taken part in a process which confers authority on someone who otherwise would not have it. He may bitterly regret the election of the successful candidate and may not even have expected it, but if the election was free and he freely took part in it, he consented to the authority of the man elected”. Plamenatz (1968), pp. 170–171. Similar ideas are expressed in Walzer (1970), pp. 9–10 and Singer (1973), pp. 45–59.
62
F. Dietrich
the risk of secession. This line of argument needs, of course, further elaboration; whether it can ultimately succeed is far from certain.
1.2 Hypothetical Consent The aforementioned problems have led many proponents of contract theory to focus on hypothetical consent. Their argument has a counterfactual structure; political institutions are deemed legitimate, if the individuals would have opted for them in a fictitious choice situation. In what follows I will distinguish between two methods of generating hypothetical consent.7 Firstly, I will examine contractarian theories that – following the Hobbesian tradition – describe the choice situation in a rather realistic manner. Thus, James Buchanan takes the natural distribution of power that results from the bodily strength and intellectual capabilities of the individuals as a starting point for his contractarian argument.8 Afterwards, I will consider contractualist theories that – following the Kantian tradition – specify the choice situation in view of certain moral ideals. Most famously John Rawls imagines a decision behind a veil of ignorance that debars the individuals from all information they could use to their advantage.9 1.2.1 Morally Unconstrained Choices Theories of the first kind do not impose moral constraints on the choice of the contracting parties. They aim to demonstrate that the implementation of certain political institutions is in each individual’s interests. However, having an interest and having an obligation is quite a different matter. The interests that may possibly be revealed by the hypothetical consent argument give the individuals no reason to feel politically obligated. Let me illustrate this point with an example that I owe to Eric Rakowski.10 Jim wishes to offer his friend John a very favorable bet. He tosses a coin: If heads turns up, John will win $ 1.000; if tails is facing upward, John will loose $ 10. Unfortunately, the bet comes into his mind during the evening and Jim cannot reach John to ask him for his acceptance. But, since he knows his proposal to be very attractive he decides to rely on John’s hypothetical consent. He flips the coin and it lands tails up. The next day Jim rings at John’s door to tell him about the bet and to cash the money in. John knows his friend as a trustworthy person and has no doubts about his sincerity. He even admits that he would have agreed to the bet, if Jim had reached him yesterday. Now, do you think John is morally obliged to pay $10? Certainly, John may have good reasons to satisfy Jim’s demand. He may wish to honor the benevolent intentions of his friend or speculate on further favorable bets 7
Morris (1996), pp. 219–221; Stark (2000), pp. 314–316.
8
Buchanan (1975), pp. 23–25.
9
Rawls (1971), pp. 118–123.
10
Rakowski (2001), pp. 252–253.
Consent, Obligation, and Legitimacy
63
which Jim might offer. But I do not think John would act in a blameworthy manner, if he refused to pay the required amount. The example gives an important insight into the relationship between interests and obligations. Interests can provide strong motives for undertaking moral obligations, but taken as such they do not impose moral obligations on the individual. The analogy between the aforementioned example and the justification of political obligations is not hard to draw. Let’s assume that according to contract theory the implementation of a kind of majority rule is in each individual’s interests. The minority in any particular voting would then be in a similar position as John; being outvoted in a majority decision is very much like loosing a favorable bet. As in John’s case there may be weighty reasons for the minority to accept the outcome and to fulfill the resulting requirements. For instance, they may not wish to frustrate the expectations of their fellow-citizens or to destabilize a decision procedure that by and large is working to their advantage. But if I was right in claiming that John is under no moral obligation, the same seems to hold for the outvoted individuals. Their hypothetical interest to be governed by a kind of majority rule does not put them under an obligation to comply with its results. However, the case is not as clear-cut as it first appears. Under certain conditions it seems to be warranted to base moral obligations on hypothetical interests. Think, for example, of a physician who gives an unconscious person first aid. Since there is no way to obtain the patient’s consent, the physician cannot help but refer to her supposed interests. Imagine further that the health care system requires all patients to reimburse physicians for their treatment up to a certain amount. If the patient’s interests have been carefully investigated and no medical malpractice is involved, she seems to be under an obligation to pay her share. Apparently, a hypothetical consent can in exceptional cases give rise to a moral obligation. I shall now try to clarify, in which respects the situations of John and the patient differ. Some reflections on the meaning of autonomy may help to understand, why we deem the latter, but not the former, to be morally obliged. In my view autonomy is appreciated for two reasons. Firstly, we regard the ability to make our own choices as instrumentally important. It allows us to act in accordance with our interests and to pursue our respective aims. Secondly, we usually attach intrinsic value to autonomy.11 We want to take our own decisions and to translate them into actions. Even if our interests are understood correctly, we do not want other people to act on our behalf. Unless we have somebody explicitly authorized as our representative, we tend to feel disparaged, if decisions are made for us. Hypothetical consent arguments of the first kind retain the instrumental value of autonomy. They justify only acts, which the individuals themselves would have chosen. The prospect of achieving a goal is not lessened because someone else has made the decision. Of course, the construction of a hypothetical consent carries the 11
Though it is put forward in a slightly different context, Robert Nozick’s thought experiment of the experience machine illustrates the intrinsic meaning of autonomy very well. Perhaps the most important reason why we would abstain from using the experience machine is that we wish to decide and to do certain things ourselves. Nozick (1974), pp. 42–45.
64
F. Dietrich
risk of misinterpreting the individuals’ interests. However, in the two examples that I have cited the interests are properly described. John admits that he would have accepted the bet; the patient acknowledges that she would have approved of the medical treatment. There was no misinterpretation involved that could explain why we judge their cases differently. But hypothetical consent arguments fail to take the intrinsic value of autonomy into account. They are negligent of the importance that is attached to the activity of decision-making. However, in extraordinary cases only the instrumental aspects of autonomy matter. If the individuals are unable to make deliberate choices, deciding in their best interests is the only way to respect their autonomy. In this regard the two examples differ considerably. Jim could have waited until the next day to ask John whether he agrees with the bet. In making the decision for John, he showed a lack of concern for his autonomy. The physician on the other hand had to decide immediately which medical measures to take. Since the patient was not responsive, referring to her presumed interest was the best thing to do. In view of the preceding discussion my former judgment on hypothetical consent arguments has to be qualified. They can impose moral obligations on the individuals, if the interests are carefully investigated, and if an actual consent is not accessible. Does this conclusion shed new light on the justification of political obligations? Of course, citizens are not like unconscious patients; they can in principle be asked whether they approve of certain norms or rules. But one has to concede that it is much more difficult to establish the opinion of a whole population than to inquire about the attitude of a single person. The citizenry usually consists of millions of people and its composition is constantly changing. One may possibly argue that not every effort has to be made to achieve an actual consent. The criterion of accessibility may be interpreted in a wide sense; it may be understood to exclude particular burdensome and expensive procedures.12 This line of reasoning will, certainly, not go uncontested; whether it could survive a thoroughgoing examination is open to question. 1.2.2 Morally Constrained Choices Let me now turn to the second variant of hypothetical consent arguments. Contract theories of the Rawlsian type describe a highly idealized choice situation. The individuals are deprived of all knowledge that would allow them to act in a selfinterested manner. Their decisions for a set of political institutions or principles are largely determined by the moral features of the choice situation. In contrast to the contract theories that I have discussed before, the focus is not on the individuals’ interests. Rather the hypothetical consent argument intends to elucidate what justice or fairness demands in the realm of politics. 12 This line of argument is pursued by Paul Menzel in his reflections on medical rationing: “(. . .) A crucial condition of presuming consent is that obtaining actual consent be either impossible or in some sense prohibitively costly. In the rough sense, “prohibitively costly” simply means what the actual individual, with his or her actual values, would agree is too high a cost for upgrading from presumed to actual consent”. Menzel (1990), p. 31.
Consent, Obligation, and Legitimacy
65
Again the question arises whether the institutions or principles that would have been chosen under counterfactual conditions may entail political obligations. First of all it is important to be aware of the role that consent plays in the theories under consideration. Against the core idea of contract theory it is not regarded as an act with which the individuals voluntarily undertake obligations. The hypothetical consent that is given behind a veil of ignorance purely serves an illustrative purpose; it shall bring to light what the requirements of justice or fairness are. Now I think it should be conceded that apart from self-imposed obligations we may have so called natural duties. For instance, we commonly assume that everybody is under a moral duty not to harm other people without reason. The validity of this principle does not depend on prior consent; the offender cannot excuse his behavior by stating that he has never promised to obey it. Thus, the second variant of hypothetical consent arguments does not give rise to self-imposed obligations, but it may possibly reveal natural duties. However, the loyalty that states demand of their citizens cannot be regarded as natural duty. Natural duties are general in kind, whereas political obligations concern a special relationship between the individuals and their state. The contractualist approach may perhaps illuminate what the features of a just polity are. On this basis one may possibly argue for a moral duty to support political institutions that satisfy the criteria of justice. However, the political obligations that we are concerned with are commonly thought to bind the individuals to a particular state. We feel, if at all, obliged to serve in the army of our country; we do not consider ourselves to be under a duty to join the armed forces of any other just state. Therefore, hypothetical consent arguments of the second type fail – if they aim at justifying political obligations. One might object that my account of Rawls’ theory leaves an important detail out. According to Rawls we are not bound to a multitude of just states; we only have a natural duty to comply with just institutions which apply to us.13 The addition of “which apply to us” cannot solve the aforementioned problem. Instead of providing an argument for the particular relationship between the individuals and their country, he relies on convention. Moreover, Rawls seems to assume that it is always clear which laws or institutions apply to the individuals. This view, however, is as I shall demonstrate in the next section mistaken.
2 Legitimacy Without Consent In his recent book “Justice, Legitimacy, and Self-Determination” Allen Buchanan has proposed an alternative to contract theory. At the core of his account is a terminological distinction between political legitimacy and political authority.14
13 14
Rawls (1971), pp. 293–301; Waldron (1993).
Buchanan’s use of words has changed in time. In a previously published article he has distinguished between justifiable enforcement legitimacy and legitimacy as the right to be obeyed. Buchanan (1999), pp. 56–59.
66
F. Dietrich
According to Buchanan “(. . .) an entity has political legitimacy if and only if it is morally justified in wielding political power, where to wield political power is to (make a credible) attempt to exercise supremacy, within a jurisdiction, in the making, application, and enforcement of laws”.15 In his view states are morally justified in using coercive force, if they fulfill two criteria of justice.16 Firstly, they have to respect a number of basic human rights, such as the right to bodily integrity or the right to freedom from religious persecution.17 Secondly, they have to meet fundamental standards of democratic government, for example they have to provide for representative, legislative institutions.18 Buchanan’s notion of political authority is more demanding. According to his terminology “(. . .) an entity has political authority if and only if, in addition to (1) possessing political legitimacy, it (2) has the right to be obeyed by those who are within the scope of its rules; in other words, if those upon whom it attempts to impose rules have an obligation to that entity to obey it”.19 Buchanan takes the view that political philosophy should be primarily concerned with the legitimacy of states. Whether political entities possess authority is deemed irrelevant for two reasons. Firstly, states can be morally justified in wielding political power, even if their citizens have not consented to their authority. They merely have to satisfy the two criteria of justice referred to above. Secondly, since the individuals usually have sufficient prudential, religious or general moral reasons for obeying the law, the stability of legitimate states does not depend on the existence of political obligations. According to Buchanan the various forms of contract theory, therefore, are mistaken in focusing on political obligations. They should direct their attention not to the authority, but to the legitimacy, of states. Buchanan’s preoccupation with the notion of political legitimacy has an obvious advantage; it avoids all problems that are connected with the principle of consent. In assessing Buchanan’s proposal I shall first take a closer look at his concept of political legitimacy. According to Buchanan the legitimacy of states solely depends upon the fulfillment of the aforementioned criteria of justice. States are morally justified in making, applying and enforcing laws, if they meet basic standards of human rights protection and democratic government. Political obligations do not play any role whatsoever. Neither does the legitimacy of a state entail political obligations for its citizens, nor is the existence of political obligations a precondition for legitimate rule. This total disregard for political obligations, however, leads to a number of serious problems. Consider first the significance of Buchanan’s distinction between legitimate states and illegitimate states. The former are entitled to rule over their
15
Buchanan (2004), p. 235.
16
Ibid., pp. 247–249.
17
Ibid., p. 129.
18
Ibid., pp. 146–147. Buchanan leaves the question open, whether democratic government should be regarded as a basic human right. If democracy is included in the list of protected rights, then there remains, of course, only one criterion of justice. Ibid., pp. 142–144. and pp. 278–280.
19
Ibid., p. 237.
Consent, Obligation, and Legitimacy
67
subjects, whereas the latter are not justified in exercising political power. Therefore, one would expect that the question of legitimacy does matter to the individuals. Whether the state’s use of political power is morally justified, or not, should make a difference to their moral commitments. In Buchanan’s account, the classification of a state as legitimate or illegitimate respectively has no consequences at all. Citizens of legitimate states do not have necessarily more compelling reasons to obey the law than citizens of illegitimate states.20 Furthermore, the notion of political legitimacy, as it is used by Buchanan, appears to imply political obligations. Legitimate states are morally justified in wielding political power; they have in other words a moral right to make, apply, and enforce laws. Following a widely accepted definition, the term right is intimately connected with the term duty or obligation.21 For instance, a person can only be said to have a right to perform action X, if everybody else is under a duty not to interfere with X. Likewise, the assertion of a right to good Y implies that somebody must have a duty to supply the person with Y. Consequently, a state can only be said to possess a right to impose rules, if its citizens have an obligation to comply with them. An entitlement to make, apply and enforce laws appears to be meaningless, if there are no corresponding obligations to obey the law. Finally, Buchanan himself assumes that the classification of a state as legitimate gives rise to two kinds of obligations. First the notion of political legitimacy plays an important role in his normative theory of state recognition. Buchanan stipulates three conditions for state recognition: the satisfaction of a non-usurpation requirement, the fulfillment of minimal standards of international justice, and the possession of political legitimacy.22 If a political entity meets these conditions, it has to be credited with the full rights of a sovereign state. Thus, the ascription of political legitimacy (along with the satisfaction of the non-usurpation and the international justice requirement) entails moral obligations for other states. Moreover, the concept of political legitimacy is an integral part of Buchanan’s theory of secession. According to Buchanan the right to secession has to be understood as a remedy of last resort against unjust states.23 For instance, separatists are morally justified in seeking political divorce, if their basic human rights are seriously violated. But they do not have a right to secede from a state that exercises its political power in a legitimate way. In other words, they are obliged to refrain from breaking away, if the state they belong to meets the aforementioned criteria of political legitimacy. If the concept of political legitimacy entails an obligation to remain loyal to the state, why does it not imply an obligation to comply with its laws? If the individuals are
20
Legitimate states, which satisfy the requirements of justice, give their citizens more general moral reasons to obey the law. Illegitimate states on the other hand impose in many cases harsher punishments on lawbreakers. Their citizens usually have stronger prudential reasons to conform with the law.
21
Raz (1986), p. 166.
22
Buchanan (2004), pp. 266–268.
23
Ibid., pp. 353–373.
68
F. Dietrich
obliged to obey a law against secession, why are they not obliged to obey the law in general? Buchanan outlines, as far as I can see, no argument for treating these questions differently. Let me now turn to Buchanan’s assertion that the existence of political obligations is irrelevant for lawful behavior. In his view, consent-based obligations are not needed because the individuals usually have sufficient prudential, religious, and general moral reasons for complying with the law. I shall briefly examine whether Buchanan’s account is satisfactory. The individuals have prudential reasons for obeying the law in so far as they wish to avoid punishment. Though this is certainly a strong motive for compliance, if the risk of detection is high, it has no effect in situations where one can expect to get away with a violation of law. Therefore, prudential reasons could only be sufficient, if the state had full control over its subjects. A state that keeps its citizens under constant surveillance is hardly conceivable (and certainly not desirable). The individuals always have opportunities for breaking the law without taking any serious risk of detection and punishment. Consequently, no state can rely exclusively on prudential reasons. According to Buchanan the individuals have religious reasons to obey the law, if the scriptures require them to do so. But, evidently, in modern, pluralist, and secularized societies this motive applies only to a part of the citizenry, possibly a rather small one. Not all religious communities expect their members to conform with the law; some are indifferent to political matters or even require their followers to disregard laws they consider to be sinful. And, more importantly, many individuals do not adhere to religious beliefs at all. Buchanan does not explain his use of the term general moral reasons, the third motive for conformity with the law he refers to. I take him to mean moral reasons that do not depend on the existence of a law. For instance, one could say that since it is always wrong to commit murder, one should refrain from killing even if the law does not prohibit it. As far as the law codifies norms that are morally binding Buchanan rightly holds political obligations to be dispensable. Since the individuals are already morally obliged to abide by these laws, there is no need for an additional obligation. But the law of modern states consists of a variety of different norms. Some of them, for example tax laws or traffic rules, do not bind the individuals in virtue of their moral quality. States, however, normally want their subjects to respect the whole corpus of law; therefore, they cannot be satisfied with general moral reasons. To sum up, each of the three reasons for lawful behavior Buchanan mentions is in one respect or the other deficient. It appears doubtful whether they can sufficiently motivate the individuals to comply with the law. Consequently, the idea of political obligation remains attractive. Political obligations concern all laws, not just moral ones, and apply to everyone irrespective of her religious convictions. If the citizens feel politically obliged to obey the law, it will certainly benefit the state. Up to now I have made two points. Firstly, I have argued that Buchanan’s concept of political legitimacy is unconvincing because it does not include individual obligations. Secondly, I have contended that political obligations are not irrelevant for the individuals’ motivation to conform with the law. If I have been right so far,
Consent, Obligation, and Legitimacy
69
the question arises whether Buchanan’s concept of legitimacy can – contrary to his intention – establish political obligations. (This is by no means evident from what I have said. To state that the concept of legitimacy has to take individual obligations into account does not imply that it provides a plausible justification for them.) Essentially, Buchanan’s theory runs into the same difficulties as the second variant of hypothetical consent arguments. As I have mentioned before, political obligations bind the individuals to a particular state. If it is asserted that the legitimacy of a state gives rise to political obligations, this must hold generally. The individuals would then be obliged to every state that meets the criteria of legitimacy. Therefore, Buchanan’s theory cannot provide an explanation for the particular relationship between the citizenry and its state.24 At first sight the aforementioned problem may seem to have no practical relevance. Normally, the individuals are only subordinated to the law of their own country; they do not encounter competing claims of two or more legitimate states. However, practical problems can easily be imagined. Consider once again the case of secession, which is at the center of Buchanan’s theory. Many democratic states are currently confronted with demands for independence. The separatists usually do not intend to change the political system. Their striving for independence is rather motivated by the wish to escape the cultural dominance of the majority population. Now imagine that a region, for example Quebec, Catalania, or the Faeroes, secedes without permission of the central government. In this case competing claims for sovereignty would emerge. Both political entities, the established state as well as the newly proclaimed one, would meet Buchanan’s criteria of legitimacy. Consequently, the question to which authority the individuals are politically obliged could not be answered in a non-arbitrary manner. It might be tempting to settle this dispute by inquiring the preferences of the people concerned. But this would mean to reintroduce the core idea of contract theory. If a referendum is considered the best solution, the principle of consent appears to be decisive.
3 Conclusion Let me briefly summarize my two main points. Firstly, the contract-based justification of political obligations faces in all its variants serious problems. Secondly, Buchanan’s concept of legitimacy that dispenses with political obligations provides no convincing alternative. This leaves us with a dilemma. We neither have a plausible justification for political obligations nor can we do without them. How can we proceed? In my view contract theory cannot be evaded. If there is a solution at all, it has to be found within the framework of contract theory. Since the second version of
24
Buchanan addresses the particularity problem solely in regard to the government. He convincingly argues that particular individuals are justified in forming a government, if they are democratically elected. But he fails to realize that the particularity problem concerns the state as well. Ibid., pp. 254–256.
70
F. Dietrich
hypothetical consent arguments is closely related to Buchanan’s theory, I can only see two options. We have to strengthen either tacit consent or hypothetical consent arguments of the first kind. However, I am by no means sure that the difficulties, which I have outlined before, can be surmounted.
References Buchanan, Allen (1999): “Recognitional Legitimacy and the State System” in: Philosophy & Public Affairs, 28, pp. 46–78. Buchanan, Allen (2004): Justice, Legitimacy, and Self-Determination. Moral Foundations for International Law, Oxford: Oxford University Press. Buchanan, James (1975): The Limits of Liberty. Between Anarchy and Leviathan, Chicago: University Press. Hume, David (1993): Selected Essays, edited by Copley, Stephen and Edgar, Andrew, London: Dent. Locke, John (1993): Two Treatises of Government, edited by Goldie, Mark, London: Dent. Menzel, Paul T. (1990): Strong Medicine. The Ethical Rationing of Health Care, Oxford: Oxford University Press. Morris, Christopher W. (1996): “A Contractarian Account of Moral Justification” in: SinnottArmstrong, Walter and Timmons, Mark (eds.): Moral Knowledge? New Readings in Moral Epistemology, Oxford: University Press. Nozick, Robert (1974): Anarchy, State, and Utopia, Oxford: Blackwell. Plamenatz, John P. (1968): Consent, Freedom and Political Obligation, Oxford: Oxford University Press, 2nd edn. Rakowski, Eric (2001): “Taking and Saving Lives” in: Harris, John (ed.): Bioethics, Oxford: Oxford University Press. Rawls, John (1971): A Theory of Justice, Cambridge: Cambridge University Press. Raz, Joseph (1986): The Morality of Freedom, Oxford: Oxford University Press. Simmons, A. John (1979): Moral Principles and Political Obligations, Princeton: University Press. Simmons, A. John (1993): On the Edge of Anarchy. Locke, Consent, and the Limits of Society, Princeton: Princeton University Press. Singer, Peter (1973): Democracy and Disobedience, Oxford: Oxford University Press. Stark, Cynthia A. (2000): “Hypothetical Consent and Justification” in: The Journal of Philosophy, 97, pp. 313–334. Waldron, Jeremy (1993): “Special Ties and Natural Duties” in: Philosophy & Public Affairs, 22, pp. 3–30. Walzer, Michael (1970): Obligations. Essays on Disobedience, War, and Citizenship, Harvard: Harvard University Press.
Part II
Which Role can Rationality Play for State Legitimacy?
On the Rationality and Stability of a Minimal Consensus Reinhard Zintl
Consensus is a central feature of normative constitutional thinking: If it is possible to interpret an institution as the reasonable result of voluntary agreement among those who have to live with this institution, then we have a prima facie legitimization of that institution. This is the normative core of the idea of a social contract, if it is used as an instrument of critical scrutiny of rules.1 The central questions, which are dealt with by contractarian arguments, are the question of why people might rationally prefer political order to anarchy, and second, the question of what order they might rationally choose. The answer to the first question is well known and it is the easier one: A state of nature without the existence of a peace keeping authority is not an attractive idea. Much harder to answer is the second question, which will be dealt with in this paper. The answers to the question of what order should be chosen differ widely, as the classic examples of Hobbes2 and Locke3 prove – for Hobbes, a sovereign power which is above the Law is the constitution which rational people will choose; for Locke, only a constitution guaranteeing the Rule of Law makes it rational to leave a state of anarchy. The root of the problem lies in the fact that there are plausible reasons why every rational person should prefer peace to the war of all against all (hence the possibility of a contractarian answer to the first question), whereas there is no argument why every rational person should prefer one specific order to any other conceivable order (hence the difficulties in giving a contractarian answer to the second question). Of course, we have always the option to answer the second question by implanting our own normative judgments into the contractarian argument. The typical way to do so is a description of the pre-contractual situation in a specific way – for
1 As done by modern contractarians like Buchanan (1975), Nozick (1974) or Rawls (1971); cf. also Gordon (1976); Koller (1987). In general, contractarian arguments have two faces: On the one side a contract implies voluntariness and consensus; on the other hand a contract implies duties. The use of the idea of a social contract as legitimizing duties has been criticised thoroughly by Hume (1964 [1748]). 2
Hobbes (1991 [1651]).
3
Locke (1988 [1689/90]).
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
73
74
R. Zintl
instance, by introducing a veil of ignorance, as it is done by Rawls. But in this case, the contractarian argument is not the base on which our judgments on rules are founded, but rather an instrument of their presentation.4 If we do not want to take this escape, we have to ask whether we can find any reason why the possible diversity of ideas on rules among rational persons might be overcome or at least reduced by voluntary agreement. Is there anything which we might identify as a rational compromise or a minimal consensus5 among rational actors? What, if any, rules or institutions might be generally acceptable or necessarily accepted in a hypothetical contract situation? I will first try to make explicit the concept of a minimal consensus (Part I). In the next steps I will ask, first, what can be said about the rationality of such a consensus, if it exists (Part II), and then, and on its stability (Part III). As far as the rationality of a possible consensus is concerned, the figure of a minimal consensus refers to a thought experiment ex ante: What, if any, design of institutions might find consent among rational and well informed constitution makers? As far as the stability of a possible consensus is concerned, the figure of a minimal consensus does refer to a real problem of integration. The question here is whether some institutional arrangement might find ongoing support of the actors living with it, be they well informed or not.
1 Minimal Consensus 1.1 Constellations The simplest way to get an idea of what is meant by a minimal consensus is to look at three different and extremely simplified descriptions of hypothetical constellations of constitutional choice. We consider constellations where only two distinct groups of people, group A and group B, with different ideas on the desirable institutions try to find out whether it is attractive and possible or not to reach some consensus involving all. 1.1.1 Constellation I In the first constellation, three constitutional proposals are considered: The first, CA , makes the members of group B the slaves of group A; the second, CB , makes the members of group A the slaves of group B; the third, CE , establishes equal rights among all. The default option is the status quo, SQ. The preference orders of the groups look like this: Group A: CA > CE > SQ > CB Group B: CB > CE > SQ > CA 4
For a discussion cf. Zintl (1983), p. 29 ff.
5
Cf. Albert (1976).
On the Rationality and Stability of a Minimal Consensus
75
If we assign ordinal numbers from 3 (best) to 0 (worst) to the outcomes, we get the following description of the choice situation (in cases of disagreement the Status Quo and its payoff are preserved): Table 1 A Definite Minimal Consensus B A
CA
CE
SQ
CB
CA CE SQ CB
3,0 1,1 1,1 1,1
1,1 2,2 1,1 1,1
1,1 1,1 1,1 1,1
1,1 1,1 1,1 0,3
We expect the two groups to reach agreement on CE . This constitution is not the desired constitution for either of them, but neither of the respectively favoured constitutional proposals CA and CB can be realized by its adherents separately. CE is the minimal consensus.6 1.1.2 Constellation II Next, consider this situation: The alternatives to the Status Quo are “Socialism”, CS , and “Capitalism”, CC . We now assume that both arrangements can be realized by their respective adherents separately.7 Suppose the following preference orders of two parties: Party A: CS > CC > SQ Party B: CC > CS > SQ Then the possible states of the world are given by the following matrix: Table 2 Separate Ways B A
CS
CC
SQ
CS CC SQ
2,1 1,1 0,1
2,2 1,2 0,2
2,0 1,0 0,0
Both parties rank both types of constitutions higher than the Status Quo, but since “Socialism” as well as “Capitalism” can be realized by its adherents in separate polities, no minimal consensus exists. Instead, party A chooses constitution CS , and party B chooses constitution CC . 6 7
As can be seen in the figure, it would be somewhat more precise to call it a maximin consensus.
They may, for instance, agree to disagree and to form distinct communities, settling at distinct places. This was not uncommon among the first settlers in North America.
76
R. Zintl
1.1.3 Constellation III Finally, in between these two cases lies a constellation where the groups have similar conflicting constitutional preference orders as in “Constellation II”, but where no separate clubs can be formed and where no clearly visible and unique intermediate solution as in “Constellation I” is possible. The constellation then has the basic properties of the Battle of the Sexes Game:8 Table 3 A Minimal Consensus Range B A
CA
CB
SQ
CA CB SQ
2,1 0,0 0,0
0,0 1,2 0,0
0,0 0,0 0,0
Here, for both groups the founding of a common polity is better than taking separate ways, but the terms of cooperation in that common polity are contested. The concept of a minimal consensus can still be used in this situation, but in a weaker sense than before: The region of efficient solutions does not contain one specific constitution but rather a group of constitutions which are preferred to the status quo, but which are contested among each other and where it is not ex ante clear whether and what compromises might be found.9 What is the empirical relevance of these stylized descriptions?
1.2 Problems of Constitutional Choice The real choice among constitutions is more complex than these simplified pictures suggest: Many features of constitutions can be separately chosen and combined, at least to some degree. Therefore a less simplistic description of the choice situation and its outcomes is more adequate: Whenever we have reasons to describe a specific feature of a constitutional arrangement as a minimal consensus feature we may safely consider it as a rational and stable constitutional choice; whenever we find only a minimal consensus class of constitutional features we cannot make an assertion of this kind, but at least we may safely consider all constitutional arrangements which do not fall into this group as neither rational nor stable constitutional choices. Empirically, we then expect different societies to be similar with respect to all constitutional features where a definite minimal consensus can be identified; we expect a limited variety among political communities with respect to those traits 8 9
Cf. Scharpf (1997), p. 72 ff.
The classic example is the exodus of the Roman plebes to the mons sacer, 287 B.C. and the then following constitutional reforms (lex hortensia; Titus Livius, Ab Urbe Condita, 2, p. 32). Cf. also Roemer (1985, 1996).
On the Rationality and Stability of a Minimal Consensus
77
where at least a consensus range can be identified; and finally, we have to expect unrestricted variety among political communities where neither is the case. In matters where we are able to identify a minimal consensus we have to expect not necessarily one encompassing political community, but at least uniformity across different political communities. Whether there might exist only few big or many small ones is open – all we know is that in the matter at hand lies no specific incentive to go different ways. To get an idea of which constitutional features may belong to which group, it is best to distinguish among different types of problems which are solved or addressed by the choice of specific constitutional features. Some of these features shall be excluded right at the beginning from our discussion, namely, all constitutional arrangements which organize the political game itself. The most basic of them – democracy and the rule of law – are not really contested in principle, so that it would be somewhat artificial to look for a minimal consensus argument for or against them. Others – like having a federal system or not, or like having a system of proportional representation or not – are certainly contested, but it is obvious that we do not have theories about them which suggest the existence of some rational minimal consensus or rational consensus range with respect to them. It is more promising to look at those institutions and rules which determine the way in which people live together in a society, i.e. a society’s “normative order”. Here on the one hand we do have some theoretical knowledge about the performance of arrangements, and on the other hand we know the most salient differences of opinions which might enter hypothetical contractual deliberations. Three different types of problems of choice can be discerned here: 1. Rules of cooperation have to be chosen and made reasonably safe against corruption by opportunistic behaviour. This is the first level of questions of justice: principles of commutative and correctional justice have to be adopted here. 2. Principles of the distribution of the gains of social cooperation and principles of the distribution of duties and burdens in a society have to be chosen. On this level, the questions of distributive justice arise: the state’s role not only as protective instance but also as provider and distributor of public goods has to be defined here. 3. One has to choose principles of coping with the potential diversity of the people’s conceptions on the Good Life. In contrast to the first two problems, which were problems of regulating interpersonal relationships, here the individual’s way of life is at stake: To what degree and in what respect is it considered as a matter of private or of public concern? We will now look at each of these three types of problems in turn to find out about the properties of solutions to them, i.e. to find out whether certain solutions or properties of solutions can be singled out as a minimal consensus:
78
R. Zintl
1.2.1 Minimal Consensus and Rules of the Game These rules basically consist in the specific structure of property rights chosen, including the rules of what people may do with their property in their cooperation with other people. With respect to the content of such rules we have no reason to expect a specific minimal consensus: Rules of cooperation can and do differ among different polities (in part certainly depending on the prevalent ideas on the Good Life). We can, however, place different systems of such rules on a continuum of efficiency, depending on their degree of restrictiveness and their degree of group particularism (the rules may make a distinction between members and non-members of the polity or they may make no such distinction). This fact will not directly prevent different contracts from being possible, but it may play a role in the contractual deliberations of rational people. What should be more or less similar across all societies is not the content of the rules of conduct, but rather the ideas on what makes the adopted rules work. Rules, which are not senseless must restrict behaviour at least sometimes and therefore are ridden with a well known dilemma: It may be in the general – constitutional – interest of all members of the group to have a certain rule, but still their concrete – situational – interest may make it attractive to break that rule.10 The temptation to behave opportunistically can partly be kept under control by external sanctions and incentives, but these external controls rest on self control and work together with it. The idea of a person which has the ability to control – and if necessary, to punish – herself, a virtuous person, is the same across societies and cultures: Rights should be mutually respected, promises should be kept, and reciprocity should be honoured. A virtuous person is trustworthy, and being trustworthy is an indispensable ingredient of any non-trivial cooperation. Like the rules themselves, virtue can be socially defined and practised in a universalistic way – “be trustworthy against everyone!” – or in a particularistic way – “do not cheat your own people!”. A society where the rules are particularistic and the dominant conception of virtue is congruent with this feature is a closed society. On the other end we can imagine a perfectly open society where the rules are universalistic in their content and where virtue is conceived in the same way. In between these extremes we find those societies where the relative weights of universalism and particularism are mixed in varying shades, depending on the character of spheres of life.11 In particular, universalistic formal rules and particularistic conceptions of virtue are not a rare combination.12 In this area, then, we have no strong reasons to describe some definite content of a social contract as a minimal consensus among rational actors; at the same time we have reasons to argue that ideas on virtue will be present in any reasonable contract – not as a compromise in the face of divergent ideas, but as convergence ex ante.
10
Vanberg and Buchanan (1988, 1989).
11
Walzer (1983).
12
Zintl (1997).
On the Rationality and Stability of a Minimal Consensus
79
1.2.2 Minimal Consensus and Principles of Distribution Here the problem is not a problem of iustitia commutativa as before, but a problem of iustitia distributiva: The questions are, how gains from social cooperation should be distributed in general, what ideas of solidarity and positive duties should be incorporated in the constitution, how burdens should be shared, and how the access to socially valued goods should be determined. In abstract terms here we can delineate the boundaries of a minimal consensus range which are determined by the exit costs of conflicting groups; i.e. their best available alternative: Since distributional conflicts on a constitutional level are conflicts among societal groups, any group will enter the contract only if the conditions of the contract are not worse than segregation. However, that does not mean that the limits to intergroup redistribution are clearly visible in reality: People’s ideas about where these limits lie – and whether there are any limits at all – depend on their theories on how institutions work and what the respective bargaining positions of the different groups are. Certainly we have no single theory about such matters which is accepted by everyone. Therefore, we find different and conflicting normative theories on what constitutes justified entitlements. For instance, income generated by property may be regarded as legitimate income or as a kind of theft. In this area no specific content of a possible social contract can be identified as a minimal consensus, but that does not mean that an unlimited diversity of social contracts is possible; rather a limited range of admissible systems of rules exists.
1.2.3 Minimal Consensus and Conceptions of the Good Life In this field we have to accept unrestricted diversity. Not only the conceptions may differ in many ways, but also the ideas on how to cope with the differences among conceptions: Any mixture of indifference, toleration, respect, conflict and segregation may be practised. We have no reason to describe some arrangement as a minimal consensus among rational actors, not even a minimal consensus range. The most basic choice which people have here is the choice to live either in separate and internally homogeneous communities or in communities where heterogeneity is accepted. Among polities where people choose segregation and internal homogeneity, any kind of differences in the publicly defined content of the conceptions of the “Good life” are possible, but we can expect similarities in the rules which preserve their internal homogeneity. Among polities where people choose respecting diversity, we expect an absence of a public conception of the Good Life – combined with the public respect for the individual conceptions on the Good Life – and no major differences in the rules which make a peaceful diversity possible. Here we have no reason to single out some basic decision on privacy or publicness of ideas of the “Good Life” as minimal consensus, but again – given that decision – we can expect a consensus on the secondary rules which are implied by the respective choices. If we apply the concept of a minimal consensus in a strict way, namely referring to what societal arrangements are possible or impossible, then we have to admit
80
R. Zintl
that the concept is not very helpful – it makes not a great difference, since it does not exclude much. In other words: So far, we have not made much progress over classical contractualism: We find strong contractual arguments for order instead of anarchy, but there are no similarly strong contractual arguments for a specific order. Things might look better if we use the concept of a minimal consensus in the weaker sense and ask only whether we can at least single out classes of efficient institutions. Now, we do not ask whether or not it is possible to do something “your own way”, but whether it is rational and whether it can be expected to be stable in the long run. In order to answer such questions we have to be more explicit in our description of the contracting actors – on their preferences and dispositions. We will look first at questions of the rationality of a specific consensus.
2 Rationality Problems 2.1 The Concept of Rationality I will use the following weak concept of the rationality of subjects: First, they are considered as being formally rational in the sense that they will never settle for an option they consider as only second best in a given situation. Second, they are considered as being materially rational in the sense that their own economic well being is a strong, but not necessarily the only, ingredient of their evaluation of alternative options. This minimal concept leaves open a wide range of possible properties of actors and accordingly a wide range of possible ways to model rational actors. What model we use should depend on our theoretical problem: It makes a difference whether we have in mind rational persons as constitution makers, with whom we argue or whether we have in mind rational persons as the actors which live under a set of rules, about whom we argue. That the same persons might play both roles does not imply that we have to model them the same way in both roles. An argument why one and the same person may be modelled differently as a rule maker than as an addressee of the rule can be made and it is simple: People might well accept an argument on the level of constitutional thinking, i.e. as members of the sovereign, without being able to live up to it afterwards, i.e. as subjects. This is a valid reason to ask different questions, depending on the context. Where rational subjects figure as the constitution makers of our arguments the relevant questions are: – How, why, to what extent do we conceive of them as capable of understanding the reasoning? – Do we view them as capable of developing norms which are in their enlightened interest or as capable of developing only norms which are in their immediate interest?
On the Rationality and Stability of a Minimal Consensus
81
Where rational subjects figure as the actors who live under the rules we have to answer these questions: – Do we view them as capable of understanding their society well or not so well? – Do we view them as capable of acting norm oriented or not? These questions suggest three dimensions of properties of the actors which are relevant for our descriptions – their information processing capacities, their virtue, and their values and ethical convictions. Concerning the information processing capabilities we can say: Subjects may either be perfect calculators or only endowed with bounded rationality. The difference is not between perfect and imperfect information, but between high and limited information processing capabilities. Subjects endowed with high information processing capabilities perform adequate problem analyses, use well reasoned theories, argue coherently, and choose optimal courses of action, meaning they are able to maximize their benefits. Subjects endowed with low information processing capabilities use simple theories, are not particularly interested in forming coherent pictures of the world, and choose in a satisficing13 instead of a maximizing manner. Concerning virtue the following can be said:14 Actors may either always behave opportunistically and try to optimize with respect to the specific situation given now and here.15 Alternatively, they may be endowed with a capacity to act according to long term considerations combined with a capability and propensity to follow general norms which take into account more than their immediate and short term interest. This capacity to control oneself, irrespective of the specific content of the rule which is followed, we may call formal morality. Finally, with respect to values and ethical convictions, actors may either care only for their economic interests or they may also consider other aspects of life as important. These values refer to their own non-economic interest or to the welfare of other persons or to general convictions about what is just or unjust. These latter convictions we call their morality in a material sense.
2.1.1 First Conclusions The second and the third properties of subjects are conceptually distinct – we can certainly think of someone who has a formal moral capacity and is at the same time interested only in income maximization; we can also imagine a person endowed with strong substantive ethical convictions, but no capacity or propensity to act accordingly. If we combine these dimensions we get eight possible cases:
13
Cf. Simon (1957, 1972).
14
Cf. Baurmann (1996).
15
Cf. Williamson (1985), p. 64 ff.
82
R. Zintl Table 4 Types of Actors
Good calculator Limited calculator
No virtue Only Also income values Old homo oeconomicus
Virtue Only Also income values Philosophical spectator New homo Complex actor oeconomicus
The four types which are given names are of particular interest for the following argument, depending on the context in which we use them: If we talk about the rationality of some minimal consensus, then our actors should be modelled as rich as possible in their capacities – this is fulfilled by the philosophical spectators; if, on the other hand, we look at the working of institutions, we should use as model the old homo oeconomicus,16 who is a perfect calculator and a virtueless materialist. For more real descriptions our model should be closer to reality – we should use as our model an actor whose calculating capacities are limited and who at the same time has at least some capacity to practice virtue. What has to be found out here is whether values and ethical convictions make a difference in this case. Therefore it is best that we look at both the new homo oeconomicus, who has virtue and is essentially a materialist, and at the most complex type who is not confined to materialist preferences.
2.2 The Rationality of a Minimal Consensus In dealing with questions of the rationality of a minimal consensus it is the rationality of the “constitution makers” which is of interest to us. For the hypothetical constitution builders the following description seems adequate:17 First, it is assumed that the theories the constitution makers have about the world are correct. This is justified since it would make no sense to model those who choose a constitution as badly informed on the working of institutions: The whole argument is hypothetical and it would be bizarre to take as the foundation of our thought experiment knowledge which is not the best knowledge available. Second, the constitution makers take their own and everyone’s income interests very seriously. This is a simplification which is justified since income interests are the only type of interest which is not controversial as such. The income of individuals is of instrumental value to whatever final goals people may have. The simplifi16
For the distinction between the old and the new homo oeconomicus cf. Baurmann (1996). Strictly speaking, the “old” economic man is not the standard model of the classics as Baurmann seems to suggest: For the Bernard Mandeville of the Fable of the Bees it might be true, but not for the Adam Smith of the Theory of Moral Sentiments.
17
Although one has to mention that this description is normatively highly loaded, and in some aspects obviously counterfactual.
On the Rationality and Stability of a Minimal Consensus
83
cation is not overly restrictive, since it contains some dose of universalist morality: not only their own interest is taken seriously by them, but also the same interest of other people. Third, the constitution makers respect the convictions of other people concerning the Good Life, which means that they take – in forming institutions – a neutral position among the convictions of different persons and insist on mutual respect. In any case, they resist the temptation to impose their convictions on other people. This description of the constitution makers inevitably pre-empts the decision on the content of the constitution in essential aspects. Finally, the constitution makers know that most people, including themselves, are not complete consequence-oriented opportunists, but have some capacity to respect and internalize norms which are in their well informed common interest, but that saints still are a minority in every society. This is normatively harmless and empirically sound. As these people know they will have to live under the rules they consent to, they will have a propensity to choose the following institutions:18 The rules of the game which they choose contain only those restrictions which are necessary to ensure peaceful coexistence between the members of the society and which do not put additional restrictions on their voluntary cooperation. The rules are neutral with respect to the personal conceptions of the Good Life. The rules are universalistic rather than particularistic.19 The reason why they consent on minimal and universalistic rules is that such rules and this conception of virtue are efficient and therefore attractive with respect to the income interests of the members of the society: A more encompassing and therefore a potentially more profitable societal cooperation is possible under these rules than in the case of particularistic and restrictive norms and a particularistic interpretation of virtue.20 The principles of the distribution of benefits and duties which are chosen by our constitution makers are built on ideas about legitimate entitlements (resulting basically from voluntary cooperation), combined with some transfer arrangement which keep the risks of autonomous cooperation limited. The transfer schemes take into account long term incentive effects. Whether they incorporate a Rawlsian maximin rule or insurance elements or minimal endowment schemata is open;21 but neither full public control of the distribution as process nor equality as the standard will be adopted – again for efficiency reasons. Not surprisingly, constitution makers who are endowed with the opinions described above will agree on such legal rules for dealing with diversity of convictions
18
We use the distinctions introduced in Part I.
19
Also, the conception of virtue the constitution makers favour is universalistic, but this is a matter of the culture they would like to have in their society, not a matter of explicit rules.
20
The basic argument here is Friedman’s argument that nepotism and discrimination is irrational since it implies foregone opportunities. cf. Friedman (1962), pp. 108 ff.
21
Cf. Ibid., pp. 190 ff., on negative income tax or Hayek (1960), pp. 285 ff. on social security provisions.
84
R. Zintl
that encompassing and heterogeneous political communities are supported, preferably based on social norms of mutual respect, or at least tolerance, of the citizens. All this is very similar to what Rawls calls an overlapping consensus.22 This decision against constitutions which favour small and homogeneous communities is not only normatively attractive for them, but it is also attractive with respect to their material interests because an open society provides more opportunities of profitable cooperation than a closed society. The constitution described is “efficient” and it has traits of a minimal consensus: Other constitutions might give some privilege to this or that specific sub-group of the constitution makers, and therefore may be more attractive to the members of these specific groups, but no other group has reasons to agree to this. In contrast to our results in the preceding section we now have a somewhat more definite picture. However, as already has been said, this picture is strongly impregnated with normative pre-decisions. Other descriptions of the contract partners and their preferences would lead to other constitutional choices. Certainly we can find good arguments for the description of the actors used here, but nevertheless it is a description which rests on normative premises which no one is forced to accept. Therefore, we do well to interpret this normative version of a minimal consensus as not more than a possible argument about the merits of certain rules which we now have to confront the consensus which might be stable among real actors. It may well be that our philosophers’ design has the defect that it works only among philosophers and not among real people.
3 Stability Problems 3.1 Models of Actors The question now is about another kind of social contract, not the explicit social contract which is considered by impartial subjects in a hypothetical situation, but the implicit and at the same time real social contract detectable in every day behavior.23 The validity of this contract can be seen in the respect or disrespect of rules, in the severity of social conflict, and in the processes of integration and segregation. As has been said above, three different models of the actors will be used here – the classical or old homo oeconomicus, the new homo oeconomicus and a complex actor. The old homo oeconomicus is a good or even perfect information processor who is predominantly income oriented and whose behavior is strictly consequence oriented; the new homo oeconomicus is also predominantly income oriented, but closer to reality with respect to two other properties: His information processing capabilities are limited and he has a capacity to behave norm-oriented. The last 22
Rawls (2001), 11, pp. 54–60.
23
This is the social contract Binmore (1994) talks about.
On the Rationality and Stability of a Minimal Consensus
85
model is closest to reality: Complex actors are restricted in their information processing capabilities, they have a capacity to behave norm-oriented, and they are not necessarily only income oriented, but also may have other values, including strong ethical convictions. Each of these descriptions is theoretically useful since each of them brings into focus different aspects of our problem as we shall see presently. We will examine the different constitutional areas discerned above one by one:
3.2 Rules of the Game and their Enforcement The old homines oeconomici as perfectly calculating and income oriented opportunists easily identify their common constitutional interests and have no problem in agreeing on the content of the needed rules; they also see the advantages of universalistic over particularistic rules. In all these respects their views are identical with the views of the rational constitution makers. Nevertheless, the old homines oeconomici are harassed by serious stability problems of the order they establish: Since they are opportunists they see no reason to follow the rules even if they accept them, leading to a highly expensive but never completely effective external enforcement of the rules among them.24 The new homines oeconomici, in contrast, as imperfect calculators, have problems in identifying their constitutional interests and may find efficient rules only by trial and error; they also may have a myopic propensity to prefer particularistic practices. In these respects their situation is worse than the situation of the perfect calculators. On the other hand, due to their capacity to behave norm-oriented, they encounter less stability problems than perfect calculators. On balance, the advantages of their capacities of virtue outweigh the costs of limited intellectual capabilities, since the latter can be reduced by learning, whereas the former are just given. The third type of actors, whose distinguishing feature is the weight they put on ethical, religious or other non-materialistic convictions, may come to a different and stable kind of consensus on the rules of the game they want. First, they judge the rules not only by efficiency criteria. Second, the rules they prefer may make a difference between their comrades and the rest of the world. In both cases the potential material gains from more extended cooperation are seen as either irrelevant or are even detested as a sort of corruption. Where this is so, restrictive and particularistic rules and particularistic habits cannot be interpreted as imperfections which should sooner or later be weeded out by learning processes (as was possible in the case of the new homo oeconomicus), but as deliberate choices. Inside a given group, no stability problem exists, as long as the convictions are intact. If
24
If opportunists are imperfect information processors, the overall consequences are similar – on the one hand, enforcement is even less effective, but, on the other hand, the potential rule breakers are less cunning and more prone to follow habits.
86
R. Zintl
groups with different convictions and strong internal loyalties and consequentially particularistic interpretations of rules of conduct live together in one and the same legal order, stability problems of the society at large are endemic, not only because particularistic interpretations of virtue lead to frictions between groups, but also since the people may not have the feeling of belonging to the same society in the first place. As far as the problem of rules and order is concerned, we can summarily describe the picture in this way: Information and information processing capabilities are not decisive in this context. What counts are the “non-rational” features of our actors – the presence or the absence of virtue, the presence or the absence of ethical convictions. These features work in opposite directions: Virtue, on one hand, is a decisive ingredient of the stability of rules – if there is consensus on the content of the rules and the principles behind them. The capacity to strive for more than pure material gain, on the other hand, makes this consensus less plausible and is a potentially divisive force.
3.3 The Distribution of the Gains of Cooperation The old homines oeconomici, as interest oriented and perfectly calculating consequentialists, acquiesce without too much conflict and delay to some societal bargaining equilibrium on principles of distribution and risk sharing, since they have adequate theories about the working of institutions. These principles are similar to those applied by rational constitution makers in that they take into consideration entitlements on the one hand and incentive and efficiency arguments on the other hand which put limits to politically enacted transfers and redistribution. This consensus on principles does not prevent them from trying to exploit the institutions and also to modify the principles of distribution whenever the power balance in their society shifts. But the latter temptation is not too strong – power games are tamed by embedding efficiency considerations. Therefore, individual shirking inside the institutions is endemic, but the institutions are not seriously challenged. The situation of our second type of actors, the new homines oeconomici, is substantially different: By assumption, they have no serious stability problems resulting from opportunistic behaviour, which would make their institutions more stable than those of the old homines oeconomici. The bad news is that their limited information processing capacities do easily outweigh this advantage. Not only the bargaining among them will be cumbersome, much more fatal is the fact that they do not reach a common definition of the situation: No consented and adequate theory will embed and tame their distributional conflicts, but rather their theories will be simplistic and self-serving. There is no mechanism by which bad theories are weeded out, rather bad theories drive out good theories. If the actors reach some equilibrium at all it is only a temporary cease fire, open to challenge anytime. Things could be even worse in the case of the third type of actors, where bad theories are combined with strong ethical convictions. Here compromise solutions
On the Rationality and Stability of a Minimal Consensus
87
or a cease fire could be nearly impossible. If in distributional matters people have fundamental ideas of fairness which collide with efficiency and incentive considerations, they may willingly forgo compromise opportunities which they consider as humiliating.25 This may well lead to a breaking up of the cooperation even if people know that this is not in their economic interest – which would be impossible in the case of people who consider their economic interest and not their convictions as the ultimate criterion. Concerning the choice and stability of distributional principles the overall picture therefore is this: For the old homo oeconomicus the range of consensus is clearly visible and consented. In the case of the new homo oeconomicus the idea of some range of admissible distributions is still valid, but its boundaries are blurred and distributional conflicts therefore are more intense. Finally, in the case of complex actors it is even possible that the distributional conflict is not tamed at all – people may willingly accept losses for the sake of principle. In clear contrast to what we saw in the case of rules of conduct, the information processing capabilities of the actors are the decisive feature here. This is obvious if we compare the two homines oeconomici, but it is also relevant for complex actors: Among good information processors, who have adequate theories about the working of social cooperation, ethical convictions and ideologies are somewhat restricted by their knowledge and may even help in stabilizing a consensus. Among bad information processors, on the other hand, where things are contested all the way, moral convictions may deepen the conflict among constitutional preferences.
3.4 Conceptions of the Good Life Both varieties of economic man have no problem at all with differing conceptions of the Good Life, since ex definitione they do not hold such convictions, at least not as public convictions with political consequences. Since economic men measure the quality of life mainly or only by economic success and see anything else as a question of private consumer sovereignty, they do have no divergences or conflicts in these matters and they do not have to care about minimal consensus or compromise in that region. All this is obviously different in the world of complex actors. The severity of the problems which confront them depend on their priorities and on their information endowments. If their values and ethical convictions are only a part of their thinking and do not dominate everything, then we can still expect mutual respect and tolerance in the limits of the general legal order. We also can expect limits to segregation, since here too, more profit generating cooperation is possible if people do not restrict ex ante their cooperating opportunities.26 If, however, the convictions of the actors are more important to them than their economic interests, then we 25
Think of the typical behaviour in Ultimatum Games (G¨uth and Tietz, 1985, 1990).
26
Cf. Friedman (1962) on discrimination.
88
R. Zintl
have to expect conflict, forced assimilation, massive segregation, even expulsion. Information and information processing capabilities play a similarly mitigating role here as in the case of problems of distribution: The better informed and the fitter to digest information the actors are, the more opportunities of conflict resolution and compromise do they see and the less stereotyped and hostile will their perception of people with other convictions be. Conflicts and differences may persist, but their intensity is lowered and unnecessary conflicts may vanish altogether.27
4 Normative Consequences Problems of the stability of a minimal consensus are problems of social integration. We have seen that these problems differ in kind and severity, depending on the properties of the actors. We also have seen that it depends on the nature of the matter which properties of the actors are critical: When, on one hand, the stability of the rules of the game and their enforcement is the task, virtue is the decisive property, whereas the information processing capabilities of the actors and their moral convictions play a lesser role. When, on the other hand, questions of just distribution and of values are at stake, morality plays a central and often divisive role and the information processing capabilities of the actors are a crucial intervening variable. The better these capabilities, the greater is the chance that the ethical convictions of the actors are not a divisive force. It is a theoretically interesting fact in itself that a more complex and rich model of the actors has ambiguous consequences. The infamous homo oeconomicus is not responsible for all evils in society and our picture of the world does not become friendlier whenever we give this homunculus a more human face. The consequence which should be drawn from this result, however, should not be an argument about which – if any – of the descriptions given above is on balance most accurate or realistic. It is much more fruitful to focus on the different traits separately and then to ask what kind of actors we should be and in which of our properties we should therefore invest. Hayek’s answer to this question, for instance, is basically that we should take as given our very limited information processing capabilities, that we should strive to be virtuous people in the framework of abstract rules of just conduct, and that we should try to suppress our “moral” instincts with respect to the overall order of society.28 This advice fits into the picture given on the preceding pages, and, if successful, would obviously solve the stability problems described. Nevertheless, it is probably somewhat paradoxical in the proposed combination of traits. Is it plausible that someone is at the same time a virtuous being and still indifferent with
27
This assertion needs a qualification: It is well possible that the protagonists of ideological or religious conflicts, the moral entrepreneurs, are well informed and highly ‘rational” – but their chances to mobilize people into conflict are the smaller the less ignorant their audience is.
28
Hayek (1976); Zintl (2003).
On the Rationality and Stability of a Minimal Consensus
89
respect to material moral questions? Is it not psychologically much more plausible that virtuous persons tend to have also material ethical convictions, and that persons who are ethically indifferent tend to be consequence oriented rather than virtuous actors? If this is so, then it would be more promising to respect our moral convictions and values as given and to invest in the other two properties – in virtue (like Hayek proposes) and in information processing capabilities (in some contrast to Hayek). Admittedly, this last point is speculative and may be wrong. However, what should be clear is that the rational legitimization of norms among real actors is a multi-layered enterprise in which it is by no means obvious whose and what rationality is to be taken as the measuring rod.
References ¨ Albert, H. (1976): “Politische Okonomie und rationale Politik”, in: Aufkl¨arung und Steuerung, idem: Hamburg: Hoffmann und Campe, pp. 92–122. Baurmann, M. (1996): Der Markt der Tugend, T¨ubingen: Mohr. Binmore, K. G. (1994): Game Theory and the Social Contract, Vol. I: Playing Fair, Cambridge (MA): MIT Press. Buchanan, J. M. (1975): The Limits of Liberty, Chicago: The University of Chicago Press. Friedman, M. (1962): Capitalism and Freedom, Chicago: The University of Chicago Press. Gordon, H. S. (1976): “The New Contractarians”, in: Journal of Political Economy, 84, pp. 573–590. G¨uth, W. and Tietz, R. (1985): “Strategic Power versus Distributive Justice. An Experimental Analysis of Ultimatum Bargaining”, in: Brandst¨atter, H. and Kirchler, E. (eds.):Economic Psychology, Linz: Trauner. G¨uth, W. and Tietz, R. (1990): “Ultimatum Bargaining Behavior. A Survey and Comparison of Experimental Results”, in: Journal of Economic Psychology, 11, pp. 417–449. Hayek, F. A. (1960): The Constitution of Liberty, Chicago: The University of Chicago Press. Hayek, F. A. (1976): Law, Legislation and Liberty. Vol 2: The Mirage of Social Justice, London: The University of Chicago Press. Hobbes, T. (1991 [1651]): Leviathan, edited by Tuck, R., Cambridge: Cambridge University Press. Hume, D. (1964 [1748]): “Of the Original Contract”, in: The Philosophical Works, edited by T. H. Green and T. H. Grose, Bd. 3. Aalen: Scientia, pp. 443–460. Koller, P. (1987): Neue Theorien des Sozialkontrakts, Berlin: Duncker & Humblot. Locke, J. (1988 [1689/90]): Two Treatises of Government, edited by Laslett, P., Cambridge: Cambridge University Press. Nozick, R. (1974): Anarchy, State, and Utopia, New York: Basic Books. Rawls, J. (1971): A Theory of Justice, Cambridge (MA): Harvard University Press. Rawls, J. (2001): Justice as Fairness: A Restatement, Cambridge: Harvard University Press. Roemer, J. E. (1985): “Should Marxists be Interested in Exploitation?”, in: Philosophy and Public Affairs, 14, pp. 30–65. Roemer, J. E. (1996): Theories of Distributive Justice, Cambridge: Harvard University Press. Scharpf, F. W. (1997): Games real Actors Play. Actor Centered Institutionalism in Policy Research, Boulder: Westview. Simon, H. A. (1957): Models of Man, New York: Wiley and Sons. Simon, H. A. (1972): “Theories of Bounded Rationality”, in: McGuire, C. B. and Radner, R. (eds.): Decision and Organization. New York: Elsevier, pp. 161–176. Vanberg, V. and Buchanan, J. M. (1988): “Rational Choice and Moral Order”, in: Analyse & Kritik, 10, pp. 138–160.
90
R. Zintl
Vanberg, V. and Buchanan, J. M. (1989): “Interests and Theories in Constitutional Choice”, in: Journal of Theoretical Politics, 1, pp. 49–62. Walzer, M. (1983): Spheres of Justice, New York: Basic Books. Williamson, O. E. (1985): The Economic Institutions of Capitalism, New York: The Free Press. Zintl, R. (1983): Individualistische Theorien und die Ordnung der Gesellschaft, Berlin: Duncker & Humblot. Zintl, R. (1997): “Social Relationships and Economic Theories of Clubs” in: Associations, 1, pp. 96–116. Zintl, R. (2003): “Moral: Wissen, Meinungen, Interessen”, in: Pies, I. and Leschke, M. (eds.): F.A. von Hayeks konstitutioneller Liberalismus, T¨ubingen: Mohr, pp. 241–258.
A Commentary on Zintl Rafaela Hillerbrand
Zintl’s approach on how to solve a variant of the Hobbesian problem of social order – concerned not about why people prefer political order to anarchy, but what kind of political order they choose rationally – differs significantly from Hobbesian approaches: Zintl does not reduce the ‘natural’ conditions of social interaction to the Hobbesian assumption of selfishness. Instead of being interested only in their own economic wellbeing,1 more realistic actors whose actions might be also guided by other (formal and material) interests are considered.2 This approach indicates a middle way between the discussions in social decision theory and many approaches discussed in the political sciences which, to some extent, give up a systematic treatment. While at present the former are restricted to the analysis of oversimplified models and thus do not give any information concerning the problems of constitutions for real people, the latter are quite often no more than a description of empirically visible problems of constitutions among real people. Due to the consideration of more complex actors, Zintl’s approach is much less able to be formalized than the problems dealt with in social welfare theory.3 Due to the simplifications he nonetheless makes, the normative power of his results is weaker than that of the approaches concerning more realistic actors. Yet this middle way that Zintl indicates seems to be a fruitful approach of how to tackle the problem of a rational legitimatization of a stable political community. We begin with a short summary of those aspects of Zintl’s analysis that this comment focuses on (I). The way in which he takes into account those features of the addressees of the arguments on political norms that cannot be reduced to interests in their own economic wellbeing is investigated in more detail (II). We conclude with the way Zintl distinguishes different constitutional features where a minimal consensus can be identified (III).
1
This is identical to what Zintl refers to as ‘material rationality’.
2
This aspect of Zintl’s account be seen as a generalization of the distinction between constitutional and compliance interests as it was introduced by Vanberg and Buchanan (1988). 3
See, for instance: Arrow (1963).
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
91
92
R. Hillerbrand Table 1 Stability of a minimal consensus
Old homo oecomomicus New homo oecomomicus Complex person
Rules of the game
Distribution
Good life
I S I
S I I
S S I
The horizontal axis distinguishes different types of choices concerning a society’s normative order, while the vertical axis distinguishes different idealized members of the society (‘S’ corresponds to stability, ‘I’ to instability).
1 Three Sets of Problems for Three Different Actors: A Short Review of Zintl’s Approach When choosing a constitution, the constitution makers – in the ‘ideal’ case these constitution makers are purely rational actors – are aware of the fact that the addressees of their arguments on political norms might obey certain non-rational features. As such non-ideal features, limited capabilities concerning the processing of information as well as formal and material moral convictions are mentioned. These attributes may distinguish different actors. Zintl implements those non-rational features by analyzing and comparing three types of idealized actors that differ according to (a) their intelligence and information, (b) their capability to act norm oriented, and (c) the kind of preferences they want to realize with their actions. Here one can further distinguish between the case where people have interests only in their own economic well being, and the case where they have other interests guiding their actions. For the latter, Zintl mentions material ethical convictions. The results of Zintl’s investigations concerning the stability of a minimal consensus among groups consisting of only one type of actor – ‘old homo oeconomicus’, ‘new homo oeconomicus’, and ‘complex person’ – are shown in Fig. 1.4 The three types of actors listed in the vertical axis face different problems vis-a-vis the three questions Zintl distinguishes when dealing with the reasoning of political norms, listed in the horizontal axis: the so called ‘rules of the game’, the ‘principles of distribution’, and the ‘conception of a good life’. ‘S’ indicates that according to Zintl, good chances for a stable consensus are visible, while ‘I’ indicates instability, that is the different parties will most likely not come to a long lasting agreement concerning this question.
4 Zintl distinguishes between problems related to the stability of a minimal consensus and those that connect to the visibility of a minimal consensus.
A Commentary on Zintl
93
2 Distinction between Constitution Makers and Addressees of Political Norms We do not want to comment on Zintl’s approach in context and thus will not discuss the entries in the matrix sketched in Fig. 1. A more detailed methodological analysis of Zintl’s approach is given instead because we consider Zintl’s approach of how to tackle the problem of stability of a minimal consensus among real members of a society as promising. Nonetheless it still seems to have some conceptual shortcomings.
2.1 Distinction between Material Morality and Economic Interests A very popular approach in rational choice theory proceeds on the basis of preferences (or interests) which are not determined any further.5 This has the main shortcoming that – despite all formal precision – it completely obscures any differences between (a) purely material egoistical interests and therefore interests that are able to be generalized to a certain extent and (b) preferences which are affected very strongly by private convictions and are thus much less able to be generalized.6 By blurring these differences, the evaluation of costs and benefits can become vague and somewhat arbitrary.7 Zintl assumes the preferences (b) to be identical to material ethical convictions.8 Although this distinction seems to be of central importance, especially when considering constitutions, additional differentiations might seem appropriate in order to make the term ‘benefit’ more precise. This would lead to an approach that parallels preference approaches in welfarism.9 Concerning the question what kind of preferences the constitution maker shall take into account (see Section 2), reasoning in the context of preference utilitarianism could be of great value for the discussion of the political order.
2.2 Drawing Normative Consequences Inconsequence: According to the review of Zintl’s ideas, the constitution makers have to take into account that the addressees of the political norms are real people. Zintl thus suggests already implementing inherent deviations from an ideal, purely 5
Cp., for example, Gauthier 1986.
6
See Vanberg and Buchanan (1989), p. 50. We do not hold the opinion that the generalizability of the material egotistical interests is very far reaching. 7
See, for instance: Hillerbrand (2005).
8
Zintl’s distinction can be understood as a refinement of Vanberg’s and Buchanan’s distinction between ‘interest-’ and ‘theory-component’. 9
See, for instance: Singer (1984), Griffin (1986).
94
R. Hillerbrand
rational person at a very early stage of the theoretical reasoning about constitutions. This approach can be seen in contrast to the so called ‘two-level models’ as they were first proposed by R.M. Hare and which found their way to different fields of ethical reasoning.10 In the present context where we reason about a minimal consensus, a two-level approach would first develop a constitution for ideal actors. Only in a second step is it asked how to accomplish the aims of the ideal norms under more real circumstances which allow for the fact that the addressees of the norms are real persons. The purely functional relation between norms on the level of ideal actors and norms on the level of real actors leads to serious difficulties in two-level approaches. These problems could and can be overcome by Zintl’s approach. Nevertheless, applying the analyzed behaviour of the three different idealized actors on how to solve existing problems11 suggests that Zintl’s approach is intentionally fairly similar to two-level models: Zintl’s ultimate aim seems to be a constitution that is stable in some, not further specified, way; looking at the stability problems that different idealized actors face and then deciding what features they have to develop in order to reach a stable consensus parallels the approach on the second level. Zintl’s approach therefore faces to some extent the same difficulties as two-level models.12 For example, if, as in the case of two-level models only the norms for ideal actors are justified, it is not clear to the actor why to obey the rules on the second, ‘intuitive’ level. Solving the problems of stability using Zintl’s approach thus does not result in the desired success. This shall be demonstrated for instability problems that due to Zintl occur among good calculators which lack ‘formal ethical convictions’. Such actors will accept ‘virtue’ only when it yields overall benefits for the individuals. This is usually not the case – as in general one does not face coordination problems, but rather problems of the type of the prisoners’ dilemma: If under some circumstances it is better to settle for cooperation from a collective point of view, an individual or a small subgroup of the society might nevertheless be better off if it or they do not cooperate. Thus inventing or rehabilitating virtue can only be successfully done in a state that consists only of good calculators if it is indeed enforced coercively. It is questionable as to whether the assertion of principles that cannot be explained to the majority really yields a stable constitution. Applicability: Some further serious difficulties arise when one applies the results of Zintl’s considerations in Subsection 3.1 to existing conflicts in order to obtain a stable constitution: When uniform constitutional interests can be identified, Zintl pleads for an investment in virtue. Problems of distribution and of dealing with a diversity of convictions otherwise call for an investment in knowledge and in calculating capabilities. Even if we accept Zintl’s analysis on the stability of constitutional
10
Hare: (1979, p. 146; 1992, p. 70).
11
Zintl indeed seems to suggest to apply his approach to existing problems (e.g. Zintl: this volume, subsection 3.4).
12
For a short review on the problems related to two-level approaches see Hillerbrand (2005), subsection 5.4.
A Commentary on Zintl
95
arrangements as it is sketched in Fig. 1, this yields serious difficulties. In many conflicts we simply cannot know with any degree of certainty whether or to what extent an observed disagreement reflects deficient information or deficient handling of available information, or whether the conflicts are due to different material ethical concepts.13
3 Determining the Rules of Conduct The two previously mentioned problems seem to imply not only that analyzing the status quo, i.e. how real people indeed act, is not sufficient for a normative analysis.14 It implies furthermore that before asking the question of what kind of actors we should be, we have to analyze in more detail what kind of society we want. To obtain a stable minimal consensus it might be more fruitful to continue further along the lines Zintl has taken: This implies to look closer at the social contract argument and look even more closely at what a stable constitution has to look like, while taking into account that the addressees of political reasoning are real persons with certain motivational and cognitive shortcomings.
3.1 Principles of Distributions For Zintl, a stable minimal consensus with respect to the ‘principles of distribution’ only means that the actors agree on a common formal scheme according to which duties and burdens as well as gains from cooperation are distributed among the members of a society. In this case, the stability of a minimal consensus among old hominem oeconomici follows trivially from the definition of an ‘old homus oeconomicus’. The relevant question for political philosophy with respect to a stable consensus concerning ‘distributions’ is however the question if the actors can at least to some extent agree on actual distributions in context. We hold the opinion that concerning this problem, a minimal consensus among purely rational actors if at all, is not easily visible. The question whether rational actors are able to reach a stable distribution has been discussed mainly in the framework of social choice theory.15 At present, the answer seems to be, that – in contrast to Zintl’s analysis – a stable consensus cannot be found unless we assume some capability to obey general rules which
13
Compare Vanberg and Buchanan (1989), p. 59.
14
In a side remark Zintl mentions this himself: this volume, subsection 2.2.
15
At present, game theoretical models seem to be too simplified to give an adequate answer to the question as they cannot treat a large number of participants with different (egoistical material) interest.
96
R. Hillerbrand
amounts to what Zintl terms formal morality or virtue.16 For a detailed review we refer to the classical literature on social decision theory.17
3.2 A Pragmatic Approach to a Rational Legitimization of Political Norms Concerning a social contract argument, the most fundamental question seems to concern what Zintl terms the ‘rules of the game’: These include (i) rules that certain rights have to be mutually respected. But these include as well (ii) the necessity of enforcement of such rules. Concerning the ‘rules of the game’ one might be able to find a consensus ex ante among rational actors, but the consensus that might exist is probably not very far-reaching. With respect to the first point, Zintl mentions this himself: Rules might comprise certain ceteris paribus conditions which yield a particularistic interpretation of the rules in question. Concerning the second point, it should be stressed that even within the same civilization one finds different opinions about how to enforce various rules. Just to mention the different attitudes the majority of people in the United States and in Europe seem to have on the death penalty, or the changing attitude towards capital punishment only within the US since the 1960s. Dividing the question on what a minimal consensus can look like in three subquestions like Zintl does has conceptual advantages. Nevertheless one has to keep in mind that questions concerning the rules of conduct, the way of distributing duties and burdens as well as questions concerning the ‘good life’ have strong mutual dependencies. People who have a strong tendency to interpret rules in a particularistic way will most likely accept other distributions than those who interpret rules more universalistically. Thus a necessary preliminary work has to determine the relation between the three sets of problems more closely – without blurring their conceptual differences. Further, it is necessary at this formal level to look closer at the ‘rules of the game’ and figure out what kinds of rules and what kinds of enforcement of these rules might be acceptable for rational actors – irrespective of their social, cultural, or individual background. A more detailed investigation than that concerning the three idealized actors summarized in the matrix is required. The answer to the question among which types of persons a non empty intersection of acceptable rules exists, will determine in a second step to what extent a constitution is actually capable of taking into account questions concerning the ‘good life’. That is, it will determine whether or to what extent material moral convictions – and other preferences that cannot be reduced to purely egoistical material interests18 – have to be of constitutional concern or to what extent these have to be purely private. A plea for such 16
K¨otter (1984).
17
See, for instance: Feldman (1980).
18
Zintl: this volume, subsection 1.2.
A Commentary on Zintl
97
a constructivistic approach which takes pragmatic19 considerations – the (formal) rules actual societies actually follow – as a starting point is not new. But so far this approach has not been applied systematically to ethics in general or to the question of how to legitimize political norms in particular. Summarizing briefly this comment, we can state two major points. First, concerning the distinction between constitution maker and addresses of the constitution, it is fruitful to continue along the line Zintl seems to suggest initially: Instead of asking what type of people we should be in order for a specific constitution to work, we should take into account from the very beginning of the constitution making process that the addresses of norms have some non-ideal features. Secondly, on the basis of Zintl’s differentiation of the three sets of problems that have to be dealt with by political norms, a more pragmatic approach to problems concerning the legitimization of political power seems promising for tackling the problems connected with a stable moral consensus. Continuing the middle way indicated in Zintl’s paper might indeed contribute considerably in answering the question what political order should be chosen rationally.
References Arrow, Kenneth J. (1963): Social Choice and Individual Values, New York. Feldman, Allan M. (1980): Welfare Economics and Social Choice Theory, Boston/Den Hague/London. Gauthier, David P. (1986): Morals by Agreement, Oxford. Griffin, James (1986): Well-Being, Oxford. Hare, Richard M. (1979): “Utilitarianism and the Vicarious Affects”, in E. Sosa (ed.), The Philosophy of Nicholas Rescher. Dordrecht: Reidel. Hare, Richard M. (1992): Moralisches Denken: seine Ebenen, seine Methode, sein Witz, Frankfurt a. M. ¨ Hillerbrand, Rafaela C. (2005): Technik, Okologie und Ethik: ein normativ-ethischer Grundlagendiskurs u¨ ber den Umgang mit Wissenschaft, Technik und Umwelt, Paderborn: Mentis. K¨otter, Rudolf (1984): “Distributive Gerechtigkeit und Wohlfahrt. Zum Grundproblem einer utili¨ taristischen Ethik und Wohlfahrts¨okonomie”, in: Okonomie und Gesellschaft, 2, pp. 67–105. Singer, Peter (1984): Praktische Ethik, Stuttgart. Vanberg, Viktor and Buchanan, James M. (1988): “Rational choice and moral order”, in: Analyse und Kritik, 10, pp. 138–160. Vanberg, Viktor and Buchanan, James M. (1989): “Interests and theories in constitutional choice”, in: Journal of Theoretical Politics, 1, pp. 49–62.
19
The term ‘pragmatic’ is used as opposed to ‘semantic’ and ‘semiotic’.
Rational Egoism, Morality and Human Rights Matthias Kaufmann
1 The Republican Challenge It was only a few years ago that liberalism could be seen as the only dominant and (mostly) unchallenged political theory to have survived the struggles of the 20th century. All of the remaining differences between serious political theorists and/or political philosophers were simply viewed as variations within the various branches of liberalism. Richard Flathman – in his work Reflections of a WouldBe-Anarchist – drew the theoretical line between a so-called virtue liberalism, and an agency liberalism which he himself favored: “On this construal, agency liberals idealize human beings as restless, pulsating sources and loci of singular ideas and ideals. . .There is no individual and certainly no collective Summum Bonum”,1 so the state’s task consists in the arrangement of conflicting purposes and desires. For virtue liberalism, on the contrary, “life is or can be just and humane,. . ., only if or to the extent that all public and much private thinking and acting are governed by beliefs that are arrived at under the discipline of intersubjective reason or ‘deliberative rationality’. . .Passion, desire and interest must submit to the discipline of reason and reason-based rules.”2 The historical source of virtue liberalism can be traced back to Kant, while its chief contemporary exponents are Rawls and Habermas. Flathman has even gone so far as to characterize the discussions between Rawls, Habermas, and the communitarians like Sandel and Taylor as a “family quarrel” within “virtue” liberalism.3 I have referred to Flathman’s position at some length because he demonstrates quite impressively the vast realm of theories and political positions attributed to liberalism. He himself speaks again and again of family resemblances between the theories he discusses. So, obviously, the ordering can also be done differently, which is of relevance to our topic. The first point we may maintain is that, according to many authors, liberalism also includes political theories in which both negative and positive freedom are seen as essential for modern democracy. Therefore it should come as no surprise that authors like Richard Dagger, who wanted to make the case 1
Flathman (1998), p. 4 and p. 6.
2
Ibid., p. 4 and p. 8f.
3
Ibid., p. 4 and p. 11.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
99
100
M. Kaufmann
for republicanism, tried, above all, to find a way of reconciling republicanism and liberalism.4 One of the authors who (in a certain sense) withdrew from liberalism was Philip Pettit. He claimed that the kind of freedom relevant for republican political theory is neither a negative liberty nor a positive liberty but rather liberty considered as nondomination.5 Domination is defined as the “capacity to interfere. . .on an arbitrary basis. . .in certain choices” of the dominated person.6 Identifying liberalism with a concept of freedom as non-interference, as mere silence of the law, Pettit saw a certain gap (at least) between mainstream liberalism and his version of republicanism. There is no reason for me to doubt either the importance of non-domination or the reasonableness of Pettit’s arguments. The only question is whether he has actually left the realm of liberalism. It might well be the case that he just favors a certain type of liberal political theory. Nonetheless, firstly there may be different ways to order the bundle of political theories, some of which are labeled liberal theories; secondly this ordering does not seem to be his main concern. We shall return to the important concept of non-domination later in the paper. One of the few authors, assuming I have appraised the situation correctly, who really wishes to leave liberalism behind in favor of a new form of democracy based on virtue is Jos´e Eisenberg, a political theorist from Rio de Janeiro. He identifies liberalism with the arrangement of interests (although not necessarily those of selfinterest). Such things as group-interests and the fair regulation of competing interests might be of concern to liberalism. Yet, as long as it remains tied to this concept it will be unable to conceive human rights, peace, and democracy in an adequate manner; for these are “human values” which a “theory of collective action that is centered in the concept of collectively organized interests, is unable to enclose.”7 The problems of liberalism that still hold us within the “elastic prison” of modern political theory seem, according to Eisenberg, to be twofold: Firstly, the contingent historical development – which he describes with the main steps in Spanish Scholastic, in modern republicanism, and in political economy – leads us into an “imperative of representation” for everyone, which established a moral and political gap between the ego of the Lebenswelt and the I as an entity which has respectable interests. This imperative of representation separates the persona naturalis of our everyday practice from the persona moralis of “institutionalized practices which constitute the life of the political community”. It therefore implies a “radical separation” of the discursive practices which constitute our everyday life – these are the practices in which our passions are formed and satisfied – from the institutionalized practices of ethics in which the passions are channeled into interests.8 The outcome is a political situation in which only those who have a voice in political discourse 4
Dagger (1996).
5
Pettit (1997), p. 35ff.
6
Ibid., p. 52.
7
Eisenberg (2003), p. 67.
8
Ibid., p. 50.
Rational Egoism, Morality and Human Rights
101
are seen as relevant and are heard. Originally, this meant only free white men with an independent income. And although the social transformations of the 20th century had removed this limitation, the central problem has remained the same. Even the institutions of civil society, originally designed to escape the dichotomy of state and economy via a revival of republican virtues, run into this logic: On the one hand, their influence is rather limited. On the other, they form yet another sphere in which the persona naturalis has to be able to find representation of her interests in order to be considered present and relevant. Although, I think this criticism is correct, it is not specific to political representation. In many types of modern ethics we also have differences similar to that of homo phaenomenon and homo noumenon in Kant’s ethics with the corresponding problem that the good will of the homo noumenon is free, while I can never be certain whether my will, as the will of a homo phaenomenon, is a good will. Secondly, it will be difficult to escape this situation within the political reality of the common citizen, if she is not satisfied with a new verbalization created by political philosophers – who, e.g., quit speaking of public reason in favor of talk about public use of reason – which is still irrelevant for the everyday life of citizens. The second and even more fundamental point in Eisenberg’s criticism of liberalism seems to be that it is centered around the concept of interests. According to Eisenberg, while it is certainly possible to describe and explain the behavior of contemporary political actors in terms of interests, it is quite impossible to write a normative theory of a real democracy (uma verdadeira democracia) using this terminology. At least, the results of those liberal dynamics, “intentional or not, have been a dangerous threat to the project of human emancipation,” for which we need the virtuous and equal citizen who strives for the common good and not for the articulation of particular interests, be it those of a minority or those of the majority.9 The interests that are to be viewed critically are not only those of myopic egoism. Additionally, long-term interests as well as the codex of honor deduced by Jon Elster utilizing methods of rational choice theory prove insufficient, because it deduces only honesty, whereas a theory of virtue contains several different ethical demands, the most important of which pertain to the cardinal virtues.10 Eisenberg admits that his normative premises are more demanding than those of liberalism but he hopes that with them we might be able to enlarge our elastic prison or perhaps even succeed in stepping outside of it.11 Eisenberg, however, is not so na¨ıve – and not so dangerous – as to postulate a republic of plain virtue or one where politics is only seen as an end-in-itself, neither instrumental, nor “strategic” in the sense of Habermas. He even admits that many of the political debates in any republic will primarily be concerned with economic problems. He just thinks that the instrumentality of politics should not mean that politics serves the pre-figured interests of individuals and groups. His vision sees politics as a means to create different types of normative consent 9
Ibid., p. 74.
10
Elster (1990).
11
Eisenberg (2003), p. 74f.
102
M. Kaufmann
of which the participants may be proud because they consider them virtuous, and therefore they are willing to follow them because they are legitimate.12 On the whole, there seems to be a clear tendency toward (re-)moralizing political theory in two senses: firstly the moral obligation of the citizen to the political community is confirmed as well as the protection of the individual against state interference restricted to protection against arbitrary interference. Secondly, a just community has to fulfill exigencies especially in favor of the underprivileged – the “invisibles” in an extreme version13 – which are more or less founded in egalitarian morality. Nevertheless, there are good reasons to discuss whether a political community can be founded on rational egoism: From a theoretical point of view this might be promising because a political theory could be founded on an anthropology which does not demand too high of a moral qualification from the citizens. A moral aspect/dimension is also implied: A political system which does not set too high a moral standard for its citizen shows, so we may hope a reduced tendency in the persecution and punishment of those who do not succeed in meeting these standards. My defense of rational egoism will, therefore, not consist of a criticism of Eisenberg’s views. Rather I will try to show that, on the one hand, from a rationalist point of view (in the sense of long term egoism), there are good reasons to live and to participate in a moral society. On the other hand, the reference to the demands of rational egoism may provide an instrument to protect the individuals against an intrinsic tendency of virtuous communities to burden the individual with exaggerated moral demands. In a certain sense rational egoism could be seen as more fundamental than a moral approach to political theory.
2 Rational Egoism and Morality I will argue for my first point, namely that rational egoism is able to give reasons for moral engagement via a discussion of the question whether it is – from a purely rational egoistic point of view – a realistic and attractive option to abolish morality. This question contains at least two others: – Is it a realistic and attractive option to live in a society without moral norms? – Is it a realistic and attractive option to live as a permanent free rider in a moral society? As I will try to show, the answer to the first question is more or less clearly “no,” and the answer to the second question is not as clearly “yes” as one might expect. A first consideration shows us that the way in which we show our approval or disapproval towards the behavior of our contemporaries in our ordinary language also implies moral judgments. Approval and disapproval are used in the hope that 12
Ibid., p. 84f.
13
Cf. Pievatolo (1999).
Rational Egoism, Morality and Human Rights
103
this will contribute to making the behavior of our contemporaries more reliable. The use of moral evaluations seem to make this hope more realistic than the mere reference to individual preferences because it provides the acting participants with a common basis for their judgments. The moral judgments of ordinary language with their use of moral concepts do not necessarily contain an ontological commitment to objective moral values as John Leslie Mackie had suggested,14 since we do not have to look at the concepts of ordinary language as labels for (moral) things. So, there seems to be no need to clear an “ontological slum”15 within moral theory. Furthermore it is not at all clear that a revision of everyday language via the removal of these moral implications from the standpoint of rational egoism is as attractive an end as Mackie (and several other authors) seems to think in his work on Hume’s ethics.16 In another of his works entitled “Ethics,” Mackie gives us good reasons why the invention of right and wrong was a good piece of work.17 There have been different attempts, at least since Hume’s “A Treatise of Human Nature”,18 to show the advantages of “artificial” virtues, like justice, for the welfare of a group. And, as man is a zoon politikon, and therefore not in the position to decide whether he or she wishes to live amongst or outside of any group – at most she can decide within which group she wants to live – the persistence of the group is normally useful for the individual. A more modern approach speaks of the evolutionist argument according to which groups with moral values tend to be more fit for survival than those without these values.19 Variants of the prisoner’s dilemma have been used to show or at least to illustrate this point. Some versions even claim that moral behavior is of direct advantage to the acting individual. So, let us take the situation of two gangsters caught after a failed bank-robbery with their weapons still in their hands (unfortunately I have forgotten who first used this example). Furthermore, let us assume that if one of them confesses and the other keeps quiet, the one who confesses will be sentenced to one year of prison, while the other will receive ten years. If both confess to the crime, then they will be sentenced to five years each. Finally, if both stay silent they will receive two years in jail for unlicensed possession of firearms. For the rational egoist it would be most rational to avoid the worst case (ten years), and accept the five years by confessing. If both are members of a moral group of – say – Mafiosi, then they know that the other would never confess, and therefore they would receive only two years of punishment. In this case the advantage of the group is identical with the advantage of the members who are involved: the mafia-gang can make use of these two highly qualified gangsters for a total of six more years if both of them act reliably and are only imprisoned for two years each. Although we know that the advantage of the group is not always 14
Mackie (1977), p. 31ff. and p. 42ff.
15
Quine (1953).
16
Mackie (1980).
17
Mackie (1977), p. 190f.
18
Hume (1978), p. 484ff.
19
Mackie (1977), p. 115.
104
M. Kaufmann
so closely connected to the advantage of its individual members – on the contrary occasionally the interests may run contradictory – we can conclude that not only modern liberal societies and cities of god but also groups of gangsters live better with moral principles. Regardless of whether we ourselves behave morally or not, it clearly seems better for us, if we live in a society of morally acting individuals. This can easily been seen if we look at our own security: Although one can readily admit that morality is in most cases not enough to deter everyone from committing crimes, it would surely serve as an additional motive, and perhaps even succeed in cases where the long-term calculus whether this crime might pay would not prevent theft or robbery. To look at morality from the point of view of rational egoism, and to reflect whether it is better for a person to live in a moral society, whether there are good reasons for her to be moral does not mean that a person can be moral and amoral at the same time and in relation to the same problem. It just means that she is able to exercise a certain kind of abstraction which allows her to think about which attitude towards morality she wants to choose. At least if we draw on the evolutionist argument we may say that it is obviously advantageous for the individual to live in a society where the virtue of justice is respected. Clearly, it is better if there are brave, today we might rather call courageous individuals within “my” society capable of defending me against others. Everyday life is much more convenient if people are of moderate temper, and prudence in political and moral topics is intrinsically good in a perspective of a good life. Therefore, an ethics of virtue cannot be deduced from rational egoism. However, it can be shown that it is better for the group if its members act virtuously – and it is usually better (also in a non-moral sense), for the members of the group themselves. The question of whether in this situation it is better for me to act morally or more attractive to be a free-rider will be discussed later. Now, I would like to briefly touch on this point in the following consideration. There are other ways, in which rational egoism is able to support the “good” behavior of the group members. Christian Thomasius, one of the leading figures of early German Enlightenment, who is sometimes seen as the first to establish the difference between morality and legality, posited, in addition to the iustum and the honestum, i.e. the just and the moral, a third mode of social regulation. To designate this mode he used the Ciceronian term decorum, roughly: “that which is decent”, although there are significant differences to Cicero’s “De Officiis”: For Cicero, the concepts are more or less equivalent, quod decet, honestum est et, quod honestum est, decet,20 “what is decent is moral and what is moral is decent”. The distinction of the terms relates more to thought than to the modes themselves.21 For Thomasius, the concepts cover two overlapping, however different realms of social life: Whereas the honestum deals with the ways to attain happiness which necessarily entail the virtues, the Decorum, the decent, tells us how to behave if we want to be socially
20
De officiis, lib. I XXVII, p. 94.
21
Ibid., p. 95.
Rational Egoism, Morality and Human Rights
105
accepted.22 There is, on the one hand, a “natural” decorum which is valid for all human beings and teaches us many morally relevant things, such as to act with modesty, to exercise gratitude, and so forth. On the other hand we have the so-called Decorum politicum which is subject to historical and social contingency. It differs with respect to the various social layers, and depends on the respective social role, e.g. what is decent at city hall is not necessarily decent at the king’s court.23 This shows us that Kant’s “rational devils”24 are not only able to create states, but also of behaving with good manners because they prefer to be socially accepted. The enormous importance of social esteem finds support in a recent book by Geoffrey Brennan and Philip Pettit. They show that one of the enormous advantages of social control via the “intangible hand” of social esteem is its enormous efficacy stating that social pressure is often a much stronger motivation than financial stimuli or even sanctions.25 They also confirm that the search for social acceptance may coincide with moral motivations: the intangible hand is “virtue compatible”.26 Unfortunately, things do not appear as favorable for morality in a modern liberal and egalitarian sense as it might seem: If we return to Thomasius’ Decorum, it is by no means clear that all of these manners are in accordance with impartial and liberal modern morality. Some parts of the Decorum politicum were explicitly created to preserve the hierarchies between casts, races, and genders. Furthermore, they might correspond to social rules that are taken as seriously and defended as vigorously as we take the rules of morality. A similar objection applies to the evolutionary argument, which states that groups with moral individuals are more persistent than those whose members act only egoistically: A social group, a tribe, a people made up of a dominant caste of warriors might have moral principles which entail very cruel practices. However, they could be more effective and forceful than another with a more mild and humanitarian morality of devotion and care. This was Machiavelli’s criticism of Christianity which had – in his eyes – “effeminated,” and therefore weakened the Europeans.27 Similarly, Rousseau complained that Christian soldiers knew how to die but not how to win the battle.28 In this situation, despite a common tendency, it does not help to say that these discriminating or cruel norms are “not really” moral, nor is it really helpful to say that they belong to views of the good, but not of the right (Rawls) or that they are views of ethics, but not of morality (Habermas). Recently, more than 10% of the voters of the most powerful and stable democracy in the world (and more than 20% of those who voted for the winner of the election), declared that they had given their vote to protect moral values. It is well known that their values are closely connected 22
Thomasius (1718), p. 177.
23
Thomasius (1713), p. 385ff.
24
Kant (1996): Toward Perpetual Peace, p. 335.
25
Brennan and Pettit (2004), p. 255.
26
Ibid., 260f.
27
Machiavelli (1992): Discorsi, II.2, p. 148ff.
28
Rousseau (1991): Du contrat social, IV.8, p. 460ff.
106
M. Kaufmann
to their religious beliefs and in direct contradiction to the normative demands which many moral philosophers consider the implications of well accepted moral principles. Drawing on yet another example stemming from the second half of the 20th century we may remember the discussion surrounding Lord Devlin’s thesis that a society had the right to defend its morals with the instrument of punishment.29 At least one of his cases – prostitution and homosexuality – was still at stake during the campaign leading up to the U.S. election. Although from the perspective of modern moral philosophy it seems clear that any discrimination based on sexual orientation is morally unacceptable, it would clearly be precipitous to deny that the voters were motivated by moral principles. Furthermore, it would be precipitous to deny that those who cling to values of tradition or to subordination, to the welfare of the group are acting in accordance with principles which they consider to be moral. Thus, in addition to the universal morality of the Enlightenment, which advocated the protection of those subject to suffering, and the universal respect for autonomous equals (because any attempt to justify discriminatory practices for morally irrelevant reasons have failed), we have the following types of moral norms: – Norms of tradition and customs which one follows because “this is how we have always done it”. – Norms of religion which are right because they are more or less in accordance with the will of god or of the gods and goddesses. Usually tradition and religion are rather close to each other, but not necessarily. Differences between local tradition and religion can be found in times of religious reform or in situations of diaspora or when religious women and men try to preach and to act against the general decline of morality. – A third form of moral commitment is the dominance of group interest, which in its extreme forms implies that the individual commits offenses against all of its other moral and religious beliefs, if it is deemed necessary for the benefit of the group. Although we are committed to respect traditions and religious beliefs as well as group loyalties, we modern western intellectuals are of the firm conviction that in cases of conflict between differing norms humanitarian, impartial, and universal morality will overrule the norms of other types to avoid cruelties and discrimination. Still, we have to realize that this is not a universal belief in the sense that it is self-evident to everyone. In any case, it is hard to imagine that any larger social group could be completely free from all forms of moral regulation; for one of the things they all share is a stabilizing effect on society. Philosophical anthropologists like Arnold Gehlen insisted that there is an inner tendency of social institutions toward moralization.30
29
Cf. Dworkin (1978), p. 240ff.; Bayles (1992), p. 194ff.
30
Gehlen (1975), p. 26f. and p. 60.
Rational Egoism, Morality and Human Rights
107
Therefore, someone who is skeptical toward all forms of morality, and would prefer to live according to rules of rational egoism is not actually in a position to choose between a society with moral obligations and one without. Rather she is compelled to choose between the different types of moral obligations. Thus, she would seem to have good reasons to engage herself in favor of a humanitarian morality, which accepts her moral right to be left in peace as long as she does not harm anyone, even if the supporters of this type of morality do not consider her a moral person. For this reason I think that it is neither a realistic nor an attractive option to live in a society without moral norms. Up until this point we have discussed rational egoism as a general theory of a moral skeptic who is in favor of removing the oppressive elements of morality, and is ready to concede the same freedom to all others. Those who use moral norms within their particular group with the intention of oppressing, abusing or committing fraud against others show some similarities to the phenomenon of free riders. At first glance and from a perspective that takes only non-moral arguments seriously the optimal choice would suggest living in a society of moral persons as a free rider. However, there are several reasons to doubt the truth of this claim. The first involves Plato’s reply to Thrasymachus’ challenge presented in the ninth book of the Republic. Plato does not accept the suggestion that the best way to lead a good life is to be a successful tyrant because tyrants live in constant fear of revenge by those they oppress;31 a fear further compounded by the uncertainty regarding who among them will carry out the deed. This argument has persisted over the ages, and can be found repeated in various types of literature – even if the majority of it is trivial or written for young people with an obvious intention to educate them. One could respond that it is not only the tyrant who must live in constant fear, but the people living under the tyrant as well. In this respect the most we can say is that he is not much better off than they are. A second point is that to be socially successful in a moral world someone who does not want to be moral – not in the sense that she sometimes does things which she is not supposed to do for moral reasons, but that she intentionally lives a life without moral motives – sometimes has to act as if she were a moral person. This is the case because someone who declares – as Max Stirner – only to rely “on my power” is in hopeless opposition to the majority in any society, and therefore in a much less powerful position. The strategy to escape this desperate outcome in using a double identity would create something like a double intentionality, which in the long run might seriously harm the person’s psyche. Therefore, it is not clear why it should be so attractive for a person to be moral in foro externo – in Hobbesian terms – but amoral in foro interno. In reality it seems much more likely that this person gives an ideological argument that claims to give (moral) reasons for amoral behavior and/or hypocrisy, but in any case something that accepts the relevance of morality. The personal and social costs of this kind of attitude towards morality are the subject of psychoanalysis and different kinds of literature. Therefore, it is not at
31
Plato (1990): Republic, 578a–579e, pp. 744–750.
108
M. Kaufmann
all clear that rational egoism favors amoral behavior, at least if the rational egoist is really clear about its advantages and disadvantages. Lastly, the problem of free-riding loses much of its severity if we realize that, from the perspective of political philosophy, the inside-outside distinction is not as clear as was supposed by some Enlightenment philosophers (at least it would not make that much difference regarding politically relevant behavior). Certainly, if people behave in accordance with the Decorum naturale it will never be completely clear whether they are acting utterly morally because they want to be socially accepted or because they have respect for the law of morals. If we remember, Kant admits that even we ourselves never really know whether we have acted morally: “From love of humankind I am willing to admit that even most of our actions are in conformity with duty; but if we look more closely at the intentions and aspirations in them we everywhere come upon the dear self, which is always turning up; and it is on this that their purpose is based, not on the strict command of duty”.32 It was Hannah Arendt who underlined the crucial role of appearance in politics and political theory.33 For a republican theory, the honesty and sincerity of one’s “final” motives are of little importance compared to the existence of an institution of public reasoning and public justification. This does not mean that we approve hypocrisy.34 It just means that in public discourse there is no need to complain about the fact that we never really know why a person shows morally acceptable behavior. And since a rational egoist has good reasons to live in a virtuous and moral society – even if deep down she tries to be a free rider – there is no reason to believe that rational egoism ultimately destroys political virtues and morality. As I see it, it is not the view of interests, normally associated with rational egoism, which is problematic in modern political theory. Many more problems are raised by the modern tendency to reduce morals to a theory of balancing interests directly, whereas for political theory it may often be more important to judge the behavior of politicians, citizens and perhaps foreigners as either virtuous or vicious, even if we are unable to determine their real intentions which might be governed by interests. After having shown that the reference to rational egoism entails no threat to a republican political theory, I wish to push forward and show why we should maintain rational egoism as a measure for the rationality of political institutions.
3 Rational Egoism and Human Rights So far we found that although incapable of deducing morality itself, rational egoism can provide hints as to why it is useful – actually inescapable – to engage in the moral discourse, especially those favoring a type of morality which protects 32
Kant (1996): Groundwork of the Metaphysics of Morals, p. 62.
33
Arendt (1982), p. 73f.; Benhabib (1996), p. 111; (Assy), p. 2005.
34
On this topic see for example Elster (1998).
Rational Egoism, Morality and Human Rights
109
minorities as well as “amoral” individuals such as rational egoists. Let us take up a moral discussion, and try to show why we should, for moral reasons, defend rational egoism. While acknowledging the existence and the importance of moral rules for the maintenance of social life, rational egoism – taken as a general theory and not as an individual search for advantages – favors a type of morality that limits the impact of any kind of social rule on the individual, be it religious, moral, cultural, or any other form. This would serve to hinder the tendency of those who see themselves as moral from destroying not only immoral behavior, but immoral persons as well. But does such a tendency exist? At least according to some influential authors, it does exist. An early version of this argument can be found in Hegel’s critique of Enlightenment-morality.35 However, various forms are still in use today. For instance, Flathman makes reference to the Theory of Justice in order to show that this tendency can also be found in Rawls: “Recall what Rawls famously (or infamously) says of those ‘who find that being disposed to act justly is not a good for them’: ‘their nature is their misfortune’ and is no reason for the more favorably endowed to hesitate before punishing them for their failings”.36 There are similar criticisms of the alleged or real despotism within discourse ethics. Obviously, this is a moralist critique of hypocrisy and moral fanaticism which may become as dangerous as religious or national fanaticism. And although we know that there is a theoretical gap between moral rigor and fanaticism, it will not always be easy to draw the line in political, cultural, and educational contexts. Furthermore, it is not even theoretically beyond discussion whether strictness or clemency is the morally better attitude in cases of doubt. But what does rational egoism have to do with this? I think it provides us with a general rule for the limits of moral interference in the lives of other persons, especially when the possibility for coercion exists (as is always the case in the realm of political theory). Political community should, except in cases of self-defense and urgent need, take as a maximum of interference what the relevant person might do out of rational egoism. However, it cannot be denied that this position already concedes too much, because it does not give clear “a priori” or “intrinsic” limits for paternalism and even re-education. It simply relies on the empirical fact that resistance will be encountered whenever people feel oppressed by measures of this kind. And within this context it is possible to discuss whether their resistance is rationally legitimate or not. Nevertheless, this position provides an intuitively simple criterion for a degree of interference that should only be transgressed under extreme circumstances. As a result of the continuous doubt it casts upon the legitimacy of any kind of repressive measures against the individual, rational egoism exercises a permanent anarchist pressure that attempts to reduce any and all repression and coercion to an unavoidable extent; for it entails the permanent question: Why should I obey you?
35
Hegel (1955): Grundlinien der Philosophie des Rechts, § 140, p. 125ff.
36
Flathman (1998), p. 8, Rawls (1972), p. 576.
110
M. Kaufmann
Furthermore, it is much less open to ideological abuse than the language of virtues. And while it might very well be that more crimes have been committed out of egoism, it cannot be overlooked that some of the worst crimes have been committed in the name of morality, religion, and political ideals. Some even contend that these ideals are responsible for many more deaths than both greed and ambition. Thus, although I completely agree with Eisenberg with respect to the need for favoring and encouraging virtuous behavior, there are nonetheless good reasons why we should never try to transform virtue into a legal duty. The usual legal and political device which places limits on the force exercised by the state and other political organizations is human rights. They provide both the conditions and limits under which any use of force by political authorities can be deemed legitimate (assuming it can ever be legitimated). But even talk of human rights may not be without its difficulties. Currently, there is a lively discussion regarding the views of some very influential and powerful politicians who claim that it is morally right to implement human rights by military means which extend beyond mere defense in cases of genocide. While there can be no doubt about the central importance of human rights, we can now see that the discussion concerning the most adequate way to realize them has become too difficult to provide an easily applicable tool for intuitive judgment. As a simple instrument of control against expanding and exaggerated claims of moralists we may use the principle that, with the exception of a few special cases, a person should be conceded what she might fairly claim in accordance with rational egoistic interests. If we bring together the ideas, (1) that preliminarily, the easiest way to find out what is in a person’s interest is to let her decide for herself, and (2) admit to a condition of fairness, what we are left with is more or less the Kantian definition of “right”: “Any action is right if it can coexist with everyone’s freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone’s freedom in accordance with a universal law”.37 What we need from the part of the individual is the right to use one’s abilities according to one’s plans without being arbitrarily hindered as long as one does not keep anyone else from doing the same. This right, according to Kant, is the only innate right, and is the right of freedom understood as “independence from being constrained by another’s choice” (Unabh¨angigkeit von eines anderen n¨otigender Willk¨ur).38 Two remarks: First, this is, of course, not a random connection between the realm of law, the field of morals, and the criterion provided by rational egoism. Remember that there are references to rational egoism as an instrument for finding a criterion for just political institutions, not as a means to find a reason for being moral, in many contractual theories, from Rousseau to Rawls. According to Rawls, people in the original position would arrive at their principles because they were rational for them. In Rousseau’s social contract we find that institutions are chosen in such a
37
Kant (1996): Metaphysics of Morals, Doctrine of Right § C, p. 387.
38
Ibid., p. 393.
Rational Egoism, Morality and Human Rights
111
way that one is given the chance to act according to the general will because their rational interest is in agreement with it.39 My second point is that Kant’s innate right – the independence from arbitrary coercion – is very close, if not identical, to Pettit’s liberty understood as nondomination. This, again, is anything but a random similarity; for Kant’s philosophy of law and right was strongly influenced by the republicanism in Rousseau. In this context, what makes Kant’s work interesting is a certain ambivalence in this field (at least at first glance). On the one hand, he explicitly declares that (positive) freedom “in a dependence upon laws” which “arises from his own lawgiving will” is the only freedom in a state after the individual “has relinquished entirely his wild, lawless freedom”.40 On the other hand, his definition of right seems to make a strong case for a supposition in favor of non-interference as long as the individual acts in accordance with the laws. The right to freedom as independence seems to provide the missing link because it demarks the degree of non-interference to which the individual has a right, and therefore sets a limit to the lawgiving will of the individual as a citizen. A remaining difficulty relates to the fact that those who might be called rigorous moralists by some and moral fanatics by others would not say that the restrictions under which they wish to place the individuals were arbitrary. They would insist that these modes of coercion were strictly rational, and if themselves placed in the position of the person currently subject to coercion, they would accept them. This means that they would not feel hindered by the different attempts to avoid arbitrary coercion via the criteria of strict reciprocity and generality neither either as found in Rousseau’s and Kant’s works, or via the construction of an original position and the veil of ignorance in Rawls. Furthermore, they would hold that their aim is just freedom as non-domination. It is this moral attitude which Flathman aligns – correctly or incorrectly – with “virtue liberalism,” and to which he objects in stating his preference for “agency liberalism.” Although I share his concern on this particular point, my strategy is different to that of Flathman. Whereas he rejects the so-called virtue liberalism altogether, I attempt to fill the gap between an individualist view that is open for discussion regarding the use of morality itself and the claim that citizenship entails moral behavior to some degree. I show that, on the one hand, even from this originally amoral point of view it is better to act morally, and yet, on the other hand, from a moral perspective it is morally better to avoid the dangers associated with rigidity and fanaticism. Only those contributions are asked of others – excluding special instance – which are at least in some sense to their own good. In my view, it is this close and indispensable connection to self-preservation that signaled the being of modern political philosophy – partly as a reaction against religious persecution and the religiously motivated civil wars. It can certainly be viewed as one of the pivotal points within liberalism – despite all moral implications – and, as I have tried to show, one that should not be given up on too easily.
39
Rousseau: Contract social, II 4, II 11.
40
Kant (1996): Metaphysics of Morals, § 47, p. 459.
112
M. Kaufmann
References Arendt, Hannah (1982): Lectures on Kant’s Political Philosophy, edited by Beiner, Ronald, Chicago: The University of Chicago Press. Assy, Bethania (2005): “Framing a Cosmopolitan Existence: Cultivating Public Feelings”, in: Jos´e Casanova (ed.): Ethics of finitude – Martin Heidegger, Hannah Arendt, Emanuel Levinas, Sao Paulo: Manole. Bayles, Michael D. (1992): Hart’s Legal Philosophy, Dordrecht: Kluwer. Benhabib, Seyla (1996): The Reluctant Modernism of Hannah Arendt, London/New Delhi: Sage Publications. Brennan, Geoffrey and Pettit, Philip (2004): The Economy of Esteem: An Essay on Civil and Political Society, Oxford: Oxford University Press. Cicero, Marcus Tullius (1994): De officiis, Oxford: Oxford University Press. Dagger, Richard (1996): Civic Virtues. Rights, Citizenship, and Republican Liberalism, Oxford: Oxford University Press. Dworkin, Ronald (1978): Taking Rights Seriously, 2nd edn, London: Duckworth. Eisenberg, Jos´e (2003): A democracia depois do liberalismo. Ensaios sobre e´ tica, direito e politica, Rio de Janeiro: Relume Dumar´a. Elster, Jon (1990): “Selfishness and Altruism”, in: Mansbridge, J. (ed.): Beyond Self-Interest, Chicago: The University of Chicago Press. Elster, Jon (1998): “Introduction”, in: Elster, J. (ed.): Deliberative Democracy, Cambridge: Cambridge University Press. Flathman, Richard (1998): Reflections of a Would-Be Anarchist. Ideals and Institutions of Liberalism, Minneapolis/London: University of Minnesota Press. Gehlen, Arnold (1975): Urmensch und Sp¨atkultur, 3rd edn. Frankfurt am Main: Athen¨aum. Hegel, Georg Wilhelm Friedrich (1955): Grundlinien der Philosophie des Rechts, Hamburg: Meiner. Hume, David (1978): A Treatise of Human Nature, edited by Nidditch, Oxford: Clarendon Press. Kant, Immanuel (1996) Practical Philosophy, in: The Cambridge Edition of the Works of Immanuel Kant, translated and edited by Gregor, Mary, Cambridge: Cambridge University Press. Macchiavelli, Niccolo (1992): Tutte le opere, Firenze: Sansoni. Mackie, John Leslie (1977): Ethics. Inventing Right and Wrong, Harmondsworth: Penguin. Mackie, John Leslie (1980): Hume’s Moral Theory, London/Boston/Henley: Routledge. Pettit, Philip (1997): Republicanism. A Theory of Freedom and Government, Oxford: Oxford University Press. Pievatolo, Maria Chiara (1999): La giustizia degli invisibili. L’identificazione del soggetto morale, a ripartire da Kant, Roma: Carocci. Platon (1990): Werke Bd. 4, Darmstadt: Wissenschaftliche Buchgesellschaft. Quine, Willard Van Orman (1953): “On What There Is”, in: Quine, W.V.O. (ed.): From a Logical Point of View, Cambridge (MA)/London: Harvard University Press. Rawls, John (1972): A Theory of Justice, Oxford: Oxford University Press. Rousseau, Jean-Jacques (1991): Oeuvres completes III, Paris: Gallimard. Thomasius, Christian (1713): H¨ochstn¨otige Cautelen welche ein studiosus juris Der sich zur Erlernung Der Rechtsgelahrtheit auff eine kluge und geschickte Weise vorbereiten will/zu beobachten hat, Halle: Renger. Thomasius, Christian (1718): Fundamenta juris naturae et gentium, Halle/Leipzig: Salfeld, 41718 (reprinted Aalen: Scientia, 1979).
A Commentary on Kaufmann Stephan Schlothfeldt
Kaufmann’s article examines the relationship between rational egoism and political morality. Since his detailed account might not be easy to follow I will try to offer a reconstruction of what I take to be his main argument. After a critical discussion of some of the crucial steps the commentary concludes with a general remark on attempts to reconcile rational egoism and morality. As Matthias Kaufmann states at the beginning of his paper, liberal political institutions might run into the following problem: They need morally motivated citizens in order to function properly, but they seduce citizens to pursue mainly their self-interest. This is one of the reasons why critics believe that liberalism has to be backed up – or even substituted – by the moral virtues of republicanism. Kaufmann himself does not pursue this kind of solution. He rather argues that there is a sufficiently strong connection between rational egoism and a liberal political morality. In particular, Kaufmann tries to defend the following two claims: – Even for rational egoists who are not morally motivated, there are good reasons to live and to participate in a moral society (discussed in part I). – From a moral point of view, it is better to demand from others only what they would accept as rationally self-interested actors (discussed in part II). I will devote some space to Kaufmann’s quite interesting argument for the first claim since it seems to be the more provoking one. My remarks on the second claim will be somewhat shorter.
1 Would Rational Egoists Support Moral Norms? In order to establish his first claim, Kaufmann distinguishes two questions that have to be answered: – Would rational egoists prefer to live in a society which is not governed by moral norms? – Would they choose to act permanently as free riders in a society governed by moral norms? J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
113
114
S. Schlothfeldt
Kaufmann thinks that the first question should be answered in the negative, and that the response to the second question is at least not clearly affirmative. I agree that both questions have to be answered in order to establish that rational egoism would not undermine morality: It is a first and necessary step to show that rational egoists support moral norms “officially”; but this does not suffice since it might still be rational for them not to act according to those norms. I also agree that it would be sufficient to argue for both the answers Kaufmann advocates: If almost everyone in a society supports moral norms and only a small number of people act as free riders, the moral code would work well enough. However, I am not convinced by Kaufmann’s arguments as to why his answers to the questions are the right ones. Let us start with the first question: Why would it be better for rational egoists to live in a society with a moral code? Before we look at the argument, it is necessary to specify the claim: For the purposes of the article it is obviously not enough to show that rational egoists would prefer a morality – rather, it has to be the right kind of morality, a morality which supports liberal institutions (or in short: a liberal morality). As Kaufmann puts it, he is concerned with a morality of equal respect that protects the vulnerable and especially minorities. As far as I can see, Kaufmann’s argument why it is no realistic and attractive option for the rational egoist to do without a liberal morality consists essentially of the following three steps: i. Moral attitudes are good for the welfare of a group, and moreover groups with moral values are fitter for survival than those without such values. That is the reason why there is usually an established moral code within a group. ii. In general, it is advantageous for individuals as well to live in a society which is governed by moral norms. So, rational egoists do have reasons to support the moral code of their group. iii. Since there will be moral norms in any case, the rational egoist cannot avoid living within the realm of a morality; he can only choose between different moral codes. And there are good reasons for a rational egoist to prefer the liberal alternative since a liberal morality will not restrict him as much as other moral codes: if he does not harm others he will be able to act according to his convictions. How convincing is this argument? I do not have major difficulties with the first two steps – it is plausible that groups and societies need norms in order to function properly. And I am also prepared to accept that those norms will usually be of advantage for the group members, too – though group welfare and individual welfare may not always coincide, and the norms might not be advantageous for all group members alike. The main problem is obviously set up by the third step: Is it really true that a rational egoist will in general prefer a morality which protects basic rights for all? If he only has the choice between different moral codes, it might be better for the rational egoist to opt for the liberal alternative. However, whether this is the case depends on the circumstances. If he is in a position in which he has to fear that others might
A Commentary on Kaufmann
115
suppress him, he does have reasons to prefer a liberal morality. However, if he has a good chance to suppress others and might gain by a morality which discriminates in favour of a feature he possesses, it appears to be rational to choose a non-liberal alternative. A strong majority of rational egoists might be able to secure privileges at a minority’s expense.1 Hence, it is not clear that rational egoists would favour a morality which protects the rights of minorities. Kaufmann does not disprove that suppressing minorities might be a realistic and attractive alternative; he only shows that a morality will be established in each society. Which one it will be is not decided by the argument. What about the second question: Are there reasons for rational egoists to act on an established moral code? Let us suppose for the sake of the argument that rational egoists support a liberal morality. The question is still open why they should follow the moral rules when it is not in their interest – instead of acting as a “free rider”. Kaufmann’s main argument why free-riding is not an attractive and realistic option could be reconstructed as follows: i. A rational egoist cannot act as a free rider in public if he wants to be socially successful. At least, he has to pretend to be a moral person when he is watched by others. ii. The double strategy “act according to the moral rules as long as someone watches”/“break the norms when you are not observed” might lead to serious psychological harm. In my view, this argument is again not totally convincing: If someone is in a powerful position, he might very well violate moral norms in public and still be (socially) successful. Furthermore, I do not know whether the psyche is harmed by following the mentioned double standard. But even if we grant that, a rational egoist might gain so much by breaking the moral norms that it could be worth it to put up with psychological harm – just think of Patricia Highsmith’s talented Mr. Ripley.2 We should recall that Kaufmann’s claim concerning the second question is rather modest: He states that the answer to the question whether rational egoists would act as free riders on moral norms is not clearly affirmative. However, in order to show that a (liberal) morality will not be undermined, he needs the result that not too many people will act as a free rider. I have doubts whether his argument suffices to show that.
1
It seems that Kaufmann does not accept such a kind of morality as a serious option since he states that oppressing others by moral means is related to the behaviour of free-riding. This statement is rather surprising as Kaufmann has pointed out shortly before that there is no reason to deny that discriminating norms might be called moral. 2 A similar objection applies to the related argument that rational egoists want to be socially accepted (which is surprisingly placed by Kaufmann into the context of the first question): It might be true that they are interested in social acceptance, but it is not at all obvious that this interest occupies a superior place in their hierarchy of preferences. One might as well give up on social acceptance in order to gain something more important.
116
S. Schlothfeldt
2 Rational Self-interest as a Criterion for Morality? Let us have a short look at Kaufmann’s second general claim – the claim that it is better from a moral point of view to demand from others only what they would accept as rationally self-interested actors. Kaufmann’s argument for this position is essentially the following: There is always the danger that moral fanatics might suppress others by excessive demands. We want to defend ourselves against this kind of moral exploitation, and the appropriate means is to restrict moral demands to what rational egoism recommends. I basically agree with this point if it is spelled out properly, but we have to avoid a certain ambiguity: A strong interpretation of the claim would be that we can only demand from others what is to their own good anyway. A much weaker interpretation states that the actors should choose moral principles according to their self-interest when it is already assured that they are in a symmetrical position. The second interpretation appears to be a convincing idea if we want to prevent that moral claims interfere with the strictly private decisions of individuals – our choice of moral principles should only be concerned with our own life since we would not accept that our private affairs are dominated by others’ preferences. However, this position does not have much to do with rational egoism in the true sense of the word which was meant in the former part of the article: Choosing according to rational self-interest is only a technical device here; the actors are already placed into a setting which guarantees that they are accepting moral demands of fairness – by being forced into a reversal of roles or, as in Rawls’ case, by being put under a veil of ignorance which covers information they might use to their own advantage. On the other hand, the stronger interpretation – that we should only demand from others what is to their own good anyway – corresponds to rational egoism in the sense of the first claim of Kaufmann’s article. However, this position is much less convincing since it might lead to the result that we could demand almost nothing from strong and powerful individuals – and it is clearly not shared by the liberal political philosophers (Rousseau, Kant, Rawls) Kaufmann mentions. In my view, the moral progress obtained by liberalism is connected to the weaker claim that we should not interfere with the private sphere of other people but concentrate moral demands on what is important for our own life. This has not much to do with rational egoism – though it can be illustrated by a technical procedure which makes use of rational self-interest within a certain setting. Though some details of Kaufmann’s argument in favour of a relatively strong connection between morality and rational egoism are original, the shortcomings could count as typical for related attempts in the philosophical literature: First of all, a couple of moral and political philosophers do not realize that a morality which is concerned with self-centred interests has to be distinguished from a “morality” which is based solely on self-interest. Hence, quite different moral and political approaches are often labelled as “contract theories”. Secondly, if “contractarians” try to spell out the moral implications of rational egoism they do not realize sufficiently that it depends on contextual details whether the result will be desirable. If power is equally distributed, the risk to be harmed
A Commentary on Kaufmann
117
is the same for all, and the hierarchy of preferences is more or less similar, moral norms which conform to liberalism might be the preferred option. However, those conditions are very special. It is at least necessary to explain in detail the necessary and sufficient assumptions for the desired outcome. But if we would specify those conditions, the following problem could arise: Reconciling liberal morality with self-interest might presuppose a well-ordered society – and especially a social setting where moral norms are already established. If this turned out to be true, the defence of rational egoism would free-ride on morality.
Political Contractarianism and Equally Distributed Basic Rights ¨ J¨org Kuhnelt
1 Pluralistic Societies and Political Contractarianism The large diversity of values and interests, which are of importance for individuals living within pluralistic societies, prompts the question of how to design a convincing argument to legitimize a state. It appears to be important to find a compelling answer to this question because existing legal states are, in general, equipped with enough coercive power to ensure that all individuals normally obey the law, regardless of whether or not they consider the state to be legitimate in relation to their interests and values. At least in pluralistic societies, different reasons exist to criticize the legitimacy of a state: Some individuals, typically described as rational egoists, deem values to be irrelevant or unsuitable for the legitimization of a state. However, even if the remaining individuals do not focus solely on their egoistic interests and accept the relevance of values in principle, they do not necessarily consider similar values to be appropriate for the legitimization of a state. In this context, I will briefly discuss some problems of normative accounts of state legitimacy and analyse whether rational accounts can be more attractive for pluralistic societies. Normative accounts of legitimacy, even political ones, are based on specific values. The tasks and institutions of a state are more or less directly derived from these normative premises. Problems arise, because normative theories can be founded on different values, for instance, on an idea of equality, on a conception of natural rights or on a utilitarian principle.1 Even if normative theories provide a justification for why their underlying values are considered to override egoistic interests
1
These examples show that I am using a broad definition of the term value. This definition shows that normative accounts of legitimacy can be criticised by individuals who support different values regardless of whether the legitimacy of a state is derived, for instance, from a normative principle or a conception of natural rights. In the introduction of this volume I also sketched John Rawls’ conception of a reasonable overlapping consensus between different normative reasonable comprehensive doctrines, firstly fully developed in: Rawls (1993): Political Liberalism. However, his conception of an overlapping consensus is also based on a “value” defining fair cooperation. Thus, individuals in pluralistic societies only necessarily agree to this consensus when they already support the Rawlsian idea of fair cooperation, meaning they support only reasonable doctrines.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
119
120
J. K¨uhnelt
or different normative claims, at least in relation to pluralistic societies two serious problems arise: Firstly, rational egoists are systematically excluded from normative arguments: If, for whatever reasons, individuals consider normative arguments to be irrelevant, they are also not convinced of a normative argument which is supposed to justify overriding egoistic interests. Secondly, all remaining individuals only have an incentive to agree to a normative legitimization of a state if differences or possible conflicts between the normative foundation of the legitimization and their personal values are not too grave. Different normative positions on the content and importance of distributive justice in a legal state are good examples of conflicts between values, which can arise even if individuals share certain core values and agree, for instance, on the legitimacy of a legal prohibition to kill, hurt or threaten others. As already mentioned in the introduction of this volume, it is surely not easy for a Libertarian to accept an Egalitarian conception of legitimacy and vice versa as both hold incompatible conceptions of distributive justice.2 For an illiberal normative theory it is probably not a substantial problem that rational egoists and individuals holding conflicting values will not agree with the underlying normative premises. However, at least for liberal theories a problematic trade-off exists: On one hand, liberal theories are usually based on an idea of individual freedom and claim that individuals should keep the authority to decide how they want to live their lives. Thus, legal constraints are to be avoided, where possible, in order to ensure this individual freedom. On the other hand, if a state in a pluralistic society is legitimized with a certain liberal theory, not only rational egoists but also individuals holding different conceptions of freedom have reasons to disagree with this legitimization, for instance, because their idea of freedom is underlined by a different conception of distributive justice. Nevertheless, legal sanctions can still be efficient in providing an incentive for most individuals to obey the law. In consequence, not only individuals being systematically excluded from the legitimization of this state, but also individuals holding conflicting values have good reason to complain that the problems arising from pluralistic values are not solved by justification but by coercion. As long as those individuals are not better integrated within the legitimization of the state, they can argue that the state constrains their individual freedom illegitimately because its legitimization is incompatible with their personal interests or values, for instance with their idea of freedom. If the description of pluralistic societies above is correct and some individuals are egoists while others hold quite different values, every normative legitimization will be refused by a considerable number of citizens and it becomes difficult or impossible to decide which normative legitimization is best for a pluralistic society. In relation to these problems, I believe rational accounts of legitimacy to be more attractive for pluralistic societies because these accounts try to avoid contentious normative premises and refer primarily to the assumptions of rationality and basic interests. Within the contractarian tradition, which is firstly fully developed in
2
Robert Nozick’s Anarchy, State, and Utopia (1974) can be seen as an example for a Libertarian theory and Ronald Dworkin’s Sovereign Virtue (2002) as an example for an Egalitarian position.
Political Contractarianism and Equally Distributed Basic Rights
121
Thomas Hobbes’ Leviathan, a state is considered to be legitimate if it can be shown that it is advantageous for all citizens.3 This legitimization is supposed to be convincing even for rational egoists because contractarians argue that rational egoists can fulfill their basic interests better within a legal state than in a state of nature without any legal or moral regulations. At least implicitly, contractarians also assume that even individuals who hold certain values and thus consider themselves not to be egoists still share the need to fulfill certain basic interests. In consequence, even the remaining individuals who hold different and possibly conflicting normative perspectives are supposed to be convinced of the contractarian argument.4 On a general level, the argument can be described as follows: Firstly, contractarians illustrate why rational individuals have to cope with substantial problems of interaction in the state of nature, causing even difficulties for the satisfaction of basic interests. Secondly, contractarians argue that only a legal state is capable of changing the framework of interaction in a way to solve the problems of the state of nature and is therefore considered to be beneficial for everybody. However, both parts of the argument must be discussed in further detail to analyse the potential of contractarian theories for pluralistic societies. In the state of nature, no legal regulations exist to constrain the actions of individuals: Each individual is on his own when it comes to fulfilling his aims and protecting his person and his goods. Furthermore, contractarians assume that the scarcity of goods in the state of nature hinders individuals ability to satisfy not only their further aims but also their basic interests. This uncertainty regarding the fulfilment of basic interests leads to a central conclusion of contractarian theory: Within the state of nature, almost everybody is meant to behave as an egoist, as contractarians assume that not only egoists but also individuals with altruistic interests firstly try to secure the goods necessary to their survival.5 As long as no legal regulations exist, it is supposed to be rational for everybody to use force and fraud if these are helpful means to satisfy their basic interests. However, if a sufficient number of individuals actually kill, steal goods or threaten others, the problem of insecurity will worsen and all individuals have to fear becoming victims of force and fraud. Moreover, this insecurity leads to the problem that the interaction in the state of nature is generally inefficient because it is usually irrational for individuals to cooperate with others as long as trade or other forms of mutual beneficial cooperation are easily exploitable.
3
Hobbes, Thomas (1991 [1651]): Leviathan.
4
Later I will argue that the normative views of the individuals living in pluralistic societies should not be completely ignored. Still, I think that conflicts between different normative views can be handled better within the contractarian argument than within normative arguments. 5 A different reading of the Contractarian argument is possible: One could argue that all individuals are predominantly egoists and Contractarians show why it is rational for egoists to establish a state or to act morally. This is not only a more difficult argument, as altruism is supposed to be explained somehow in relation egoism; it is also based on the assumption that individuals are generally egoists. As I do not believe that all individuals are normally egoists, I propose a different reading that is, for instance, compatible with James Buchanan’s description.
122
J. K¨uhnelt
This context helps to understand Hobbes famous description of the life in the state of nature as “solitary, poor, nasty, brutish and short”.6 A modern interpretation of his argument usually refers to systematic problems of strategic behaviour planned on the individual level:7 The well known prisonersdilemma illustrates why, under certain conditions, rational individuals act in a way which is rational in relation to their personal advantage but nevertheless leads to a sub-optimal outcome for everybody. As long as these conditions persist, rational individuals will not change their behaviour, even though the situation could be improved for everybody if all individuals would change their behaviour: Within the state of nature it is not rational for individuals to stop their aggressive behaviour or to cooperate because the risk of exploitation is too high, as they cannot be sure that the other individuals hold their promises. The analysis of this dilemma supports the Hobbesian argument and shows not only why it is considered to be rational for each individual to use force and fraud to secure his life and goods; it becomes also understandable why Hobbes assumes that it is rational for every individual to use force and fraud to increase his power, because this can reduce the risk of becoming a victim. However, this behaviour enlarges the core problems of the state of nature – uncertainty and inefficiency – and the individuals remain in a sub-optimal situation: Even powerful individuals have to cope with a certain level of insecurity, especially on a long-term perspective. Furthermore, because of the risks concerning cooperative behaviour, individuals will barely realize any gains of cooperation. As all individuals more or less have to cope with the problem that they might be attacked or that their cooperative behaviour is exploited by others, the situation in the state of nature is considered to be neither secure nor efficient. The contractarian argument is based on the idea that all individuals would be better off in cooperating on a general level to change of the framework of interaction. Hobbes argues that rational individuals are capable of realizing that everybody’s situation can be improved by establishing a legal state to set up laws to stop force and fraud.8 This change of the framework of interaction is considered to be beneficial for everybody because everybody’s life is more secure and efficient within a stable 6
Hobbes (1991 [1651]): Leviathan, p. 89.
7
The prisoners-dilemma is discussed in various contexts. A good overview on some general ideas and on the background of this discussion can be found, for instance, in Yanis Varoufakis and Shaun Heap (2004): Game Theory – A Critical Text. In this volume Reinhart Zintl also discusses some central aspects of the prisoners-dilemma. However, not everybody agrees on the direct analogy between Hobbes’ argument and the prisoners-dilemma. One example is Allan Ryan’s paper: Ryan (1996): “Hobbes’s Political Philosophy”. A good general analysis of Hobbes theory can be found, for instance, in Bernd Ludwig (1998): Die Wiederentdeckung des Epikureischen Naturrechts. 8 In this context it is important to point out a possible distinction between two forms of cooperation. Firstly, cooperation is possible on an individual level: Prominent examples are trade of goods or services, coordination of the use of resources or mutual help to fulfill aims that are not achievable for single individuals. However, the establishment of a state is based on a more general cooperation that involves all individuals in the relevant territory. Nevertheless, I do not discuss whether such cooperation is possible in the state of nature, in other words, whether the individuals in the state of nature would be actually able to establish a state. I focus only on the question which legal state is
Political Contractarianism and Equally Distributed Basic Rights
123
and peaceful coexistence. Living in a legal state, each individual no longer lives in fear or uncertainty and furthermore saves energy and expenses otherwise spent in protecting his life and goods.9 However, it is only rational for everybody to change their behaviour if a legal state protects its laws by police forces and punishes illegal behaviour with sanctions. Otherwise, a reduction in the protection of life and goods would, at least in principle, only increase the potential for exploitation. Only within a certain framework of interaction provided by a legal state, it is finally irrational for individuals to behave aggressively; at least as long as they realistically have to fear punishment.10 Obviously, establishing a state creates some costs, because everybody’s life is constrained by certain legal regulations and each citizen has to pay taxes. Still, contractarians argue that changing the conditions of interaction – from the state of nature to a legal state – is rational for everybody, because every citizen realizes more gains from security and new possibilities to cooperate than losses from legal constraints and payment of taxes.
2 Three Problems of the Contractarian Argument To analyse the potential of contractarian theories for pluralistic societies, it is helpful to discus three general problems: Firstly, even contractarian theories are usually based on the following normative assumption: A state is considered to be legitimate if it enables an improvement for everybody in relation to the problems in the state of nature.11 With reference to these advantages, contractarians argue that even rational egoists could agree to a state and its legal regulations. I will not discuss whether this normative assumption is really less contentious than possible alternatives.12
more advantageous for all in relation to the state of nature and can therefore be considered to be legitimate. 9 Of course everybody would have to give up murder and theft, but contractarians assume that individuals have more benefits from security than from the possibility to steal or kill. I believe this assumption to be convincing in general, though it is necessary to mention that it is based on a certain hierarchy of interests. For instance, a sadist would ascribe his interests different priorities. 10
However, as soon as the risk to be punished is sufficiently low, at least egoistic benefit maximizers would analyse whether illegal behaviour is more advantageous than legal behaviour. This problem is discussed in the context of the freerider problem. For example, Russell Hardin (2003) provides a good introduction to the freerider problem in the Stanford Encyclopedia of Philosophy.
11
This normative assumption can be specified in different ways. For instance, if rationality is understood as benefit maximizing, there is a difference between an outcome which can be described as rational for an individual and an outcome which is only an improvement. I do not discuss different conceptions of rationality in detail, but I think improvement is a more general specification and can be understood as a minimal condition of the contractarian argument.
12
One could argue that this normative assumption is also based on a certain idea of justice: Only legal states which improve the situation for everybody in relation to the state of nature can be considered to be just. This also illustrates the problem, that purely egoistic rational individuals would not accept any idea of legitimacy. They would be focused only on the status quo and try to improve their situation as much as possible, even if this is disadvantageous for others. Consequently, they
124
J. K¨uhnelt
Still, I would like to outline at least two consequences: On one hand, a normative assumption of that type is not necessarily convincing for rational egoists because reforms can be even legitimized if they are only beneficial for all individuals in relation to the state of nature and not in relation to the status quo. For instance, it is possible that egoistic individuals in the status quo are powerful enough to maintain certain privileges which cannot be explained with reference to the state of nature. However, it is only rational for egoists to accept these reforms if they also accept the normative premise that a state is only legitimate if it improves the situation for everybody in relation to the state of nature. Otherwise, they would have no incentive to give up their privileges as they would not consider them to be illegitimate. On the other hand, improvement in relation to the state of nature is defined primarily with reference to basic interests, because contractarians assume that only such premises convince all individuals including rational egoists. Still, it is at least questionable whether altruists and egoists always consider the same legal state to be an improvement in relation to the state of nature.13 Nevertheless, for the following discussion I assume that egoistic as well as altruistic individuals in pluralistic societies generally accept the normative premise that a state is legitimate, or at least acceptable if it is beneficial for everybody in relation to the state of nature. Secondly, difficulties arise because numerous different legal states can fulfill the mentioned normative condition. Hobbes may have had good reason to assume that the individuals in the state of nature have more or less equal power:14 If the others have equal strength and each individual has to cooperate to improve his situation, everybody is motivated to agree to reciprocal constraints of behaviour. In a simplified situation with two equally powerful individuals living on an island, each individual stops his aggressive behaviour only if he can be sure that the other does the same. Under these conditions it is rational for both to cooperate and agree on a legal or minimal state which guarantees everybody the same set of basic rights and establishes, for instance, a general prohibition to kill and to steal.15 However, if individuals are not equally powerful, even unequal cooperation can be beneficial for both: If there is strong inequality, the weaker individual might still benefit from enslavement because he no longer has to fear being murdered in the state of nature. Thus, even a slavery state can be a result of the contractarian argument even though
have no incentive to give up any privileges they are able maintain. For the further discussion I will not suppose that all individuals are “purely egoistic rational actors”. The question of how legitimacy can be ascribed to an analysis that shows the advantages of a minimal state in relation to a state of nature is discussed in more depth in my thesis Pluralismus und Vertragstheorien. 13
This appears to be no problem for the legitimization of the prohibition to kill or to steal. However, the situation becomes more difficult as soon as the necessity and the aims of further governmental tasks are discussed. I will come back to that point at the end of this paper.
14
Hobbes (1991 [1651]): Leviathan, pp. 86–87. Nevertheless, within this paper I primarily focus on the structure of the contractarian argument and not on a detailed interpretation of Hobbes’ Leviathan.
15
For the further discussion I define basic rights as the right not to be hurt, threatened or killed by others (individual rights) and the right to own property (property rights).
Political Contractarianism and Equally Distributed Basic Rights
125
such a state is commonly considered to be illegitimate from the external perspective of normative theories. In general, it becomes questionable if the contractarian argument is specific enough because not only slavery states, but also egalitarian or libertarian states can fulfill the condition to be advantageous for everybody in relation to different but possible distributions of power and interests in the state of nature. Thirdly, contractarians usually focus on the idea that a convincing legitimization has to be even valid for a society of rational egoistic benefit maximizers. However, it is questionable whether the assumption of benefit maximization is realistic and, as stated above, not everybody in a pluralistic society can be described as a rational egoist. Furthermore, in a society of rational egoists the freerider problem leads to the question of whether a state can really become sufficiently stable: Rational benefit maximizers always calculate whether the risk to be convicted multiplied with the disadvantages of the sanctions is smaller than the advantages of illegal behaviour. It is not rational for benefit maximizers to obey a law that is in principle advantageous, if illegal behaviour is realistically more beneficial. Moreover, rational egoists do not care whether illegal behaviour leads to disadvantages for others. Of course, it is possible to try to ensure stability by establishing a police state with high sanctions for illegal behaviour. Still, it is questionable as to whether or not a police state really solves the freerider problem.16 In general, I think these three problems show a systematic difficulty: Current contractarian approaches are usually based on premises that are not specific enough to derive convincing results. Under certain circumstances it appears to be realistic that rational individuals would agree to a minimal state that is sufficiently stable, and that ascribes everybody the same set of basic rights.17 Nevertheless, in relation to the common contractarian premises discussed above, this is not a necessary result. Furthermore, the tasks of a legal state are limited because the contractarian argument is focused only on systematic problems being dominant in the state of nature. Two options exist to specify the assumptions of the contractarian argument: On one hand, it is possible to work with more substantial normative premises and restrict the content of the social contract to ensure sufficient reciprocity. To exclude the possibility of a slavery state, it would be necessary to define reciprocity as a demand for a minimal gain for everybody. Additionally, these more substantial normative premises could be used to derive further governmental tasks or to ascribe every citizen an obligation to obey the law to deal with the freerider problem. However, in this
16
Additionally, rational egoists could be deterred from too much surveillance or high sanctions, as they realize the risk that a necessarily powerful police state could be abused. Still one could argue that even a group of rational egoists has a strong incentive to solve at least the problems of stability, because they have, from a general point of view, a strong interest in a sufficiently stable cooperation within a legal state to avoid the problems of the state of nature. However, I do not think that an argument which is capable of convincing a group of rational egoists is also necessarily convincing for all individuals in pluralistic societies.
17
A minimal state is a legal state which guarantees every citizen stable basic rights. A slavery state is an example for a “legal” state which does not guarantee everybody basic rights.
126
J. K¨uhnelt
case contractarian theories are hardly distinguishable from contractualist theories or other normative accounts of state legitimacy.18 Consequently, this option is not really attractive for pluralistic societies. On the other hand it is possible to specify a relativistic contractarian argument, which can rely on more detailed descriptive premises. Within a relativistic contractarian argument it is neither necessary to assume that all individuals really try to maximize their benefits or that everybody is a rational egoist. The idea is to analyse which problems arise if the individuals of a pluralistic society would interact in the state of nature and which tasks of a legal state would be an improvement for those individuals. In this context James Buchanan’s theory provides some interesting changes to the classical contractarian argument.
3 James Buchanan’s Contractarian Theory In The Limits of Liberty, James Buchanan enhances contractarian theory into a twostep argument to avoid the difficulty that the tasks of a legal state are focused only on the systematic problems of interaction within the state of nature. Buchanan argues that certain problems of coordination and cooperation exist even in a minimal state, for instance, the lack of an incentive for individuals to produce public goods. His enhanced theory has the advantage to show which basic and further tasks should be legitimized for a legal state by referring to the problems of the state of nature in the first step and to the problems in a minimal state in the second step. However, the question of whether or not further governmental tasks are beneficial for everybody arises only for individuals already living in a minimal state. Within a slavery state, further cooperation of that kind is not relevant for the egoistic rulers because they are only concerned with the improvement of the situation for everybody if this is a necessary condition to improve the fulfilment of their personal aims. However, as long as a slavery state is a possible result, it is questionable if contractarian theories are really more convincing for pluralistic societies than normative arguments. For this reason, I will discuss the first step of Buchanan’s argument in detail to analyse how he tries to show that a minimal state is a realistic result for contractarian theories.19 Buchanan avoids contentious assumptions, for instance, equal strength or egoism of all interacting individuals. He analyses a state of nature in which rational
18
At this point the distinction between contractarian and contractualist theories, which is also discussed in the introduction of this volume, becomes relatively clear. The former theories try to avoid any contentious normative premises whereas the latter theories frame the possible outcomes for the social contract with normative assumptions. Contractarians are often described as members of a Hobbesian tradition and Contractualists usually argue within the Kantian tradition. For instance, Robert Sugden gives a good overview on the distinction of Hobbesian and Kantian contract theory. Sugden (1993): “The Contractarian Enterprise”.
19
For the following description of Buchannan’s arguments I refer primarily to the chapter 4 “Constitutional Contract: The Theory of Law” in James Buchanan (1999 [1975]): The Limits of Liberty.
Political Contractarianism and Equally Distributed Basic Rights
127
individuals act without any legal regulations but do not necessarily have equal power or similar interests. Buchanan explicitly takes into consideration that individuals are not necessarily egoists. Nevertheless, in this state of nature powerful egoists will certainly secure more advantages than weaker individuals. Furthermore, as long as everybody’s survival is in direct danger, most altruists behave probably quite similar to egoists. However, altruistic individuals do not behave exactly like egoists because they respect the interests and aims of other individuals to a certain extent: In the state of nature altruists will predominantly hesitate to kill, threaten or steal if this is avoidable.20 As soon as their basic interests are sufficiently satisfied, they probably even actively help weaker individuals. It is a general advantage of Buchanan’s argument that he does not need to presuppose egoistic behaviour: Regardless of how everybody acts in detail, Buchanan shows that all individuals will reach a state of relative equilibrium, in which each individual produces, protects and steals efficiently in relation to his current power and interests.21 Equilibrium will always be reached, no matter which distribution of interests or power exists in the state of nature. Within this equilibrium, every individual knows which personal freedom of action and which goods he can demand or defend in relation to his power and interests. As individuals cannot improve their situation any more without cooperation on a general level, everybody will agree to a legal state as soon as this state is capable of improving the situation for everybody. Nevertheless, it is important to analyse possible situations within the equilibrium in detail, because different legal states can be advantageous depending on different distributions of interests and power. It is possible that weak individuals cannot even defend a basic freedom of action if a strong inequality of power exists and a sufficient number of the stronger individuals are egoists. No matter how rationality is defined in detail, if the strong individuals are sufficiently egoistic, they have no incentive to establish a minimal state as long as being a master in a slavery state is more advantageous. Furthermore, because of this strong inequality within the equilibrium, even enslavement will be an improvement for the weaker individuals. Therefore, at least under certain conditions, a slavery state needs to be considered as a legal state, because it can fulfill the condition to be advantageous for everybody in relation to the state of nature. Buchanan is aware of that problematic result but he
20
I work with a quite simple distinction of altruism and egoism: The satisfaction of egoistic interests leads primarily to an improvement of the situation of the actor. The fulfilment of altruistic interests is, of course, stronger related to other individuals. I assume that altruistic interests are somehow derived from values, which of course vary within pluralistic societies. Therefore, individuals will not only act differently to egoists. Their actions also vary depending on how much importance each individual ascribes these values or which values they consider to be relevant for their altruistic interests.
21
Ibid., pp. 34–35. Although Buchanan mentions the relevance of interests at this point, he primarily focuses on the power of individuals in his further analysis. However, I believe the interests of individuals equally important. I will come back to that point later.
128
J. K¨uhnelt
assumes that there is usually not enough inequality in the state of nature.22 A different situation within the equilibrium is realistic if the inequality of power is smaller, or if a sufficient number of powerful individuals also have altruistic interests. In this case it is likely that all individuals secure certain basic abilities to act and only a minimal state improves the situation for everybody. Hence, only in certain situations all individuals have an incentive to establish a minimal state with equally distributed basic rights.23 Buchanan’s argument is interesting, because he can show the advantages of a legal state no matter which freedom of action and which goods individuals can secure in the prevailing equilibrium in relation to their interests and power. As soon as equilibrium is reached, all individuals, even strong ones, have no possibility to enlarge their benefits any further as long as they do not cooperate. For example, if an individual would steal more, he could not protect his goods as well, leading to others stealing more from him than he would gain by stealing more from others. Therefore, every individual, no matter if weak or strong, necessarily has costs for the protection of his goods and his personal freedom of action. Consequently, it is advantageous and thus rational for everybody to cooperate and to invent certain rights that are secured by a state to reduce his costs for protection. Buchanan calls this a protective state, but I think state of protected relative equilibrium is more precise, because all relevant relations between individuals that occurred in the state of relative equilibrium are now protected by legal rights.24 These rights protected by a state can be divided into individual rights, defining the personal freedom of action and property rights, defining the goods owned. Nevertheless, Buchanan’s state of protected equilibrium is funded only to guard these rights, because the distribution of rights is the outcome of the distribution of abilities and power in the state of relative equilibrium. Therefore, the state of protected equilibrium is not legitimized to change the distribution of legal rights.25 These rights can change only in two
22
Ibid., p. 34. Nevertheless, there is a certain contradiction in his book as he also assumes the possibility that strong individuals kill weaker individuals before they start negotiating in the state of nature (pp. 77–78).
23
As mentioned above, the term basic rights is defined to include individual rights and property rights. The term legal rights just illustrates that these rights are protected by a legal state. However, even in a slavery state, a slave could have legal rights but not all basic rights. The aim of this distinction is to show the difference between a minimal state in which all individuals have basic rights, and for instance a slavery state in which slaves just have certain legal rights.
24
Buchanan does not use this terminology: He just distinguishes state of nature, natural distribution or equilibrium, and protective state. As power and interests of individuals can change over time, I think it is clearer to distinguish state of nature, state of relative equilibrium, and state of protected relative equilibrium (short: state of protected equilibrium). Furthermore, Buchanan does not distinguish between individual rights and property rights. His reasons can be found, for instance, in: Buchanan (1999 [1975]): The Limits of Liberty, p. 14f. For his argument this distinction is not really necessary, but it turns out be relevant for my criticism. 25
This shows that any kind of normative argument for a redistribution of rights to improve the situation of the slaves cannot be considered to be legitimate within Buchanan’s argument.
Political Contractarianism and Equally Distributed Basic Rights
129
ways: Either individuals trade them, or the distribution of interests or power in the state of protected equilibrium changes significantly. Without discussing Buchanan’s theory in depth, I focus on two problematic aspects. Within his argument, the distribution of rights is derived and therefore linked to the distribution of power and interests. This leads to the first consequence that a significant change in these parameters is supposed to influence the distribution of legal rights. Individuals who increase their relative power want to have more rights, regardless of if others loose these rights. Buchanan argues that the potential position in the state of relative equilibrium lies somehow “underneath” the status quo and creates a certain incentive to reshape the distribution of rights in the status quo, as soon as the distribution of power and interests changes. Consequently, he assumes it to be rational for individuals in the status quo to offer a reduction of their rights, as soon as they have significantly more rights in the status quo than achievable in the state of relative equilibrium.26 Thus, Buchanan at least implicitly assumes that, at a certain point, individuals who gained power can decide to quit the state of protected equilibrium and renegotiate in a situation similar to the state of nature. In both cases, citizens have to fear to lose their property or, at worst, to become slaves.27 However, it is not clear if individuals in a state of protected equilibrium have a “right” to renegotiate. Therefore, it might be difficult to say under which condition a renegotiation is realistic. Nevertheless, after a significant shift in the distribution of power, for instance, the possibility of a revolution arises leading probably to a different distribution of individual rights. Terrorism could also be a means to force other individuals to renegotiate the actual distribution of rights. Even smaller shifts in power might be relevant, for example, if less wealthy individuals are powerful enough to demand a redistribution of goods from richer ones. I do not think it is necessary to analyse these possible renegotiations in detail, to show the second consequence: As long as a renegotiation of legal rights is generally possible, the crucial question arises of whether or not Buchanan’s state of protected equilibrium can really be considered to be sufficiently stable. The origin of the problem is Buchanan’s idea to derive the distribution of rights more or less directly from the distribution of power and interests: If a slavery state is realistic in relation to certain distributions of interests and power in the state of nature, it can be also a possible result after a renegotiation of the distribution of rights in a legal state, because in both situations the distribution of rights is not restricted by any moral or legal constraints. Consequently, no state of protected equilibrium can be considered to be sufficiently stable, because even a minimal state can, at worst, change into a slavery state after a significant change in the distribution of power. This shows the crucial point in Buchanan’s argumentation: He does not provide an argument
26 27
Ibid., ch. 5, esp.: p. 103, 105 and 109f.
One could also imagine a state in which some individuals have no right to own property. These persons are not owned by masters, but they would be strongly dependent on others. However, I think the general problems become clear, if only minimal state and slavery state are discussed.
130
J. K¨uhnelt
to establish stable basic rights which are ascribed to all individuals no matter how much (or less) power they have. Therefore, even within a minimal state that guarantees everybody basic rights, citizens still have to fear that the distribution of power changes significantly and that they lose some of their basic rights. It is not only possible to become a slave but also to lose property or the right to own property, depending on how much power each individual loses or others gain. Hence, I think that there is not enough difference between Buchanan’s state of protected equilibrium and the state of nature, which is ruled by force and fraud, and therefore a state of protected relative equilibrium does not really solve the problem of uncertainty: The drastic fear of death is only replaced by a more subtle fear of slavery or other disadvantageous future developments. In consequence, the state of protected equilibrium cannot solve the problem of efficiency either. As long as individuals have to cope with too much uncertainty, they do not know whether or not they can keep the gains of their investments. Therefore, it will mostly be irrational for them to invest on a medium or long-term perspective. Within these conditions, individuals will, for instance, hesitate to invest time and effort in a specialisation of their work force, leading to an inefficient division of labour. For the same reason they will also hesitate to invest in production facilities or infrastructure. In relation to this result, I will examine if no other arguments can be found to show the advantages of a minimal state which ascribes everybody stable, meaning nonnegotiable basic rights.
4 Can Basic Rights solve the Problem of Uncertainty? The contractarian tradition is based on the premise that individuals have an interest in a legal state which provides security, improves the efficiency of interactions or trade and consequently enables them to fulfill their interests better than in a state of nature. In my criticism I tried to show that a minimal state is not a necessary result in relation to the usual premises of the contractarian argument: Rational egoists, as long as they are strong enough, will consider it to be rational to establish a slavery state or other forms of oppressive states, if this improves their situation even more.28 Furthermore, benefit maximizers will try to freeride as often as possible, leading to problems for the stability of legal states. However, I believe that contractarians primarily refer to these premises because they want to provide an argument with universal validity. They usually refer to basic egoistic interests because they assume this to be an uncontroversial premise as everybody has such interests. If it could be shown that a state is beneficial even with reference to egoistic interests, everybody could accept that argument. If
28
I will use the slavery state as a paradigmatic example of an oppressive state that does not guarantee every citizen basic rights. Of course, different forms of oppressive states are possible in relation to different distributions of power and interests. However, I will not refer to other oppressive states, as I think the problem becomes clear by comparing a slavery state with a minimal state.
Political Contractarianism and Equally Distributed Basic Rights
131
contractarians would assume substantial altruistic interests within their premises, they would either convince only individuals with similar interests or they would have to claim normatively that everybody should have these altruistic interests, but would be unable to justify this claim within their theory. In this context, Buchanan’s theory is interesting because he does not explicitly assume rational behaviour to be benefit maximizing, although he does not discuss possible alternatives in detail. Furthermore, he tries to avoid the general assumption that all individuals are egoists, although he does not explicitly give up the universal validity of his theory.29 In the last part of this paper, I will argue that these problems are not necessary relevant within pluralistic societies: The contractarian argument refers to the interests of individuals and therefore it is important to analyse which interests individuals really have and how they try to satisfy them. At least in relation to pluralistic societies, I believe it to be neither necessary to assume that all individuals are egoists, nor that they always maximize or try to maximize their benefits from actions. Nevertheless, the attempt to specify the premises of the contractarian argument in relation to pluralistic societies has the important consequence that the result no longer has universalistic validity. Instead of a universalistic contractarian theory I will argue for a relativistic argument, but I try to specify the premises in a way that is convincing for a sufficiently high number of pluralistic societies. Furthermore, a relativistic argument has the advantage to be based on a more definite set of premises leading to a more convincing conclusion. Within this relativistic framework, I will argue that a minimal state can be considered to be sufficiently advantageous for everybody, even for egoists. For this argument, I will specify three contractarian premises in further depth: Firstly, I suggest an alternative model of individual behaviour. Secondly, I propose a certain setup defining the distribution of individual interests within pluralistic societies. Thirdly, I argue that contractarians need to refer more explicitly to economic results, especially the relevance of stability for long-term investments. The assumption of benefit maximizing within the classical model of homo oeconomicus is still a common fundament for contractarian arguments, although it has been criticized by many scholars.30 In his paper “The Virtual Reality of Homo Oeconomicus”, Philip Pettit suggests to describe individuals not as benefit maximizers
29
Ibid., p. 4 and p. 71. This might be the reason why he does not say more about the distribution of interests within the state of nature. Although, one could argue that he constrains the validity of his theory when he assumes that an inequality in the state of nature leading to a slavery state is possible but unrealistic. However, as he does not advocate a relativistic theory, he does not analyse any consequences for such an argument.
30
Amartya Sen is a prominent example for such criticism: Sen (1977): “Rational Fools – A Critique of the Behavioural Foundations of Economic Theory”. A current reference to this paper is, for instance, Hartmut Kliemt and Werner G¨uth (2008): “The Rationality of Rational Fools – The Role of Commitments, Persons and Agents in Rational Choice Modeling”. Michael Baurmann (2002): The Market of Virtue can be mentioned in this context as well as Pettit (2002 [1995]) “The Virtual Reality of Homo Oeconomicus”. Mark Granovetter argues that rational behavior is more strongly influenced by social relations than, for instance, Hobbes assumed. Granovetter (1985): “Economic Action and Social Structure – The Problem of Embeddedness”. However, this is just a narrow selection of texts within this discussion.
132
J. K¨uhnelt
but as virtually self-regarding.31 I will sketch Pettit’s suggestion briefly to show its advantages: Firstly, self-regarding individuals have a broad spectrum of interests from egoistic interests to stronger or weaker altruistic interests. Still, Pettit assumes that individuals usually have a higher motivation to fulfil their egoistic interests than to fulfil their altruistic interests. Secondly, virtually self-regarding individuals usually act in accordance with social rules. Therefore, they do not try to maximize their benefits in every act. However, if individuals recognize that acting in accordance with a common pattern would reduce their advantages significantly, certain alarm bells ring and motivate them to consider their interests. For example, if an individual realizes that he is frequently exploited by one of his friends, he will think about this friendship and might behave differently to that person to secure his individual advantage. Pettit’s suggestion is interesting because he tries to integrate the economic view that individuals act in accordance with their personal advantage into the sociological view that individuals usually respect cultural or social norms. I believe the model of virtually self-regarding individuals to be fitting for contractarian theories.32 Self-regarding individuals not only recognize the advantages of a legal state in relation to a state of nature, they are also capable of modifying their patterns of behaviour and to obey the law if that is generally beneficial for their interests. Within Pettit’s model it is assumed that individuals have the ability to analyse the consequences of their behaviour rationally without necessarily trying to maximize their benefit in every act. If he is right, freeriding is not a central problem for contractarian theory, because individuals would only act illegally if they would expect a quite substantial advantage. Nevertheless, I will propose two modifications of Pettit’s model. Firstly, I assume that all individuals are virtually self-regarding and therefore have a higher motivation to secure, at least certain, important egoistic interests which I dub primary interests. For example, Hobbes shows that all individuals have the primary interest to avoid the fear of death in the state of nature. As long as they have this fear, they will not be concerned with any other interests including further egoistic or altruistic interests. Therefore, all or at least most individuals behave as egoists in the state of nature because they want to fulfil their primary interests first. Still, individuals with altruistic interests consider the interests of others as 31 32
Pettit (2002 [1995]): “The Virtual Reality of Homo Oeconomicus”
However, the reference to Pettit is not necessary for the proposed argument. The idea is to use a model that does not assume egoism and benefit maximization and is therefore closer to the reality of individual behaviour. In this context, for instance, the works of Reinhart Selten or Herbert Simon are surely interesting as well. Simon’s conception of satisficing is, for instance, explained in: Simon (1997): Models of Bounded Rationality or Simon (1958): Organizations (together with James March), ch. 6. Selten’s evaluation of rational choice theory is nicely outlined in: Selten (1991): “Evolution, Learning, and Economic Behavior”. Both scholars and many others are searching for a conception of rationality that is closer to actual individual behaviour. An interesting overview of different suggestions presented at a conference on bounded rationality can also be found in Wolfstetter and Demougin (eds.) (1994): Journal of Institutional and Theoretical Economics. However, no matter which model is used for a modified contractarian argument, it should be clear that the definition of rationality is always depending on this particular model of individual behaviour.
Political Contractarianism and Equally Distributed Basic Rights
133
soon as their own survival is secured. Obviously this behavioural difference between egoists and altruists becomes more relevant the more altruistic individuals are able to satisfy their primary interests. Nevertheless, it can still be explained why the uncertainty in the state of nature has the consequence that even altruistic individuals behave mostly like egoists. Furthermore, this analysis can show that altruistic individuals are probably interested in a different kind of legal state, as soon as they no longer have to cope with the problem of fulfilling their primary interests. Secondly, I assume that all individuals have secondary interests which can be egoistic and altruistic. This allows the possibility to specify the general description mentioned above, in which some individuals in pluralistic societies are considered to be egoists while the remaining individuals support different normative views leading to different altruistic interests. In this context, I believe the assumption to be plausible that the majority of individuals within pluralistic societies have a broad spectrum of interests, including egoistic and stronger or weaker altruistic interests, while a minority of individuals have only a narrow spectrum of interests that can be reduced to egoistic interests.33 I will not specify the content altruistic interests in detail, although I suppose that most altruists would not establish a slavery state, even if they would be powerful enough. This assumption appears to be compatible with pluralistic societies because it is not based on the premise that everybody shares, for instance, a similar conception of liberalism. In consequence, the strong focus on egoistic interests of current contractarian theories can be avoided in this more complex setup. Nevertheless, even within this setup in the state of nature, the scarcity of goods and the problems of interaction will lead to a situation that can be described as uncertain and inefficient. Egoists will behave aggressively, leading to conflicts that motivate more and more altruistic individuals to behave aggressively in the way described above.34 Furthermore, this setup still allows the possibility that the minority of egoists establish a slavery state, at least as long as the majority does not have enough power to avoid this outcome. Even if egoists are not considered to be benefit maximizers, the exploitation of others in a slavery state could still be sufficiently advantageous, as egoists are not directly concerned about the interests of others. However, if the assumption of egoists existing in a minority is plausible, only a slavery state which oppresses the majority needs to be discussed further. Under this condition, the following argument becomes interesting. Thirdly, I think contractarian theories can be more convincing if they are based explicitly on a rational analysis of medium and long-term perspectives which already play an implicit role in the argument. Firstly, the oppressors of a majority 33
This is, of course, a very general understanding of altruistic individuals. In my thesis Pluralismus und Vertragstheorien I provide a more detailed analysis.
34
Because of this uncertainty, at least individuals with altruistic interests will not be able to fulfill their broad spectrum of interests. I dub this the problem of biased interests. Those individuals have an additional incentive to establish a minimal state to provide a framework of interaction in which they can fulfill more than only egoistic interests.
134
J. K¨uhnelt
must always calculate the risk of a revolution. One could object that slaves have no strong incentive to overthrow the rulers because they have at least some advantages from a slavery state. Moreover, rational rulers of a slavery state will try to prevent any attempt to organize revolutions, for instance, with sanctions, limitations of knowledge or constraints of communication. Nevertheless, the risk of a revolution still persists as the rulers will always be in the minority and a smaller shift in the distribution of power could already endanger their position. Secondly, the implicit contractarian reference to economic knowledge should be extended. For example, even Hobbes notes that the problems in the state of nature lead to inefficiency in trade and interaction.35 If my argument above is convincing, even Buchanan’s state of protected equilibrium is not sufficiently stable to motivate its citizens to invest within a medium or long-term perspective. In the already described slavery state the majority of individuals have no incentive to increase the efficiency of their actions, as only the masters will profit from such behaviour. Thus, the efficiency of their actions depends on the orders of the masters. However, a basic analysis of current industrialized societies shows the importance of investments for an efficient production of goods or efficient interaction. Examples are not only investments in production facilities or infrastructure, but also individual investments in personal specialisation. Furthermore, investments in education, science or political integration are also primarily efficient on a medium or long-term perspective. However, even the masters in a slavery state cannot be sure to be sufficiently powerful in the future. Hence, at least long-term investments will generally not be rational for them, as the risk persists that other egoists become strong enough or the majority overthrows the state in a revolution. Furthermore, even if the masters would try to invest, they still face the problem to find an efficient allocation of resources.36 Therefore, I conclude that even powerful individuals are only able to access medium and long-term advantages of an industrialized society if they live in a minimal state with nonnegotiable basic rights. Only in such a state, everybody has at least the security that he does not loose the property of his investments. In relation to the discussed setup, I believe a good position in a minimal state to be equally advantageous to the leadership of a slavery state. In consequence, powerful egoists have no strong incentive to establish a slavery state, as long as they are not sadists or fanatics. Of course, it is necessary to discuss this argument in more detail to prove whether or not it is really convincing within pluralistic societies. However, the aim
35 36
Hobbes (1991 [1651]): Leviathan, p. 89.
The decentralized structure of markets is one important origin for their efficiency. In relation to socialist societies, for instance, Friedrich von Hayek shows some general advantages of decentralized markets in relation to a centralized production, even though he his arguments are sometimes quite polemic. An overview on Hayek’s ideas can, for example, be found in: Hayek (1967): Studies in Philosophy, Politics and Economics.
Political Contractarianism and Equally Distributed Basic Rights
135
of this paper is restricted to outline the assumptions and the structure of an argument that has the potential to fulfill this claim.37 Still, one could object that the proposed argument is less convincing, as the majority of individuals are assumed to share the value that a slavery state is not acceptable. However, this assumption appears to be realistic as the conflicts arising between different normative views held in pluralistic societies are still taken seriously. Nevertheless, this assumption can only explain why the majority of individuals have one reason to reject a slavery state; it cannot explain why everybody should consider a minimal state to be better than other alternatives. However, supported by this premise, contractarianism has a better potential to show the advantages a minimal state has for the fulfilment of the interests of everybody. I tried to show that, under certain circumstances, a minimal state that secures stable basic rights will be more attractive than other possible states: Stable basic rights are supposed to ensure a stable and more efficient interaction and, hence, provide a framework in which everybody can better fulfil his egoistic and altruistic interests. Consequently, while rational egoists – at least as long as they remain a minority – have a reason to agree to a minimal state, the majority of individuals can accept this state because of its advantages and its accordance with their refusal of suppression. Additionally, as long as most individuals can also be considered to be self-regarding, this level of contentedness will further support the stability of the minimal state. However, some problems still remain unsolved at this stage of the argument. Most conflicts between normative conceptions are linked to the questions of whether or not and which further governmental tasks should be legitimized. Still, I think even in this case the proposed contractarian argument can lead to convincing results. Buchanan stands in the economic tradition and mentions the production of public goods as a possible gain of cooperation that can be established by an expansion of a minimal state. In relation to the setup discussed above, a further extension of the argument can at least be indicated at the end of this paper: Within current contractarian arguments only certain further governmental tasks can be legitimized because of the focus on egoism or an economic understanding of efficiency. However, given a constitution that secures everybody the realistic chance of further gains of cooperation in relation to his position in a minimal state, altruists and egoists would no longer be restricted to cooperate only in certain areas. As they will negotiate with reference to their broader spectrum of interests, even questions concerning the implementation of social justice can be discussed. Even though an extended minimal state is surely less efficient in economic terms as soon as social justice is established, it could still be more beneficial for the individuals living within this state, depending on their interests and results of their negotiations.
37
Moral or religious fanatics can cause problems, as they could try to establish a slavery state or other forms of oppressive states, if they are strong enough. I think that they can be convinced to a certain extent, but the reference to the efficiency of a minimal state appears not to be sufficient. A more detailed argument for the proposed relativistic contractarian argument can be found in my thesis Pluralismus und Vertragstheorien.
136
J. K¨uhnelt
Acknowledgments I am very grateful to Stephan Schlothfeldt, Michael K¨uhler and Sebastian Schleidgen for critical comments on earlier versions of this paper. Furthermore, I would like to thank Julia Day and Michelle Wilson for very good comments on style and language.
References Baurmann, Michael (2002): The Market of Virtue – Morality and Commitment in a Liberal Society, Boston: Kluwer Academic Publishers (Springer). Buchanan, James (1999 [1975]): The Limits of Liberty – Between Anarchy and Leviathan, Indianapolis: Liberty Fund. Dworkin, Ronald (2002): Sovereign Virtue, Cambridge (MA): Harvard University Press. Granovetter, Mark (1985): “Economic Action and Social Structure – The Problem of Embeddedness”, in: The American Journal of Sociology, 91. pp. 481–510. Hardin, Russell (2003): “The Freerider Problem”, in: Zalta, Edward (ed.): The Stanford Encyclopedia of Philosophy,
. Hayek, Friedrich von (1967): Studies in Philosophy, Politics and Economics, Chicago: University of Chicago Press. Heap, Shaun and Varoufakis, Yanis (2004): Game Theory – A Critical Text, London: Routledge. Hobbes, Thomas (1991 [1651]): Leviathan, edited by Tuck, Richard, Cambridge: Cambridge University Press. Kliemt, Hartmut and G¨uth, Werner (2008): “The Rationality of Rational Fools: The Role of Commitments, Persons and Agents in Rational Choice Modelling”, in Peter, Fabienne and Schmid, Hans Bernhard (eds.): Rationality and Commitment, Oxford: Oxford University Press K¨uhnelt, J¨org (2009): Pluralismus und Vertragstheorien, forthcoming. Ludwig, Bernd (1998): Die Wiederentdeckung des Epikureischen Naturrechts, Frankfurt am Main: Vittorio Klostermann. Nozick, Robert (1974): Anarchy, State and Utopia, New York: Basic Books. Pettit, Philip (2002 [1995]): “The Virtual Reality of Homo Oeconomicus”, in: Pettit, Philip (ed.): Rules, Reasons, and Norms, Oxford: Oxford University Press. Rawls, John (1993): Political Liberalism, New York: Columbia University Press. Ryan, Alan (1996): “Hobbes’s Political Philosophy”, in: Sorell, Tom (ed.): The Cambridge Companion to Hobbes, Cambridge: Cambridge University Press. Selten, Reinhard (1991): “Evolution, Learning, and Economic Behavior”, in: Games and Economic Behavior, 3, pp. 3–24. Sen, Amartya (1977): “Rational Fools – A Critique of the Behavioural Foundations of Economic Theory”, in: Philosophy and Public Affairs, 6, pp. 317–344. Simon, Herbert (1997): Models of Bounded Rationality, Volume 3: Empirically Grounded Economic Reasons, Cambridge (MA): MIT Press. Simon, Herbert and March, James (1958): Organizations, New York: Wiley. Sugden, Robert (1993): “The Contractarian Enterprise”, in: Gauthier, David and Sugden, Robert (eds.): Rationality, Justice and the Social Contract. Themes from Morals by agreement, Ann Arbor: University of Michigan Press. Wolfstetter, Elmar and Demougin, Dominique (eds.), (1994): Journal of Institutional and Theoretical Economics, 150.
Part III
How to Ensure the Stability of a Legitimate State?
Value-Mistaken and Virtue-Mistaken Norms Philip Pettit
Many norms appear to arise, or at least to stabilize and fixate, as a result of an error on people’s parts as to the attitudes of others. I think that this is a phenomenon of more than marginal interest and my aim here is to put it in the limelight. In doing so I draw in part on work done elsewhere.1 My discussion will be in five sections. I look in turn at the definition of norms; at how, under this definition, a norm can be based on an error shared among those who observe it; at the psychological plausibility of such an error; at the sorts of norms that are likely to be supported, for good or ill, in this way; and at the lessons for institutional design.
1 The Definition of Norms The word ‘norm’, as used in characterization of society, has two more or less obvious connotations. First, anything that deserves to be described as a norm of a society – an actual, not a would-be, norm – has to be a regularity that prevails there: a pattern of behavior that characterizes the society, marking it off from actual or potential competitors. But, second, the regularity in question cannot be a matter of indifference amongst the people who sustain it. In order to have the status of a norm, a social regularity has to be seen, however tacitly, as something with a certain sort of normative claim on people’s allegiance: as something that, for whatever reason, is appropriate in relevant contexts; it has to attract approval or the breach of the regularity disapproval. But these two connotations do not exhaust the associations of the word ‘norm’, as that is used in social contexts. Suppose that a regularity was behaviorally and attitudinally supported among the members of a certain society but that there was no connection between the attitudinal approval and the behavioral compliance; suppose, in other words, that the attitudinal support was epiphenomenal in relation to the behavioral compliance. In that case the regularity would certainly be something that people had reason to welcome and celebrate, like an aspect of their biology; 1
Pettit (1990); Brennan and Pettit (1993, 2000, 2004).
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
139
140
P. Pettit
but it would scarcely count as a regularity that they treated as normative. This suggests that we ought to build a third connotation into our use of the word ‘norm’. We ought to stipulate that in order for a regularity to count as a social norm, it should not only be instantiated as a general rule, and not only seen in general as an appropriate regularity to instantiate; in addition, the fact that it is seen as appropriate – the fact that it is approved – should help to explain why it is generally instantiated. This characterization of social norms is fairly rough, since it leaves open a range of questions. Does a regularity count as generally instantiated if it is a regularity that applies only to those holding a certain office or meeting a certain qualification? How extensive is the pattern of approval envisaged when it is said that the regularity must be seen as appropriate: appropriate morally, or at least prudentially, or at least for someone with this or that goal in mind? Moreover, must the approval be associated with the type of behavior, considered in general, or will it suffice if it just happens instance by instance that the behavior is seen as appropriate? And what, finally, is required for the pattern of approval to help to explain the pattern of behavior? Must it contribute in some measure to the production of the behavior, at least among a number of those complying? Or will it do if it is there to reinforce the behavior, should the motives that normally produce it fail for one or another reason? Will it do, in other words, if it is a virtual or standby force that is triggered to support the behavior only on a need-to-act basis, when the ‘red lights’ are illuminated?2 I am happy to leave aside most of these questions, taking the inclusive view that we should be ready to describe as a social norm any of the large range of regularities in people’s behavior that meet some version of our three conditions. Thus a regularity among the members of a society will constitute a norm just in the event that: – nearly everyone conforms; – the behavior is nearly always thought appropriate in some way; – and this attitude helps to explain the general conformity. I am happy also to leave aside the further question, often raised in this context, as to whether it is essential for a social norm that the fulfillment of the three core conditions is a matter accessible to common knowledge. It will be commonly known that the conditions are fulfilled if each is aware that the conditions are fulfilled; each is aware that each is aware of this; and so on indefinitely, for any question of higher awareness that may arise. I am content that a regularity may count as a social norm even if it is not accessible to common knowledge in this sense. The upshot is a generous, perhaps deflationary sense of social norm. But it is supported by most recent authors on the subject and ought not to generate any deep controversy.3
2
Pettit (1995).
3
Hart (1961); Winch (1963); Coleman (1990); Sober and Wilson (1998); Elster (1999).
Value-Mistaken and Virtue-Mistaken Norms
141
Norms in this sense often have very welcome effects. They dictate the ways in which people relate to one another in discourse, generally embracing patterns of honesty, trustworthiness and sincerity; the ways in which they otherwise seek to influence one another, eschewing resort to violence, theft, fraud, and coercion; the ways in which they commit themselves conscientiously to various collaborative causes, playing the part that is collectively required of them; and the ways in which they conduct their business and professional lives according to relevant codes of practice. Such norms are the motors of civil society, leading people to deal well with one another, even when they are beyond the reach of the law, are unconstrained by the discipline of self-interest, and are free of the incentives provided by family and related ties. They exemplify the idea of the rule of obligation introduced by H.L.A. Hart in his classic study of ‘The Concept of Law’.4 He characterizes such rules by the fact that they generally prevail in the relevant group, they are supported by serious social pressure, they are thought useful in some way for the life of the group, and they are individually burdensome, however beneficial in group terms. But norms in the sense defined – the sense that answers to our rough account – may also be socially neutral or even socially counterproductive, unlike Hart’s rules of obligation. Norms that are socially neutral will include the sorts of norms that barely enter consciousness, such as those governing eye-contact and turn-taking in conversation, and the distance at which it is appropriate to stand in relation to an interlocutor.5 Socially counterproductive norms come in a variety of shapes. Some will serve subcultures well, for example, while serving the society as a whole badly. Some will have a mutually destructive effect as in the norms whereby it becomes obligatory for people to exact revenge in kind for any harm done to a member of their family.6 And some will impose fashions and fads on people who would generally prefer not to be motivated to embrace them; they represent a tyranny of majoritarian esteem.7 One final comment on this definition of norms, before turning to the possibility of norms sustained in error. There is no conflict in saying that a regularity that is a matter of law, being supported by legal penalty, is also a norm; it can satisfy the conditions for being a norm while satisfying also the conditions for being a law. And neither is there any conflict in saying that a regularity that is a matter of convention, say because it solves a coordination predicament to everyone’s satisfaction,8 may also be a norm; again it can consistently satisfy the conditions for belonging in each of those categories. The class of norms to which our definition directs us can overlap, and surely does overlap, with those distinct categories.
4
Hart (1961), pp. 84–85; see also Ullmann-Margalit (1977), pp. 12–13.
5
Goffman (1975).
6
Elster (1990); Hardin (1995); Nisbett and Cohen (1996).
7
McAdams (1997). See also Dharmapala and McAdams (2001) and McAdams (1995).
8
Lewis (1969).
142
P. Pettit
2 The Possibility of Error-Dependence The definition of norms just given allows for a variety of normative regularities. Norms will vary, for example, depending on factors like the following: – whether compliance is approved, non-compliance disapproved, or both conditions hold; – whether the approval involved is a matter of egocentric comfort or advantage or engages a more moralized disposition; – whether the approved compliance involves successfully achieving something or just trying to do so, as in a norm of aspiration. I am interested here, however, in a different source of variation in norms. Under the definition offered, it is required that there is a general pattern of approval and that this helps to explain the compliance of each with the regularity in question. But there are a number of ways in which the general pattern of approval may connect with individual compliance and, correspondingly, there are different sorts of norms that the approval may support. The standard connection, as we might describe it, involves each individual in internalizing the general pattern of approval – coming to share the attitude of approval present in the society at large – and in being led by his or her personal approval into instantiating the regularity. When such an internalized norm is in place, those who conform to it will do so out of personal attachment or conviction to the value in question and will count, by traditional criteria, as agents of virtue. Some theorists give almost exclusive attention to internalized norms, emphasizing how far internalization supports the stability of normative behavior.9 But even if internalization aids stability, it seems scarcely deniable that virtuous, normobservant behavior may often be supported by something distinct. It may be reinforced – it may be protected from weakness of will and the like – by the fact that if someone does not act in accordance with a received norm, particularly a norm that is socially beneficial, then he or she is going to lose the approval of others, and perhaps attract their disapproval. Assume that for whatever reason, intrinsic or instrumental, people care about enjoying the positive approval of others and avoiding their disapproval.10 In that case it should be clear that if failures of virtue lead to failures of behavior, then people are liable to be punished, however involuntarily, by those others who observe and understand what they have done; they will be punished, at the least, by the withdrawal of approval, or the appearance of disapproval. Once we recognize that people may be motivated by the pursuit of others’ approval, rather than by the internalization of that approval, then we are in a position to see that not all norms, however widely instantiated, may be internalized. There are two varieties of non-internalized norms that are possible, in particular.
9
Cooter (1994; 1996).
10
Pettit (1990); Brennan and Pettit (2004).
Value-Mistaken and Virtue-Mistaken Norms
143
Given that people are motivated by the desire for the approval of others, the first possibility is that each acts out of that desire, and out of that desire only, without anyone actually having the attitude of approval in question. Everyone, in other words, is mistaken in thinking that others generally approve of a certain form of behavior – they are mistaken about the relevant value adopted by others – yet everyone displays that behavior, believing falsely that this will attract the approval of others. This possibility has been described by psychologists as a case where pluralistic ignorance supports a norm.11 A good example of a norm supported in pluralistic ignorance – better, perhaps, pluralistic error – is that whereby, according to one study, a group of students tended to comply with a certain regularity in the amount of alcohol consumed on a night out. The study revealed that almost all of the students disapproved of the relatively high level of drinking required under the regularity but abided by that regularity because of mistakenly thinking that others disapproved of lower levels of consumption.12 Should a regularity supported in such error about the values of others count as a norm, under our definition? It may seem not, since the definition presupposes that the behavior involved attracts approval, or its absence disapproval – that it actually answers to an espoused value – and all that is available in this case is the mistaken expectation of approval or disapproval. But if I believe that everyone else approves of some behavior, even a sort that I myself don’t approve of, there is a sense in which I believe that that behavior is appropriate or valuable: it is appropriate-accordingto-local-standards. So perhaps this regularity should count as a norm under the definition offered. Even if that is not accepted, however, it is surely reasonable to recognize how close to a norm in the strict sense this sort of regularity is and to treat it as a limit case. And if that seems too relaxed a view to take, we can go back and define a norm so that what is required is not necessarily general approval of the behavior involved but the general expectation that there will be such general approval.13 So much for the first sort of non-internalized norm. The second variety has not been explicitly recognized in the same way but becomes visible once we recognize that people care about enjoying the approval of others and avoiding their disapproval: for short, care about their approval. It involves error, like the first, but error about a somewhat different issue from that of whether others approve of a certain behavior. In this case, people believe that others approve of the relevant type of behavior, and they are right to do so: they get the values of others right. But here they mistakenly think that it is this personal approval that leads others, in fidelity to their values, to display that behavior. They rightly believe that the others see the behavior as appropriate or valuable, even perhaps required, but they wrongly think
11
Miller and Prentice (1994, 1996).
12
See Prentice and Miller (1993); Schroeder and Prentice (1998).
13
Brennan and Pettit (2004), pp. 267–8.
144
P. Pettit
that others perform the behavior out of a recognition of its merits in that regard; they mistakenly think that the others are virtuous. This scenario of error about the virtue of others, as distinct from error about their values, is possible under the hypothesis that people care about the approval of others. For while everyone thinks that others are displaying the normative behavior out of personal virtue – as a result of their personally approving of the behavior – the fact may be that they display that behavior out of a desire to enjoy the approval of others or avoid their disapproval. The scenario is one in which everyone is ‘continent’ rather than virtuous, doing the right thing for reasons other than that it is right: doing the right thing for the sake of the approval of others. A norm is sustained in people’s behavior but it is not internalized by anyone. The first case of a non-internalised norm not only involved error; it involved error essentially. Did students in the drinking example know that others don’t actually hold by the pattern of approval imputed by each, that knowledge might be expected to undermine the norm; and this appears to have been borne out in that actual study. Adherence to the norm in this case is, as we might say, error-dependent: more exactly, it is dependent on the near universality of the error. Does the second sort of non-internalized norm involve error essentially, in the same way as the first? Is compliance with this norm also likely to be errordependent? I believe that there is a possible scenario in which it would be. Imagine that the soldiers in a military unit all display courage in action, that they all approve of courageous behavior and that those two facts are correctly registered in common belief amongst them: each believes that each believes this, and so on. Imagine, next, that they each hold by the equally common belief that the courageous behavior of others testifies to the presence of personal courage – that is, virtue – and that others act as they do out of such courage. But imagine, finally, that this belief is false, since they each act out of the belief that if they act courageously they will be taken to be courageous like the others and that they will attract approval on that account: they will avoid the stigma of being seen as cowards. In such a situation compliance with the norm is certainly error-dependent: that is, dependent on the near universality of the error. For suppose that this error was not in place, so that it was a matter of common belief that no one is really courageous or virtuous and that they all behave courageously, if they do, out of a wish to be seen as courageous. In that case, not only will there not be a motive of courage to drive the soldiers to behave courageously, neither will the motive of seeking a reputation for courage be present; or neither at least will it be effective. For no one will be able to think that others will take them to be courageous just because they behave courageously. Behaving courageously will not be supported, then, in the motivations of the parties and, short of an awareness of this absence of motive generating some further transformation – which it well may do – courageous behavior will vanish. We can imagine an informationcascade in which the awareness that there is no reputational gain in acting courageously – in particular, no shame in failing to do so – rapidly spreads to the point where almost no one remains disposed to act that way; there is mutiny in the ranks.
Value-Mistaken and Virtue-Mistaken Norms
145
One final comment. In both the cases we have considered the error committed is made by everyone. That is a particularly dramatic and clear-cut sort of situation but it is worth mentioning that error-dependence may also come in degrees. A norm may come to be maintained, not in virtue of universal error on some matter, but in virtue of a certain level of error, whether about attitudes or dispositions. In this paper I abstract from a consideration of such cases, however, as they would take me too far afield. I focus on the more clearly paradoxical case where everyone is affected by error and is led by that error into conforming with a norm. This focus is dicated primarily by a concern for keeping things simple and manageable.
3 The Plausibility of Error-Dependence These cases of error-dependent norms may seem, on the face of it, to represent logical but very implausible scenarios. For, it may be asked, how could people be led into the egregious sort of error required for the norms to emerge or at least stabilize? Isn’t it willful and ad hoc to postulate such a rank degree of proneness to error about the values or virtues of others? I shall try to argue that it is not. Social psychologists have documented a consistent tendency in people’s attribution of motives to others that they describe as the fundamental attribution error; I prefer to think of it as a bias, since it may not invariably lead to mistakes. E.E. Jones presents it in the following terms. ‘I have a candidate for the most robust and repeatable finding in social psychology: the tendency to see behavior as caused by a stable personal disposition of the actor when it can be just as easily explained as a natural response to more than adequate situational pressure’.14 The fundamental attribution bias consists in a preference for explanations of what people do, and of what people say, that emphasizes the contribution of character over context. Let someone do or say something and there will be many possible explanations. Some will suggest that people adapt in very finely tuned ways to differences of circumstances, so that there are no easily predictable patterns in evidence. Others will depict people as possessed of dispositions that are stable across different contexts and that dictate predictable patterns of response: patterns that materialize reliably across quite different sorts of circumstances. The fundamental attribution bias consists in a preference for this second sort of explanation, in which character receives relatively more weight, context relatively less. The finding about the fundamental attribution bias – and I shall assume that it is a well-documented finding – bears in an obvious way on issues of the kind that arise with non-internalized norms. Suppose that your drinking companions routinely drink a certain amount and routinely acquiesce in a general acceptance of that level of drinking, perhaps even offering general applause for going that far: we may well expect such applause, given that under the story presented it too can be expected – mistakenly – to earn 14
Jones (1990), p. 138.
146
P. Pettit
approval. What more natural explanation of your companions’ behavior, then, than that they approve of that level of drinking – in particular, perhaps disapprove of a lower level – and that this approval plays a role in explaining their behavior? To explain what they do in those terms is to project on them a stable complex of attitude and motive. The disposition imputed suggests that regardless of who they are drinking with, they should each be expected to maintain the same level of consumption; it is to downplay the role that circumstances might be expected to play. Contrast this sort of explanation, however, with that which appeals to the desire of each to be well regarded by others. Under that account, each has to explain the behavior of the others as the product, not of a simple attitude-motive complex, but in terms that give a much larger role to context. Each will take others to be stable of disposition at the high level of seeking the approval of others. But the disposition imputed will be expressed in action in very different ways, depending on differences of circumstances. Let the context be one where others are expected to approve of a high level of drinking, and they will tend to drink to that level. Let it be one where others are expected to disapprove of such drinking, and they will tend to drink less. The fundamental attribution bias is bound to lead people to explain the presentation of others in the simpler, more context-invariable way. And it is precisely that sort of explanation that each is expected to give of the behavior of others in the story about the fixing of the drinking norm. True, the explanation given turns out to be a mistake. But it is not an implausible explanation to posit. It is the sort of account that is bound to appeal to anyone who is susceptible – as we are all said to be susceptible – to the fundamental attribution bias. The comments made on this case carry over to the second case of non-internalized norms. Imagine that you are a soldier among soldiers, involved in a series of dangerous military engagements. There is no doubt in your mind but that you and your fellows all approve of courage. Nor is there any doubt about this being a matter of shared awareness. After all, you all talk about courage in the most positive terms, and listen to regular harangues on the topic. So now you find that the others in the ranks with you do indeed behave very courageously. What more natural explanation than that this is due to the fact that they put their lives where their words are: they march to the drum of their values? This explanation is straightforwardly in line with the fundamental attribution bias, imputing to others a stable disposition – again, a simple attitude-motive complex – that can be expected to produce similar behavior across many fine differences of context. Contrast with that explanation the account that you would have to give of others’ behavior, were you to grasp the facts of the case, as my hypothesis presents them. As in the other case, you would posit a stable, high-level disposition – the concern with approval – that is liable to produce quite different behaviors in different contexts. You would downgrade character and upgrade context in a way that runs deeply against the grain of the fundamental attribution bias. I conclude that in these respects the stories told in illustrating the possibility of error-dependent norms are not at all implausible. The alleged mistakes about the values and virtues of others are precisely the sorts of error that we should expect them to be liable to make, according to received psychological analysis.
Value-Mistaken and Virtue-Mistaken Norms
147
But there is a second way in which the stories told may be thought to be implausible. They suppose that people can be moved by one incentive but can expect at the same time that others will explain their behavior by reference to another. The motive that actually operates, according to those stories, involves the desire to enjoy the approval of others and escape their disapproval. But the motive that agents expect to be ascribed involves a much lower-level, context-insensitive disposition. Is it reasonable to suppose that people could expect others to make that sort of mistake? I believe it is, again because of the presence of the fundamental attribution bias. The motive people expect to be attributed is the disposition to act in a way that answers to how they are (really or apparently) disposed to approve and disapprove of behavior: in the one case the disposition to drink to the level they make a show of endorsing; in the other the disposition to act courageously after a pattern they clearly admire. As there is no surprise about people’s explaining the behavior of others in the erroneous manner described, then, so there should be no surprise about their expecting others to make the same mistake in relation to them. If the fundamental attribution bias is really a feature of human psychology, then it is likely to be a salient feature and one that people are likely to expect others to display. One final question. On the account given, people explain others’ behavior by ascribing a stable, dispositional motive, and they expect others in turn to explain their behavior by reference to a similar motive. But they themselves, according to our account, act on the basis of a pattern of motivation that gives the lie to the notion that low-level, dispositional motives are the more or less ubiquitous sources of action. And they may more or less consciously act this basis. Does this make the account inconsistent? Or does it at least mean that those to whom the account is applied are inconsistent? No, it does not. I might consistently think that I am motivated differently from others, that others don’t see this, and that actually I am exceptional; there is independent evidence, indeed, that this is how we each tend to think of ourselves, avoiding the attribution bias in our own case.15 Unlike others, so I may feel, I lack the required attitude of approval, or I lack the disposition to act as I approve. But I do of course see the importance of securing approval and so fall back myself on this other motive. On the account offered, we might suggest that people generally think they are different from others in these respects. And that would not be a particularly difficult hypothesis to embrace, given the frequent asymmetry in how people view themselves and others. There is a second fairly plausible hypothesis that would also explain how I could act on one motive but expect to be ascribed another. This is that I am not always aware of how my own motives work; it is not always the case that I more or less consciously act out of a desire for esteem or a fear of disesteem. While being sensitive primarily to the forces of approval and disapproval, I may often imagine, whether out of self-ignorance or self-deception, that I am not like this: that I am, in fact, exactly like others in being moved by the same values or the same virtues.
15
Jones and Nisbett (1971).
148
P. Pettit
To sum up, then, there is no deep implausibility in holding by the assumptions that would predict and explain the appearance of error-dependent norms. The fundamental attribution bias, powerfully supported as it allegedly is, should lead us to expect exactly the pattern of motivational attribution that would account for the presence of such norms.
4 Varieties of Error-Dependent Norms There are two broad types of error-dependent norms identified in the story told so far, one sort exemplified by the drinking case, the other by the case of military courage. The first I describe as value-mistaken norms, the second as virtuemistaken norms. Value-mistaken norms will be liable to emerge only so far as two conditions are fulfilled. The first is that those among whom the norm emerges care greatly for one another’s approval; in particular, care enough to be ready to act against their own values for the sake of the approval of others. And the second is that conditions are such as to allow the parties to be mistaken about one another’s values, and so about what they are each likely to approve or disapprove. The first condition suggests that there will be considerable peer pressure associated with membership of the group in question. And the second implies that members of that group do not come together out of a search on the part of each for those of a common mind: a search for people who share the same values and the same attitudes of approval and disapproval; the group forms on some other basis. These conditions are likely to be fulfilled quite commonly, so that we should not be surprised if there are a variety of value-mistaken norms in place in any society. Two salient possibilities are worth mentioning. One involves what we might think of norms of corruption: that is, corruption from the point of view of the wider society. Take a group of police officers who have a defensive attitude towards outsiders, expecting them to be somewhat hostile. Membership and acceptance in the group will tend to matter greatly to officers, particularly so far as they expect outsiders not to accept them fully. Given that they do not invariably join the force out of any particular values, there will be room in such a group for members to be unsure about what exactly others care about. In such a scenario it is very possible that a number of value-mistaken norms may emerge or stabilize. Imagine that someone in the group is seen by others as breaching police rules – that is, the rules imposed by higher authority – in some perhaps not very significant manner. Will any of those others report the breach? Very possibly not, since they are each liable to be afraid of alienating themselves from the offender and his or her friends; and this, even if they each actually disapprove quite strongly of the breach. But suppose now that this pattern of not welching on fellow-officers begins to get established as a routine of behavior. As it does, this will suggest to each that others disapprove of welching and approve of turning a blind eye to an
Value-Mistaken and Virtue-Mistaken Norms
149
individual officer’s offences, at least within certain limits. And so we can imagine that a norm of not welching may emerge, with almost everyone conforming, with almost everyone expecting almost everyone to approve of conformity, and with this expectation helping to keep the norm in place. But for all that our story entails, the norm may be value-mistaken in character and so not internalized by the parties. The members of the force may each disapprove of not welching – they may embrace the value of whistle-blowing – and may each keep to the pattern of not welching only out of an erroneous expectation as to how others would respond to whistle-blowing. If there is a norm of not welching in place among the members of such a police force, of course, then there will be room for the emergence of other value-mistaken norms as well. Suppose that while members of the group think that others disapprove of whistle-blowing, they believe that there are certain limits to what will be tolerated: there are forms of breach such that no one would disapprove of their being publicized. And suppose now that a certain form of breach that goes beyond the ascribed limits begins to emerge in the group – say, one of taking bribes – yet is not exposed by fellow officers. It may be that each is wrong in thinking that this is beyond the limits of tolerance; it may be that the behavior is reported by no one, out of fear of incurring disapproval as a whistle-blower. But so far as there is a general mistake made about the limits of tolerance – surely, a real possibility – each may then be led to think, again mistakenly, that actually the behavior in question is not disapproved of by others, perhaps even that it attracts a degree of approval. The error about attitudes to whistle-blowing can generate other errors too. Value-mistaken norms may multiply and propagate. The group of police officers imagined exemplifies a range of real-world possibilities. Instead of police officers we might have imagined the lower-level workers in any enterprise, the students in a school, the inmates in a prison, or the members of any professional association. In every such case there will be motives in place that allow for a value-mistaken norm of not whistle-blowing to emerge and, as a result, for the appearance of other value-mistaken norms. These other norms may lead to a tolerance of shirking or bullying or closing ranks against outsiders, even when everyone in the relevant group disapproves of those behaviors. I do not say, of course, that whenever there is a norm against whistle-blowing, or whenever any of the associated norms prevails, that is because people are mistaken about one another’s values. But I do say that even if individuals hold values that would support whistle-blowing and would outlaw a variety of other breaches, those values need not give rise to corresponding norms. The value-mistaken norms illustrated can be described as norms of corruption, since they support behaviors that fall away from standards that others expect members of the group to honor. But it is worth mentioning that they do not exhaust the possibilities for value-mistaken norms. A second, more or less salient possibility is that norms of correctness, as I will call them, may emerge on the basis of mistaken values. I take the word ‘correctness’ from the way in which we speak of political correctness, though I believe that norms of correctness may run far beyond the limits of any explicitly political context.
150
P. Pettit
With norms of corruption people are misled by the action or inaction of others into ascribing values that none of them actually endorses. With norms of correctness, they will tend to be misled by the speech or silence of others into making similar mistakes. Suppose that a group of a political or religious or cultural character is established and, as in the earlier sort of case, that it commands great allegiance among members; they each care greatly about belonging. And now imagine how people are likely to respond to the words of some authorities or would-be authorities within the group, when they declare what the attitudes and values of the group are and what they require. Such a declaration will carry influence, so far as no one opposes it. Yet people may each fail to oppose it, not because of sharing the values declared, but because of assuming that others do share them and because of fearing the ostracism that would go with resisting the common line. No one may oppose the declaration, in short, because no one may dare oppose it; no one may think that the risk of being the only dissident voice is worth taking. Nor is this all. It is also possible in such a case that the person or persons who assume the role of authority and declare the values of the group are not themselves sincere. Just as those who fail to oppose them may not share in the declared values, so the authorities themselves may fail to share in them. The people involved may mistakenly think that others do share the values and may put themselves forward as spokespersons for those values, not out of an attachment to the ideals, but out of a wish to ingratiate themselves with other members of the group. This possibility is all too easy to envisage. Any number of political and religious and social movements are liable to generate the sort of pressure under which words can engender norms that no one dares to dishonor, even while they are each opposed to the norms in their hearts. Insincerely endorsed words may be cheap for those who utter them; indeed, they may promise a positive reward in the approval that they are expected to earn. And insincerely accepted or unopposed words may be powerful; they may give rise to a cascade of behaviors that no one approves but that everyone displays. So much for value-mistaken norms, whether of corruption or correctness. The other error-dependent norms that we identified are virtue-mistaken rather than valuemistaken. They arise, not on the basis of an error about what others approve or disapprove, but on the basis of an error about why others display the behavior of which they approve and avoid the behavior of which they disapprove. The error here consists in thinking that others are virtuous – they are disposed to act according to their values – and that it is their virtue rather than any ulterior motive that explains why their actions conform to what they say and think. Virtue-mistaken norms presuppose that membership and acceptance matter to people just as much as they do in the other case, so that each is loathe to lose the approval of others or earn their disapproval. But it is a matter of common awareness in this case that such and such patterns of behavior are approved of or disapproved of – the values of people in the group are manifest. Here error can kick in, then, not in the perception of how others approve or disapprove, but only in the perception of their motives. Why, however, might it kick in? Why might it be the case that although everyone thinks that others are possessed of the virtue of living up to their values, no one actually has that sort of virtue?
Value-Mistaken and Virtue-Mistaken Norms
151
The obvious answer is that while such virtue is a natural disposition to ascribe in explanation of value-conforming behavior – this, in view of the fundamental attribution bias – the value in question is very demanding and is unlikely to be capable of motivating the ordinary run of people, all on its own. In order to identify examples of virtue-mistaken norms, then, we need only reflect on cases where the values that people proclaim – and proclaim sincerely – are really very demanding and are unlikely to have such an imaginative and emotional appeal that they will routinely gird agents against temptation. People will see others generally conforming but will have to struggle against temptation to conform themselves. And when they succeed, if they succeed, that will not be because they are as virtuous as they think others are. It will be because the cost of failure, as they see it, would be too grave to bear: it would consist in being seen, to their shame, as more or less isolated defectors. The case of military courage exemplifies this structure faithfully. It goes against the grain of human nature, by the testimony of history and introspection, to expose oneself to grievous physical danger; or to do this, at any rate, when there is no immediate goal like that of aiding a comrade in trouble. But in the military context everyone will clearly regard action in face of such danger as a supreme value and everyone will see others as having enough virtue – enough courage – to be able to honor that value in practice. In such a context, the prospect of being the only one not to act virtuously – the only one not to act as if they had the virtue – will promise ignominy and stigma. It should be no surprise if, under the force of that motive, everyone does indeed act virtuously. No one has the virtue of courage: no one has an attachment to the value that is strong enough to see them through adversity. But, this being the only way to avoid the prospect of a crushing burden of shame, everyone simulates the presence of such virtue and seeks to hide what they wrongly see as their untypical, perhaps unique, cowardice. The structure that courage exemplifies, according to this story, is liable to be replicated across many different contexts. Take any of the values that are universally acknowledged, difficult in practice to honor, yet fairly generally conformed to. In any such case, there will be a possibility that the norm is maintained on a virtue-mistaken basis. There is going to be a question, then, as to how far various norms may not be due to mistakes on people’s parts about one another’s virtue; in particular, mistakes sourced in a credulous acceptance that others are made of better stuff than they are themselves. We have focused in this essay, for simplicity, on cases where error is wholesale rather than coming in degree. There may not be many norms where it is plausible to think that they are virtue-mistaken in that wholesale way but there are likely to be many where such error plays at least a partial role. Think of values that are commonly proclaimed, whether in society as a whole or in certain groups, such as punctiliousness in making one’s tax return, steadfastness in upholding a religious faith, or conscientiousness in preparing oneself for committee deliberations. All of these values make demands that may prove hard to sustain as one is assailed by the temptation to fudge some financial details, to raise doubts about a church’s teaching, or to cut corners in reading background material. People may well overcome such temptations by dint of attachment to the relevant value. But there is always a possibility that the real force that enables them to achieve this success is not their natural
152
P. Pettit
virtue but the perhaps credulous belief that others do have such virtue combined with the desire to avoid the shame of appearing to be more or less isolated deviants. The examples of value-mistaken and virtue-mistaken norms that we have canvassed in this last section should be sufficient, I hope, to show that error-dependent norms are not only a possibility; it is very likely that some such norms obtain in any society or any grouping. That people conform to norms, then, does not mean that those norms are internalized amongst them: that they internalize the values and the corresponding virtuous dispositions. Many norms may prevail – whether for good or for ill – as a result of the simulated adherence to certain values or the simulated display of certain virtues. The social patterns that obtain among people in aggregate may give only very misleading cues as to the patterns that obtain in their individual souls.
5 Towards Institutional Design If there is likely to be such a cleavage between patterns in social life and the patterns in people’s souls, of course, then that raises a serious question as to how far we should deplore this phenomenon, or how far we should try to turn it to good. The moralist’s response will be to say that we should deplore it. But the institutional designer’s will be that, on the contrary, we should try to build on it. Recognizing that the task in institutional design is to construct something straight out of what Kant called the crooked timber of humanity, we should try to see how far mistakes about value and virtue can be exploited for social good. I favor the response of the institutional designer. It may seem that the response involves condoning and encouraging hypocrisy. But the hypocrisy involved is special. Hypocrisy in the normal sense means dissimulation: pretending to have done something approved of, or to have avoided something disapproved of, when this is not actually the case. Hypocrisy of the sort displayed in upholding error-dependent norms is quite different. It means doing something good out of a love of approval, or out of a fear of disapproval, while pretending to do it out of a love of the value relevant to the approval or disapproval. This is not the hypocrisy of dissimulation but, in the old phrase, the hypocrisy of simulation: the hypocrisy involved in simulating the behavior of those who internalize a norm, without having internalized it oneself. Simulation is a very benign sort of hypocrisy, since it leads to the same behavior as internalization of the norm; the shortfall associated with it is not behavioral but only motivational. Assuming that we should not baulk at the prospect of condoning such simulation, what steps in institutional design would the observations of this paper support? The paper supposes that people care about the esteem of others, and shrink from their disesteem; and that mutual exposure is sufficiently assured to engage this concern and motivate the simulation of internalized, normative behavior. But it directs our attention to two phenomena in particular that may be exploited in mobilizing the forces of esteem and using them to generate or reinforce socially desirable patterns
Value-Mistaken and Virtue-Mistaken Norms
153
of behavior; I shall assume for simplicity that there is no issue about what patterns count as desirable, what not. The first is the tendency of people to be guided by a desire for correctness, in particular correctness according to what are taken as the expectations of a group. And the second is people’s credulity in being prepared to accept that others are virtuously attached to a value that they themselves find it very hard to internalize in their motivations. We can easily imagine situations in which the culture of correctness gives life – and so, can be exploited to give life – to some socially attractive norm. Without really recognizing or internalizing the value of not smoking in the company of others, or putting one’s recycling in a separate bin from one’s trash, or using language that respects proprieties of gender or race, it seems entirely plausible that people should be moved to act in a way that supports such patterns. But there is no block to this being true of everyone. And so we can equally imagine that while each acts in a way that will attract the approval of those who recognize the appropriate value, actually there may be few if any members of the society who do recognize it. The norms we imagine emerging under a culture of correctness, we can equally easily imagine emerging as a result of people’s credulity in thinking that others are more virtuous than they are themselves. Consider the old emphasis on the importance of giving a good example to others, in particular giving a good example to those who might be expected to be less attached to a relevant norm. This underscores the common belief that if people have faith in the virtue of their fellows – including the virtue of their betters – then this sharpens the incentives they have to simulate that virtue, thereby winning esteem or avoiding disesteem. The belief is also underscored by the common fear of revealing to the members of a community just how many of them fail to live up to certain standards. It is a rare society in which government will be happy to reveal the fact that fewer people comply with the tax laws than is generally believed. It is a rare church in which the authorities will concede that fewer members conform to certain norms of faith or morals than might be generally expected. This being so, there can be little doubt about the possibility of exploiting people’s credulity in order to muscle certain norms into existence. But could we really hope to establish a norm by putting it about that correctness requires such and such behavior; or by making it seem that many people act out of a love of the value involved in the norm? Wouldn’t any norm established in that way be very unstable, being susceptible to collapse at the point where the truth comes out: the Emperor has no clothes? I don’t think so. Norms that begin in a culture of correctness or a habit of credulity can gain a more solid hold if the value involved – unlike perhaps the value of becoming intoxicated with others – is one that can really engage human affections. The point is particularly persuasive if we consider a case, not where absolutely no one holds the value at the start, but where a portion of the society do. Correctness and credulity may cause the class of those who hold by the value to increase, as more and more of those who pay it lip service only – behavioral service only – come to register and feel its affective grip. This claim should not be surprising. There is a long and plausible tradition of thinking that internalized adherence to norms can be supported and stabilized by
154
P. Pettit
collateral incentives and my claim about the power of a culture of correctness, or a habit of credulity about the virtue of others, fits easily within this. The tradition may have its origin in the Aristotelian idea that continence – compliance with the right for reasons independent of its rightness – can support and even give rise to virtue proper: to a compliance with right behavior that is a matter of second nature and bespeaks an attachment to the right.16 Doing the right thing for the wrong reasons may be a vice: to quote from T.S. Eliot’s ‘Murder in the Cathedral’, it may be ‘the greatest treason/To do the right deed for the wrong reason’. But if it is a vice, then it is a saving vice; it represents a form of motivation and behavior that is friendly to virtue, not inimical.17 We should not rail against the resort to institutional design, then, not even against the resort to mobilizing correctness and credulity for design purposes. It is now widely recognized that people become responsible citizens through being ‘responsibilized’.18 If they are held responsible for certain failures on a basis of more or less strict liability, and they know that this is so, then they are likely to become sufficiently attentive to ensure that any failure which occurs will actually be a matter of fault; they will be blameworthy in the ordinary sense, not in the sense of being just strictly liable.19 People may be made fit to be held responsible by being treated as if they were fit to be held responsible. The reason we should not rail against institutional design of the sort envisaged here is that what is true of responsibility may also be true of normative sensitivity. As we may make people responsible by treating them as if they were responsible, so we may make people normatively sensitive by getting them to behave, out of a love of esteem, as if they were sensitive. As we may responsibilize people, so we may equally hope to be able to ‘sensitize’ them: that is, to sensitize them to the claims of suitable values. This may sound like the expression of a government-house mentality, in which we the institutional designers think of ourselves as different from those on whom impose our designs. But that need not be so. We should each be aware of the complexity and fragility of our own motivations; none of us can be complacent, for example, about the possibility of remaining virtuous while wearing the ring of Gyges. And being aware of our own nature in that way, we should embrace the prospect of being scaffolded in our fragile commitment to relevant values. We should embrace this prospect even as we recognize that that scaffold may be the product of human design, and perhaps the product of our own designing initiatives.
16
Aristotle (1976).
17
Lovejoy (1961); Brennan and Pettit (2004).
18
Garland (2001).
19
Pettit (2001), Chap. 1.
Value-Mistaken and Virtue-Mistaken Norms
155
References Aristotle (1976): The Nicomachean Ethics, Harmondsworth: Penguin. Brennan, G. and Pettit, P. (1993): “Hands Invisible and Intangible”, in: Synthese 94: pp. 191–225. Brennan, G. and Pettit, P. (2000): “The Hidden Economy of Esteem”, in: Economics and Philosophy 16: pp. 77–98. Brennan, G. and Pettit, P. (2004): The Economy of Esteem: An Essay on Civil and Political Society, Oxford: Oxford University Press. Coleman, J. (1990): Foundations of Social Theory, Cambridge (MA): Harvard University Press. Cooter, R. D. (1994): “Structural Adjudication and the new Law Merchant: A Model of Decentralized Law”, in: International Journal of Law and Economics 14: pp. 215–231. Cooter, R. D. (1996): “Decentralized Law for a Complex Economy: The Structural Approach to Adjudicating the New Law Merchant”, in: University of Pennsylvania Law Review 144: pp. 1643–1696. Dharmapala, D. and McAdams, R. H. (2001): “Words that Kill: An Economic Perspective on Hate Speech and Hate Crimes”, in: University of Illisnouis Law and Economics Research Papers. Champaign: Urbana. Elster, J. (1990): “Norms of Revenge”, in: Ethics 100: pp. 862–85. Elster, J. (1999): Alchemies of the Mind: Rationality and the Emotions, Cambridge: Cambridge University Press. Garland, D. (2001): The Culture of Control: Crime and Social Order in Contemporary Society, Chicago: University of Chicago Press. Goffman, E. (1975): Frame Analysis: An Essay on the Organization of Experience, Harmondsworth: Penguin Books Ltd. Hardin, R. (1995): One for All: The Logic of Group Conflict, Princeton (N.J.): Princeton University Press. Hart, H. L. A. (1961): The Concept of Law, Oxford: Oxford University Press. Jones, E. E. (1990): Interpersonal Perception, New York: Freeman. Jones, E. E. and Nisbett, R. E. (1971): The Actor and the Observer: Divergent Perceptions of the Causes of Behavior, New York: General Learning Press. Lewis, D. (1969): Convention, Cambridge (MA): Harvard University Press. Lovejoy, A. O. (1961): Reflections on Human Nature, Baltimore: Johns Hopkins Press. McAdams, R. H. (1995): “Cooperation and Conflict: The Economics of Group Status Production and Race Discrimination”, in: Harvard Law Review 108: pp. 1003–1084. McAdams, R. H. (1997): “The Origin, Development and Regulation of Norms”, in: Michigan Law Review 96: pp. 338–433. Miller, D. T. and Prentice, D. A. (1994): “Collective Errors and Errors about the Collective”, in: Personality and Social Psychology Bulletin 20: pp. 541–550. Miller, D. T. and Prentice, D. A. (1996): “The Construction of Social Norms and Standards”, in: Social Psychology: Handbook of Basic Principles, E. T. Higgins and A. W. Kruglanski (eds.), New York: Guilford Press: pp. 799–829. Nisbett, R. E. and Cohen, D. (1996): Culture of Honor: The Psychology of Violence in the South, Boulder: Westview Press. Pettit, P. (1990): “Virtus Normativa: A Rational Choice Perspective”, in: Ethics 100: pp. 725–755; reprinted in: Pettit, P. (2002): Rules, Reasons, and Norms, Oxford: Oxford University Press. Pettit, P. (1995): “The Virtual Reality of Homo Economicus”, in: Monist 78: pp. 308–329; Expanded version in Maki, U. (ed.), (2000): The World of Economics, Cambridge: Cambridge University Press; reprinted in Pettit, P. (2002): Rules, Reasons, and Norms, Oxford: Oxford University Press. Pettit, P. (2001): A Theory of Freedom: From the Psychology to the Politics of Agency, Cambridge and New York: Polity and Oxford University Press. Prentice, D. A. and Miller, D. T. (1993): “Pluralistic Ignorance and Alcohol Use on Campus”, in: Journal of Personality and Social Psychology 64: pp. 243–256.
156
P. Pettit
Schroeder, C. M. and Prentice, D. A. (1998): “Exposing Pluralistic Ignorance to Reduce Alcohol Use Among College Students”, in: Journal of Applied Social Psychology 28: pp. 2150–2180. Sober, E. and Wilson, D. S. (1998): Unto Others: The Evolution and Psychology of Unselfish Behavior, Cambridge (MA): Harvard University Press. Ullmann-Margalit, E. (1977): The Emergence of Norms, Oxford: Oxford University Press. Winch, P. (1963): The Idea of a Social Science and its Relation to Philosophy, London: Routledge.
A Commentary on Pettit Gottfried Seebaß
Pettit draws attention to an important social fact. To a large extent behaviour conforming to certain standards is not carried out out of a personal attachment to those standards, if there is any such attachment at all. Rather, the real motive is a prevailing desire for the approval of others combined with a twofold belief: (1) that this approval depends on being regarded as a person attached to the relevant standards and motivated by them, and (2) that people are prone to take factual conformity by words and deeds to be evidence of this. Now, this scenario is quite common and understandable in cases of prudential individual adaptation, say of a dissenter simulating allegiance to the moral standards of the majority. So one may well ask whether there are corresponding generalized cases of social adaptation, too. Are some if not many standards present in a group or society accepted and conformed to only by some kind of general mutual mistake and hypocrisy? Pettit suggests they are. This is a disquieting and provocative idea raising a number of intriguing questions of which I shall take up three.
1 First, is Pettit right in claiming that such cases are instances or at the very least limit cases of normativity.1 Here it is advisable to distinguish between mere conformity to a norm and normative action. A bunch of apples, e.g., may conform to the agricultural norms set by the European Union. But apples certainly do not act normatively. Normative action implies intentionally trying to meet standards taken to be norms. But surely this may also be done by someone not endorsing these norms personally or being mistaken in believing that they are socially well-established. So one may easily say that trying intentionally to live up to the standards (taking up Pettit’s examples) of student drinking or military courage is normative action even if no member of the relevant groups is attached to these standards personally or does act for this reason. Surely it is one thing to ask whether a person follows a norm and quite another thing to ask why. 1
Pettit: this volume, p. 143.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
157
158
G. Seebaß
2 Secondly, there is the question under what conditions a standard taken to be a social norm really is one. In part this is a matter of conventional definition. So Pettit is right, I think, in taking a liberal stance at this point.2 Still it is necessary to face the problem of the social establishment of norms, by which I mean both their validity and their social entrenchment as well. A given norm, say a law, which is not reinforced any more and to which nobody tries to conform may be valid and existent in a juridical sense though inexistent in terms of social practice. Conversely, suspicion or knowledge that a norm is invalid may induce people not to care about it any more. Whether derived from valuations, interests or other motives: first and foremost the validity of social norms rests on the combined will of men, if even indirectly and restricted by majority rules (as in democratic legislation) or merely by way of an implicit, tacit consent to existing traditions or the normative declarations of an unquestioned common authority. A norm that is not supported by a fairly general common will loses its social validity, and to the extent this becomes public it is very likely to lose its binding force and practical meaning, too. A fortiori there will be no gain in social reputation by trying earnestly or pretending to meet it. Accordingly, such norms are instable and likely to be given up sooner or later, implicitly at least or even explicitly (as in a functioning democratic society). To be sure, this process will slow down if people do not notice that volitional support is fading or missing in the minds of other people or even within themselves. Yet it is not likely that a situation of general mutual volitional error will last for long (except in very special, e.g. pathological, cases). Even a child with normal intelligence learns to distinguish early between words and deeds backed by a serious will and those which are not. Hypocrisy may be prolonged a bit if the social climate is such that the individuals do not dare to take the first step in coming out, shunning public ostracism even from those who they know are also volitional dissenters. Nevertheless the situation remains unstable even here and may change rapidly if a few candid and courageous persons take the lead. So while I agree that there can be “value mistaken norms” in Pettit’s sense,3 I would like to insist that they are strongly dependent on an unusual kind of error and will be temporary, being ill-established in terms of validity and social entrenchment. “Virtue mistaken norms” are different. Here validity is secured by the hypothesis that there exists a relevant common will. Still it is necessary to differentiate. Take Pettit’s example of courage. “Courage” may be defined “thinly” as a personal disposition to show courageous behaviour no matter on what grounds. Courage in this sense can be endorsed sincerely by everyone even if no one is mistaken about the fact that most people are not really brave-hearts but merely act out of a strong desire not to be ostracized as cowards. Moreover, despite this disillusioning fact,
2
Ibid., p. 140.
3
Ibid., p. 148.
A Commentary on Pettit
159
courage defined “thinly” may well be entrenched as a norm and socially stable. As has been diagnosed ironically by the humorist Wilhelm Busch with respect to politeness, social norms may be well-known to be based on hypocrisy for the most part, yet valued and enjoyed by all as a game of civilization. Thus, norms of this kind are not dependent on error at all. Indeed, I think this is the regular case with most everyday norms of morals and etiquette and nearly all norms of law. Now, courage and many other virtues or behavioural norms supported by common will may be defined in a more demanding manner, requiring not only that people try to meet them in fact but that they do this out of a relevant state of mind. If so, people unqualified for this but eager to gain a good reputation may try to deceive others (or even themselves) about what is really the case with them. But as before, it is not likely that they will succeed in this enterprise generally and in the long run. Suppose, then, it is known or strongly suspected among the members of some group or society that all or most of them do not live up to these demanding standards. Surely this may lead to their successive retraction. But it need not and will not in many cases. An obvious alternative reaction is intensified moral education and habituation, a response that has long been recommended, e.g., by Aristotle and Cicero. And if this also proves ineffective for the majority one can still maintain high standards as an ideal the attainment of which is highly exceptional and therefore admired by all as supererogatory. In sum, then, although I can agree that Pettit’s suggestion is a theoretical possibility, I cannot see that there is much room for stable social norms which are dependent essentially on mutual error and hypocrisy, a result which to my mind is agreeable both to our ordinary and theoretical view of normative action.
3 Thirdly, however, isn’t it still true that many norms depend on the motive of gaining social approval? Sure it is. Gain and loss of reputation are general and strong motives, much stronger even in some groups or societies (e.g., Islamic ones) than classical standards of utility and individual interest. Therefore they are a challenge to standard contractualist argumentation. Many people – children and adults as well – are inclined to consent readily (i.e. not out of hypocrisy) to certain standards merely because they realize (without error) that these are accepted as general social norms. This is the main source of traditionalism, authoritarianism or charismatic dominion in Weber’s sense. And norms established that way can well be deeply entrenched and stable for a long time. Moreover, they can be socially valid insofar as they are supported continuously by the combined will of all or most members. Now, this validity and later on entrenchment, too, may become undermined gradually if individuals come to realize, not that they have been mistaken about the opinions of others, but that their own will and corresponding willingness to conform have grown unreflectively, unfreely or even as a result of social manipulation. If so, they are in a position to weigh their desire for social approval against other desires they have or can develop. This can, but need not normally lead to a very substantial
160
G. Seebaß
change of mind. So a more sophisticated contractarian who is not fixed on primary interests and utilities might argue that the supposed generality of meta-interests in social reputation guarantees that individuals who are sufficiently free and rational are willing to enter in “negotiations” about particular interests they would not have been willing to possibly give up otherwise. To what extent social norms are dependent on considerations of this kind, or could be based plausibly on them, is unclear to me as there are many unanswered questions. Still it would certainly be worthwhile trying out this scheme, starting from Pettit’s very helpful reference to the central role which the desire for social reputation could play in establishing stable social norms.
Political Norms, Markets and Social Capital Michael Baurmann
1 Introduction: Democratic Interests and Democratic Virtues The viability of a democratic political order depends at least on two preconditions: (i) the citizens must recognise the fundamental principles of the political constitution of this order as being in their common interest; (ii) the fundamental principles of the political constitution must be secured by an intrinsic commitment of the citizens. These two conditions are not identical and their fulfilment is not necessarily linked with each other. Whereas it is plausible that an intrinsic commitment to the principles of a democratic order can only evolve when the political constitution is in the interest of the citizens, a political constitution which is in the interest of the citizens will not automatically be supported by an intrinsic commitment to obey its principles. Citizens who are driven by purely opportunistic behaviour will disobey rules and norms if it is advantageous for them in the concrete situation, even when the general efficacy of these rules and norms is in their personal interest. In recent years the research on the impact of social capital on the working of democracy has shed new light especially on the determinants for the fulfilment of the second precondition: the focus of this research is on the societal factors which promote an intrinsic commitment to a democratic constitution and thereby contribute to the stability of its principles and norms. This research has provided a lot of evidence that widespread social networks and well-functioning private associations in a vibrant civil society are necessary foundations for the development of essential civic virtues such as the readiness to participate actively in the democratic process and to contribute one’s share to those public goods which cannot be provided by formal institutions.1 However, social capital theory is weak in two respects: first, the exact mechanism by which private associations promote civic engagement and democratic commitment is not yet fully understood. It is in particular unclear which forms and variants of social capital produce the desirable effects and which do not. Second,
1 Cf. Banfield (1958); Putnam (1993, 2000, 2001); Newton (1997); Braithwaite and Levi (eds.), (1998); Warren (ed.), (1999); Baurmann (1999, 2000a, 2002b); Fukuyama (2000b); Ostrom and Ahn (eds.), (2003).
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
161
162
M. Baurmann
social capital theory has not yet addressed the first and more fundamental problem that the evolvement of an intrinsic commitment to democratic political norms is hardly possible at all if the members of a society do not recognise a democratic constitution as being in their genuine interest. Instead, social capital theory takes it more or less as granted that a democratic political order is in the interest of the vast majority of the population of all modern societies. But this is not self-evident. Actual developments in the world remind us that identifying one’s interest with a democratic political order is far from automatic. In my paper I will argue that social capital theory has the potential to make progress in both dimensions. It can advance the analyses of the varieties of social capital and of the different impact of the variants for the democratic process. It can also help to answer the question as to which factors are relevant for ensuring that a democratic political order is in the common interest of the members of a society. The potential of social capital theory in regard to the second problem is connected with the fact that the shape and distribution of political interests in a population is significantly influenced by the social structure of a society. A society with deep cleavages between certain classes or groups, a society which is dominated by a powerful aristocratic or clerical elite or a society which is divided by unsolvable ideological or religious conflicts is not favourable to the emergence of a strong and unquestioning sense of common interest in a democratic constitution which guarantees pluralism und liberalism. An important part of the social structure of any society is the amount and form of social capital existing in that society. Therefore social capital is also a genuine part of the factors that are relevant for the kind of political interests which are dominant in a society as a whole as well as in its different groups and parts. On this view social capital is not only a resource by which individual and collective interests can be realized. It is also a social force that to a great extent shapes individual and collective interests. Social capital is not only relevant to making democracy work, but also to making democracy first of all a political order which is in the common interest of the members of a society. In this regard it would be not very favourable if, for example, a society is characterized by a multitude of organized pressure groups which compete for political power for the sole purpose of redistributing economical wealth in their own direction, if separated ethnic or religious groups fight for political and cultural supremacy, or if functioning social networks are only available to a small elite which dominates an isolated and powerless population. If in a society dense social networks are centred solely around competing groups and exclude access to these networks for outsiders then this form of social capital will create structures in which the respective group interests will not so easily harmonize around a democratic constitution. A democratic political order has many features which potentially can come into conflict with the interests of some or even the majority of the members of a society. But if we look especially at the principles of a liberal democracy in a modern society then one characteristic of a democratic constitution seems to be particularly significant. The rules and norms which structure the institutions in such a democracy share an important property: they incorporate the principle of political and legal equality. The citizens of a modern, pluralistic democracy despite their many
Political Norms, Markets and Social Capital
163
potential differences in culture, ethnicity, religion, wealth or abilities enjoy the same fundamental rights and privileges. They are all included within the set of beneficiaries of the constitutional order. It is a central feature of the political principles and norms of a liberal democracy that they are universalistic. They consider the interests of all citizens with equal weight and they are applied to all members of a society in a non-discriminatory way.2 From this fundamental quality of a modern democracy it follows that an interest in the existence of such a political order must include as one element an interest in the existence of universalistic political norms. That a democratic political order will be in the interest of all citizens, therefore, does not only presuppose that every citizen has a personal interest in the benefits of the political and civil rights which are typical of a democratic society, but that every citizen also has a personal interest that all the other citizens also should enjoy the benefits of those rights. Each and every citizen must have a personal interest in universalistic political norms which do not discriminate between individuals and groups, but treat all citizens the same – despite their cultural, religious, ethnic, economic, or intellectual differences. This demand can lead especially in a pluralistic and heterogeneous society to a considerable tension between the two preconditions of a working democracy: an intrinsic commitment to norms may be easier to produce and to maintain if there is a direct and recognizable connection of these norms to the homogenous interests of a certain group of individuals. But if a democracy demands an acceptance and commitment to norms which promote indiscriminately the interests of “all” in a pluralistic society with many heterogeneous interests then such acceptance and commitment may appear as a sacrifice one has to make for people with whom one has no direct social bond. So if both conditions are met we may face a fragile constellation that is dependent on an equilibrium of many interrelated factors. It seems that in this bundle social capital plays a crucial role. The question whether the members of a society develop a genuine interest in a political order with universalistic norms is apparently largely influenced by the social relations and networks in which the individuals are embedded. Commitment to the norms and rules of a society is learned through participation in social capital. This is the lesson social capital theory teaches us. But if this learning process should include the commitment to universalistic norms and rules then we must be aware of the fact that not any variant of social capital will do the job. We must take a closer look at which kind of social capital is relevant here. This is the question I will discuss in my paper. I will start with a short recapitulation of the arguments which speak in favour of the relevance of civic virtues und intrinsic commitment for the working of democracy (Section 2) and why the social capital of a society is a main factor in the production of such virtues and attitudes (Section 3). After this I explain that it is of central importance to distinguish between different variants of social capital (Section 4) and that the most relevant dimension for classification is the distinction between
2
Cf. Baurmann (1997b).
164
M. Baurmann
universalistic and particularistic social capital (Section 5). In the conclusion I argue on this basis for the thesis that in order for social capital to become a supportive factor for democracy it is crucial that it is embedded in a strong market-economy (Section 6).
2 Economising on Virtue or Taking Virtues Seriously? Virtue and morality are scarce goods. It is therefore expedient to be sparing with them. This principle of “economising on virtue”3 not only applies to the relationship between single individuals but, as the Scottish moral philosophers have taught us, should also be a guideline for the creation of social institutions. The market serves as a paradigmatic example of an arena where the participants’ virtues and morals are largely dispensable, and yet where the result of their actions serves everyone’s interest and, thereby, the public welfare. Institutions of this kind relieve individuals of the burden of moral duties and reduce the need for moral norms as well as for investments to enforce them. The classical authors of the Scottish Enlightenment were optimistic that this principle could also be transferred to political norms and institutions. Even within the difficult realm of state power, it seemed possible to invent institutions through which an “invisible hand” would aggregate the general pursuit of individual interests to a common good.4 This prospect was particularly attractive, as one could discard the – possibly futile – Platonic task of controlling the personal ambition of state rulers by instructing them in virtuousness and morality. If, instead, there were ways of shaping the institutional framework of political action so that it would be to the rulers’ own advantage to take care of their subjects and the common weal, then trust in politics would become independent of trust in the character of the politicians. In this case the particularly difficult task of instigating moral norms for politicians would be superfluous. The hope of being able to rely on the “morality” of the political institutions rather than on the morality of the politicians still plays a prominent role in modern political science and social theory and, moreover, in public opinion too. Especially the modern democratic state with its institutionalized possibility of voting politicians out of office, its protection of basic rights, and its ingenious system of the separation of powers and “checks and balances” seems to be the perfect example of a system, which by means of cleverly constructed mechanisms, prevents state rulers from misusing their power for their own private aims. In recent years, however, the insight has grown among social theorists that the principle of “economising on virtue” has its limits and that we cannot solve all the problems of social and political order by well-designed institutions and their
3
Cf. Brennan (1995).
4
Cf. Hirschman (1977).
Political Norms, Markets and Social Capital
165
incentives.5 This is especially true of a democratic society. The functioning of the political norms of a democracy is, to a large extent, not only dependent on the behaviour of politicians or civil servants acting directly under the rules of state institutions, but more on the attitudes and the spontaneous behaviour of the citizens outside formal institutions. Many social scientists today believe that because of this democracy must be rooted in genuine civic virtues and commitment which cannot be traced back to rational opportunistic behaviour under some artificially created extrinsic incentives.6 Civic virtues and commitment seem to be especially important in regard to three core-areas of the democratic process: I. Public deliberation. II. Political participation. III. Collective decisions. It is scarcely conceivable that a democracy can work well if all citizens would only act as purely opportunistic actors in these arenas. Public deliberation about common issues demands informed participants who are ready to invest time, energy, and good will. Deliberative processes concerning political issues will be more efficient the more the participants are motivated to argue according to general standards instead of solely promoting their individual interests. Active participation in political processes presupposes the motivation to contribute to public goods both in the context of individual action and in the context of collective action. Collective decisions in democracies must be protected against a tyranny of the majority on the one hand and, on the other hand, accepted and observed by the minority. Both demands imply commitment to the constitutional order, to political norms, and to substantive ethical principles.7 However, there can be no doubt that institutions and the incentives they create matter and that different institutions will produce different outcomes. Institutional rules influence the behaviour of actors inside and outside the institution – direct democracy, for example, has consequences for the behaviour of voters and politicians which differ significantly from the consequences of representative democracy. But the effects of institutional design are dependent not only on the properties of the institutions themselves. Every institution is embedded in a social environment and the overall impact of an institution is not the result of an endogenous equilibrium produced only by the incentives of the institution and the given preferences of the actors. This impact is always a result of an equilibrium which emerges from the characteristics of the institution and exogenous forces and conditions. So the same institutional system can have very different outcomes depending on the social
5
Cf. Baurmann (2000b).
6
Cf. Putnam (1993); Fukuyama (1995); Pettit (1997); Warren (ed.), (1999); Brennan and Lomasky (1993); Brennan and Hamlin (2000); Cook (eds.), (2001); Dekker and Uslaner (eds.) (2001); Brennan and Pettit (2004). 7
Cf. Buchanan and Congleton (1998); Baurmann (2003).
166
M. Baurmann
context in which it is implemented. The “rules of the game” always include more than deliberately created rules of institutions. “Design principles” for institutions are clearly relevant for institutional stability and performance – but their exact consequences are not context-independent.8 Emphasizing that civic virtues are important for making democracy work is therefore not tantamount to assuming that institutions and institutional design are irrelevant. To some degree the opposite is true: civic virtues do not make institutions superfluous, but can serve as a basis for making institutions even more successful. We can reach more efficient equilibria by institutional devices if we can trust in the virtues and intrinsic motivation of the actors: it becomes easier to create and change institutions, the demand for hierarchy and control in institutions decreases, the tension between formal und informal institutional processes diminishes, institutional norms and rules are more readily followed and the commitment to collective decisions under institutional rules increases.9 If the outlined thesis is right, the working of democracy and the viability of its political norms demands a stable equilibrium between proper institutional design and a suitable social environment in which supporting civic virtues play a central role. Institutions can bring about a lot of things – but whether they do so in a desirable way is greatly influenced by factors outside the institutions themselves. The efficiency of democratic institutions, their stability, their legitimacy and conformity to their norms and rules can only be realized if they are properly implanted in their social soil. It is true that societies can be changed and shaped with the help of institutions, but how successful this is and what kinds of institutions are necessary cannot be answered in general terms. We cannot simply replace the moral fabric of a society and its spontaneous forces by the incentives of a cleverly designed institutional framework. The working of a democracy cannot only be based on extrinsically motivated compliance with formal rules but also requires an intrinsically entrenched commitment to fundamental political norms and substantial ethical principles: we have to take virtues seriously!
3 Bowling Together: Democracy and Social Capital The view that civic virtues are essential prerequisites of a stable political order and a good government has a long history. The same is true of a family of theories about the factors which promote the desired virtues in a society. These theories, which go back to Aristotle, were ingeniously renewed in Tocqueville’s analysis of democracy in America and in our time have been put in the context of social philosophy by the communitarians.10 In the last 10 years, however, a new and promising variant of these theories has been developed by the political scientist Robert Putnam in his 8
Cf. Ostrom (1990); Ostrom and Ahn (2003).
9
Cf. Baurmann (2002a).
10
Cf. MacIntyre (1981); Etzioni (1993).
Political Norms, Markets and Social Capital
167
pioneering books Making Democracy Work and Bowling Alone which initiated a large number of theoretical and empirical studies on the social and cultural fundaments of democracy.11 Put in a nutshell, these theories share the assumption that civic virtues are the product of a particular sort of social relationship between the members of a society. According to this assumption these relationships constitute a special area of a “civil society” whose dynamics are rooted in the aspirations and values of the citizens as private actors. As participants in this kind of private relationship people will develop capacities and dispositions which are beneficial to the society as a whole and will spill over into the public sphere. To Aristotle this function is fulfilled by friendships which motivate individuals to behave altruistically towards each other and to jointly promote the values of their community. Tocqueville extended Aristotle’s view to include all personal relationships which are part of a collective enterprise that people privately and voluntarily initiate to realize a common aim. From his observations he draws the conclusion that by taking part in such associational groups, individuals will overcome short-sighted egoism and will learn to contribute to collective goods, trust each other, and discuss and peacefully solve issues of common interest. For Tocqueville the concrete aims, sizes, and structures of associational groups are secondary. Whether they are established to build a bridge for the village, to come together to pray or to collect money for an opera they will all have beneficial influences on the behaviour and character of their members turning them into virtuous citizens who feel responsible for the common welfare. Whereas the communitarians in some respect go back to Aristotle in emphasizing the importance of common values, uniform convictions and shared traditions in a society as a whole as the basis of civic virtues, the modern theories in the political and social sciences are more in the spirit of Tocqueville focussing as they do on the variety and diversity of associational activities. They have coined the term “social capital” to summarize the different forms of association civil society can produce through the private initiative of the citizens. The exponents of social capital theory believe that there are manifold kinds of social relationships which – although, maybe to different degrees – have the capacity to create those special bonds between their participants which promote the development of civic virtues: from the weak ties of loose social networks in neighbourhoods, from bowling and bird-watching, soccer-clubs and bible-circles to political parties, NGOs and spontaneous social movements. Social capital theory assumes that the varieties of civil society provide the most important resources for making democracy work because only in small settings can people learn what is relevant for the society as a whole. Informed deliberation, active participation, and producing and accepting collective decisions are as important for the working of a small private association as for democracy in its entirety. Without being able to overcome the free-rider problem and act successfully as a collective,
11
Putnam (1993, 2000).
168
M. Baurmann
being fair towards minorities and feeling committed to the rules of a group, most joint enterprises would not get off the ground. So at the heart of modern social capital theory is the link between the vibrancy of different forms of associational life and a high level of civic engagement and democratic participation. Social capital theorists name at least three social phenomena which they view as typical results of a flourishing civil society and which are assumed to be directly connected with the development and reinforcement of civic virtues: I. Social networks of strong and weak ties. II. Norms of reciprocity and trust. III. Commitment to common aims. According to this view, social networks are not only important to provide individuals with access to different kinds of valuable resources.12 Networking also teaches the virtue of sociability and the capacity to create and maintain cooperative, friendly and sometimes even altruistic relationships. Effective norms of reciprocity and trust promote the virtue of carrying out exchange relations and of maintaining cooperation under risk and uncertainty.13 Commitment to common aims embodies the virtue of not behaving as a free-rider but contributing to collective goods even when the individual contribution is marginal and insignificant.14 It is a plausible assumption that virtues like these are, in principle, also of a high value for democracy as a whole. The quality of public deliberation, political participation and collective decisions will improve if people are connected by the ties of encompassing social networks, practice mutual reciprocity, trust each other and feel an intrinsic commitment to common aims. The crucial premise of social capital theory is, however, that there is indeed a spill-over, a transfer from the context of the privately organized associational life to the society as a whole: the virtues which are learned in the context of 10 will be generalized to the context of 10 Million! This assumption is dependent on two separate premises: the first presupposes that personal attitudes and dispositions are better learned in small groups than in (very) large groups; the second implies that what is learned and developed in small groups to the advantage of these groups and their members will keep its positive impact in the context of large groups and for the society as a whole. Whereas the first thesis seems reasonably plausible, the second needs further clarification and explanation. The exact mechanisms by which membership in associations of civil society leads to a high level of civic engagement and high quality democratic politics are not yet clearly understood. We must get more insight into which forms and elements of private associations promote the desirable transfer and which do not. Of course, there is undeniable empirical evidence that there are important differences between
12
Cf. Coleman (1987, 1988); Granovetter (1973, 1985).
13
Cf. Gambetta (eds.), (1988); Misztal (1996); Fukuyama (1995); Woolcock (1998); Woolcock and Narayan (2000); Dasgupta (2000); Lahno (2002); Uslaner (2002).
14
Cf. Baurmann (2000a, 2002b).
Political Norms, Markets and Social Capital
169
various forms of social capital in this respect and that not every jointly celebrated Bible or Koran study is conducive to democracy.
4 Making Democracy Worse: the Dark Side of Social Capital Timothy McVeigh and his co-conspirators in the Oklahoma City bombing were members of a bowling league: they were not, unfortunately, “bowling alone”.15 Osama Bin Laden is not acting as an isolated mad man, but is firmly embedded in a well-functioning network of internationally acting terrorists. These extreme examples make clear that being intrinsically devoted to a common aim, developing trust relationships and overcoming free-rider problems by membership in some kind of group is not automatically desirable for people outside the group or the political order of a society. The public good for the group could be a public bad for the community. Even when we think of less dramatic possibilities than in the Oklahoma City bombing or in the case of Al Quaida: differential mobilization of the population by ethnic, racial, religious, or other ascriptive criteria can lead to very particularistic demands and will undermine, rather than support democracy.16 A rich network of civic activities alone is no guarantee of a flourishing democracy. It can be both a source of trust and a source of distrust. Instead of promoting the recognition and realization of common goods it can produce insurmountable conflicts by shaping and organizing antagonistic interests and locking them in an inextricable equilibrium of continuous power struggle and mutual hostility. So if we want to have insight into the potentially positive relationship between democracy and social capital, we have to learn more about the special kind of social capital which is necessary here. Putnam claims as a central result of his studies in Italy that the malfunctions of democratic institutions in Southern Italy were chiefly a consequence of a low level of social capital.17 To generalize this correlation would be misleading. It is not the case that societies with no democracy or with a malfunctioning democracy always display a low level of social capital. The stability of autocratic and despotic regimes often has two faces: on the one side there may exist a fragmented civil society in which more or less isolated individuals live within weak social networks and must endure an underdeveloped associational life – a situation which is often the intentional outcome of a political strategy of the rulers who want to prevent the emergence of a strong civil society. But on the other side the members of the ruling oligarchy themselves may be integrated in a social and political network which guarantees a sufficient degree of mutual trust and reciprocity inside the political elite to enable the efficient realization of their collective goods. On this basis the commitment among them can be strong enough to overcome short-term opportunistic and
15
Cf. Levi (1996).
16
Cf. Hardin (1995).
17
Cf. Putnam (1993).
170
M. Baurmann
selfish behaviour and achieve beneficial cooperation – which does not exclude the fact that the aim of this cooperation is to suppress and exploit the rest of the society. Instead of being supportive of democratic processes, high levels of social capital can also be a difficult obstacle in the transitional phase from traditional societies to modern democracies. Afghanistan and Albania, for example, are not societies with an especially low level of social capital. In both societies there are at least partially well-functioning social networks, relations of trust and reciprocity, and the capacity for collective action embodied in traditional structures of families, kinship, clans, and tribes – all of them embedded in a highly respected social and religious tradition which contains values and norms with a considerable degree of legitimacy. The problem for democracy here is clearly not a problem of lacking social capital – the problem is the lack of the right kind of social capital. Moreover the problem is worsened by the fact that the “wrong kind” of social capital not only prevents the development of civic virtues which are necessary for the democratic process. The wrong kind of social capital can also structure the society in a way that a democratic political order is not at all in the common interest of the people. Established traditional institutions can secure particularistic rights and privileges which would be removed by the universalistic norms of a democratic political order. Social capital is not only a relevant factor for the existence and distribution of civic virtues and intrinsic commitment in a society, but also for the existence and distribution of certain political interests. It both helps (and hinders) democracy working, but also helps (and hinders) democracy operating as a public good! We have to acknowledge that social capital can have a dark, even sinister side.18 Networks, reciprocity, trust and altruistic behaviour are good only in the right context. Indeed, some of the communities that have been able to educate their members successfully to behave unselfishly and to sacrifice their individual interests to the common cause are responsible for the largest catastrophes in the history of human mankind. Equally, democratic suboptimality is not always combined with low levels of social capital – as is maybe the case in Southern Italy. Even there the low level of social capital in one area is not by chance connected with a high level of social capital in another area: the Mafia is a form of social capital, it embodies a highly efficient social network, creates strong norms of trust and reciprocity, and very successfully overcomes collective action problems of all sorts.19 However, it is not necessary to refer to the Mafia to prove that well-developed forms of social capital could make democracy worse instead of making it work. Nepotism, corruption, rent-seeking, partisanship, or free-riding are all forms of behaviour which are detrimental and destructive to democracy. And they are all forms of behaviour which are, as a rule, more successful if carried out in a group as a collective enterprise than as an individual effort.20 Therefore we find that empirically 18
Cf. Hardin (1995, 1999); Levi (1996); Portes and Landolt (1996); Adler and Kwon (2000).
19
Cf. Gambetta (1993).
20
Cf. Baurmann (2005).
Political Norms, Markets and Social Capital
171
all manner of social capital is built around those activities – ranging from loose networks which bring a few people together for a short period of time, to small associations with horizontal relations between members connected by trust and reciprocity, up to large organisations with formal rules and a strict hierarchy. The more developed and the more efficient those forms of social capital are, the worse for the outsiders and for democracy as a whole. The dark side of social capital is not always connected with obviously condemnable behaviour like trying to free-ride, bribe, or seek rents at public expense. Negative externalities where social capital is used to facilitate collusion among a group can also be generated when particularistic demands are put forward which cannot always be judged as morally wrong at the outset. Mobilization of people to realize their religious visions or to promote the interests of their race or ethnicity can be rooted in moral convictions and personal virtues and can create social capital in a paradigmatic form. Groups and associations like these will often embody dense social networks, high levels of personal trust, altruistically driven reciprocity and generosity, and a strong intrinsic motivation to make sacrifices for the common good. These forms of social capital will, nevertheless, more often subvert rather than strengthen democracy. The reason for this is obvious: associations like these are not “bridging” and “outward-orientated”,21 but centred around people of the same kind or origin and promoting goods which are exclusively valuable to the members of the group. The more successful these associations are the less their members will have the incentive to cooperate and bargain with other groups on a common basis, but will see the chance to enforce their particularistic interests at the expense of other. Thus social capital in this variant erects barriers of mistrust between people instead of uniting them and contributes to aims and goods which can very easily conflict with the aims and goods of the society as a whole. Associational groups of this kind will trigger a vicious circle because they undermine shared interests in a society and thereby create incentives for other groups – who, by themselves, would have no genuine reason to develop in this way – also to concentrate exclusively on their members and their particularistic interests. Even if an association and the activity of its members have no negative effects on the surrounding society, it is not easy to answer the question under which conditions they will have positive effects. The assumption that membership in one kind of group leads to overcoming free-rider problems in another is not self-evident. The causal chain between bird-watching and political activism is not very close. There is a wide gap between various kinds of social clubs and organizations for political action. A number of comparative studies on social capital of recent years suggest indeed that the kind of social capital that is possibly typical of an “individualistic” society – informal activities and “events” which are restricted to single issues and
21
Cf. Putnam and Goss (2001).
172
M. Baurmann
do not demand a lasting commitment – may hardly produce any positive spill-over into other areas of civic or political engagement.22 To make the theory of social capital more precise regarding this crucial point, it is necessary to identify the characteristics of the associational groups that are valuable to society and democracy as a whole.23 Three modifications to our previous list seem to be important in this respect. To contribute to a commitment to the political norms of a democratic order and to promote the overall democratic process social capital should: 1. create bridging social networks; 2. establish norms of generalized reciprocity and generalized trust; 3. promote commitment to public goods. Social capital can have a dark side because, contrary to these demands, it can embody networks which are not bridges between different kinds of people bringing them together to promote joint interests, but are tools of separation erecting borders and barriers, and providing an exclusive resource to a special group. Instead of encouraging reciprocal and trustworthy behaviour beyond the confines of a group or association, social capital can contribute to a restriction of reciprocity and trust and lead to an increase of opportunism and distrust outside the respective groups. Social capital can lack positive spill-over effects because it only promotes commitment to the “club” good of a group rather than to the public good of the surrounding society as a whole.24
5 Mafia and Mazda: Particularism and Universalism Social capital is always embodied in social relationships which do not encompass a society as a whole – at least in societies with a large number of members. This was already true for the Aristotelian polis: not every citizen of Athens could be a friend of everyone else. The circle of friends is always limited. Therefore Aristotle’s idea that friendship was important for the community of Athens already presupposes that people are formed in their behaviour and character by their intimate personal relations in a way that is also beneficial for people with whom they are not befriended. The same point is central to Tocqueville’s theory: associational life in America could foster democracy because in small settings people learn social skills and adopt virtues which are also of central importance to their behaviour as citizens of the “big” country as a whole. Equally the communitarians or the social capital theorists must rely essentially on the thesis that members of communities or associations are shaped in their behaviour in a way that is advantageous beyond the group of which they are members. So if social capital be deemed useful for society 22
Cf. Putnam (ed.) (2001).
23
Cf. Stolle and Rochon (1998); Curtis et al. (2001); Warren (2001); Paxton (2002).
24
Cf. Stolle (1998).
Political Norms, Markets and Social Capital
173
or democracy as a whole, the crucial question is not only how networks, reciprocity, trust or commitment might develop, but how bridging networks, generalized reciprocity and trust, and commitment to genuine public issues can be developed and maintained. We can summarize the crucial differences by referring to the alternative between particularism and universalism. A group is all the more particularistic, the more its networks, its norms of reciprocity and trust and its aims are confined to the members of the group, whereas a group is all the more universalistic, the more its networks, its norms of reciprocity and trust and its aims transgress the confines of the group and encompass other citizens and groups in a society.25 The different qualities of the respective forms of social capital can be illustrated by two paradigmatic examples which represent extremes on the continuum between particularism and universalism: the Mafia and the international company Mazda. The Mafia creates two sorts of social networks: one is strictly confined to the members of the group and is carefully isolated from outsiders. As a rule, it is important that the members of the in-group share essential personal traits and already belong to a common social framework: the same family, kinship, clan or village (“Corleone”). The Mafia tries to utilize the already existing “strong” ties between them as building blocks for the Mafia-network. The very function of this network is to offer social capital only to the members of the “family” and to form a constant threat to outsiders. This kind of network does not connect different sorts of people in a mutually beneficial way, but is rather an instrument to divide people and to produce benefits for one group at the expense of the other. The same holds true for the second sort of social network which is created by the Mafia: this network is designed to guarantee a reliable contact to external allies, helpers and victims of the Mafia. Although this network transgresses the confines of the organization, it does not help to create social bonds or reciprocal exchange between insiders and outsiders. It is also only a tool of suppression and extortion and thus does not help to create but to destroy forms of a spontaneously emerging civil society. Moreover, the Mafia has a manifest interest to undermine all kinds of social networks in its environment which are not controlled by its own forces. Norms of reciprocity and trust play a key role in the fabric of the Mafia. Its “code of honour” is legendary and compels the members of the Mafia to adhere to a strictly enforced normative order. It is no contradiction in this respect that reciprocity and trust inside the Mafia can become very fragile. But the main thing here is the fact that, like the Mafia-created networks, the Mafia-engendered reciprocity and trust and their potential to promote benevolent social relations are also strictly confined to the Mafia-members and designed to exclude outsiders. That is, of course, a trivial consequence of the fact that the Mafia acts illegally and secretly and that its aim is not to cooperate fairly with other people but to use its power to exploit and suppress them. Therefore its internal reciprocity and trust are not only beneficial to the Mafia itself but also hinder reciprocal relations outside the Mafia thereby producing
25
Cf. Baurmann (1997b); Baurmann and Lahno (2002).
174
M. Baurmann
massive mistrust in the social environment. This process likewise creates an artificial demand for trust which, in turn, is perversely traded by the Mafia, thus reinforcing the problem by means of a vicious circle. Last but not least is the common good for the Mafia a public bad for the society. Insofar as the Mafia successfully promotes personal virtues and commitment to overcome collective goods problems, these dispositions are not beneficial to public goods in general. The virtue of a Mafia member must be a strictly biased virtue which is solely orientated towards the particularistic good of the Mafia. Benevolence towards the Mafia and its members must be combined with hostility towards people and potential victims outside the Mafia. It is conceivable that the Mafia could restrain opportunistic behaviour among its members to a certain extent and in this respect actually produce “real” virtues based on genuine intrinsic motivation. This however would not be an advantage for the society. The more effective the Mafia is in producing particularistically orientated virtues and commitment, the more effective it will be as a collective and the more harmful to all others. It should be clear that an established Mafia-organization not only promotes virtues and commitment of a kind which are not desirable for a democracy. If the Mafia is powerful enough to serve the interests of its members successfully the Mafiosi will not belong to the group of people who naturally develop a special interest in the universalistic principles of a democratic order. Universalistic principles of peaceful cooperation and equal rights are not in the interest of individuals who belong to a powerful collective that realizes its aims and aspirations by force and fraud. Now let’s look at the other side of the continuum. Mazda, as a worldwide operating company, also creates two sorts of social networks: one is the firm-internal network which embodies special ties between the members of the company. But unlike Mafia membership, Mazda membership is not restricted to people who already belong to a common social framework and possess identical personal traits of kinship, origin, or culture. In contrast, Mazda exemplifies social capital with a high degree of “bridging” qualities. That means that members of the Mazda-“family” include people from different social backgrounds, nations, ethnicities, and races uniting many of them for the first time. The company-network of Mazda is highly inclusive and serves as a device to overcome manifold differences between people in order to create a mutually beneficial cooperative relationship. The same applies to the second sort of social network which is established by Mazda. This network serves to facilitate a reliable contact to business partners, consumers, competitors, public administrators, or politicians. This network transcends the confines of the company and must encompass all sorts of people in a myriad of social contexts. Because the network is used to build cooperative relationships, it fosters social bonds and beneficial exchange between the participants. By its weak ties it creates a kind of skeleton on which the flesh of a civil society can grow – far beyond the borders of the company itself. Norms of reciprocity and trust are as fundamental to a company like Mazda as they are to the Mafia. But there is an essential difference: the norms of reciprocal and trustworthy behaviour in a company cannot be particularistic in the sense that
Political Norms, Markets and Social Capital
175
they only include people of a narrowly defined nature. Mazda as an internationally operating company is dependent on its capacity to employ a wide range of people, whether they are of different nationalities, cultures, races or social classes. It is of vital importance to Mazda that its chances to employ persons despite their social or cultural diversity will be enhanced and protected by their inclusion in the norms of reciprocity and trust. The same applies to persons outside the company, who as consumers or business partners, are relevant to the company’s ends. In regard to them too, Mazda is interested in reciprocal and trust-based cooperative relationships. Under these conditions, when looking for suitable employees, a company like Mazda has good reasons not to look for people who practice their virtues only with respect to a particular group of persons, but for people who are disposed to behaving reciprocally and in a trustworthy manner generally – and therefore Mazda also has good reasons to promote norms of reciprocity and trust with universal scope. The common good for the Mafia is a public bad for society and vice versa. Mafia-style personal virtues and commitment are therefore not desirable for the community as a whole. The common good for Mazda is not simply identical with the common good for society either. But the common good for Mazda is not dependent on creating a public bad for the rest of society – as in the case of Mafia. Producing public goods for society as a whole is not threatening for Mazda. On the contrary, in regard to many genuine public goods – secure property rights, political stability, sound politics, rule of law, efficient public administration – companies like Mazda belong to the group of direct beneficiaries of these goods. Thus Mazda does not have incentives to promote intrinsic motivation and commitment to overcome collective action problems solely in regard to company-specific goods. Of course, a specific loyalty to the aims of the firm will be demanded. But this kind of loyalty is not necessarily combined with disloyalty to the common welfare and is not discarded by others who themselves show a special loyalty to some other groups or associations. There is no reason for a company like Mazda to promote a strictly particularistic commitment to the interests of the firm alone. As the case of the Mafia makes clear, such particularistic commitments have their own risks for the beneficiaries because they can be easily shifted from one sub-group to the other. If this analysis is accurate then companies like Mazda incorporate social capital which promote virtues and commitment of a kind which are desirable for a democracy and support the stability of its political order. Mazda-like social capital also shapes the political interests of its members in a way that is fundamental for the existence of a durable democracy: the members of a successfully operating firm will realize their aims and aspirations by exchanges and transactions in the market-place and not by force of arms or political power. Therefore, in contrast to the members of the Mafia, the members of Mazda will belong to the group of people who naturally develop a genuine interest in the universalistic principles of a modern democratic order and the rule of law. These principles guarantee peaceful cooperation and equal rights as necessary preconditions for an efficient and expanding market. The examples of Mafia and Mazda illustrate by means of extreme cases the mechanisms by which social capital may produce detrimental or beneficial spill-over effects on the surrounding society depending on its place on the continuum from
176
M. Baurmann
particularism to universalism. The more particularist associational groups are in regard to their networks, norms, and commitment, the less they will contribute to social relations and personal dispositions serving the whole society. Particularist variants of social capital tend to be dangerous for the rest of the society and undermine political stability and democracy. The more universalistic social capital is, on the other hand, the more it will produce networks, norms and virtues that will not only serve the purposes of its direct beneficiaries, but also shape and promote a general interest in democratic political order.26
6 Bowling, Bombing and Booming: Why Markets Matter The vast majority of social capital theorists hesitate to consider market relations and firms as variants of social capital relevant to the vitality of the democratic process. This seems to be a serious shortcoming as it is obvious that networks, norms of reciprocity and trust, and commitment to common aims play an essential role in the functioning of market exchanges and for the personal relations in firms. It is not true that reciprocity, trust and commitment in economic relations are just parasitic on the production of these virtues in other social areas. Closer inspection reveals that the relations of people in economic transactions embody endogenous forces to produce reciprocity, trust and commitment. This is true both of market exchanges as such and for relations inside commercial organizations. Indeed, the idea that the kind of highly universalistic social capital of special relevance to society is predominantly produced in economic contexts has much force.27 On the other hand, the general neglect of “economic” social capital by the social capital theorists needs to be taken seriously. They make clear that social capital in the non-commercial civil society can not be entirely replaced by social capital in the economic realm. Mazda is not enough. The bird-watchers are still indispensable. There are a couple of aspects which are relevant here. Firstly, firms are not usually created by their employees to realize their personal aims, but by an entrepreneur. Secondly, the professional and social relations in a firm are not egalitarian, but more or less hierarchical. Thirdly, the aim of a company is not to produce a public good, but to make profit in the marketplace. These three typical features of firm-related social capital are assumed to limit its positive side effects for society as a whole. The fact that the employees of a firm act as agents on behalf of a principal at best creates a kind of indirect loyalty to the aims of the firm. Such loyalty is mediated by a contractual relation and triggers virtues like honesty, reliability, or a sense of duty. What is missing is the special virtue which is embodied in the commitment to a self-chosen aim and a self-organized association to realize this aim. Additionally, acting in hierarchical
26
Cf. Baurmann (1997b, 2000a, 2002b).
27
Cf. Baurmann (1997a, b, 2002b).
Political Norms, Markets and Social Capital
177
structures does not create mutuality and equality of participation and so weakens reciprocal exchange and the possibilities of developing trust relations. It is assumed that relationships within vertical networks of hierarchy and dependence are not able to create experiences of reciprocity and trust to the same extent as relationships in horizontal networks which bring together agents of equivalent status and power.28 Last but not least the production of public goods is not the aim of firms and companies. This means that commitment to common aims is primarily restricted to the “club” goods of the firm. The virtue to contribute to a genuine public good and the even more demanding virtue to contribute voluntarily and out of altruistic concern for the individual goods for others who are in need – the virtue of “solidarity” – plays no significant role in this context. So what seems to be true in regard to these arguments is that the democratic process can indeed gain considerably if social capital is developed (also) in the context of a civil society where self-governance and self-organization is independent of commercial motives and where people are sometimes ready to provide voluntarily individual or public goods for others and where they create reciprocity and trust in egalitarian relationships. It is plausible that these qualities cannot so easily be produced in the marketplace and in commercial business. As bird-watchers we do not realize our aims in competition with others and are not dependent on the good will of an authority. By watching birds with others we do as we like and do not perform tasks which are defined by someone else. It is no less important that civil society be embedded in a flourishing market economy. Universalism is learned, valued, and enforced by the expansion of economic exchanges. Only if civil society is embedded in markets can it be prevented from collapsing into particularism und isolationism. Bowling without booming may lead to bombing. A market economy creates a unique environment for the associational groups of civil society. In a society with a vibrant market economy people are not restricted to stable, strictly limited communities and to relationships based on personal ties. Social groups are flexible and “osmotic”, their membership fluctuates, the boundaries and composition of groups are constantly shifting. Instead of continuity and stagnation, there is private, social, political, economic and geographic mobility. A modern market society engenders non-local, supra-regional cooperative interests. With the expansion of market relations and a well-developed division of labour, group confines become permeable. Social groups and communities are no longer isolated from their environment, their members are not bound to each other by unbreakable ties. The lack of static social ties leads to a relatively frequent change of partners in cooperative ventures. In such an “unbound” society, one cannot rely on temporarily existing barriers and boundaries. Often, market-societies with a great number of members, with prosaic and impersonal relations between people, with frequent fluctuation between social groups and communities, are seen as destructive powers undermining and dissolving once intact
28
Cf. Putnam (2000); Offe and Fuchs (2001).
178
M. Baurmann
communities and personal relationships. But societies of this sort also give rise to the phenomenon that people, who at first have little in common, can come into contact and establish relations. They encourage cooperation and association regardless of racial, national, social, or cultural differences. Only when this is the case will people who adopt social norms of unlimited universal scope become valuable for their fellow-men. The anonymity, dynamism, and mobility of an open market-society mean that the reasons which speak in favour of enacting universalistic norms acquire maximal importance. The destruction of traditional structures and relationships in modern market societies is therefore highly congenial to the establishment of the kind of interests and virtues that are necessary for making democracy work. When searching for a suitable partner for a cooperative enterprise in such an anonymous, mobile, changing society will one not look for persons who are moral and trustworthy only in relation with their particular circle of people, but for persons who in general have a moral attitude: persons, that is, who take a moral point of view, in the sense of being impersonal and impartial towards the interests of others. Thus, the kind of associational life characteristic of mobile and dynamic societies also produces a demand for a special kind of people as suitable partners. The required qualification is no longer unconditional loyalty to a certain class of people, but a general adherence to the norms of reciprocity and trust. Therefore the development of a market-society contributes to a universalistic orientation and thereby to a kind of social capital which is beneficial to society as a whole and especially the democratic process. In the first place, as it develops an open society with inclusive cooperative structures, transcending natural and artificial borders and demarcations of all kinds, it enhances the chances of “bridging” social capital that promotes and enforces norms of reciprocity and trust beyond the confines of some well-defined groups and shapes interests in favour of a universalistic democratic order. Second, commercial enterprises in a market society themselves create a kind of social capital which is an important source of open networks and universalistic norms of generalized reciprocity and general trust. In this way the market-embeddedness of the civil society is an important foundation for the development of “civic” social capital which overcomes the dangers of particularism and group-selfishness. On the other hand, without the genuine “civil” social capital created mainly outside the market-place the important resources of solidarity, altruistic engagement and autonomy by self-government and self-organization would dry up. The overall result is: we need social capital to shape the kind of political interests and to produce the kind of civic virtues which together make democracy work. But to enhance the chances to get the right form of social capital, civil society should be embedded in a well established market-economy. There is not only one direction of influence here. What is needed is a virtuous equilibrium between markets, politics and civil society. Acknowledgments I would like to thank Margaret Birbeck and Geoffrey Brennan for helping me to prepare the English version of this paper.
Political Norms, Markets and Social Capital
179
References Adler, P.S. and Kwon, S.W. (2000): “Social Capital: The Good, the Bad, and the Ugly”, in: Lesser, E.E. (eds.): Knowledge and Social Capital: Foundations and Applications, Boston: pp. 89–113. Banfield, E.C. (1958): The Moral Basis of a Backward Society, New York. Baurmann, M. (1997a): “Liberal Society and Planned Morality?” in: Associations 1, pp. 81–102; reprinted in: Morscher, E. and Neumaier, O. (eds.), (1998): Applied Ethics in a Troubled World, Dordrecht, pp. 203–223. Baurmann, M. (1997b): “Universalisierung und Partikularisierung der Moral. Ein individualistisches Erkl¨arungsmodell” in: Hegselmann, R. and Kliemt, H. (eds.): Moral und Interesse. Zur interdisziplin¨aren Erneuerung der Moralwissenschaften, M¨unchen, pp. 65–110. Baurmann, M. (1999): “Solidarity as a Social Norm and as a Constitutional Norm” in: Bayertz, K. (ed.): Solidarity, Dordrecht, pp. 243–272. Baurmann, M. (2000a [1996]): Der Markt der Tugend – Recht und Moral in der liberalen Gesellschaft, T¨ubingen. Baurmann, M. (2000b): “Legal Authority as a Social Fact” in: Law & Philosophy 19, pp. 247–262. Baurmann, M. (2002a): “Vertrauen und Anerkennung. Wie weiche Anreize ein Vertrauen in Institutionen f¨ordern k¨onnen” in: Maurer, A. and Schmid, M. (eds.): Neuer Institutionalismus. Beitr¨age zur soziologischen Erkl¨arung von Organisation, Moral und Vertrauen, Frankfurt, pp. 107–132. Baurmann, M. (2002b): The Market of Virtues. Morality and Commitment in a Liberal Society, Dordrecht. Baurmann, M. (2003): “Majority Without Morality? Why Democratic Decisions Demand Ethical Principles” in: The Tampere Club (ed.): The Future of Democracy. Essays of the Tampere Club, Tampere, pp. 97–127. Baurmann, M. (2005): “Korruption, Recht und Moral”, in: Alemann, U. von (ed.): Dimensionen politischer Korruption, Politische Vierteljahresschrift, Sonderheft 35, pp. 164–182. Baurmann, M. and Lahno, B. (2002): “Vertrauen, Kooperation und große Zahlen”, in: Schmalz-Bruns, R. and Zintl, R. (eds.): Politisches Vertrauen, Baden Baden, pp. 191–220. Braithwaite, V. and Levi, M. (eds.), (1998): Trust and Governance, New York. Brennan, G. (1995): “Economising on Virtue”, in: Constitutional Political Economy 6, pp. 35–56. Brennan, G. and Lomasky, L. (1993): Democracy and Decision, Cambridge. Brennan, G. and Hamlin, A. (2000): Democratic Devices and Desires, Cambridge. Brennan, G. and Pettit, P. (2004): The Economy of Esteem, Oxford. Buchanan, J.M. and Congleton, R.D. (1998): Politics By Principle, Not Interest, Cambridge. Coleman, J.S. (1987): “Norms as Social Capital”, in: Radnitzky, G. and Bernholz, P. (eds.): Economic Imperialism: The Economic Approach Applied Outside the Field of Economics, New York, pp. 133–155. Coleman, J.S. (1988): “Social Capital in the Creation of Human Capital”, in: American Journal of Sociology 94, Supplement, pp. 95–120. Cook, K.S. (ed.), (2001): Trust in Society, New York. Curtis, J.E., Baer, D.E. and Grabb, E.G. (2001): “Nations of Joiners: Explaining Voluntary Association. Membership in Democratic Societies”, in: American Sociological Review 66, pp. 783–805. Dasgupta, P. (2000): “Economic Progress and the Idea of Social Capital”, in: Dasgupta, P. and Serageldin, I. (eds.): Social Capital: A Multifaceted Perspective, Washington, World Bank, pp. 325–424. Dekker, P. and Uslaner, E.M. (eds.), (2001): Social Capital and Participation in Everyday Life, London–New York. Etzioni, A. (1993): The Spirit of Community. Rights, Responsibilities, and the Communitarian Agenda, New York. Fukuyama, F. (1995): Trust: The Social Virtues and the Creation of Prosperity, New York. Fukuyama, F. (2000): “Social Capital and Civil Society”, International Monetary Fund Working Paper, WP/00/74, pp. 1–18.
180
M. Baurmann
Gambetta, D. (ed.), (1988): Trust: Making and Breaking Cooperative Relations, Oxford-New York. Gambetta, D. (1993): The Sicilian Mafia: the Business of Private Protection, Cambridge. Granovetter, M.S. (1973): “The Strength of Weak Ties”, in: American Journal of Sociology 78, pp. 1360–1380. Granovetter, M.S. (1985): “Economic Action and Social Structure: The Problem of Embeddedness”, in: American Journal of Sociology 91, pp. 481–510. Hardin, R. (1995): One for All. The Logic of Group Conflict, Princeton. Hardin, R. (1999): “Do We Want Trust in Government?” in: Warren, M. (ed.): Democracy and Trust, Cambridge, pp. 22–41. Hirschman, A.O. (1977): The Passions and the Interests. Political Arguments for Capitalism before its Triumph, Princeton. Lahno, B. (2002): Der Begriff des Vertrauens, Paderborn. Levi, M. (1996): “Social and Unsocial Capital: A Review Essay of Robert Putnam’s ‘Making Democracy Work”’, in: Politics & Society 24, pp. 45–55. MacIntyre, A. (1981): After Virtue, A Study in Moral Theory, London. Misztal, B.A. (1996): Trust in Modern Societies, Oxford. Newton, K. (1997): “Social Capital and Democracy”, in: American Behavioral Scientist 40, pp. 575–586. Offe, C. and Fuchs, S. (2001): “Schwund des Sozialkapitals? Der Fall Deutschland”, in: Putnam, R.D. (ed.): Gesellschaft und Gemeinsinn. Sozialkapital im internationalen Vergleich, G¨utersloh, pp. 417–514. Ostrom, E. (1990): Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge. Ostrom, E. and Ahn, T.K. (eds.), (2003): Foundations of Social Capital, Cheltenham/Northampton: Edward Elgar. Ostrom, E. and Ahn, T.K. (eds.): Foundations of Social Capital, Cheltenham/Northampton. Paxton, P. (2002): “Social Capital and Democracy: An Interdependend Relationship”, in: American Sociological Review 67, pp. 254–277. Pettit, P. (1997): Republicanism: A Theory of Freedom and Government, Oxford. Portes, A. and Landolt, P. (1996): The Downside of Social Capital, in: The American Prospect 26, pp. 18–22. Putnam, R.D. (1993): Making Democracy Work. Civic Traditions in Modern Italy, Princeton 1993. Putnam, R.D. (2000): Bowling Alone. The Collapse and Revival of American Community, New York. Putnam, R.D. (ed.), (2001): Gesellschaft und Gemeinsinn. Sozialkapital im internationalen Vergleich, G¨utersloh. Putnam, R.D. and Goss, K.A. (2001): “Einleitung”, in: Putnam, R.D. (ed.): Gesellschaft und Gemeinsinn. Sozialkapital im internationalen Vergleich, G¨utersloh, pp. 15–43. Stolle, D. (1998): “Bowling Together, Bowling Alone: The Development of Generalized Trust in Voluntary Associations”, in: Political Psychology 19, pp. 497–525. Stolle, D. and Rochon, T.R. (1998): “Are All Associations Alike? Member Diversity, Associational Type, and the Creation of Social Capital”, in: American Behavioral Scientist 42, pp. 47–65. Uslaner, E.M. (2002): The Moral Foundations of Trust, Cambridge. Warren, M.E. (ed.), (1999): Democracy and Trust, Cambridge. Warren, M.R. (2001): Dry Bones Rattling. Community Building to Revitalize American Democracy, Princeton. Woolcock, M. (1998): “Social Capital and Economic Development”, in: Theory and Society 27, pp. 151–208. Woolcock, M. and Narayan, D. (2000): “Social Capital: Implications for Development Theory, Research, and Policy”, in: The World Bank Research Observer 15, pp. 225–49.
Do Multinationals Create Social Capital Just Like That? A Commentary on Baurmann Christoph Schmidt-Petri
Michael Baurmann’s stimulating paper tries to fill a lacunae in the social capital literature. He rightly observes that while a lot of work has been done to explore the impact of private associations on the creation of social capital, the role of markets has been widely neglected. Baurmann argues that multinational companies (he uses the example of Mazda) are in fact very efficient in creating social capital of the best variety. They i) create bridging social networks, ii) establish norms of generalized reciprocity and generalized trust, and iii) promote commitment to public goods.1 These are all, let us agree, highly desirable. Even without doing a detailed analysis of any underlying mechanisms, it is clear that achieving these aims would be highly effective in promoting democratic values. If Baurmann is right, we would therefore have made a gigantic leap towards finding a way how to legitimize democratic political norms by relying on institutions that are not primarily driven by political or moral concerns. In this brief paper I will give reasons to believe that the case is not as simple as Baurmann suggests. He does not give good reasons to believe that markets actually create, rather than need, presuppose, or rely on, the social capital desired. It should be noted at the outset, however, that whether or not social capital is efficiently created in multinationals is an entirely empirical question. Without proper evidence, it is impossible to decide on this issue. Gathering such evidence would, I think, amount to a very substantial piece of empirical research – for which, to begin with, the idea of ‘social capital’ would have to be made so precise as to be measurable – which neither Baurmann nor myself have undertaken. In other words, we are both relying on indirect evidence and abstract argument. But this is the only thing we have to work with for the moment.
1 The Mafia and Mazda Baurmann considers the Mafia, representing particularistic social capital formation, and contrasts it with the universalistic social capital created at Mazda. His observations about the Mafia are certainly right and to the point: despite exhibiting 1
Baurmann: this volume, p. 172.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
181
182
C. Schmidt-Petri
an extremely high degree of internal cohesion and trust (comparable to a ‘family’), it is hostile to democracy and openly in conflict with democratic and universal values (but not just them). But I have serious doubts that Baurmann is right in claiming that a multinational like Mazda does create social capital of the desired kind. Already I don’t see very well what the assertion that the Mafia is a locus of high social capital – but of the wrong kind – adds to our understanding of what is wrong with the Mafia. In fact, to even emphasize its ‘high social capital’ seems misleading as it suggests, more or less explicitly, that strongly hierarchical associations formed to exploit the weakest members of society using violent and unlawful means possess a great number of good-making characteristics which are, in some sense, unfortunately employed to the wrong ends, and contingently so. If we want to say that close-knit criminal organisations do exhibit ‘high social capital’, that seems to indicate that our conceptual apparatus is probably not well adjusted. The fact that we already have to make apparently subtle distinctions between different types of social capital at such a coarse level of theorising illustrates how unhelpful this terminology is. But this is just a matter of rhetoric, of course, and only highlights the need to further differentiate, just as Baurman does. So let us focus on the good variety of social capital from now on. Baurmann’s main argument concerns multinationals like the car maker Mazda. Mazda is essentially the opposite of the Mafia: Mazda membership is not restricted to people who already belong to a common social framework and possess identical personal traits of kinship, origin or culture. In contrast, Mazda exemplifies social capital with a high degree of ‘bridging’ qualities . . . . The company-network of Mazda is highly inclusive and serves as a device to overcome manifold differences between people in order to create a mutually beneficial cooperative relationship.2
Although I also believe, as I will explain below, that multinationals go some steps towards creating capital, I think that Baurmann is overstating his case. As far as I can see, the opposite might just as well be true. As I have mentioned, whether a company such as Mazda does create social capital is an empirical matter. Baurmann doesn’t provide any direct evidence in support of his claims. This is a serious drawback since whether or not Mazda does create social capital depends crucially on whether Mazda actually is ‘highly inclusive’. Even so, this might be a necessary condition, but it is far from sufficient – maybe the different cultures represented at Mazda don’t get along very well. Maybe people meet at work to build cars, and then go home to watch tv. In other words, maybe people at Mazda are professionals: they work together but bowl alone. We just don’t know this – but it would be interesting to find out more. However, there is a more abstract argument. Baurmann claims that: the norms of reciprocal and trustworthy behaviour in a company cannot be particularistic . . . [Mazda] is dependent on the possibility to employ a wide range of people, whether they are of different nationalities, cultures, races or social classes.3
2
Ibid., p. 174.
3
Ibid., pp. 174–175, italics added
Do Multinationals Create Social Capital Just Like That?
183
But clearly it is one thing to be dependent on something, and a different thing to be able to produce the thing one is dependent on. The fact that Mazda needs a workforce that is diverse in the sense required surely does not entail that Mazda actually does anything to provide the conditions required for potential members of the workforce to exhibit this diversity. It is not outlandish to suppose that a company will decide on where to locate their factories based on whether or not the potential workforce present already does exhibit the desired characteristics; that is, that the company will try to take advantage of conditions it has done nothing to build up. In fact, this is plausibly true. Besides, Baurmann doesn’t give any argument to show that Mazda does need such a workforce. Considering how hierarchical the Japanese management style traditionally is, my impression is that Mazda primarily wants to hire people who will perform well at the assembly line (or whatever job the employee is hired to do). One could think that the issue is different at the middle and top management level. I have no evidence on Mazda’s hiring policies, but it is worthwhile to note that from 1996 till 2003 Mazda was mostly run by CEOs taken from Ford. Ford and Mazda started to collaborate in the late 1970ies, and Ford now owns about a third of Mazda’s shares. These issues are much more complex than portrayed here, but the fact that Mazda would systematically import foreign leadership tends to disconfirm the hypothesis that by just running the business of car manufacturing, any significant amount of trust is being generated. A major factor in hiring a CEO, besides the hard skills involved in running a company that size, is likely to be whether he or she can provide leadership to the workforce, and transport the company’s values. If Mazda was so good at creating relations of trust among its employees – and this presumably applies to the top management too – it is curious that they would have to look for foreigners to lead the company. A similar fallacy seems to be the basis of the following section. Baurmann argues that higher geographical mobility, as required by modern market economies to an ever increasing extent, will create a demand for people who in general have a moral attitude . . . . The special kind of associational life in mobile and dynamic societies also produces a demand for a special kind of people as suitable partners.4
That is almost certainly true. But it does not suffice to establish that this demand is met, that is, that such universally trustworthy people are being supplied due to the increase in demand. Quite the contrary. It seems natural to suppose that as the proportion of short-term or one-shot interactions rises in a more mobile society, so does the proportion of dishonest interactions (keeping a supposed ‘individual honesty level’ constant). If cheating gets easier, it will occur more frequently, even if people’s characters don’t change to the worse – they are just tempted more frequently. This would in fact explain why one would perceive an increase in ‘demand’ for universalistic values: they are in rather short supply in highly mobile and anonymised societies. This is the problem, not its solution. It is clearly not a priori
4
Ibid., p. 178.
184
C. Schmidt-Petri
true, since plausibly false, that the increase in occasions for dishonesty will be offset by an automatic increase in honesty. Having said that, I should add that Mazda in fact seems very active in creating social capital, in particular with the Mazda Foundation. For instance, the Mazda Foundation gives scholarships to minority students at selected universities, encourages the transition of women into the workforce, and promotes cross-cultural relations between the US and Japan. These are the kind of activities that, in my opinion, do create the social capital Baurmann is talking about. And they happen at Mazda. But, and this is important, not in the ordinary day-to-day business of car manufacturing. And to me, that doesn’t seem surprising.
2 General Considerations Maybe I was focussing too much on the specifics of Mazda. But it is hard to see what the general point is. What is it about Mazda that makes it so praiseworthy as hotbed of social capital? Its being a multinational? Its being a car manufacturer? Or just its being a company? Does the same argument work for Daimler-Chrysler? For Fiat? For Toshiba? In Bowling Alone, Putnam himself gives good reason to doubt that Baurmann’s thesis holds in general. In chapter five, entitled ‘Connections in the Workplace’ he writes: I know of no evidence whatever that socializing in the workplace, however common, has actually increased over the last several decades . . . Workplace ties tend to be casual and enjoyable, but not intimate and deeply supportive.5
After considering at some lengths the effects of modern management tools such as ‘reengineering’, ‘right sizing’ etc. he concludes: In the end, ‘work’ entails time and effort destined to serve primarily material, not social, ends. Work-based networks are often used for instrumental purposes, thus somewhat undercutting their value for community and social purposes.6
The primary aim of any normal firm – profit-making – does not lead to conditions particularly favourable to the creation of social capital. But what is more, firms actively interfere with processes that normally would lead to the creation of social capital: two-thirds of [American] employers record employee voice-mail, e-mail, or phone calls, review computer files or videotape workers. And such surveillance is becoming more common. Rights of free speech and privacy that are essential to public deliberation and private solidarity are, to put it mildly, insecure in the workplace. Substantial reforms in public law
5
Putnam (2000), p. 87.
6
Ibid. p. 91.
Do Multinationals Create Social Capital Just Like That?
185
and private practice would be necessary before the water cooler could become the equivalent of the back fence or the town square . . . . The workplace is not the salvation for our fraying civil society.7
Given that the most important proponent of the concept of social capital is sceptical about trying to rely on firms and the marketplace to bolster social capital creation, what hope is there for the project of finding the legitimization of universalistic political norms in the head offices of multinationals? At first sight it seems even uncontentious to claim that to the extent that you want to do global business, you are interested in the global rule of law.8 But just considering such a straightforward claim renders apparent the pitfalls of the project: probably some companies are quite happy with the situation, currently prevailing, that some rules and laws are interpreted somewhat more liberally in certain countries – just think of the examples of differential legislation on tax, personnel safety, environmental protection, and employment. Any normal company would go to where they get the best overall deal, and which country will offer that will depend on whether you produce chemicals, sneakers, or hedge funds. What does seem true is that global players would like to be protected in whatever country they do business, and probably ideally by the law, since that is typically very cost-efficient for them. But this by itself does not entail that the exact form of this protection should either be the same in all countries, nor that it should cover everyone else in any of the countries concerned – that is, that it be universalistic in Baurmann’s sense. Strong markets and favourable conditions to international trade are entirely compatible with a systematic discrimination of certain groups of the population. Besides, such commitments to political norms are clearly instrumental rather than intrinsic, and will depend on the current business interests. So even such a straightforward argument will not, without further ado, give the desired result. To run the same argument with a detour via social capital seems even less promising. This is not to say that market economies do not have the tendency to contribute to social capital creation. I am convinced that the socialisation in the workplace makes a big difference to how people conceive of larger associations and maybe society as a whole, and I am also convinced that this effect very often runs in the ‘right’ direction. International cooperation in any field and on any level is likely to raise cultural awareness and reduce prejudice, and should be encouraged for that reason – and this also applies to commercial cooperation of course. However, I think it is very difficult to successfully argue that working for and in a multinational company contributes to the legitimization of democratic values just like that. The contexts are simply too different to endorse such far-reaching claims without strong empirical evidence. As
7
Ibid. p. 92.
8
Baurmann: this volume, pp. 174 and 175.
186
C. Schmidt-Petri
Baurmann points out, what is required is a virtuous equilibrium between markets, politics and civil society.9 I think he has overestimated the part the market will play in this equilibrium.
Reference Putnam, R.K. (2000): Bowling Alone, Simon and Schuster.
9
Ibid., p. 178.
Cultural Diversity and Liberalism Russell Hardin
Political and economic liberalism descend to us from ancient times after passing through the Scottish Enlightenment and the English Whigs. Political and economic liberalism were directed at government intrusions, which were to be stopped. Now in highly developed liberal societies that face diversity, much of it through immigration, people assert a liberalism of group rights. Indeed, Will Kymlicka and Yael Tamir insist that the protection of groups per se is justified on liberal grounds.1 The assumptions of political and economic liberalism are those of spontaneous individual creation and therefore liberation from the control of others; the assumptions of group liberalism are very different. Commonly, the defense of this liberalism requires that government intrude into individual lives to overcome deficits – economic, political, and cultural – that particular groups suffer. In particular, they require government support for forceful control of group members themselves or, at least, special exemptions from certain laws that otherwise apply to everyone.2 With John Dewey one could speak of stages of liberalism:3 liberation from despotic and oligarchic political control, liberation from government economic decisions over who is to work or produce or trade, liberation from the depredations of institutionalized private power, liberation from the dead hand of many social conventions, and perhaps other liberations. Dewey would put the first three of these in this historical order, although I think it is misleading to put economic liberalism in the order in which it was first articulated as opposed to the order in which it began to work its way on the scene.4 To date, we have genuinely articulate accounts of only the first two of these and a still emerging account of the third. What we need for giving a liberal gloss to the protection of groups as such would be a theory that trumps Dewey’s largely still unarticulated social liberalism. Social liberalism frees individuals from the dead hand of custom and social pressure; group liberalism requires subordination of individuals to group customs and values. 1
See: Kymlicka (1989) and Tamir (1993).
2
But see Miller (1995), Chap. 5.
3
Dewey: (1987 [1935]), Liberalism and Social Action.
4
See further, Hardin (1999), Chap. 2.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
187
188
R. Hardin
When political and economic liberalisms were joined in the same governments, especially beginning with that created by the US Constitution, they were joined without subordinating one to the other or curtailing the application of one on behalf of the other. Dewey’s vague prescription – substantial government control of the economy – for a new liberalism in the 1930s would likely have curtailed economic liberty and perhaps, therefore, political liberty on behalf of a nascent welfare liberty. The welfare state that has grown up mostly after World War II in the West (although much earlier in Wilhelmine Germany) can, as the best experiences suggest, similarly be built alongside, instead of on the partial ruins of, the old political and economic liberalisms. Apart from making education and culture relatively available, no one has seriously proposed any scheme for generally breaking the hold of perverse social conventions. Consider the strategic natures of these further liberalisms in comparison to political and economic liberalism. There are two issues. First is the role of government, whether it is to be constrained or put to use. Second is the game theoretic structure of the larger interactions at stake. All of the later liberalisms differ from the earlier political and economic liberalisms in that they virtually require government action in their support. On the other issue, however, they differ. Economic and political liberalisms, when they work, are coordinations on mutual advantage regimes. Because they are coordinations, they are self-enforcing. In this respect, these share the strategic structure of constitutionalism and democracy.5 The regime of institutional liberalism, at least as it would affect large institutions, might readily be mutually advantageous and self-enforcing in a democratic society. Social liberalism, which involves the breaking of destructive social conventions, would also be self-enforcing if once achieved. Unfortunately, group liberalism is not strategically analogous to political and economic liberalism. The maintenance of group autonomy for selected groups is not likely to be mutually advantageous for the most politically important groups in liberal societies. They do not serve the interests of the middle class and the politically influential, wealthy entrepreneurial class. Nor are resolutions of group problems likely to be self-enforcing the way the old liberalisms, once in place, are self-enforcing or the way a workable constitution is self-enforcing. The greatest threat to the survival of a group’s ways is the next generation, whose interests are often not served by the group’s static values and norms. A striking fact about many of the new labels for ostensibly variant liberalisms is that those labels and their categories have been around for a long while – most of, or more than, a century – but that the liberalisms they represent have yet to have much effect. The original liberalisms reversed this history: Elements of them were long in effect before they were well understood, although there was arguably more invention in the case of political liberalism. Economic liberalism had been working piecemeal for centuries before Mandeville, Hume, Smith, and others began to figure it out. One might say with less conviction that political liberalism also had a past
5
Ibid.
Cultural Diversity and Liberalism
189
history that eased the task of Hobbes, Locke, Hume, and Mill in coming to formulate its theory. This different history is indicative of an important strategic difference between the original liberalisms and some of the later ones. The original liberalisms were and are self-enforcing because they are mutually advantageous to important, politically efficacious, large groups in the societies they influence. Because they were self-enforcing, they had survival power and they could take root and grow over time without their being yet understood. This is not true of the group liberalism of our time. Such a liberalism is still in want of an intellectual grounding, of a theory of how it can be made to work. We can be confident that merely constraining government cannot be a major part of any program of group liberalism, as it was for economic and political liberalism. In group liberalism, government is not the source of the problem but it must, rather, be a major part of the solution. Out of the welter of categories of liberalism that have filled twentieth-century debate, three are fairly widely mentioned and are clearly relevant to fundamentally important aspects of liberal societies, and a fourth is widely asserted and at least arguably important. These are the following. First is what we can call social liberalism, which is liberation from the deadening weight of burdensome social conventions. Second is the liberalism that Dewey wanted and that we may call institutional liberalism, which is liberation from the snares of large private organizations on analogy with the liberation from the snares of government under political and economic liberalism. Third is what is commonly called welfare liberalism, which is primarily liberation from poverty and its concomitants. Fourth and newest is group liberalism, which focuses not on liberty for individuals but for groups. All of these sound close to the welfarist vision of traditional political and economic liberalism, whose point is to make life better, but they require very different strategic devices that go beyond constraining government. I will discuss social and institutional liberalism briefly in order to set up discussion of the very different problems of group liberalism. Indeed, it is not merely different, it is fundamentally contrary to the other liberalisms, perhaps most especially social liberalism, whose point is largely the ending of group control over individuals.
1 Social Liberalism Social liberalism has had a long history, with articulate concern for it in Mill’s On Liberty and other classical liberal works. It perhaps has some trace in the US Constitution in the prohibition of a state religion. Its greatest impact on any political program, however, has probably been in French revolutionary moves against the Catholic Church, Communist efforts to break the hold of religion and various customary constraints, the turn-of-the-century Chinese move to break the coercive custom of women’s foot-binding, and other efforts, some of them worse than the ills they were intended to cure, as in Pol Pot’s destruction of everything he could destroy in Cambodian culture, whether good or bad, at the cost of upwards of a million lives
190
R. Hardin
and the radical impoverishment of virtually the entire population. Most of these effects required government action, although the foot-binding in China was broken by creating an opposite norm from the ground up in one of the most remarkable social changes on record.6 Social liberalism is typically contrary to notions of group autonomy. Breaking the hold of a social norm may mean loosening the hold of a particular community on its members. For example, undercutting destructive religious norms is likely to undercut ties to a religious community and even to create conflict within such a community, in either case causing a decline in group cohesion. Social liberalism therefore should be anathema to communitarians. Strangely, however, many of the Anti-Federalist opponents of the US constitution favored the introduction of a bill of rights that would specifically protect individuals. Such rights might indirectly protect communities through the protection of individuals who have communal values. The Supreme Court protected the Amish as a group, however, by ruling that individual Amish children could have their legal requirement (in the state of Wisconsin) to complete at least a tenth grade education reduced in order, somewhat forcibly, to keep them loyal to their community. Immanuel Kant argued that for one generation to stifle the intellectual and moral development of a later generation in this way is to commit a crime against human nature.7 In this instance, the Court abused future generations in order to satisfy demands of the current generation of adult Amish or, arguably, primarily the leadership of the Amish order. One might have expected the Anti-Federalist communitarians to favor such group rights.
2 Institutional Liberalism Dewey thought that the central problem in the established liberal democracies of our – or his – time, was the liberation of people from the impositions of large private organizations on individual liberty and welfare. He supposed that this was the new problem of liberalism. He also supposed that the problem must be handled by government action. We had long been liberated from governments that imposed aristocratic control of society and mercantilist control of the economy, and now we needed to be liberated from private power that had arisen under the regime of old liberalism. He wrote that, “after early liberalism had done its work, society faced a new problem, that of social organization”.8 Because he was writing in the depths of the Depression, it is plausible that most of what Dewey thought we needed was what could be handled by social welfare programs that do not infringe old economic liberalism beyond the standard infringement of taxation, with which advocates of the old liberalism were always content.
6
Mackie (1996).
7
Wisconsin v. Yoder, et al.; see further, Hardin (1995), pp. 201–3. Also Kant, (1983 [1784]).
8
Dewey (1987 [1935]), p. 39; see also Lindblom (1977), pp. 49–51, McConnell (1966).
Cultural Diversity and Liberalism
191
Institutional liberalism was a response to crude aspects of economic life in the brightest moments of capitalism from, say, 1840 to 1929 in the United States and roughly the same period in England. The triumph of capitalism did not end grotesque poverty and inequality but, in the view of many, exacerbated them. Or, at the very least, one can say that economic liberalism and the market have benefited some far more than others, that they are not neutral in their impact. Dewey’s most articulate statement of the need for institutional liberalism was delivered in 1935 during the darkest days of capitalism when, oddly, it was arguably beside the point for the problems that were most urgent then. These problems were still poverty and inequality, especially as aggravated by unemployment. But their solution was not, as in institutional liberalism, liberation from the intrusions of large private organizations. The unemployed of the 1930s did not need to be liberated from such institutions. They would, rather, have benefited from greater success of these institutions. How can we fit Dewey’s institutional liberalism with the earlier liberalisms? It was, of course, motivated by a concern for welfare, and in this it is similar to all liberalisms. The conceptual analogy with political and economic liberalisms is that it liberates. The earlier liberalisms liberated from arbitrary government intrusions into people’s lives and from government control of the economy. The arbitrary intrusions that provoked political liberalism were star chambers, bills of attainder, arrests without warrant, billeting of troops without permission or recompense, seizure of presses, political imprisonment, and virtually anything else an uncontrolled government might choose to do or demand. The intrusions that were against economic liberalism were the panoply of practices of government economic control in the heavy-handed system of mercantilism, in which friends and relatives of the crown were given economic privileges and in which workers and producers were hassled by destructive regulations on what they could do and where they could do it. These regulations restricted mobility, closed off cities from independent artisans and traders, required long apprenticeships to qualify for work, gave strict monopolies to some, and blocked trade with foreign enterprises. Dewey’s institutional liberalism was intended to overcome the similarly grim intrusions of large institutions other than government. It would liberate from the control of large private organizations.
3 Group Liberalism In recent decades, there have been many demands for attention to group “rights” or group protections of various kinds. Virtually any other liberalism could be called group liberalism, but I will reserve the term for protections of specifically namable groups. For example, protection of an immigrant group’s use of its native language in its dealings with government and in the education of its children would be an instance of group liberalism. All the other liberalisms canvassed here work by protecting individuals. Group liberalism is very odd in that it somehow elevates the
192
R. Hardin
relevant group above its members by protecting the group, plausibly against its own members. We could characterize demands for group protections in two ways. First, it might be an extension of some of the earlier demands for institutional liberalism to protect workers or consumers against private institutions. For example, workers are a group who can claim that they need general enforcement of a rule to enable them to mobilize against corporations.9 Similarly, government may determine limits on what can be the terms of contracts covering either relations between unequal parties or relations that have significant external effects on those not party to the contract. This facilitates what groups can do or protects them against harms. For such protections, a liberal government might adopt something akin to Mill’s harm principle. But, unless group liberalism is to conflict with economic liberalism, government should not avoid harms by manipulating specific aspects of the economy. For example, government might protect workers as a group against the harms brought by economic change. But it should do this with worker specific programs rather than by artificially keeping a failing firm or an obsolescent industry in business. Second, it might be an extension of the descriptive theory of interest group liberalism, which characterizes American politics in the quasi-Madisonian system of a plurality of interests engaged in trying to influence national policy. But in the pluralism of interests, the groups are typically contending for favor directly from the government. In the newer group liberalism, groups are demanding protections against government and private agencies. For example, they demand protection against government requirements on how to educate their children and against the freedom of speech of movie makers and television programmers. Strategically, such liberalism is a hotch potch. One of the demands of groups in our time is for the official protection of minority languages. In the United States such protection probably makes the first generation speakers of Spanish, Korean, or Vietnamese better off. But it might partially cripple the next generation because, typically, it means making sure that the next generation is educated in the minority language and plausibly made less able to assume a full role in the larger community. Hence, protecting the supposed group interest requires action against the interests and incentives of some group members. At the very least, this makes group liberalism a very complex version of liberalism. It can hardly be defended either on standard welfarist or autonomy grounds. And it conflicts with social liberalism and possibly with institutional liberalism. Finally, advocates of group liberalism in its stronger variants demand impositions on the larger society and even their own members, and they often want government to manage these impositions. Hence, group liberalism is often profoundly illiberal in any sensible prior reading of that notion. Many who are not members of groups that 9
Mill argued, as an example, that workers might require legal backing to enforce their unanimous preference for reduction from a 10-h to a 9-h day, because without legal enforcement, individual workers would have incentive to freeride on the abstinence of others and to work an extra hour for bonus wages, thus destroying the 9-h day. See Mill (1965 [1848]), book 5, Chap. 11, Sec. 12, p. 958; see also Hardin (1988), pp. 92–94; Hobhouse (1948 [1910]), pp. 32–33, 37–39.
Cultural Diversity and Liberalism
193
want group autonomy defend group protections despite the illiberal implications. Their positive argument for group rights is, roughly, that giving groups status, even with some controls over individual group members, allows the group members to enjoy benefits that would otherwise be at risk from the corrosive effects of the larger society. Hence, government protection of a group is merely a means to protecting its members. If the central meaning of liberalism is that it liberates, the meaning of group liberalism is, of course, that it liberates one group from the hegemony or control of another group or of other groups in combination. This would make it unlike any of the other liberalisms discussed here because it would be analogous to liberation of one individual from the control of another individual, rather than liberation of individuals from control by large social entities, such as governments, institutions, and powerful, widespread norms. We might conceive of group liberalism as protecting groups from government intrusions in their lives, and that can be an issue. It is only in this vision that it is analogous to other liberalisms. Any “group” right that essentially protects a group against other groups or against government can be seen either as a particular application of the ordinary rights of all individuals to the rights, individually, of the members of the group or as a protection of the norms or rules of the group largely against its own members’ violation of those norms or rules. For the first category, there need be no constitutional provision other than heightened attention to the claims of individuals in a relevant group, as in the case of the rights of former slaves and blacks more generally in the United States. Such protection seems likely to be little different from the protection of individual rights more generally. The major difference is that the group whose members are protected might suffer simultaneously from prejudicial norms of the larger society, so that the government must intervene to help break that social imposition, as when it opposes and attempts to end racism. Many groups – or at least many groups’ leaders – insist on having group rights of the latter, stronger kind. The leaders of Inuit native populations in Canada and of Old Order Amish populations in Wisconsin have won the latter kind of group rights with the implicit or explicit enforcement of those rights by the Canadian and US governments, respectively. The Federal government of the US enforced the group right of the Amish to block their children’s education against the government of the state of Wisconsin. The Canadian government much more actively supports the special legal claims of the Inuit and also funds the tribe and its tribal government. Some philosophers and many advocates of the strong form of group rights claim that they are essentially liberal or are required by liberalism. Because liberalism is such a protean term, these claims might be nothing more than definitional. But some writers insist that traditional liberals should support the enforcement of such rights for certain groups.10 I will not argue against them here other than perhaps implicitly in showing how such rights conflict with individual rights that traditional liberals
10
Kymlicka (1989), Tamir (1993), Young (1990).
194
R. Hardin
support. When there is such a conflict, then the claim that traditional liberals should support such group rights is specious.
4 Illiberal Groups It is instructive to see how these various liberalisms come into play in an actual case. The Civil Rights Movement in the United States captures the whole range of concerns of the various liberalisms. That movement in the 1950s and 1960s was primarily a movement to extend political liberalism to cover a previously excluded group, and in this it was first directed at government: at Jim Crow laws and at courts that refused to enforce liberal laws that would give blacks easier access to politics, education, and the market economy. Substantial success in this movement was inadequate to overcome the deeper problems of racism and, therefore, the movement also pushed for laws to force private institutions to end discrimination of many varieties. In this, its program was that of Dewey’s institutional liberalism. Even this program, however, would be inadequate to overcome the inequalities of blacks in American society. Two further projects would be needed: ending the pervasive, non-institutional racism of social conventions and ending poverty. And some in the black community would go further and demand group rights, although they typically would want autonomy as a way of escaping racism and white institutional controls rather than as a way of protecting religious or other group-level values. Other subnational groups have a very different agenda. They, or their leaders, want recognition as separate groups with special status in the larger society. That special status might go no further than to make it possible for a group to survive as such into future generations. The group, given such protection, would then take care of attempting to secure its own members’ loyalty and of maintaining its values and its members’ commitment to those values. But the requisite special status might be almost that of a state within the state with substantial powers over its members, especially powers to coerce them. Such a subgroup is very unlikely to have a democratic structure, although its norms and values might be widely shared among many, especially older, members of the group. Clearly, the urgent problem in the life of a subnational group that wishes to have state protection in sustaining itself is keeping its members loyal to it, so that they do not leave the group for the blandishments of the larger society. Those blandishments are likely to be especially attractive to the young of the next generation, who have yet to settle into the routines of the group’s norms. The power of Yoder was that it secured the next generation’s loyalty to the local Amish community by cutting off any chance of finding attractive employment in the larger society outside. In a society in which a high school education is virtually required even to be an unskilled worker or clerk, forcibly ending a child’s education at age fourteen is a powerful move. When these are the concerns of group liberalism, it is impossible for it not to conflict with traditional liberalisms that protect individuals first and foremost.
Cultural Diversity and Liberalism
195
Group liberalism with such a program is inherently illiberal on the traditional theories. Hence, we face the problem of justifying it despite its implication of sometime hostility towards and suppression of its own members. Let us divide the problem into two categories. The first is a liberal society into which illiberal groups immigrate; the second is an older society in which there are long extant illiberal subnational groups and into which a liberal constitution is introduced. The way we deal with an illiberal population may depend on whether that population precedes a liberal constitution. If the population immigrates into the nation and a liberal constitution is in place, then the new population can be assumed to accept life under the extant constitution. This does mean, of course, that the members of that population genuinely support the terms of the constitution but only that they have chosen to immigrate – perhaps despite that constitution – even though they could in almost all cases readily have stayed in their prior nation. If the immigrant population wishes to maintain some practice that violates the extant constitution, they can choose either to attempt to get the constitution changed to permit their practice or they can act illegally. If they act illegally, they are then subject to the law of their newly chosen nation and may be punished for acting according to their illegal practice. For an indigenous population that precedes the liberal constitution, one might expect that the constitution would make special provisions for their illiberal practices. For example, the current Indian constitution allows for Muslim practices of marriage and divorce by Muslims, while not permitting those practices for nonMuslims. Even then, however, the constitution might require liberal treatment of individuals who wish not to follow their family’s practices. Our chief issue therefore is how to handle illiberal immigrant groups. Consider a particularly harsh example. In early 2002 in Sweden, Fadime Sahindal, at age 26, was shot dead by her father in a so-called honor killing. He and her brother had threatened to kill Fadime for several years because she refused to give in to her family’s plan to marry her to a Kurdish cousin. She had been in love with an Iranian Swede who had died in a car crash in 1998. After his death, she still refused to marry the cousin and she lived more or less as a widow. She became a public figure as an advocate of the right of immigrant women to choose their own lives.11 On a claim to protect his family’s honor, Sahindal’s father murdered her. In some Muslim nations, Sahindal’s father’s action would not be criminal and in many Muslim societies he would receive high praise and approval for his honorable action. In Sweden that action was murder. In Sweden, Norway, and Denmark, with their relatively open immigration policies, such honor killings have become frequent events and forced marriages are very common. Forced marriages are apparently not punished, but honor killings are. And, Nadia, a woman who is a Norwegian citizen, was protected against forced marriage even when her family had abducted her to take her to Morocco for an arranged marriage that she did not want.12
11
Personal communication with Unni Wikan, February 2002.
12
Wikan (2000).
196
R. Hardin
Such cases of honor killings and the kidnapping of young women who do not conform to their parents’ values are especially alarming to traditional liberals. Indeed, it seems inconceivable that a traditional liberal would want the state to grant group rights to do such things to individuals who do not conform to the group’s norms. When Kymlicka and others defend group rights, they do not typically take up such issues, perhaps because their cases have more benign cultural practices than honor killing for refusal to marry a particular person. Even in their cases, however, there are commonly sexist practices that no contemporary liberal could defend. The very logic of traditional liberalism is contrary to allowing honor killings and coerced marriages for some groups in a society while governing the rest of the society through ordinary laws prohibiting murder and coercion. The values of liberalism are inherently universalistic, not hand-crafted to different groups according to their labels. Indeed, a central appeal of liberalism is its universalism. The defense of group rights that imply coercion of the next generation is illiberal, and the arguments of Kymlicka and others in defense of group rights cannot, contrary to their sometime claims, be inferred from liberal principles. At most, groups can be granted modified rights that still exclude such actions as honor killing and marital coercion. If a group’s members are systematically disadvantaged, for example economically, a liberal state can readily provide support to the group. It is in fact part of the universalism of liberalism that such a policy could be commended. To claim such support while, however, insisting on blocking universalistic principles in other realms is duplicitous. Such duplicity is much of the stuff of politics, but it wrecks any effort to give a justification of the contradictory policies.
5 Concluding Remarks Where should an otherwise liberal state stand on its dealings with illiberal immigrant groups? I do not think there is a correct answer to such a question if it is essentially normative or moral. But there can be an answer practically, which is that for a universalistic liberal state to support illiberal practices in selected subpopulations is incoherent and likely to be destructive of the survival of the liberal order. We must choose a generally liberal order or a generally illiberal order. It is not incoherent or immoral or impolitic to choose a generally liberal order if that is what we already have and if it is the likely choice of a majority of our society. Indeed, we can even choose not to allow substantial immigration of any population that would be profoundly illiberal and that would be destructive of our society. If we hear many stories such as those of Fadime and Nadia, we are likely to wonder at the cost generally of allowing migration of those who cannot adapt to moderately liberal customs and mores. The conclusions here do not depend on any claim that liberalism is right or good in ways that trump other principles for social organization and interpersonal relations. Indeed, it is the absence of any such justification that makes the defense of liberalism (or any other social theory) inherently a matter of defending our life because it is good for us in particular. Any claim that we should give up that way of life on behalf of others who wish to have a very different way of life is as morally
Cultural Diversity and Liberalism
197
ungrounded as liberalism itself. Devout religious believers have as much claim to the morality of their vision as liberals do. But they have no claim to disrupt a liberal society. For liberals this is largely a welfarist concern. Our lives are good because we are in a liberal society. Liberalism is a welfarist principle. It serves the mutual advantage of those who have the relevant concern with individual liberty and control of their lives. Many of the communal value systems, such as that which ruined the life of Fadime and nearly ruined that of Nadia are at least in part anti-welfarist. If it requires government action to secure those values, there is a contradiction. We cannot in the name of liberalism protect illiberalism. Those who insist that we must, as liberals, make a place for illiberalism in our society are incoherent. In general, this means in practice that we cannot allow, for example, kidnapping or murder on behalf of communal values. It is perhaps because they had grown up in liberal Scandinavia that Fadime and Nadia suffered reprisals and coercions. The prior generation (and apparently the younger generation of males) of their families was still immersed in communal values of a very illiberal society; Fadime and Nadia were not. If one wanted to assign responsibility for the fates of Fadime and Nadia and thousands of others in similar positions, one would have to note that it was the prior generation who chose to rear them in a liberal society with all its freedoms (or, if one prefers, with all its license). They chose to mold their daughters in ways that then offended them. In Nadia’s case Liberal Norway was forced to decide between the two generations. A liberal can make no other choice than to protect the adult or near-adult children of these families just as it would protect children from ethnically Norwegian families. A liberal society must be liberal in the treatment of those citizens and residents who wish to be treated liberally. All of the talk of the rights or interests of groups per se has so far failed to bridge the generation gap. The Yoder case sharply focused on exactly that issue, and the US Supreme Court was in woeful default on the issue. Until it grapples with the next generation and the status of its members, the defense of group rights is dishonest and vacuous. Norwegian and other liberal societies cannot make room for state defense of honor killing by fathers who are offended at their daughters’ behavior. The suppression of honor killing of various kinds and the transfer of social control in many often violent matters from the community to the state – including the invention of the legal idea of murder – arguably initiated the long historical development of liberalism. The original introduction of law to handle matters that were formerly handled by feud and vengeance was surely a change that served mutual advantage for almost everyone. Some might initially have defended the prior practice of, say, vengeance as somehow inherently moral or right, but subsequent generations can only count the change to the law of murder as a good one. For example, the Corsican norm of vendetta was arguably part of the reason for Corsican economic backwardness and it brought enormous suffering and pain to many families, yet many in that society defended the practice as morally required.13 At a slightly higher level of social control, 13
M´erim´ee (1989 [1840]), see also Hardin (1995), pp. 133–136.
198
R. Hardin
is the Greek practice of punishing people on the spot for various public offenses, which has led to the Islamic practice of instant punishment for women who reveal skin in the wrong places, even if accidentally, a practice that was especially harsh under the Taliban in Alghanistan. The Greek official was “inspector of the market,” or agoranomos. This was translated directly into the “Arabic amil al-suk or sahib al-suk, who had a limited civil and criminal jurisdiction; it was later, under the early Abbasids, to develop into the Islamic office of the muhtasib”.14 At a later development of law, one would expect to see such unregulated forms of enforcement to be superseded by more nearly routinized, depersonalized systemic devices. Such depersonalization in the law is the height of liberalism. It would be a strange betrayal of the long history of the development of liberalism and depersonalized law to justify, in the name of liberalism, contemporary cultural practices that are brutally personalized and illiberal.
References Dewey, John (1987 [1935]): Liberalism and Social Action, in: The Later Works of Dewey, 1925– 1953, Vol. 11, 1–65, Carbondale, IL: Southern Illinois University Press. Hardin, Russell (1988): Morality within the Limits of Reason, Chicago: University of Chicago Press. Hardin, Russell (1995): One for All: The Logic of Group Conflict, Princeton, NJ: Princeton University Press. Hardin, Russell (1999): Liberalism, Constitutionalism, and Democracy. Oxford: Oxford University Press. Hobhouse, L. T. (1948 [1910]): Liberalism, Oxford: Oxford University Press. Kant, Immanuel (1983 [1784]): “An Answer to the Question: What is Enlightenment?” In: Kant, Immanuel: Perpetual Peace and Other Essays on Politics, History, and Morals, trans. by Ted Humphrey. Indianapolis, IN: Hackett. Kymlicka, Will (1989): Liberalism, Community, and Culture. Oxford: Oxford University Press. Lindblom, C. E. (1977): Politics and Markets. New York: Basic Books. Mackie, Gerald (1996): “Ending Footbinding and Infibulation: A Convention Account,” in: American Sociological Review, Vol. 61: 999–1017. McConnell, Grant (1966): Private Power and American Democracy, New York: Knopf. M´erim´ee, Prosper (1989 [1840]): “Colomba”, In: M´erim´ee, Prosper: Carmen and Other Stories, Oxford: Oxford University Press. Mill, John Stuart (1965 [1848]): Principles of Political Economy, edited by Robson, John M., Toronto: University of Toronto Press. Miller, David (1995): On Nationality, Oxford: Oxford University Press. Schacht, Joseph (1982 [1965]): An Introduction to Islamic Law, Oxford: Oxford University Press. Tamir, Yael (1993): Liberal Nationalism, Princeton, NJ: Princeton University Press. Wikan, Unni (2000): “Citizenship on Trial: Nadia’s Case.” Daedalus (fall): pp. 55–76. Wisconsin v. Yoder, et al., (1972) 406 U.S., pp. 205–249. Young, Iris Marian (1990): Justice and the Politics of Difference, Princeton, NJ: Princeton University Press.
14
Schacht (1982[1965]), p. 25. I owe this reference to Paul Bullen (personal correspondence, 24 March 2002).
Redistributing Liberty A Commentary on Hardin Julius Sch¨alike
Can we, in the name of liberalism, protect illiberalism? Russell Hardin claims that we cannot, that this would be incoherent.1 He points to some examples of illiberal practices in the light of which his claim appears quite plausible because of the drastic level of the violence involved. But Hardin not only thinks that a liberal should not tolerate these practices; he rejects the concept of group liberalism altogether, which he takes to be incompatible with the central ideas at the heart of liberalism. I would like to stress that we face two different questions here, the question of the justification of group rights generally, and the question of the justification of toleration or support of illiberal groups. These two projects of justification don’t stand and fall together: one could be in favor of group rights in general, but against the protection of illiberal groups. Not all rights that groups demand are privileges to coerce their members. Will Kymlicka introduces the useful distinction between group-claims for internal restrictions and for external protections. The former are claims for illiberal practices, but the latter need not involve the group’s illiberalism in any way. They may be claims of one of several kinds: one possibility is a claim for special group representation within the political institutions of the larger society in order to make it less likely that a minority will be ignored in the making of decisions on a countrywide basis. A second is the claim for polyethnic rights to protect specific religious and cultural practices which might not be adequately supported by the market.2 Obviously, group liberalism of the internal restriction variety is more problematic. I agree with Hardin that practices like forced marriages and honor killings should not be tolerated, and certainly not be supported by a liberal society. But my reason is not that I think group liberalism in general is incompatible with the core idea of liberalism. Quite the opposite, I think it to be an acceptable liberal concept for the same reasons I think a liberal should endorse the other liberalisms. Group liberalism is indeed likely to conflict with social liberalism, as Hardin points out, but this would only prove that it is not a coherent liberal concept if social liberalism were to follow analytically from the core ideas of liberalism. But I think the various 1
Hardin: this volume.
2
Kymlicka (1995), p. 37f.
J. K¨uhnelt (ed.), Political Legitimization without Morality? C Springer Science+Business Media B.V. 2008
199
200
J. Sch¨alike
liberalisms are better conceived as spellings out of the core ideas of liberalism in the face of specific empirical situations. They generate only pro tanto claims for liberation, which in case of conflict need to be balanced against each other in the light of the core ideas. This needs explanation. Hardin says the central idea of liberalism is that it liberates. The liberty that interests us in the present context is liberty from restrictions imposed by other persons. As Hayek said, the extreme case of a person lacking this kind of freedom is a slave, not a paralyzed man or a rock climber trapped in a crevasse.3 Hardin points out that the original liberalisms, of the political and the economic kind, as well as some of the newer ones, much as social, institutional and welfare liberalism, can be understood as liberating individuals from the control of other persons or institutions. Furthermore, all of them share the game theoretical structure of self-enforcing mutual advantage regimes which gives them survival power. In these respects group liberalism differs. It does not liberate individuals, but quite the contrary subordinates them to group customs and values; and it is not self-enforcing because it is not to the advantage of the most important groups in liberal societies and it conflicts with the interests of the next generation. While the other liberalisms can be combined harmoniously, group liberalism is different: it obviously points in the opposite direction to social liberalism, and this makes them incompatible. I grant that these observations of Hardin’s are correct: in all these respects group liberalism does not fit in with the other liberalisms. But should we conclude from this that it is incoherent as a variety of liberalism? I don’t think so. To explain why, I want to direct attention to a fact that could easily be overlooked: the liberations that the liberalisms of the uncontroversial kinds call for are not liberations that can be accomplished for free. They come at some cost, the currency of which is liberty again. For example, to liberate a slave obviously is an act required by political liberalism. But even in this case, the liberation is accomplished at the cost of some loss of freedom – the freedom of the slaveholder to command. The same holds for welfare liberalism: the aim is to liberate people from the constraints that result from the lack of economic means, especially when the poor are not responsible for their misery. These constraints can be interpreted as restrictions imposed on the poor by other persons, the rich, because the economic means are in their hands and their refraining from passing them on to the have-nots prevents the liberation of the latter. The poor cannot choose as they wish because the fact that other people refrain from helping blocks the options prerequisite for choice. The liberation can be brought about by redistributive taxation. But since taxation reduces the liberty of the rich to do and buy things they want, it needs a justification. There are libertarians who describe redistributive taxation as a kind of robbery or, like Robert Nozick,4 as on a par with forced labor. Liberation comes at some cost, which is to be paid in the currency of liberty, and we need to know whose freedom
3
Hayek (1960), p. 12.
4
Nozick (1974), p. 169.
Redistributing Liberty
201
may be restricted how much in order to increase the liberty of someone else. In a world where all resources are already distributed, liberation means redistributing liberty. What are the principles of the redistribution? How can we decide who has to give, and who may receive something? Obviously, we are in need of a concept of distributive justice. The call for liberation alone does not settle the matter, because it leaves us without orientation.5 Like most liberal philosophers, I think the distributive principle should be constructed around the idea of equality. Everybody should be treated with the same respect, or be treated as equals, or their interests should be credited with the same value. Liberal egalitarian philosophers like Rawls, Dworkin and Kymlicka claim that equal respect calls for the rectifying of undeserved, unchosen inequalities.6 If there are no good reasons for an unequal distribution of freedom, we should aim at equality. In case of welfare liberalism, this means that, since the freedom of the rich exists at the expense of the freedom of the poor, there should be some redistribution by taxation.7 What does this mean for the issue of group rights? One could argue that if everybody’s individual rights are equally respected, no further claim for group rights can be justified. Would it not mean granting the members of specific groups undeserved advantages, which would contradict equality? The answer is no, because a state which grants its citizens only individual rights cannot be culturally neutral: it will make some days official holidays, some language the official language, and make some dress codes obligatory for government uniforms, and thus promote certain cultural identities and disadvantage others.8 In Germany and France for example religious symbols are forbidden at school in order to guarantee the religious neutrality of the state; but obviously this means a disadvantage for people who think that their religion prescribes a dress code, like the Muslim head scarf. It may be questioned if this disadvantage weighs so heavily that Muslim women should be allowed to wear head scarves at school or if the interests that are protected by religious neutrality weigh heavier. Anyway, there can be no doubt that Muslim women would be disadvantaged if they were not granted a
5 The common differentiation between “libertarian liberalism”, as concerned with individual liberty, and “egalitarian liberalism”, as concerned with principles for the mitigation of inequalities in political and economic power, is to my mind illusory. I agree with Ernst Tugendhat that “there is something in the concept of liberty itself which points beyond libertarianism to liberalism in the egalitarian sense.” Tugendhat (1988), p. 353. 6
Kymlicka (1995), p. 109.
7
We don’t have to decide in the present context whether we should go so far as to strictly give everybody the same amount of material goods, or allow inequalities if this would give a boost, for example, to the total amount of welfare or to the well being of the least advantaged (as Rawls thought). The important point is that liberalism has to embrace some distributive principle, and that it is difficult to see why this principle should not be constructed around the idea of equality in such a way that we need good reasons to depart from equal distribution. 8
Ibid. p. 108.
202
J. Sch¨alike
special right to wear a head scarf. They would lack the liberty to reconcile their job as a teacher and their religious obligations, a liberty that their Christian colleagues, if they are not monks or nuns, could enjoy. Another example illustrates the same point: Will Kymlicka tells of Canadian aboriginal fishermen for whom fishing is an important aspect of their culture. The Canadian Supreme Court guaranteed them special fishing rights to “ensure that they are not outbid or outvoted by the larger society on decisions regarding access to fishing.”9 Fishing for them is more important than for the other fishermen, who could easily be compensated with money for a temporal prohibition of fishing. The group right does not privilege the aboriginals but only puts them on an equal footing regarding access to culturally important activities with the members of the larger society whose culture is not endangered by fishing regulations. That’s why I think that group rights are in principle compatible with egalitarian liberalism. But what about the hard cases Hardin has in mind when he rejects group liberalism, cases of illiberal groups that demand the right to restrict the liberty of their group’s members? I completely agree with Hardin that honor killings, forced marriages and the prohibition of apostasy cannot be tolerated by a liberal, but for different reasons. Groups may demand internal restrictions because they think that without them their groups would decay, and this would deprive them of the basis of their culture. This loss of cultural basis would be a very serious disadvantage because a cultural membership cannot be changed as easily as a job. And it would be a disadvantage in terms of liberty, because the people affected would no longer be able, unlike the members of the larger society, to engage in cultural practices that presuppose community participation. This gives rise to a pro tanto reason for group rights that allow internal restrictions. But the liberal has to take into account the people who would suffer the restrictions, too. To meet the requirements of egalitarianism, the liberal has to put these interests on the scales and, for reasons to be explained, they outweigh the interests of the traditionalists who want to live according to the rules of their culture and to protect its continuity. Obviously, the illiberal traditionalists don’t agree with this evaluation of interests, because they don’t agree that interests have to be evaluated on a neutral basis. But their agreement is irrelevant for the liberal’s moral judgment. The liberal may not ignore the interests of the illiberal traditionalist, for example the interest that everybody obeys the rules proclaimed by some god, but these interests are not the normative standards for his moral judgment. The liberal’s normative yardstick is equal respect, and so the traditionalist’s interest is only to be taken into account as one interest among others, as an interest with a certain content and a certain strength. But under this description, it is to be considered and may not be ignored simply because it is illiberal in its content. When it is taken into account in this way, it generates a pro tanto right for the liberty to realize the illiberal interest – a right that may be outbalanced by the rights of other people involved, and that will, in all contexts realistically to be expected, be outbalanced if the competing interests contain the interest to autonomously decide
9
Ibid. p. 44.
Redistributing Liberty
203
whom to marry, which god to pray to, and how to deal with one’s sexuality. One way to vindicate this claim is suggested by John Stuart Mill, who says that direct harm is worse than indirect harm: it is much worse to suffer coercion to live in a way that pleases others, or even be killed in the name of some archaic concept of honor, than to bear, as a bystander, the fact that someone else lives in a way that one dislikes.10 A second rationale is that, as Hardin reminds us, a lot of people will suffer, large parts of the next generation. And since we have to take into account the generations still to be born as well, the number of people coerced, but who would enjoy autonomy in the absence of the illiberal culture, would tend towards the infinite. This makes the dying-out of illiberal culture highly desirable. The liberal will expect the traditionalist to tolerate the claim of the group’s members for autonomy, whether it leads to apostasy and disintegration of the group or not. And he will not only reject the direct blocking of choice by coercion, but also indirect blocking by means of restricting access to the perception of options through educational bias and refusal, as practiced by the Old Order Amish. The consideration of future generations puts a new light on the group rights of the external protection kind: these rights may be costly, because they involve subsidies which the members of the larger society have to pay. This means they lose some of their liberty by losing their money. So it would be best to set a deadline for the subsidies, combined with favorable options for the younger generation to leave the group and become integrated into the larger society. This should be done in a way that guarantees a high degree of autonomy for the “dropouts” and equips them with acceptable live chances by means of education. In this way, hardships for the traditionalists whose ties to their culture are so strong that they are not able to live autonomously outside of it are prevented, while, in the long run, liberty is maximized.
References Hayek, F. A. (1960): The Constitution of Liberty, Chicago. Kymlicka, W. (1995): Multicultural Citizenship. A Liberal Theory of Minority Rights, Oxford. Mill, J. S. (1975 [1859]): On Liberty, Harmondsworth. Nozick, R. (1974): Anarchy, State, and Utopia, New York. Tugendhat, E. (1992 [1988]): “Liberalism, Liberty, and the Issue of Economic Human Rights”, in: Philosophische Aufs¨atze, Frankfurt/M.
10
Mill (1859), Chap. 1. – “There is no parity between the feeling of a person for his own opinion, and the feeling of another who is offended at his holding it; no more than between the desire of a thief to take a purse, and the desire of the right owner to keep it.” Ibid. Chap. 4.
The Authors of the Volume
Christopher W. Morris is Professor of Philosophy at the University of Maryland, College Park (USA). His main interests are moral and political philosophy, philosophy of law, practical rationality. He is the author of An Essay on the Modern State (Cambridge University Press, 1998), the editor of several collections, including Practical Rationality and Preference – Essays for David Gauthier, (co-edited with Arthur Ripstein, Cambridge University Press, 2001) and The Social Contract Theorists – Critical Essays on Hobbes, Locke, and Rousseau (Rowman and Littlefield 1999). He is at present writing about the state system and international order, legitimacy, and justice. Neil Roughley is presently Visiting Professor for Practical Philosophy at the University of Muenster (Germany). His areas of interest are action theory, ethics, philosophical anthropology, philosophy of mind and aesthetics. He has edited several books on philosophical anthropology, including Being Humans – Anthropological Universality and Particularity in Transdisciplinary Perspectives (de Gruyter 2000) and Anthropologie und Moral – Philosophische und soziologische Perspektiven, (co-edited with M. Endress, Koenigshausen und Neumann 2000). His Wanting and Intending. Elements of a Philosophy of Practical Mind is soon to be published by Springer. Wilfried Hinsch is Professor of Practical Philosophy at RWTH Aachen University (Germany) and has a regular Visiting Professorship at the Coll`ege d’Europe in Brugge (Belgium). Since February 2006 he is a member of the German Science Council. His main fields of research are moral and political philosophy with a current focus on human rights, international justice, and military intervention. Three recent publications are Severe poverty – Weak and strong (with M. Stepanians, in: Føllesdal/Pogge (eds.): Real World Justice, Springer, 2005); Human Rights as Moral Claim Rights (with M. Stepanians, in: Martin/Reidy (eds.): Rawls’s Law of Peoples, Blackwell, 2006), and Menschenrechte milit¨arisch sch¨utzen – Eine Verteidigung der humanit¨aren Intervention (C.H. Beck 2006). ¨ Michael Kuhler is a postdoctoral research fellow at the Department of Philosophy and the Collaborative Research Centre (SFB) 485 “Norm and Symbol” at the University of Konstanz (Germany). His main research interests are located in the 205
206
The Authors of the Volume
field of practical philosophy, especially moral philosophy, metaethics, and political philosophy. Recent publications are Moral und Ethik – Rechtfertigung und Motivation. Ein zweifaches Verst¨andnis von Moralbegr¨undung (Mentis 2006), Did you do something wrong if you couldn’t have done otherwise? – Deontic act evaluation and some doubts concerning “ought implies can” (Pre-Proceedings of the 29th International Wittgenstein Symposium, 2006) and Political contract theories and freedom of speech (in: Evgeny Arinin (ed.): Politics and Liberties, forthcoming). Frank Dietrich is Assistant Professor of Practical Philosophy at the University of Leipzig (Germany). His main areas of research are political philosophy, moral philosophy and medical ethics. He is the author of Dimensionen der Verteilungsgerechtigkeit (Lucius and Lucius 2001) and is currently working on a book on the justification of secession. His articles on questions of political self-determination have appeared in Analyse and Kritik (2005/2), Deutsche Zeitschrift f¨ur Philosophie (2006/4) and Leviathan (2007/1). Reinhard Zintl is a political scientist (Ph.D., University of Freiburg 1972) and an economist (Master, University of Regensburg 1976) and has been a Kennedy Fellow at Harvard University 1977/78. He holds the chair of political theory at the University of Bamberg (Germany). Research areas are constitutional political economy, work on applicability and limitations of Rational Choice in political theory, normative theories of political institutions and liberal theories of justice. His publications include Politische Philosophie (co-author with Michael Becker and Johannes Schmidt,UTB, 2006), Politik und Markt (co-editor with Roland Czada, Vs Verlag 2004), Political Equality and the Prospects of Democracy (in: The Tampere Club (ed.): The Future of Democracy, Tampere University Press, 2003), Rational Choice as a Tool in Political Science (Associations, 5, 2001), Constitutions as External Constraints? (in: J. Nida-R¨umelin, W. Spohn, (eds.): Rationality, Rules, and Structure, Powell, 2000), Social Relationships and Economic Theories of Clubs (Associations, 1, 1997). Rafaela Hillerbrand is currently working as a senior researcher at the Faculty of Philosophy of the University of Oxford (UK). She holds a PhD in philosophy (University of Erlangen/N¨urnberg, 2003) as well as a PhD in theoretical physics (University of M¨unster, 2007). Her current work ranges from the use of models within the sciences to epistemic and moral questions related to decisions under risk and uncertainty. A particular focus of her work is on ethics of technology. ¨ Her thesis in philosophy Technik, Okologie und Ethik – Ein normativ-ethischer Grundlagendiskurs u¨ ber den Umgang mit Natur, Umwelt und Wissenschaft (Mentis 2005) was awarded the Lilli-Bechmann-Rahn-Preis. This book includes an in-depth analysis of the moral issues related to global warming. She has edited an Ebook Beyond the Global Village – Environmental Challenges inspiring Global Citizenship on environmental issues. Matthias Kaufmann studied mathematics (diploma 1980), philosophy and political science (Dr. phil. 1986). He is currently professor in philosophy (ethics) at the
The Authors of the Volume
207
Martin-Luther-University Halle-Wittenberg (Germany). He was visiting Professor at the Humboldt University Berlin (1992), at the University of M¨unster (1994) and at the UECE (Universidade Estadual do Cear´a) in Fortaleza (Brazil) (2005). His fields of work are political philosophy, philosophy of law, philosophy of language (including problems of intercultural translation) and bioethics, medieval and modern philosophy. He is author of the books Begriffe, S¨atze, Dinge – Referenz und Wahrheit bei Wilhelm von Ockham (Brill 1994) and Aufgekl¨arte Anarchie – Eine Einf¨uhrung in die politische Philosophie (Akademie Verlag, 1999) Stephan Schlothfeldt (University of Konstanz) is currently Visiting Professor for Practical Philosophy at the Saarland University (Germany). His areas of interest are Applied and Theoretical Ethics, Political Philosophy, and Philosophy of the Social Sciences. Major publications are Erwerbsarbeitslosigkeit als sozialethisches Problem [Unemployment as a problem of social ethics], (Alber 1999), and Pflichten von ¨ Individuen und Gruppen bei der Behebung gravierender Ubel [Duties of individuals and groups concerning the elimination of grave evils], (habilitation; publication in preparation). ¨ J¨org Kuhnelt has been a research fellow at the Collaborative Research Centre (SFB) 485 Norm and Symbol at the University of Konstanz (Germany). He received his PhD in philosophy from the University of Konstanz in 2007 with the thesis Pluralismus und Vertragstheorien [Pluralism and Contract Theories]. In this book, he discusses the potential of current contractarian theories in relation to pluralistic societies and develops a modified version of the contractarian argument. His areas of interest are political and moral philosophy, economic theory, theory of film-analysis and aesthetics. Starting in September 2008 he will be a Visiting Scholar at the New York University. Philip Pettit is Laurance S. Rockefeller University Professor of Politics and Human Values (USA). He works in moral and political philosophy and on background issues in philosophical psychology and social ontology. His books include The Common Mind (Oxford University Press, 1996), Republicanism (Oxford University Press, 1997), A Theory of Freedom (Oxford University Press, 2001), Rules, Reasons and Norms (Oxford University Press, 2002) and Penser en Societe (PUF, 2004). He is the co-author of Economy of Esteem (Oxford University Press, 2004), with Geoffrey Brennan; and Mind, Morality and Explanation (Oxford University Press, 2004), a selection of papers with Frank Jackson and Michael Smith. A new book, Made with Words: Hobbes on Mind, Society and Politics appears in 2007 from Princeton University Press and he is currently working on a book on group agency with Christian List, LSE. Gottfried Seebaß is Professor for Philosophy at the University of Konstanz (Germany), Chair for Practical Philosophy and the Theory of Social Sciences. His main research fields are the philosophy of action and motivation, the theory of freedom and the philosophy of language. His book publications are Das Problem von Sprache und Denken (co-editor with R. Tuomela, Suhrkamp 1981), Social
208
The Authors of the Volume
Action (Reidel 1985), Wollen (Klostermann 1993), Handlung und Freiheit (MohrSiebeck 2006) and Die Bedeutung des Willensfreiheitsproblems, (Akademie Verlag, 2007). Michael Baurmann is Professor of Sociology at the University of D¨usseldorf (Germany). Main research interests are General Theory of Sociology, Rational Actor Theory, Theories of Trust, Sociology, Law and Ethics. Since 1979 he is founder and joint editor of Analyse & Kritik, Journal for Social Theory. He published books and articles in Social Sciences, Ethics and Theory of Law. Recent publications in English include The Market of Virtue. Morality and Commitment in a Liberal Society (Springer, 2001). Christoph Schmidt-Petri is a lecturer at the Chair for Economics and Philosophy at Witten/Herdecke University (Germany). He received a PhD in philosophy in 2005 from the London School of Economics, and has held temporary appointments at the universities of Konstanz and Glasgow. He has published in the Philosophical Quarterly and Philosophy of Science and is currently editing an anthology on Formal Epistemology (with Franz Huber) and an interview book on the Philosophy of Social Science (with Diego Rios). Russell Hardin is Professor of Politics at New York University (USA). Among his publications are One for All: The Logic of Group Conflict (Princeton University Press, 1995), Indeterminacy and Society (Princeton University Press, 2003), Liberalism, Constitutionalism and Democracy (Oxford University Press, 1999) and David Hume: Moral and Political Theorist (Oxford University Press, forthcoming). He writes on moral and political theory, with emphases on law, constitutionalism, social order, cooperation, conflict, and the cultural bases of these. Julius Sch¨alike is Assistant Professor of Philosophy at the University of Konstanz (Germany) He studied in Munich and Berlin. His topics of research comprise ethics, theory of action, moral responsibility and political philosophy. He is author of the book W¨unsche, Werte und Moral – Entwurf eines handlungstheoretischen und ethischen Internalismus (K¨onigshausen and Neumann 2002), and various articles on akrasia, ethics, moral responsibility and political philosophy.