UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The standpoint of justice and conflicting liberties : John Rawls’s conception of freedom Dalibor, Marek 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-0091.pdf [ 4.77MB ]
Metadata
JSON: 831-1.0088516.json
JSON-LD: 831-1.0088516-ld.json
RDF/XML (Pretty): 831-1.0088516-rdf.xml
RDF/JSON: 831-1.0088516-rdf.json
Turtle: 831-1.0088516-turtle.txt
N-Triples: 831-1.0088516-rdf-ntriples.txt
Original Record: 831-1.0088516-source.json
Full Text
831-1.0088516-fulltext.txt
Citation
831-1.0088516.ris

Full Text

THE STANDPOINT OF JUSTICE AND CONFLICTING LIBERTIES JOHN RAWLS'S CONCEPTION OF FREEDOM by MAREK DALIBOR B.A. The University of York, GB, 1996 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Political Science) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA February 1998 © Marek Dalibor, 1998  In  presenting  degree freely  at  this  the  thesis  in  partial  fulfilment  University  of  British  Columbia, I agree that the  available for  copying  of  department publication  this or of  reference  thesis by  this  for  his thesis  and study. scholarly  or  her  for  of VO^TicAL-  financial  SUEMCB  The University of British Columbia Vancouver, Canada  DE-6  (2/88)  I further  purposes  gain  the  requirements  agree  may  representatives.  permission.  Department  of  It  shall not  be is  that  Library  an  advanced  shall make  permission for  granted  by  understood be  for  allowed  the that  without  it  extensive  head  of  my  copying  or  my  written  11  Abstract: In A Theory of Justice, John Rawls attempts to develop an alternative conception of justice to that provided by utilitarian theories. His critique of the latter consists mainly in the conflation of individuals into one which needs to take place in order to extend the concept of individual satisfaction of desires (utility) to a social concept. Rawls instead postulates the "inviolability of the person" against being considered merely a part of a greater whole and potentially being sacrificed for the benefit of others. From this "inviolability of the person', Rawls constructs his "Original Position", which I interpret as the "Standpoint of Justice", the position which specifies what kinds of arguments are acceptable as justifications for particular principles of justice. Most importantly, the inviolability of the person is expressed through the absence of knowledge of one's particular conception of the 'good' in the Original Position. Conceptions of the good are understood to belong to the domain of the individual, and no such conception can serve as a universally shared basis of justification. "Freedom", in this respect, consists in the right not to have another's conception of the good imposed on oneself. This is "freedom as an end". However, in order to deduce specific principles of justice from the Original Position, Rawls needs to introduce assumptions about what ends the parties there are following. He thus introduces the notion of "rational plans of life" and of "primary social goods" which serve as means to the realization of plans of life, whatever these may in particular be. Freedom, here, consists of a list of "basic liberties", which are part of the set of "primary social goods". Rawls thus introduces the notion of "freedom as a means". This gives rise to a conflict between the two aspects of freedom which Rawls is unable to solve.  iii  T A B L E OF CONTENTS Abstract  ii  Table of Contents  iii  Chapter One: Introduction  1  Chapter Two: Utilitarianism vs. Intuitionism  7  2.1: 2.2: 2.3: 2.4:  Intuitionism Utilitarianism Reflective Equilibrium Summary  8 11 15 18  Chapter Three: A Theory of Justice and Values 3.1: 3.2: 3.3: 3.4:  Teleological vs. Deontological Theories Individual Conceptions of the Good Publicly Shared Values Summary  Chapter Four: Rawls's Individualism: The Inviolability of the Person 4.1: Pure Procedural Justice 4.2: The Hypothetical Contract 4.3: Summary Chapter Five: The Standpoint of Justice: The Parameters of the Original Position 5.1: The Veil of Ignorance 5.2: The Parties in the Original Position 5.3: Summary  19 19 21 22 23 25 25 29 38 40 40 45 50  Chapter Six: Rational Choice in the Original Position: The "Thin Theory of the Good" 52 6.1: 6.2: 6.3: 6.4: 6.5: 6.6:  Rational Choice The "Thin Theory of the Good": Rational Plans of Life Primary Goods The "Thin Theory of the Good" as an Account of Human Nature Maximizing One's Share of Primary Goods Summary  54 57 60 65 67 69  iv  Chapter Seven: The Two Principles of Justice and Freedom  71  7.1: Rawls's Two Principles of Justice 7.2: The Original Position and Risk 7.3: Rawls's Conception of Freedom 7.3.1: The First Principle of Justice 7.3.2: Freedom as an End-in-itself 7.3.3: Freedom as a Means 7.4: Conflicting Liberties 7.4.1: The Constitutional Convention 7.5: Summary  71 74 79 79 82 85 90 92 95  Chapter Eight: Conclusion Bibliography  97 101  1 Chapter 1: Introduction  "Each member of society is thought to have an inviolability founded on justice [...] wh  even the welfare of every one else cannot override. Justice denies that the loss of free some is made right by a greater good shared by others." (TJ, p.28) The inviolability of the person, the notion that there are limits to what can justifiably be done to and expected from a person, lies at the heart of John Rawls's Theory of Justice. Rawls attempts to develop an account of justice that provides a basis by which citizens can adjudicate their claims against each other while at the same time imposing boundaries on what such claims can consist of and how far they can go. Whilst justice is a social concept and necessarily entails both rights and duties of citizens against each other, the "inviolability of the person" implies that what these rights and duties are must not be determined solely with reference to some "greater good of all". In this way, the concepts of freedom and of justice are strongly linked in Rawls's theory. To the extent that freedom is part of the inviolability of the person it must be expressed and protected by the principles of justice to be developed. Part of my argument will be to show how Rawls attempts to develop a conception of justice that protects the individual from infringements in the name of others' welfare and how well he achieves it.  1  I shall argue that in addition to a conception of freedom as the protection of the inviolability of the person, which is freedom as an end in itself, Rawls must use a conception of freedom as a means to the end of realizing what he calls one's "rational plan of life". Freedom as a means, however, I shall argue, gives rise to the potential for conflict between liberties (which freedom as an end does not).  *A more complete description of the meaning of the "inviolability of the person" is to be found in my analysis of Rawls's critique of utilitarianism in chapter 2.2.  2 As various commentators have pointed out, Rawls does not provide a solution to the question as to how such conflicts should be resolved. Brian Barry, for example, points out  that: "Rawls gives little usable guidance about the way to aggregate the different liber  as to arrive at an estimate of the total amount of liberty generated by alternative combinations of these liberties." (Barry, p.34) The main aim of this paper is to show the following three points: 1) Why and how conflicts between liberties arise in Rawls's account of justice and what they consist of; 2) that Rawls does not offer a solution to this problem; and 3) that this problem cannot possibly be solved within Rawls's framework. In order to do this I shall attempt a close reading of the derivation of Rawls's two principles of justice. This will allow me to unravel the underlying thread that goes through his argument. By taking seriously the assumptions he makes and the deductions he proposes it will be possible for me to single out what the necessary implications for Rawls's conception of freedom are. It is my intent to offer a critique of Rawls's argument by the very standards he sets himself. I hope to show that the conception of freedom developed in the Theory of Justice is unsatisfactory in a way that Rawls himself would have to admit. This method of looking at how different parts of Rawls's argument have necessary implications for the meaning and relevance of other parts leads me to an interpretation of the Theory of Justice, particularly of the role of the "original position", which differs significantly from a lot of the prominent literature on Rawls. I shall argue against the view that Rawls's "original position" represents primarily a rational choice situation in which transcendental  3 subjects choose the principles of justice they wish to agree upon by means of a contract. By contrast, my interpretation of Rawls's argument is closer to T.M.Scanlon's, who sees Rawls's aim as that of finding the "methods of reasoning acceptable to all". The importance of this different interpretation will become clear in my discussion of the "original position" in chapter 5. Some important criticisms that arise out of interpreting the "original position" solely as a rational choice model can thus be answered.  Part of the problem with analyzing Rawls's conception of freedom in the Theory of Justice is that it is not summed up in one place, but rather dispersed throughout the argument. It is therefore going to be necessary to draw upon very different sections of the text. In order to make this intelligible I shall give a brief summary of Rawls's argument: Rawls proposes an original position characterized by the free association of individuals on the basis of equality. In this original position, he argues, the principles of justice would rationally be chosen. In order not to have the principles of justice contaminated by purely particularistic motivations, but rather to insure that the principles chosen would be justifiable to any rational individual, Rawls introduces a "veil of ignorance" behind which all knowledge of particularities is concealed. Specifically, the parties in the original position are to make their deliberations without reference to their particular conceptions of the good, their status in society, or their talents and inclinations. These, Rawls argues, are morally irrelevant in a conception of justice. The only knowledge the parties in the original position have is that they possess rational plans of life, whatever their content may be, and that for the execution of any such plans they require "primary social goods", namely "rights and liberties, power  See for instance Sandel: Liberalism and the Limits of Justice, or Wolff: Understanding Rawls. Other examples where this is explicit include Barber, Kukathas and Pettit, Dworkin, and Frazer and Lacey 2  4 and opportunities, and income and wealth". The original position thus becomes an instance of rational choice, where the participants are motivated by their desire to maximize their share of primary goods. The outcome of this choice, Rawls argues, would be his two principles of justice, which state that a) every member of society has an equal right to the greatest possible liberty, and that this liberty cannot be sacrificed for material gains for others or even for themselves (First Principle plus "priority rule"); and b) that inequalities in the distribution of primary goods are justified if and only if they are to the benefit of the least advantaged (maximin rule: Second Principle). Liberty thus features twofold in these principles: in the first principle explicitly, and implicitly as amongst the primary goods in the second principle. How this particular conception of freedom is arrived at, and what it entails, will be the focus of this paper. My argument will proceed as follows: Chapters 2 to 5 shall be concerned mainly with the significance of Rawls's insistence on the "inviolability of the person". In chapter 2: "Utilitarianism vs. Intuitionism", I shall analyze Rawls's critique of utilitarianism in which the inviolability of the person plays an essential role. Chapter 3: "A Theory of Justice and Values" contains a brief discussion of the basis for Rawls's insistence on this inviolability and the relationship between it and "individual conceptions of the good". This sets the stage for chapter 4: "Rawls's Individualism: The Inviolability of the Person", in which I argue that the inviolability of the person is one of the main determinants of the structure of Rawls's argument. In this context the role of the concept of "pure procedural justice" which Rawls attempts to produce is examined and explained. That Rawls wishes to provide an account of pure procedural justice has strong implications for an understanding of  5 the original position, as I shall argue. Chapter 5: "The Standpoint of Justice: The Parameters of the Original Position", then, contains an in-depth discussion of the significance of the original position and the way the inviolability of the person impacts upon how and why the veil of ignorance is introduced. Chapter 6: "Rational Choice in the Original Position: The Thin Theory of the Good" proceeds from my interpretation of the original position. Here, the original position is taken as established and I consider how Rawls proposes to deduce the principles of justice from it. I shall argue that the notion of rational choice, albeit useful, constitutes a break in the argument by introducing ends other than that of securing the inviolability of the person. I shall look at Rawls's "thin theory of the good" which is required to make rational choice in the original position possible and discuss how this "thin theory" relates to individual conceptions of the good. Finally, in chapter 7: "The Two Principles of Justice and Freedom" I shall be in a position to offer an interpretation of Rawls's conception of freedom. I shall briefly discuss Rawls's two principles of justice, but focus my attention on the conception of freedom expressed through these. I shall argue that whilst the lexical ordering of the two principles can only be justified on the basis of freedom as an end-in-itself, the two principles of justice can only be understood as an outcome of rational choice if freedom as a means is also included. It is the latter that gives rise to conflicts between liberties, since their usefulness as means will depend on the particular ends an individual maintains. The relative weights placed on different liberties will therefore be contingent on individual preferences. On the other hand, a universal method by which to arrange the different liberties-as-ends into a "total system of liberty", as Rawls proposes, cannot be found unless the individual preferences regarding the relative  6  weights of liberties are brought together and somehow aggregated into one system. This, however, would violate people's freedom-as-an-end. I shall therefore argue that Rawls's Theory of Justice  cannot resolve conflicts between liberties and must remain indeterminate.  This, in turn, is not an acceptable outcome from the point of view of the original position. My argument will be that Rawls can do nothing to avoid this problem within his framework. A conceptualization of how freedom-as-a-means and freedom-as-an-end relate to each other and can be unified is necessary but lacking.  7 Chapter 2: Utilitarianism vs. Intuitionism  In the preface to his Theory of Justice, John Rawls situates his theory with respect to a conflict between utilitarian and intuitionist accounts of justice. "During much of modern moral philosophy," he writes, "the predominant systematic theory has been some form  utilitarianism. [... Its critics] pointed out obscurities of the principle of utility and noted th  apparent incongruities between many of its implications and our moral sentiments. But t  failed, I believe, to construct a workable and systematic moral conception to oppose it. T  outcome is that we often seem forced to choose between utilitarianism and intuitionism. M  likely we finally settle upon some variant of the utility principle circumscribed and restri  in certain ad hoc ways by intuitionistic constraints. Such a view is not irrational; and there no assurance that we can do better. But this is no reason not to try." (TJ, p.viii) Rawls attempts to provide a theory that is, on the one hand, not prone to implications which contradict our intuitive feelings on justice (like utilitarianism does), but is, on the other hand, systematic and comprehensive (within the scope of justice), enabling us to resolve conflicts between our intuitive principles of justice (unlike intuitionism)". Given this, it seems helpful to begin a discussion of Rawls with a closer look at his critiques of both utilitarianism and intuitionism in order to bring out some of the aims of his theory of justice. I shall therefore consider the critiques of intuitionism and utilitarianism in turn, and thence describe Rawls's strategy of "reflective equilibrium" which he offers as a method of ensuring the adequacy of his theory as a systematic alternative to utilitarianism. This will set the stage for my interpretation of Rawls's theory of justice. I believe it is important to  3  TJ, p.34  8  understand Rawls in the context of the struggle between our shared intuitions and the systematic account offered by utilitarianism.  2.1 Intuitionism  "/ shall think of intuitionism in a more general way than is customary: namely, as th  doctrine that there is an irreducible family offirstprinciples which have to be weigh against one another by asking ourselves which balance, in our considered judgement,  most just. [...] While the complexity of the moral facts requires a number of distinc  principles, there is no single standard that accounts for them or assigns them their wei (TJ,p.34) In Rawls's view, intuitionist theories of justice are marked by the fact that they rely on a pluralism of first principles with no systematic account of how these principles are to be weighed against each other. Such weighing must be done intuitively on a case-by-case basis. The intuitionists' first principles can be of various levels of generality, but ultimately fall short of a level of abstraction where all claims can be weighed. It is not the case that 4  intuitionist accounts of justice can never contain methods for resolving conflicts between intuitions, only that there is no one final level at which this can be done. As an example, Rawls mentions conflicting intuitions concerning just taxation (TJ, p.36). It is possible for an intuitionist account to maintain that such conflicting intuitions should be weighed with reference to their effect on more general principles such as allocative efficiency, full  I am not sure whom, in particular, Rawls has in mind when he speaks of "the intuitionists". Most importantly, perhaps, it should be asked whether Kant is supposed to fall into this category. If so, then the term "intuitionist" may be considered a misnomer, for the Kantian first principles are surely not meant to be simply intuitive. For Rawls, however, what is important is in what way these first principles are to be compared and balanced against each other. As long as Kant does not provide a systematic method for doing so, he should be understood as one of the "intuitionists" in Rawls's sense. 4  9  employment, a larger national income, and its equal distribution (ibid.). However, these higher-order ends of social policy themselves cannot systematically be weighed against each other where they conflict. At some level, intuitionist accounts of justice must always fall back on the intuitive resolution of conflicts. "[The intuitionist] contends that there exists no expressible ethical conception which underlies these weights. A geometricalfigureor a mathematical function may describe them, but there are no constructive moral criteria that establish their reasonableness." (TJ, p.39)  5  The main justification for such a view is that moral problems are often too complex for us to solve them by reference to one simple criterion: "[The intuitionist] contends that attempts to go beyond these [first] principles either reduce to triviality, as when it is said that social justice is to give every man his due, or else lead to falsehood and oversimplification, as when one settles everything by the principle of utility. " (ibid.)  Rawls acknowledges the strength of this claim. There is no reason why we should assume that all moral problems can ultimately be solved. By contrast, upholding a single fundamental principle over all others may, if there is some truth to the intuitionists' claim, conflict with and violate our considered judgements on particular questions of morality. This, as I shall describe below, is Rawls's main criticism of utilitarianism. However, the intuitionist view of morality leaves much to be desired. It offers at best partial answers to moral questions. In the end, such questions must be answered with reference to intuition. One of the main drawbacks of this, most debilitating in an account of justice, is that individuals may not agree on how first principles are to be intuitively weighed. Since there is nothing systematic about these weights, there is also no basis for justifying them to others. In Rawls is alluding here to the representation of weights by indifference curves, as is done in economic theory. 5  10  Rawls's example (TJ, p.37), whereas one person may be indifferent between two possible 6  distributions of equality and total welfare (which are assumed to be first principles by which just taxation is to be determined), another may not be. And this can happen even though both agree that equality and total welfare are important determinants of what "just" taxation should consist of. Both individuals can only appeal to their intuition in making these judgements of weights and are therefore unable to convert each other. Again, Rawls admits that this may be an insurmountable result of the complexity of moral questions, and in itself this imperfection of intuitionist accounts of justice does not amount to a repudiation. However, if it were possible to come up with a systematic account of justice that enabled us to account for our intuitive weights, this new account would be preferable to the intuitionists'. A "systematic" account of justice, then, would be one that entailed rules for weighing the different relevant aspects of justice. It must be an account that explains the basis of our judgements and provides guidelines by which these judgements can be arranged into a coherent whole. In order to provide such an alternative to the intuitionist account, the new account must not only be systematic; it must also fit our intuition. That is, it should not, in general (see below), supplant and contradict our intuitive feelings about justice, but rather explain them in a systematic way. This will give rise to what Rawls labels the process of "reflective equilibrium" which I shall describe below. First, let me turn to Rawls's critique of the most prominent systematic alternative to intuitionism, namely utilitarianism, to underline that not  "Indifferent", here, does not mean "disinterested" but strictly that the two outcomes are valued equally.  6  It should be noted that this line of thought falls neatly into the category of what Gilligan calls the "ethics of justice", and excludes the "ethics of care". Rawls does not consider this charge. His answer would probably be that the "ethics of care" represents an example of intuitionism and suffers from the same 7  11  any systematic account is acceptable.  2.2 Utilitarianism The concept of utility plays a major role in modern moral philosophy. It has engendered a g  vast amount of sophisticated literature in an ongoing debate. The main appeal of this concept is that it provides a basis for comprehensive, systematic accounts not only of morality, but of rational choice and human behaviour in general. At its most basic, "utility" means the satisfaction of desires, whatever these may be. The point is that there is one and only one standard of measurement by which any such satisfaction can be measured: utility. This is the common basis of all forms of utilitarianism. In spite of the different forms of utilitarianism, Rawls proposes to discuss only one form, classical utilitarianism. His justification for doing so is that his objective is to develop an alternative to all forms of utilitarianism using classical utilitarianism as an example (TJ, p.22), in other words to the use of the concept of utility as such. Thus he writes: "The kind of  utilitarianism I shall describe here is the strict classical doctrine which receives perhap  clearest and most accessible formulation in Sidgwick. The main idea is that society is r ordered, and therefore just, when its major institutions are arranged so as to achieve  greatest net balance of satisfaction summed over all the individuals belonging to it." (T p.22 ) 9  Utilitarianism, as a moral theory, takes the concept of individual utility (satisfaction of drawbacks as other forms of intuitionism. Whether this is a satisfactory reply or not I shall leave open. (Gilligan, C : In a Different Voice. Cambridge: Harvard University Press, 1982) See for instance Samuel Scheffler: Consequentialism and its Critics. Oxford: Oxford University Press, 1988 8  Rawls is referring to Sidgewick's The Method of Ethics, 7th Edition; (London, 1907). (See his footnote 9  9  12 desires) and aggregates it over different individuals. This, in turn, allows utilitarians to determine the "best" outcome, that is, the one yielding the greatest utility, which it defines as the morally preferable and thus just outcome. This structure is common to all forms of utilitarianism, although they may differ in their account of how utility is to be aggregated. Sidgwick's is an example of "total utility". Alternatively, one could be concerned with average utility, or weigh the utilities of different individuals in any number of ways. In order for utility to be a concept that allows for a comprehensive morality, however, it is essential that some systematic method of aggregation exists. Aggregation is thus common to all forms of utilitarian theories of morality. It is this that Rawls objects to, using Sidgwick as an example. Rawls's main criticism of utilitarianism is that aggregating utility across individuals is unacceptable in a theory of justice, since it can produce results that contradict our (considered) moral intuition. For example, slavery can in principle be justified by the standards of utilitarianism (see TJ, p. 158). In general, almost anything can be done to an individual as long as the respective decrease in utility is outweighed by a greater increase in utility for others.  10  By contrast, Rawls tells us, by the "convictions of common sense", i.e. intuition, "each  member of society is thought to have an inviolability founded on justice [...] which even welfare of every one else cannot override. " (TJ, p.28) Rawls's argument is that utilitarianism fails as a systematic alternative to intuitionism on p.22) Note that this is true whether we consider total or average utility or any other method of aggregation. In fact, total and average utility yield identical results as long as the number of individuals in a given society is considered constant (average utility = total utility divided by the number of individuals). For a discussion of the difference between total and average utility when the number of individuals changes see for example: 10  13  because, albeit systematic, it cannot account for our moral intuition. Instead of explaining the underlying system which gives rise to our moral intuition, utilitarianism proposes a system which contradicts at least some of our considered convictions."  This is an important criticism of classical utilitarianism. However, in order to reject all forms of utilitarianism, it is not enough for Rawls to show that one kind of utilitarianism can sometimes produce unacceptable results such as slavery. When Rawls claims that what makes his account of justice different from classical utilitarianism applies to all forms of utilitarianism (TJ, p.22), his critique must go deeper than just pointing out undesirable results of the one form. In other words, his critique must be directed at something that is common to all forms of utilitarianism. Indeed, a large part of Rawls's critique of utilitarianism is concerned not so much with its outcomes, but with the method by which they are produced, namely that of aggregating across individuals: "[The utilitarian] view of social cooperation is the  consequence of extending to society the principle of choice for one man, and then, to m  this extension work, conflating all persons into one through the imaginative acts of  impartial sympathetic spectator. Utilitarianism does not take seriously the distinc between persons." (TJ, p.27) The "principle for one man" is that people (not just men) act in such a way as to satisfy their desires (utility). These desires, however, are purely individual. In order to make use of the notion of utility in a social context, it is necessary to assume that these desires can be known  Bayles, M.D.: Morality and Population Policy. Alabama: University of Alabama Press, 1980 1  "For the meaning and importance of our intuitions being "considered" see the next section.  14 by others. As a social theory, utilitarianism must therefore stipulate an "impartial observer" into whom all individuals are conflated so that they can know all individual desires as if they were their own (hence the term "impartial"). Only such an observer could accurately aggregate and compare individual utilities. In the eyes of the "impartial observer", then, the individual is nothing but a part of a larger whole, another term of the sum of utilities. If this is to be understood as a criticism of utilitarianism, Rawls must assume that our intuition tells us that there is something wrong with conflating persons in such a way. Indeed, this seems to be what Rawls is getting at. Consider, for instance, his reply to Mill's claim that in practice the precepts of justice and liberty will tend to maximize utility and should 12  therefore be upheld (TJ, p.26 ). By implication, in those few cases where utility maximization would legitimize the violation of liberty we ought to change our intuition. Rawls is quick to point out the contingency of Mill's point: "There is no reason in principle  why [...] the violation of the liberty of a few might not be made right by the greater good  shared by many. It simply happens that under most conditions, at least in a reason  advanced stage of civilization, the greatest sum of advantages is not attained in this wa (ibid.) If, again, we are to understand this as a criticism of utilitarianism rather than just a comment, Rawls must mean that our intuition tells us that this contingency is objectionable. This clarifies Rawls's point when he states our intuition as insisting on an "inviolability founded on justice" (TJ, p.28; see above). Our intuition, according to Rawls not only tells us that sacrificing the liberty of some for the benefit of others is wrong, but that there is something wrong with viewing liberty simply as a result of maximizing utility. Rawls is referring to Mill's Utilitarianism, ch.V (1863). See also Rawls's discussion on pp.209ff.  12  15 If Rawls is understood in this way, it becomes clear in what sense his criticism of utilitarianism applies to all its forms not just the "total utility" version. It applies to all those theories in which morality is considered from the standpoint of an "impartial sympathetic observer" into which all persons are conflated. What is wrong with such theories is not only that they may propose outcomes that are intuitively objectionable, such as slavery, but more fundamentally that they make justice and liberty contingent on calculations in which people are conflated into one. To present an alternative to the utilitarian account of justice, Rawls must thus develop a theory of justice that does not aggregate across people. To what extent he achieves this remains to be seen. For now, let me turn to the role intuitions play in this theory and analyze Rawls's concept of "reflective equilibrium" as a conceptual tool for ensuring that our intuitions are given their proper merit.  2.3 Reflective Equilibrium As we have seen, Rawls claims that utilitarianism fails as a systematic basis for a satisfactory account of justice because it violates our intuitions in an unacceptable way. His alternative, he attempts to show, matches and explains these intuitions to a far better degree and is therefore preferable. However, to provide strength to this line of reasoning, it needs to be shown to what extent our intuitions themselves are valid. Rawls does not go so far as to claim that all intuitions are necessarily correct. In fact, if they were, there would be little need to present an alternative to purely intuitionist accounts of justice. As Rawls argues, however, our intuitions do not always provide complete answers to questions of justice (TJ, pp.35-39). Intuitions may be vague and may contradict each other. Therefore, we must be careful to what  16  extent to trust them. Rawls's objective in developing an alternative to both intuitionism and utilitarianism is thus not simply to provide a systematic account of all our intuitions, but rather to engage our intuitions in a deliberative process. Ideally, what he is aiming at is what he calls a state of "reflective equilibrium". This state is reached by balancing the conflicting claims deriving from our intuitions on the one hand, and our systematic account of justice on the other. In other words, neither our systematic principles nor our intuitions are regarded as necessarily correct. If they do not match, they must both be reconsidered. By this process, some of our intuitions might actually be changed if we are convinced of the validity of our principles. Alternatively, if our particular intuitions are so strong that we fail to be convinced by the principles (as is the case with utilitarian aggregations across individuals, according to Rawls), then we must amend those principles. Thus, Rawls hopes, "by going back and forth,  sometimes altering the conditions of the contractual circumstances [Rawls's systema  basis], at others withdrawing our judgements and conforming them to principle, I ass  that eventually we shall find a description of the initial situation that both express reasonable conditions and yields principles which match our considered judgements pruned and adjusted. This state of affairs I refer to as reflective equilibrium." (TJ, p.20) The concept of reflective equilibrium thus serves as a basis of justification for a theory of justice. If the theory results in a reflective equilibrium, it would seem to provide a good alternative to both a purely intuitionist account of justice, which must take all intuitions as inherently valid and cannot ultimately resolve conflicts between intuitions, on the one hand, and a utilitarian account, which fails at convincing us of the invalidity of all the intuitions that it contradicts, on the other.  17  Note Rawls's use of the concept of reflective equilibrium as a "testing device'. It serves as a test to see whether the principles of justice a given theory proposes succeed in bringing together our intuitions in a systematic whole. As such it can be used to test various theories, whether these have been developed with the notion of reflective equilibrium in mind or not. Most importantly, in Rawls's case, it can be used to test utilitarian theories. This underlines, again, what Rawls's objection to the latter fundamentally consists of. It is not enough for him to maintain that utilitarian theories of justice can, for instance, justify slavery in particular circumstances. There is no reason, in principle, why we should not wish to modify our intuition about slavery after having carefully considered the particular utilitarian argument. Rawls instead maintains that our intuitions about slavery are so strong and so well-considered that we will under no circumstances be convinced to conform them to calculations of utility. One way he does this, as we have seen, is by stating that we intuitively envision an "inviolability of the person founded on justice." (TJ, p.28; see above) Utilitarianism categorically denies this by defining justice itself as the maximization of utility. If however it is our considered judgement that we wish to uphold the inviolability of the person founded on justice, then we will never be convinced by utilitarian justifications for slavery or other sacrifices of the person. This is part of what Rawls must show in his theory of justice if it is to provide an alternative to utilitarianism. He must show how our intuitive concern for the inviolability of the person can give rise to systematic principles of justice that are intuitively acceptable. I attempt to show that his use of the concept of freedom plays an important role in expressing this inviolability of the person, and that an understanding of this aspect of freedom is crucial for his conception of freedom as a whole.  18  2.4 Summary Neither utilitarianism nor intuitionism, according to Rawls the main alternative bases for theories of justice, are satisfactory. Intuitionism, on the one hand, leaves important conflicts open. In particular, since it ultimately relies on the intuitive weighing of principles, it cannot resolve conflicts between different people's intuitive weights. As a social theory it is therefore of limited use, for it provides no method by which we can justify principles of justice and their weights to one another. Utilitarianism, on the other hand, produces one single standard, the satisfaction of desires, as relevant for weighing principles. In this way, the resulting conception of justice can be objectively justified. However, in order to reach this level of "objectivity", Rawls argues, utilitarian accounts of justice must sacrifice the separateness of the person. In order to be able to aggregate utility, it must be assumed that individual desires and their satisfaction can objectively be known by an "impartial sympathetic observer", to whom the individual is nothing but part of the greater whole. This, according to Rawls, is intuitively unacceptable. Instead, he proposes, we envision an "inviolability of the person founded on justice". The task he sets himself, then, consists in the first instance in developing a "systematic" theory, one in which all principles can ultimately be weighed against each other in a manner that we can justify to each other, while at the same time upholding the inviolability of the person and thus not conflating individuals into one "impartial observer". Only if Rawls can achieve this will his theory result in a reflective equilibrium between our principles and our intuition and hence represent a workable alternative to both utilitarianism and intuitionism.  19 Chapter 3; A Theory of Justice and Values  As I have described above, Rawls's Theory of Justice does not originate in a normative void. It is not Rawls's intention to derive his account of justice from self-explanatory and self-justifying first principles. Instead, he takes some normative landscape, namely the set of our intuitions (whether considered or not) as given, and attempts to engage with it in a deliberative process. Before I go into an analysis of his particular account of justice, it will be helpful to take a closer look at the way Rawls treats normative content for a) it is important to understand the special role "conceptions of the good" play in his theory; and b) as I intend to show, there is a distinction between different kinds of values in Rawls's theory that impacts upon his conception of liberty.  3.1 Teleological vs. Deontological Theories As we have seen, one of the objectives of the Theory of Justice is to avoid the counter-intuitive methodology of utilitarianism, by which people are conflated into a single "impartial, sympathetic observer". One of Rawls's solutions to this problem is to construct a deontological rather than a teleological theory (TJ, p.30). According to Rawls, teleological theories, utilitarianism being one example, are those theories that define the good independently of the right, and, in turn, the right as maximizing the good. (Other examples of teleological theories include perfectionism and hedonism depending on how "the good" is specified, (TJ, p.25.)) In other words, in teleological theories, there must be a complete conception of all that is valuable, the good, before questions of justice, the right, can be asked. Justice itself has no independent value.  20  Specifically, questions of distribution, falling under the concept of right (ibid.), have no impact on the good and are therefore irrelevant in teleological theories. If they did carry weight, the good would no longer be prior to and independent of the right. By contrast, Rawls explains that his is a deontological theory, i.e. "one that either does not  specify the good independently from the right, or does not interpret the right as max the good." (TJ, p.30) He directly goes on to say that his theory is deontological in the second sense, in other words that its principles of justice will not necessarily maximize the good. However, only a couple of paragraphs later, Rawls also states that "in justice as fairness the concept of right is prior to that of the good." (TJ, p.31) Thus it is also the case that the good is not specified independently from the right, in particular "the principles of right [...] put limits on which satisfactions have value." (ibid.) Rawls's theory is therefore deontological in both of the above senses. One of the implications of Rawls's not being a teleological theory is that he does not need to presuppose a complete system of values. For a teleological theory do be able to provide a conception of justice, its conception of the good must be comprehensive. That is, there must be a systematic account of the good by which each individual's "good" (i.e. their utility or their level of perfection or their pleasure, etc.) can be measured and compared. A deontological theory does not need such a complete, universal account of the good. It is therefore possible to allow for a pluralism of incommensurable individual values.  21 3.2 Individual Conceptions of the Good Since a deontological theory does not require a universal account of the good, it is possible to allow for different individuals having different conceptions of the good. This means that each individual can have their own set of values, which are not necessarily commensurable with others'. Contrast this with utilitarian theories, where although each individual has a unique set of preferences or desires which can be satisfied to varying degrees, there is one standard, utility, by which all these individual values/preferences can be compared. This makes it possible to conflate them all into one, as we have seen. Rawls's introduction of individual and incommensurable conceptions of the good plays a major part in avoiding such a conflation and taking the separateness of persons seriously. As we shall see, this idea of individuals having their particular conception of the good is a main determinant of Rawls's conception of freedom and is to a large part responsible for giving Rawls's theory its particular individualistic flavour. However, it should be noted that the notion of individual conceptions of the good can only be taken so far. If Rawls were to claim that the separateness of the person consists in there being no objective grounds for inter-personal comparisons, then there would not be much room for a theory of social justice. In order to be able to even think about justice as a social concept, as Rawls does, there must be some shared meanings and values. Thus when Rawls proposes that every individual has their own conception of the good, he should not be taken to mean that all values are purely individual. At least some values are shared, whilst others are not. Note that it is not necessary for a deontological theory to provide a conception of the right strictly prior to that of the good. It is enough that the two should be interdependent, that the conception of the good includes a conception of the right.  22  3.3 Publicly Shared Values When Rawls talks about intuitionism and how utilitarian accounts of justice do not match our intuitions, it is evident that he must have in mind such a thing as "our" shared intuitions. He is not talking simply about his own personal intuitions which have no bearing upon anybody else. Instead, he appears to assume a set of commonly shared intuitions, particularly concerning the value of the person. In order for his process of reflective equilibrium to make sense where more than Rawls's personal intuition is concerned, there must be some values that are shared. But this creates some confusion in Rawls's argument. He is never quite explicit about which values (or even which kinds of values) are shared and which are left to the individual. As we shall see, the idea that there is no one conception of the good which applies to all persons is central to Rawls's theory and the derivation of his two principles of justice. On the other hand, it is equally essential that some normative concepts, that is concepts that would be part of a conception of the good, are shared. In particular, the conception of justice must be  shared for it to be workable: "Among individuals with disparate aims and purposes a shared  conception ofjustice establishes the bonds of civic friendship [...] One may think of a pu  conception of justice as constituting the fundamental character of a well-ordered hum association." (TJ, p.5) This conflict between publicly shared and individual values points to two rather disparate tendencies in Rawls's theory. As I have already mentioned, his idea of strictly individual conceptions of the good gives his theory a strongly individualistic character. So much so that he has been criticized by more communitarian commentators for having an a-social  23 conception of the person that does not allow for the deductions Rawls attempts to make. Michael Sandel, for instance, argues that "Rawls' conception of the person can neither  support his theory of justice nor plausibly account for our capacity for agency and  self-reflection; justice cannot be primary in the way deontology requires, for we ca coherently regard ourselves as the sort of beings the deontological ethic requires us (Sandel, p.65) On the other hand, the concept of reflective equilibrium presupposes a community with shared intuitions about justice. To the extent that these are relevant in shaping Rawls's theory, this latter acquires a rather more communitarian character. Kukathas and Pettit thus  argue that: "Despite the criticism he [Rawls] has always attracted from commentators  Sandel, who draw their ammunition from Hegel's arsenal, Rawls's developed theory  out to have a decidedly Hegelian flavour. [... This] is because he takes as his startin  existing social practices [...] and looks to examine them in the light of the values em in them." (Kukathas and Pettit, p. 144) The existence of such opposed interpretations of Rawls suggests that there is indeed a great deal of ambiguity in his work with respect to the relationship between publicly shared and purely individual values. I shall argue that Rawls depends heavily on both kinds of values, but is ultimately unable to combine them in a coherent fashion.  3.4 Summary The first instance by which Rawls attempts to secure the inviolability of the person is by establishing that each individual has their own particular conception of the good and that this must be respected. In other words, values are the domain of the individual. A theory of justice  24 that wishes to take the inviolability of the person seriously, like Rawls's, must therefore not presuppose a universally valid system of values. On the other hand, for him to be able to treat justice as a shared rather than an individual concept, Rawls must show that there is some basis for interpersonal comparisons. The conception of justice itself must be shared across society, and it must allow for the resolution of conflicts between individuals when these are concerned with justice. Rawls must therefore presuppose that some values are not part of the individual conceptions of the good, but rather of a set of shared intuitive values, such as the inviolability of the person. In doing so, Rawls potentially sets the stage for combining the notions of individuality and community into one. Both the idea that the individual and his/her values are prior to and independent of society on the one hand, and that the individual forms part of a society with shared values on the other are introduced into Rawls's theory. That he does not provide a conceptual framework which explains how these two notions relate, however, lies at the base of the problem he encounters with conflicts between liberties, as I shall argue.  25  Chapter 4: Rawls's Individualism: The Inviolability of the Person As we have seen, one claim Rawls makes concerning our shared intuition is about the value of the individual. In his critique of utilitarianism, he makes it clear that he believes in a strong conviction of the inviolability of the person. As I shall argue below, Rawls's individualism is to a large extent expressed through his claim that individuals should be free to pursue their own conception of the good. It follows from this that the publicly shared conception of justice should not be based on any one such conception of the good. In attempting to eliminate conceptions of the good from the derivation of the principles of justice, Rawls turns to the notion of "pure procedural justice". It is not, however, immediately clear from his writing what role pure procedural justice exactly plays in the theory of justice. Examining this role will bring out the importance of his assumptions about our intuitions about the person.  4.1 Pure Procedural Justice  "Pure procedural justice obtains when there is no independent criterion for the right re  instead there is a correct or fair procedure such that the outcome is likewise correct or f whatever it is, provided that the procedure has been properly followed." (TJ, p.86) In chapter II of his Theory of Justice, Rawls describes his account as one of pure procedural justice. As the above quote indicates, pure procedural justice is a matter of following a fair procedure. The question of justice here refers not so much to the outcome, but rather to the way in which it is produced. Rawls's example is gambling (ibid.). If all players in a gamble participate of their own volition and accept the terms of the gamble, then the outcome is just regardless of who wins and who loses. The final distribution of gains and losses says nothing  26  about justice. The only thing that matters is whether the gamble has been carried out according to the accepted rules. By contrast, Rawls argues, in perfect procedural justice, and its correlate imperfect procedural justice, there is an independent criterion for the right result and the fairness of the procedure is measured by the extent to which it can produce this result. Rawls's example here is dividing a cake (TJ, p.85). The intuitive idea is that an equal division is just. Thus justice becomes a matter of ensuring that this outcome will be reached. Depending on whether a procedure can be found that guarantees the just outcome, independently specified, we have perfect procedural justice; if not imperfect. (Letting the person who divides the cake pick their share last is Rawls's example of a perfectly just procedure; whereas a criminal trial is of imperfect procedural justice. In the latter case, the just outcome is independently specified to be that a defendant should be declared guilty if and only if they have indeed committed the crime under investigation. There exists, however, no known procedure to guarantee this result.) Perfect and imperfect procedural justice are thus geared towards the outcome, by which the fairness of the procedure can be judged.  Rawls makes it clear that his conception of justice is one of pure procedure (TJ, pp.87ff.). However, it is less clear, I believe, why he should want this to be so. Of course, there are the practical difficulties with achieving perfect rather than imperfect procedural justice, but this in itself is no reason to reject the idea of perfect procedural justice, unless it can be shown that the negative consequences of merely imperfect procedures are so grave as to outweigh the benefits. One possible answer to why Rawls should attempt to provide a pure procedural account of  27  justice is that this better suits the idea of the inviolability of the person. Consider this: In utilitarian theories of justice, the person is supposed to have their own individual preferences. Therein consists their individuality. However, these can be known by others, most importantly the "impartial sympathetic observer", and used to determine the "value", and hence justice, of different distributions of utility (satisfaction of desires) across different persons. In this way persons, in spite of their individuality, are conflated into one and their individuality is violated. By contrast, in Rawls's account of justice, individuality is supposed to be secured against such a conflation. But this can only be achieved by excluding the individual preferences or conceptions of the good from the principles of justice. It is this seeming paradox that distinguishes utilitarian theories of justice from Rawls's: In utilitarianism, individual preferences are considered relevant determinants of justice, and because of this individuality ends up being violated (according to Rawls). In Rawls's account of justice, individuality is supposed to be inviolable, and therefore individual conceptions of the good must be seen as irrelevant from the standpoint of justice. Declaring that individuality consists (at least in part) in the existence of incommensurable individual conceptions of the good implies, from the standpoint of justice, that none of these is better than any of the others (subject to the constraints to be expressed by the principles of justice) and therefore that none of them may be imposed on others (TJ, pp.18/19). The only way to take the individuality of conceptions of the good seriously is to exclude them from the account of justice. But this makes it difficult to create an account of justice that is concerned primarily with outcomes. These can no longer be described with reference to personal preferences. On the  28 other hand, it is equally unacceptable to impose some "objective" conception of the good instead of the individual ones. This would constitute an equal violation of the person. Introducing a universal conception of the good would defy the purpose of excluding individual conceptions of the good in the first place. This problem can be avoided by opting for an account of pure procedural justice. Here it is not necessary to compare outcomes. These are just, no matter what they are and whose individual preferences they satisfy to what extent, if they arise out of the fair procedure. "The  great practical advantage of pure procedural justice," Rawls thus writes, "is that it is no  longer necessary in meeting the demands of justice to keep track of the endless vari  circumstances and the changing relative positions of particular persons. One avoids th  problem of defining principles to cope with the enormous complexities which would ar 13  such details were relevant." (TJ, p.87) Perfect procedural justice is thus better suited to do without particular conceptions of the good. However, just because it is better suited to the task does not necessarily mean it fulfills it adequately. In particular, it is not the case that any account of pure procedural justice necessarily secures the inviolability of the person. (Although Rawls is not entirely explicit about this.) In order to do so, the inviolability of the person must be included in the conception of "fairness" appropriate to the procedure. Pure procedural justice will only produce just outcomes if the procedure itself is just. If, for example, Rawls wants to maintain that there is an "inviolability of the person founded on justice" (see above), then this inviolability should Note that this quote also seems to imply that the problem with making justice refer primarily to outcomes is mainly the practical one of "keeping track" rather than that doing so implies a violation of individuality. This admittedly confuses me. Maybe it can be understood as yet another stab at utilitarianism. l3  29 be constitutive of the just procedure. This raises the question how we determine whether a procedure is just or not, given that we assume no prior conception of justice. It is here that Rawls's notion of a contract comes in. The fair procedure is to be one of free association between equals in whose rational interest it is to agree on the terms of their cooperation. Rawls thus envisages an "original position" that encapsulates these elements and gives rise to a social contract similar to other contractarian writers. The idea behind the contract is that it can serve as a basis of justification, since it is 14  founded on everybody's rational agreement.  "The merit of the contract terminology is that it conveys the idea that principles ofju  may be conceived as principles that would be chosen by rational persons, and that in t way conceptions of justice may be explained and justified." (TJ, p. 16) Imagining the just procedure as a contract between rational persons thus secures the inviolability of the person. Rational persons would surely not enter into a contract that violated their very essence. Nor would they enter into a contract that allowed them to be conflated into one "impartial observer". In this way, the inviolability of the person gets incorporated into the just procedure.  4.2 The Hypothetical Contract It must of course be observed that Rawls's contract is purely hypothetical. At no point does he wish to claim that any real persons actually find themselves in a position where this contract is agreed to. This, however, creates another difficulty, which, in turn, brings out 15  Examples include such prominent writers as Hobbes, Locke, and Rousseau.  14  See for instance TJ, pp. 120/121  l5  30 some important questions about Rawls's use of the notion of pure procedural justice. In the discussion of pure procedural justice, we have seen that what makes an outcome just is that it arose out of a just procedure. Justice resides in the procedure, not the outcome itself. But this implies that an outcome cannot be just unless it was actually produced by the just procedure. For example, just because a distribution of gains and losses could have come about as the result of fair gambling does not make it just. Rawls acknowledges this: "A  distinctive feature of pure procedural justice is that the procedure for determining the  result must actually be carried out; for in these cases there is no independent crite  reference to which a definite outcome can be known to be just. [...] A fair proce translates its fairness to the outcome only when it is actually carried out." (TJ, p.86) How, then, can a hypothetical contract, which is never actually carried out, constitute pure procedural justice? Ronald Dworkin, for example, writes: "[Rawls's] contract is  hypothetical, and hypothetical contracts do not supply an independent argument fo  fairness of enforcing their terms. A hypothetical contract is not simply a pale form of  actual contract; it is no contract at all. [...] There must be reasons, of course, why I  have agreed if asked in advance, and these may also be reasons why it is fair to enfor  rules against me even if I have not agreed. But my hypothetical agreement does not c  a reason, independent of these other reasons, for enforcing the rules against me, as m agreement would have." (Dworkin, p. 151) In other words, the notion of a hypothetical contract is superfluous. If Rawls asks us to imagine an original position where the parties agree on the terms of their cooperation, based on fair terms between them, and he then goes on to argue what principles these parties would agree to, he must be arguing with reference to independent reasons. Only if such reasons  31 exist, can it be possible for Rawls to say anything about how the parties in the original position would decide. There is another problem related to this. Suppose there exist independent reasons for why the parties in the original position should decide in a particular way. If Dworkin is right, their hypothetical agreement in the original position does not add anything to the strength of their obligations arising out of that decision. Hence all the argument is saying is that in a particular situation (rational) parties would decide in a certain way. However, "the fact [...] that a  particular choice is in my interest at a particular time, under conditions of great uncerta  is not a good argument for the fairness of enforcing that choice against me later und conditions of much greater knowledge." (Dworkin, p. 153)  Similarly, Kolm argues that "the fact that the imaginary ex ante self-interest, and hence t  judgement and the choice in the original position, are not biased by a specific actual po  and self-interest, does not imply that this choice is just or fair with regard to the actua post individuals." (Kolm, p. 191) Hence there is no reason why what is a rational choice in the original position, i.e. under what Rawls claims would be the parameters of a fair procedure, should be rational or even fair under actual circumstances. Let me make this clear by considering an example (borrowed from Dworkin, p. 152): Suppose Ms. A owns a painting which she would be willing to sell for $100. One Monday, Ms. A finds out that the painting is actually much more valuable. Given this knowledge, of course, she is no longer willing to sell the painting for a mere $100. Now although there may have been perfectly good reasons for Ms. A to sell the painting for $100 prior to finding out its hidden value (she did not like it, for example), it would be strange indeed to claim that Ms.  32 A now has an obligation to sell the painting at this low a price even after that Monday. It does not matter that the selling of the painting before Monday would have taken place under completely fair conditions. (We may even assume that the buyer of the painting did not know its true value before Monday, either, thus achieving symmetry of information.) The rationality of the decision to sell at $100 is confined to the circumstances before Monday, and the hypothetical agreement (hypothetical from the point of view after Monday) does not bear upon different circumstances. If these arguments are correct, it would seem that Rawls's notion of a hypothetical contract cannot lead him very far. Since the contract is hypothetical, it cannot make an impact on the justice or injustice of outcomes that merely could possibly have come about through it. And if there are independent reasons for why a particular outcome should be chosen under the fair conditions specified by the original position, there is no reason why the same outcome should arise from different conditions. The idea of a contract does not seem to be helpful in constructing an account of pure procedural justice.  I believe, however, that the above criticisms, to the extent that they were meant as such, are somewhat mistaken. Dworkin is right in pointing out that a hypothetical contract by itself cannot give rise to obligations for actual persons. It is not a contract in the legal sense. However, it may be the case that there is another sense of contract which Rawls uses, and in which the hypothetical contract can indeed serve to produce the principles of justice. These principles will not be legally binding, but they may be morally so in that they specify what we ought to consider as just. In political philosophy, the notion of a contract has a tradition as a basis of justification for  33 particular social systems and the corresponding conceptions of justice. In this, let me call it social, sense, the contract is meant to convey two things: First of all, showing that a particular social order would arise out of a contract between free and rational persons, is meant to ground that social order in the rational interest of the individual rather than "divine right" or historical contingency. Secondly, the social contract can give rise to obligations if and only if the circumstances under which it is made can be shown to be the appropriate ones for the question at hand. Whatever the parameters of the original position in which the social contract is supposedly agreed to, it must be shown that these represent all and only the relevant aspects of "human nature" or "social cooperation" or whatever is supposed to be the rational basis for answering the question. In other words, the original position must be shown to embody those circumstances which we ought to put ourselves in when asking the question "what is the just society?" Consider, for example, Hobbes's Leviathan. Here, Hobbes argues that the individual is obliged to obey a supreme ruler, since in the State of Nature, the original position before the creation of society, it is rational for every individual to install such a ruler. A large part of the argument, of course, is taken up by showing that to do so is indeed rational in the State of Nature, but it is equally essential for Hobbes to convince us that his State of Nature is the kind of state we must imagine ourselves in when asking what kind of society is justified in demanding our obedience. Hobbes's State of Nature is just as hypothetical as Rawls's original position. "It may  peradventure be thought, there was never such a time, not condition of warre as this;  believe it was never generally so, over all the world [...] Howsoever, it may be percei  what manner of life there would be, where there no common Power to feare; by the m  34 of life, which men that have formerly lived under a peacefull government, use to degene into, in a civill Warre. "  I6  The validity of this State of Nature consists entirely in its embodying all and only the relevant aspects for the question at hand. Thus in Hobbes's case, the question is why should there be a sovereign and what kind? Hobbes's argument is that when we strip away and discard as irrelevant all those features of human life that are contingent on our living in a social order, we find ourselves in the State of Nature as he specifies it. I do not wish to claim that Hobbes convincingly shows that his State of Nature is a correct representation of "human nature" prior to society as he claims. The point is precisely that it is a valid criticism to argue that his State of Nature does not correctly describe the circumstances faced by people prior to the establishment of a sovereign. If it does not, then it is entirely irrelevant to actual persons' obligations to society that a supreme ruler would rationally be chosen in the State of Nature. If, however, we did agree that Hobbes's State of Nature represents the fundamental parameters of the question "what is the right society?", (supposing that the rest of his argument was convincing, too,) then we would have to accept the obligations arising thereof. This highlights some of the differences between a legal and a social contract. A legal contract is a binding agreement made between parties of their own volition (and other constraints). It is valid and legally binding only if it was actually made. By contrast, a social contract is hypothetical. It is valid and morally binding if it would be made under conditions that specify the appropriate and relevant standpoint for the question at hand. In Rawls's case, this means that he must convincingly argue that the perspective of actual  Leviathan, ch.13, pp.89/90; ed. R.Tuck. Cambridge: Cambridge University Press, 1991  I6  35 persons, "ex post individuals" as Kolm calls them, is not the appropriate perspective when asking questions of justice; that instead all and only the relevant factors are compiled to produce the original position. "The idea [behind the original position] is simply to make vivid to ourselves the restrictions that it seems reasonable to impose on arguments principles of justice, and therefore on those principles themselves." (TJ, p. 18) The principles that arise out of the contract made in this original position are binding because they are a necessary implication of the complete set of relevant considerations. Returning to Dworkin's example of the selling of a painting, there would be a moral (but not legal) obligation to sell at $100 if and only if it could be shown that the circumstances before the Monday were more appropriate than after, in particular that the knowledge of the true value of the painting is irrelevant to the transaction. It is because presumably this information is not irrelevant, that the circumstances before Monday are not the appropriate ones to the question at hand, and the hypothetical contract is therefore not morally binding. By contrast, let us imagine Ms. A in her job as a public employee contracting out a government building project to the private sector. She is interviewing a prospective contractor, but is not satisfied with that contractor's prices. At one point, Ms. A finds out that she and the contractor have a friend in common. Since she owes that friend a favour, it now becomes in her interest to give the job to this particular contractor. In this case, it may well be that we would like to say that this information is irrelevant to the question at hand, and the right solution would be for the Ms. A to act upon her initial considerations, i.e. reject this contractor. Admittedly, we would not say that Ms. A is legally bound by the decision she only would have made, but we may well wish to maintain that she ought to (is morally bound to) act according to her original reasoning.  36 Even if this may not be a good example of a moral decision, the distinction should be clear. In the case of a social contract (moral) obligation may arise out of a hypothetical decision if the circumstances under which this decision would be made can be shown to be the only appropriate ones for the question at hand. The contract itself would represent a method by which to bring together these considerations. All this, I believe, is implicit in the notion of pure procedural justice. Remember that there an outcome is just if it arose out of a just procedure. But "just" procedure does not simply mean that it be fair. It must also mean that it be appropriate in the above sense. Consider Rawls's referral to gambling as an example of pure procedural justice (TJ, p.86). Fair betting  procedures are specified as "those having a zero expectation of gain, that the bets are ma voluntarily, that no one cheats, and so on." (ibid.) What Rawls seems to forget, creating confusion, is that these conditions must also be appropriate. For the outcome of a gamble to be just (not simply fair), one of the conditions of the gamble must be that gambling is at least one legitimate way of distributing the particular respective gains and losses. We would probably not consider the rolling of dice the just procedure for all other distributions, for example of the tax burden. For his hypothetical contract to be binding, Rawls must convince us that the original position is the appropriate situation for the determination of the principles of justice.  "If, however, the initial situation is to delineate the neutral framework in which the  are supposed to analyze and accept as binding the conditions of their prospective  well-ordered society, it would have to be the case that the conditions of the framewo 17  themselves are justifiable as the necessary point of view to be adopted." (Brehmer, p.9) "Wenn die Ausgangssituation jedoch den neutralen Rahmen absteckt, in dem Parteien die Bedingungen einer kuenftigen wohlgeordneten Gesellschaft analysieren und sie als fuer ihr Handeln verbindlich l7  37 Although Rawls does not articulate it this clearly in his own discussion of the role of pure procedural justice, I am convinced he is aware of it. In fact, he states twice, once on page 21 and then again on the very last page of his Theory of Justice, that "the hypothetical nature of the original position invites the question: why should we take any interest in it, moral  otherwise? Recall the answer: the conditions embodied in the description of this situat  ones that we do in fact accept. Or if we do not, then we can be persuaded to do so by  philosophical considerations of the sort occasionally introduced. [...] Thus what we a  doing is to combine into one conception the totality of conditions that we are ready up  reflection to recognize as reasonable in our conduct with regard to one another (§4). " (T p.587) Rawls does believe that his original position and the contract made therein function as pure procedural justice because we accept the respective conditions as appropriate to the question of justice. This also highlights, once again, the role of the notion of reflective equilibrium. The appropriateness of the original position is determined by whether, upon due consideration, it accords with our shared intuition or not. If it does, if we agree that the original position represents the correct standpoint for the determination of the principles of justice, then we will have to accept our moral obligation to follow them under the actual conditions we might find ourselves in. What Rawls therefore has to show is that his original position correctly represents the relevant conditions and bring them into a reflective equilibrium with our considered intuitions.  One of the conditions of the original position must, as we have seen, be the inviolability of anerkennen sollen, so muessten gerade die Bedingungen dieses Rahmens als notwendig einzunehmender Standpunkt allgemein gerechtfertigt werden koennen." (my translation)  38 the person. Unless the original position expresses this concern for the individual there will be no reflective equilibrium, and the contract will not be binding. In the next chapter, I shall turn to Rawls justification for the other parameters of the original position.  4.3 Summary In pure procedural justice, an outcome is just if and only if it arose out of a just procedure. No prior notion of what the just outcome should consist in is needed. This conception of justice is a useful one for Rawls, for he attempts to arrive at the principles of justice without reference to particular conceptions of the good, of which a prior conception of the just outcome would be part. Rawls therefore attempts to find a just procedure, from which his principles of justice would arise. What he comes up with is a social contract made in a hypothetical original position. This contract would be based on the rational interest of the parties in that position and would therefore be justifiable to each of them. I have argued that such a contract can indeed produce a moral obligation to uphold its terms, but only if it could be shown that the circumstances under which it is made have themselves moral force. In other words, Rawls must show that these circumstances are the only correct ones for asking what the principles of justice should be. There must be a moral obligation for us to place ourselves in that position, otherwise the contract would fail to be binding. This moral obligation arises, in Rawls's framework, from our shared intuitions, which thereby acquire their importance in his theory. The original position is relevant and can perform its function only if our shared intuitions about justice tell us that it indeed embodies all and only the relevant circumstances. Rawls's account is an instance of pure procedural  39 justice only if the procedure, that is, the contract made in the original position, is indeed just itself, and this can only be established by reference to our prior shared intuitive sense of justice.  40 Chapter 5: The Standpoint of Justice: The Parameters of the Original Position As I have interpreted Rawls so far, he must, in order for his original position to be a valid basis for the derivation of the principles of justice, convincingly show that this position entails all and only the relevant aspects of matters of justice. As we have already seen, the inviolability of the person is one of these. By including this inviolability, Rawls's theory acquires a decidedly individualist character. One of the implications of maintaining the inviolability of the person is that the individual must be regarded as more than just a member of a society, completely defined by his or her social environment. But there is also a strong egalitarian tendency in Rawls's argument. Much of the remaining parameters of the original position are concerned with the equality of the parties. Although these are less obviously connected with his conception of liberty, they are nonetheless relevant. The principles of justice, which include a conception of freedom, supposedly arise logically out of a contract made in the original position. It is thus to be expected that the parameters of that position strongly influence the eventual conception of freedom. Looking at those parameters can therefore be helpful in understanding the meaning of that latter conception.  5.1 The Veil of Ignorance "It seems reasonable to suppose that the parties in the original position are equal. That is, all have the same rights in the procedure for choosing principles [...] Obviously the purpose of these conditions is to represent equality between human beings as moral persons, as creatures having a conception of their good and capable of a sense ofjustice." (TJ, p. 19)  41 Although the original position is specified both by those factors that are included as well as those that are excluded, Rawls puts the emphasis on those that are excluded. This, I believe, is largely due to the fact that some of the most startling characteristics of the original position concern the knowledge that Rawls argues is irrelevant from the standpoint of justice.  "The aim is to rule out those principles that it would be rational to propose for accept  [...] only if one knew certain things that are irrelevant from the standpoint ofjustice. [.  represent the desired restrictions one imagines a situation in which everyone is depriv  this sort of information. One excludes the knowledge of those contingencies which se  at odds and allows them to be guided by their prejudices. In this manner the veil of ign is arrived at in a natural way." (TJ, p. 19) Equality in the original position consists primarily in the absence of knowledge that differentiates persons. As we have already seen, the knowledge of particular conceptions of the good has to be excluded from considerations of justice if the separateness of the person is to be taken seriously. Let me repeat the reasons: If we assume that every individual has a particular conception of the good, and that these particular conceptions cannot be conflated into one universal conception of the good, as in utilitarianism, (i.e. that conceptions of the good are incommensurable), then society and the publicly shared conception of justice must not be founded on any one particular conception of the good. The freedom to pursue one's conception of the good is made possible if particular conceptions of the good are excluded by the veil of ignorance. The concepts of freedom and equality are thus strongly related in Rawls's theory. The parties in the original position are equal, in the first instance, by their equal lack of knowledge of their particular conceptions of the good. Similarly, the veil of ignorance excludes other  42  considerations which would "set men at odds", i.e. make them particular individuals. In this way, the parties in the original position become not only equal, but essentially identical. This conflation of the parties in the original position into one, however, functions on the basis of the knowledge that as individuals they do have differentiating characteristics, such as conceptions of the good. In other words, knowledge of those characteristics is excluded, but not supplanted by "universal" characteristics. In this way, at least in principle, none of these characteristics are imposed on the actual individual. The point is, that the characteristics to be excluded would not only differentiate people, but introduce particularities which should have no relevance to the question of justice. In addition to conceptions of the good, Rawls therefore goes on to exclude "one's place in society, one's class position or status" (TJ, p. 137). This seems intuitively correct. But if social status should be excluded, Rawls argues, so should one's "fortune in the distribution of natural assets and  abilities, [one's] intelligence and strength and the like." (ibid.) These are outcomes of a natural lottery, i.e. arbitrary from the standpoint of justice. The other two particularities explicitly excluded are "the special features of [one's]  psychology such as [one's] aversion to risk or liability to optimism or pessimism. [...An  parties do not know the particular circumstances of their own society. That is, they  know its economic or political situation, or the level of civilization and culture it has able to achieve." (ibid.) One of the underlying reasons for these exclusions is to be found in Rawls's conception of  society as a cooperative venture: "The intuitive idea is that since everyone's well-being  depends upon a scheme of cooperation without which no one could have a satisfactor  the division of advantages should be such as to draw forth the willing cooperation  43 everyone taking part in it, including those less well situated. " (TJ, p. 15) In other words, society, being a cooperative venture, produces a surplus of advantages to be distributed among its constituents. However, no one has any special desert to those advantages (For example because, "the marginal product of labor depends on supply and demand," (TJ, p.308) and is therefore contingent; see also in particular TJ, pp. 103/4). Their distribution should be such that it can be justified to everyone taking part. This, again, brings out the close relationship between Rawls's conceptions of freedom and equality. From the standpoint of justice, everybody is seen as an equal contributor in the scheme of social cooperation and their free and willing participation must be guaranteed. There is, however, another reason why Rawls should want to exclude particular individual knowledge from the original position. Recall that the contract made in the original position is supposed to be an instance of pure procedural justice. Since the original position is hypothetical, and there are no independent standards of what is just, it must be possible to infer the principles of justice from the parameters of the original position alone. Such an inference is only possible if the original position specifies a unique solution. As we have seen, it is not a valid claim that an outcome could have come about by a just procedure. It must be shown that the procedure is just, including appropriate, and that the outcome would necessarily follow from that procedure for it to be binding. But this means that in the original position there can be no bargaining. The principles of justice produced in the original position must not be contingent on who is in it. The solution is to make everybody the same. Of course it must also be shown that this is intuitively acceptable (at least upon due consideration), but if it were not, it would no longer be possible to use the hypothetical contract as a theoretical basis for pure procedural justice.  44 Let us reconsider the example of gambling. Suppose we are faced with a fair and appropriate bet into which all the players willingly enter. We thus have a case of pure procedural justice. But since the procedure in this particular case is one of random determination, it is never possible to deduce the outcome. Any particular outcome is still contingent on the random result of the dice. Just because this is accepted as just does not mean it is also deducible in advance. Thus a hypothetical gamble could not, even if it were just, produce results that could be known in advance. The only reason why it is possible for Rawls to argue that his principles of justice are just because they arise out of a just procedure is if it is possible to uniquely determine what results the just procedure will produce. The principles of justice are binding only if they are shown to be the necessary consequence of accepting as appropriate and just the parameters of the original position. If the parties in the original position were allowed to retain differentiating characteristics that had an effect on the outcome, then this latter would depend, for example, on the relative number of parties with each of those characteristics. Rawls would be able to argue only that his principles would be chosen given specific relative numbers. But unless there were reasons why these specific relative numbers should themselves be just, the principles would cease to be binding. Supposing however, that Rawls convincingly shows that differentiating characteristics should be hidden by the veil of ignorance, it is possible for him to make the parties in the original position identical and thence to consider the deliberations of the parties as if they were one.  "The veil of ignorance makes possible a unanimous choice of a particular conceptio  45 justice. Without these limitations on knowledge the bargaining problem of the orig position would be hopelessly complicated." (TJ, p. 140) The only knowledge the parties in he original position have, therefore, is general knowledge that is the same for all of them. It follows that the only factors relevant for questions of justice must be universal ones not pertaining to particular persons or contingent upon social circumstances.  18  5.2 The Parties in the Original Position It should be noted that the circumstances of the original position are not initial assumptions beyond questioning. Rawls's argument is not axiomatic. On the contrary, "in arriving at the  favored interpretation of the initial situation there is no point at which an appeal is m  self-evidence in the traditional sense either of general conceptions or particular convic  I do not claim for the principles of justice that they are necessary truths or derivable such truths. A conception of justice cannot be deduced from self-evident premises  conditions on principles; instead, its justification is a matter of the mutual support of considerations, of everything fitting together into one coherent view." (TJ, p.21) And this "everything" includes both our intuitions and the theoretical argument Rawls puts forward. He must not only show that his proposed conclusions follow if we accept his initial assumptions, but also that his initial assumptions (the parameters of the original position) must precede if we accept his conclusions (the principles of justice). A better way of saying this is that the original position itself is not the starting point of Rawls's argument, but part of the conclusion. As Dworkin puts it, "the original position, far from being the foundation of  Note that this entails Rawls's adoption of the Kantian Categorical Imperative, by which moral principles must be universal. 18  46  his [Rawls's] argument, or an expository device for the technique of equilibrium, is on the major substantive products of the theory as a whole." (Dworkin, p. 158) But if the original position is a product of Rawls's theory, then it makes sense to ask "a product of what exactly?" The answer to this question will make it clear what the parties in the original position represent and dispel, I believe, a substantial confusion about Rawls's theory. If my interpretation of Rawls so far is valid, then the answer should be pretty clear: The parties in the original position represent the standpoint of justice. The original position is the product of developing the acceptable conditions under which questions of justice should be considered. Most importantly, it is not the outcome of an attempt to develop an account of the "true essence of human nature". This means that the parties in the original position are not a vision of actual persons "as they really are, fundamentally", but of the correct point of view for a particular question. They represent not persons, but a standpoint. It follows that the frequent criticism, that Rawls's view of human nature is untenable, is invalid, in so far as it refers to the parties in the original position. (Note that Rawls almost consistently refers to them as "parties" rather than "persons".) Thus consider, for example, Sandel's assertion that "[Rawls claims that] no commitment  could grip me so deeply that I could not understand myself without it. No transformati  life purposes and plans could be so unsettling as to disrupt the contours of my identi  (Sandel, p.62) Or, in the same vein: "The independence of the subject [means] that my value  and ends do not define my identity, that I must regard myself as the bearer of a self d from my values and ends whatever they may be." (Sandel, p. 12) What Sandel is criticizing is Rawls's attempt to exclude the knowledge of conceptions of  47 the good, i.e. of values and ends, from the original position. Sandel seems to believe, correctly or not, that a person without their values is no longer a person. But, as I have argued, that is besides the point for Rawls, since the parties in the original position are not persons, nor do they need to be. Another example of this misunderstanding is provided by Benjamin Barber who claims that  "men in the original position are defined by rough equality and freedom [...]" (Barber p. 194, my emphasis)  And even Dworkin writes that "[the parties in the original position] are men and women  with ordinary tastes, talents, ambitions, and convictions, but each is temporarily igno  these features of his own personality, and must agree upon a contract before his self-awareness returns." (Dworkin, p. 150) The confusion is partly understandable when one remembers that a large part of the contractarian tradition has been concerned with arriving at a good account of "human nature". (Hobbes is one example, so are Rousseau and Locke.) This, however, should not distract us from the fact that Rawls does not fit in this mould. His question, to which the parties in the original position are the answer, is substantially different. What he is asking, again, is what point of view one should take when asking questions about justice. His answer entails that some of what makes us individuals and persons is irrelevant from the standpoint of justice. What he is trying to do is not, however, to transform us into non-persons when talking about justice, but rather to find the right basis of justification for that issue. The original position provides the circumstances in which arguments about justice between persons with, for example, different conceptions of the good can be made. "The idea here is  48 simply to make vivid to ourselves the restrictions that it seems reasonable to impose on arguments for principles of justice." (TJ, p. 18)  Because individuals have different conceptions of the good, different values and goals, and our sense of justice demands that these not be violated, it is necessary to find a common basis of justification by which these individuals can justify conceptions of justice to each other. Since it is a requirement of justice that it be a publicly shared value, we must be able to justify the eventually developed principles of justice to persons with a variety of conceptions of the good and regardless of their particular talents and position in society. Rawls's argument is that the correct point of view from which such justifications can be made is one where no reference is made to such particularities, the original position. What Rawls is asking us to do, if my interpretation is correct, is not to imagine ourselves without our particular conceptions of the good; he is not suggesting that we regard ourselves as beings detached from particular values. What he is arguing is that in finding the principles of justice we must justify these on grounds that are acceptable to persons with different values. His argument is not about the essence of the person, but strictly about what kinds of arguments are acceptable as grounds for a conception of justice. This, I believe, not only takes the force out of Sandel's charge, but also addresses the feminist critique that Rawls simply represents the position of a "white, middle-class, liberal American male." (Frazer and  Lacey, p.55) In my interpretation, Rawls attempts to find the common basis by which we can justify conceptions of justice to one another regardless of gender, class, race or age. That is not to say that these are not important determinants of our identity. On the contrary, because they are we must find principles of justice that are acceptable to individuals with any and all of the possible combinations of these determinants. Rawls's proposed method by which to  49  achieve this is to exclude such factors from the original position. Admittedly, it may be argued that the "thin theory of the good" Rawls then introduces is gender- and class-biased, but this is not, strictly speaking, part of the original position as such.  Note that this interpretation seems to be contradicted by Rawls's self-professed affiliation to Kant. Linking his theory to Kant's, Rawls writes that "when persons act on these [his own] principles [of justice] they are acting in accordance with principles that they would choose as rational and independent persons in an original position of equality. The principles of their actions do not depend upon social or natural contingencies, nor do they reflect the bias of the particulars of their plan of life or the aspirations that motivate them. By acting from these principles persons express their nature as free and equal rational beings subject to the general conditions of human life." (TJ,  pp.252/3)  This seems to be saying that Rawls's principles derive from a conception of human nature as free and equal human beings, similar to Kant's noumenal selves. The contradiction is only apparent, however. It does not follow from this comparison with Kant that Rawls's principles are arrived at in the same way, only that they can be understood to fulfill the requirements set out by Kant's Categorical Imperative. In my reading of Rawls, the standpoint of the original position is one in which the parties justify to each other the principles of justice, treating each other as free and equal rational beings, i.e. the kinds of arguments allowed in the original position must not violate Kant's Categorical Imperative. The original position specifies how we should treat each other when considering conceptions of justice rather than how we should treat ourselves. And "treat" here means primarily what kinds of justifications we can reasonably expect to be acceptable. To be aware that others  50 may not share our particular conception of the good is in this sense an expression of what it means for Rawls to be a "rational person with a sense of justice" (see TJ, p. 19).  5.3 Summary The parameters of the original position consist primarily in the absence of particularities. The parties in the original position are stripped, through the veil of ignorance, of the knowledge of their particular conceptions of the good, their talents and status in society, and all other purely contingent factors that make them particular individuals. In this way, the original position comes to represent the standpoint of justice. Rawls argues that, in accordance with the requirements of pure procedural justice, we must find a common basis for justifying the principles of justice to each other without reference to our particularities. When looking for the principles of justice, we therefore have a moral obligation to imagine ourselves in the original position and argue with reference only to the information not excluded by the veil of ignorance. Not to do so would be equivalent to violating each others particularity and assuming that there can be one conception of the good, one position in society, or one set of talents that is more appropriate and thus valuable than others. The veil of ignorance, furthermore, is set up bearing in mind that as real persons we do have conceptions of the good, status, etcetera. That is, Rawls does not suppose that these particularities are irrelevant to the conception of a person. He simply asks us to exclude these from justifications for the principles of justice. Rawls does therefore not need to make any particular claims about individual autonomy. He does not need to claim that the individual is independent from society nor the opposite. All he needs to convince us of is that the imposition of the veil of ignorance is in accordance with our considered judgements about  51 justice, and that it upholds the inviolability of the person. If we do indeed agree that it is just to protect the pluralism of conceptions of the good, as Rawls believes we do, we should also accept the obligation to place ourselves under the veil of ignorance when justifying to each other the principles of justice.  52 Chapter 6: Rational Choice in the Original Position: The "Thin Theory of the Good"  Having argued that Rawls's original position is supposed to represent the correct standpoint of justice, it now makes sense to return to the notion of pure procedural justice and ask to what extent and in what sense principles of justice arrived at through the original position represent an instance of pure procedural justice? In pure procedural justice, an outcome is just if and only if it arose out of a just procedure. If we agree with Rawls that his original position embodies all and only those aspects by reference to which justifications for particular principles of justice ought to be made, then we must also agree with Rawls that this original position is an example of a just procedure. It is appropriate to imagine ourselves in that position when asking questions of justice. Therefore, we have a moral obligation to follow whatever principles of justice may arise from the original position. They are just because they are those principles which can be defended exclusively with reference to what we agree is all and the only relevant information from the standpoint of justice. If (and only if) we accept Rawls's description of the original position as the appropriate and just standpoint, the principles of justice that arise out of the original position will be instances of pure procedural justice. This imposes a logically distinguishable two-part structure on Rawls's argument. First, he must show that the original position represents the correct standpoint of justice. Up to now I have mainly analyzed what this means. Second, however, Rawls must also show that his particular principles of justice do indeed follow from the original position. This requires a separate argument. Remember that in pure procedural justice the outcome is just only if the procedure is actually carried out. But, as we have seen, the original position is purely  53 hypothetical and never actual. However, if we understand the original position as the standpoint of justice it becomes clear what the procedure to be carried out must consist of. Rawls does not claim that the contract made in the original position ever actually takes place. What he does ask us to do, though, is to carry out the thought experiment of imagining ourselves in the original position, to accept the limitations on justifications for the principles of justice, and to see which such principles can be justified only with reference to the relevant information. Therein consists the procedure of Rawls's account of justice. And this procedure must actually be carried out. Only if Rawls can convincingly show that his principles of justice follow necessarily when we imagine ourselves in the original position does his account satisfy the requirements of pure procedural justice. In other words, Rawls must argue that his principles of justice not only could be chosen in the original position, but more strongly that, in fact, they would. He must convince us that we, the readers of his argument, the ones who carry out the thought experiment of imagining ourselves in the original position, would choose his principles and not others. Ideally, then, Rawls must show that his principles would be chosen over all possible alternatives. It should be noted that Rawls's argument falls well short of this. He brings out a number of difficulties with the representation of "all the possible alternatives" (TJ, p. 122) and thence limits himself to showing that his principles would be chosen over a specific short list of alternatives, notably two forms of utilitarianism, perfectionism and mixed theories:  "Admittedly this is an unsatisfactory way to proceed. It would be better if we could def  necessary and sufficient conditions for a uniquely best conception ofjustice and then ex  a conception that fulfilled these conditions. Eventually one may be able to do this. For  time being, however, I do not see how to avoid rough and ready methods. Moreover,  54 such procedures may point to a general solution of our problem. [...] For the present,  attempt is made to deal with the general problem of the best solution. I limit the arg  throughout to the weaker contention that the two principles would be chosen from th conceptions ofjustice on the following list." (TJ, p. 123) Thus Rawls accepts the possibility that there may be principles of justice that would be chosen over his. These, however, have yet to be put forward. With respect to pure procedural justice the point is not so much that his principles represent the uniquely best solution to the problem, but rather that they would necessarily be chosen even if only under the restrictive assumption that a limited list of alternatives is available. There is, I believe, no great theoretical problem with Rawls's accepting that his principles may one day be superseded. If only he can show that they offer a better account of justice than utilitarianism and related theories, he has already achieved a great deal. Nonetheless, the point remains that in order to fulfill the requirements of pure procedural justice, Rawls must show that his principles follow in some sense necessarily. Only if this sense of necessity is evident can Rawls claim that his principles would indeed arise out of following the just procedure. Rawls's main tool for showing the necessity of his two principles is the notion of rational choice.  6.1 Rational Choice I have pointed out, in chapter 5, that by the veil of ignorance all differentiating characteristics are excluded from the original position. This implies that the parties in the original position are essentially identical. It is a necessary step in ensuring that the outcome of the hypothetical contract can be deduced without knowledge of who in particular carries out  55 the thought experiment. As I have argued in chapter 5, the principles of justice must not arise out of a process of bargaining. Since, however, we essentially need to consider only one party in the original position it is possible to frame that situation as one of the rational choice of one entity and deduce the outcome. There are several reasons for making the problem of the original position one of rational choice. One of them is that if Rawls can successfully describe his two principles of justice as an outcome of rational choice he may, as we shall see, thus be able to provide a basis for the claim that the two principles arise in some sense necessarily.  The theory of rational choice is complex and has attracted a lot of inquiry in its own right. For my present purpose, however, it should be enough to concentrate on some of the implications of the  notion of rationality included in rational choice. "The concept of  rationality invoked here, with the exception of one essential feature, is the standard familiar in social theory. Thus in the usual way, a rational person is thought to have  coherent set of preferences between the options open to him. He ranks these optio  according to how well they further his purpose; he follows the plan which will satisfy m his desires rather than less." (TJ, p. 143) The conception of rationality Rawls is using is essentially that of utility maximization. This may appear striking since Rawls attempts to provide an alternative to utilitarian accounts of justice. We should, however, bear in mind that what Rawls objects to is not so much the principle of utility as such, but the way it is used in constructing a theory of justice. In particular, as we have seen, Rawls's focus is on the aggregation across individuals that utilitarians need to make in order to extend the principle of utility for one individual to one of  56 "social utility". In the case of rational choice, however, no such aggregation need be made. Rational choice as Rawls uses the term consists strictly in the utility maximization of and by one subject.  Making the decision process of the original position one of rational choice has important advantages. First of all, rational choice is likely to lead to a unique solution. If we assume that the parties in the original position have certain preferences then it is very possible (although not necessary) that there will be one set of principles of justice that best satisfies those preferences. If such a set can be shown to exist, this will be the one unambiguously chosen in the original position. Second, and more importantly, if the principles of justice are such that they would be chosen by a rational individual under the constraints of the original position, Rawls can successfully identify the sense in which they arise necessarily out of the original position. Given the preferences of the parties in the original position, any rational individual who imagines themselves in that position will agree that those principles of justice would be chosen and therefore indeed represent the result of carrying out the just procedure. At the very least, if we should disagree with him, there is a basis for Rawls to convince us that his principles are indeed the correct ones. If they are rational given the preferences of the parties in the original position, then they must be objectively so, and Rawls should be able to show us. In other words, if Rawls can successfully argue that his principles are the rational choice of the parties in the original position, he has also shown that they are the necessary result of correctly carrying out the just procedure and therefore represent an instance of pure procedural justice.  57 There is, however, a severe problem with applying rational choice to the original position. In order to deduce what would be rational for the parties to choose, it is necessary to objectively know what their preferences are. Unless these are unambiguously known, Rawls cannot claim that his principles are the only choice the parties in the original position can rationally make. And if he cannot, then his argument ceases to fulfill the requirements of pure procedural justice. In fact, an essential characteristic of the original position is, of course, that the parties there do not know what their particular conception of the good consists in. In other words, they are from the outset barred from referring to their own preferences. On the other hand, if the parties' preferences are not known, we cannot say what it is they are trying to maximize, nor which principles they would choose. Rawls therefore needs some kind of assumptions about what the parties' preferences are, while at the same time maintaining that they do not know their particular conception of the 19  good (which they nonetheless must ultimately be free to pursue).  It is in order to overcome  this seeming paradox that Rawls falls back on what he calls the "thin theory of the good". 6.2 The "Thin Theory of the Good": Rational Plans of Life  "In contrast with teleological theories, something is good only if itfitsinto ways o  consistent with the principle of right already at hand. But to establish these principles  necessary to rely on some notion of goodness, for we need assumptions about the par  motives in the original position. Since these assumptions must not jeopardize the prior  of the concept of right, the theory of good used in arguing for the principles of justice This last condition is crucial in avoiding to impose a "universal" theory of the good upon them (see chapter 4). 19  58  restricted to the bare essentials. This account of the good I call the thin theory." (TJ, p. Rawls is clearly aware of the problem with introducing assumptions about the parties' preferences (theory of the good) into the original position. His solution is to provide a theory of the good that is at the same time limited enough not to interfere with individual conceptions of the good, and at the same time substantial enough to allow for rational choice. What Rawls attempts is to find something that the parties can maximize regardless of what their individual preferences are. Since conceptions of the good are assumed to be different for different individuals it would be unacceptable for Rawls to generalize any such conception. Instead what he attempts is to single out common characteristics of all conceptions of the good, whatever they may be. Specifically, as Rawls puts it: "In entering into the original  agreement [...] the parties suppose that their conceptions of the good have a certa structure." (TJ, p.397) The thin theory of the good is supposed to bring out what this structure, common to all conceptions of the good, consists of. To this end, Rawls introduces the notion of "rational plans of life". The idea behind rational plans of life is that although we are barred from making assumptions about the particular content of individual conceptions of the good, it is possible to at least put some restrictions on these conceptions. Specifically, it is possible to rule out those conceptions of the good that are not in themselves coherent when considered over the long term. The assumption is that rather than just following their arbitrary and ever-changing desires, persons can arrange their desires in such a way that they form part of a coherent, and thus rational, plan, and that it is just from the point of view of the original position that we expect them to do so. Even though a person may sometimes experience conflicting desires, it  59 is possible for them to solve such conflicts and determine which desires are most relevant in their overall, long-term plan. Rawls backs this claim up by referring to what he calls the "Aristotelian principle", which postulates that human values have a tendency to increase as the level of skill and complexity involved in their pursuit increases, thus promoting specialization over the long term rather than the arbitrary following of desires as they arise:  "The Aristotelian Principle runs as follows: other things equal, human beings enjoy  exercise of their realized capacities (their innate or trained abilities), and this enjoym 20  increases the more the capacity is realized, or the greater its complexity." (TJ, p.426) Note that this principles says something about the structure of conceptions of the good, rather than their content. The idea is that whatever activities a person may perceive as worthwhile, they will tend to prefer being more skilled at them rather than less, and they will tend towards activities that require more skill rather than less in their assessment of worth, ff this principle is accepted then it makes sense to assume that persons will tend to regard the pursuit of values as stretched out in time and therefore engage in long-term planning, since skill usually comes with experience. Thus people's sets of values, whatever they may be, will tend to have a common structure in that they will be arranged systematically to form part of a long-term plan.  21  "The main idea is that a person's good is determined by what is for him the most rationa  long-term plan of life given reasonably favorable circumstances. A man is happy when h more or less successfully in the way of carrying out his plan. To put it briefly, the good is the Rawls is alluding to Aristotle's Nicomachean Ethics, bk.VII, chs.l 1-14, and bk.X, chs.1-5; see his footnote 20 (ibid.) 20  It should be stressed that this is a tendency and need not always hold, as long as everyone can agree that it is generally the case. 21  60 satisfaction of rational desire. [...A person's plan] is designed to permit the harmoniou satisfaction of his interests." (TJ, p.93) This is not to say that a person's desires can never conflict, or that plans of life are fixed from the beginning of a life-time to its end. It does imply, however, that the person him/herself, whatever their particular desires (conception of the good), can come up, at any one point in time, with a system by which their desires can be ranked. In brief, Rawls assumes that persons are rational, and that their rationality consists in their being long-term individual utility maximizers. This is important because it implies that individuals not only maintain values, about which nothing much can be said, but that they have long-term ends the realization of which they are trying to maximize. This "thin theory of the good", Rawls claims: "is a familiar one going back to Aristotle,  and something like it is accepted by philosophers so different in other respects as Kant  Sidgwick. It is not in dispute between the contract doctrine and utilitarianism." (TJ, p.92 It is a "thin" theory because it is , Rawls claims, generally agreed upon, and does not infringe upon the individuality of conceptions of the good. However, once the "thin theory of the good" is accepted, Rawls goes on to say, we can define a set of what he calls "primary goods", the share of which every individual would like to maximize, whatever their particular conception of the good. In this way, the "thin theory" makes it possible to specify the rational choice facing the parties in the original position.  6.3 Primary Goods Although in the original position it is illegitimate to make claims about the particular content of the parties' conceptions of the good, Rawls would like to establish that the parties  61 nonetheless know that whatever their conception of the good, they will have rational plans of life, and that there is a set of primary goods that are useful for the execution of any such plan. Thus, in the absence of more specific knowledge, it is in the rational interest of the parties in the original position to attempt to maximize their share of primary goods. If Rawls can establish this, then the original position can indeed be understood as a problem of rational choice under the given constraints. It is not clear, however, that the notion of primary goods arises out of a stringent logical deduction from the assumption about rational plans of life. Perhaps Rawls's reasoning is best understood by looking at what the class of "primary goods" contains. In this way it may be easier to see why Rawls believes he can convince us that maximizing one's share of primary goods is the rational course of action in the original position.  "Suppose that the basic structure of society distributes certain primary goods, tha things that every rational man is presumed to want. These goods normally have a  whatever a person's rational plan of life. For simplicity, assume that the chief primary g  at the disposition of society are rights and liberties, power and opportunities, income a wealth." (TJ, p.62) The essential point about primary goods is that they are means to unspecified ends. Nothing needs to be known about the ends in order to agree that primary goods are desirable means. Thus in maximizing one's share of primary goods, one is not restricted in one's choice of ends (conception of the good). If Rawls is correct then the parties in the original position, knowing that they have rational long-term plans of life but not what these specifically are, will agree that maximizing their share of primary goods is as close as they can get to ensuring the realization of their particular plans. It would be irrational for them not to maximize their  62 share of primary goods. As straightforward as this may seem, there is nonetheless a great deal of vagueness about Rawls's conception of primary goods which has implications for the content of the "thin theory". In particular, there are really three different conceptions of primary goods at work in Rawls's argument, each of them warranted only if specific assumptions about what it means to be rational are made: First, there is the conception that primary goods are such that everyone wants more of them rather than less. (TJ, p.92) Second, Rawls claims that primary goods "have a use" whatever one's plan of life. (TJ, p.62; see above) Third, at several points Rawls goes so far as to claim that primary goods are "necessary means' whatever one's plan of life. (TJ, p.93; also p.433) These three conceptions are not at all identical. I have ordered them according to the strength of their implicit empirical claim. In the first case, all Rawls is saying is that primary goods cannot have a negative impact on the execution of one's plan of life. Whatever that plan is, it is always better to have more wealth, for example, than less. Primary goods, according to this conception, are such that they do not restrict the realization of one's ends. At the very least, they are easy to get rid of (wealth) or not make use of (opportunities and rights). No claim need be made about the positive impact on plans of life. The second conception of primary goods, however, makes a stronger empirical claim. Here it is maintained that primary goods always have a positive impact on the execution of rational plans of life. This empirical claim about the relationship between means and ends implies a restriction on what plans of life can be considered rational. It can only be maintained if all rational plans of life are indeed such that they benefit from one's having a greater share of  63 primary goods. Plans of life that do not benefit from one's having, for instance, more liberties cannot be considered rational (although they may be entirely coherent) or this particular conception of primary goods is not warranted. If, for example, my plan of life consists mainly in developing a cure for cancer it may well be that greater liberty does not at all further this plan. Even as a slave I may be able to achieve my end. Of course, I might agree that I would prefer greater liberty rather than less (first conception of primary goods), but if Rawls wants to claim that primary goods always have a use whatever my rational plan of life (second conception) he must exclude plans such as mine from the list of "rational" plans of life. It may be possible to do so by showing that rational plans of life never consist in one single goal such as finding the cure for cancer. Be that as it may, the second conception of primary goods contains a much stronger claim than the first. The third conception of primary goods is even more restrictive. According to this conception primary goods are necessary whatever one's rational plan of life. Thus a decrease in any of the primary goods will always have strong negative effects on the execution of one's plan. In the above example, this would imply that a decrease in liberty would always make it less likely that I may find a cure for cancer. If this conception of primary goods is to be used, then we must not only exclude plans of life that do not actually benefit from a greater share of primary goods from the set of "rational" plans, but also those that do not depend entirely upon access to primary goods. Consider another example. If my end is to dig a hole in the ground, then I would most probably agree that I would prefer having a spade to not having one (first conception). I would also agree that a spade is a good means to my end since it makes digging the hole easier (second conception). I would not, however, say that a spade is necessary for digging a hole (third conception). If I do not have a spade then I use a shovel, or at the worst a spoon or  64 even my hands. Although "digging a hole" is not a good example for a plan of life, it brings out the difference in claims about the relationship between means and ends implied by the different conceptions of primary goods. None of these conceptions implies those lower down on the list (although the reverse is true). On the other hand, there is a proportional relationship between the conceptions of primary goods and the strength of the claim that the parties in the original position will act so as to maximize their share of primary goods. This claim is strongest if Rawls can establish the third conception of primary goods, and much weaker for each of the "weaker" conceptions. This serves to show that the logic of the "thin theory of the good" is not quite as tight as we might hope. Rawls is not clear on which of the conceptions of primary goods he is subscribing to and seems to jump from justifying the first conception with regards to its impact on the pluralism of conceptions of the good to defending the maximization of their share of primary goods by the parties in the original position on the basis of the third conception. This is not to say that Rawls is necessarily wrong in postulating primary-goods-maximizing behaviour in the original position but the ambiguity of the "thin theory" deals a severe blow to the strength of Rawls's argument. For we must remember that the task of this "thin theory' is to show the necessity of Rawls's two principles of justice as required by the demands of pure procedural justice.  65 6.4 The "Thin Theory of the Good" as an Account of Human Nature I have argued above that the original position is best understood not as an account of human nature but as the standpoint of justice. This does not mean, however, that Rawls can make do without claims about human nature. In fact, that is precisely what the "thin theory of the good" represents: empirical claims about human nature as rational beings. As we have seen, Rawls must introduce the "thin theory" in order to be able to deduce the outcome of the choice in the original position. He must also be careful to make the "thin theory" thin indeed so as to avoid infringing upon the pluralism of conceptions of the good, which is an essential feature of his conception of justice. Nonetheless, the "thin-ness" of the "thin theory" consists in its impact on conceptions of the good, not in its foundation. The "thin theory of the good" constitutes a set of claims about that which is common to all rational human beings in the real world (not just in the original position). But since it forms part of the carrying out of the "just procedure", i.e. imagining ourselves in the original position, it is a requirement that the "thin theory' be unambiguously accepted. Otherwise Rawls could not claim that his principles of justice arise necessarily out of following the just procedure. In this vein, Brian Barry  comments: "Either it [the Aristotelian Principle] is an empirical generalization or it is  partially constitutive definition of rationality. If it. is an empirical generalization it appe  me open to serious doubt. Moreover, Rawls requires not merely that it should be true it should be so unquestionably true that the actors in the original position will treat  axiom in their reasoning about the choice of principles. Alternatively, let me suppose th put forward as a partially constitutive definition of 'rationality'. This would mean that  whose choice ran counter to it would be said not to have a 'rational plan'. [...] We m  say that, contrary to Rawls's professed intention, a substantive idea of human excell  being advanced under cover of the neutral-appearing concept of rationality." (Barry, p.28  J  66 Although Barry is commenting primarily on Rawls's "Aristotelian Principle' (which states that people value activities that are more complex and require more skill more highly than others), I believe what he points out holds true for the whole of the "thin theory". Either we understand Rawls's definition of rationality, and his conception of primary goods, as a set of claims about empirical facts, which are not backed up by any data, or we understand it as mere definition, in which case there is an infringement on what conceptions of the good (and thus plans of life) are to be considered "rational". In both cases, however, Rawls must assume that his claims are so "self-evident" as to be universally and unambiguously accepted ("treated as axioms"). Barry relates this problem only to the empirical claim, which, I believe, is mainly due to the fact that he does not consider the "thin theory" in the light of its role within pure procedural justice. In that light, it should by now be clear that the "thin theory" can only fulfill its task if we, carrying out the procedure of imagining ourselves in the original position, accept it unequivocally. If we do not then Rawls cannot claim that his two principles of justice arise necessarily out of the just procedure. Only if we agree that his "thin theory" is a true representation of human nature can we consider the hypothetical choice made in the original position as morally binding. In other words, in the light of its role in pure procedural justice, Barry makes a distinction concerning the Aristotelian Principle where there is none. The "thin theory" (including the Aristotelian Principle) must be understood as an empirical claim about our shared definition of rationality. This is a serious problem which Rawls can do little to resolve. On the one hand, he needs, at this point, to make claims about our shared conception of human nature as rational beings; on the other hand, this conception needs to be vague in order to infringe as little as possible upon the pluralism of conceptions of the good.  67 6.5 Maximizing One's Share of Primary Goods I am not satisfied that Rawls walks the thin line between assuming too much and assuming too little well. The "thin theory" seems at best a "least of all evils" that we have to accept it we want to make rational choice in the original position possible. But maybe this is its best justification in the original position. In the absence of any knowledge of the parties' particular conceptions of the good, maximizing their share of primary goods seems to be one kind of behaviour that they can all agree upon. Unless someone can provide a more satisfactory "thin theory of the good", Rawls's will have to do. We should remember that it is assumed that people have a "sense of justice" and realize that in order to achieve a well-ordered society they must find a shared conception of justice, acceptable to all (TJ, pp.4/5). Hence we should acknowledge that rejecting the "thin theory" without an alternative is not a good option. In the absence of anything else to maximize, we may concede that maximizing their share of primary goods is rational for the parties in the original position. There is, however, another, rather different foundation for primary goods in Rawls's theory. Rawls repeatedly speaks of primary goods as "at the disposal of society" (for instance TJ, p.62 above). Remember that in Rawls's view: "justice is a virtue of social institutions, as truth is of systems of thought." (TJ, p.3) It is social institutions that he is concerned with. And  "the primary subject of justice is the basic structure of society, or more exactly, the way  which the major social institutions distribute fundamental rights and duties and determine t division of advantages from social cooperation." (TJ, p.7) The division of advantages from social cooperation is what is at stake. A crucial part of Rawls's justification of the veil of ignorance is that such things as status and one's position in society are arbitrary from the standpoint of justice. The advantages from social cooperation are no one's desert, since they are by definition produced by social cooperation rather than  68 purely individual effort. And these advantages from cooperation include rights and liberties, power and opportunities, and to a large extent power and wealth (the latter through establishing the framework in which trade takes place and wages and prices are determined; see Rawls's comment on the dependence of the marginal product of labour upon supply and demand (TJ, p.308)) Thus another way of looking at primary goods is that these are the "advantages of social cooperation" which have to be distributed justly by the social institutions. It follows directly that the parties in the original position are concerned with their share of primary goods, and the most likely rational objective is to maximize one's share of them. Understanding primary goods in this way provides additional support for the claim that the original position can be understood as an instance of rational choice. Rawls still needs the "thin theory", though, in order to strengthen his claim that the parties in the original position will necessarily want to maximize their share of primary goods (rather than, for example, satisfy some minimum level). Additionally, in this latter case, too, Rawls must be understood as making an empirical claim which must be accepted unambiguously by his readers for his argument to fulfill the requirements of pure procedural justice. We must all agree that the primary goods are indeed produced primarily by social cooperation and not by individual effort. In both cases Rawls needs to bring in assumptions that need to be shared for the argument to work. And as with all such claims to universal validity there is a grave danger of introducing normative statements  69 that may infringe upon the pluralism of values. For example, many a strong individualist would disagree with Rawls's view of social cooperation . 1  6.6 Summary Having established the veil of ignorance in the original position, Rawls is faced with a serious problem. He now needs to be able to show that his principles of justice would necessarily be chosen over those of aggregate utility maximization or perfectionism, which he hopes to achieve by making them the outcome of rational choice. But in order to introduce the notion of rationality as individual utility maximization it must be objectively known what the preferences of the parties in the original position exactly are. These preferences, on the other hand, have explicitly been excluded from the original position. Rawls needs to introduce a "thin theory of the good" in order to make rational choice in the original position possible. This 'thin theory' must at the same time be thin enough not to infringe upon the pluralism of individual conceptions of the good, and extensive enough to produce unique results. In my mind, this is the weakest part of Rawls's argument. I do not find his claim that primary goods must unambiguously be maximized by the parties in the original position particularly convincing, especially bearing in mind that this claim must not infringe upon individual conceptions of the good and be universally shared. As far as I can see, the strongest support for the assumption of primary-goods-maximizing behaviour stems from our alleged "sense of justice" by which we realize that some solution must be found, that not to deduce any principles of justice is not a good option for the parties in the original position.  'This point comes out more clearly in Rawls's later work, in particular Political Liberalism, where he shifts his emphasis from "conceptions of the good" to "comprehensive moral, religious and philosophical doctrines", thus widening the notion of "values" to include frameworks for justification and explanation.  70  Nonetheless, the introduction of primary goods is part of the uniqueness of Rawls's theory, and it is through using the notion of such goods that he hopes to achieve what Kant could not, namely to deduce particular principles that obey the Categorical Imperative. I am not convinced that Rawls does so satisfactorily. Be that as it may, the introduction of primary-goods-maximizing behaviour is absolutely necessary for Rawls to proceed with his account of justice, and I shall therefore accept the assumption for the sake of the argument.  71  Chapter 7: The Two Principles of Justice and Freedom  7.1 Rawls's Two Principles of Justice Assuming that Rawls can convince us that what the parties in the original position are concerned with is to maximize their share of primary goods, it is now possible to examine the claim that they would choose Rawls's two principles of justice as the best way to coordinate their cooperation. In particular, as I have pointed out, Rawls is concerned with showing that his principles would be chosen over that of total (or average) social utility. Again, the best way to follow Rawls's argument, I believe, is to look at the result first, and then to examine the reasons for why it should be reached. Rawls presents his principles of justice at various stages throughout the Theory of Justice, amending them as he thinks is required by the argument. Their final formulation is as follows: "First Principle  Each person is to have an equal right to the most extensive total system of equa liberties compatible with a similar system of liberty for all. Second Principle Social and economic inequalities are to be arranged so that they are both:  (a) to the greatest benefit of the least advantaged, consistent with the just savings p and (b) attached to offices and positions open to all under conditions of fair equality opportunity. First Priority Rule (The Priority of Liberty)  The principles of justice are to be ranked in lexical order and therefore liberty ca restricted only for the sake of liberty. There are two cases:  72  (a) a less extensive liberty must strengthen the total system of liberty shared by all; (b) a less than equal liberty must be acceptable to those with the lesser liberty. Second Priority Rule (The Priority of Justice over Efficiency and Welfare)  The second principle ofjustice is lexically prior to the principle of efficiency and to th  maximizing the sum of advantages; and fair opportunity is prior to the difference prin There are two cases: (a) an inequality of opportunity must enhance the opportunities of those with the opportunity;  (b) an excessive rate of saving must on balance mitigate the burden of those beari hardship. General Conception  All social primary goods - liberty and opportunity, income and wealth, and the bas  self-respect - are to be distributed equally unless an unequal distribution of any or all o goods is to the advantage of the least favored." (TJ, pp.302/303) This formulation of the two principles of justice is the final outcome of Rawls's discussion of the original position. It contains a number of technicalities, such as the "just savings principle", designed to take care of potential problems. Since I am mainly concerned with 2  the conception of liberty that arises out of the two principles, I shall examine the technicalities in depth only where they relate to liberty. The first principle, together with the first priority rule, contains Rawls's definition of liberty as chosen in the original position. I shall consider this in detail below. At this point let me The "just savings principle", for example, is designed to resolve the problem of justice between generations. Rawls contends that any generation is bound by the precepts of justice to provide the basic infrastructure for following generations and not use up all its resources to maximize the share of primary goods only of its present members. The main justification for this is that the parties in the original position do not know which generation they will belong to. (see for instance TJ, p. 140) 2  73 simply point out that Rawls's introduces a qualitative distinction between "rights and liberties" and the other primary goods. This, I believe, is largely reflective of Rawls's  fundamental assertion that "each member of society is thought to have an inviolability  founded on justice [...] which even the welfare of every one else cannot override." (TJ, quoted also in ch.2) One of the fundamental assumptions of Rawls's theory as a whole is that a conception of justice must not sacrifice the individual to the "greater good". As I shall discuss more fully below, establishing that individual liberty cannot be sacrificed for other primary goods (in particular income and wealth) is a major step in protecting said inviolability of the person. The second principle of justice seems to be the more striking one. Whereas the notion of equal liberty and inalienable rights is common, the notion that inequality must be justifiable particularly to those who end up relatively worst-off is somewhat more original. The "general conception" (see the above quote) sums up this idea: Equality is fundamentally just and needs no further justification; inequality can be preferable from the standpoint of justice, however, if it works to the advantage of the least favoured. Contrast this with the principle of utility maximization: Given the constraints of the original position, in particular the absence of knowledge of the particular individual ends, the principle of utility maximization would indicate that social institutions should be arranged so as to maximize the total (or average) amount of primary goods available. Inequalities could be justified if they raised this amount. Rawls's principles, on the contrary, in addition to excluding rights and liberties from this calculation, imply that not only must the amount of primary goods increase, but it must do so in a way that benefits the least advantaged. What,  74 then, are the reasons why the parties in the original position should choose Rawls's principles over that of social utility maximization?  1.2 The Original Position and Risk First of all, it should be noted that both potential choices can be justified only under the simplifying assumption that the parties in the original position are free from envy. If they were concerned with relative shares of primary goods it would be much more difficult to justify any kind of inequality. However, both Rawls's principles and that of social utility maximization are intended to justify inequality in certain cases. What distinguishes them is not their concern with absolute rather than relative shares. Now the situation is that the parties in the original position are supposed to want to maximize their total share of primary goods, not knowing which position in society they will occupy, or which particular skills they may have. The reasoning behind the principle of social utility maximization would be that the parties have to assume that they are equally likely to end up as anybody. Thus they would calculate their expected share of primary goods as the average of all the possible shares and then try to maximize this expected share. Rawls must 3  argue that there is a mistake in this line of reasoning. He does so by pointing out that it is not the case that the parties in the original position should assume that they are equally likely to turn out as anybody: "The veil of ignorance  excludes all but the vaguest knowledge of likelihoods. [...] Thus they [the parties in t  original position] have strong reasons for being wary of probability calculations if any ot Mathematically, the expected value of a random event is the sum of all the possible values multiplied by their probability: 3  75 course is open to them. [...] These considerations are strengthened by the fact that the  parties know very little about the gain-and-loss table. Not only are they unable to con  the likelihoods of the various possible circumstances, they cannot say much about wha possible circumstances are, much less enumerate them." (TJ, pp. 155/156) The main reason why the parties in the original position, according to Rawls, would choose his principles over that of utility maximization is that they are not making a simple calculation, multiplying and adding probabilities and outcomes they know nothing about. Another way of putting this is that the parties in the original position are assumed to be "risk-averse". They are not willing to take the risk of ending up badly off if another alternative is open to them. They will opt for Rawls's principles of justice because these ensure that even if they should end up in the worst-off position, their share of primary goods would be as large as it could be. At first sight, it may seem that the introduction of risk aversion is just as unwarranted as the reverse. In fact, Rawls explicitly excludes knowledge of one's aversion to risk from the original position (TJ, p. 137). However, the unwillingness to take risks is the main reason why the parties in the original position should choose Rawls's principles over the alternatives. Without this assumption it is hard to see why rational choice would demand that the parties opt for the maximin rule embodied in Rawls's second principle. (Maximin is a rule for choice under uncertainty that says one should attempt to maximize the minimum, that is, the worst possible outcome should be the "least bad" of all the alternatives. ) 4  E[x]= (p *x ), where p=l. Since here all the outcomes are equally likely (p =p =l/n Vij), this amounts to the sum of all the possible values divided by the number of possible outcomes, the arithmetic mean or average. (E[x]=p* x =l/n* x) Rawls refers to W.J.Baumol and R.D.Luce and H.Raiffa for a fuller discussion; see his footnote 18 on p.152. 4  76 Rawls offers his "strains of commitment" argument as the main justification for the assumption of risk aversion. However, there is another, more powerful, argument why the parties in the original position should opt for maximin which arises out of the interpretation of the original position as the standpoint of justice. The "strains of commitment" argument, in brief, states that: "when we enter an agreement  we must be able to honor it even should the worst possibilities prove to be the case." ( p. 176)  And therefore: "the two principles of justice have a definite advantage. Not only do t  parties protect their basic rights but they insure themselves against the worst eventu (ibid.) In other words, the parties in the original position ought to choose Rawls's principles because they are entering into a binding agreement which they must keep whichever position in society they may end up in. Thus they must make sure that even if the worst case should happen they will still be able to honour their agreement. Now according to Rawls, if the parties opted for any other rule than maximin, in particular maximizing the average share of primary goods (social utility), they might end up in a position where they know they could have been better off had they chosen maximin. For some reason Rawls thinks that therefore they could not keep their agreement. Part of what Rawls is trying to point out, I believe, is that the potential losses involved in risking one's freedom and other primary goods in this way can be very severe. So severe, in fact that it would be irrational to take that risk.  5  That the potential losses are hardly acceptable is indeed one of the conditions under which maximin is generally a rational rule for choice under uncertainty. See TJ, pp.152-154 and Rawls's reference to W.Fellner in footnote 19 on p.154. 5  77 But I do not think this is entirely convincing since the parties do not know what the potential losses actually are. Maximizing the average share of primary goods, too, may lead to a distribution where even the worst off can still cope. On the other hand, maximizing the minimum does not guarantee that the worst-off live above the subsistence level however defined. These are contingent upon the actual circumstances a society finds itself under. Thus the "strains of commitment" argument at best serves to justify the choice of a minimum share of primary goods for everyone, enough for them to get by. Unless Rawls wants to claim that people can keep agreements only if the situation they end up in is the best possible, which would mean that it is not much of a binding agreement in the first place, he cannot maintain that the potential losses make any choice but maximin irrational. I believe what is happening here is that Rawls is getting caught in his own argument. He is trying to justify maximin strictly in terms of rational choice for one person. However, the reason why he is speaking about rational choice for one person in the first place is that the parties in the original position are essentially identical (see ch.5). Therefore we can treat them as if there was only on party. This is a simplifying assumption. It is not the case that there really is only one party. Bearing this in mind, I believe, it is possible to arrive at a much stronger justification for the assumption of risk aversion: If we understand, as I maintain we should, the original position as the standpoint of justice it is much easier to see why the parties in the original position should be so concerned with the worst possible outcome. The original position is an argumentative tool to highlight what the acceptable grounds for justification in matters of justice are. It brings out the way in which we can justify the fundamental institutions of society to each other. Now, although we can treat all the parties in the original position as one for lack of differentiating characteristics, it is still a fact that in the actual world somebody is going to be in the worst-off position. The parties  78  are not concerned so much with that position because they might end up there, but because some of them will. If the original position is understood as the standpoint of justice from which we can justify conceptions of justice to one another, then we must be able to do so even to those who end up worst off. The principles of justice are not simply the outcome of the rational choice of each party in the original position who may or may not fear ending up in the worst off position. They are the outcome of a process by which it must be shown to all those involved that they can accept said principles. They are the result of a rational choice under the condition that with full certainty some will be worst off. The only principle of choice that the worst off can accept is maximin, not because of the risk involved but, on the contrary, because of the certainty that the outcome must be justifiable to all, including the worst-off. Of course, it is important that Rawls can show that the principles of justice would be rationally chosen by each single party in the original position. But the parties' attitude towards risk is one of the parameters of the original position which are determined prior to rational choice taking place. Like the veil of ignorance, the preoccupation with the worst-off derives from general considerations about justice, in particular our "sense of justice" which requires that we come up with a conception of justice which is acceptable to all, not our rational choice. The principle of maximizing average shares must be rejected because it is not justifiable to those who are disadvantaged by it.  79  7.3 Rawls's Conception of Freedom Supposing, then, that Rawls can convince us of the validity of maximin as the rule for choice in the original position, we should agree that his second principle for the allocation of primary goods by the social institutions should be chosen. However, as I have already pointed out, Rawls makes an important distinction between "rights and liberties" and the other primary goods in separating the former from the rest. The first principle of justice, which applies to "rights and liberties" only, together with the first priority rule set them apart: "The distinction between fundamental rights and liberties and economic and social benefits a difference between primary social goods that one should try to exploit." (TJ, p.63) Rawls believes that rights and liberties indeed fall within the class of primary social goods in that they are means to the end of successfully pursuing one's rational plan of life. They are nonetheless different from the other primary goods and should never be traded-off against them. The essential difference that marks rights and liberties, as I shall argue, is that they are not only means to an end but also ends in themselves.  7.3.1 The First Principle of Justice Rawls's first principle of justice contains two claims about how freedom should be allocated. First, everyone has an equal right to freedom, and second, this freedom, defined by a "total system of liberties", should be as extensive as possible. The second aspect is rather 6  straightforward. A greater total system of liberties is preferable to a lesser. The crucial factor is how we define what is "possible". In this respect it is the first priority rule that is relevant. This rule specifies that the possible decrease in some people's share of the other primary  I shall discuss the significance of there being a "total system" of liberties below.  5  80  goods has no force against someone's claim to a greater liberty. The only limit on what are considered "possible" increases in liberty are infringements upon other people's liberty. The'total system of liberty should be "as extensive as possible" without reducing others' total systems. More significant is the declaration that everyone has an equal right to freedom. This means that anybody's claim to a greater total system of liberties is inherently as valid as anybody else's. It does not matter who makes the claim. Again, it is the first priority rule that specifies the meaning of the first principle: What matters is the extent of the system of liberties they already have. Thus, in accordance with the maximin rule, the first priority rule, part (b), states that the claim of those with the lesser total system of liberty is stronger than the claim of those with the greater system. It is the first priority rule that really defines the meaning of the first principle of justice. This rule, as I have pointed out, demarcates what are to be considered "possible" increases in liberty and how claims to greater liberty are to be adjudicated. It is the first priority rule, also, that marks the distinction between rights and liberties and the other primary goods. In justifying his conception of freedom, therefore, Rawls must pay particular attention to how he justifies the first priority rule.  "The basis for the priority of liberty is roughly as follows: as the conditions of civiliza  improve, the marginal significance for our good offurther economic and social advant  diminishes relative to the interests of liberty, which become stronger as the conditions  existence of the equal freedoms are more fully realized. Beyond some point it become  then remains irrational from the standpoint of the original position to acknowledge a l liberty for the sake of greater material means and amenities of office." (TJ, p.542)  81  One of the reasons for why the parties in the original position should decide upon the first priority rule is that once the basic physical needs are met, people develop a strong desire to be "their own masters". In other words, Rawls acknowledges that there may be circumstances under which the first priority rule does not apply, namely, when the civilization in question is so under-developed that the exercise of freedom is severely hindered by the lack of the basic means to survive (TJ, p.543). On the other hand, Rawls maintains that once these basic needs are met, liberty becomes an important determinant of self-consciousness and self-respect and should therefore be considered lexically prior to material goods in the principles of justice. This, by itself, is not a very good justification for the priority of liberty. For one, as David  Schaefer has pointed out: "The statement [of the lexical ordering of the two principles] see  to be a clear violation of the principle that the parties in the original position are n  choose in the light of any particular conception of the good. Rawls has now assumed fo that they will find 'equal liberty' to be of greater value than any other good.' (Schaefer, Furthermore, with Rawls's admittance that the first priority rule does not hold under certain circumstances he casts a doubt on his subsequent claim that under "favourable" circumstances liberty may never be traded-off against material gains. Thus Norman Bowie argues that:  "Another problem is presented once Rawls hedges the lexical ordering of the two pri  ofjustice. To salvage plausibility, Rawls must hedge [...] But now the theory has cons  indeterminacy. What is needed are some guidelines which tell us when the lexical o should be given up. The necessity for guidelines will be even more evident once it is  that tradeoffs between economic betterment and liberty are desired not only in so underdeveloped countries but even in the most affluent countries." (Bowie, p. 128) Even though Rawls might be correct in asserting that the marginal benefits of economic welfare are decreasing from a certain level onwards relative to the value of freedom, it in no  82 way follows that liberty therefore has to be ascribed the priority Rawls would like to ascribe to it. There is, however, a much stronger reason for the priority of liberty, and this reason brings out that there are two important conceptions of liberty Rawls is referring to.  7.3.2 Freedom as an End-in-itself In my interpretation of Rawls, the original position represents the standpoint of justice. This standpoint is defined by what, in our considered judgement, we consider appropriate bases for justifying the principles of justice to one another. As I have pointed out above, however, the original position is also part of pure procedural justice. In this sense it is necessary that we, the recipients of Rawls's argument, do in fact agree that the original position correctly represents the standpoint of justice. If we do not, the hypothetical contract has no binding force on us. But if this is correct, then it follows that the parties in the original position, that is, we, imagining ourselves in that position, are bound by the values that gave rise to the original position in the first place. The original position is not some arbitrary construct that Rawls postulates and then speculates what might follow from there. It is a product already, of a normative argument, including the inviolability of the person. If Rawls's claim about our shared values concerning this inviolability is incorrect, in other words if we do not feel that the inviolability of the person should be upheld against utilitarian calculations, then there is no reason to be dissatisfied with the utilitarian conception of justice in the first place, and Rawls's whole argument is futile. If, on the other hand, we do indeed share this value, then this is a relevant determinant of what principles of justice we can justify to each other in the original position. Herein lies, I believe, the main justification for the first priority rule. It is this rule which reflects our concern with the inviolability of the person.  83 As we have seen in the discussion of Rawls's critique of utilitarian accounts of justice, the distinctness of persons is violated when it is assumed that there exists one standard of value (utility) by which interpersonal comparisons can be made. The inviolability of the person consists at least in part in the existence of a pluralism of conceptions of the good. Thus each person is assumed to have their own individual conception of the good, and there is no method of commensurating values across individuals. This gives rise to a fundamental individualrightto maintain one's personal conception of the good. To infringe upon this right means to offend against the inviolability of the person since it amounts to either denying the worth of that person completely, or to postulate a "superior" standard of goodness of which the individual conception of the good is but a part. If Rawls's critique of utilitarian theories of justice is to be taken seriously, his alternative account of justice must be understood to include the individual's right to maintain their own conception of the good. This gives rise to a conception of freedom as the freedom to have one's own rational plan of life. The first priority rule, then, is the device by which this freedom is ensured. No potential economic gain can justify an infringement on a person's right to have their own plan of life and to promote their individual values. Liberty in this sense is not merely a means to the fulfillment of one's rational plan of life, but a precondition for maintaining such a plan in the first place. It is an end in itself in that it is the necessary expression of the inviolability of the person. The conception of liberty as an end in itself has its origins not so much in the original position, although anyone who willingly imagines themselves in that position would acknowledge the first priority rule, but is prior to the original position. Thus Ronald Dworkin, for example, argues that: "The original position is well designed to enforce the  abstract right to equal concern and respect, which must be understood to be the funda  concept of Rawls's deep theory. [...] The right to equal respect is not, on his [Rawl  84 account, a product of the contract, but a condition of admission to the original position (Dworkin, p. 181) The 'right to equal concern" lies at the very foundation of Rawls's theory. And as I have argued throughout, in Rawls's context this right must be understood as the right, and hence liberty, to maintain one's own conception of the good. In this sense, liberty is "negative liberty" as Berlin uses the term (Berlin, p. 122). It is a liberty from infringement upon one's conception of the good. This liberty ensures that there is  an "area within which a [person] can act unobstructed by others," (ibid.) namely the area individual values and conceptions of the good. It is an abstract liberty that has implications on what kinds of justifications are acceptable for the principles of justice and what kinds of interpersonal comparisons can be made. As such, it does not necessarily entail the right to actually carry out one's plan of life.  This conception of liberty, imbedded in Rawls's theory of justice, has granted some commentators to conclude that liberty is purely abstract and can never give rise to conflicts. Serge-Christophe Kolm, for instance, argues with reference to Rawls that: "Equal and maximal basic liberties are limitless basic liberties because they are act-freedoms  essentially no rivalry across liberties and across individuals [...] This actual structure  basic liberties makes irrelevant a number of qualifications proposed by Rawls (such as  proposal of a maximin in basic liberties) and a number of criticisms that could be addre to his formulation." (Kolm, p. 176) By "act-freedom" Kolm means that they are freedoms to particular acts but not to the necessary means required for those acts. For example, freedom of speech as an act-freedom does not entail that anybody has a claim to resources such as time and space so that they may  85 actually be heard. Freedom of speech in this sense could not possibly conflict with another freedom or with another person's freedom of speech, for whoever actually gets to speak depends entirely on who has access to the resources, which is not a matter of act-freedom. Act-freedom thus means freedom from infraction once one has access to the necessary means. Nobody has any claim against my free speech once I have got a megaphone. As restrictive as this may seem, I believe Kolm is partially correct. The freedom to maintain one's own conception of the good as an expression of the inviolability of the person is indeed an act-freedom in this sense. It does not give rise to the right to actually carry out our plan of life, only to the right not to have anybody else, including society as a whole, impose their conception of the good on us. What Kolm does not see, however, is that Rawls does not use only this conception of freedom. There is another element of freedom that plays an important part in Rawls's theory, namely freedom as a means to carry out one's rational plan of life. And the fact that Rawls uses both aspects of freedom, freedom as a means and freedom as an end-in-itself, does indeed, as I shall argue, give rise to conflicts between liberties.  7.3.3 Freedom as a Means In order to see that Rawls is using a conception of freedom that is not exclusively act-freedom, compare his treatment of freedom to that of Hobbes: "By Liberty, is understood  [...] the absence of external impediments [...] In such a condition, every man has a rig every thing; even to one another's body." (Hobbes, p. ) In Hobbes's State of Nature, people are free in that there exist no impediments to their actions other than their own physical strength. Freedom is characterized purely by the absence of constraints (act-freedom), not by what this freedom actually achieves. And because there is therefore no limit to what one can do to others with one's physical power, according to  86  Hobbes, this freedom traps humankind in a "State of War". By appointing a sovereign and thus establishing the basis for social cooperation, people agree to give up their freedom. This is clearly not Rawls's view. Admittedly, Hobbes's is an extreme example (although it seems precariously close to Kolm's description of act-freedom as limitless). What separates Hobbes and Rawls is much more than their conceptions of freedom. Nonetheless, we may ask ourselves whether the parties in Rawls's original position might agree to a conception of freedom like Hobbes's. If all they were concerned with was to protect their right to maintain their individual conceptions of the good, why should they not accept a conception of freedom that gives everyone the right to use their (justly allocated) share of primary goods against each other completely as they please? Part of the answer must surely be that the protection of their right to maintain individual conceptions of the good is only part of what motivates the parties in Rawls's original position. They also have an interest in actually being able to realize their plans of life. And there is more they can do to ensure this than simply insist that the means for doing so be allocated according to the maximin rule. Seeing that the inviolability of the person requires the freedom to maintain one's individual conception of the good does not imply that freedom must stop there. In his discussion of the choice made by the parties in the original position Rawls introduces "rights and liberties" as amongst the primary social goods. This suggests that he is aware of the potential use rights and liberties have as means to fulfilling one's individual plan of life. For example, if we look at the resulting first priority rule, it is clear that in addition to lexically ordering the two principles of justice Rawls addresses the issue of trading-off liberties against  each other. Let me recite: "(a) A less extensive liberty must strengthen the total system  liberty shared by all; (b) a less than equal liberty must be acceptable to those with th liberty." (TJ, p.302)  87 As Kolm correctly points out, such qualifications (in Kolm's terms "maximin for liberties"; see above) are unnecessary if freedom is understood purely as act-freedom. And if freedom is conceived of as purely an end-in-itself, the protection of one's right to have an individual conception of the good, such act-freedom may be enough. That freedom be a matter of degrees only makes sense in Rawls's theory if freedom is understood as a means to something else. The fact that Rawls does mention possible trade-offs between liberties suggests that there must be a standard by which these liberties can be compared, for instance by their use as means in realizing one's plan of life. By contrast, if liberties are regarded purely as ends-in-themselves it is hard to see how their value could be compared. This is particularly relevant since Rawls speaks of a "total system of liberties". On page 61 of the Theory of Justice Rawls provides a list of the basic liberties: "The basic liberties of citizens  are, roughly speaking, political liberty (the right to vote and to be eligible for public off  together with freedom of speech and public assembly; liberty of conscience and freedo  thought; freedom of the person along with the right to hold (personal) property; and fr from arbitrary arrest and seizure as defined by the concept of the rule of law. "  7  If this list of liberties is to give rise to a "total system" of liberties, the liberties cannot be regarded only as ends in themselves. Rawls's first priority rule states that any of the liberties from this list may be circumscribed if doing so gives rise to a greater total system of equal liberties for all. This seems reasonable from the point of view of the original position if liberties are regarded as means to the end of fulfilling one's plan of life. It does not, however, make sense if freedom is an end-in-itself only.  Scanlon stresses that this list is only "indicating 'roughly speaking' what this class is to include. " (Scanlon, pp.181/182) The relevance with regards to the "total system", however, remains the same.  88 To make this clear, let me consider the contrast Rawls points out between a liberty and the worth of that liberty: "Liberty and the worth of liberty are distinguished as follows: liberty is represented by the complete system of the liberties of equal citizenship, while the worth of liberty to persons and groups is proportional to their capacity to advance their ends within the framework the system defines." (TJ, p.204)  Rawls rightfully, I believe, draws an important distinction between the liberty to do something and having the resources to do so. These two are distinct issues and the former does not necessarily give rise to a right to the latter. For example, the liberty of free speech does not give rise to a right to be listened to whenever one feels like it; nor does freedom of movement imply the right to an airplane ticket whenever one feels like going to Hawaii. This is different from the distinction between freedom as an end-in-itself and freedom as a means-to-an-end, however. Freedom as an end-in-itself, in Rawls's case, is, as I have argued, an expression of the inviolability of the person. It is not a means to the end of guaranteeing this inviolability but rather its equivalent. Insisting on a right to one's own conception of the good is the same as saying that "the person" is inviolable, for what being a "person" essentially consists in , in Rawls's framework, is to have a conception of the good. By contrast, freedom as a means-to-an-end implies that there is something else which this freedom makes possible. Thus freedom of speech, for instance, can be very valuable for being a journalist. This distinction is different from the one between liberty and its worth. Of course, material resources are still important in determining the extent to which a journalist can actually make use of their freedom of speech, but the distinction between freedom as an end-in-itself and as a means-to-an-end marks something else. This latter points at a discrimination that can be made in the worth of a liberty regardless of access to material resources. Only if freedom is regarded as a means-to-an-end does it make sense to say, for  89  example, that an increase in the freedom of speech is more valuable to a journalist than an increase in the right to be eligible for public office, or that their "total system of liberty" is greater when the former is increased at the expense of the latter. As long as Rawls refers to liberty as an end-in-itself only, he must maintain, as he initially does, that "all the liberties of  equal citizenship must be the same for each member of society." (TJ, p.204) However, dur the course of his discussion, he drops this strict demand and introduces "maximin for basic liberties" as expressed in the final formulation of the two principles of justice: "A less than  equal liberty must be acceptable to those with the lesser liberty," (TJ, p.302) under th constraint that a loss of liberty can never be justified by a gain in other primary goods (lexical ordering of the two principles). This amounts to the acknowledgement that the parties in the original position, who are to choose these principles, realize that "rights and liberties" can have a use not only in-themselves but for other things, too. Rawls's use of liberty as a means-for-an-end prompted at least one commentator to go so far as to say that "although freedom as a means is thefirst(and lexically prior) principle of procedural justice, freedom as an end receives little attention." (Barber, p.299) If my argument is correct both this and Kolm's comment quoted earlier are mistaken. It is true that a distinction must be made between freedom as an end and freedom as a means. But it does not follow that Rawls's conception of liberty falls into either of these categories. Instead, as I have argued, he must make use of both, and his conception of liberty can only be properly made sense of if both these aspects are borne in mind.  90 7.4 Conflicting Liberties In order, as we have seen, to be able to deduce the outcome of the deliberations of the parties in the original position, Rawls must make at least some substantive claims about conceptions of the good. By introducing the "thin theory of the good", however, he brings into the original position values other than the inviolability of the person. Because of this, it is no longer to possible for him to exclusively talk about freedom as a means-in-itself. Freedom has a potential as a means for other ends, too, namely the fulfillment of one's rational plan of life. In this light, freedom turns into something quite different from the purely abstract act-freedom Kolm discusses (see above). Freedom as a means is no longer unlimited and v  conflicts amongst liberties do indeed arise. In particular, since the parties in the original position are no longer concerned simply with "having" a plan of life but also with actually executing this plan, they will come to pay attention to the potential effects of another's use of their liberty on their own plan of life. As H.L.A.Hart correctly points out: "Any scheme  providing for the general distribution in society of liberty of action necessarily does tw  things:first,it confers on individuals the advantage of that liberty, but secondly, it ex  them to whatever disadvantages the practices of that liberty by others may entail for th (Hart, p.247) Admittedly, someone else's use of their freedom limiting the extent to which I am able to execute my own plan of life does not constitute a violation of my right to stick to my individual plan of life (inviolability of the person). It does, however, constitute a restriction which the parties in the original position may wish to prevent by agreeing to limit their own liberties. According to Rawls this should be done in a way that maximizes the "total system of liberties" available to all (first priority rule). By introducing "maximin for liberties", Rawls  91 acknowledges that the use a liberty has as a means for executing one's plan of life is a relevant factor is determining the "just" distribution of liberties. Nonetheless, Rawls seems reluctant to follow through with this idea; rightly so, as I shall argue. Norman Bowie thus comments: "[Thefirstprinciple] seems excessively general and  totally ignores the weighing and balancing of the various liberties which must occur in  for an individual to have the most extensive total system of basic liberties." (Bowie, p.l 1  And similarly, Brian Barry criticizes that: "Rawls gives little usable guidance about the way  to aggregate the different liberties so as to arrive at an estimate of the total amount of li  generated by alternative combinations of these different liberties [...This] would obviou  of crucial importance if one were seriously to attempt to apply the 'two principles ofju in a real society" (Barry, p.34) In other words, although Rawls admits the concept of a "total system of liberties", thus making room for freedom as a means, he does not offer a good solution to the problem of how to compare liberties and calculate the maximum "total system". One of the reasons why Rawls does not provide a solution to this problem may be that he cannot. He must introduce ends other than the inviolability of the person into the original position in order to derive the principles of justice. This does not imply, however, that freedom as an end in itself is no longer important or indeed but one of many ends the parties in the original position wish to pursue. As I have argued, freedom as an end in itself, the inviolability of the person, is a crucial determinant of the parameters of the original position. As such, it must be maintained and secured by the parties in that position. It is Rawls's use of freedom as a means that gives rise to conflicting liberties. It is his dependence upon freedom as an end in itself, however, that forbids him to provide a system by which liberties can be evaluated and compared. To  92 make this clear, let me consider the role of the constitutional convention as part of the four-stage sequence by which the principles of justice are to be applied.  7.4.1 The Constitutional Convention  "So far I have supposed that once the principles ofjustice are chosen the parties ret  their place in society and henceforth judge their claims on the social system by t  principles. But if several intermediate stages are imagined to take place in a defini  sequence, this sequence may give us a schema for sorting out the complications that  faced. Each stage is to represent an appropriate point of view from which certain ki  questions are considered. Thus I suppose that after the parties have adopted the princ  justice in the original position, they move on to a constitutional convention. [...] I i  then a division of labor between stages in which each deals with different questions o  justice. [...] Thefirstprinciple of equal liberty is the primary standard for the constitut  convention. Its main requirements are that the fundamental liberties of the person and of conscience and freedom of thought be protected." (TJ, p. 199) According to Rawls, once the principles of justice have been agreed to, these impose some restrictions on conceptions of the good. The principles of justice, themselves arrived at without reference to any particular conception of the good, take priority over individual such conceptions. That is, the principles of justice are binding even if they should conflict with someone's conception of the good or plan of life. It is therefore possible to allow the gradual lifting of the veil of ignorance and introduce knowledge that is necessary to apply the principles of justice and develop the just institutions of society. As part of a four-stage sequence, therefore, Rawls proposes a constitutional convention as the second stage (after the original position). Since the parties to the constitutional convention are bound by the  93 principles of justice, it would seem, the constitution they establish will be equally in accordance with the demands of justice. At the stage of the constitutional convention, some commentators have therefore suggested, it may be possible to resolve the potential conflicts between liberties and to specify how the basic liberties are to apply to citizens: "Rawls tackles  this theoretical issue [of how to weigh various liberties] when he considers how the pr  of equal liberty for all is to be actualized in the institutions of society. Rawls' means fo achieving this is the constitutional convention." (Bowie, p.l 17) The idea behind this is that since the principles of justice are by now firmly in place, it may be possible to allow knowledge about the particular plans of life the citizens of a given society entertain and upon this basis measure the relative value of basic liberties as means: "It is true  that at the stages in the four-stage sequence where such conflicts [between liberties]  be resolved there is no veil of ignorance to prevent those who have to take decisions k Q  what proportion of the population favour which alternative," (Hart, p.242) Once we allow the parties to the constitutional convention to have access to the knowledge of how the citizens of a given society (on average) value different liberties, it should be possible to establish how potential conflicts between liberties are to be resolved, subject to the limitations set out by the principles of justice. There is, however, a mistake in this line of reasoning if we are to understand that because the principles of justice themselves are indeterminate with respect to conflicts between liberties, they do not impose limitations on how this indeterminacy is to be resolved. There is a reason why the indeterminacy exists in the first place, and this is a reason that retains its strength even after the principles of justice  have been agreed to. "[Some] conflicts between basic liberties will be such that differ Note that Hart immediately goes on to say: "But I do not think Rawls would regard such knowledge as relevant." (ibid.) 8  94 resolutions of the conflict will correspond to the interests of different people who wil  diverge over the relative value they set on the conflicting liberties. In such cases, the  no resolution which will be uniquely selected by reference to the common good." (H p.241) The conflict that is to be resolved by the constitutional convention is not simply one between liberties in the abstract, but fundamentally between the different values that are placed upon them by different people. In other words, it is a conflict between conceptions of the good. The only way to resolve the conflict is by somehow aggregating across these conceptions of the good. (Since it would be inadmissible for the parties in the constitutional convention simply to take sides with a particular conception of the good.) Such an aggregation, however, would be in violation of the conception of freedom-as-an-end. I have argued that Rawls's first principle is chosen by the parties in the original position primarily as an expression of the inviolability of the person. If my argument in this respect is correct, then this inviolability must be upheld by the parties in the constitutional convention even though its expression in the first principle of justice may be obscure. According to my interpretation of Rawls's argument, the first principle says that every person has an inviolable right not to have another's conception of the good imposed on them. This is the reason why the principles of justice have to be indeterminate with respect to conflicting liberties. The resolution of this indeterminacy by means of an aggregation across conceptions of the good (for example via majority rule) is unacceptable at whatever stage, because it would constitute a violation of the first principle of justice (its spirit, if not its letter). Indeed, what Rawls envisages with his four-stage sequence is a procedure for arriving at the just outcome as independently specified by the principles of justice. The gradual lifting of the  95 veil of ignorance is supposed to be an instance of perfect procedural justice (or as "perfect" as feasible), not pure procedure. The just outcome has to be independently specified prior to  the constitutional convention: "Inframinga just constitution I assume that the two principle  of justice already chosen define an independent standard of the desired outcome. If ther such standard, the problem of constitutional design is not well posed." (TJ, p. 198) If my argument up to this point is correct, then the principles of justice do not provide a determinate content for the just distribution of liberties. The constitutional convention is not well defined and cannot resolve conflicts between liberties. Note that Rawls is well aware of this when, in discussing the gradual lifting of the veil of ignorance, he points out that: "The  flow of information is determined at each stage by what is required in order to apply th  principles intelligently to the kind of question of justice at hand, while at the same tim  knowledge that is likely to give rise to bias and to set men against one another is ruled (TJ, p.200) The inviolability of the person requires that particular interests, values, or point of view do not determine the outcome of justice. While the conception of freedom as a means introduces the need to think of a "total system" of liberties and to weigh liberties against each other, the conception of freedom as an end in itself absolutely makes it impossible for Rawls to produce a method by which the required weights can be chosen.  7.5 Summary The two principles of justice which, according to Rawls, would be chosen by the parties in the original position make a distinction between "rights and liberties" and the other primary goods. "Rights and liberties" are never to be traded-off against gains in the other primary goods. As I have argued, this is an expression of the inviolability of the person upon which  96 the whole of Rawls's theory rests. Freedom in this sense should be interpreted as the freedom not to have someone else's conception of the good imposed on oneself. It is the freedom to have one's own conception of the good and is inalienable. This is freedom as an end-in-itself. However, Rawls does not use the concept of freedom in this sense alone. "Rights and liberties" are also acknowledged as amongst the primary social goods which are means to ends other than themselves. In addition to referring to freedom as an end in itself, Rawls also makes use of the conception of freedom as a means. This aspect of freedom, however, necessitates that it be thought of as a "total system" of liberties rather than limitless and purely abstract freedom. It thus becomes possible and necessary that individuals rank different kinds of liberties, such as freedom of thought or freedom to be eligible for public office, according to their usefulness as means in actualizing their different plans of life. This implies that liberties can conflict, at least with respect to their relative value, and in order for the parties in the original position to maximize their "total system" of liberty there must be agreed standards on how to compute this "total system". In  this  way  freedom-as-a-means  produces  conflicts  between  liberties,  whilst  freedom-as-an-end requires that the pluralism of conceptions of the good, and thus of the weights assigned to different liberties, not be violated. The importance of freedom-as-an-end prohibits the resolution of conflicts between liberties. If Rawls wants to maintain the inviolability of the person, his conception of liberty must remain indeterminate.  97 Chapter 8: Conclusion In his Theory of Justice, Rawls relies on two different aspects of freedom: Freedom as a means and freedom as an end. Freedom as an end features heavily in determining the structure of the argument. This aspect of freedom is an expression of the inviolability of the person that Rawls wishes to preserve. Not everyone, according to Rawls, shares the same conception of the good. In other words rather than there being one single system of values that holds universally, at least within a given society, there exists a pluralism of incommensurable values. A theory of justice, in order not to violate the sanctity of the person, must therefore not infringe upon the pluralism of values. It must not be based on any particular conception of the good. Freedom is thus an end in itself in that it guarantees every person's right to their own conception of the good. It is because of this aspect of freedom, I have argued, that Rawls chooses to develop a theory of pure procedural justice, which does not require a prior conception of what the just outcome should be, rather than one of perfect procedural justice. It is because of freedom-as-an-end, too, that the veil of ignorance must hide particular conceptions of the good from the parties in Rawls's original position. Finally, freedom-as-an-end is expressed in the two principles of justice that Rawls proposes through the lexical ordering of the two principles. Liberty must not be sacrificed for gains in other primary goods. The sanctity of one's own conception of the good cannot be violated, that is the claim that there exists a conception of the good that is more valid than others can at no point be justified to those that do not share this particular conception of the good. However, freedom-as-an-end is rather abstract and indeterminate. It does not, for instance, guarantee that one's personal conception of the good can actually be put into practice. Furthermore, upon the basis that comparisons across conceptions of the good are inadmissible no theory of social justice can be constructed. If people can never agree on any normative  98  issue, then there will never be a conception of justice that all accept. More specifically, if all knowledge of desires and interests is to be excluded from the original position, then it is impossible for the parties there to choose the principles of justice. They have no idea what it is they want. For this reason, Rawls must introduce a "thin theory of the good" which provides the basic shared preferences for the parties in the original position to be able to make a rational choice between alternative principles of justice. Even if we accept Rawls's claim that the "thin theory" is compatible with the inviolability of individual conceptions of the good, this theory introduces ends other than the protection of this inviolability into the account of justice. This is a necessary implication of a necessary step. Once ends other than the inviolability of the person are introduced, however, it becomes necessary for Rawls to consider the second aspect of liberty, liberty-as-a-means. Since freedom can have a use in pursuing one's rational plan of life it now becomes possible to measure the value of freedom with respect to this pursuit. Different kinds of freedom, for example freedom of speech and freedom of property, can have very different value as means depending on one's plan of life, and it may be possible for the parties in the original position to justify to each other decreases in one kind of liberty if the overall value of the "total system" of liberties-as-means is thereby increased. It would be unconvincing for Rawls to claim that the parties in the original position are not allowed to consider this (which he does not). But since the weighing of liberties-as-means now becomes necessary, there arise conflicts between liberties. This conflict consists on the one hand in the fact that the value of the "total system" of liberties is not necessarily maximized when all liberties are realized to an unlimited extent. The value of a liberty as a means depends not only on its positive impact on the pursuit of one's own plan of life, but also on the negative impact due to other people's exercise of their liberty. This is the reason why it is necessary to consider the "total system" of  99  liberties. More importantly, on the other hand, the conflict between liberties consists in a conflict between the different relative weights assigned to different liberties by different people. These weights depend on one's particular plan of life and one's conception of the good. And since these are not the same for everyone there will be disagreement over the "correct" weights. Rawls does not tackle this issue. At best he can be understood to offer the four-stage sequence, in particular the constitutional convention (stage two), as a procedure for resolving the conflict. This, however, is unacceptable. The value of liberty-as-a-means is important, but this does not mean that liberty-as-an-end no longer is. That the inviolability of the person be upheld is absolutely required. If Rawls at any stage dropped this inviolability from his account of justice he would be undermining his whole theory and could no longer claim to offer a better alternative than the utilitarian account of justice, where the inviolability of the person is indeed not guaranteed. Also, from the point of view of the original position, Rawls's principles of justice would not be chosen if they violated the individuality of conceptions of the good. The parties in the original position are in part characterized by the fact that they maintain the inviolability of the person as a principle we are morally obliged to uphold. The original position can fulfill its task as the standpoint of justice only if this assumption holds true. It is an essential part of Rawls's argument that it is impossible for us to justify to each other as free and rational persons that there should be one universal system of values that determines the weights of the different liberties. Rawls cannot solve this conflict. Furthermore, if this conflict cannot be resolved within Rawls's framework, then the principles of justice contain considerable indeterminacy. This is more than a small inconvenience because part of the whole point of the Theory of Justice is to provide a systematic alternative to intuitionism, in other words, to explain the underlying structure of  100 our intuitions about justice so that conflicts between intuitions can be resolved. If now Rawls ends up with considerable indeterminacy which can only be resolved intuitively by each individual, he has failed at completing one of the main tasks he set himself. However good we may find his analysis of our intuitions and their underlying structure, in the end we still have to fall back upon utilitarian calculations as the only alternative available to resolve the conflicts between intuitions. Having said all this, I do believe Rawls is on the right track and his attempt is by no means futile. I do believe that his notion of a "standpoint of justice" from which we can in a pluralistic society justify conceptions of justice to one another is a very important one for political philosophy. Furthermore, I do not think there is anything intrinsically wrong with considering freedom both as an end in itself and as a means to other ends. On the contrary, in my opinion it would be naive not to do so. However, what is needed is a conceptualization of how these two aspects of freedom relate to each other which Rawls does not offer. An important task that remains would be to analyze how individual conception of the good on the one hand, and shared values on the other are connected. Rawls relies heavily on both kinds of values but does not offer an account of why some values should be purely individualistic whereas others are shared. Providing a comprehensive theory of this relationship may contain insights into how we can justify to each other the relative weights of different liberties while at the same time not violating the individual's right to have their own conception of the good.  101 BIBLIOGRAPHY Anderson, C.W.: Pragmatic Liberalism. Chicago: University of Chicago Press, 1990 Arrow, K.J.: "Some Ordinals Notes on Rawls's Theory of Justice", in: Journal of Philosophy, no.9, 1973: pp.245-263 Barber, B.: "Justifying Justice: Problems of Psychology, Politics and Measurement in Rawls", in: Reading Rawls, by Daniels, N. (ed.)- Stanford: Stanford University Press, 1989: pp.292-318 Barry, B.: Liberal Theory of Justice. Oxford: Clarendon Press, 1973 Berlin, I.: Four Essays on Liberty; Oxford University Press; Oxford; 1969 Bowie, N.: "Equal Basic Liberties for All", in: John Rawls' Theory of Social Justice, by Blocker, H.G. and Smith, E.H. (eds.). Athens: Ohio University Press, 1976: pp.110-131 Brehmer, K.: Rawls' "Original Position" oder Kants "Urspruenglicher Kontrakt". Koenigstein, Ts: Verlag Anton Hain Meisenheim GmbH, 1980 Darwall, S.L.: "Is There a Kantian Foundation for Rawlsian Justice?", in: John Rawls' Theory of Social Justice, by Blocker, H.G. and Smith, E.H. (eds.). Athens: Ohio University Press, 1976: pp.311-345 Dworkin, R.: Taking Rights Seriously. Cambridge, Mass.: Harvard University Press, 1977 Frazer, E. and Lacey, N.: The Politics of Community. London: Harvester Wheatsheaf, 1993 Graham, K.: The Battle of Democracy. Brighton: Wheatsheaf Books, 1986 Hart, H.L.A.: "Rawls on Liberty and its Priority", in: Reading Rawls, by Daniels, N. (ed.). Stanford: Stanford University Press, 1989: pp.230-252 Kolm, S.-C: Modern Theories of Justice. Cambridge, Mass: MIT Press, 1996 Kymlicka, W.: "Individual and Group Rights", in: Group Rights, by Baker, J. (ed.). Toronto: University of Toronto Press, 1994: pp. 17-33 Levine, A.: "Rawls' Kantianism", in: Social Theory and Practice, vol.3, no.l, 1974: pp.47-63 Martin, R.: Rawls and Rights. Lawrence: University Press of Kansas: 1985 Mill, J.S.: On Liberty. London: Penguin Books, 1974 Nagel, T.: "Rawls on Justice", in: Philosophical Review, no.82, 1973: pp.220-234  102 Norman, W.J.: "Towards a Philosophy of Federalism", in: Group Rights, by Baker, J. (ed.). Toronto: University of Toronto Press, 1994: pp.79-99 Pettit, P. and Kukathas, C : Rawls. Cambridge: Polity Press, 1990 Pogge, T.W.: Realizing Rawls. Ithaca: Cornell University Press, 1989 Rawls, J.: A Theory of Justice (TJ). Cambridge, Mass.: Harvard University Press, 1971 Rawls, J.: Political Liberalism. New York: Columbia University Press, 1993 Raz, J.: The Morality of Freedom. Oxford: Clarendon Press, 1986 Sandel, M.: Liberalism and the Limits of Justice. Cambridge: Cambridge University Press, 1982 Scanlon, T.M.: "Rawls' Theory of Justice", in: Reading Rawls, by Daniels, N. (ed.). Stanford: Stanford University Press, 1989: pp. 169-205 Schaar, J.: "Reflections on Rawls' Theory of Justice", in: Social Theory and Practice, vol.3, no.l, 1974: pp.75-100 Schaefer, D.: "A Critique of Rawls' Contract Doctrine", in: The Review of Metaphysics, vol.28, 1974: pp.89-115 Taylor, C : Multiculturalism. Princeton: Princeton University Press, 1994 Waldron, J.: Liberal Rights. Cambridge: Cambridge University Press, 1993 Wolff, R.P.: Understanding Rawls. Princeton: Princeton University Press, 1977  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088516/manifest

Comment

Related Items