Chapter 5 Chapter 2 Principles of Justice-1

A theory of justice can be divided into two parts: (1) explaining the original position and proposing principles that could be used to make choices in the original position, and (2) demonstrating which principles might in fact be adopted.This chapter discusses two principles of justice that apply to institutions and several principles that apply to individuals, and explains what these principles mean.For the moment, therefore, I am concerned only with one aspect of the first part of the theory.It is not until the next chapter that I proceed to explain the original position problem and begin to develop arguments that the principles considered here would in fact be recognized.Among the various issues to be discussed are: the concepts of institutional and formal justice as subjects of justice; the three types of procedural justice; the status of the theory of the good; the notion that principles of justice are egalitarian principles, and others.The purpose of each discussion is to illustrate the meaning of these principles and their application.

The basic theme of the principles of social justice is the basic structure of society, that is, the transformation of the main social institutions into a cooperative arrangement.We have seen that the purpose of these principles is to guide the distribution of rights and duties in these institutions.Determine the appropriate distribution of the benefits and burdens of social life.The principles of justice that apply to institutions must not be confused with the principles of justice that apply to individuals and their actions in particular circumstances.These two principles apply to different objects and must therefore be discussed separately.

According to my understanding, the so-called institution is a system of public rules, which stipulates the positions and positions and their rights and obligations, stipulates the powers and immunities, and so on.These rules specify that certain actions are permissible and others are prohibited; they also prescribe certain penalties and defenses for possible violations, and so on.We can think of games and rituals, trials and parliaments, markets and property institutions as examples of institutions or social conventions more generally.Institutions can be viewed in two ways: first as an abstract goal, that is to say, as a possible way of behaving expressed by some system of rules; The actions specified by these rules are effectively realized in thought and behavior.It is somewhat ambiguous, then, which is just or unjust, as a realized system or as an abstract goal.It seems better to say that the systems that have been realized and that are effectively and fairly administered are just or unjust.The so-called system as an abstract goal is just or unjust, just means that the realization of this system may be just or unjust.

An institution existed at some time and place if the actions prescribed by the institution were normally carried out according to a general agreement that the system of rules governing the institution was observed.A parliamentary system, for example, is regulated by some system of rules (or such systems which may vary).These rules enumerate certain ways of doing things, from calling a parliamentary meeting, to voting on a bill, to raising a point of order, and so on.Various general principles are combined into a coherent arrangement.If certain persons perform appropriate actions, perform them in a prescribed manner, and recognize each other's agreement that their actions conform to the rules to which they are bound, then there is a certain time and place a parliamentary system.

Therefore, when I say that a certain system, the basic structure of society, is a system of public rules, I mean that everyone participating in the system knows what he can know if these rules and his participation in the activities prescribed by these rules It is the result of some kind of agreement.A person who is part of a system knows what the rules require of him and of others.He also knows that others know it, and that they know that he knows it, and so on.Of course, this condition is not always fulfilled in terms of actually existing institutions, but it is a reasonably simple assumption.The principles of justice must apply to the general arrangements of society understood in this sense.If the rules of a minor part of an institution are known only to those belonging to that part, then we can presume that there is an agreement that, as long as the purpose of these rules is to achieve recognized ends, no one else will adversely affected, then this part of the population can make rules for themselves.Making the rules of the system public ensures that those who participate in the system know what limits are placed on each other's behavior and what kinds of behavior are permissible.This provides a common basis for establishing mutual expectations.Moreover, in a well-ordered society, that is, a society effectively governed by a common conception of justice, there is also a general agreement on what is just and what is unjust.I will also assume below that the principles of justice are chosen with the knowledge that they are general principles (Section 23).In contract theory, this condition is a natural condition.

It must be pointed out that there is a difference between the basic rules that define the various rights, duties, etc. of a system, and the strategies and guidelines for how best to use the system to achieve certain ends.Sound strategies and principles are based on an analysis of which permissible actions individuals and groups will select based on their interests, beliefs, and presumptions about each other's plans.These strategies and norms are not part of the system per se, but belong to a theory about the system, for example, to the theory of parliamentary politics.Like theories of games, theories of institutions generally assume that the ground rules are given, analyze how power is distributed and describe how those who participate in the system are likely to take advantage of the opportunities that the system provides.In designing and reforming social arrangements, one must of course study carefully the various schemes and strategies that the system allows, and the various modes of behavior that the system tends to encourage.Ideally, rules should be formulated so that people, guided by their primary interests, act in ways that promote the desired goals of society.The actions of individuals are guided by their rational plans and must therefore be as coordinated as possible to achieve outcomes that are not expected, perhaps not even foreseen, but from a social justice perspective.They are still the best results.Bentham saw this harmony as an artificial coincidence of interests, and Adam Smith thought it was the work of an invisible hand.This is the legislative goal of the ideal legislator, and it is the goal of moralists who urge reform of the law.However, the strategies and tactics that individuals follow, while essential to evaluating institutions, are not part of the system of public rules that dictate them.

We can also distinguish a single rule (or set of rules), a system (or major parts of the system) from the basic structure of the social system as a whole.The reason for this is that one or a few rules of an arrangement may be unjust, but the system itself is not.Likewise, a system may be unjust, although the entire social system is not unjust.Not only may single rules and institutions not be of sufficient importance in themselves, but it is also possible within the structure of an institution or social system that one apparent injustice compensates for another.If the whole contains only one unjust part, the whole is less unjust than it could be.Further, it is conceivable that a social system could be unjust even if none of its institutions were individually unjust: the injustice would be the combination of many institutions into a single one. result of the method.One system may encourage and seem to justify expectations that are denied or ignored by another system.These differences between regimes are quite stark.They simply reflect the fact that, in evaluating institutions, we may be looking at them in a wider or narrower context.

It should be noted that the concept of justice is generally not applicable to some systems.Rituals, for example, are not usually regarded as just and unjust, although examples can no doubt be conceived where this is not the case, eg, rituals in which first-born children or prisoners of war are sacrificed.A general theory of justice might consider when rituals and other practices not usually thought of as just or unjust do come under such criticism.Presumably they must somehow involve the distribution of certain rights and values ​​among people.However, I do not plan to do further research in this area.We are concerned only with the basic structure of society and its principal institutions, and thus with social justice in general.

For now, let's assume that there is some basic structure.Its rules conform to a certain conception of justice.We ourselves may not accept its principles; we may even consider them repugnant and unjust.But they are principles of justice in the sense that they assume a just role for the system: they define the distribution of fundamental rights and duties, and they determine the distribution of the benefits of social cooperation.Let us also imagine that this conception of justice is generally accepted in society and that the system is administered fairly and consistently by judges and other officials.That is, the same situations are all treated in the same way, and the relevant similarities and differences are confirmed by the current norm.The correct rules laid down by the system are often observed and properly accounted for by the authorities.This just and consistent administration by laws and institutions, whatever their real principles may be, we may call formal justice.If we think of justice as always showing an equality, then formal justice requires that laws and institutions, when administered, should apply equally (that is, in the same way) to persons belonging to the classes they define.As Sidgwick emphasizes, once this equality is conceived as an arrangement of general rules, it is included in the notion of law or institutions.Formal justice is adherence to principles, or, as some say, obedience to institutions.

Sidgwick went on to say that clearly laws and institutions can be applied fairly and still be unjust.It is not enough for the same situation to be treated the same to ensure true justice.This depends on principles, because the basic structure is constructed according to these principles.There is nothing contradictory, though perhaps improbable, in assuming that a society of slaves or castes, or one that encourages the most arbitrary forms of discrimination, is governed fairly and consistently.Nevertheless, formal justice, or the idea that justice is uniform, excludes gross injustice.If the system is assumed to be fairly just, it is important that the authorities deal with certain particular situations impartially, independent of personal, pecuniary, or other irrelevant considerations.As far as the legal system is concerned, formal justice is only one aspect of the rule of law that supports and guarantees legitimate expectations.There is an injustice resulting from the failure of judges and other authorities to adhere to or account for the proper rules in adjudicating claims.A man is unjust if his character and inclinations should incline him to such an action.Furthermore, even when laws and institutions are unjust, they tend to be better if they are consistently enforced.Thus those who are subject to these laws and institutions may at least know what they require, and may endeavor to protect themselves; And the injustice is even greater if, in some cases, it is treated arbitrarily.On the other hand, in some cases, deviating from existing norms to alleviate the plight of those who have been wronged may be a good thing.How much justification we have in doing so, especially at the expense of expectations built in good faith on the existing system, is a complex question of political justice.In conclusion, what can be said is that the strength of the demand for formal justice, that is, the demand for compliance with the system, obviously depends on the real institutional justice and the possibility of institutional reform.

Some have argued that real justice and formal justice often coincide in fact, and that, therefore, institutions that are at least largely unjust are never, or at any rate seldom, justly and consistently administered.Those who uphold and benefit from unjust arrangements, and who contemptuously deny the rights and liberties of others, are said to be unlikely to allow scruples about the rule of law to interfere with their interests in particular cases.The inevitable ambiguity of general law, and the ample latitude possible for its interpretation, encourage arbitrary behavior in decision-making, which can be moderated only by an adherence to justice.Therefore, it has been argued that as long as we find formal justice, the rule of law and respect for legitimate expectations, we may also find real justice.The desire to follow the rules fairly and consistently, to have the same situations treated alike, and to accept the consequences of the application of universal norms.is closely related to the desire (or at least the willingness) to recognize the rights and liberties of others and to share equitably the benefits and burdens of social cooperation.One desire is often linked to another.This argument certainly seems plausible, but I do not intend to examine it here.For only until we know what are the most sound principles of true justice.This argument can be properly evaluated by knowing the conditions under which these principles are finally recognized as a way of self-discipline.Once we understand the content of these principles and their grounds in reason and in the attitudes of men, we may be able to determine whether true justice is closely related to formal justice. I shall now state, in a provisional manner, two principles of justice which I think might have been chosen in the original position.In this section I intend to make only the most general remarks, so this first account of these two principles is only an attempt.As the review progresses, I will touch briefly on several formulations, before approaching the final notes below.Doing so, I believe, allows the exposition to unfold naturally.A first statement of these two principles follows: First Principle: Every individual has an equal right to the broadest range of fundamental liberties that can coexist with similar liberties of others, Second Principle: Social and economic inequality should be arranged so that it is not only (1) reasonably expected to be in everyone's interest, but (2) is linked to positions and offices open to all. Together.In the second principle, there are two ambiguous terms, "everyone's interest" and "open to all".A more precise determination of the meaning of these two terms will lead in Section 13 to a second statement of this principle, to which Section 46 will give its final statement; Section 39 will examine the interpretation of the first A description of the principle. A general remark about these two principles, as I have already said, is that they apply primarily to the basic structure of society.They can guide the distribution of rights and duties and the distribution of socioeconomic benefits.The elaboration of these two principles shows that they presuppose that the social structure can be divided into two more or less distinct parts, the first applying to one part and the second applying to the other.These two parts are divided into those aspects of the social system which prescribe and guarantee the equal liberty of citizens, and those which prescribe and establish social and economic inequalities.Roughly speaking, the basic liberties of citizens are political liberties (the right to vote and the right to hold public office) as well as freedom of speech and assembly, freedom of conscience and thought, liberty of the person and the right to own (personal) property, according to the concept of the rule of law Freedom from arbitrary arrest and detention is provided for.The first principle states that all these liberties are equal, since citizens in a just society should have the same basic rights. The second principle presumably applies first to the distribution of income and wealth, and to the design of organizational structures or chains of command that exploit differences in power and responsibility.While the distribution of wealth and income need not be equal, it must be beneficial to everyone, and positions of power and command must be accessible to all.Applying the second principle is to make positions of power open to all, and then to arrange social and economic inequalities according to this constraint so that everyone benefits. These two principles are to be arranged in a sequence, with the first principle taking precedence over the second.This order implies that even greater social and economic interests cannot justify or compensate for a departure from the regime of equal liberty enshrined in the first principle.The distribution of wealth and income, and the hierarchy of power, must be consistent with equal civil liberties and equal opportunity. Obviously, the content of these two principles is quite specific, but the acceptance of these two principles depends on certain assumptions that I finally have to try to explain and prove.As the discussion progresses, it becomes increasingly apparent that theories of justice depend on social theories.The point that should be made at this point is that these two principles (which apply to all accounts) are special cases of a more general conception of justice which can be formulated as follows: Broadly, the only reason for doing so is: These equal rights provided for above may interfere with each other. Another thing to keep in mind: if a principle refers to people, or requires everyone to benefit from some kind of inequality, it means having a representation of the various social positions or positions, etc., dictated by the basic structure. people.In applying the second principle, therefore, I assume that certain expectations of well-being can be assigned to representative individuals occupying these positions.This expectation indicates their life prospects in terms of their social status.In sum, the expectations of the representative individual are determined by the distribution of rights and duties throughout the structure.If that changes, expectations change with it.I therefore postulate that people's expectations are linked: if we improve the life prospects of a representative person in one position, we may make the life prospects of a representative person in another position either better or worse. worse.Since the second principle applies to institutional forms, it (or rather, its first part) concerns the expectations of representative individuals.As I shall argue below, neither of these principles applies to the assignment of particular goods to particular individuals who can be named.As for someone who is considering how to distribute certain commodities among the poor he knows.This situation falls outside the scope of application of these two principles.All they have to do is adjust the basic institutional arrangements.We must not think that, from the standpoint of justice, there is much resemblance between the administrative distribution of the good to particular persons and the proper design of society.Our commonsense intuitions about the former are not good guides to understanding the latter. The second principle is the insistence that everyone should benefit from the permissible inequalities in the underlying structure.That is to say, if every representative person concerned, determined by the basic structure, regards the basic structure as a constant concern, he would rather have the prospect of life with inequality than a life without inequality Prospects, this is presumably reasonable.One cannot justify differences in income or organizational power on the grounds that the greater benefits to those in another position outweigh the losses to those in one position.Still less can this method be used to offset violations of the right to liberty.The application of the principle of utility to the basic structure enables us to maximize the total amount of expectations of representative persons (measured, in the classical sense, by the number of persons they represent); .On the contrary, these two principles require that everyone can benefit from economic and social inequality.It is evident, however, that if the original arrangement of equality is taken as a basic standard, there are innumerable ways in which all can be benefited.How, then, do we choose among these possibilities?These two principles must be elucidated so that they lead to a definite conclusion.Now, let me talk about this issue. I have already mentioned that, because of the ambiguity of the terms "everyone's interest" and "equally open to all," both parts of the second principle have two proper meanings.Since these meanings are independent of each other, the principle has four possible meanings.Assuming that the first principle of equal liberty remains the same throughout, we have four interpretations of these two principles.These interpretations are represented by the following table: "Everybody's Interest" "Equal Openness" —————————————————— principle of difference ——————————————————————————————— Equality, natural liberty, system of opening professions to talents, natural aristocracy Equality is equality of opportunity equality freedom equality democracy equality I will outline these three explanations in turn, the institution of natural liberty, liberty equality, and democratic equality.In some respects this sequence is more of an intuitive sequence, but the sequence arranged by the interpretation of natural aristocracy is not unimportant, so I will also comment briefly on it.In advancing the idea that justice is fairness, we must decide which explanation is preferable.I will adopt the interpretation of democratic equality and show what this concept means in this chapter.The reasons in favor of accepting this concept in its original position are not discussed until the next chapter. I shall call the first explanation (in both sequences) the institution of natural liberty.In this formulation, the first part of the second principle is understood as the principle of efficiency, adapted to apply to institutions, in this case, to the basic structure of society; the second part is understood as an open The social system, according to the traditional saying, under this social system, all kinds of undertakings are open to talents.In all these explanations I assume that the first principle of liberty about equality is realized and that the economy is largely a free market system, although the means of production may or may not be privately owned.Thus, the system of natural liberty holds that the basic structure that realizes the principle of efficiency and makes positions open to those who can and will work hard to gain them will lead to a just distribution.This allocation of rights and duties is said to produce an arrangement.Such an arrangement distributes wealth and income, power and responsibility, in whatever manner is justified by the outcome of that distribution.This theory includes an important component of pure procedural justice, and the issue of procedural justice will be explained later. Here it is necessary to use a few off-topic words to explain the principle of efficiency.This principle is nothing more than the (what economists call) Paredo optimization principle proposed for application to basic structures.However, we'll always use the word "efficiency" because it's the correct word in its sense of the word, and the word "optimization" suggests a much broader concept than it actually does.Of course, this principle is not originally intended to apply to institutions, but to a certain structure of economic institutions, for example, to the distribution of goods among consumers, or to modes of production.This principle holds that a structure is efficient as long as it cannot be changed to make some people (at least one person) happier without making others (at least one person) less fortunate .Thus, the distribution of existing goods among some individuals is efficient if there is no redistribution of good that improves the situation of at least one person without disadvantaging another person.An organization of production is efficient if there is no way to vary the inputs so that more of one good is produced without causing less production of another good.If we could produce more of one kind of good without having to produce less of another good, then more of the existing good could be used to improve the situation of some without harming the situation of others deterioration.These applications of this principle show that it is indeed a principle of efficiency.A distribution or production arrangement of good is inefficient if there is also a way to make some people better off without making others worse off.I will assume that parties in the original position accept this principle in order to judge the efficiency of economic and social arrangements (see additional discussion of the efficiency principle). efficiency principle Assume that the available quantity of goods that can be distributed between X1 and X2 is fixed.When using the points on the AB line to represent the benefits of X1 at the corresponding level, except for the point indicated by the curve, it is impossible to make the distribution of goods favorable to the points of X2.Let point D = (a,b).Then let X1 be in the position of plane a, then the best result that X2 can get is horizontal line b. In Figure 3, the origin O represents the position before any commodity allocation. Each point on the AB line is the efficiency point. Every point on the AB line can be seen as meeting Paredo's criterion: no redistribution can make either of the two people better off without making the other worse off.This is indicated by the downward slope of the AB line to the right.Since the quantity of goods available is fixed, one person gains and the other must lose (of course, if the basic structure is a cooperative system that produces a certain amount of positive benefits, this assumption no longer exists).Generally speaking, the OAB region is regarded as a kind of convex set, that is, if there are any pairs of points in the set, the points on the line connecting the two points are also in the set.Circles, ellipses, squares, triangles, etc. are all convex sets. Obviously, there are in fact many efficiency points, all of which lie on the line AB.The principle of efficiency does not itself select a particular distribution of goods as efficient.It is necessary to select some other principle in efficient distribution; that is, the principle of justice. Among these two points, if one point is northeast of the other, then according to the principle of efficiency, this point is the better point.Points in the northwest or southeast are not comparable.The order prescribed by the efficiency principle is only a partial order.Therefore, in Figure 4, although C is better than E, and D is better than F, there is no question of whether the points on the AB line are better than each other.Efficiency points cannot be graded.Even the endpoints A and B, where one of the parties involved owns everything, are as efficient as any other point on the line AB. Please note: We can never say that any point on the line AB is better than all points in the region OAB. Every point on the AB line is superior only to those within the southwest region.Therefore, point D is superior to all points within the rectangle represented by the dashed line connecting D with a and b. Point D is no better than point E.These points are not graded.However, point G is superior to point E, so all the points on the AB line that belong to the small triangle shaded area with point E at the corner are also superior to point E. On the other hand, if the 45-degree line is viewed as representing the locus of equal distribution (which appears to be a basic interpretation of the axis of interpersonal pairs, an assumption not mentioned in the previous comment), if this As another basis for making a decision, then.All things considered, point D may be preferable to both points C and E.It's much closer to this line.One can even decide that a point in the region, say F, is preferable to the efficient point c.Indeed, on the justice-as-fairness view, principles of justice take precedence over considerations of efficiency, so that, roughly speaking, points in the region representing a just distribution are generally preferable to efficient points representing an unjust distribution .Of course, Figure 4 describes a very simple situation.It is not applicable to basic structures. However, there are many valid structures.For example, an allocation in which one person gets all available goods is efficient because there is no rearrangement that makes some people better off without making anyone worse off.He who possesses all the goods is bound to fail.But, of course, not every distribution is efficient; this may be explained by the large disparity.So long as some distribution makes some people willing to exchange goods with others, it cannot be efficient, because this willingness to exchange indicates the existence of a condition that improves the situation of some people without harming the situation of anyone else. rearrange.In fact, an efficient allocation is one in which it is impossible to find a more favorable exchange.In this sense, the distribution of the good in which one person owns everything is valid, since all others have nothing to exchange for him.Thus, the efficiency principle recognizes that there are many efficient structures.Every efficient arrangement is better than the others, but no efficient arrangement is better than another efficient arrangement. Efficiency principles can now be applied to basic structures through representative human expectations.We can say, for example, that if the rules cannot be changed, the arrangement of rights and duties cannot be redefined so as to raise the expectations of any representative person (at least one person) without lowering the expectations of some other representative person (at least one person). people) expectations, then the arrangement of rights and obligations in the basic structure is valid.Of course, this change must be consistent with other principles.This means that, in changing the basic structure, we must not violate the principle of equal liberty or the regulation of open positions.What can change is the distribution of income and wealth and the way in which organizational power and various other powers manage cooperative activities.The distribution of these primary goods, consistent with restrictions on liberty and access to liberty, can be adjusted to alter the expectations of representative individuals.A basic structural arrangement is efficient if the distribution cannot be changed so as to raise the hopes of some without lowering the hopes of others. I will assume that there are many valid arrangements of the basic structure.Each of these arrangements clearly stipulates the specific distribution of the benefits of social cooperation.The problem is to choose among these arrangements, to find a conception of justice in order to pick out of these valid distributions a distribution that is also just.If we can do this, we also go beyond pure efficiency in a way that is still efficient.Now examine the assumption that as long as social institutions are efficient, there is naturally no reason to worry about distribution.All valid arrangements are declared equally just in this case.Of course, such a scenario might be ludicrous for assigning certain goods to known individuals.No one assumes that whether any one of several happens to own everything is a matter of indifference from the point of view of justice.But this assumption seems equally unreasonable for the basic structure.It may thus be the case that, under certain conditions, serfdom cannot be significantly reformed without lowering the expectations of certain representative persons, such as the landowners, according to whom serfdom was Effective.However, under the same conditions, it may also happen that the system of free labor cannot be changed without lowering the expectations of certain representative reasonable.More generally, so long as a society is properly divided into classes, then we may assume that it is possible at each time to give the widest possible representation to every representative person in the society.This broadest representation at least affords this individual many valid positions, since it is impossible to leave any one of them to raise the expectations of any one representative without lowering the expectations of the other, namely the It sets out the expectations of the broadest representative representative.Thus, each of the broadest representations is valid, but they cannot all be just, or equally just.These arguments are only intended to find for social institutions the same case of distributing certain goods to known individuals, in which case it is valid that one person possesses all of this distribution. These considerations, however, only show what we have always known, namely, that the principle of efficiency alone cannot be a conception of justice.So it has to be supplemented somehow.In the system of natural liberty, the principle of efficiency is constrained by certain background institutions; so long as these constraints are realized, any subsequent efficient distribution is recognized as just.自然自由权制度大致上是以下列方式来选择一种有效分配的：假定我们从经济理论知道，根据规定竞争性市场经济的一般假定，收入和财富将以一种有效的方式进行分配，在任何一段时间产生的某种有效分配都决定于资产的原始分配，就是说，决定于收入和财富以及天生才智和能力的原始分配。随着每一次原始分配，产生了一种明确的有效结果。因此，如果我们承认上述结果不仅是有效的，而且也是正义的，那么我最终将必须承认资产的原始分配始终赖以决定的基础。在自然自由权制度中，这种原始分配受到事业向人才开放（如前面所规定的）这个观念所包含的各种安排的支配，这些安排以自然自由权为背景（由第一个原则明确规定），并以自由市场经济为先决条件。它们要求有一种形式上的机会均等，使所有的人至少拥有得到各种有利的社会地位的同等合法权利。但是，除了为维持必要的背景体制而必须做的以外，并没有为维持社会条件的平等或相似而作出任何努力，所以资产在任何时期的原始分配都要受到自然和社会的意外事故的强烈影响。例如，现有的对收入和财富的分配，就是先前对自然资产——天生才智和能力——的分配的累积结果，因为这些资产或者已经开发，或者尚未实现，而由于社会环境以及诸如意外和幸运这类偶然因素，对于这些资产的利用或者一直得到赞同，或者始终遭到反对。从直觉上说，自然自由权制度的最显而易见的不正义，就是它允许分配份额受到这些因素的不适当的影响，而从道德的观点看，这些因素太没有道理了。 正如我将要谈到的那样，对这两个原则作自由主义的解释，是试图给事业向人才开放的规定再加上公平的机会均等原则这个条件来改变这种情况。这里的思想是：地位不仅在正式的意义上是开放的，而且所有的人都应该有取得这些地位的公平机会。这里的意思一下子看不清楚，但我们可以说，具有相似能力和技艺的人应该具有相似的生活机会。说得更明确点，假定要对自然资产进行分配，那么那些具有同等才智和能力并同样愿意利用这些资产的人，都应该有取得成功的同样前景，而不管他们在社会制度中的初始地位如何，就是说，不管他们出生于什么收入等级。在社会的各个部门，每一个具有相似动机和天赋的人，都应该有大致平等的文化和成功的前景。具有相同能力和抱负的人的期望，不应受到他们的社会阶级的影响。 therefore.对这两个原则作自由主义的解释，是为了极力减少社会意外事故和天生运气对分配份额的影响。为了达到这个目的，必须对社会制度从基本结构上规定进一步的条件。自由市场的安排必须符合政治和法律体制的结构，因为这个结构支配着经济事件的总趋势，并维持着为公平的机会均等所必要的社会条件。这个结构的各个成分是人们所熟知的，虽然去回想一下防止财产和财富的过分积累的重要性，回想一下维持人人受教育的平等机会的重要性，也许是值得的。获得文化知识和技能的机会，不应决定于一个人的阶级地位，因此，学校制度，不管是公立学校还是私立学校制度，都应以打破阶级界线为其目的。 虽然这种自由主义的观念似乎明显地比自然自由权制度更为可取，但从直觉来看，它仍然显得不完全。首先，即使它能完全消除社会偶然因素的影响，它仍然允许财富和收入的分配由能力和才智的自然分配来决定。在背景安排所许可的范围内，分配份额是由自然不测之事的结果决定的；而从道德的角度看，这种结果是毫无道理的。没有理由要让历史和社会命运来决定收入和财富的分配，同样也没有理由要让自然资产的分配来决定收入和财富的分配。此外，公平机会的原则只能不完全地实现，至少在家庭体制存在期间是如此。自然能力发展到什么程度和实现到什么程度，要受到各种社会条件和阶级态度的影响。甚至是努力奋斗的意愿，从而一般说来应该受到奖赏的意愿，其本身也决定于良好的家庭和社会环境。保证具有相似天赋的人在成就和文化方面具有同等的机会，这实际上是不可能的，因此我们可能需要采用一种原则，它既能承认这个事实，又能减轻自然不测之事本身的毫无道理的影响。这种自由主义的观念不能做到这一点，从而促使人们去寻找对这两个正义原则的另一种解释。 在转而讨论民主的平等这个观念之前，我们应该注意一下自然贵族政治观。根据这个观点，除了形式上的机会均等所规定的以外，并无任何要控制社会偶然因素的意思，但是具有更大天赋的人的利益，必须以能促进社会上较不幸部分的善为限。贵族政治的理想适用于一种开放的制度，至少从法律观点来看是如此，而受这一制度之惠的那些人的较优越的地位，只有在居上位者得到较少的利益才使居下位者也可能得到较少利益的情况下，才被看作是正义的。这样，位高任重的思想在自然贵族政治的观念中得到了延续。 因此，自由主义的观念和自然贵族政治观都是不稳定的。一旦我们在决定分配份额时为社会的偶然因素的影响或天然机会的影响而感到苦恼，我们在思考时也不免要为另一种影响而感到苦恼。从道德的观点看，这两种影响似乎是同样毫无道理的。因此，不管我们对自然自由权制度如何敬而远之，但如果没有民主的观念，我们是不会感到满足的。关于这个观点，我仍需予以说明。此外，前面的所有论点都不是赞成这个观念的论据，因为严格说来，根据契约理论，全部论据只有按照原始状态中可能作出的合理选择才能提出。但我这里所关心的是要为对这两个原则作出有利的解释作好准备，以使这两个标准，尤其是第二个标准，不会使读者产生过于古怪或莫名其妙的印象。我曾试图证明，一旦我们努力找到了实现这两个标准的办法，把每一个人都作为一个有道德的人来同等对待，并且不是根据人们的社会命运或在自然不测之事中的运气来决定人们在社会合作中利益和负担的份额，那么用民主的平等来解释，显然就是这四种选择办法中的最佳选择。有了这些评论作为引子，我现在可以转而讨论民主的平等这个概念了。 如图所示，把公平的机会均等原则和差别原则结合起来，就可以作出民主的平等这种解释了。差别原则挑出某种地位，从这一地位判断基本结构中社会和经济安排是否不平等，从而消除效率原则的不确定性。假定体制的结构是由平等的自由和公平的机会均等来规定的，那么地位较优越的人的较高期望，只有在其成为提高地位最不利的社会成员的安排的一部分时才是正义的。这方面的直觉观念是：社会等级不是为了确立和保证境况较好的人的更美好的期望，除非这样做符合命运较差的人的利益（参见下文关于差别原则的讨论）。 差别原则 假定无差异曲线表示被判定为同样正义的分配。那么，差别原则就是一种具有强烈色彩的平等观，除非有一种分配能使两个人的境况更佳（为了简便起见，我们只以两个人为例），否则宁可选择一种平等的分配。只有在这个意义上，差别原则才是一种强烈的平等观。无差异曲线的形状如图5所示。这些曲线实际上是由成直角相交于45度线的垂直线组成的（又一次表现为对一种人际对轴线所作的基本解释）。不管两人中任何一人的地位得到了多大的改善，从差别原则的观点看，除非另一人也得益，否则此人无益可得。 设X1为基本结构中受惠最多的有代表性的人。他的期望提高了。x2，即受惠最少的人的期望也随之而提高。在图6中设曲线OP表示X1的更大期望对X2的期望的差异。原点O表示所有社会基本善平等分配的假定状态。OP曲线始终在45度线之下，因为X1始终境况较好。因此，无差异曲线的唯一有关部分就是这条曲线以下的那些部分，由于这个缘故，图6的左上部分就没有包括进来。显然，只有当OP曲线与无差异曲线的最高点相切时，差别原则才能完全得到实现。在图6中，这个最高点就是a点。 请注意：差异曲线，即曲线OP，假定基本结构规定的社会合作是互利的。对于固定不变的现有的善来说，不存在重新分配的问题。同样，如果对利益进行准确的人际比较是不可能的，那么就什么也没有失去。只要能够认出受惠最少的人并确定他的合理选择机会，这就够了。 有一种比差别原则较少平等主义色彩，而且也许初看起来似乎比较有道理的观点。按照这种观点，如图7所示，代表正义的分配（或者代表所有考虑到的情况的分配）的无差异曲线就是凸向原点的平滑曲线。代表社会福利函数的无差异曲线。通常用这种方式来表示。这种曲线的形状表明，当两人中的一人相对于另一人来说已经获益，从社会角度看，则此人所得之更多的利益对他来说就变得只有较少的价值。 另一方面，古典的功利主义者对于如何来分配恒量利益问题是漠不关心的。他们只是为了打破僵局才求助于平等。如果只有两个人，那么在对人际对轴线作出基本解释时，功利主义者的代表分配的无差异曲线就成了垂直于45度线的直线。然而，既然X1和X2是有代表性的人，那么他们所得到的利益必须按照他们各自所代表的人数来衡量。既然X2所代表的人数可能大于X1所代表的人数，那么这种无差异曲线就如图8所示变得较平。地位有利者的人数与地位不利者的人数之比，规定了这些直线的斜度。如前图画出同样的差异曲线，则可看出，按照功利主义观点，在b点外OP曲线所达到的极点即为最佳分配。由于差别原则选择了b点，而b点又始终处于a点的左方，所以在其他条件相等的情况下，功利主义允许较大的不平等。 为了说明差别原则，可以考虑一下社会各阶级的收入分配。让我们假定，不同的收入集团是与有代表性的个人相互关联的。参照这些人的期望，我们就能对分配作出判断。假定那些开始在财产占有的民主制社会里是企业阶级成员的人，比在不熟练工人阶级中开始生活的人有着更美好的前景。即使在消除了现有的社会不正义之后，这种情况看来仍然可能是真实的。那么，怎样才能证明生活前景中这种原始的不平等是正当的呢？按照差别原则，只有在期望中的差异有利于境况较差的有代表性的人，这里也就是有代表性的不熟练工人的时候，这种不平等才可以证明是正当的。只有在减少这种不平等会使工人阶级境遇更糟时，期望中的不平等才是可以允许的。考虑到关于地位开放的第二个原则的附加条款以及一般的自由权原则，企业家们可以有的较大期望大概会鼓励他们去做一些可以改善劳动阶级长远前景的事情。他们的较美好的前景起了刺激作用，使经济过程更加有效，创造发明也会以更快的速度进行，等等。最后，由此而产生的物质利益扩展到整个制度，使地位最不利的人也能得益。我不打算考虑这些情况究竟有多大真实性。重要的是，如果按照差别原则，这些不平等是主义的，那么对这类问题必须加以论证。 关于这个原则，我现在要发表几点看法。首先，在应用这个原则时，必须区别两种情况。第一种情况是：地位最不利的人的期望确实被提高到最大限度（当然也要受到上述限制）。境况较好的人的期望的任何改变，都不能改善境况较差的人的地位。于是，最佳安排得到了公认，我将把这称之为完全正义的安排。第二种情况是：所有境况较好的人的期望，至少对较不幸的人的福利产生了影响。就是说，如果他们的期望被降低了，那么，地位最不利的人的期望也可能降低。这时还仍然没有达到最大值。甚至地位较有利的人的较高期望，也可能会提高地位最低的人的期望。我将认为，这种安排是完全正义的，但却不是最佳的正义安排。如果较高期望（不论是一种期望还是多种期望）是过分的，那么安排就是不正义的。如果这些期望被降低了，那么受惠最少者的地位就可能会得到改善。一种安排不正义到什么程度，要看这种较高期望过分到什么程度，以及它们对违反其他正义原则，如公平的机会均等原则，依赖到什么程度；但我本打算用任何精确的办法去衡量不正义的程度。这里需要指出的是，虽然差别原则严格说来是一种最大值原则，但在上述尚不是最佳安排的两种情况之间存在着重大的差别。社会应该努力避免出现这样的领域：在那里，境况较好的人的边际贡献是负的，因为，在其他条件相等时。这似乎比虽未达到最佳安排但边际贡献是正的这种情况更糟。富人与穷人之间的甚至更大的差别，使穷人的境况变得更糟，这不但违反了民主的平等，而且也违反了互利的原则（第17节〕。 还有一个问题是：我们已经知道，自然自由权制度和自由主义的观念试图超越效率原则，缩小这个原则的作用范围，用某些背景体制来限制它，并把其余的事交给纯粹的程序正义去做。这种民主观认为，虽然至少在某种程度上可以求助于纯粹的程序正义，但是前此的种种解释在这样做时仍然使很多情况决定于社会和自然的偶然因素。但应该指出，差别原则和效率原则是一致的。如果差别原则得到充分的实现，那么，使任何一个有代表性的人的境况变得更好，而又不使另一个人，即我们打算提高其期望的地位最不利的有代表性的人的境况变得更糟，这事实上是不可能的。因此，对正义的规定应能使正义符合效率，至少在这两个原则得到完全实现时应该如此。当然，如果基本结构是不正义的，这两个原则将会允许作出改变，以便降低某些境况较好的人的期望，因此，如果效率原则是用来表示只有改善每一个人的前景的改变才是可允许的，那么民主的概念就与效率原则是不一致的。正义优先于效率，因此它要求作出某些改变，而这些改变在这个意义上是无效的。一种完全正义的安排也应是有效的设计，也只有在这个意义上，才谈得上一致。 其次，我们可以考虑一下关于差别原则含义的某些复杂情况。如果这个原则实现了，每个人就都得到了利益，这一点一直被认为是理所当然的。认定这一事实是由于人们明显地感觉到，和原始的平等安排相比较，每一个人的地位改善了。但是，事情显然并不取决于能否确定这种原始安排；事实上，原始状态中的人的境况究竟好到什么程度，这一点对运用差别原则并不起关键作用。我们只是按照规定的限制，最大限度地提高地位最不利的人的期望。只要像我们所假定的那样，这样做对每个人都有好处，那么来自这种假设的平等地位的估计利益就是毫不相干的，虽然不是根本不可能确定这种利益。然而，可能还有一种感觉，即如果差别原则实现了（至少如果我们作某些自然的假定），那么每一个人也就都得到了利益。让我们假定，期望的不平等是一种连锁关系，就是说，如果一种有利的地位对提高地位最低的人的期望产生了影响，那么它也提高了地位介乎两者之间的所有人的期望。例如，如果企业家的较大期望使不熟练工人得到了好处，那么它们也使半熟练工人得到了好处。请注意：这种连锁关系没有提到地位最不利的人没有得到利益这一情况，所以它并不意味着全面的影响。进一步假定各种期望是紧密结合的，就是说，提高或降低任何一个有代表性的人的期望，而又不提高或降低其他每一个有代表性的人的期望，尤其是地位最不利的人的期望，这是不可能的。各种期望紧密结合，司以说是无法拆开的。不过，由于有了这些假定，人们就会觉得，如果差别原则实现了，每一个人就都得到了好处。在任何双向比较中，境况较好的有代表性的人由于给予他的有利条件而获得了利益，而境况较差的人也由于这些不平等所产生的差益而获得了利益。当然，这些情况并非始终如此。但在此情况下，境况较好的人不应拥有对受惠最少的人可以得到的利益的否决权。我们仍然应该最大限度地提高地位最不利的人的期望（参见关于连锁关系的附加讨论）。 连锁关系 为了简明起见，假定有三个有代表性的人。设x1为受惠最多者，X3为受惠最少者，X2介乎两者之间。沿水平轴线划出X1的期望，沿垂直轴线划出X1和X2的期望。表示受惠最多者对其他集团的差异的曲线，从代表假想的平等地位的原点开始。此外，假定受惠最多者可以得到最大限度的利益，其根据是：即使差别原则可以承认这一点，自由权优先的观念也会拒绝考虑对政治制度等等所产生的不正义的影响。 差别原则所选定的点就是代表X3的曲线所达到的最高点， 如图9中的a点。 所谓连锁关系是指：代表X3的曲线向右方上升所达到的任何一点，代表X2的曲线也能达到，如图9中a点和图10中b点左边的区间所示。连锁关系没有提到的一个情况是，代表X3的曲线向右边下降，如图9中a点右边的区间所示。代表X2的曲线可能上升，也可能下降（如虚线X2所示）。连锁关系对图10中b点的右方是不适用的。 代表X2和X3的曲线上升的区间，就是代表地位差异的区间。向右边的任何上升，都会提高平均期望（如果功利是用期望来测量的，即为平均功利），同时也实现了作为一种衡量变化的标准的效率原则，就是说，右边的各个点改善了每一个人的地位。 在图9中，平均期望的上升可能超过a点，虽然受惠最少者的期望下降了（这一情况决定于各个集团的影响）。差别原则排除了这一情况而选定了a点。 所谓紧密结合是指：代表X2和X3的曲线都不是水平延伸的。在每一点上，这两条曲线或者上升，或者下降。如此表示的曲线都是紧密结合的。 我不打算去研究连锁关系和紧密结合可能在多大程度上是适用的。差别原则并不决定于这些关系是否实现。然而，人们可能注意到，如果受惠较多者的地位所产生的差异普遍地扩展到整个社会，而不是限于社会的某些部门，那么，如果地位最不利的人得到利益，处于中间地位的人也会得到利益，这看来是可能的。此外，基本结构所体现的体制的两个特征，也有助于利益的普遍扩散。第一个特征是：建立体制的目的是为了促进人人共有的某些基本利益；第二个特征是：职务和地位是向所有人开放的。因此，如果立法者和法官的特权和权力改善了受惠较少者的地位，他们也就改善了一般公民的地位，这一点看来是可能的。倘若其他正义原则得到了实现，连锁关系就可能经常适用。如果是这样，那么我就可以说，在积极差异区域（所有处于有利地位的人的利益提高了最不幸的人的期望的区域），为实现完全正义的安排而进行的任何活动，不但增进了平均福利，而且也提高了每一个人的期望。考虑到这些附加的假定，差别原则就具有了同平均功利及效率原则同样的实际效果。当然，如果连锁关系几乎是不适用的，而上述情况又是无关紧要的，那么这两种原则的一致反而显得令人奇怪了。但我们往往假定，在正义的社会安排内，像利益的普遍扩散这种事情确实发生了，至少从较长远的观点看是会发生的。如果情况确实如此，那么这些论点就是表示，差别原则是怎样来说明这些作为特例的人们较为熟悉的观念。不过，仍然需要指出的是：从道德的角度看，这个原则是更基本的原则。 另外还有一种复杂情况。假定紧密结合这种情况，是为了简化对差别原则的说明。不管在实际上有多少可能性或有多大重要性，显然可以想象的是，地位最不利的人并不因为境况最佳的人的期望的某些变化而受到某种影响，尽管这些变化使别人受益。在这种情况下，紧密结合这种假定就不适用了，而为了对这种情况也能适用，我们可以把一种更普遍的原则表述如下：在有n个有关代表的基本结构中，首先，最大限度地提高境况最差的有代表性的人的福利；其次，为了境况最差的代表的平等福利，最大限度地提高第二个境况最差的有代表性的人的福利，因此类推，直到最后为了前面n－1个代表的平等福利，最大限度地提高境况最佳的有代表性的人的福利。我们可以把这个情况看作是词典编纂式的差别原则。然而，我们将经常使用形式比较简单的差别原则，因此，作为前面这几节的讨论结果，第二个正义原则的内容如下： 对社会和经济不平等的安排，应能使这种不平等既（1）符合地位最不利的人的最大利益，又（2）按照公平的机会均等的条件，使之与向所有人开放地位与职务联系在一起． 最后，应该指出可以很容易地使差别原则或这一原则所表明的思想与普遍正义观相适应。事实上，普遍正义观只是适用于包含图由权与机会在内的所有的社会基本善，从而不再受这个特殊观念的其他组成部分限制的差别原则。从前面对正义原则的简短讨论来看，这是显而易见的。正如我将不时指出的那样，随着社会状况的改善，序列中的这些原则也就是普遍正义观最后采取的形式。这个问题是和我将在下文（第39节和第82节）予以讨论的自由权优先问题联系在一起的。某种形式的差别原则自始至终都是基本的原则，目前只要说这么一点就够了。