We have asked about how the justification of the rule of law is related to the justification of punishment.  Our weak conclusion is a surprising one given the hint that led us to the issue.  We thought that the purpose of law was tied to that of punishment.  Law and punishment had to be justified together (as Confucius had argued). What we conclude reverses the order of justification. We don’t know if punishment is ever justified, but if it is, then it is only when combined with the rule of law. 

Even if punishment is never justified, we can still justify law on the empirical assumption that powerful people have a tendency to use their monopoly on coercive force to “get their way.” We can also not the empirical tendency of ordinary people to vengeance for perceived wrongs.  In the absence of genetic strategies to remove these two human tendencies (which does not seem probable in the foreseeable future) we can mount a justification of the rule of law. The rule of law serves to rationalize, control, and limit both tendencies.  Rationalizing them makes them more predictable so the violence they involve becomes avoidable. Rational beings can choose courses of action (daos or ways) that minimize their exposure to such violence.

Is there a stronger or deeper justification of law?  Any justification will have to be from some point of view other than legal--since it is a legal point of view we are trying to justify.  The weak justification we have offered is from a prudential (self-interested) point of view.  Any more interesting justifications would be from a moral point of view.  We could use a moral theory of desert, (that is, the convict deserves the punishment) but as we have seen, retributive or desert morality is still controversial and difficult. (Desert intuitions are one of the interesting contrasts between Western and Chinese morality. Chinese moral reasoning is much more consequential.  That, in part, explains Confucius’ resistance to the idea of law.)

A natural line to pursue for a moral justification of the rule of law is the one Fuller hints at -- justice. One sense of "justice" is merely another word for "desert" or "retribution." (E. g., when we speak of criminal justice or justice to the victim.)  However, we are interested in the other sense--where justice has to do with equity and fairness.  (Morris tried but failed to get retributive conclusions from equity intuitions.  We are asking if we can get rule of law conclusion from equity or fairness considerations.) Brian Barry draws attention back to Fuller’s implicit notion of procedural fairness.  He asks us to consider the wide concept of a social decision procedure--a way of making social decisions. We can then focus on whether a social decision procedure is fair or not.  Then we can ask, in particular, if the social decision procedure we call the rule of law is a fair one.

We have discussed some of these social decision procedures before. Combat is one example.  It is widely used in other species and has been used throughout human history. The winner gets her way! Discussion on the merits is also a social decision procedure. It is used prominently in science and in morality.  Bargaining or negotiating are used in commercial and business dealings as well as in government and diplomacy.  Voting, chance (lotteries), and contests are other examples of social decision procedures.  Law belongs to the social decision type known as authoritative determination.  Some person or body (judge or jury) decides performatively who is right and settles the dispute.

Barry makes several points about how the notion of fairness may be applied to social decision procedures:

Barry argues that procedures may be fair or unfair (1) in general and (2) may be fair or unfair in particular cases. For example, bargaining is generally fair, but when you have three times as much money as I do, it is less fair than when we have the same amount.  Combat is an unfair procedure when one party is large and powerful and the other small, aged and weak or when one has weapons or skills that the other lacks.  Chance is unfair when the dice is weighted.  The rule of law is less fair when skilled lawyers charge money and some people have much more than others and so forth.

We may also ask if a procedure is fairly applied.  Did the participants follow all the rules constituting the procedure?  For example, in a duel, did both parties walk ten paces counting before turning to shoot?  In an election, was the vote counting rigged? Did someone bribe or threaten the judge or jury?  A fair decision procedure that is fair to apply in this particular case, may still be improperly (unfairly) executed in a particular case.  Someone breaks the implicit rules of the procedure.

One crucial aspect of fairness is that if you accept a decision procedure and use it when it benefits you, then it is only fair to adhere to it and accept its results when they are not to your advantage.  This is one of the reasons that we look for decision procedures that are not only fair in particular cases (combat between equally armed, strong and skillful combatants.)

The suggestion is that legal decision (authoritative determination following competitive advocacy and following rules of evidence procedure and etc.) is a fair or just decision procedure. It may be unfairly employed.  (That is, it may lack both background or procedural fairness in particular circumstances.)  Having some decision procedure is warranted, as Gray had argued, because we do want means to settle our disputes. Further, as Fuller argued, we want a morally fair or just procedure and we will think the procedure is justified only if it does tend to produce what we consider substantive justice. 

This means that we have to distinguish between two conceptions of justice.  One, which Barry is analyzing, is formal or procedural justice.  The basic formal just outcome is one that treats like cases alike and different cases dif­ferently.  This is a weak condition unless we presuppose some way to specify what counts as a relevant likeness and difference.  But intuitively, it seems right.  If we commit the same crime but mine is on a Tuesday and yours is on a Wednesday, we would not normally regard that as a reason for a different sentence. (We might think that the same crime on Christmas is more serious than on a date less “joyous.”)

Barry’s arguments presuppose the now familiar crucial appeal to respect for people as rational agents.  People are animals who make deliberate choices after reflection and planning at least much of the time.  We should treat people as capable of adopting projects, planning, making rational choice of under­takings etc. The idea is that people of such a description can plan so as to avoid punishment consequences if she chooses to. Barry notes an important consequence of this way of justifying the social decision procedure.  It only allows you to limit the class of relevant differences we worried about to those properties the agent can do something about. He argues that this entails that an implicit injunction against certain kinds of discrimination—against woman, the handicapped, racial differences, and perhaps homosexuality is built into the rule of law. It also explains our repugnance at ex post facto laws.  It's too late to do anything about your behavior after the fact.

The value at issue here can also come from a utilitarian argument as well as from some Kantian respect for rational decision-making.  Having the rule of law, by reducing unfulfilled expectations and improving predictability makes for a more efficient, productive society.  The difference is whether we focus on our respect for such beings or on the consequential importance of efficient use of human rational planning capacities. Either justifies the rule of law.

 What about a substantive just outcome?  What do we mean by that?  Basically a just decision is one that gets the right answer--convicts the right person, gives the just sentence, distributes the costs equitably, etc.  A voting procedure is just in this substantive sense if it produces the best government.  A procedurally fair vote with candidates with equal starting points and advantages does not necessarily have this result. The weaker team may win the contest, say by accident or luck. We tend to approve of a decision procedures more when we think it tends in general to yield the right answer. If we think its typically less reliable than another procedure, we may prefer the alternative. 

The majoritarian fallacy in justifying democracy is making the assumption that the fair popular election of someone proves she was the right candidate. It seldom does. The justification for democracy might be on purely procedural grounds – it’s a fair procedure whether or not it gets the best government more often than the alternatives. Most democrats probably think that it does so at least as often as the alternatives (heredity, appointment, longevity, etc.) They need not, however, think that it guarantees the correct outcome. 

The same goes for the judge’s authoritative decision in the rule of law.  As we saw earlier, we can make sense of the judge’s being wrong on the law and still counting it as a valid outcome.  Agreeing to use the system does not mean agreeing with all particular outcomes.  It does means abiding by them, although we may think them mistakes.  If mistakes become common or we have evidence of political domination of the process, however, we will lose respect for the decision procedure (as we may, say, in Singapore or Malaysia).

Barry merely starts us thinking about a more detailed account of the substantive concept of justice is. Here again he draws on the two contrasting ethical outlooks: utilitarian (which he calls aggregate) and deontological (egalitarian or distribu­tive). 

Rawls provided a richly detailed theory of substantive justice and with it became the central figure in modern ethics. He adapts his argument partly from Thomas Hobbes and partly from Immanuel Kant. Hobbes pioneered the social contract theory of politics and Kant's 3rd categorical imperative envisioned a "kingdom of legislators." who collectively “willed” the moral law.  The article I have asked you to read is an influen­tial earlier version of the key argument in Rawls’ book.

 The basic idea was to give a normative theory of justice--to state the highest principles of justice.  Justice, Rawls notes, is only part of morality.  It is that aspect of morality that evaluates our social arrangements and institutions (our practices). The concept of a social arrangement is of an activity conducted according to rules.  Here we are considering the practice of criticizing or justifying institutions.

As Rawls formulates the question, we should ask ourselves what considerations or ideals are we justified in using to praise or criticize those social practices.  He suggests that the principles we should appeal to are those that we would un­animously choose under certain constraints.  He calls these constraints the original position. Rawls appeals to Barry's fairness principle. The considerations we appeal to when we criticize institutions must be such that we would accept them as valid when others use them to criticize institutions that benefit us. The Kantian element is found in the assumption that the justification must be one the other person could not reasonably reject. Hence the agreement in the original position on the principles must be unanimous.

This feature of Rawls' theory reflects a common feature of moral theories.  They are implicitly universal.  The conventional models of this universality include Confucius' and Christ's golden rule, slogans like “put yourself in others shoes,” and the notion of an ideal observer or God as the source of moral judgments.  The point is that we distinguish a moral judgment from others, such as prudence and mores, by its impartiality or its neutrality.  It takes everyone's interests and perspectives equally into account.

Rawls achieves this neutrality through convention in an imagined pre-existence --the original position.  Imagine if everyone who ever lives gets together before humans evolve on earth and tries to work our a “logical constitution” – the principle by which we can criticize social institutions.  He treats the principles of justice as those chosen freely and rationally by individuals who do not yet have a justice concept.

The conditions of choice concern mainly the knowledge the participants in the convention have at their disposal when they meet.  Some kinds of knowledge are denied to them, for example, any knowledge that would allows you to skew the principles in your favor. For example, if I know I will be a male, I might prefer a principle that gives preference to male judgments about justice.  Ditto for races, classes, wealth, intelligence, size etc.  Therefore, Rawls imagines a veil of ignorance about just who you will be in the society you are designing.  You have to select principles of justice without knowing if you are going to be male or female, rich or poor, smart or dumb, a member of an outcast class or a privileged class. Rawls sometimes characterizes the veil of ignorance as assuming that our enemy assigns you your position in the subsequent real world.  You make a free rational choice about the basic structure of society on the assumption that you might be the least well off in that society.

On the other hand, he allows us to know a great deal about general things. Ideally, we should have correct social science information in general.  We do not want our principles of justice to be blind to economic, psychological or sociological laws. It would be important to our decision to know the psychological or deterrent effect of punishment. Nor would we want to be ignorant of the natural sciences such as nutrition, physics and so forth.  One way to put Rawls’ two conditions of knowledge is that we have all general knowledge and no indexical knowledge (knowledge with an “I” in it).

Rawls outlines a number of other conditions called the circumstances of justice.  These include that:

We are rational;

We know our own interests;

We can plan and trace consequences, adhere to course of action against momentary temptation;

We have enough similarity of needs and interests to justify cooperation;

There is a rough equality of power and intelligence;

There is moderate scarcity in the goods we all desire (disagreements will therefore arise. We would not, he thinks need justice if we lived in paradise).

Rawls argues that we must make these and other minimal but important assumptions about human nature.  Certain features make us appropriate subjects for a conception of justice. One is that adopting and following a just system will tend to strengthen our sense of justice and our willingness to continue following it. Justice should be psychologically reinforcing – it should be a chang^constant dao^way.

 Under these complicated circumstances, Rawls argues, we would adopt two principles.  They are the principles of (1) the equal maximal liberty principle and (2) the difference principle (a distribution principle).

1. Each person should have as much liberty as possible, consistent with everyone else having the same liberty. 

2. Goods should be distributed equally except where an unequal distribution benefits everyone (or the least well off group). 

Much of the debate about Rawls’ theory concerns the second principle.  Our interest here is in the first one. Rawls argues that legal institutions are justified mainly by the first principle. The justification we have offered so far can be restated in liberty terms. The rule of law gives individuals an equal liberty or freedom from arbitrary coercion or punishment. Before looking at that derivation, let us see why Rawls thinks we would adopt the first principle.

Rawls argues that no matter what place we eventually take in society, whatever our interests or status, we would be better off in a system that allowed us to pursue those interests insofar as they did not interfere with others similar pursuits.  A society could, of course, favor one group's interests (say Christians or Buddhists' interests) and compel the rest of us to convert.  Why not allow that possibility? (Suppose you knew that a majority of people would be Buddhist and reasoning probabilistically, decide that you might be among the majority.  Wouldn’t that justify you in restraining liberty to get to live in a community that reinforced and participated with you in your spiritual interests?)

Rawls argues that it would be irrational to choose on the basis of probability in the original position. That is why he asks us to think of our enemy assigning us our eventual positions in society. He justified his analysis using game theory.  Consider rational gaming or gambling strategies.  We may pursue three strategies:

Maxi-max strategies (like betting on the lottery--make the win as large as possible if you do win). 

Maxi-average strategies strive for the best average expected outcome over the long run (accepting some losses and counting in some big wins).

Maxi-min strategies are defensive, conservative strategies.  We play not to lose.  We adopt the strategy that makes the worst outcome as good as it can be.

Rawls argues that maxi-min is the only rational strategy (1) when the bet is a one time only bet and (2) where literally everything (your whole life) is at stake.  The more common max-average strategy is rational strategy in other circumstances, e.g., when you are going to bet many times and with only a portion of your goods. It is a rational principle for someone who visits the racetrack or Macao or plays the HK stock or property market. But it is less rational when the bet is for your family’s entire life savings and you get only one roll of the dice!  Thus, he says, in the original position deciding on the principles of justice, the rational strategy is maxi-min.  We would rationally seek to ensure the best outcome for the least well off segment of the society. In the case of a Buddhist society, we would want to protect the religious minority and allow them their minority religious practice even on the small chance that we might belong to that religion.

Rawls launches a furious attack on utilitarianism as being based on the max-average principle.  The main focus of his analysis here is in the distribution principle where inequalities could be justified as long as the average went up. The contrast applies to the first principle as well. As we saw, utilitarianism may justify Maoist study sessions rather than the rule of law.  (It’s important to note that, given human psychology, economic laws, sociology etc. it may not too.)  Rawls want not to have to rely on such empirical matters.  Such socially directed, educational techniques designed to alter our individual sense of value and the good life are inimical to our dignity, our liberty and our freedom of choice.