A theory of Justice: Revised Edition

Games and Decisions
(New York, John
Wiley and Sons, 1957), pp. 278–306.
392
Goodness as Rationality

---

are not forthcoming. It is only in the thin theory of the good that we have
to worry about these matters. Here the notion of rationality must be
interpreted so that the general desire for the primary goods can be estab-
lished and the choice of the principles of justice demonstrated. But even
in this case, I have suggested that the conception of justice adopted is
insensitive with respect to conflicting interpretations of rationality. But in
any event, once the principles of justice are chosen, and we are working
within the full theory, there is no need to set up the account of the good so
as to force unanimity on all the standards of the rational choice. In fact, it
would contradict the freedom of choice that justice as fairness assures to
individuals and groups within the framework of just institutions.
A second contrast between the right and the good is that it is, in
general, a good thing that individuals’ conceptions of their good should
differ in significant ways, whereas this is not so for conceptions of right.
In a well-ordered society citizens hold the same principles of right and
they try to reach the same judgment in particular cases. These principles
are to establish a final ordering among the conflicting claims that persons
make upon one another and it is essential that this ordering be identifiable
from everyone’s point of view, however difficult it may be in practice for
everyone to accept it. On the other hand, individuals find their good in
different ways, and many things may be good for one person that would
not be good for another. Moreover, there is no urgency to reach a publicly
accepted judgment as to what is the good of particular individuals. The
reasons that make such an agreement necessary in questions of justice do
not obtain for judgments of value. Even when we take up another’s point
of view and attempt to estimate what would be to his advantage, we do so
as an adviser, so to speak. We try to put ourselves in the other’s place, and
imagining that we have his aims and wants, we attempt to see things from
his standpoint. Cases of paternalism aside, our judgment is offered when
it is asked for, but there is no conflict of right if our advice is disputed and
our opinion is not acted upon.
In a well-ordered society, then, the plans of life of individuals are
different in the sense that these plans give prominence to different aims,
and persons are left free to determine their good, the views of others
being counted as merely advisory. Now this variety in conceptions of the
good is itself a good thing, that is, it is rational for members of a well-or-
dered society to want their plans to be different. The reasons for this are
obvious. Human beings have various talents and abilities the totality of
which is unrealizable by any one person or group of persons. Thus we not
only benefit from the complementary nature of our developed inclinations
393
68. The Right and the Good Contrasted

---

but we take pleasure in one another’s activities. It is as if others were
bringing forth a part of ourselves that we have not been able to cultivate.
We have had to devote ourselves to other things, to only a small part of
what we might have done (§79). But the situation is quite otherwise with
justice: here we require not only common principles but sufficiently simi-
lar ways of applying them in particular cases so that a final ordering of
conflicting claims can be defined. Judgments of justice are advisory only
in special circumstances.
The third difference is that many applications of the principles of
justice are restricted by the veil of ignorance, whereas evaluations of a
person’s good may rely upon a full knowledge of the facts. Thus, as we
have seen, not only must the principles of justice be chosen in the absence
of certain kinds of particular information, but when these principles are
used in designing constitutions and basic social arrangements, and in
deciding between laws and policies, we are subject to similar although
not as strict limitations. The delegates to a constitutional convention, and
ideal legislators and voters, are also required to take up a point of view in
which they know only the appropriate general facts. An individual’s con-
ception of his good, on the other hand, is to be adjusted from the start to
his particular situation. A rational plan of life takes into account our
special abilities, interests, and circumstances, and therefore it quite prop-
erly depends upon our social position and natural assets. There is no
objection to fitting rational plans to these contingencies, since the princi-
ples of justice have already been chosen and constrain the content of
these plans, the ends that they encourage and the means that they use. But
in judgments of justice, it is only at the judicial and administrative stage
that all restrictions on information are dropped, and particular cases are to
be decided in view of all the relevant facts.
In the light of these contrasts we may further clarify an important
difference between the contract doctrine and utilitarianism. Since the
principle of utility is to maximize the good understood as the satisfaction
of rational desire, we are to take as given existing preferences and the
possibilities of their continuation into the future, and then to strive for the
greatest net balance of satisfaction. But as we have seen, the determina-
tion of rational plans is indeterminate in important ways (§64). The more
evident and easily applied principles of rational choice do not specify the
best plan; a great deal remains to be decided. This indeterminacy is no
difficulty for justice as fairness, since the details of plans do not affect in
any way what is right or just. Our way of life, whatever our particular
circumstances, must always conform to the principles of justice that are
394
Goodness as Rationality

---

arrived at independently. Thus the arbitrary features of plans of life do not
affect these principles, or how the basic structure is to be arranged. The
indeterminacy in the notion of rationality does not translate itself into
legitimate claims that men can impose on one another. The priority of the
right prevents this.
The utilitarian, on the other hand, must concede the theoretical possi-
bility that configurations of preferences allowed by this indeterminacy
may lead to injustice as ordinarily understood. For example, assume that
the larger part of society has an abhorrence for certain religious or sexual
practices, and regards them as an abomination. This feeling is so intense
that it is not enough that these practices be kept from the public view; the
very thought that these things are going on is enough to arouse the
majority to anger and hatred. Even when these attitudes are unsupport-
able on moral grounds, there appears to be no sure way to exclude them
as irrational. Seeking the greatest satisfaction of desire may, then, justify
harsh repressive measures against actions that cause no social injury, To
defend individual liberty in this case the utilitarian has to show that given
the circumstances the real balance of advantages in the long run still lies
on the side of freedom; and this argument may or may not be successful.
In justice as fairness, however, this problem never arises. The intense
convictions of the majority, if they are indeed mere preferences without
any foundation in the principles of justice antecedently established, have
no weight to begin with. The satisfaction of these feelings has no value
that can be put in the scales against the claims of equal liberty. To have a
complaint against the conduct and belief of others we must show that
their actions injure us, or that the institutions that authorize what they do
treat us unjustly. And this means that we must appeal to the principles that
we would acknowledge in the original position. Against these principles
neither the intensity of feeling nor its being shared by the majority counts
for anything. On the contract view, then, the grounds of liberty are com-
pletely separate from existing preferences. Indeed, we may think of the
principles of justice as an agreement not to take into account certain
feelings when assessing the conduct of others. As I noted before (§50),
these points are familiar elements of the classical liberal doctrine. I have
mentioned them again in order to show that the indeterminacy in the full
theory of the good is no cause for objection. It may leave a person
unsettled as to what to do, since it cannot provide him with instructions as
to how to decide. But since the aim of justice is not to maximize the
fulfillment of rational plans, the content of justice is not in any way
affected. Of course, it cannot be denied that prevailing social attitudes tie
395
68. The Right and the Good Contrasted

---

the statesman’s hands. The convictions and passions of the majority may
make liberty impossible to maintain. But bowing to these practical neces-
sities is a different thing from accepting the justification that if these
feelings are strong enough and outweigh in intensity any feelings that
might replace them, they should carry the decision. By contrast, the
contract view requires that we move toward just institutions as speedily as
the circumstances permit irrespective of existing sentiments. A definite
scheme of ideal institutions is embedded in its principles of justice (§41).
It is evident from these contrasts that in justice as fairness the concepts
of the right and the good have markedly distinct features. These differ-
ences arise from the structure of contract theory and the priority of right
and justice that results. I do not suggest, however, that the terms “right”
and “good” (and their relatives) are normally used in ways that reflect
these distinctions. Although our ordinary speech may tend to support the
account of these concepts, this correspondence is not needed for the
correctness of the contract doctrine. Rather, two things suffice. First,
there is a way of mapping our considered judgments into the theory of
justice such that in reflective equilibrium the counterparts of these con-
victions turn out to be true, to express judgments that we can accept. And
second, once we understand the theory, we can acknowledge these inter-
pretations as suitable renderings of what on reflection we now wish to
maintain. Even though we would not normally use these replacements,
perhaps because they are too cumbersome, or would be misunderstood, or
whatever, we are prepared to grant that they cover in substance all that
wants to be said. Certainly these substitutes may not mean the same as
the ordinary judgments with which they are paired. How far this is the
case is a question that I shall not examine. Moreover, the replacements
may indicate a shift more or less drastic from our initial moral judgments
as they existed prior to philosophical reflection. Some changes anyway
are bound to have taken place as philosophical criticism and construc-
tion lead us to revise and extend our views. But what counts is whether
the conception of justice as fairness, better than any other theory pres-
ently known to us, turns out to lead to true interpretations of our consid-
ered judgments, and provides a mode of expression for what we want
to affirm.
396
Goodness as Rationality

---

CHAPTER VIII. THE SENSE OF JUSTICE
Having presented an account of the good, I now turn to the problem of
stability. I shall treat it in two stages. In this chapter I discuss the acquisi-
tion of the sense of justice by the members of a well-ordered society, and
I consider briefly the relative strength of this sentiment when defined by
different moral conceptions. The final chapter examines the question of
congruence, that is, whether the sense of justice coheres with the concep-
tion of our good so that both work together to uphold a just scheme. It is
well to keep in mind that much of this chapter is preparation and that
various topics are touched upon only to indicate the more basic points
that are relevant for the philosophical theory. I begin with a definition of a
well-ordered society and with some brief remarks about the meaning of
stability. Then I sketch the development of the sense of justice as it pre-
sumably would take place once just institutions are firmly established and
recognized to be just. The principles of moral psychology also receive
some discussion; I emphasize the fact that they are reciprocity principles
and connect this with the question of relative stability. The chapter con-
cludes with an examination of the natural attributes in virtue of which
human beings are owed the guarantees of equal justice, and which define
the natural basis of equality.
69. THE CONCEPT OF A WELL-ORDERED SOCIETY
69. A Well-Ordered Society
At the beginning (§1) I characterized a well-ordered society as one de-
signed to advance the good of its members and effectively regulated by a
public conception of justice. Thus it is a society in which everyone ac-
cepts and knows that the others accept the same principles of justice, and
the basic social institutions satisfy and are known to satisfy these princi-
ples. Now justice as fairness is framed to accord with this idea of society.
The persons in the original position are to assume that the principles
397

---

chosen are public, and so they must assess conceptions of justice in view
of their probable effects as the generally recognized standards (§23).
Conceptions that might work out well enough if understood and followed
by a few or even by all, so long as this fact were not widely known, are
excluded by the publicity condition. We should also note that since prin-
ciples are consented to in the light of true general beliefs about men and
their place in society, the conception of justice adopted is acceptable on
the basis of these facts. There is no necessity to invoke theological or
metaphysical doctrines to support its principles, nor to imagine another
world that compensates for and corrects the inequalities which the two
principles permit in this one. Conceptions of justice must be justified by
the conditions of our life as we know it or not at all.
1
Now a well-ordered society is also regulated by its public conception
of justice. This fact implies that its members have a strong and normally
effective desire to act as the principles of justice require. Since a well-or-
dered society endures over time, its conception of justice is presumably
stable: that is, when institutions are just (as defined by this conception),
those taking part in these arrangements acquire the corresponding sense
of justice and desire to do their part in maintaining them. One conception
of justice is more stable than another if the sense of justice that it tends to
generate is stronger and more likely to override disruptive inclinations
and if the institutions it allows foster weaker impulses and temptations to
act unjustly. The stability of a conception depends upon a balance of
motives: the sense of justice that it cultivates and the aims that it encour-
ages must normally win out against propensities toward injustice. To
estimate the stability of a conception of justice (and the well-ordered
society that it defines), one must examine the relative strength of these
opposing tendencies.
It is evident that stability is a desirable feature of moral conceptions.
Other things equal, the persons in the original position will adopt the
more stable scheme of principles. However attractive a conception of
justice might be on other grounds, it is seriously defective if the princi-
ples of moral psychology are such that it fails to engender in human
beings the requisite desire to act upon it. Thus in arguing further for the
principles of justice as fairness, I should like to show that this conception
is more stable than other alternatives. This argument from stability is for
the most part in addition to the reasons so far adduced (except for consid-
1. It follows that such devices as Plato’s Noble Lie in the