A theory of Justice: Revised Edition

particular the good of everyone’s willingly acting from the public concep-
tion of justice in affirming their social institutions. When we consider
these questions we can work within the full theory. Sometimes we are
examining the processes by which the sense of justice and moral senti-
ments are acquired; or else we are noting that the collective activities of a
just society are also good. There is no reason for not using the full theory,
since the conception of justice is available.
However, when we ask whether the sense of justice is a good, the
important question clearly is that defined by the thin theory. We want to
know whether having and maintaining a sense of justice is a good (in
the thin sense) for persons who are members of a well-ordered society.
Surely if the sentiment of justice is ever a good, it is a good in this special
case. And if within the thin theory it turns out that having a sense of
justice is indeed a good, then a well-ordered society is as stable as one
can hope for. Not only does it generate its own supportive moral attitudes,
but these attitudes are desirable from the standpoint of rational persons
who have them when they assess their situation independently from the
constraints of justice. This match between justice and goodness I refer to
as congruence; and I shall examine this relation when we take up the
good of justice (§86).
61. THE DEFINITION OF GOOD FOR SIMPLER CASES
61. Simpler Cases
Rather than proceeding immediately to the application of the concept of
rationality to the assessment of plans, it seems best to illustrate the defini-
tion I shall use by first considering simpler cases. Doing this will bring
out several distinctions that are necessary for a clear understanding of its
sense. Thus I suppose the definition to have three stages as follows (for
simplicity these stages are formulated using the concept of goodness
rather than that of better than): (1) A is a good X if and only if A has the
350
Goodness as Rationality

---

properties (to a higher degree than the average
1
or standard X) which it is
rational to want in an X, given what X’s are used for, or expected to do,
and the like (whichever rider is appropriate); (2) A is a good X for K
(where K is some person) if and only if A has the properties which it is
rational for K to want in an X, given K’s circumstances, abilities, and
plan of life (his system of aims), and therefore in view of what he intends
to do with an X, or whatever; (3) the same as 2 but adding a clause to the
effect that K’s plan of life, or that part of it relevant in the present
instance, is itself rational. What rationality means in the case of plans has
yet to be determined and will be discussed later on. But according to the
definition, once we establish that an object has the properties that it is
rational for someone with a rational plan of life to want, then we have
shown that it is good for him. And if certain sorts of things satisfy this
condition for persons generally, then these things are human goods. Even-
tually we want to be assured that liberty and opportunity, and a sense of
our own worth, fall into this category.
2
Now for a few comments on the first two stages of the definition. We
tend to move from the first stage to the second whenever it is necessary to
take into account the special features of a person’s situation which the
definition defines to be relevant. Typically these features are his interests,
abilities, and circumstances. Although the principles of rational choice
have not yet been set out, the everyday notion seems clear enough for the
1. See W. D. Ross, 
The Right and the Good
(Oxford, The Clarendon Press, 1930), p. 67.
2. As I have remarked, there is wide agreement, with many variations, on an account of the good
along these lines. See Aristotle,