TO

THE STUDENTS
WHO HAVE HELPED ME TO LEARN
DURING MY YEARS
OF TEACHING

A DIALECTIC OF MORALS
Towards the Foundations of Political Philosophy
 2

by

Mortimer J. Adler
Associate Professor of the Philosophy of
Law at the University of Chicago

THE REVIEW OF POLITICS

University of Notre Dame
Notre Dame, Indiana
1941
 3

Table of Contents
Chapter I. Introduction: The Dialectical Task.........................8
Chapter II: Preference and Pleasure: Induction of a Principle
................................................................................................20
Chapter III: Pleasure and the Order of Goods........................38
Chapter IV: The Order of Goods and Happiness ...................54
Chapter V: Psychological Presuppositions: Limitations of the
Dialectic .................................................................................75
Chapter VI: Real vs. Apparent Goods: The Reality of Virtue.
................................................................................................80
Chapter VII: From Ethics to Politics: The Common Good and
Democracy ...........................................................................117

DETAILED TABLE OF CONTENTS

CHAPTER I. INTRODUCTION: THE DIALECTICAL TASK

The obligations of philosophia perennis in the twentieth century,
1; primacy of the dialectical procedure, 3; polemical and inductive as-
pects of this procedure, 5; moral questions one field among many for
dialectical work, 6; moral skepticism provides the dialectical opposi-
tion, 7; summary of its denials, 8; its cultural causes, 9; its resem-
blance to Greek sophistry 9; origins of this dialectical process in the
exigencies of contemporary teaching, 10.

CHAPTER II. PREFERENCE AND PLEASURE: INDUCTION OF A

PRINCIPLE
The fact of preference, 11; minimal meaning of “better-than,” 12;
agreement between moralist and skeptic on fact of preference, disa-
greement on explanation thereof, 13; the skeptical explanation in terms
of liking, 14; skepticism necessarily becomes hedonism, 14; subjectiv-
ism and relativism as a consequence, 15; introduction of the criterion
of the quantity of pleasure, 16; significance of the fact that pleasure it-
self is not the only criterion of preference, 16; induction of an abso-
 4
lutely universal hedonic precept, 18; objection: that this precept is not
a moral principle or rule of conduct because absolutely inviolable, 19;
separation of two problems: the objectivity and the violability of this
precept, 19; evidence that this hedonic precept is violable, 20; signifi-
cance of future moments, and of the span of life, for hedonic calcula-
tions, 22; reformulation of the hedonic precept, now verified as a mor-
al principle because seen to be violable. 22; return to the question of
objectivity in terms of the need for hedonic casuistry, 24; the distinc-
tion between universal and particular judgments with respect to pref-
erence and action, 25; significance of this distinction: different conse-
quences of errors of judgment on two levels, 25; objective criticism of
mistaken moral judgments, 26; the ultimate moral judgment in any
case has both objective and subjective notes, 27; extremism of the
skeptic denying the objective note caused by extremism of some mor-
alists neglecting or denying the subjective note, 27; tentative agree-
ment between the moralist and the skeptic: that there can be moral
knowledge only on the level of universal judgments, in contrast to
opinion on the particular level, 28; bearing of this agreement on the
skeptic’s insistence about the relativity of mores, 29; the skeptic’s
chief objection at this point: the paucity and vacuity of so-called
“moral knowledge,”. 30; this objection answerable only if pleasure
and quantity of pleasure are not the sole criteria of preference, 30.

CHAPTER III. PLEASURE AND THE ORDER OF GOODS

The meaning of pleasure as a criterion of preference, 31; apparent
equivalence of being pleased by something and preferring it, 31; ob-
jection: pleasure may not explain preference, because preference is ul-
timately inexplicable, 32; detection of the ambiguity of “pleasure,” as
naming both an object of preference and also the satisfaction in attain-
ing this object or any other, 33; objection: pleasure and displeasure are
the only objects between which men choose when they exercise pref-
erence. 33; distinction between desire and object of desire, between
object of desire and satisfaction of desire, 34; significance of the fact
that ‘pleasure.’ cannot univocally name both object and satisfaction of
desire, 34; evidence that pleasure is only one of the objects of desire,
34; this fact cannot be inconsistent with the fact that every object of
desire can be pleasurable in the sense of satisfying desire when at-
tained. 35; pleasure cannot explain preference if it is only one among
many objects of desire, nor can it explain preference when it is identi-
fied with the satisfaction of any desire, 36; resolution of the ambiguity
of “pleasure’’ and recognition that pleasure as object-of-desire is iden-
tical with sensual pleasure, 37; re-examination of the hedonic precept
in the light of this resolution, 37; proof that men cannot desire pleasure
in the sense of satisfaction-of-desire, 39; proof that neither pleasure
nor quantity of pleasure, as objects of desire, can be the sole criteria of
 5
preference, 39; introduction of the concept of the good as object of de-
sire, 39; the skeptic’s demand for proof that objects are intrinsically
good and bad, better or worse, 40; evidence of a variety of goods, 40;
summary enumeration of this variety, 42; the skeptic’s demand for
proof that the variety of goods is the same for all men and for proof
that all men should agree in preferring one type of good to another, 42;
explanation of the fact that there is some variety of goods for all men,
in terms of the diversity of their desires, 43; the common desire of all
men: to live and to live well 43; living well consists in the satisfaction
of all desires, 44; the skeptics objection: that all men do not mean the
same thing by “living well’’ and hence that variety and order of goods
is not the same for all men, 45; summary of agreements reached so far,
before meeting this objection, 45; the three questions which remain to
be answered, 46; introduction of the distinction between end and
means, 46; solution of the problem of preference depends on determi-
nation of the end for which means are chosen, 46.

CHAPTER IV. THE ORDER OF GOODS AND HAPPINESS

Proof that everything cannot be desired as the means to something
else, 47; the skeptics demand for proof that there is only one end for
each man, and this the same for every man, 47; proof that no man can
have more than one end as the ultimate object of all his desires, 48;
conception of the end as a whole of goods, and of the various means as
partial goods constituting this whole, 48; since the end, which is the
whole of goods, must be desired if anything is desired. it cannot be an
object of choice or preference, 49; two objections, first, that it may not
be possible for a man to possess all good things because partial goods
may be incompatible with each other, and second, that there can be no
moral rule about seeking the end if no man can avoid seeking it, 49;
proof that the various sorts of partial goods cannot be incompatible
with each other, 49; consideration of two equivalent conceptions of the
end: as the whole of goods and as the totality of satisfactions, 52; first
showing that the end, so conceived, must h the same for all men, 53;
repetition of the second objection, that a precept directing men to seek
this end cannot be a moral rule because apparently inviolable, 53; dis-
tinction between a formal and a material conception of the end, 54; in-
troduction of the notion of happiness as the end identified with living
well or a good human life, 55; two ways in which men can differ in
their material determinations of the end, or happiness, though it re-
mains formally the same for all in so far as it is identified with “all
good things,” 55; evidence that men differ in their conceptions of hap-
piness, 56; thesis that there is only one right or true conception of hap-
piness, 56; distinction between the real and the apparent good in terms
of true and false conceptions of happiness 56; “Seek the real good’’ is
a moral precept because violable, whereas “Seek the good real or ap-
 6
parent” merely describes, but does not prescribe, conduct, 56; the
skeptic’s objection that it has not yet been proved that men can be mis-
taken about what is happiness for them even when they differ in how
they conceive it, 57; answer in terms of the fact that human happiness
is relative to human nature, its natural capacities and desires, 57; hence
men can misconceive happiness by misunderstanding their own na-
ture, and by failing, in consequence, to recognize all the goods it de-
mands for the fulfillment of its capacities, 58; the skeptic’s objection
in terms of doubts about the specificity of human nature, the same for
all men and essentially distinct from the nature of other animals, 58;
the skeptics question about whether happiness can be misconceived
with respect to the order as well as the variety of partial goods, 59;
proof that there is only one right order of partial goods as well as only
one adequate enumeration of their variety 59; the skeptics repeated
demand for a proof that it is possible for a man to be happy, i.e., to
possess all good things, however ordered or enumerated, 61; this de-
mand meet by the distinction between being and becoming happy, 61;
Solon’s point about happiness in a complete life, and Boethius’ defini-
tion of happiness applied to temporal beings, 61; how the means are
the end in the process of becoming, 63; the skeptics return to his con-
cern about the violability of moral rules directing men in the search for
happiness, 64, two sources of violation distinguished: false thinking
and wrong willing, 64; the skeptic’s objection to the “hypothesis’’ of
free will, without which there would be no problem of preference at
all, 65; recognition of the several “hypotheses” about human nature
and human action, challenged by the skeptic, upon which depends an
adequate proof that we have moral knowledge about happiness as the
end, 67; how such knowledge about the end, once established, would
solve the problem of preference and would determine all the remaining
principles of moral theory, 67.

CHAPTER V. PSYCHOLOGICAL PRESUPPOSITIONS: LIMITATIONS

OF THE DIALECTIC

Summary of the argument so far, 69; its incompleteness because

of unverified presuppositions about the nature of man, 70 the need for
a dialectical approach to the truths of natural philosophy 70; the type
of truth which can be dialectically or inductively established, 71; the
hypothetical conclusions so far reached are nevertheless capable of
deductive elaboration, 72; the work of the remaining chapters outlined,
73.

CHAPTER VI. REAL VS. APPARENT GOODS:

THE REALITY OF VIRTUE
 7
(1) The good as object of desire. 74; this raises the question
whether the object is good because desired or desired because it is
good, 74; to solve this problem of the distinction between real and ap-
parent goods we must distinguish between natural and conscious de-
sire, 75; the relation between the good as metaphysically and as moral-
ly considered, 77; analysis of conscious as opposed to natural desire,
77; the natural good and the natural human good, 78; the natural hu-
man good always a real, and never an apparent, good, 79; the meta-
physical status of every object of desire, 80; consequences of this fact,
81; the good of man and the good for man, 82; correlation between the
unity of a complex nature and the unity of its complex natural end, 83;
correlation between the multiplicity of natural desires and the variety
of partial goods constituting the end, 84.

(2) The kinds of goods as means and ends, 85; primary and sec-
ondary objects of desire, 85; distinction between enjoyment and use,
86; distinction between extrinsic and intrinsic goods, in relation to the
traditional distinction between bonum utile and bonum honestum, 86;
happiness is strictly neither a bonum utile nor a bonum honestum, nei-
ther used nor enjoyed, 87; the problem of bodily goods in the Aristote-
lian tripartite classification of goods, 88; the problem of sensual pleas-
ure in the Thomistic tripartite classification of goods, 88; three sets of
criteria required for an adequate classification of goods, 89; using the-
se criteria, the adequate classification is fourfold, not tripartite, 89; re-
view of an earlier enumeration of five types of partial goods in the
light of this fourfold classification, 91; the problem of the goodness of
good habits and good acts, 91; solution of this problem in terms of
natural desire, 92; distinction between constitutive and functional
means, 93; distinction between universal and particular goods, be-
tween constitutive means-in-general and -in-particular, 95; generative
and productive means, 96; the fourfold classification of means and
ends in relation to one another, 97.

(3) Virtue as principal means to happiness, 97; two problems for

those who know what happiness is, 97; the secondary principles of
moral theory are the means productive of happiness, 98; deductive
proof that the virtues are these secondary principles, 98; the end and
its productive means must be correlative, 100, the two ways in which
happiness exists practically, 100 proportionality of happiness and vir-
tue, 100; distinction between intellectual and moral virtues as second-
ary and cardinal, 101; significance of this distinction for the apparent
conflict between two definitions of happiness in Aristotle’s Ethics,
Books I and X, 102; solution of this difficulty, 103; the problem of be-
coming happy, 104; formation of virtue, 105; “inchoate virtue’’ as a
stage in the development of character, 106; government and laws as
 8
causes of “inchoate virtue,” 106; the Aristotelian insight concerning
the relation of ethics to politics, 106.

CHAPTER VII. FROM ETHICS TO POLITICS: THE COMMON GOOD

AND DEMOCRACY

The ordering of ethics to politics and of politics to ethics, 108; the

political common good as an end which is also a means to happiness,
108; an adequate understanding of happiness and virtue necessary to
establish political philosophy, to defend it against realpolitik, and
safeguard it from errors, such as “totalitarianism,” 109; the nature of
the state, 109; its naturalness in the order of final causality as an object
of natural desire, 110; the mode of being of the state, 111, the state ex-
ists as an accident in so far as it exists in and through the habitual jus-
tice of its members, 112; the goodness of the common good, 113; the
common good as identical with the state’s well-being, 113; as such,
the common good is the end of government, government being its pro-
ductive means analogous to the relation of virtue and happiness, 114;
the common good as a whole of goods, 114; the major principles of
political theory, 115; the moral classification of governmental forms as
unequally good, Democracy being best, 116; the challenge of realpoli-
tik and of “totalitarianism” can be met only by the dialectical cure for
moral skepticism, 116.

CHAPTER I. INTRODUCTION: THE DIALECTICAL TASK

I N ST. THOMAS AND THE GENTILES I tried to define the obligations

of perennial philosophy in the twentieth century. Philosophy may
be perennial, but its work changes according to the cultural con-
ditions in which the philosopher lives and thinks. In its Greek be-
ginnings, philosophy arose out of the dialectical efforts of Plato and
Aristotle to clarify and order the welter of opinion. They struggled
not only with the sophists to divide the line between knowledge and
opinion; but they also moved in the realm of opinion to distinguish
the true from the false; and, in their patient consideration of pre-
Socratic thought, they both tried, though differently, to convert right
opinion into knowledge by making it evident to reason.

Although the result of their work was the establishment of phi-

losophy as a body of knowledge, founded on principles and devel-
 9
oped by demonstrations, we must not forget that, in their day, the
mode of their work was primarily dialectical. In saying this I do not
overlook the demonstrative or scientific achievements of Plato and
Aristotle; but those must be regarded as secondary, for the first
work of pioneers is to stake out the land, to clear away the brush, to
prepare the soil, and to dig for firm foundations. Only thereafter can
a city be planned, buildings raised, and interiors decorated.

The Platonic dialogues certainly reveal an intellectual pioneer at

work; but no less do the so-called “scientific” works of Aristotle, for
they are primarily records of exploration and discovery. Rather than
orderly expositions of accomplished knowledge, they are, not only
in their opening chapters but throughout, dialectical engagements
with adversaries, wrestlings with the half-truths of error and opinion
in order to set the whole truth forth.
Under the altered cultural circumstances of the Middle Ages,
philosophy lived a different sort of life. With few exceptions, the
mediaeval philosophers dwelt in the domain Plato and Aristotle had
won from the wilderness. The fields having been cleared and the
foundations completed, the philosopher now had a different sort of
work to do. Accepting the ground-plan, he proceeded to erect the
mansions of philosophy, each well ordered to the others, and in each
orderly disposition of many rooms. The architectural achievement
the mediaeval philosophers extended even to exterior facades and
the detail of furnishings within. And in all this work, the primary
mode of procedure was demonstrative rather than dialectical.

In contrast to the writings of Plato and Aristotle, the philosophi-

cal literature of the Middle Ages is expository rather than explorato-
ry. It proceeds by steps of analysis and synthesis. The so-called “de-
ductive” character of mediaeval thought must not be taken to mean
that medieval philosophers regarded philosophy as primarily or ex-
clusively deductive, but rather as signifying that they were no longer
in the pioneering stage. The inductive work, which is necessarily
first, had already been well done by the Greeks.

Again I must point out that, emphasizing the demonstrative

mode of mediaeval thought, I am not overlooking its dialectical
phases. But the dialectical efforts of the Middle Ages were mainly
in new territory, in theology rather than philosophy, and, of course,
in the borderlands between philosophy and theology. And even
where, within the sphere of purely philosophical questions, there is
 10
the obviously dialectical procedure of objection and reply, the dia-
lectic is defensive rather than exploratory. It is not undertaken as a
way of discovering the truth, but rather as a way to purify the truth
of admixed errors, or to assimilate to knowledge the truth that is
contained in errors. In every aspect and at every stage of this under-
taking, the philosopher regards himself as having a wealth of well-
established knowledge—an inheritance he must husband against
loss or decay, a fortune he must defend against the foes of truth, an
endowment not only to live on and by, but to increase by using it
well.

Now the modern followers of Aristotle and St. Thomas—or, for

that matter, the followers of Plato and St. Bonaventure—should not
neglect the fact that the cultural situation in which they find them-
selves is neither Greek nor mediaeval. The most dismal failure of all
modern “scholasticism” is its failure to be modern. This is true not
only of the second-hand text-books which try to be even more
demonstrative and less dialectical than the great mediaeval works,
whose intellectual achievement they reflect dimly, whose living ri-
gor becomes in the copy a rigor mortis. With some exceptions, it is
true even of the work of the best Thomists, from John of St. Thomas
to the present day.1

The reason for this is the failure to see precisely the way in
which modern culture imposes upon the philosopher a situation
analogous to, not the same as, the one in which Plato and Aristotle
did their work. It is not merely that the cultural aggrandizement of
the investigative or phenomenological sciences has gradually threat-
ened the very existence of philosophy and has progressively worked
to dispossess it of its ancient home; worse, and in consequence, the
prevalence of positivism today requires the philosopher to face an
audience radically skeptical of anything he may say, doubtful even
that he can say anything worth listening to at all.

I am assuming, of course, that a philosopher who is alive today

should try to talk to his contemporaries, and by this I mean an audi-
ence much wider than the inner circle of his like-minded fellows in
the philosophical enterprise. This is not the living philosopher’s on-
ly obligation, but if he is concerned with the life of philosophy in
modern culture, it is his primary one. To discharge it, he must pro-
ceed dialectically, not demonstratively, and his dialectical efforts
must resemble the Greek rather than the mediaeval mode of argu-
 11
ment. Though he might regret the fact that history’s progressive spi-
ral seems to throw him back to an earlier stage, he must return to the
pioneer work of the Greeks. He must once again try to be primitive-
ly inductive about the basic philosophical truths.1a

I describe the motion of history as the path of a spiral, because

the same ground is never retraced. Unlike the simpler cyclical mo-
tion which returns to the same place, progress along a spiral reaches
an analogous place—both the same and different. This is illustrated
by the fact that the contemporary follower of Aristotle and St.
Thomas cannot do exclusively either the sort of work which Aristo-
tle did or the sort done by St. Thomas. He must do both sorts, and in
that very fact he at once resembles and differs from each of them.

Like St. Thomas, the contemporary Aristotelian must continue

the constructive work that the Middle Ages began so well and did so
much of—the systematic and demonstrative elaboration of philo-
sophical knowledge.2 Like Aristotle, the contemporary Thomist,
because he is living in the modern world, must undertake the prima-
ry dialectical task of making evident the most rudimentary philo-
sophical truths.3 And because we are obligated today to do both
sorts of work, we can do neither well unless as we do the one, we
are always mindful of the other.

When perennial philosophy shakes off the dead skin of scholas-

ticism, and really comes to live in a modern metamorphosis, the
event will be signified by a renewal of the dialectical enterprise with
which philosophy originated in the Greek period, as well as by the
renovation of the edifice which the Middle Ages raised upon Greek
foundations. And each—the renewal and the renovation—will pene-
trate the other.

In this essay I am going to try to exemplify—even though inad-

equately and remotely—what I mean by the modern analogue of
Greek philosophical work. I am going to try to proceed dialectically
against those who say there is no moral knowledge; who say that
good and bad, right and wrong, are entirely matters of opinion; who
say, as a consequence, that “might makes right” in the sphere of pol-
itics. My aim is not merely negative, though in an effort to establish
first principles, my arguments will usually take the form of the re-
ductio ad impossibile. The destructive force of such arguments is,
 12
however, for the sake of a positive result—the inductive perception
of the most elementary truths.

There are many other topics which offer similar occasions for
dialectical work and, in every case, there is a parallelism between
the contemporary situation and that of fifth century Greece. Thus,
where the ancient sophists denied knowledge and said that every-
thing was a matter of opinion, the modern positivists deny that there
is any knowledge beyond or outside of the so-called positive scienc-
es, or, in other words, they say that philosophy is opinion. As, in the
ancient world, there were those who said that the truth was merely
what appeared to be the case, and hence relative to each individual,
so today there are similar relativists about truth. As then there were
those who denied any way of knowing except by the senses, so now
the intellect is denied as a distinct faculty of knowing. As among the
pre-Socratic physicists there were those who regarded the sensible
world as exclusively an affair of flux and becoming, in which there
were no enduring entities, such as substances, so those who regard
themselves as philosophical interpreters of modern physics also de-
ny substances, and view the sensible world as nothing but a process
of events.

In each of these cases, the dialectical task confronting us is

analogous to the task that Plato and Aristotle faced: to establish,
inductively, the distinction between knowledge and opinion and to
show that philosophy is knowledge; to establish that truth is objec-
tive and the same for all people because it is an agreement of the
mind with reality; to establish the distinction between sensitive and
intellectual knowing, and to show that people know things that they
cannot know by their senses alone; to establish the existence of sub-
stances as the subjects of change.

I have chosen the topic of moral knowledge—the objectivity

and universality of moral standards—because it is so relevant to this
critical moment in our culture. It will not be necessary to engage in
distinct dialectical enterprises for the separate fields of ethics and
politics. If skepticism about moral truths can be overcome at all, if
any judgments about good and bad can be shown to have the status
of knowledge, then a foothold is won for political as well as for eth-
ical standards. How much of the traditional content of ethics and
politics can be drawn from the few principles we are able to estab-
lish dialectically is something which remains to be seen.
 13

Let me describe the state of mind which I call moral skepticism.

It is not a total skepticism. There is no question about the validity of
the natural and social sciences. These sciences describe phenomena;
their generalizations can always be verified by reference to particu-
lar sense experiences; and even though the truths they achieve are
not “final” or “absolute”—but always relative to the data now at
hand—these truths are, nevertheless, objective in the sense that they
are matters upon which all competent judges can be expected to
agree in the light of the evidence.

In contrast to the affirmation of the natural and social sciences is

the denial of the moral sciences—the branches of practical philoso-
phy traditionally known as ethics and politics. This denial is made
on any one of three counts: (1) It may be involved in the general
denial of philosophical knowledge, for this would eliminate the pos-
sibility of practical philosophy as a body of knowledge.4 (2) Even
though some branches of philosophy are admitted as a kind of
knowledge, such as logic and mathematics.5 There is no philosophi-
cal knowledge which reports the nature of things; and to the extent
that ethics and politics depend upon theoretic philosophy, they are
involved in this denial; (3) Whether or not theoretic philosophy has
the status of knowledge, there cannot be any practical philosophy,
for that would be “normative” or “evaluative” and such judgments
can never be more than opinion.

The position of moral skeptics can, therefore, be summarized as

follows. He says that about moral matters (good and bad, right and
wrong, in the action of individuals or groups) there is only opinion,
not knowledge. They say that moral judgments are entirely subjec-
tive, i. e., having truth or meaning only for the individual who
makes them. They say that moral judgments are relative to the cus-
toms of a given community, at a given time and place, in which
case, although the judgments made by an individual may be meas-
ured in terms of their conformity to the mores of the group, the mo-
res themselves have no truth or meaning except for the group which
has instituted them. They say that all norms or standards are entirely
conventional, whether instituted by the will of the community or by
the will of individuals; and this amounts to saying that moral judg-
ments are ultimately willful prejudices, expressions of emotional
bias, of temperamental predilection. That these several statements
all come to the same thing can be seen in the fact that in every case
 14
the same thing is being denied, namely, the possibility of making
moral judgments which are true for all people everywhere, unaffect-
ed not only by their individual differences but also by the diversity
of the cultures under which they live.6

The issue is quite clear. The dialectical task is set. It will not do
for the philosopher simply to reiterate their claims concerning the
universality of moral truths, the self-evidence or demonstrability of
the principles and conclusions of ethics and politics. Nor is it suffi-
cient for them to be passive in their defense of them, however will-
ing they may be to answer objections; for the moral skeptic, espe-
cially if he is a positivist, is not entirely wrong in charging that eve-
ry answer begs the ultimate question—the question whether any-
thing the philosopher says is more than opinion. In this situation,
philosophers must be aggressive. They must engage the moral skep-
tic on his own grounds. They must open their adversary’s mind to a
perception of the truth—if not to the whole truth, at least to certain
aspects of the truth which will function as seed to be cultivated. This
is what I mean by an inductive use of dialectic.

I have elsewhere discussed the prevalence and causes of moral

skepticism among the educated classes in America today.7 It is the
position of most of the teachers in our secular colleges and universi-
ties, and naturally enough it becomes the position of their students. I
have already mentioned one of the causes, namely, positivism; but
there are two others which, although consequences or aspects of
positivism, should be separately noted. One is the kind of psycholo-
gy that is taught: the only knowledge we are supposed to have con-
cerning human nature comes to us from the laboratory or the clinic.8
The other is the emphasis, in the teaching of all the social sciences,
upon the diversity of mores: each culture consists of its own peculi-
ar system of values, and there is no way of evaluating cultures
themselves, no way of judging them, without begging the whole
question, for such judgments would have to be made in terms of the
“postulates” or assumptions underlying a given culture.9

Though the causes may be superficially different, insofar as

they reflect peculiarly modern conditions, the ultimate sources of
our moral skepticism are essentially the same as those responsible
for the teaching of the Greek sophists.10 The parallelism is extraor-
dinary. In both cases, the issue is a matter of general concern be-
cause it deeply affects the education of youth; in both cases, the phi-
 15
losopher is opposed to the dominant elements in the teaching pro-
fession.

The dialectic of morals which I shall now proceed to outline is

not an imaginary intellectual process. It is rather a distillation of
actual arguments which President Hutchins [the late Robert M.
Hutchins, president of the University of Chicago 1929-1951] and I
have had with students in courses devoted to the reading of great
works in ethics and politics. The situation we face year after year is
the same: the students are, for the most part, moral skeptics. They
challenge us to try to change their minds. In meeting that challenge
we have found certain modes of argument most effective. The only
invention involved in the development of this dialectic is the precise
ordering of the steps. It is necessary to find those points of departure
which make contact with the minds we are trying to move; and it is
necessary to sustain the motion, once started, by linking the steps in
a tight sequence, so that no leaps are required. Most of the steps are
provided by the tradition, especially by Plato and Aristotle, but we
have found it necessary to produce an ancient play of the mind in
modern dress.

The whole dialectic cannot be accomplished in a single se-

quence. Several motions are involved, some from opposite direc-
tions, but all converging on the point to be established. What I am
going to set down in each case must be regarded as the bare plot for
a dialogue between teacher and student. To write such dialogues out
in full—to report in detail the actual sessions in which these argu-
ments took place—would require more skill than I possess, and
more space than is available. Furthermore, what is essentially the
same intellectual process can take place in countless different ways,
according to the contingent circumstances of actual discussion. The-
se dialectical plots can never be enacted in the same way, but they
are, nevertheless, common to a wide variety of conversations about
such themes.

NOTES TO THE INTRODUCTION

1. (p. 10) If we consider carefully the character of these exceptions—

their philosophical mood and temper—they illustrate, by contrast to
the rest of “scholasticism,” what it means for philosophers to re-
member the thirteenth without forgetting the twentieth century.
Confining myself to the field of moral philosophy, I should cite as
 16
striking exceptions—striking in themselves and also striking be-
cause it is only in the very recent past that such work has oc-
curred—the writings of Jacques Maritain (such as True Humanism
and Scholasticism and Politics) and of Yves Simon (especially
noteworthy in this connection is his Nature and Functions of Au-
thority); and I must also mention the work of Father Walter Farrell.

1a. (p. 11) In St. Thomas and the Gentiles, I wrote: “Far from making
every effort to join issue with those who differ from us, we have, in
my judgment, not even begun to make an effort properly directed
and properly proportionate to the task at hand. We have been loath
to absent ourselves from the felicity of moving further into the inte-
rior of philosophical thought, when there is pressing work to be
done on the border, the arduous and lowly work of the pioneer. The
borderland I speak of is marked by the issue between those who
hold, as we do, that philosophy is a field of knowledge in which
there can be perennial truth and those who deny it” (p. 20). In this
earlier work I tried to find a parallel for our task in the sort of dia-
lectical work St. Thomas did against the gentiles in the sphere of
faith. I now think a better parallel is to be found in the dialectic of
Plato and Aristotle against the sophists, because the ancient effort
was, and the modern effort must be, entirely within the sphere of
reason.
In saying that the modern effort must be entirely within the
sphere of reason, I am thinking of what I regard as the primary task
of philosophy in the contemporary world—to win respect for itself
in a culture that is predominantly positivist. I hope it will be under-
stood that this is not incompatible with the general notion of a char-
acteristically Christian philosophy—the work of reason elevated by
faith—for although faith seems to have been indispensable for the
mediaeval discovery of truths not known to the ancient pagans, the
truths, once discovered, are possessed by reason and can, therefore,
be made acceptable to the reason of modern pagans. For the most
part, Christian philosophy, because its truths are rational, can be
taught to pagans even though it could not have been initially devel-
oped by them. There is, however, one profound limitation on the
foregoing statement, which is crucially relevant to the present un-
dertaking, namely, the fact that Christian moral philosophy is not,
and cannot be, purely a possession of reason, because as practical
wisdom it is necessarily guided by faith and subalternated to moral
theology. (M. Maritain has completely analyzed this point in Sci-
ence and Wisdom, New York, 1940: Part II).
 17

The doctrines of humankind’s fall, redemption, and salvation

are theological, not philosophical. Since in the practical order we are
concerned with ends and means, we cannot neglect the difference
between the end as declared by faith and as known by natural rea-
son; nor can we ignore the fact that natural means are insufficient
for a supernatural end; they may not even be sufficient for a natural
end, if the “natural human being’’ is a hypothetical creature who
does not exist. But even though a purely natural moral philosophy is
not the whole truth, taken theoretically, and even though a purely
rational morality may be practically false because of its theoretic
inadequacy, we must nevertheless begin our dialectical undertaking
with what reason alone can accomplish. If we succeed in winning
the moral skeptics to the path of reason, and if we take them with us
as far as reason can go, it will then be time enough to ask where we
are; for then, as not now, they may be willing and prepared to con-
sider the relation of theology to philosophy, of faith to reason, in the
practical order. The reader should, therefore, understand why our
present objective is the induction of Greek, and not distinctively
Christian, moral principles.

2. (p. 11) I am not forgetting that this process cannot occur, today, in
exactly the same mood or manner as in the Middle Ages. Since the
aim is certainly not just to repeat the mediaeval construction, we
must attempt further and more detailed analyses, and these must
take account of every genuine advance in knowledge, and every
sound critical insight, which the modern world has gained. We may
even find it necessary to tear down some parts of the mediaeval
building and to reconstruct it, in order to let modern light in, to ven-
tilate it properly and to make it truly habitable by a modern mind.
And in emphasizing here the demonstrative and expository character
of such constructive, or reconstructive, work, I do not mean to ex-
clude dialectical procedures entirely, for they are necessarily in-
volved. But the kind of dialectic by which a living Thomism contin-
ues to grow is mediaeval rather than Greek in type—that is, it is not
primary and inductive but secondary and auxiliary to the deeper
penetration of truths already known.

3. (p. 12) Here, too, there is a difference in the mood and manner in
which a similar task is undertaken; for whereas Aristotle was genu-
inely exploring the philosophical field by dialectical methods, and
discovering truths by inductive procedures, we are not learning the-
 18
se elementary truths for the first time, but rather are trying to teach
them to a world which denies their possibility. We must, therefore,
use the dialectical method and the inductive procedure as instru-
ments of instruction rather than of discovery. It is highly probable,
of course, that what occurs as a discovery of truth for those whom
we try to teach may be more than a mere re-discovery for us, the
teachers. Since the cultural context of the modern world is different,
since the steps we must take in reaching the same truths are not pre-
cisely those which Aristotle took, the truths themselves may be seen
in a new light; and it is even possible that, as a result of such efforts,
new truths may be discovered.

4. (p. 14) It should be noted that what is being denied is not politics as
one of the social sciences, but politics as a branch of practical, or
moral philosophy.

5. (p. 14) They are regarded as regulative disciplines, as formal scienc-

es, whereas the natural and social sciences are regarded as sciences
of the real, even though the only reality be phenomenal.

6. (p. 15) Two other denials are implicit here: (1) the denial of a natural
moral law, in consequence of which morality becomes entirely con-
ventional; and (2) the denial that moral judgments are expressions
of reason, rather than of will or passion.

7. (p. 16) In This Pre-War Generation, Chapter 1, Reforming Educa-

tion: The Opening of the American Mind, (Edited by Geraldine Van
Doren), Macmillan Publishing, New York, (1977).

8. (p. 16) The neglect or denial of what, in contrast, I would call

philosophical psychology results in the denial or, what is just as bad,
the misconception of mans rationality and freedom. The relevance
of such denials or misconceptions to moral skepticism will become
apparent in the course of the dialectic.

9. (p. 16) This can be most strikingly exemplified by the position of

those political scientists who are willing to urge us to fight for de-
mocracy, but who refuse to argue that the principles of democracy
are intrinsically, and absolutely, right, or even objectively better
than the principles of totalitarianism. Adopting the views of realpo-
litik, they must regard this issue as nothing more than a struggle be-
tween “ideologies”—the one to which we are devoted not being ob-
 19
jectively better than the other, but better-for-us because it is ours by
the accident of cultural location.

Let me add here that all the facts of cultural anthropology must be
admitted. The moral skeptic often supposes these facts to be abso-
lutely incompatible with the position that some moral judgments are
true for all people everywhere. But this is not the case. The truths of
moral philosophy, the principles of ethics and politics, do not re-
quire us to shut our eyes to any facts about human life and human
society. The precise relation between the universality and absolute-
ness of moral truth, on the one hand, and the diversity and relativity
of the mores, on the other hand will become apparent, I hope, in the
course of the dialectic.

10. (p. 16) The position of Thrasymachus in The Republic, and the
views attributed to Protagoras and other sophists, in the writings of
Plato and Aristotle, are perfect expressions of moral skepticism.
Although the thing we call “positivism” is typically modern, be-
cause it arises in modern times with the gradual distinction of sci-
ence from philosophy, there is a Greek analogue in so far as the
sophists were not total skeptics. All but the most extreme among
them, such as Cratylus, were willing to admit that we had
knowledge of the physical world; in fact, they used such knowledge
to make their point that in moral matters only opinions prevailed.
They were fond of saying that fire burns in the same way in both
Greece and Persia, both a hundred years ago and today, but the laws
of Greece and Persia are not the same nor are the customs of antiq-
uity and of the present. Of nature, because it is nature and has a per-
sistent uniformity independent of human will, there can be
knowledge, but there can be only opinions on moral matters, be-
cause they are not natural, because they are entirely conventional,
entirely dependent on human institution, entirely expressions of
will. The sophists knew a great deal about the variety of customs;
obviously impressed by the relativity of mores, they made the same
false supposition that is made today, namely, the incompatibility of
such facts with the possibility of universal moral principles.

Finally, it can even be said that the sophists’ view of human nature,
without benefit of experimental research or clinical investigation,
emphasized, as does our current scientific psychology, the will or
passions, rather than the reason, and made the sensitive faculty the
primary, if not the exclusive, principle of human knowledge. The
 20
main points of this analogy between the ancient sophists and the
contemporary moral skeptics is confirmed, from the other side, by
the late Professor F. C. S. Schiller, the follower of William James
and John Dewey who, more explicitly than they, avowed the moral
skepticism which is implicit in pragmatism. cf. his essay, “From
Plato to Protagoras” in which Schiller sides with Protagoras (in
Studies in Humanism, N. Y., 1907: Ch. II).

CHAPTER II: PREFERENCE AND PLEASURE: INDUCTION OF A

PRINCIPLE

L ET US BEGIN with an indisputable fact. No one can deny the

fact of preference. If there are people who say they have never
preferred one thing to another, never done one thing rather
than another, we must inquire, then, whether they have ever experi-
enced desire at all, of any sort. And if they admit having had the
experience of desire, they can certainly be made to understand the
difference between something which would satisfy that desire and
something which would not. Hence, they can at least imagine a situ-
ation in which, given a certain desire, they would prefer one thing to
another. But it is unlikely that we shall be compelled to persuade
anyone about the fact of reference—certainly not about its exist-
ence, though, perhaps, about its significance. That, then, can be our
starting point. 11
 21

11. I should like to observe here that the fact of preference plays a role in
the dialectic of morals like the role played by the fact of change in the dia-
lectic of substance. If anyone persist in denying the existence of change, it
will be impossible, I think, to induce that person to see the necessity for
there being a multiplicity of individual substances. So, too, if anyone really
persist in denying that people exercise preferences, it will be impossible to
carry him or her any distance at all into the field of morals.

The fact of preference can be set forth in a simple formula

which describes every case: X, who is a human being, prefers A to
B, and here A and B can either be objects or courses of action. In
fact, whatever A and B stand for, whoever prefers A to B is saying
that A is better than B. The fact of preference is thus seen to be
equivalent to the judgment of better-than.

But a student may object, of course, that he does not know what
“better-than” means; he has admitted the fact of preference, but he
has not admitted that there is anything really good and bad, or better
and worse. If “better-than” means no more than “preferred-by-me,”
says the student, then the equivalence of the fact of preference with
the judgment of better-than can be conceded; but not otherwise.

At this point let us focus the whole issue on the fact of prefer-
ence. Let us consider two people, X and Y, both of whom, as a mat-
ter of fact, prefer A to B. Let X be a moral skeptic, such as the stu-
dent is, who claims that in expressing this preference he is express-
ing nothing more than his private opinion; X, furthermore, denies
that there are any principles behind this judgment of preference
which might lead any other person, in the same situation, to judge in
the same way. And, for the sake of contrast, let Y be a moralist who
claims that his reasons for preferring A to B include universally val-
id principles which set up an order of goods, of things as better and
worse, for any person at any time and place.

Now it will be observed that the two people, X and Y, agree up-
on the fact of preference, though they disagree in the explanation
they give in answer to the question, Why do you prefer A to B? We
have not yet heard the moral skeptic’s explanation of his preference,
but we know it must be different from the moralist’s. It should be
noted, moreover, that it makes no difference whether X and Y both
 22
prefer A to B, or whether they make opposite choices here, for in
either case the fact of preference remains to be explained, and it is
the difference in the explanations which matters. Let there be no
doubt on this point, for if the explanation given by the moral skeptic
is not radically and irreducibly different from the explanation given
by the moralist, there is no issue.

We must, therefore, ask the student to explain preference. He

may, of course, answer that there is no explanation, that he never
has any grounds whatsoever for preferring one thing to another. If
he says this, he must be asked why, then, does he prefer one thing to
another. Should he reply that, in fact, he does not really prefer one
thing to another—that, when he appears to choose A rather than B,
it is only in the way in which one tosses a coin to make a decision,
or in the way in which one makes a blindfold choice between the
right hand and the left—it will be necessary to remind him that he is
now denying what before he admitted. He was not originally asked
to agree that he, in fact, did one thing rather than another, but that he
preferred to do this rather than that. In short, he cannot admit the
fact of preference and deny that he regards one thing as better than
another, even if that means only better-for-him. Hence, he cannot
refuse to give us some explanation of his preferences, some account
of how or why he regards one thing as somehow better than another.

At this point the student can be helped to a decision by being

presented with the following dilemma: either what is preferred is
something which any rational being would prefer under those cir-
cumstances, something which in the nature of the case is better than
the rejected alternative, or the preference expresses nothing more
than this individual’s feelings at the moment. The student will rec-
ognize at once that if he take the first horn of the dilemma, he is
conceding the existence of moral knowledge, a rational judgment
about what is good and bad, which has truth for any person. Since
the existence of moral knowledge is to be proved, the student quite
properly takes the other horn of the dilemma.

Let us now make the student’s position explicit. He is saying

that he prefers A to B, because he likes A. Furthermore, he wishes
to be understood as saying that his liking A is entirely a matter of
his present state of feelings about A and B; tomorrow he might like
B. And he would not be at all surprised to find that other people
liked B when he liked A, or conversely; nor would he attempt to
 23
argue with them about this difference in their tastes, for about liking
and disliking there can be no argument.

We have now discovered an interesting point, which the student

should recognize. The moral skeptic, when urged to explain the fact
of preference, becomes a hedonist. In order to avoid saying that he
prefers A because his reason tells him it is really better, he says that
it is entirely a matter of his feelings—feelings of pleasure and dis-
pleasure. Nothing new has been introduced into the discussion by
the use of the words “pleasure” and “displeasure” for the student
will admit that “A pleases me” or “A gives me pleasure” is the ver-
bal equivalent of “I like A.” Hence, with the student’s consent, we
can conclude that a moral skeptic is one who explains preference in
terms of feelings of pleasure and displeasure—feelings which are
entirely subjective, operating for this individual and at this moment
in this situation.

If, now, we ask the student why he likes A, why it pleases him,
he may protest the question. There is no why for liking. The feeling
of pleasure is an immediate experience which determines prefer-
ence, and that is all there is to it. The student may even tell us that
we have no right to ask why, for the very question implies that there
are reasons; whereas he has already told us there are none unless the
feeling of pleasure itself be called a “reason” for preference. If we
wish to use the word “reason” that way, then pleasure and displeas-
ure, he reiterates, are the only reasons for preference.

But there is still some room for inquiry about these feelings of
pleasure and displeasure. We admit that there is no problem if A
pleases and B displeases. In this simple case, the principle of prefer-
ence is clear: pleasure is preferred to displeasure. And no further
explanation need be given of this principle, for we can agree with
the student that it is a principle of animal conduct: animals embrace
what they like, and avoid what they dislike. That can be taken as a
scientific fact. And although with some of the lower animals their
likes and dislikes are instinctive (and so common to all members of
the species), in the case of humankind, instinct is either weak or
non-existent, and human likes and dislikes are matters of individual
conditioning. Hence, we cannot as a matter of scientific knowledge
declare what all people will like or dislike. Therefore, on moral mat-
ters there is only opinion.
 24
All cases are not, however, so simple. We must ask the student
to consider a situation in which he has often found himself; he likes
both A and B. Whereas in the simple case first given, B was posi-
tively displeasing, here B is pleasing. Now what is the principle of
preference? The student will answer, as it seems he must, that in this
case he prefers A because A is more pleasing—he likes A more than
B.

We have thus arrived at a second principle of preference. The

first principle was: A is considered better-than-B-for-me whenever
A gives me pleasure and B displeasure. The second principle is: A is
considered better-than-B-for-me whenever A gives me more, and B
less, pleasure. The question now is whether a genuinely new criteri-
on has been introduced. According to the first principle, pleasure
was the only criterion of preference. The second principle appears to
add a new criterion: quantity of pleasure. To be sure we understand
this new criterion, let us consider another case in which the alterna-
tives are A and C, on the one hand, and B, on the other. Let it be
supposed that B is more pleasing than either A or C taken separate-
ly, but that together A and C will give more pleasure than B. Apply-
ing the standard of quantity, the student tells us that in such a situa-
tion he will prefer A and C to B.

Would any other person make the same judgment? we ask. Yes,
says the student, faced by a choice between more and less pleas-
ure—whether the greater quantity be simply the greater intensity of
one pleasure over another, or the summation of two pleasures which
exceeds a single pleasure—any person would prefer more or less. Is
this, we ask, a matter of human instinct or of human reason? Why is
more of what we like better than less? The student replies that he
doesn’t know whether it is instinct or reason, but that it makes no
difference. Animals not only seek pleasure and avoid displeasure,
but they also prefer more pleasure to less. This is simply the fact,
and it applies to human beings as well as other animals. It is an ul-
timate fact, about which no further whys can be asked.

But, we persist, the criterion of quantity as a principle of prefer-

ence raises further questions which must be faced. In the first place,
the student must now admit that pleasure is not the only criterion of
preference. Quantity is an additional criterion, and a more ultimate
criterion, since one pleasure is preferred to another because of quan-
tity, not one quantity to another because of pleasure. The student
 25
objects, saying that more pleasure is better simply because it is more
pleasure, not because it is more.

To argue this question, let us consider a case. One is faced with

a choice between a bag containing three apples and a bag containing
two. One likes apples. Both bags are obtainable with equal ease. Let
us further suppose that one’s appetite for apples is equal to eating
three of them in succession. The preference, then, for the bag of
three must be based on the difference in quantity, on the fact that
more of the same is better than less. Hence whenever there is an
alternative between two things which please in the same way, pleas-
ure itself cannot determine preference, but only something which
measures the pleasure, namely, quantity. And if quantity measures
pleasure, and if it is on such measurement of pleasure that prefer-
ence is based, then quantity is a more ultimate criterion than pleas-
ure.

But the student counters by asking us to consider an opposite

case, in which pleasure appears to measure quantity. In this case,
one is faced with a choice between two bags, containing an equal
number of objects, let us say, three apples and three bitter pills. Of
course there is no problem here, we hasten to admit, because here
the choice will be made in terms of pleasure as against displeasure
The student then revises the situation, supposing the bags to contain
three apples and three bars of chocolate, both of which give pleas-
ure, and let us even add, he says, that the pleasure they give is of the
same sort. The student will soon realize that his case has now be-
trayed him, for if any preference is to be expressed it will have to be
in favor of the greater pleasure to be obtained from the unit of apple
as against the unit of chocolate, or conversely. Given an equal sum
of such units in the two bags, and given the same rate of diminish-
ing increment of pleasure from successive units, he must, according
to his own principles, prefer the bag which contains the object, any
unit of which gives him greater pleasure.

That pleasure never measures quantity, as quantity measures

pleasure, is thus summarily seen in the fact that there is no ground at
all for preference between equal quantities of the same pleasure, and
in the fact that whenever one quantity is preferred to another it is
because the one preferred gives more pleasure, not simply pleasure.
 26
Granted, the student may now be willing to say, but what is the
significance of all this? There are two answers: first, that pleasure
and displeasure are by themselves, taken without qualification or
measurement, insufficient to explain all the facts of preference; se-
cond, the criterion of quantity, as irreducible to the criterion of
pleasure, and as more ultimate than pleasure because measuring it,
may help us to modify the extreme character of the student’s moral
skepticism. To show him this, we go on to the next point.

If pleasure, as against displeasure, were the only criterion of

preference, the student could persist in holding his original position
that every moral judgment (every judgment of A-better-than-B-for-
me) was entirely individual, made by him at this moment according
to the state of his feelings, and hence subjective, hence an opinion
that has no relevance to anyone else faced with the same alterna-
tives. But if instead of A representing a source of pleasure and B a
source of displeasure, we let A represent a greater, and B a lesser,
pleasure, then is the judgment of preference for A over B subjective
in the same way? Yes, says the student, because the fact that I find
greater pleasure in A at this moment does not mean that anyone else
does, or need to, or even that I will tomorrow. This we must grant,
but that the principle itself is not subjective is our real contention.

We are not trying to say that two different individuals, or the

same individual at different times, will find greater pleasure in A.
We are saying, however, that whenever anyone finds greater pleas-
ure in one thing than in another, that is the thing he will prefer. And
this principle of preference is absolutely universal. It holds for all
people everywhere and at all times. One might formulate this prin-
ciple as follows: if anything at all is good, a larger amount of good
is better than a smaller. Even people who say that the only good is
pleasure are nevertheless compelled to agree that they would be
fools if, in pursuing such goods, they ever took less pleasure when
more was available.

Here, then, is a moral rule binding all people. Let us state it as a

moral rule, in the imperative mood: Always choose the greater
good. Agreeing for the moment that pleasure is the only good, this
command can be stated declaratively: A person should always
choose more pleasure in preference to less. And this moral judg-
ment, however stated, and with whatever meaning is assigned to the
word “good,” appears to be universally true, a matter of knowledge,
 27
not opinion. Hence when A stands merely for “more pleasure” and B
stands for “less pleasure,” the words “for me” can be omitted from
the judgment that A is better than B.

Not so fast, says the student. Either you did not need the criteri-
on of quantity to make this point, or I do not understand its signifi-
cance. You could have made the same point, he goes on to explain,
in terms of pleasure and displeasure. For if A stands for “source of
pleasure” and B for “source of displeasure,” then the words “for
me” can also be omitted from the statement that A is better than B.

Here, too, there is a universal moral rule, if you wish to call it

such: Always choose pleasure rather than displeasure. And if you
want to substitute the words “good” and “evil” as verbal equiva-
lents, you can say: Always choose good rather than evil. But such
statements are either tautologies, or they do no more than merely
report the facts of animal behavior, namely, that all animals seek
pleasure and avoid displeasure, or seek more pleasure rather than
less. All that you have done, he tells us, is to disguise a scientific
fact by putting it into the linguistic form of a command, or a moral
injunction, using the word “should.” What is the point of saying that
people should do what they cannot fail to do? Is there any meaning
to a moral rule which cannot be violated? In fact, have we the right
to call anything a moral rule, a rule of conduct, unless it can some-
how be violated? For otherwise the moral rule would not be a basis
for judging people as good and bad, right and wrong in their actions,
according as they conform to or transgress the rule.

The usual conception of the moralist’s position certainly in-

volves not only universal rules, but the possibility of making such
judgments about people in terms of them. Furthermore, the whole
discussion is off the point, because the real judgment of preference
is made by me here and now in this situation, and is determined not
by such universal principles as “pleasure is always better than dis-
pleasure” or “more pleasure is always better than less pleasure,” but
by my present, thoroughly individual feelings about objects I like
and dislike, or like more and less intensely.

By such objections, the student has brought the issue into clear-
er focus. He has raised two questions, not one, and these must be
separated. The first has to do with the point about the violability of
moral rules. In a sense he is right that an inviolable moral rule is not
 28
a statement of what should be done, but of what in fact is the case
about the nature of human conduct. There must be some distinction,
he rightly insists, between moral and natural necessity, between a
moral statement and one made by the psychologist as a descriptive
scientist. The second question concerns the subjectivity of any actu-
al preference; and here again the student is right if the preference is
solely determined by how he feels about A and B. Even if the judg-
ment, that people should always prefer a greater good, were truly a
moral rule, because violable, it would have no significance practi-
cally if, as between A and B, preference were entirely determined
by how an individual felt about A and B, which he liked more, for
example. Let us consider these two points in order.

The student’s objections, it will be remembered, arose from his

inability to see why we were so insistent about the criterion of quan-
tity. That can now be explained to him, perhaps, in terms of the fact
that it makes it easier to formulate a moral rule which shall be at
once both universal and capable of violation. If we had used the cri-
terion of pleasure, as against displeasure, to formulate a rule (e.g.,
that pleasure should always be preferred), it would have been ex-
tremely difficult, perhaps even impossible, to show that this rule
was not a statement of observable fact, confirmed by all psychologi-
cal investigations; for even the pathological cases of masochism are
generally understood as people taking pleasure, as opposed to dis-
pleasure, in sensations of pain. Let us see, therefore, whether the
criterion of quantity helps us.

We must take a more complicated case than any we have so far

considered. Let A and C stand for a sum of pleasures greater than
the single pleasure B. But let the conditions be such that whereas A
and B are pleasures capable of immediate enjoyment, C is a pleasure
that cannot be enjoyed until some time in the future, though it can
be imagined now. Furthermore, let the future enjoyment of C de-
pend upon the present choice of A rather than B; in fact, let the pre-
sent enjoyment of B exclude the possibility of a future enjoyment of
C. Finally, let us state the facts about quantity: B is a greater pleas-
ure than either A or C taken singly, though the sum of A and C is
greater than B. According to quantity as a criterion of preference,
the student must admit that the rule of anyone’s conduct in this case
must be that he should prefer A and C to B. But, as a matter of fact,
will everyone behave accordingly?
 29
To obtain the student’s answer to this question, we take a con-
crete case in which the choice is between the pleasure of going to
sleep as against the pleasure of further conviviality. Now the latter
pleasure may be regarded as greater than the former taken by itself;
but the former entails a future pleasure—the pleasure of feeling
rested on the morrow, here set against the displeasure of weariness
when there is work to be done. Let it even be supposed that the
pleasure of feeling rested on the morrow, as now imagined, is less
than the presently enjoyable pleasure of further carousing. It is only
when the two pleasures—of sleep now and feeling rested tomor-
row—are taken together, that they exceed the alternative which is
involved.

Will the student deny that a person who made such calculations
as these might sometimes violate the universal rule, and choose the
lesser pleasure? The student will undoubtedly admit that he has
made such a foolish choice himself; he will remember moments of
repentance for having made the wrong choice, moments of resolu-
tion not to be so foolish again. But wherein lies the folly, unless it is
wisdom to follow a true rule of conduct? And how could one ever
repent, in cases of this sort, if the rule we have stated is strictly invi-
olable?

Shall we not, therefore, now ask the student to admit that by his
own criterion of preference we have formulated a universally true
rule of conduct, true for any person and yet also frequently violated?
The student may still demur, saying that at the time of the choice,
the lesser pleasure actually seemed the greater; and that repentance,
with its recognition of folly, occurred at a later time when a more
accurate calculation of the opposed pleasures was made. Thus, he
may continue, it remains true as a matter of fact that people always
prefer what at the time appears to them to be the greater pleasure,
although the apparently greater may not be really so.

Undoubtedly, we must admit, such mistakes in calculation are

sometimes made, but that is not always the case. We can regret two
sorts of mistakes: on the one hand, mistakes of calculation; on the
other, mistakes of acting contrary to our calculations. It does not
require much effort of thought to add to the pleasure of going to
sleep now the consequent pleasure of feeling refreshed in the morn-
ing; but it does require strength of will, as is popularly said, to give
sufficient weight to a future pleasure against a present one. That is
 30
why many people have violated the sound rule which prescribes the
choice of greater pleasure (the sum of A and C, against B). At the
moment of the choice, they like B more than A, and even though
they fully realize that the alternatives do not consist of A against B,
but of A, along with C, against B, they foolishly put the morrow out
of mind. They set up as the maxim of their conduct, “Eat, drink, and
be merry, for tomorrow we die.” But if that maxim be a moral truth,
then the rule about always preferring the greater good must be
false—on the condition, of course, that we do not die on the mor-
row. Since, as a matter of fact, most of us make choices in the ex-
pectation of a normal span of life, the maxim which permits us to
take the greater pleasure at the moment is false precisely because it
is not the greater pleasure in that larger framework of moments
which constitutes a whole life.

We must ask the student at this point whether he is willing to

agree that a person, who has both memory of the past and imagina-
tion of the future, exercises preferences not only for the present
moment, but for the future, and in view of his or her life as a whole.
If he says No, we need only remind him that he is neglecting obvi-
ous facts with which he is acquainted, for example, the many cases
in which he and other people have preferred a momentary displeas-
ure for the sake of a future pleasure. As between going to the dentist
now to have a cavity filled, when the tooth is not yet decayed
enough to hurt, and waiting for toothache to set in, most of us make
the choice of what is at the moment unpleasant for the sake of
avoiding a greater unpleasantness later. If, in the light of cases of
this sort, the student now admits that the criteria of preference re-
quire us to consider future moments as well as present ones, then we
can formulate a principle of preference, which subsumes the other
two. This rule of conduct is: In any case in which a choice can be
made, people should prefer the alternative, which, in the long run or
viewing life as a whole, maximizes pleasure and minimizes dis-
pleasure.

We must remind the student here that, so far, we have adopted

his own criteria of preference—pleasure against displeasure, or the
greater quantity of pleasure—and that we have succeeded in show-
ing him, in terms of his own criteria, that he himself must
acknowledge the truth of a moral rule, which is of universal applica-
tion; and we have also now shown him that such a rule, especially in
its most general formulation, is normative, saying how people
 31
should behave, not descriptive, saying how they do, the evidence for
this being the obvious violations of the rule, and the experience of
repentance for folly in so doing, whether it results from bad thinking
or weak willing. In other words, the operations of people in exercis-
ing preferences cannot be simply instinctive, even though it be in-
stinctive to humankind’s animal nature to seek pleasure and avoid
displeasure. We cannot ask why human beings should prefer pleas-
ure to displeasure, for the student is right in replying that there is no
reason for this except the fact of instinctive determination itself. But
if in a complicated situation, involving sums of pleasure and dis-
pleasure, some present and some future, we ask why a human being
should prefer one set to another, instinct by itself will not suffice as
an answer.

Here it is necessary to say that, in view of humankind’s instinc-

tive preference for pleasure over displeasure, and in the light of
memory and imagination, human beings have developed a rule of
calculation which goes beyond the momentary promptings of in-
stinct. Since this rule is not itself instinctive, it can be misapplied by
bad thinking in particular cases, and even when the calculations are
well performed, it can be violated by contrary choices. A violable
rule of this kind, developed as the result of thinking about the prob-
lems of preference, can be called a rule of reason. It satisfies all the
requirements of a universally true moral judgment, providing as it
does both a prescription for conduct and a standard whereby to
judge people’s choices as wise or foolish, right or wrong. Hence we
can say to the student that, accepting his own explanations of the
fact of preference, we have removed one of the unqualified nega-
tives in his moral skepticism, namely, that no universally valid mor-
al judgment, no rule which directs all people everywhere, is possi-
ble. The possibility is more than proved by the existence of at least
one such rule.

It is now the student’s turn to remind us that we have another

question to answer before we have really won our point. Granted
that there is such a rule, it does not determine actual preferences in
particular situations, for they are determined by the feelings of
pleasure and displeasure, remembered, imagined, or presently expe-
rienced, which vary among individuals according to their tempera-
ments, their biographical conditioning, and their social environment.
Hence, the rule that A should be preferred to B whenever A repre-
sents a greater pleasure, is an empty formula, which does not oblige
 32
two people to agree in their actual judgments. One can say that he
likes A better, and the other can say that he likes B better and so,
without violating this so-called universal moral rule, the two people
can make quite opposite choices in the same situation. Each per-
son’s preference expresses his or her own private opinion, and noth-
ing more, for according to the rule itself, he or she has no grounds
for saying that the other person has made a wrong choice.

Certainly we must admit, the student tells us, that if moral

judgments are worth anything at all, they must be practical: they
must decide our conduct. Now the kind of judgments which decide
our conduct are the actual judgments we make in particular cases,
the judgment that this A is better than this B, here and now, and for
me. The universal moral judgment that any A, which is a greater
pleasure than any B, should be preferred, decides no one’s conduct,
for in particular situations, wherein we act, we do not find any A
and any B, but this A and this B, and the whole question is whether
we like this A better than this B. And although the universal judg-
ment, that the greater pleasure should always be preferred to the
less, is true for anyone, the particular judgment that this pleasure is
greater than that may be true only for me, and certainly need not be
true for everyone. Hence, the particular judgment, which must al-
ways carry the qualifying words “for me,” is strictly an opinion,
guiding only my own conduct, and if true in any sense at all, true
only for me in this situation. But such particular judgments are the
only ones which operate practically, and so, the student concludes,
for all practical purposes moral questions are decided only by opin-
ion. The moral skeptic is right, and the moralist wrong.

Much that the student has said is right, and yet his conclusion is
wrong. Let us concede at once that, so far as our discussion has
gone, all particular moral judgments, which express an individual’s
preference for A over B because more pleasing to him or her in the
light of all calculable circumstances, are subjective, are opinions
true for that individual only at the time they are made. Let us, fur-
thermore, admit that such particular judgments are the most practi-
cal in the sense that they directly determine a choice and ensuing
conduct. But instead of saying that they are the only really practical
judgments, and that universal judgments are not practical at all, let
us see if we can show the student that the universal judgments are
also practical, though in a sense not so obviously or directly.
 33
Here are two people, facing the same alternatives under the
same circumstances. The two people differ as individuals in many
ways, and so whereas one likes this A better than this B, the other
likes this B better than this A. Now suppose the situation to be com-
plicated by the fact that both A and B involve future as well as pre-
sent pleasures. What, then, does it mean to say that A is liked better
than B, or B better than A? It must mean that each person, according
to his or her individual nature, has made a different calculation here
of which is the greater-good-for-him-or-her. But, as we have already
seen, a person can act contrary to such a calculation, and in so doing
violate the universal moral rule that the greater good should be cho-
sen. Hence, there are the following possibilities: (1) if both people
violate the universal moral rule, it can be truly said that each should
have made the opposite choice; (2) if the first person obeys the uni-
versal rule, and the second transgresses it, then it can be said that the
second person’s judgment is wrong, even though it now will agree
with the first person’s. The first person’s judgment is not right be-
cause this A in fact gives a greater pleasure than this B to anyone;
on the contrary, this B gives a greater pleasure to the second person;
so that if the second person had acted wisely in his or her own be-
half he or she should have chosen B rather than A.

What this all comes to can be summarized simply enough by

pointing out that the act of preference follows from two judgments,
not from one, a universal judgment and a particular judgment. With
respect to the universal judgment, a person can be objectively right
or wrong; thus, a person who says that a greater pleasure ought not
to be preferred—pleasure and the quantity of pleasure being the on-
ly criteria of preference—speaks as falsely as a person who says two
plus two does not equal four. With respect to the particular judg-
ment, a person can only be subjectively right or wrong, according as
they correctly or incorrectly calculates what, for them in this situa-
tion, is the greater pleasure. Their being right in the particular judg-
ment has no relevance to the choices of other people; whereas their
being right in the universal judgment indicates what is right for eve-
ry other person.

But, the student persists, how does the universal judgment have
any practical bearing? The question can be answered in two ways.
The first is difficult to imagine, though possible: the case of a per-
son who actually was in error about the universal principle, who
somehow thought that the greater pleasure ought not to be preferred.
 34
Such a person, however accurately they calculated their present and
future pleasures in any particular situation, would, if they put their
universal and their particular judgment together into practice, make
a choice which could be called wrong—and objectively so, in the
sense that it was not only wrong for them, but wrong for any person,
because their error lay in an erroneous general principle.

The second case is one we have already discussed: the case of

the person who violates the true universal rule as a result either of
wrong calculations in this particular situation, or as a result of not
following the calculations according to the prescription of the uni-
versal rule. Whichever of these two things they do, their preference
can also be objectively criticized. It was wrong not only for them,
but for any person in the same situation. These facts indicate con-
clusively that having the right universal rule and, more than that,
applying it accurately to the circumstances, and, even more than
that, putting the combination of the universal and the particular
judgments into practice, are indispensable conditions of reaching a
sound conclusion in the particular case. And any person who fails to
satisfy all of these conditions can be criticized objectively, as he or
she could not be if the only factors which determined actual prefer-
ences were entirely subjective.

If that is so, the student then asks, why did you admit earlier in
this discussion that one person can prefer this A to this B, and an-
other prefer this B to this A, and both be quite right? Was not that
admission tantamount to conceding the subjectivity of actual prefer-
ences? Again, we must repeat that actual preferences, expressed in
the particular judgments which immediately precede action, are sub-
jective in the sense indicated, namely, that two people can make
opposite judgments in the same situation and still both be right. The
only point the student failed to see, when he asked the question, was
that these opposite judgments are not entirely subjective, for both
can be wrong if both were reached in the wrong way, i.e., in reliance
upon a false universal rule, or in violation of a true one, through
miscalculation or willful transgression.

We have now arrived at a point favorable for summarizing our

discussion so far. Let us submit this summary to the student for his
approval before we go on.
 35
There are two extreme errors which are equally wrong. (1) The
error of the moral skeptic who says that actual preferences are en-
tirely subjective, that there is absolutely no way of pointing out to
a person that he or she is wrong in a particular moral judgment in a
manner which would make any other person wrong in the same
situation. (2) The error of the moralist who says that actual prefer-
ences are entirely objective, that there is absolutely no way in
which a person can regard their particular judgments as right for
them and for themselves alone, since if they are right at all, they
must be right for any other person in the same situation.

The truth, which corrects these errors, can be succinctly summa-

rized in the following propositions: (1) two people can make oppo-
site preferences in the same situation, and both be wrong; (2) two
people can make opposite preferences in the same situation, and
both be right. And if there is any moralist who makes the error just
described, the moral skeptic is thoroughly right in attacking them. It
may even be that the student has been led to espouse moral skepti-
cism because of the error he has attributed to the moralist. Once the
student is told that this error is no part of the moralist’s position, a
stumbling block may be removed. So far as we have gone, the mor-
alist’s attack upon skepticism can be justified only with respect to
the error that is a blemish on the skeptical position, just as much as
the opposite extreme error is a blemish on the position of the moral-
ist. With both errors removed, the moralist and the moral skeptic are
drawn a little closer.

With both errors removed, what can teacher and student (or
moralist and moral skeptic) now positively agree upon? If they will
examine together the two truths, stated above as corrections of the
two extreme errors, they will find an explanation for these truths.

On the one hand, the reason why two people can make opposite
preferences in the same situation, and both be wrong, is that each
can violate in his or her own way a rule that is equally obligatory on
both. That there can be any universal moral truths at all, such as the
rule for always preferring the greater pleasure, arises from the fact
that, in so far as they are human, all people are the same, at any time
or place.

On the other hand, the reason why two people can make oppo-
site preferences in the same situation, and both be right, is that both
 36
are not simply human beings, for each is a uniquely differing indi-
vidual person, whose individual nature, constituted by the accidents
of birth, biography, and environment, belongs to him or her alone.
That two people, both adhering to the same universal moral rules
and following them equally well, should be able to reach different
conclusions arises from the fact that they differ as individuals; and
the rightness of their opposite conclusions is a rightness relative to
their individual natures.

In short, whatever is universally true or objectively right in the

making of a particular moral judgment is something relative to the
human nature common to all people; whereas whatever is only indi-
vidually true or only subjectively right in the making of such a
judgment is something relative to the individual nature uniquely
possessed by each person .

Now the moralist can claim to have moral knowledge, in the

strict sense of objectively true moral principles or rules, only on the
level of universal judgments. If he claims more than this, the moral
skeptic is right in opposing him. The moral skeptic, on his side, can
claim that moral judgments are subjectively true, or mere opinions,
only on the level of particular judgments. If he claims more than
this, the moralist is right in opposing him. The fact that the particu-
lar judgment is the one which is directly proximate to action does
not mean that the universal judgment is not practical, for it is indi-
rectly practical in so far as it is operative in the formation of the par-
ticular judgment. And although the particular judgment, taken as a
whole, is subjective and has the status only of opinion, it contains
implicitly the universal judgment which has been operative in its
formation. It is necessary, of course, to extricate this universal
judgment and to make it explicit, in order to discover a moral prin-
ciple which has objective truth, obliging all people, and applicable
to every situation.

There should be no difficulty about getting the student to ap-

prove this summary, for it says no more than what the student him-
self had admitted in the course of the preceding discussion. Making
it, however, enables us to make two further points. The first looks
backward. If the student, as a moral skeptic, still holds that although
all moral standards are not individual, they are at least all conven-
tional (relative to a social group at a given time and place), we can
now begin to suggest to him that just as what is individual in moral
 37
judgments, because they are made by individual people, does not
exclude the possibility of a universal element, because individual
people are also all human beings, so what is conventional in moral
judgments, because they are made by human beings living under
certain social conditions, does not exclude the possibility of a uni-
versal element for the same reason, namely, that despite every dif-
ference of social origin, the people of different societies are still all
human beings. We can promise the student to return to this point
later, and show him, after a larger number of moral truths have been
discovered, that these moral truths not only hold for every individu-
al, but for every society as well; and that there is no inconsistency
whatsoever between the unity and absoluteness of moral principles,
on the one hand, and the plurality and relativity of mores, on the
other.

The second point looks forward. It will be made by the student

himself, after he has reviewed the ground we have so far covered.
We have claimed, he will say, to have established the existence of
moral theory, as a body of knowledge rather than a set of opinions,
by getting him to admit the truth of one, or at most two, universal
judgments, such as “men ought to prefer the greater pleasure.” But
if that is all that moral theory comes to, morality is not a very im-
pressive body of knowledge. What other moral truths can we show
him, and induce him to accept as such? If there are none other than
this one, or its like, he does not regret his indifference to the study
of moral philosophy, for at best it consists of the most obvious
common sense, which all people already possess, and even at that its
offering of acceptable truths is hardly elaborate enough to be worth
more than a page, or the back of a card.

The challenge is utterly fair. We are now prepared to meet it.

But, first, we must remind the student that we did not spend all this
time on the principle, that people should prefer the greater pleasure,
for its own sake, but rather for the sake of getting him to recognize a
universal principle, a true but violable precept. And we had to do
that in the student’s own terms, by accepting at the outset his own
answer to the question, Why is anything preferable to any other? He
told us that the only criterion was pleasure as against displeasure;
and then added a second criterion, the quantity of pleasure. At the
time, we did not question these criteria. But now we can tell him
that the paucity and obviousness of the principles we have so far
 38
reached are due to the two criteria of preference which he claimed
were the only ones.

Now that the first stage of the argument is completed, and he

admits the existence of some universal truths, we can go further on-
ly if he will permit us to re-examine the original premises of the ar-
gument. They were not entirely wrong: pleasure and quantity of
pleasure are criteria of preference. But, though not wrong, these cri-
teria are inadequate. There are other and more fundamental criteria
which, when seen, will not only bring us to the induction of much
more significant moral generalizations, but also will significantly
alter our understanding of the two criteria already discussed. In or-
der to correct the error of supposing that the only criteria of prefer-
ence are pleasure and quantity of pleasure, we must make a fresh
start. The best way to do this is to re-examine some of the state-
ments already made about pleasure, for in them much truth is con-
tained that we have not yet seen.

CHAPTER III: PLEASURE AND THE ORDER OF GOODS

T O MAKE A FRESH START, let us ask about the meaning of

pleasure as a criterion of preference. Precisely what does the
student mean when he says that he prefers A to B because A
pleases and B displeases him, or because A pleases him more than
B?

The student may be somewhat bewildered by this question, for

he has already told us that such judgments as “A pleases me more
than B” are equivalent to saying “I like A more than B.” In fact, he
confesses, much of the discussion we have had so far has seemed to
him to consist in making verbal substitutions of this sort. We started
out by admitting that the fact of preference was equivalent to the
judgment of “A-better-than-B-for-me” and that this in turn became
equivalent to two other forms of statement: “A pleases me more
than B” and “I like A more than B.” What has been gained by say-
 39
ing the same thing over and over again in different words? Pleasure
and displeasure, it would seem, do not explain the fact of prefer-
ence; far from explaining it, the fact of being pleased (or displeased)
seems to be identical with the fact of preferring (or not preferring).

One thing the student says is false, but one thing is true. The
falsity arises from his failure to remember that something was
gained by introducing the notion of pleasure into our discussion.
That, with the addition of considerations about quantity of pleasure,
enabled us to formulate a universal rule of conduct, which he him-
self admitted should direct people’s choices, though in fact people
do not always choose as they should, according to this rule. This
very discrepancy, between what should be and what is, certified the
character of the rule we formulated as moral rather than descriptive.

The truth in the student’s remarks was his observation that

pleasure does not explain preference. That is precisely why we are
now engaged in re-examining the connection between preference
and pleasure. And the first thing we must try to discover is whether
pleasure is the object of every choice, or merely the result of every
choice which is successfully executed. We say “every” here because
pleasure does seem somehow involved in every act of preference. If
pleasure results from getting what we prefer, then pleasure cannot
be the cause of preference, since preference precedes execution, and
we may not always be successful in getting what we prefer. The
student should be inclined to agree with the conception of pleasure
as a resultant, since it was he who recently insisted that pleasure was
not a cause.

Perhaps, says the student, but then I also return to my original

insistence that there is no cause of preference. I only appear to ex-
plain my choice by speaking of pleasure, but preference is really
inexplicable. And anyway, he adds, I don’t see what difference it
makes whether pleasure is the object of choice, as against displeas-
ure, or the result of getting what I have chosen, as against displeas-
ure as the result of failure. Why can’t I say that I prefer something
because I anticipate the pleasure I shall derive from getting what I
want?

We must warn the student that in asking the last question he

used the word “because” and thereby relaxed his resistance to our
efforts at explaining preference. If he will stay relaxed for a moment
 40
longer, we may be able to get new light by following the lead of his
last question. He must admit that making a choice precedes carrying
it out in action, that deciding what one wants precedes getting it.
Everyone knows, furthermore, that people do not always get what
they want. Hence at the moment of choice, a person who has learned
anything at all from experience must acknowledge the possibility of
failure to possess what he or she has chosen and must therefore an-
ticipate the displeasure of failure as well as the pleasure of success.
Even though people may be able to calculate the probabilities of
success and failure in particular cases, and even though it is true that
people sometimes avoid choosing things they really want because
they wish to avoid the displeasure of likely failure in seeking what
is a little beyond their present reach, pleasure and displeasure as
anticipated resultants of successful or unsuccessful seeking are, at
most, only one factor in the determination of every choice.

The student was right in supposing that a person might prefer

something because the person anticipated the pleasure to be derived
from getting what he or she wanted; but he was wrong if he sup-
posed this to be the only cause, for it is now also evident that unless
a person preferred this thing to that, the person would not be pleased
to get it, nor could he or she therefore anticipate the pleasure of suc-
cessful seeking. The fundamental truth, which is slowly becoming
apparent, is that the object of our preference is never the same as the
satisfaction we experience in getting the object we prefer. Pleasure
may be the object, or it may be the satisfaction, but it cannot be both
without treacherous ambiguity in our use of words, and if it is not
both there must be other factors than pleasure in the explanation of
preference.

Or maybe preference cannot be explained, the student reminds

us. But even if I waive that alternative to permit this discussion to
go on, the student says indulgently, I am now at a loss to understand
many points we have already agreed upon. Didn’t we agree that A is
preferred to B when A is pleasing and B displeasing, or when A is
more pleasing than B? Doesn’t that mean that A is a pleasure or a
greater quantity of pleasure, and is not A the object of my choice
when I prefer it, whether or not I succeed in getting it? If all this is
so, then why can’t I stick to my original statement that pleasure and
displeasure, or unequal quantities of pleasure, are the only objects
between which people choose when they exercise preferences?
 41
The student’s questions cannot be answered without begging
him to be more attentive to words, for unless we now clarify our
language we cannot accurately express our thought. At one moment,
the student said “A is pleasing” and at another he said “A is a pleas-
ure.” Pointing this out to him, we must ask whether it makes no dif-
ference which we say. If he replies, as he is likely to, that he sees no
difference here, we must try to explain, for upon the discernment of
this difference much depends.

Let us begin by reminding the student that, at the very opening

of our discussion when the fact of preference was first introduced,
we pointed out that the only people who could say they never pre-
ferred anything would be people who had never experienced desire
of any sort. And we said: if people “admit having had the experi-
ence of desire, they can certainly be made to understand the differ-
ence between something which would satisfy that desire and some-
thing which would not. They can at least imagine a situation in
which, given a certain desire, they would prefer one thing to anoth-
er.” Let us now call the thing they prefer the object of their desire.
The object-of-desire is certainly not the same as the desire itself, nor
is either of these the same as the satisfaction of the desire which oc-
curs when the object is attained.

There are three terms, then, which any careful analysis of pref-
erence must distinguish. They are irreducible to one another. And it
is in the light of this fundamental distinction that the student can be
made to see the difference between saying “A is a pleasure” and “A
is pleasing.” The latter statement means that A is pleasurable or a
source of pleasure. The former statement means that A is itself iden-
tical with pleasure. But if A is both, then we are saying that that
which is itself pleasure is a source of pleasure. If A is not both, then
we must decide which A is, and upon this decision will depend
whether we regard pleasure as the object of desire, or as the satisfac-
tion which results from attaining the object of desire, for the object
of desire is, when possessed, the source of satisfaction.

But, says the student, why cannot pleasure be both object and
satisfaction? And even if we decided that pleasure was always one
and not the other, what difference would it make?

The difference it would make is great. For if A is not itself a

pleasure, and B a displeasure, then A and B as the objects between
 42
which preference is exercised must have some other determinate
character. Let us suppose that A stands for wealth, or a course of
action leading to its acquisition, and B stands for health, or a course
of action leading to its preservation. Many people have been faced
with these as alternatives to choose between. Let us further suppose
that we use the words “pleasure” and “displeasure” to name the sat-
isfaction and dissatisfaction of desire. Then the reason why, for a
given person, wealth may be more pleasing than health, or converse-
ly, is that he or she desires it more. We obviously cannot say that
the person desires it more because it is more pleasing, for unless the
person initially desired wealth more than health, the person could
not anticipate being pleased or satisfied if the course of action he or
she pursued eventuated in its acquisition even at the expense of loss
of health.

Should we make the contrary supposition, however, that pleas-

ure and displeasure are the objects of desire, rather than its satisfac-
tion and dissatisfaction, then we can return to our original explana-
tion of preference, namely, that we desire A more than B because A
actually is a pleasure and B a displeasure, or because A is a greater
pleasure than B. Here we do not say that A is a greater pleasure be-
cause we desire it more, but rather that we desire it more because it
is a greater pleasure. The crucial question, in short, is whether desire
is to be explained in terms of pleasure, or pleasure in terms of de-
sire. If the latter is the case—and the student himself seems to have
rejected the former in his earlier remarks about the failure to explain
preference by identifying the preferred object with pleasure—then
we must push further to explain why one object is desired more than
another.

But, says the student, I still don’t see why pleasure cannot be
the object of desire, as well as its satisfaction. I see nothing wrong
in saying that I desire or like pleasure and that pleasure pleases me.
Before you go on to any further explanations, I’d like this point
cleared up.

The student’s insistence is justified, for there is a meaning of the

word “pleasure” in which it does name an object of desire, and our
whole problem here is to distinguish that meaning from another
meaning of the word in which it names every satisfaction of desire.
Once this basic ambiguity of the word “pleasure” is eliminated, and
two quite distinct notions are distinctively expressed, we shall be
 43
able to proceed. It should be noted at once that “pleasure” cannot be
used to name every object of desire, but only one sort of object
among many others; in contrast, “pleasure” can be used to name
every experience of satisfaction.

As objects of desire, wealth and health are not the same as

pleasure, although wealth and health can be pleasurable, i.e., they
can be sources of pleasure in the sense that when possessed they
satisfy the desire which led us to seek them. As pleasurable (i.e.,
pleasing, a source of pleasure), pleasure as an object of desire is no
different from health and wealth, for every object of desire is pleas-
urable. But to say that every object of desire is pleasurable in this
sense is not to say that every object of desire is pleasure. If one were
to say that pleasure is the only object of desire, one would be deny-
ing that such things as wealth and health are desirable objects.
This denial is not avoided by saying that wealth and health are
desirable only because they are pleasurable, for, in the first place,
that would apply to pleasure itself as an object of desire; and in the
second place, it would amount to saying that an object of desire is
desired because it will satisfy the desire when possessed. Since this
applies to every object of desire, it cannot explain the preference for
one over another; hence if wealth is preferred to health, it must be
due to some difference between wealth and health as diverse objects
of desire. Nor will quantity of pleasure help us here, for to say that
we find wealth more pleasurable (i.e., a source of greater satisfac-
tion) than health is to say no more than that we desire it or like it
more; and we still have to explain why we do or should, in terms of
something about the nature of these two objects, in themselves and
in relation to ourselves.

Finally, pleasure itself as an object of desire is sometimes op-

posed to other objects, so that we are forced to choose between
pleasure and other things. Thus, the person who seeks to gain great
wealth must often forgo pleasure, as the person who seeks certain
pleasures often sacrifices health in the process. In both these cases,
pleasure can be regarded as an object of desire, competing with oth-
er objects which, as such, are simply not pleasure at all. Let A stand
for pleasure as an object, B for wealth and C for health. Then to say,
in the first instance, that a person prefers B to A is not to say that B
is the greater pleasure, for it is not pleasure at all; it is rather to say
that B will give the person more pleasure, in the sense of more satis-
faction, because the person desires it more. Similarly, in the second
 44
instance, to say that A is preferred to C is to say that pleasure is
more pleasurable than health, i.e., it will give greater satisfaction
because it is more desired. And since to say that “pleasure is more
pleasurable” is to say that “pleasure will give more pleasure” we are
here plainly confronted with the ambiguity of the word “pleasure.”
It cannot mean the same thing when it names one object of desire
(obviously one, since there must be some other object which gives
less pleasure), and when it names any satisfaction (obviously any,
since both objects give pleasure though in different degrees).

Not only is the ambiguity of the word “pleasure” thus revealed,

but we can now help the student to understand what sort of object is
named by “pleasure” when it is used in that sense. Pleasure as an
object of desire is a bodily condition, the opposite of which is the
bodily condition known as pain. For want of better words, let us
refer hereafter to sensual pleasure and sensual pain. Using words
this way, we are certainly reporting the facts of human preference
when we say that sensual pleasure is only one of the objects people
desire, or that people often prefer other objects to sensual pleasure,
or that some people actually prefer sensual pain because, under
pathological conditions of desire, they derive greater pleasure from
it. Furthermore, to call pleasure (as object-of-desire) sensual does
not mean that pleasure (as satisfaction-of-desire) is inexperiencea-
ble. We experience satisfaction and dissatisfaction as states of desire
itself, but not as directly sensed conditions of our body as a whole
or of its members. A satisfied desire is experienced as one which no
longer impels us to action; a dissatisfied one remains a motivating
force. If the student finds this account of the two meanings of
“pleasure” satisfactory, we can now return to the problem of prefer-
ence and see how this clarification helps us.

The student will certainly concede that the ambiguity of the

word “pleasure” has been sufficiently demonstrated, and he will
probably admit that the suggested distinctions in meaning are genu-
ine. He may even agree that the facts of human preference cannot be
accurately described unless sensual pleasure (or sensual pain), as
one among many objects of desire, is distinguished from what
makes us regard either sensual pleasure or sensual pain as more or
less pleasurable than other objects, namely, the strength of diverse
desires and the resultant degrees of satisfaction to be obtained. And
here the student makes one last effort to hold the position he once
took—that pleasure, or quantity of pleasure, are the only explana-
 45
tions of preference. He tells us that he will use the word “pleasure”
as equivalent to “satisfaction-of-desire,” and using it this way, he
claims that the general rule of conduct we have already formulated
remains unaltered. That rule was: “In any case in which a choice can
be made, people should prefer the alternative which, in the long run,
or viewing life as a whole, maximizes pleasure and minimizes dis-
pleasure.”

We must inform the student that it is not our intention to argue

against the truth of this rule, but rather to criticize its insufficiency
as a guide for human conduct. We must remind him that it was he
who complained about the barrenness of moral knowledge if it went
no further than this single universal rule. It was precisely in order to
answer his complaint that we have been trying to show him that
pleasure, taken in either of its senses, cannot account for preference.
That being done, we may then be able to discover the real criteria
which determine what a human being should prefer, and in terms of
these criteria formulate more specific rules of conduct.

If the only rule of conduct were the one we have so far formu-
lated, the student would be right, for the most part, in maintaining
his moral relativism, and his skepticism about moral knowledge;
different people might abide by this one rule and yet in every partic-
ular seek different things or make different choices. So far as this
rule goes, it does not prevent us from supposing that one person
could maximize pleasure by a set of actual choices quite different
from those made by another person following the same rule; one
person might always prefer wealth to sensual pleasure and honor,
and another always prefer virtue to fame and fortune, and yet it
would be conceivable that both could maximize pleasure in the
sense of satisfying their differently oriented desires. When we say
that pleasure is insufficient to explain preference, we mean, of
course, not merely that it is insufficient to describe the fact of pref-
erence, but more fundamentally that, unless we go beyond pleasure,
we can never say, of two objects different in kind, which should be
preferred. Failing to do this, we fail to establish a practically signifi-
cant body of moral rules, both universally valid and also violable.

If I understand you, says the student, you are at last agreeing

with me. Pleasure being the only criterion, there is no moral
knowledge worth bothering about, certainly no set of rules which
would direct all people to follow the same general course of life.
 46
And I am now surer of this than I was before the discussion started.
Distinguishing the two meanings of pleasure has helped to make it
clear. For considering pleasure, in the first sense, as one object of
desire, there appears to be no reason why people should concur in
preferring it, or not preferring it, to other things. And considering
pleasure, in the second sense, as equivalent to the satisfaction of
every desire, all people do in fact concur in desiring as much satis-
faction as they can get, but this fact does not obligate them to agree
in preferring one sort of object to another. On the contrary, accord-
ing as different people have different desires, it would seem as if
they had to exercise quite different preferences in order to maximize
pleasure in the sense of satisfaction.

Unless we can correct two errors which the student has made,
we are barred from proceeding. The first may have been a slip of the
tongue. The student spoke of “people’s desiring as much satisfac-
tion as they can get.” This statement seems to regard satisfaction as
an object of desire, which is strictly impossible. If satisfaction were
an object of desire, then satisfaction would result from fulfilling
such desire, but the resultant satisfaction could not be the same as
the satisfaction which, being desired and then possessed, gave rise
to it. And there would be nothing to prevent the second satisfaction
from being in turn an object of desire, thus giving rise to a third sat-
isfaction in the same way, and so on in an endless progression. To
make satisfaction an object of desire is, paradoxically, to condemn
desire to endless dissatisfaction. Satisfaction, then, can never be an
object of desire; nor can it ever explain why we desire one object
rather than another, since given the desire for either object, its pos-
session produces satisfaction.

The student’s second error was his failure to note that our dis-
cussion has expanded to take in two new factors, namely, objects of
desire other than sensual pleasure and pain, and a variety of desires
of different strength. Pleasure is no longer the only criterion of pref-
erence; in fact, as object, it is only one among many things to be
chosen; and as satisfaction, it is entirely insufficient as a criterion,
since what will satisfy us depends upon our desires.
To make this clear, let us now introduce the word “good” to
name any object of desire. The relation between good and pleasure
is at once clear: sensual pleasure is a good, but not the only one; and
every good is a source of pleasure in the sense of satisfying a desire
when possessed. Hence, the earlier formula, that A is preferable to B
 47
whenever A is more pleasurable than B, must now be restated as
follows: treating A and B as goods, both of which are desired, A is
the greater good, and hence preferred, whenever the desire for A is
greater than the desire for B. In short, the good is the desirable, and
the better of two goods is the more desirable.

No, says the student, your last way of putting the matter is mis-
leading. You have made it sound as if one object were in fact better
than another, and your desire was determined accordingly; whereas
so far as you have been able to show, one object is better than an-
other only in so far as it is the object of a stronger desire. Thus you
have not escaped the criterion of pleasure, since the preferred object,
as the object of the stronger desire, is always the more pleasurable.
Unless you can explain why all people should desire one object
more strongly than another, you cannot avoid subjectivity and rela-
tivism. And how will you be able to show what people should desire
and what they should prefer, unless you can show that the objects
themselves are intrinsically good and bad, better and worse?

The student’s challenge is fair. We have succeeded in showing

him that pleasure will not explain either what people do prefer or
what they should prefer, but we have not yet succeeded in establish-
ing other criteria which are both adequate and objective. Some pro-
gress has been made, however, in so far as the student will now ad-
mit that there are a variety of goods, different in kind, where before
he insisted that there was only one good, pleasure.

What do you mean by a variety of goods? the student interrupts,

and whence comes this variety?

To answer these questions, let us examine the facts of life. For

the moment we shall be content to enumerate the different sorts of
objects which people do in fact desire. They desire food and drink,
clothing and shelter. Each of these is a kind of good, just as sensual
pleasure is a kind of good. We are here enumerating different sorts
of objects which people in fact desire, and of each sort there are, of
course, particular instances. Thus, “food” names a class of objects,
including not only many subordinate varieties, but ultimately this or
that particular item of food—this slice of bread, that slab of butter.
 48
But, the student interrupts again, how do you know whether a
particular object belongs to one class or another? One person may
desire this thing as food, and another desire it as sensual pleasure.

No, that is not so. Remember that sensual pleasure is a certain

type of bodily condition. It is not the same, for instance, as another
type of bodily condition which we call health. Now, food is neither
sensual pleasure, nor is it health, but it may in fact be the cause of
either, and hence, it may be desired as a means to the one or to the
other. The student is quite right in anticipating the point that food
(and, perhaps, also drink, clothing, shelter and all similar objects)
are seldom desired for their own sake, but rather as means for ob-
taining other goods, other objects of desire, such as sensual pleasure
and health. The fact that one kind of good is usually desired as a
means for obtaining another kind of good does not obliterate the
distinction between the two kinds; for if it did, we could never dis-
tinguish between objects desired as means and objects desired as
ends.

Let us proceed with the enumeration, and make it briefer by

naming more general classes of objects. The student has helped us
to achieve this generality, for he has enabled us to see that all bodily
goods (including strength and rest, as well as health and sensual
pleasure) are of one large sort, just as food, drink, clothing, shelter
and all similar things are of one large sort which we can call wealth.
Wealth, it would appear, consists of all the physical things which
human beings can use for the sake of their bodily well-being—for
their health, sensual pleasure, etc. It includes everything the econo-
mist calls consumable goods and the instruments productive of
them, and it includes money as an economic instrument involved in
both the production and distribution of consumable goods. Now, in
addition to such large classes of goods as wealth and bodily well-
being, there are such things as friendship, social peace and security,
public honor, political status, and perhaps also fame and power. In
fact people do desire such things. Let us group them all together
under the head of social goods.

Furthermore, some people, at least, seem to desire knowledge of

various kinds and different sorts of skill. This group of goods re-
sembles the bodily goods in one important respect: when a person
possesses them he or she possesses them as an altered condition of
his or her own nature, whereas the goods of wealth, in contrast, are
 49
all external goods, existing actually apart from human nature. But
knowledge and skill do not exist actually apart from the human be-
ings who possess them, and even if they may be said to exist poten-
tially, prior to actual possession, they exist potentially in human be-
ings who have the capacity for developing them.

It is difficult to find a name for this new class of goods. Despite

their resemblance to bodily goods, they must be distinguished there-
from. The student would probably object to their traditional name—
goods of the soul. Let us, therefore, call them habits, for the student
will agree that skill in doing any sort of operation is an acquired
habit. If the skill were native rather than acquired, it could not be an
object of desire. Knowledge, like skill, is something we acquire,
something we possess as a result of our own activity. Hence, for the
time being, let us regard knowledge as a habit also. Certainly the
student will admit that most people desire to be educated, and edu-
cation is the process whereby people are helped to form habits of
various sorts. The common desire for education can, therefore, be
interpreted as the desire for a class of goods we have now grouped
together as habits. If we ask why people want habits, such as
knowledge and skill, the obvious answer is that they can act more
efficiently as a result of possessing them. Hence, efficient activity
must be still another kind of good, since whenever one kind of good
is desired for the sake of another, the latter must also be regarded as
a kind of good.

Without claiming that this enumeration is either precise or ex-

haustive, we can now ask the student whether he will accept the five
types of goods we have named (viz., wealth, bodily goods, social
goods, habits, activity) as a rough indication of the variety of goods
which people do in fact desire.

Yes, says the student, people do in fact seem to desire all these
objects, and I will admit that it is possible to divide them into the
groups you have named. But I am not sure I understand why there is
this variety of goods; or to put my question another way, is there
any reason why this variety is the same for all people? Unless it is,
you are not going to be able to show that all people should exercise
the same preferences in choosing among goods of these various
sorts. And even if it is, an objective ordering of these goods still
remains to be shown, for people do in fact seem to make quite dif-
ferent choices—some people desire health and knowledge for the
 50
sake of wealth and power; others desire wealth for the sake of sen-
sual pleasure and fame; and there may even be some who desire
wealth and the social goods for the sake of habits and efficient activ-
ity.

One thing at a time. Let us first explain to the student why some
such variety of goods is the same for all people. The first part of the
answer should be obvious at once. As human beings, having human
nature, all people are the same, even though they differ in many
subordinate ways as individuals. The deeper question, however, is
why there is a variety of goods, not why it is the same for all people.
If the good is simply any object—whether an external thing or an
aspect of human nature itself—which a person desires, then the plu-
rality of diverse objects, which we have classified as a variety of
goods, must be due to a plurality of diverse desires. What the stu-
dent really wants to know is why all people should have the same
set of desires. We cannot rely upon the fact that all people do have
the same plurality of desires, for the fact may be questionable, and
even if it is not, the student is justified in asking why he, for one,
should not make an exception of himself and limit his desires to
fewer objects. If there is no reason why he should not do this, then
regardless of the facts about what most people desire, the variety of
goods is a subjective, not an objective, enumeration.

There is a further crucial consequence: the objects we have

called goods are good only because they are desired. Hence there is
always a relativity of the good to actual desire, and we shall never
be able to say what people should desire, which is central to moral
knowledge as normative or prescriptive. In order to get beyond a
mere description of what human beings do desire, we must some-
how show the student that the objects human beings desire, they
desire because they judge them to be good. Paradoxical though it
seem, we must begin to do this by getting the student to admit one
fact: all human beings desire to live.

Yes, says the student, I’ll admit that fact. Even if there were ex-
ceptions, it would certainly be true that a person who does not desire
to live desires nothing else, and for him or her there is no further
problem.

Will you admit one further thing? we must ask. Will you admit
that all people desire to live well, or as well as possible?
 51

Yes again, says the student, although I am not sure I know what
is meant by “living well” nor do I think that all people would agree
about what living well consisted in. I’ll say Yes, therefore, if all you
mean is that every person wants as much satisfaction as he or she
can get. To say this is to say no more than what we have already
agreed upon—that every person wishes to maximize pleasure, or, in
our new terms, every person seeks the utmost satisfaction of which
he or she is capable.

In saying this, the student has helped us to our conclusion.

Though perhaps inadvertently, he has introduced an indispensable
notion, that of human capacity. If living well consists in fulfilling a
person’s capacities, (and in so far as these capacities are the same
for all people because they are rooted in a common human nature),
then it follows that whatever objects are necessary to accomplish
such fulfillment must be desired by any human being who desires to
live well. And such objects are no longer to be called good simply
because they are in fact desired; we can now see that they are good
because it is necessary to desire them if one desires to live well.

We can say people should desire whatever is necessary for

achieving what they do in fact desire—namely, a good life. And the
objects they should desire, as means to the end they do desire, are
good, not because they do desire them, but because they are means
to the desired end. If the end is living well, we can say that the five
kinds of good we have named are all objectively good because they
are indispensable means. People should desire them if they seek to
live well; if in fact they do not, they are clearly in error. That people
can make such errors, for one cause or another, indicates the viola-
bility of this prescription and verifies its character as a moral rule, a
rule as universal as the commonness of the desire for a good life,
and the commonness of human nature as the root of certain capaci-
ties to be fulfilled.

We need not pause here to show in detail how the variety of

goods enumerated corresponds to the diversity of capacities to be
realized. In general, it should be clear that living consists in activity,
that the capacity for activity is more fully realized according as we
are able to act more efficiently, that habits are the immediate condi-
tions of such efficiency, that bodily and social goods are its remote
conditions, and that wealth is indispensable to the maintenance of
 52
bodily well-being. Or, to put it another way, we have capacities for
health and sensual pleasure, for social and intellectual activity, for
work and play—and the variety of objects enumerated corresponds
to these capacities. They are good for this reason, and we should
desire them accordingly. In short, the student now has the answer to
the question, whence comes the variety of goods? It comes from the
variety of capacities which human beings can fulfill, and which they
should fulfill in order to live well.

I may have helped you make all these points by mentioning ca-
pacity, the student says, but you have gone much further than I can
follow. I am not quarreling with your point of view that, in general,
human capacities are the same—so far as they are rooted in a human
nature which is the same. But, remember, I did not agree that all
people meant the same thing by such words as “living well.” Even if
the variety of goods is the same for all people in some sense, the
fact remains that different people place different values on the vari-
ous goods; and although you may have shown that all people should
desire them, you have not shown that all people should concur in
desiring them in the same way—with the same emphasis, to the
same extent, in the same order. And you must show this, since you
have to admit the contrary fact—namely that people do in fact differ
in the way they exercise desire with respect to the same variety of
goods. My guess is that they differ because they mean quite differ-
ent things when they all admit they want to live well. Furthermore,
if the various goods we have been talking about are objects we
should desire because they are indispensable means to the end we
do desire, they why should we desire the end itself? If you tell me
that there is no point to this question, and we can rest in the fact that
we do desire something as an end, then I say, in terms of all your
reasoning so far, that the end is not desired because it is good, but
rather good only because it is desired. Unless you can show it is good
apart from being actually desired, you cannot show that people should
desire it.

The student has accurately indicated what remains to be seen.

Though there is some ground yet to be covered, we have come a
long way from the initial suppositions of our discussion. Let us
summarize the advances we have made: (1) we agree that pleasure
does not, in either of its two senses, explain the fact or justify the
exercise of preference; (2) we agree that an object is good when it is
desirable, not simply when it is actually desired, and that it is desir-
 53
able as somehow related to the fulfillment of human capacities; (3)
we agree that there is a variety of such desirables, i.e., goods which
should be desired by all people, because they are indispensable as
means to an end all people do in fact desire, namely, to live well; (4)
we agree that this variety somehow corresponds to the variety of
capacities common to human nature, and that the diversity among
our desires is determined by the diversity of desirables, or goods. In
terms of all this, we have been able to formulate a universal (and
quite violable) moral rule: all people should desire every sort of
good which is an indispensable means to a desired end.

But, as the student rightly points out, three questions remain. (1)
Why should any end be desired, simply as an end and not as a
means? (2) Why should all people desire the same end, not only
verbally named in the same way, by such a phrase as “living well,”
but really understood in the same way. (3) Why should all people
desire the means (consisting of whatever kinds of goods should be
desired for the sake of the end) in the same way—i.e., in the same
order, with the same emphasis upon each kind, etc.?

Since anything which is desired must be desired either as a

means or as an end or as both, our analysis of goods or desirables
will be complete—if only in a general way—when we succeed in
answering these three questions, for then we will know why any-
thing at all should be desired, either in itself or in relation to some-
thing else. And since the problem of preference is concerned with
the reasons for choosing between one sort of good and another, as
alternative means to some end, the problem will be completely
solved when we know the order of all the goods which are means to
an end which should be the end all human beings seek.

The first question to answer is the one about the end. Beginning
with the fact of preference, our discussion began with a considera-
tion of means—alternative goods between which choice must be
exercised. But now we see that we cannot solve the problem of any
preference unless we first solve the problem of the ultimate criterion
of all preferences, namely, the end, which is itself never preferred,
because it is not a good, opposed by alternative goods, but the good,
having no alternatives. It is necessary, therefore, to make another
fresh start.
 54

CHAPTER IV: THE ORDER OF GOODS AND HAPPINESS

WE MUST NOW BEGIN by asking the student whether he is willing to

agree that we cannot desire everything as a means. Some things we
may desire simply as means, and some as ends which are in turn
means to further ends, but must we not desire at least one thing
simply as an end, and in no sense as a means?

“I suppose so”, the student will probably say, “but I don’t see
why.”

The reason is not hard to find. You admit that whatever is de-
sired as a means is sought for the sake of its end, and unless the end
is desired, whatever may be a means to it is not desirable. Now alt-
hough an end is the last thing we actually achieve in the course of
our conduct (for if an end could be achieved before some of its
means, those means would be utterly dispensable), the end must be
the first thing we actually desire, for unless we desire the end, we
have no reason for desiring things which are good only as means to
it. Hence if every good which we regard as an end could also be re-
 55
garded as a means to some further end, and so on indefinitely, there
would be no beginning. Just as you cannot begin to walk in a defi-
nite direction unless you know where you are going before you
start, so you cannot desire anything all as a means unless you desire
something simply as an end.

“Granted”, says the student, “something must be desired as an

end. You seem to be implying that there is only one end for each
person, the same for everyone. Why cannot a person have several
distinct ends? And why cannot different people have different
ends?”

Let us first consider the problem of several distinct ends for a

single man. Suppose A and B to represent two objects, each of
which is desired for its own sake, simply as an end. Now either the
man seeks both ends together, as parts of a whole we shall call X, or
he chooses between them. But if he chooses between them, he is
exercising preference. And he must exercise this preference before
he desires any other goods which are means either to A or to B, for
until his end is determined, he cannot select the means. But on what
ground shall he choose between A and B as ends? Since neither is a
means, he cannot decide between them as he might be able to decide
between alternative means to the same end. Therefore, if he prefers
one end to another, he must do so because—because—

“Because he likes it more”, the student answers, “and we are

right back to where we started. The preference for A over B (now
called ends) must be made as I originally said every preference was
made—in terms of a person’s likes and dislikes, in terms of the
quantity of pleasure to be obtained. Or, if you want me to avoid
such words as “pleasure”, I’ll say that the man will choose the end
which satisfies him most.”

It only seems as if we have fallen back; on the contrary, we have

made a great advance. Let us show this to the student by asking him
whether if a person could get some satisfaction out of having A, and
some also out of having B, would the person not get more satisfac-
tion out of having X, consisting in the sum of A and B. In which
case, if the criterion which determines a “choice of ends” (in our
suppositious case) is the amount of satisfaction to be derived from
possessing it, then whatever gives the utmost satisfaction should
always be chosen. And if, again in our suppositious case, A and B
 56
are the only two possibilities as ends, then neither can really be the
end, for the end must be X, the whole which includes them as parts.
If a person seeks either A or B when he or she can seek X, the per-
son is seeking less satisfaction than he or she can have. And even
though A and B are ends, they must also be regarded as means: the
parts of a whole are means for getting the whole itself. Therefore,
there is only one thing which is the end for any person. It is always
the totality of all the goods he or she is capable of possessing, and
possessing each derives some satisfaction therefrom. Complete sat-
isfaction occurs only when the totality is somehow possessed. The
end, therefore, is always the whole of goods; the parts of this whole
are the different kinds of goods, each of which as a part is a consti-
tutive means to the whole—constitutive in the sense that it is a
means whereby the whole can be constituted.

The constitutive means, or we can call them partial goods, may

be related to one another as means are to ends, but all of them, in so
far as they are parts, are equally constitutive means with respect to
the whole. That is why A and B as partial goods may be ends in re-
lation to other partial goods (C and D, let us say) serving them as
means. But in so far as they are partial goods, neither A nor B can
be the end, and both together, along with C and D, and all other par-
tial goods, must be regarded as constitutive means with respect to X,
which now stands for the complete totality of goods or desirables.
Hence there is always only one real and ultimate end, never desired
as a means, and since there is only one, it can never be the object of
choice or preference. It must be desired, because there is no alterna-
tive to it.

“You are going too fast for me”, the student confesses. “I am
particularly bothered by two things. One is that you seem to be as-
suming that all the partial goods are always compatible with one
another, so that it is always possible for a person to have all of them.
The other is that you seem to be reverting to an earlier position, or
even worse than that, you seem to be saying the opposite of what
you said before. You said before that satisfaction cannot be an ob-
ject of desire, and yet now you seem to be saying that X must be
taken as the end, rather than A or B, because it is more satisfying
than they are. Furthermore, if the totality of goods is always the end,
then it is silly to say that people should seek to achieve this totality,
since they cannot seek otherwise. In fact, I don’t see that we are say-
ing now any more than we said before when we said that a person
 57
should seek to maximize pleasure, or satisfaction—only it now ap-
pears there is no should about it.”

Let us meet the student’s first point by reminding him how the
diversity of partial goods is generated. There are many different
kinds of good because human beings have different capacities to be
fulfilled. Now, of course, it is possible for human capacities to be so
related that any attempt to fulfill one would necessarily interfere
with the fulfillment of one or more of the others. But that, as a mat-
ter of fact, does not seem to be the case. In the light of the facts
about human nature, we can say, then, that so long as the variety of
goods corresponds somehow to the diversity in human capacities,
this variety will include no incompatible partial goods. Hence, a
variety so constituted can always be summated in a totality. If A, B,
C, D . . . . N represents an exhaustive enumeration of partial goods,
each can be a part of the totality, X, for they can be taken together
as means to constitute that whole.

Before we leave this point, we can clarify another related mat-

ter. The student will recall that in the concrete cases of preference
which we considered earlier, the need for choice arose from the fact
that the alternatives were exclusive of one another. That always is
the case when we are faced with particular instances of the same
sort of good—this particular way of getting sensual pleasure, for
example, as opposed to that. It is precisely in such cases that the
criterion of quantity becomes operative—in fact, is indispensable.
As between competing particular goods of essentially the same sort,
a person should choose the greater—assuming, of course, that it is
right for him or her, all else being considered, to choose that sort of
good at all. But when the choice is between particular goods essen-
tially different in kind, the criterion of quantity no longer operates in
the same way.

In the first place, different kinds of good cannot be compared

with respect to quantity, for each yields a different kind of satisfac-
tion, in whatever amount that may be. In the second place, different
kinds of good are not competing but completing—if we have been
right in saying that each kind in a correctly enumerated variety of
goods is as indispensable as every other kind. This does not mean
that we are never forced to choose between particular goods of dif-
ferent kinds; it means rather that the criterion of quantity now oper-
ates only with respect to the end—the totality—and in view of that
 58
end, we must make this particular choice here and now in such a
way that we can ultimately obtain every kind of good, even though
at this moment we do that by giving up one particular good for a
particular good of another kind. The totality of goods is achieved
not by possessing every particular good, but goods of every kind.
We shall return to this point later in discussing the order of goods.

Let us now consider the student’s second point. He is right in

noticing a reversion to an earlier stage of our discussion, and even in
detecting an apparent contradiction of what was said before. We
said that the satisfaction derived from possessing a particular desir-
able object (a particular instance of a kind of good) could not be the
object of that desire, for then the satisfaction, not the object itself,
would be desirable. Nor can it be said that the object is desirable
only for the sake of the satisfaction to be derived, for, in the first
place, that generates an infinite regress of objects of desire; and, in
the second place, we have seen that a choice between two sorts of
objects cannot be explained or justified in this way. It should be
noted, therefore, that this earlier point was not relevant to the prob-
lem of preference in so far as it involved two kinds of goods as the
alternatives. But we have now seen that the amount of satisfaction,
without becoming the object of desire, can operate as a criterion
(though, perhaps, not the only criterion) in choosing between two
particular instances of the same sort of good. And when we come to
the question of the ultimate end, at which point there is no problem
of preference at all, the “amount” of satisfaction, now understood
heterogeneously, rather than homogeneously, as including every
kind of satisfaction, properly becomes the only criterion.

Since we have used the word “criterion” to mean that by which

we judge in our acts of choice or preference, it might be better to
say that the amount of satisfaction is the sign, rather than the criteri-
on, of the ultimate end, which is the totality of goods. The end
should be that which leaves nothing to be desired. The end, in a
sense, puts an end to desire by the completeness of the satisfaction
which results from possessing, not every particular good, which is
impossible, but every kind of good, which is quite possible.

It may be worthwhile to try to say this in another way. The end

can be described by the words “all good things,” which must be un-
derstood to mean a totality of kinds of goods, not all the particular
goods in each kind. The end, as a totality of goods, cannot itself be
 59
regarded as a good, for then the whole would belong to itself as a
part. If we refer to the end, thus conceived, as an object of desire—
and it can be an object of desire, though not of choice or prefer-
ence—we must refer to it as the good. And for every person there
can be only one object of desire which is the good, though there are
many objects of desire, subject to choice or preference, which are
particular instances of goods, either of the same kind or of different
kinds.

So far we have spoken of goods objectively—as objects of de-

sire. Now we can say the same thing subjectively, in terms of the
satisfaction of desire. Satisfaction results from the possession of any
object of desire, but the peculiar subjective sign of the object which
is the good is that satisfaction is complete when it is possessed. Be-
cause of the peculiar relation between the end objectively conceived
as a heterogeneous totality of kinds of good, and the end subjective-
ly realized as a heterogeneous sum of satisfactions, we can always
speak of the end subjectively as well as objectively. We cannot do
this in the case of any partial good, for unless we name the kind of
object desired, we cannot know the kind of satisfaction derived. But
if one were to say that the end of all desire is “complete satisfac-
tion,” one would know this to be the subjective counterpart of that
unique object of desire which consists in “all good things.”

The student was wrong in thinking that we had become in-

volved in a contradiction, but he was right in noticing that our pre-
sent discussion of the end gives new significance to our earlier dis-
cussion of pleasure. When we are concerned with objects of de-
sire—alternatives for preference—pleasure is not the only good; nor
is it ever the sufficient criterion for determining a choice between
particular objects. But when the maxim concerning the desirability
of the greatest quantity of pleasure is properly understood in terms
of the sum of satisfactions, that maxim directs us to the end of all
our desires—a totality of diverse goods. To say that every human
being wants as much pleasure as possible is to say that every human
being wants a good life—a life enriched by every sort of good. In
short, the end must be so conceived that if it could be obtained by
one decision, no-one could resist making this decision, no-one could
choose or prefer anything else because, by the very nature of the
case, everything else must be less good. And the end we have now
envisaged meets that test, whether we think of it as a totality of di-
 60
verse goods or as the utmost in pleasure, the complete satisfaction
of desire.

“I am still not sure I understand what you are driving at”, says
the student. “For one thing, what makes you think it is possible for a
person to get all good things? Everyone may want as much pleasure
as possible, but unless as much as possible can really be obtained, I
don’t see that the end you have envisaged is any better than the pot
of gold at the end of the rainbow. You’ll have to show the man from
Missouri how he can get all good things. And, in the second place,
the rule which directs a person to seek the end, as you have defined
it, no longer seems to be a moral rule by your own criterion, for it is
obviously inviolable. A person could not seek anything else. You
have really admitted what I have always suspected—that it is a nat-
ural law, not a moral law, of human nature, for people to try to get
as much as they can. You have simply described human behavior,
and so far as I can see it is no different from animal behavior.”

Let us begin with the student’s second point. But before we do

that, there is one other point to be made. The student previously
asked us to show that the end is the same for all human beings. We
have so far only succeeded in showing that for a given person there
is just one end, because the end is never a good, but always the
good. But when the good is conceived as a totality of diverse goods,
including every sort of desirable object, is it not clear at once that
the good must be the same for everyone? The reasoning here is sim-
ple: all people have the same human nature; hence they have the
same set of capacities to be fulfilled, however much they may differ
individually in the degree to which they possess this or that capaci-
ty; hence for every human being the variety of objects which can
fulfill these capacities and satisfy the plurality of human desires will
include the same diversity of kinds of good; therefore, the end, con-
ceived as all good things (and understood as an heterogeneous total-
ity of goods or satisfactions, not of course, as an undifferentiated
maximum quantity) must be the same for all people.

“The reasoning is all right”, says the student. “It should have
been obvious to me, I suppose, that if you could prove there is only
one end for each person, you could also show that the end is the
same for every person, for in both cases the reasoning depends on
human nature. But the joke is on you, because you have now
strengthened my point that the end as you have defined it simply
 61
describes what in fact everyone seeks. If there is any truth in what
you have been saying, it is a psychological truth, not a moral truth.
Remember that you yourself told me that a moral rule had to be vio-
lable. Well, if you were to phrase a rule about the end—such as,
Seek all good things—it couldn’t be violated.”

It may be appropriate at this juncture to tell the student that, in

the tradition of moral science, the rule about the end is regarded as
the first principle of morality. Whether it be expressed by such
words as “Seek all good things” or even more simply by “Seek the
good,” this rule is said to be the first precept of the natural moral
law. The student should now be able to see why the first principle of
moral knowledge must be a rule about the end, since he has come to
realize that every object which is not the good, but only a good, is
good only as a means, and hence good in terms of the end. He has
realized, though perhaps vaguely, that only by reference to the end
can a choice between means be determined. That is why the end is
the first principle, without which the problem of preference cannot
be solved. How that problem is solved remains to be seen, but first
we must meet the student’s objection to the rule about the end—not
as a natural law, but as a moral principle.

Some time back the student admitted that all people desire to
live. (Those who do not have no moral problems!) He was even
willing to say that all people desire to live well, though he added the
qualification that what one person meant by “living well” might dif-
fer considerably from what another meant. He should now be pre-
pared, however, to relax that qualification somewhat; for if the
words “living well” name the end which all people seek, and be un-
derstood as equivalent to “a life enriched by the possession of all
good things,” then it would appear that, in one sense at least, all
people must agree about what they mean by “living well.” Let us
call the sense in which all people agree about the end a formal con-
ception of it. All people subscribe to the same formula: a totality of
diverse goods, a maximum of diverse satisfactions. But the student
would be quite right in insisting that, though people may not differ
about this formula, they appear to differ considerably about how
they interpret it. Far from disputing with the student on this point,
we, too, insist upon it, because herein lies the clue to the violable,
hence moral, character of the first principle—the rule about the end.
 62
To carry the analysis further, it might prove useful here to intro-
duce a new term—happiness. The way in which people ordinarily
use the word “happiness” justifies us in identifying happiness with
the end. They regard happiness as something desirable entirely for
its own sake. No one would ever speak of wanting happiness for the
sake of obtaining some further or other good. The happy person is
one who wants for nothing more. Hence it is clear that happiness is
not a good; it is not even accurate to speak of it as “the highest
good” if such words signify that happiness is one good among oth-
ers, albeit the greatest. Happiness is the same as what we have
called the good, the supreme good; but it is summum bonum only in
the sense of being the totality or sum of every kind of good [totum
bonum]. Furthermore, we can identify happiness with living well, or
with a good human life, since the formula is the same in both cases.
The relation between happiness and pleasure is also clear: surely
happiness is not the same as sensual pleasure, which is merely one
kind of good; nor is happiness, considered objectively, the same as
pleasure in the sense of satisfaction; but when we consider happi-
ness, or the end, subjectively, it can be understood in terms of
pleasure, for in possessing all good things the happy person enjoys
every sort of satisfaction and in this sense has maximized pleasure.

That all human beings desire happiness seems to be the law of

their nature. Though this be true, we also know as a matter of fact
that people lead different sorts of lives. If we examine the matter
closely, we find that there are great differences in the accounts they
give of happiness—of what they are seeking, of what they are trying
to get out of life. What is the source of these differences? Since the
formula of happiness is the same for all—a whole of diverse goods,
a sum of diverse satisfactions—there are only two ways in which
people can differ in putting matter into the formula, i.e., in passing
from a formal to a material conception of their end, or happiness.
(1) They can differ in their enumeration of the kinds of goods. (2)
They can differ in the way they order whatever goods they have
enumerated; by “order” here is meant an estimation of the relative
worth of the various sorts of goods—some of which should be pre-
ferred to others, some of which, within the plurality of goods itself,
are related to others as means are to ends. Now according as people
differ in either of these ways, or both together, the happiness they
seek will be differently constituted, for happiness as a totality of
goods is a whole constituted by the variety and order of its parts.
Will the student now agree that we have described the facts which
 63
he had in mind when he said that though all people may seek the
same end (to live well, happiness), they do not seek the same thing?
Will he permit us to express this truth more precisely—and less par-
adoxically—by saying that the end all people seek is the same for-
mally, but different materially?

“Yes”, says the student, “your language says what I mean. But I
don’t see how this is going to show the natural law of human behav-
ior to be a moral rule—a rule (about the end) which is violable.”

We are now prepared to show the student that. Only one new
point needs to be added. Considered materially, there are many dif-
ferent conceptions of happiness. The differences are, for the most
part, with respect to the order of goods—one person emphasizing
wealth, let us say, another friendship, another knowledge, and so
forth. There may be differences in the listing of the goods; though
this is less frequently the case, some people have omitted sensual
pleasure as a good, others have omitted the social goods, and some
have even omitted knowledge and what has been called “moral vir-
tue.” Now the new point we must add (and prove) is simply this:
that among all the different conceptions of happiness which people
have recorded, there is only one right conception, in material detail,
of the variety and order of goods. If a person seeks anything other
than happiness as rightly constituted, the person is not really seeking
happiness at all, but a false or illusory version of it, even though the
wrong thing is sought as the person’s ultimate end because the per-
son conceives it under the same formula.

Let us call the end as rightly conceived the real good; let us call
the end as wrongly conceived the apparent good. Using words this
way, we can see that if the rule about the end is expressed by “Seek
the good, real or apparent,” then it cannot be violated, for it is simp-
ly a natural law, a description of how human beings must in fact
behave—and being a description it is incorrectly expressed as a pre-
scription, in the imperative mood rather than the declarative. But if
the rule is expressed by “Seek the real good,” then it is violated by
every human being who wrongly conceives his or her end, and we
have a moral law, truly prescriptive, saying what human beings
should seek. The same thing can be said in terms of happiness:
“Seek happiness properly constituted by a correct enumeration and a
right ordering of goods” can be violated in many ways; but “Seek
happiness as any collection of goods in any order” cannot be violat-
 64
ed at all. In short, if materially there is one right, and many wrong,
conceptions of happiness, the fact that all people seek the same end
formally does not mean there is no violable rule about the end. On
the contrary, there is a rule the violation of which leads away from
rather than toward real happiness.

“IF,” says the student, “—if. Everything seems to depend now

upon your hypothesis that people can be mistaken about how their
happiness is constituted. The hypothesis being granted, I can see
that much will follow. Those who have mistaken notions about what
their happiness really consists in will probably not do what they
should do in order to live well or become happy. But why should I
grant your hypothesis?”

The student is right to raise this question. Once he affirms the

hypothesis, he ceases to be a skeptic or a relativist about morals, for
every other moral truth can, in a way, be drawn from a true concep-
tion of the end. The student’s question has, however, already been
answered. Though perhaps he did not realize it at the time, he af-
firmed the hypothesis when he agreed to the reasoning by which we
proved, in terms of every human being’s having the same capacities,
rooted in the same human nature, that “the variety of objects which
can fulfill these capacities and satisfy the plurality of human desires
will include the same diversity of kinds of good” for every human
being. The truth about happiness is thus seen to follow from the
truth about human nature, and that is why the first principle of con-
duct (the rule about the end) is not only moral, because violable, but
also natural. It is not only natural for human beings everywhere and
at all times to seek all good things, but it is also in terms of their
nature that the variety and order of goods constituting this whole
should be the same for all. Human beings cannot act contrary to
their nature by wanting to be dissatisfied, or, if you will, by wanting
less than complete satisfaction. But they can make mistakes in un-
derstanding their nature, and as a result of such mistakes set up a
wrong conception of happiness which, if followed, must ultimately
lead them to frustration—the achievement of less than is possible to
their nature.

Thus, for example, if a person should make the error of suppos-

ing no essential difference to exist between human nature and that
of brute animals, and if, accordingly, the person should conceive
himself or herself as having no capacities beyond those possessed
 65
by brutes, he or she will misconceive human happiness by omitting
from its constitution those distinctively human goods which fulfill
capacities which humankind alone has. Such an error here may not
be one of simple omission; it may take the form of misunderstand-
ing the distinctive character of the specifically human goods. How-
ever the error is made, the result will be the same. A human being
cannot become happy by trying to live a good animal life. A human
being must try to live a good human life.

In a manner of speaking, one can say that animals seek “happi-

ness” in so far as they, too, live according to natural law. According
to the law of their nature, there is a sum of goods which can fulfill
their capacities, a totality which they are driven instinctively to seek.
But there are two profoundly significant differences between the
natural law which governs animal and that which governs human
conduct. One we have already seen—the difference in what is the
sum of goods for each according to its nature. The other is that, in
detail as well as generally, animal seeking is instinctively deter-
mined, and hence there can be no discrepancy between what ani-
mals do seek and what they should seek; whereas human seeking is
“instinctively” determined only with respect to the end as formally
conceived; hence human beings may in fact not seek what they
should. This is just another way of saying that human beings, unlike
brute animals, are able to think about their end, and since wherever
thinking occurs, error may happen, human beings can misconceive
their happiness. Unable to think abstractly, animals cannot conceive,
and hence cannot misconceive, their end. Therefore, there is only a
natural law, but no moral law, of animal behavior, whereas human
conduct is susceptible of direction by a natural moral law.

“I have gradually come to realize”, the student confesses, “how

important a role the conception of human nature plays in the discov-
ery of universal moral principles. But it never occurred to me before
that I had to swallow all this stuff about the essential difference be-
tween humans and animals. All the psychology I have studied—
experimental psychology, animal psychology—as well as all the
biology, and especially the business about evolution, is against such
a notion. If this new point is indispensable to the argument, you’ve
got a lot more proving to do. For the moment I don’t see that it is
indispensable, and so I’ll waive the point in order to ask another
question.
 66
“I’ll grant that all human beings have the same human nature
whether or not that is essentially different from the nature of ani-
mals. I can see how, in terms of that common nature, happiness
must be really the same for all people, in the sense of including the
same variety of goods; and I can also see that, if people misconceive
their nature, they will probably misconceive what is really good for
them. But you pointed out before that people misconceive happiness
in two ways—both with respect to the variety of parts which consti-
tute the whole of goods, and also with respect to the ordering of the-
se parts. Moreover, you said that the most frequent errors occur with
respect to the ordering of the partial goods, rather than in their enu-
meration. This requires some explanation. I don’t see why there
need be any ordering of goods. If a person should rightly enumerate
the parts of happiness, why should not he or she get the whole by
going after the parts in any order?”

To answer this very difficult question, let us begin by reminding

the student that, at an earlier point, he wondered whether it is possi-
ble for a person to get all good things. He compared the end, thus
envisaged, to the rainbow’s end. He wanted to be shown just how a
person can get all good things.

Now, in the first place, let us remember that all good things
does not mean every particular good, but only some of every kind of
good. If this were not so, it would take an infinitely long life to get
all good things, and, furthermore, the pursuit of happiness would be
competitive—as is the attempt to corner the market and possess eve-
ry piece of a certain commodity. But this is not the case; however
much in fact they do, people need not interfere with one another in
the pursuit of happiness.

In the second place, let us remember that all good things is a

possible whole because the various kinds of good which are its parts
do not exclude one another. They are all compossible with one an-
other; if they were not, the whole we have supposed to be constitut-
ed by them would be self-contradictory and impossible.

In the third place, there is a new consideration: the point about

order. Either the order in which we go after the various partial goods
makes no difference, or it does. Suppose it makes no difference.
Then happiness would be easily achieved by everyone who made a
right enumeration of the partial goods. Regardless of whether such
 67
people subordinated wealth to knowledge, or knowledge to wealth,
regardless of whether they spent a great deal of their time and effort
in search of sensual pleasures or postponed taking care of their
health until after they had achieved public honor, they would not be
prevented from becoming happy so long as they included every sort
of good among the objects of their pursuit. But this appears to be
contrary to the facts of life as we know them. The familiar saying
that “there can be too much of a good thing” applies to some of the
partial goods which enter into the constitution of happiness: too
much of some of them can disbar us entirely from others.

Not only must the degree to which we seek certain types of

good be proportioned to their worth as parts of the whole, in order
to prevent them from interfering with our possession of other types,
but each kind of good must be seen in its functional relation to every
other kind, according to the functional interdependence of the ca-
pacities of human nature, which these different kinds of objects are
able to fulfill. We must conclude, therefore, that a person cannot
become happy unless, in seeking all good things, he or she does so
in the right order and with due proportion. That is why happiness is
difficult to achieve, even for a person who has correctly enumerated
the various partial goods.

It is necessary to tell the student that we have not fully answered

his question. To do that would require a lengthy and elaborate anal-
ysis whereby we might be able to show him all the reasons for one
precise ordering of goods as the only correct disposition of the parts
of happiness. We shall have to be content with making two points
about the order and proportion of partial goods.

First, the order of these goods, like the enumeration of their va-
riety, depends upon our understanding of the various capacities of
human nature in their relation to one another.

Second, it depends upon our recognition of what is distinctively

human, in contrast to that part of his nature which humankind shares
with brute animals. In the order of partial goods, those are higher
which fulfill humankind’s rational capacities; the lower goods are
objects commonly pursued by humans and animals. The lower serve
the higher as means serve ends; in order to live well, we must first
live. We struggle to subsist, not merely to be alive, but to live as
humanly as possible, and this means subordinating and proportion-
 68
ing the goods which fulfill our animal capacities so that we shall be
able to enjoy a fuller life than animals can lead—enriched by goods
that fulfill capacities which only we, as human beings, possess. Just
as a true conception of human nature is indispensable to a true con-
ception of happiness, with respect to the variety of goods, so is it
also with respect to their order. And the student is wrong in suppos-
ing that the point about the essential difference between human be-
ings and brutes is dispensable. It is indispensable to a true concep-
tion of human happiness, and equally with respect to both aspects of
its constitution—both the variety and the order of its parts.

“I am sorry that you insist upon this last point”, the student says,
“because it is a stumbling block in the way of my agreement with
you. You simply haven’t proved the point in any way, and, without
obstinacy, I must stand on what I know—which is contrary to what
you say is the case. I can see, however, that, assuming what you say
to be true, the rest follows. And even though you have not given me
the analysis which shows the precise order of the partial goods, I
can surmise how that might be done. But I am still worried about
the possibility of happiness, as you have defined it. I still don’t see
how it is possible for a person to be happy if he or she has to pos-
sess every sort of good thing altogether and at once.”

The very language the student has used in raising this question
is crucial to the answer. Strictly speaking, a person cannot ever be
happy. He or she can only become happy. A human life is some-
thing in the process of becoming. It is a temporal whole, the parts of
which cannot coexist. A life is a whole only in the way in which a
day or a game is a whole—as an orderly succession of moments.
The becoming of the whole is not completed until the process is ac-
tually finished. That is why Solon, a wise man of ancient Greece,
made what at first seems to be a paradoxical point, namely, that you
cannot tell whether a person is happy until the person is dead. Stated
less paradoxically, the point is that happiness is the quality of a
whole life, not of its parts. Another ancient, the Roman Boethius,
defined happiness as the state of those made perfect by the posses-
sion in aggregate of all good things. The student may think that this
is the definition of happiness we have been employing. His attention
must be called, therefore, to two important differences.

First, happiness may be the state of immortal souls in eternity—

and that is probably what Boethius had in mind—but in this life,
 69
which is from beginning to end a process, a becoming, happiness is
never realized statically. In the realm of time and change, it must
exist dynamically—coming to be just as the life which it pervades
becomes complete in time. Second, in so far as the happiness we
have defined is a quality of this temporal life, the possession of all
good things must be successive; it cannot be a simultaneous aggre-
gate. Modifying the words of Boethius, we can define temporal
happiness as a whole life made perfect by the successive enjoyment
of all good things. Thus understood, there is nothing impossible
about becoming happy, any more than it is impossible to complete a
whole life by living from day to day.

It may be useful here to remind the student that a person cannot

become happy by making one decision. A person becomes happy
only through making many decisions, choosing many times between
one particular good and another, exercising countless preferences. If
a single decision could do it, the person who made it correctly could
be happy as a result. But even a person who has correctly conceived
happiness may fail to become happy unless the many choices he or
she has to make from day to day conform to the pattern of life he or
she has conceived—a whole rounded out by every sort of good.
Acts of choice or preference are always with respect to means, to
partial goods. Throughout life we are forever at work putting the
parts together to form the whole. The student should now be able to
see how the problem of preference is solved in general, if not in de-
tail. Faced with a choice between objects which are particular in-
stances of partial goods, we should choose in such a way that we
make progress toward the possession—in our life as a whole—of all
good things. The end we have in view determines our choice of the
means; our conception of the whole determines our manipulation of
the parts.

It has been said that the means are the end in the process of be-
coming. This is the sum of moral knowledge. In the building of a
life, as in the building of a house, it is true that as the parts are
properly chosen and properly put together, the perfection of the
whole gradually becomes. In every particular case, the ultimate cri-
terion of choice is the end: the choice is right or wrong according as
the realization of the end is furthered or hindered. At the beginning
of life all good things is a possibility; when life is over, the possibil-
ity either has been realized or not, and that will depend upon the
choices which have been made.
 70

The universal principles of moral knowledge consist of rules

about the end to be sought and about the means to be chosen. If the
end is properly conceived, the rules about the means will be proper-
ly formulated, since the conception of the end, when fully devel-
oped, consists of an ordering and proportioning of the means. If
people know what they should seek, they will know how they
should seek it. Based upon human nature, the rules of morality, di-
recting us toward our end and prescribing our choice of means, have
universality in the sense that they are the same for every person, but
this does not mean that any person can avoid the task of applying
these generalities according to the peculiar conditions of his or her
individual life, and the particular circumstances of each case in
which a choice must be made. And since these rules are the work of
reason, not the gift of instinct, the moral judgments of human be-
ings, about end or means, are susceptible to error.

“I think I see”, the student says, “how what you have called a
true conception of the end of life is a first principle from which all
the rest can be derived. I know I don’t see the detailed steps here,
but that would be too much to expect. I do understand how, in a
general way, the problem of preference is solved—at least in so far
as the value to be placed on different things in relation to one anoth-
er follows from what you have called the order and proportion of
goods, which in turn follows from the way in which we conceive
happiness to be constituted as a whole of parts. I am sure I don’t
understand any of this well enough to know how to think correctly
in a particular case, facing particular alternatives—assuming, of
course, that I had previously thought correctly about the end, and
through it about the means in general. But what bothers me most of
all is still the point about the violability of moral rules. Am I right in
supposing, from what you have said, that people fail to do what they
should do simply because of bad thinking—wrongly conceiving
happiness, which means making errors about the order or variety of
goods, and consequently misjudging the relative worth of objects in
particular cases of preference?”

No, the student must be told at once, to suppose that all miscon-
duct—bad choices leading to bad acts—follows from bad thinking
is itself a great error in thinking; as a matter of fact, a famous one in
the history of moral theory. The quickest way to show the student
that bad thinking is only one of the sources of misconduct is to re-
 71
mind him of something we saw toward the close of the first part of
our discussion—when we still supposed pleasure to be the only cri-
terion of preference. Remember the rule of conduct we had then
formulated: “In any case in which a choice can be made, a person
should prefer the alternative which, in the long run or viewing life
as a whole, maximizes pleasure and minimizes displeasure.” The
student should now be able to see how that rule contains in germ
almost all the truth we have subsequently discovered, for the maxim
could be thus rephrased: in any case in which a choice can be made,
a person should prefer the alternative which, considering the possi-
bility of living a whole life well, tends to realize the possibility of
happiness. Now we saw, in our earlier discussion, that the rule is
violable in two ways. We distinguished between two sorts of mis-
takes we can make—mistakes of calculation and mistakes of acting
contrary to our calculations. Even though a person has thought cor-
rectly about the end and the means in general, and in this sense
knows what is really good for him or her, the person may in a par-
ticular case be seduced by what is apparently better at that time,
even though it is really worse in the long run.

Human beings are not simply rational beings. They are rational
animals. They are creatures of passion, of animal appetites, but they
are also capable of abstract thought, by which they can form a con-
ception of happiness and of the goods in general which constitute it.
All the goods we have been talking about—happiness itself, or the
various kinds of partial goods—are rational or intelligible objects.
They certainly cannot be perceived by the senses, although particu-
lar instances of the partial goods may somehow present themselves
in that way. But precisely because human beings are creatures of
sense as well as of reason, and because their desires can be deter-
mined by what they sense as well as by their abstract thinking, the
alternatives which they face in particular cases of preference are, as
objects, both sensible and intelligible, and make their appeal both to
their animal appetites and to their wills—the latter being the desire
for objects rationally judged to be good, the former being the desire
for what is sensed as good.

We cannot here fully explain all the psychological points that

are involved. For our present purposes, suffice it to say that, in the
conflict between sensible and intelligible goods, the former may win
out because they are apparently better at the time, though not really
better in terms of the conception of life as a whole. Whenever they
 72
do win out, you have a case in which misconduct is due to weakness
of will, rather than error in thought. Instead of following the dictates
of reason, a person may choose according to the promptings of his
or her passions. Thus we see that the violation of moral truth has
two sources: one, bad thinking—misconception of the end, and of
the means in general; the other, weak willing—preference for the
apparent good, under the influence of the passions, rather than for
the real good which reason has determined. And we also see how
the very nature of human morality (revealed to us by the sources of
misconduct) depends on the nature of humankind—their essential
distinction from brute animals. For, lacking reason, animals know
and desire only through sense and instinct; lacking reason, they are
not able to conceive their end, and hence cannot misconceive it;
lacking reason, their desires are all instinctively determined and sub-
ject to whatever objects dominate the sensible present, whereas hu-
man beings have free will to choose between sensible and intelligi-
ble goods. In short, unlike animals, humans can be moral or immor-
al according as they think well or poorly, and according as, in the
exercise of their free will, they act according to what right reason
prescribes, or contrary to it.

“There is no point in going any further”, the student finally

says. “I told you before that I wasn’t prepared to accept your as-
sumption about an essential difference between humans and ani-
mals. I am even more opposed to it now that I see it includes the
notion of free will. You were quite fair to admit that you had not,
and probably could not here, explain all the psychological points
that are involved; but until you do explain them, until you do prove
that human thinking is different from animal intelligence, and above
all until you can show what you mean by free will and that there is
such a thing, it would be unprofitable to carry our discussion any
further. I am willing to agree that your conclusions have cogency for
anyone who grants your hypotheses. If humankind is peculiarly ra-
tional, if humankind has this mysterious free will, then you are right
about morality, and its principles, and the way they can be violated.
Until I agree to these ifs, my position is exactly what it was when
we started—though perhaps, I should admit that I now understand
better the theory which I, as a moral skeptic, have been rejecting. In
fact, I can now give you more clearly the basic reason for my moral
skepticism. It is simply that human beings do not have free will. If I
had not let you somehow obfuscate this point at the beginning, our
discussion would have stopped almost as soon as it started. I tried to
 73
tell you at the very start that I didn’t think there was any problem of
preference; I tried to say there was no why for any choice, no why in
the sense of a reason which justified it, but only a cause.

“Every choice a human being appears to make is just like any

choice an animal makes. It is no choice at all, but a pre-determined
event—arising from instinctive determinations, and all the acci-
dental conditionings which have occurred in the course of life up to
that point. If there is no problem of preference, because there is no
free will, then all the rest of our discussion was totally beside the
point. Or, to put it another way, there is a problem of preference, but
only for the psychologist who tries to find the causes of behavior
and to describe what humans and animals in fact do; but there is no
problem for the moralist who tries to find the reasons for human
conduct and to prescribe what people should do. If I am right about
the facts, then you must admit that the moral skeptic is justified in
thinking that all the different moral systems which people have in-
vented—yours among them—are nothing but intricate and elaborate
rationalizations, fostered by the delusion that human beings are
free.”

The student is right that there is no point in going further with-

out first satisfying him on the major psychological questions which
underlie all moral discourse. It would not be sufficient here to re-
mind him that he did admit certain facts, such as that people do ap-
pear to act contrary to their best lights and seem to suffer repentance
for their folly—facts which suggest human freedom. He rightly asks
for proof, and the task of proof in this case is long and arduous, as it
is also on the other point about humankind’s rationality as their es-
sential distinction from brutes. All of this requires another and sepa-
rate discussion, one in which we would probably find the student a
skeptic about the truths of philosophical psychology. We might then
discover that his moral skepticism was rooted in a deeper doubt—
the doubt about the validity of any philosophical knowledge.

By way of concluding this discussion, it might, however, be

worth while to remind him of one thing. He has learned one truth
which he may not have known before. All through the discussion he
has admitted seeing the connection between human nature and the
principles of human morality. Now if our hypotheses concerning
human nature and human freedom can be affirmed, then he must
admit the consequences (and he has indicated his willingness to do
 74
so)—namely, the conception of happiness, the order and variety of
goods, and the principles by which the moral problem of preference
can be solved. Furthermore, since whatever human nature is it is the
same for people at all times and everywhere, the student must also
agree that there cannot be a number of different “moral systems”
each equally acceptable. He must agree that there is only one true
doctrine, only one which accords with the truth about human nature,
just as he agreed that in the light of human nature there is only one
right interpretation of the natural moral law to seek the good, only
one right conception of happiness and of the means thereto.

If the student wonders where this discussion would turn next—

were it continued after the psychological questions had been satis-
factorily answered—we should, in parting, tell him that what re-
mains to be considered is the very heart of moral knowledge, name-
ly, good habits (which the ancients denominated “virtues”), and es-
pecially the habits of right desire and right action which are called
the moral virtues. All the principles we have so far discussed be-
come operative only through virtue. The virtues must be possessed,
not only as among the goods which are constitutive means of happi-
ness, but also as a special sort of means—generative of happiness.
And this is especially true of the moral virtues, which are habits of
right choice in particular cases, habits which have been formed in
the light of a proper ordering of goods and which enable us to act
according to reason, to prefer the real to the apparent good. The ma-
jor part of moral theory, therefore, is concerned with the definition
of these virtues, and with the rules for acquiring them.

We have previously said that the means are the end in the pro-
cess of becoming. The end is rightly understood only so far as we
rightly apprehend the means which constitute it. It is also true that
the end is possessed at any moment only to the extent that we pos-
sess the means which generate it—the habits from which our con-
duct flows. At any given moment in his life a person is more or less
on the way to becoming happy according to the state of his or her
habits, especially the moral habits—the virtues or vices—which
make his or her character what it is. Aristotle thus summarized the
whole of his Ethics when he said. “According as a man’s character
is, so does the end appear to him.” Until a life is over, you cannot
judge whether it is a happy one; but so far as you can see into a per-
son’s character, you can tell, even while life is going on, whether a
person is becoming happy.
 75

CHAPTER V: PSYCHOLOGICAL PRESUPPOSITIONS: LIMITATIONS

OF THE DIALECTIC

IN THE preceding chapters of this book, I have outlined a dialectical

procedure whereby a doubting mind might be led to the recognition
of moral truth. What has been given is the bare plot of a conversa-
tion between teacher and student. The student was, at the beginning,
a skeptic about moral matters, denying the objectivity of moral
knowledge, supposing that all moral judgments were a matter of
opinion, entirely relative to the individual or to his cultural location
at a given time and place. The teacher, by asking him to explain the
undeniable fact that men exercise preference, gradually made him
realize that his own criteria for preference—pleasure and quantity of
pleasure—had a certain universal validity; and then, as a result of
seeing the inadequacy of these criteria, the student began to under-
stand that happiness, rather than pleasure, was the ultimate principle
of moral judgments. The crucial steps in the argument were: (1) the
distinction between pleasure as one among many objects of desire
and pleasure as the satisfaction of any desire; (2) the enumeration of
the variety of goods which are objects of human desire; (3) the point
 76
that only the totality of goods can completely satisfy desire; (4) the
realization that this totality of goods, leaving nothing to be desired,
is the end of all our seeking, and that everything else is sought for
the sake of its attainment; (5) the conception of happiness as “all
good things,” a whole constituted by every type of good, the com-
plete good being the end, the incomplete good its parts or constitu-
tive means; (6) the conclusion that the end, as the first principle in
the practical order, is the ultimate criterion of preference, for prefer-
ence or choice is exercised only with respect to means, and hence
we should, in every case, prefer whatever is more conducive to the
attainment of happiness.

But, unfortunately, this dialectical process was far from being

completed. The student may have gained some understanding of the
position he had previously rejected. He was not, however, con-
vinced that happiness, rightly conceived, is the same for all men—
the same order and variety of goods. Nor did he admit that rules of
conduct, even if they are universal, can be violated by a disobedi-
ence born of man’s freedom to act for or against his own real good.
Conviction on these major points could be produced, the student
indicated, only if he could be shown the truth of certain views about
human nature, which the teacher seemed to be taking for granted.
And the teacher, on his side, had to acknowledge that unless men
were rational animals, unless in being rational they were essentially
distinct from brutes, specifically superior in their powers, and
through their rationality possessing freedom of will, unless these
things were so, the proof of moral principles could not be made.
Indeed, the very “fact” of preference, with which the whole discus-
sion had started, turned out to be ambiguous, since the teacher, as-
suming free will, had supposed preference to be a genuine choice
among alternatives, and the student, denying freedom, had regarded
preference as if it were a mechanically determined motion.

That the argument thus uncovered its own limitations is one of

the chief merits of the dialectical procedure. The student learned a
hypothetical line of reasoning; more than that, he acknowledged its
cogency: the premises, being granted, the conclusion seemed to fol-
low. But the premises were certainly not self-evident truths; and,
since it is not fitting in philosophy to make assumptions or regard
conclusions as merely hypothetical, the psychological propositions
upon which the whole argument turned must themselves be demon-
strated. A dialectic of morals cannot be made conclusive unless pri-
 77
or matters are similarly argued. I say “similarly argued” because it is
not enough to see that metaphysics and psychology provide the the-
oretical foundations for moral philosophy; it must also be recog-
nized that the psychological questions involved are for the philoso-
pher, not for the scientist, to answer, and that his mode of answering
these questions must be dialectical in the sense that dialectic is the
process of inductive reasoning whereby the mind establishes those
primary truths which are not self-evident.12 The proposition that
man is a rational animal is not self-evident. Its truth can be estab-
lished only after it has been inductively proved that a plurality of
individual substances exists and that among these corporeal sub-
stances there are differences in essence as well as in number. For if
there are no substances and if they do not differ essentially, as well
as accidentally, from one another, there is no point in attempting to
define man’s specific nature. That man exists as a distinct species of
corporeal substance is the ultimate conclusion of a dialectic which is
many times more difficult and much more elaborate in its phases
than the dialectic of morals herein described. Without undertaking
it, the teacher cannot convince the student of even the simplest mor-
al truths—that preference involves free choice or that happiness,
being the same ultimate end for all men, is the universal principle
which directs men in their choice of means.13

Since the student is justified in not considering the argument to

be conclusive until his basic objections have been met (i.e., until his
questions about prior matters have been answered), I am willing to
regard whatever conclusions we have so far reaches as hypothetical,
for that is the only way in which the student can now understand
them. I do so in order to go on, not with the dialectic itself, but with
a deductive elaboration of some of its major points. In the final sec-
tion of this essay, I shall try to show how the two fundamental con-
cepts of ethics—happiness and virtue—are indispensable to political
philosophy; for unless these concepts have objective validity, unless
there is an objective order of goods, an order of means and ends,
which enables us to distinguish right from wrong in human conduct,
by knowledge rather than by opinion, the philosopher has no de-
fense against realpolitik (which is an inevitable consequence of pos-
itivism in the sphere of politics). And in the subsequent section of
this essay, I propose to treat of three matters insufficiently discussed
in the foregoing dialectic: (1) the objectivity of the good in relation
to desire; (2) the kinds of good and the types of means-end relation-
ships; and (3) the nature of virtue as principal means to happiness as
 78
end. All of these points were implicated in the preceding discus-
sions, and would have been explicated had the discussions contin-
ued. In each case, I shall indicate a leading question the student
might have asked at a given turn in the preceding discussions—a
question which, if fully explored, would have then generated anoth-
er separate phase of inquiry. But now, for the sake of brevity, I shall
confine myself to an analytic summary, outlining in each case what
any teacher would have to do to carry on.14

12. Two meanings of “induction” as well as two meanings of “dialectic”

must be distinguished. The word “induction” is sometimes used to name
the non-discursive step by which the mind generalizes from experience:
just as it abstracts universal concepts from sensible particulars, so it some-
times forms, in the light of these concepts themselves and without the me-
diation of prior knowledge, universally true judgments. Because they are
not obtained by reasoning, these judgments are called propositions per se
nota or self-evident truths; and the intellectual act by which they are
achieved can be called an “intuitive induction.” (cf. Aristotle, Posterior
Analytics, II, 19). In contrast to intuitive induction, there is that process of
the mind which might be called “rational induction”, because it involves
reasoning, and is a discursive or mediated way of knowing, a process and
not a single step. Such reasoning or proof is inductive rather than deductive
in that it is a posteriori rather than a priori, from effects to causes rather
than from causes to effects. In contrast to deductive reasoning, which ex-
plicitly elaborates what is contained in universal truths already known,
inductive reasoning establishes those primary truths which are affirmations
of existence, truths which are neither self-evident nor capable of being de-
duced from prior universals. The ultimate grounds of inductive proof are
the facts of sense-experience. The a posteriori proof of the existence of
God is inductive reasoning in this precise sense. Whereas deductive reason-
ing is the motion of the mind from what is more knowable in itself to what
is less knowable in itself, inductive reasoning is that motion in which the
mind goes from what is more knowable to us to the existence of something
whose nature is more knowable in itself, though less knowable to us.
The word “dialectic” is frequently used, in the Aristotelian tradition, to
name probable reasoning from premises taken for granted for the sake of
argument. But that is not the only traditional meaning of the word. There
is, of course, the Platonic meaning of dialectic as the motion of the mind
toward first principles, but there is also the Aristotelian point that “dialectic
is a process of criticism wherein lies the path to the principles of all enquir-
ies” (Topics, I, 1). When dialectic is employed demonstratively and not
polemically, it is identical with inductive reasoning directed, not to all first
principles or the principles of all enquiries (for some of these are self-
evident and are known by intuitive induction), but only to those primary
affirmations of existence which are neither self-evident nor capable of de-
 79
ductive demonstration. As reasoning may be either deductive or inductive,
so demonstration may be either “scientific” (i.e., deductive) or “dialectical”
(i.e., inductive).

13. The argument which must be undertaken can be called “a dialectic of

substance, essence and man.” I think I am now able to work out the several
phases of this argument, and, having outlined !he whole of it as an orderly
sequence of parts, I am satisfied that it demonstrates, with certitude, a
number of primary propositions which have heretofore always been as-
sumed—not because anyone could have mistaken them as self-evident, but
because the way of inductive reasoning and dialectical demonstration has
been inadequately understood and too infrequently used in philosophy. I
hope to be able to publish this material shortly [cf. The Difference of Man
and the Difference It Makes (1966)], and with it I shall try to present a
more analytically refined account of inductive and deductive reasoning
than can be given in a brief footnote (cf. note 12 above). The “dialectic of
substance, essence and man” is not only important in itself as an argument
for certain conclusions which have not previously been demonstrated, but
it is also significant as an illustration of hitherto unnoted aspects of philo-
sophical method.
In one sense, the argument is miscalled a dialectic, for all of its phases
are not strictly inductive, though the denomination is justified by the fact
that all of the primary conclusions are inductively reached. Thus, for ex-
ample, the proof that, if there are a number of distinct essences, they must
be ordered in a perfect hierarchy, is deductive. (This proof, by the way,
was given only in the indirect form of a reductio ad absurdum argument in
“The Solution of the Problem of Species”, The Thomist, 111, 2, pp. 329-
332. In that form, the proposition that man is a rational animal and superior
to all other corporeal creatures had to be assumed. But the definition of
man, not being self-evident, must itself be proved, and that cannot be ac-
complished unless the perfect hierarchy of essences can itself be inde-
pendently proved. Hence the importance of a direct proof.) But that there
are a number of distinct essences embodied in the world of corporeal sub-
stances, how many there are and what they are must be proved inductively
from the observable motions and operations of sensible things, and this can
be done only if we first know that perceived objects, which seem to be
subjects of change, are truly substances composed of matter and forms, and
that among these forms one must be substantial and all the rest accidental.
From these facts, inductively proved, the truth about hierarchy of essences
can be deduced; and from the truth about hierarchy can be developed the
criteria for interpreting the sensible evidences from which we must induce
the existence of whatever essential distinctions there are among substances.

14. It should be recognized that brevity is the real reason for this change in
style. Although the full development of argument with respect to each of
 80
the three points mentioned would depend upon psychological propositions
already questioned by the student, there is no reason why the student
should not proceed hypothetically—to discover whether other moral truths
(other than the one about happiness) can be established, once it is granted
that man is a rational animal, that man has a nature and powers essentially
distinct from the nature and powers of brute animals, that man has free
will, etc. If the student had been told, at the very beginning of the discus-
sion, that these psychological propositions were indispensable to the argu-
ment, he would either have refused to begin until these propositions had
been proved, or rightly have insisted that any conclusions reached by an
argument thus undertaken must be regarded as hypothetical. That is the
way he now views the conclusion about happiness (as constituted in the
same way for all men). There is no reason, therefore, why he would be
unwilling similarly to entertain further conclusions about the order of
goods or about virtue, if they could be reached. But to deal argumentative-
ly with each of the three points, now to be considered, would require much
more time and patience than can be expected of the reader. That is why I
shall present an analytical summary of the argument instead of letting it
expand in response to the demands of an inquiring mind.

On the dependence of ethics and politics upon psychology, see Aristotle’s

Ethics, 1, 13.

CHAPTER VI: REAL VS. APPARENT GOODS: THE REALITY OF

VIRTUE.

1. THE GOOD AS OBJECT OF DESIRE

WHEN THE word “good” was first introduced to name “any object
of desire,”15 the student might have stopped the discussion, or
turned it into other channels, by asking whether this was not a mere
tautology, a mere rule of verbal substitution (i.e., the word “good”
being substitutable for “object of desire” and conversely). For such
a question would have raised a crucial issue concerning the relation
of these two basic terms: either whatever is called “good” is so
 81
called because it is in fact desired by someone; or whatever anyone
desires is desired because it is good. In the first alternative, there
could be no distinction between the real and the apparent good,
since the fact of desire itself is regarded as conferring goodness up-
on its object. In the second alternative, goodness would be a proper-
ty of things apart from their being actually desired, for as good they
would be desirable whether or not desired , in which case no object
would be really good unless it were desirable according to its being,
whereas any object whose goodness derived simply from its being
desired might be an apparent good. It is plain that the alternative
which destroys the distinction between real and apparent goods re-
sults in complete subjectivism (and the denial of moral knowledge).
Only the alternative which maintains this distinction can uphold the
objectivity of good. Only if it is possible to desire things because
they are good (and, as such, desired), can desires themselves be ob-
jectively judged as right or wrong (i.e., according to the goodness or
badness of their objects).16

Such traditional maxims as that “the good is what all desire” or

that “goodness is being in relation to appetite or desire,” do not
solve the problem. On the contrary, they raise the problem by call-
ing for an interpretation which will show how goodness is founded
in being prior to actual desire (i.e., to human desires determined by
judgments or estimations of objects as good or bad.)17 The problem
cannot be solved, nor can the familiar maxims about goodness, be-
ing, and desire, be rightly interpreted, unless we distinguish between
“natural desire” and “human (or animal) desire” and until we under-
stand that the good is being as a final cause of motion. Then it will
be seen that whatever is imperfect, through privation of being in
some determinate respect, has determined potentialities for perfec-
tion; each determinate potentiality is a natural appetite or desire for
some actualizing determination of the thing’s nature; any thing’s
natural appetite or desire is constituted by all its determinate poten-
cies, whether these be powers which flow from its essence or mere
potentialities determined by privation through the thing’s possession
of contrary forms; each potency has a tendency (and appetite con-
sists in such tendency) toward actual being; now that which actual-
izes a potency satisfies a natural desire or tendency; to be actual in a
certain respect is, therefore, the object of every natural desire and
this actuality is the final cause of the thing’s motion; hence, the
good is being as naturally desirable ; it is whatever completes or
perfects the being of a thing in respects in which it is imperfect or
 82
incomplete, the very respects which determine its potencies, and
hence its natural desires, for perfection. When “desire” is under-
stood to signify natural desire and when “good” is understood to
signify actual being in relation to potency , then it becomes intelli-
gibly true to say that “the good is what all desire” or that “goodness
is being in relation to appetite.” In fact, these become self-evident
truths, which must be affirmed as soon as their terms are thus under-
stood. It is similarly true that each thing has as much goodness as it
has being (i.e., actuality); and it might also be added that each thing
has as much desire for goodness as it is deprived of the being that is
due its nature, i.e., relative to its capacities for perfection. In these
meanings of “good” and of “desire,” there is a strict correlation be-
tween them, for potency, as material cause, is at every point correla-
tive with actuality, as final cause. Hence it is possible to say without
contradiction that the object of desire is good because it is desired
and also that whatever is desired is desirable because it is good, alt-
hough the priority of actuality and of final causes gives the second
statement a certain precedence over the first.18

But this account of the good as an object of desire is purely

metaphysical: in terms of potency conceived as appetitive or tenden-
tial, and of actuality conceived as perfecting or good This account of
desire and the desirable as correlative aspects of being in motion
would be true even if there were no human beings, even if there
were no creatures who felt pleasure and pain, who consciously
yearned for objects they deemed desirable, or who experienced sat-
isfaction or frustration.19 We must, therefore, not confuse the meta-
physical conception of the good with the moralist’s conception,
which concerns the good of man, or the variety of goods for man.
As metaphysically considered, the good is not subject to the distinc-
tion between real and apparent; but in the moralist’s consideration,
this is obviously an indispensable distinction. Our task is to discover
how this distinction can be made in the realm of the human good,
with which the moralist is concerned. The problem can be solved
only if we see the human good in relation to the natural good, and
this in turn depends upon the basic distinction between human and
natural desire. Human desire is conscious desire. The word “con-
scious” is here used in its primary signification: it means “with or
through knowledge.” A human desire is an appetite determined to
its object by knowledge of this object—not merely theoretical
knowledge, but practical knowledge consisting in a judgment that
the object is good. The appetite which is thus determined is not any
 83
potency; it is a specific power of man, the very nature of which is to
tend toward an object intellectually apprehended as good. Human
desire can be defined, therefore, as the act of man’s intellectual ap-
petite, i.e., his special power of desiring or tending toward what his
intellect apprehends as good. The distinction between human and
natural desire is thus seen to be twofold: in the first place, a human
desire is always an act (of a power,) and never identical with mere
potency or power itself, whereas every discriminable natural desire
is identical with some potency or power, having by its very nature a
determinate tendency; and, in the second place, natural desire is un-
conscious, i.e., it is totally independent of knowledge of the object
desired, whereas human desire must always be conscious, for it is
the act of a power which cannot be moved to act except by a prior
act of apprehension.20

In the light of the foregoing, we can define the natural good as

the naturally desirable, as comprising the objects of all natural de-
sire; and the human good as the consciously desired, as comprising
all the objects of human desire, of both intellectual and sensitive
appetite. Now a third term must be introduced, namely, the natural
human good, for man has, over and above his two conscious appe-
tites, as many natural desires as he has powers and potencies which
follow determinately from the essence and accidents of his nature.
The natural human good comprises all those objects toward which
human nature tends as toward perfections of its being in respects in
which it is deprived of actuality.21 The whole problem turns on the
discrepancy between the human good and the natural human good.
The natural human good is a specific case of the good that the meta-
physician defines as the act of being which can terminate a natural
desire: it is merely the good or the desirable appropriate to the na-
ture of man. But the object of human desire may or may not be natu-
rally good, by the metaphysical criterion. Men can be mistaken both
in their rational judgments and their sensitive estimations: they can
consider things to be good, which, when consciously desired and
subsequently possessed, ultimately frustrate their natural desires.
They can, in short, consciously desire objects which they do not
want or want for things they fail to desire because of deficiencies in
apprehension or errors in judgment; and even when their desires
somehow conform to their wants, the way in which they order the
objects they consciously desire may prevent them from achieving
the fulfillment of their natural appetites. That which is an object of
actual human desire (i.e., of an act of intellectual or sensitive appe-
 84
tite) may not be the real good, in the metaphysical sense of that
which, in the order of real being, perfects the nature of man. Where-
as the object of natural desire, in the case of human or any other na-
ture, is necessarily a real good, the object of conscious human desire
may be an apparent good, strictly not a good at all, but evil either
because unsuitable to nature or as relative to other objects of desire.
The apparent good is, therefore, good only because it is actually
desired, which means, of course, that the object is good only in hu-
man judgment or estimation, upon which the desire itself is conse-
quent. The real good is, in contrast, always an object desirable in
itself, as a perfection of the nature: hence when it becomes, through
knowledge and right judgment, the object of actual human desire, it
is desired because it is good. Whatever men actually desire, they
apprehend under the aspect of the good, for in order to desire any
object they must judge it to be good, whether they do it rightly or
wrongly. And what ultimately determines whether their judgment is
right or wrong—whether the object actually desired is a real or an
apparent good—is human nature itself, its natural appetites and the
natural goods toward which they tend. In proportion as his moral
judgments are founded on a correct and adequate understanding of
his nature, will a man make a true estimation of what is really
good.22

This understanding of the good as object of desire has two fur-

ther consequences, beyond enabling us to distinguish the real from
the apparent good. Both must be considered before we pass to a dis-
cussion of the order and variety of goods.

In the first place, we can define the precise metaphysical charac-

ter of every object of desire, natural or conscious. By the essence of
what the good is in relation to desire, the object of desire must al-
ways be a change or motion, for change or motion is the act of that
which is in potentiality in a respect in which it is potential . Since
the good is always the act of being toward which potencies or pow-
ers tend, and by which such tendencies are terminated, the object of
desire must be a change or motion, i.e., an actualization. The full
significance of this is seen by comparing the object of knowledge
with the object of desire: the object of knowledge is always the
completed nature of some existing thing (i.e., a substance, in its es-
sential or accidental determinations), and furthermore, the object of
knowledge is primarily the nature of another, and only secondarily,
in reflexive knowledge, the nature of the knower; in contrast, the
 85
objects of desire are always perfections with respect to which a na-
ture is incomplete, and furthermore, the objects of desire are primar-
ily perfections to be accomplished by changes in one’s self, and on-
ly secondarily by changes in other things.23 In short, the objects of
desire are always future acts. The acts, or the motions which they
terminate, must be future, for were they present or accomplished,
potency or desire would be abolished. This definition of desire’s
object holds, in the case of conscious human desires, for the appar-
ent as well as for the real good. The object which is apprehended as
good to have —whether by sense or intellect, whether truly or false-
ly —is never a substance or any determinate aspect of its completed
nature (because such an object could not possibly be desired, and
nothing except the desirable can be apprehended as good), but al-
ways a change in a substance, in some respect in which it is de-
prived of actual being.

This truth is frequently overlooked or obscured because we

speak loosely of objects of desire as if they were existing things : we
name objects of desire by using the same words we use to name ob-
jects of our theoretic knowledge. Not only does this falsify the con-
tingency of objects of desire as future acts, but it also obscures the
fact that the primary objects of desire are changes in one’s self, and
not in another. When we talk about desiring this thing or that—as if
the primary objects of desire were (like the primary objects of
knowledge) things in the world about us—we tend to forget the pro-
found metaphysical truth that everything desires its own perfection.
For any substance to desire (naturally, of course) its own perfection,
the objects it desires must be those still future changes or acts by
which its being can be perfected. Furthermore, a substance cannot
naturally desire anything except its own perfection: the good which
is relative to the potency of some other substance can never be the
object of natural desire on the part of this substance. But, as we
know, the motions of other substances may be efficient or material
causes of the actualization of this substance’s potencies. Hence, the
objects of natural desire may include, secondarily , changes in other
things, in so far as these are causally involved. The good of anything
consists in the perfections which are attainable by it: these are the
primary objects of natural desire. The good for anything depends on
the already possessed perfections of other things through which they
can operate as efficient or material causes in the changes by which
the thing itself attains perfection: these are the secondary objects of
natural desire. In both cases, the object of desire is a change; in both
 86
cases, the good is a cause; but when the change is in the thing itself,
the good is a final cause, whereas when the change is in some other
thing, the good is an efficient or material cause.

If we apply these metaphysical insights to the human case, we

see that man, like everything else in the world, naturally desires his
own perfection; and since the real good is the object judged in con-
formity to natural desire, a man must desire his own perfection if he
rightly conceive his good. Considering for the moment only real
goods, the objects of conscious human desire divide into the goods
of man and the goods for man: the former consisting in those chang-
es in himself by which a man achieves greater perfection, more
complete actuality; the latter consisting in those changes in other
things which cooperate, materially or efficiently, in the change of
man himself. Thus, a man’s own acts and habits, and even perhaps
his health, are goods of his nature, perfections which terminate his
potencies; whereas, foodstuffs and clothing materials, all of the re-
sources of physical nature which man can use for his own well-
being, are goods for him. When we speak of all these other things as
good for us, we are using the word “good” attributively, for we are
not referring to the good which these things can attain, but the per-
fection they now possess, without which they could not assist us in
changing ourselves.24

In the second place, we can now understand the distinction be-

tween happiness, as the good of man, and all the other goods which
constitute it, whether these be, in themselves, goods of or for man.
The natural desire of any thing can be regarded in two ways: either
in terms of the substance as a unified whole of potencies and pow-
ers, or in terms of each of these potencies or powers, considered by
itself. Viewed in the latter way, a thing has as many distinct natural
desires as it has determinate potencies and powers; and accordingly
there is a plurality of distinct objects which are the real goods of and
for that thing. But every substance is a one, and its nature is a one,
and so its powers and potencies (determined as they are by what it
actually is) are not a mere aggregation but form a unity of ordered
parts. Viewed in its unity, rather than as diversified, the natural de-
sire of a substance must be for the good of the substance as a
whole—the perfection, not of any single power or potency, but of
them all. When it is said that each thing seeks its own perfection,
this can be understood of each power or potency as a natural ten-
dency toward its own actualization; but it can also, and must, be
 87
primarily understood of the substance itself, as a unity of diverse
powers and potencies. Though the object peculiar to each power or
potency, as a distinct natural desire, is a good of the substance, it is
not the good of the substance as a whole, but only in part. The good
of the substance as a whole must, therefore, be a complex object—
not a simple one, but a one constituted as an ordered many. Viewing
the substance as a whole, we see that its natural desire for its own
perfection is the desire for this complex object—the unified realiza-
tion of the multiple perfections terminating its various tendencies
Since a many can become one only through order, which reduces
conflict and maximizes cooperation, the perfection of a substance as
a whole must be an ordered whole of goods, in which the object of
every distinct natural desire is represented, and each is achieved
without loss or undue displacement of the others

Now, in the case of man, the potentiality of his nature as a

whole is for living, not for living simply, but for the activities of a
specifically human life A human life is a whole of many activities,
not only multiple in number, but various in type according to the
powers and potencies from which they spring. This complex whole
does not and cannot exist at any moment, for its component parts,
being temporarily disposed, cannot all coexist.The good of man,
which is the perfection of his nature as a whole (i.e., the complex
object of all his natural desires as unified), cannot be accomplished,
in the temporal domain, by any single change, or, for that matter, by
any series of acts short of the work of a whole life. This is not true
of any type of partial good, whether it be a good of man or a good
for man, the primary or secondary object of some particular natural
desire. In the light of this basic difference between the natural desire
of the whole nature for its complete perfection in time, and the di-
verse natural desires which are identified with its distinct powers
and potencies, we can see the difference between happiness, as the
whole of human goods, and the various goods which constitute that
whole. The plurality of partial goods arises from the plurality of ob-
jects that answer to the various desires which belong to human na-
ture. The unity of happiness, as the complete good, arises from the
unity of that complex object which answers to the desire of human
nature as a whole for its complete perfection. Since the end of all
desire is that which leaves nothing to be desired, happiness, con-
ceived as the termination of man’s potentiality for living humanly,
is the end; and since each partial good, which can be achieved in
something less than a complete life, leaves much else to be desired,
 88
it must be regarded merely as a part of happiness and a constitutive
means thereto.

2. THE KINDS OF GOODS AS MEANS AND ENDS

After the student had begun to see the relation between happi-
ness, as a whole of goods, and the various partial goods which con-
stitute it, he might have inquired concerning the relation of these
partial goods to one another. He might have asked whether all the
partial goods are equally good, or good in the same way; whether
one type of partial good may be a means to another, and if so,
whether this means-end relationship is the same as the means-end
relationship between the partial goods and happiness. These ques-
tions would have started another line of analysis. Let me briefly
summarize what is involved in answering them.

A distinction we have already recognized explains why all the

partial goods are not equally good or good in the same way. We
have seen that the objects of desire divide into the goods of man and
the goods for man, according as they are changes in man which per-
fect his own nature or changes in other things which are causally
related to changes in himself. Now it is obvious at once that these
two sorts of objects are not desired in the same way: one is desired
for the sake of the other. Primarily, we desire our own perfection,
and these primary objects consist of the acts and habits by which
our powers are actualized. Secondarily, we desire other things in so
far as these are operable by us and, as operable, capable of contrib-
uting to our own perfection. These secondary objects consist of all
the physical, i.e., changeable, things which man can operate upon in
one way or another, and through such operation convert into his
own being. We do not desire the perfections things already possess,
for that which is cannot be an object of desire, though it may be an
object of love . We desire rather the changes which things can un-
dergo, not for their own sake (i.e., as further perfecting these
things), but for our sake, i.e., in so far as these changes are causally
involved in the motions by which we are ourselves perfected. If one
object is desired for the sake of another, the first must be regarded
as a means, the second as an end. Hence we see that, although all
the partial goods are, severally and collectively, means to happiness
as the last end, one sort of partial good is also a means to another
sort as an end. I shall return presently to the significance of this fact
for the account of means and ends. Here I wish to consider its sig-
 89
nificance for the difference in quality between two sorts of partial
goods.

That which is good as an end (even if it be not the last end or

the whole good) has greater goodness than that which is good mere-
ly as a means. Though both may be objects of desire, one is a good
to be enjoyed and the other is a good to be used. Enjoyment is the
fruition of desire in the possession of the good which, prior to frui-
tion, was being sought. An enjoyable good is, therefore, an object of
desire which can be possessed. The only perfections we can possess
are the perfections of our own nature—our own acts and habits. We
can never possess other things except by making them part of our-
selves.25 Even so, what we enjoy is the greater actuality in us, to
which they have contributed, and not a perfection relative to their
natures. Changes in other things are objects we desire only to use in
whatever manner is appropriate to their causal efficacy. A useful
good is, therefore, an object of desire which contributes somehow to
our possession of an intrinsic good, but cannot itself ever be pos-
sessed, because it is a change extrinsic to our own nature. We may
speak of enjoyable goods as intrinsic because they are always
changes in ourselves.

This distinction between extrinsic and intrinsic goods—or, what

is its equivalent, between useful and enjoyable goods—not only in-
creases our understanding of the difference between goods for and
goods of man, but also throws light upon the traditional distinction
between bonum utile and bonum honestum bonum honestum . This
distinction is misunderstood if it is employed exclusively to divide
happiness from the various partial goods; for, although happiness,
being the total perfection of man’s potentiality for human living, is a
bonum honestum , it is not the only bonum honestum. On the contra-
ry, it might even be said that happiness transcends the distinction
between bonum utile and bonum honestum , because whereas hap-
piness is certainly not used, neither is it, strictly speaking, enjoyed
or possessed. I say “strictly speaking” to indicate that the meaning
of “enjoyment” or “possession” is not univocal when we speak, on
the one hand, of possessing good habits or good acts, and, on the
other, of possessing happiness. A good habit or a good act is some-
thing actually present and existent; but happiness can exist only as a
human life itself exists—as a becoming, as a temporal disposition of
successive parts. It can never be wholly present any more than a
whole life can be. The distinction between bonum utile and bonum
 90
honestum should, therefore, be primarily applied to partial goods, all
of which are capable of present existence. Just as happiness should
not be spoken of as a good, which would imply that it is one among
a diversity of goods rather than all of them in a certain order (for
which reason we should call happiness the good); so happiness
should be called bonum honestum only by an analogical extension
of this signification from its primary and proper application to those
partial goods which can be fully possessed and enjoyed.26

Restricting our attention, for the moment, to the variety of par-

tial goods, we must now try to classify the types of good as extrinsic
or intrinsic, as useful or enjoyable. Aristotle, it will be remembered,
said that all goods could be divided into three sorts: external goods,
goods of the body, and goods of the soul;27 and St. Thomas ap-
proved the analysis of St. Ambrose, which divided goodness into
the virtuous, the useful, and the pleasant.28 Our first problem is to
relate our twofold division to each of these trichotomies, for that
will help us, secondly, to assign each type of good to its proper cat-
egory.

The Aristotelian division presents only one difficulty. External

goods are clearly cases of bonum utile , i.e., they are all those
changes in operable, physical things which, being extrinsic to our
own nature, can only contribute to, but cannot constitute, its perfec-
tion. Goods of the soul (i.e., the acts and habits of those powers with
which we are endowed through having a human soul) are clearly
cases of bonum honestum —intrinsic or enjoyable goods. But what
are the goods of the body, such things as health and sensual pleas-
ure? In one sense, they are obviously intrinsic to our being: they can
be enjoyed and possessed. But, even so, they are not of the same
order as good acts and good habits, because they are not specifically
human. They are goods intrinsic to our animal nature (the generic
nature we share with brutes).29 There is a further complication in
the fact that health (though not sensual pleasure) appears to be use-
ful, as well as enjoyable. As useful, health is extrinsic to the goods
of the soul (habits and acts); it is a condition of human operation,
and so contributes to man’s specific perfection by way of material
causality. Although it is intrinsic in the sense indicated, we are justi-
fied in classifying health as a useful good, along with external
goods, rather than as a bonum honestum, because from the point of
view of those other intrinsic goods which are specifically human
 91
(the goods of the soul) health does not perfect man qua man, but
merely helps him to achieve his characteristic perfections.

To understand the peculiar status of sensual pleasure, we must

turn to the Thomistic classification. With respect to two of St.
Thomas’ three terms, there is no difficulty. What he calls the useful
good includes everything we have classified as bonum utile (i.e.,
external goods and health, all the objects which are goods for man,
but not goods of man qua man). What he calls the virtuous good is
identical with what we have called bonum honestum in the restricted
sense of goods of the soul (i.e., virtuous habits and virtuous acts).
But sensual pleasure as a good is neither utile nor honestum , in ei-
ther of these precise senses, even though it is clearly an intrinsic and
not an extrinsic good, and even though it can be possessed and en-
joyed. The Thomistic discussion does not help us resolve this diffi-
culty because St. Thomas fails to distinguish between pleasure as an
object of desire (i.e., sensual pleasure) and pleasure as identical with
the satisfaction of any desire “That which terminates the movement
of the appetite in the form of rest in the thing desired is called the
pleasant.”30 The pleasant in this sense is not a category of good in
the classification of goods as objects of desire; rather it is identical
with any enjoyable good that is actually possessed. But sensual
pleasure is not only enjoyable; it is desirable: it is an object of de-
sire. To achieve a proper classification of sensual pleasure, we must,
therefore, distinguish between sensible and intelligible goods The
pleasant, in the sensible order, occupies a position analogous to that
of the virtuous, in the intelligible order. It is a good which can be
desired for its own sake, and not merely as a means to some other
partial good.31

Neither of the traditional threefold classifications is adequate,

because three separable sets of criteria are involved: the distinction
between extrinsic and intrinsic, the distinction between the useful
and the enjoyable, and the distinction between the human and ani-
mal. An adequate classification of goods (which are objects of de-
sire) must make the following divisions: (a) the extrinsic and useful,
i.e., external goods; (b) the intrinsically animal and useful, i.e.,
health; (c) the intrinsically animal and enjoyable, i.e., sensual pleas-
ure; (d) the intrinsically human and enjoyable, i.e., virtuous acts and
habits. Of these, only the last type of good is, strictly speaking,32
the bonum honestum hominis. The useful may, furthermore, be dis-
 92
tinguished according as it is a means to sensible or intelligible
goods—pleasure or virtue.

In the light of this fourfold classification, we can now review an

earlier enumeration of partial goods.33 There were, we said, five
kinds of good: wealth, bodily goods, social goods, habits, and ac-
tivities. We used the word “wealth” to signify all external things
which are useful goods for man. “Bodily goods” signified such ob-
jects as health and sensual pleasure, the status of whose goodness is
also dearclear. But what are the things called “social goods”—
friends, a peaceful community, etc.? Clearly they are human, not
animal, goods, but it is not so clear whether they are useful or en-
joyable, extrinsic or intrinsic. It seems possible to view social goods
in both ways. As I shall try to show subsequently, the social goods
are, in their primary aspect, instances of the bonum honestum homi-
nis , though, in a secondary aspect, they may also be regarded as
extrinsic and useful, i.e., as means to virtue .34 With respect to the
two remaining goods (habits and activities), there is no problem of
classification: they are intrinsically human and enjoyable. There is,
however, a problem about what makes a habit or an act good rather
than bad. The student might have raised this problem by asking why
every habit and act is not good, since by definition the good is the
actualization of a potency, and every habit and operation is such an
actualization. What, then, is it which makes some acts and habits
good (virtuous) and others bad (vicious)?

Although this is an extremely difficult problem (perhaps, the

most difficult question about the relation between the metaphysical
and moral significations of “good”), I shall briefly suggest a solu-
tion in terms of two points already made. In the first place, the real,
as opposed to the apparent, good is always the natural good as ap-
prehended. Now, although any habit or act may be apprehended as
an apparent good, only certain habits and acts can be truly appre-
hended as goods, because they alone conform to the natural good of
our powers. As themselves determinate natures, as well as being
determinate properties flowing from our essential nature, the several
powers do not tend toward any actualization as good, but only that
actualization which corresponds to each power’s nature. This is
most simply seen in the case of the intellect as a power of knowing
the truth about things. Because it is by nature a power of knowing ,
acts and habits of knowledge are its natural good, whereas acts and
habits of error are naturally evil, and hence can be apparent, but
 93
never real, goods. This accounts for the distinction, in the sphere of
intellectual habits and acts, between the virtuous and the vicious.
There is another group of habits and acts, which are called “moral”
as opposed to “intellectual” (not moral as opposed to immoral), be-
cause they are not habits and acts of knowing or thinking, but habits
and acts of desire or social conduct. Here the criterion for the dis-
tinction between good and bad lies in the fact that our powers of
conscious desire or behavior are moved by the apprehension of
some object as good. But our intellectual judgments concerning the
good may be true or false, and, accordingly, the good apprehended
may be real or apparent. Hence only those acts or habits of desire
(whether on the part of the will or of the sensitive appetite) are vir-
tuous, which are determined by the apprehension of the real good.
The same must be said for the virtuous in the sphere of social be-
havior. The traditional maxim that the virtuous, in the case of moral
acts and habits, is that which is in conformity with reason must be
interpreted to mean, not the power of reason simply, but the power
of reason as perfected by good habits, by knowledge rather than er-
ror. When it is remembered that the truth of practical judgments
(about good and bad) is by conformity with right desire, there will
appear to be circularity; but this is avoided by the fact that con-
scious desire is itself right by conformity with natural desire. In
short, the natural good is the ultimate criterion whereby the virtuous
and vicious are distinguished.

In the second place, human powers are principles of many acts

and capable of multiple habituations. Furthermore, the unity of man
as a substance implies, as we have seen, the unity of his or her pow-
ers as ordered to one another, and hence the unity of their actualiza-
tion through cooperation and co-habituation. These facts explain
why vicious habits are bad, even though each by itself is the actual-
ization of a power. For vicious habits conflict with one another, as
well as oppose the formation of virtuous habits, whereas virtuous
habits are, inter se , thoroughly harmonious and mutually support-
ing. Hence vicious habits prevent the maximum actualization of
human powers, whereas virtuous habits tend toward such maximiza-
tion. Thus we see (in the case of habits, though not with respect to
acts) how the moral criterion of virtue and vice relates to the meta-
physical criterion of goodness as actuality of being.34a The perfec-
tion of a human being as a whole requires the maximum actualiza-
tion of his or her powers. Granting that habits are actualizations, the
“metaphysical goodness” of any habit taken by itself is specious
 94
because neither powers nor their habits exist separately; hence only
morally good habits are metaphysically good as making for the per-
fection of a human being as a whole.

I have now outlined the sort of answer which must be given to

the student’s question about the kinds of good, and their inequality
due to differences in type. But he also asked about means and
ends—whether one partial good is a means to another of different
type in the same way that all partial goods are means to happiness.
This question must now be briefly answered.

At one point in our earlier discussion, we distinguished between

two sorts of means: constitutive means (parts of a whole) and func-
tional means (one partial good as means to another).35 The differ-
ence between these two sorts of means is readily grasped by com-
paring, first, their respective ends and, second, the relation of the
means thereto. In the case of constitutive means, the end is always
happiness, or the whole good, and here the means is included in the
end it serves, serving it by effecting its partial realization. In the
case of functional means, the end is always another partial good,
and here the means lies outside the end it serves (as the efficient
cause lies outside the effect), for it serves its end by working as a
condition precedent to the realization of the end. So much is clear;
but the student may wonder whether all functional means are useful
goods, and whether their ends are always one or another sort of bo-
num honestum. Unfortunately, the truth is not that simple. One type
of good may function as means to another good of the same type, as
well as be a means to goods different in type. Thus, wealth may be a
means to health, or health to wealth, and both to pleasure; and, with-
in the large group of goods we have classified as wealth, instru-
ments of production may be means to consumable goods; further-
more, virtuous acts may be means to good habits, virtuous habits
means to good acts, and both may be means to social goods, as well
as be served by them. In the order of partial goods, there is a great
complexity in the functional relationships of diverse types of goods
as means to ends. This complexity can, however, be reduced some-
what by the following observations: first, that some partial goods
are never functional means, but always ends, notably, sensual pleas-
ure and certain good acts which are good quite apart from the for-
mation of habits of conduct; second, that the same good may func-
tion as a means in relation to some other good as an end, and be an
end served by some other good as a means, as health, for example,
 95
in relation to virtue and to wealth; third, that even when functioning
as means certain goods are essentially ends, such as virtuous acts
and habits, whereas certain goods are essentially means even when
functioning as ends, such as wealth and health; finally, that only
those partial goods which are essentially ends are essentially consti-
tutive of happiness, whereas those which are essentially means are
accidentally constitutive of happiness.36

Within the domain of partial goods, it is also important to dis-

tinguish between the universal and the particular, i.e., between the
type as such and singular instances of each type. Failing to do this,
we cannot understand the constitution of happiness; for when it is
said that happiness includes all good things, what is meant is goods
of every type, and not every possible instance of each type of good.
This must not be construed, however, to mean that happiness is ac-
tually constituted by types of good, rather than by singular goods
actually possessed, for we know that universals do not really exist
in the moral order, any more than in the physical order, except
through singular embodiments. Happiness may be conceived (i.e.,
exist in the ideal or conceptual order) in terms of the order and vari-
ety of types of good (i.e., universals); but happiness cannot occur
(i.e., exist in the real order of living itself) except through the actual
possession of goods, and these must all be singular instances of the
several types. We are compelled, therefore, to define a third kind of
means. As the various types of good are the constitutive means-in-
general of happiness, so singular instances of each type are the con-
stitutive means whereby that type of good is gradually realized in
the course of life. These means-in-particular (this commodity, this
pleasure, this degree of virtuous habit, this good act, etc.) are thus
each directly constitutive of a special type of good, and through
serving as means in this way, they are indirectly constitutive of ac-
tual happiness. Whereas the constitutive means-in-general are nec-
essary means of happiness, the constitutive means-in-particular are
contingent means, for no one of them is indispensable to the realiza-
tion of its type, or of happiness as a whole, as, on the contrary, eve-
ry type of good is indispensable to the realization of happiness.37

There is, finally, a fourth type of means—the most important of

all, because it both epitomizes what is involved in the notion of
means (i.e., something practically useful) and combines the highest
degree of utility with the greatest excellence a means can have as an
intrinsic good. I shall call this fourth type of means the generative or
 96
productive means, the end in this case being happiness itself. As one
partial good may be functionally related to another as means to end,
by the productive relation of efficient causality, so there is one sort
of partial good—and only one—which is functionally related to
happiness itself by direct efficient causality. I am referring here to
the cardinal virtues—the moral virtues and prudence. This type of
good (virtue simply or absolutely, in contrast to intellectual virtue,
except prudence, which is virtue only relatively or secondarily) is at
once a bonum honestum and also the most useful sort of productive
means, precisely because what it serves functionally as an end is not
another partial good, but happiness itself. Before considering the
relation of virtue to happiness, let me summarize the four sorts of
means-ends relationships, according to the principles which deter-
mine them:38

I. Constitutive means : realizing the whole of which they are

parts.

A. Constitutive means-in-general: realizing happiness as a

whole.

B. Constitutive means-in-particular: realizing a universal type of

good directly, and happiness indirectly.

II . Functional means : productive or generative of the end they

serve.

A. Functional means, generally considered: any partial good

which is productive of another partial good as an end, whether same
or different in type.

B. Functional means, specially considered: cardinal virtue, the

only partial good which is directly productive or generative of hap-
piness itself as the end.

3. VIRTUE AS PRINCIPAL MEANS TO HAPPINESS

When the student understood happiness as the end in relation to

which all other goods are means in one way or another, he would
 97
also understand that happiness is the first principle of moral theory.
But he might still be unable to see how the first principle theoreti-
cally determines the subsidiary principles of ethics; and, more that
that, he might remain quite perplexed about how the theory worked
in practice. Knowing what happiness was, he might still wonder
how to become happy.

The theoretical question is the easier to answer. If the first prin-

ciple of any practical science is the end, the secondary principles are
the means. Just as the conclusion is implicitly contained in the
premises, so knowledge of the means is contained in knowledge of
the end. We cannot know what happiness is without knowing its
constitutive means: according as people differ in their conceptions
of the order and variety of goods, so will they differ in their defini-
tions of happiness; and the truly adequate definition of happiness
will be adequate in its enumeration of the partial goods and true in
its ordering of them. But the constitutive means, taken as a whole,
are identical with happiness: only by a purely formal definition of
happiness, can it be conceived apart from the goods which consti-
tute it.39 Hence the constitutive means are not the secondary princi-
ples of moral theory. The secondary principles must be functional
means. Just as for any partial good as an end, there must be genera-
tive means, so for the whole good which is the end, there must be
means specifically able to produce happiness. Furthermore, from
knowing what happiness is, we should be able to determine the na-
ture of the means productive of it. When we make this determina-
tion, we shall have solved the only remaining problem of ethical
theory, and this will enable us to consider its significance for prac-
tice, i.e., how human beings become happy .40

The deductive determination of the secondary principles of

moral theory can be simply accomplished. Happiness is an order of
goods. Throughout the course of life, the various partial goods
which constitute happiness cannot be adequately obtained unless
they are sought in the right order—an order which proportions them
according to their worth as goods and respects their functional rela-
tionships. But no good whatsoever can be obtained except through
activity. And activity with respect to goods as objects of desire is
determined by acts of the desiderative faculties, either the will or the
sensitive appetite. Hence we see that happiness will be obtained on-
ly if activity flows from right desire. This becomes a self-evident
truth when we understand “happiness” as a right order of desirables
 98
and “right desire” as a right ordering of desires . But good habit is
the proximate principle of the good operation of any power. Good
habituation of the appetitive faculties is, therefore, the source of
good acts of desiring and these determine good activity with respect
to objects of desire. We must conclude that good appetitive habits
are the principal means to happiness—the means directly productive
of it. As earlier analysis showed, the moral virtues are good habits
of desire and of behavior in accordance with right desire. When
each moral virtue is understood as a habit of rightly ordering one’s
desires in relation to happiness as a right order of goods or desira-
bles, it will be seen at once that no one of the moral virtues can be
possessed except in the same degree that all are possessed. Moreo-
ver, since habits of right desire are formed and become operative
with respect to particular goods, they depend upon the making of a
right choice among the particular goods which are contingent con-
stitutive means. Prudence as a virtue of practical reason is a habit of
rightly making such choices. Hence, just as one moral virtue cannot
exist without all the others (or can exist only in the same degree), so
no moral virtue can exist without prudence, or it without them: right
desire for the end (i.e., desire for the right end) is indispensable to a
right choice of particular means, since these are good only as realiz-
ing the kinds of goods which are the constitutive means-in-general
of happiness; and right choice of the particular means is indispensa-
ble to the formation and operation of habits of right desire for the
end. If now we use the word “virtue” to signify the cardinal virtues
(the several moral virtues and prudence) in their integrity, we can
formulate our conclusion by saying that virtue is the principal
means to happiness—or, more strictly, the only means adequate to
the task of directly generating the end.41

In the order of theory and intention, the end precedes the means,
for it is by knowledge of the end that we know and can intend the
means productive of it. But, in the order of practice and execution,
the means precede the end, for in order to become happy we must
first acquire the productive means. Whereas the constitutive means
are the end so far as it is realized at any time, the productive means
are the end in process of becoming realized. Because the end deter-
mines the productive means (in the order of intention) and the pro-
ductive means determine the end (in the order of execution), it is
equally true to say that as man conceives the end, so will his or her
character be; and to say that as a man’s character is, so is the end for
him.42 As reciprocally determinative of each other, the end and the
 99
productive means are correlative: the order of goods or desirables is
correlative with the ordering of desires. If the end at which a man
aims is only the apparent, and not the real good (because of some
inadequacy or falsity in his conception of the variety or order of par-
tial goods), then his habits of desire may appear to be virtuous, but
they will really be vices; and conversely, if a man be habitually vi-
cious in any respect, that disorder of his desires will be productive
of apparent, but not real, happiness. The man of really good charac-
ter is not one who possesses some virtues and some vices, for no
real virtue can be conjoined with any vice. Only the cardinal virtues
in their integrity can constitute a really good character, just as only
an adequate and true ordering of all partial goods can constitute real
happiness. Thus we see the two ways in which happiness exists
practically . Theoretically, happiness may exist in the mind as a
conception of the end, whether true or false. But practically, it exists
either (a) objectively in a human life in so far as it is perfected by
the possession of all good things, or (b) subjectively in a human
character in so far as it is perfected by real virtue. This is just anoth-
er way of saying that the productive means are the end in the pro-
cess of becoming. But saying it this way enables us to see that
though the complete actuality of happiness is attained only in a
complete life, it is virtually present, as in its causes , in a good char-
acter. Objectively, happiness exists only as a process of becoming
exists, which means that it cannot exist at any moment; whereas
subjectively, happiness exists at any moment to whatever extent its
efficient causes exist. In the first sense, therefore, we cannot say that
a man was happy until he is dead; in the second, we can say that he
is happy while he is alive, if we know that he has a virtuous charac-
ter.

Human nature, through its powers, is the principle of operation,

of all the activities of human living. A man’s character is his second
nature, especially in view of the integrity of virtue as a cohabitua-
tion of his powers. As human nature is the principle of any sort of
operation, good or bad, and hence of good or bad living, so a good
character, as a man’s second nature, is the principle of good opera-
tions, or happiness as a life well lived. In proportion as his nature is
perfected by virtue, so will a man’s life be perfected by happi-
ness.43

Only two problems remain, one theoretical, the other practical.

The theoretical problem concerns the intellectual virtues—habits of
 100
knowledge and skill. Clearly, they are among the partial goods es-
sentially constitutive of happiness; clearly, they are, in functional
relationship to other partial goods, both means and ends—both gen-
erative of good intellectual activity and generated by it, as well as,
indirectly, by other means, such as health, social goods, and moral
virtue. But just as clearly, we know, from what they are and from
what the end is, that the intellectual virtues (always omitting pru-
dence) are not productive means with respect to happiness, except
indirectly, perhaps, in so far as they are involved in the formation of
moral virtue. That is why, in the classification of good habits, the
moral virtues and prudence are called cardinal, or virtues simply,
whereas the speculative virtues and art are called secondary, or vir-
tues relatively. It is not merely that the intellectual virtues confer
only an aptitude for good work, but not an habitual performance of
good acts. The deeper reason is that the intellectual virtues perfect a
man’s nature only in certain respects, making him a good scientist
or a good artist, and able to operate well in these domains of work;
whereas the cardinal virtues perfect a man’s nature with respect, not
to some limited good, but all good things or happiness, making him
good as a man, and also making him live his whole life well. If the
essence of any virtue as a partial good lies in its being a generative
means, then those habits which are productive of happiness are vir-
tues simply; in contrast, those habits which are productive of only
another partial good (i.e., intellectual activity) are virtues only rela-
tively.44

This fundamental insight about the intellectual virtues helps us

to resolve an apparent conflict between two traditional definitions of
temporal happiness. The apparent inconsistency originates in Aris-
totle’s Ethics . In Book I, Aristotle defines happiness as activity in
accordance with perfect virtue, in a complete life, accompanied by a
minimum sufficiency of the goods of fortune. In Book X, happiness
is defined as one sort of activity, contemplative activity or the activ-
ity of the speculative virtues. If the second of these definitions be
taken as saying what constitutes happiness, then it is clearly contra-
ry to the first definition, and clearly false; for the intellectual virtues
are certainly not “perfect virtue” or virtue simply, and if happiness
consisted in nothing but contemplation or speculative activity (all
other goods being merely auxiliary or antecedent thereto), then such
happiness would falsify the formal definition of the end—that
which, when possessed, leaves nothing to be desired. The first defi-
nition of happiness is, on the other hand, verified by this same crite-
 101
rion: if, in a complete life, attended by good fortune, a man acts in
accordance with perfect virtue (i.e., the integrity of the cardinal vir-
tues), his life will become happy through the possession of all good
things. Now, “all good things” includes intellectual virtues, and con-
templation, or good speculative activity, in accordance therewith.
The man of perfect virtue so orders his desires, and so acts, with
respect to the order of partial goods, as to possess intellectual virtue
in some degree, and put these virtues to use in good activity. Hence
the second “definition of happiness” can be reconciled with the first
on the condition, of course, that it be subordinated thereto. This is
done by seeing that it is not, strictly, a definition of happiness at all,
but rather a statement of what is the highest good in the order of
partial goods.45 Because the intellect is man’s highest power, the
operations which flow from its good habituation are the highest
form of human activity. In the order of partial goods, then, contem-
plation or speculative activity is highest; intellectual virtue is better
than moral virtue because directly generative of such activity; and
all the other partial goods, both extrinsic and intrinsic, must be or-
dered to the achievement of intellectual virtues and to their exercise.
Even the moral virtues and prudence, regarded as generative means
within the order of partial goods, must be regarded as aiding the at-
tainment of wisdom and the activity of contemplation.46 But the
moral virtues and prudence do not discharge their total generative
function within the order of partial goods. As productive of happi-
ness, they also stand outside the order of goods— functioning
uniquely as the efficient cause whereby all good things become pos-
sessed in a life well lived. When we speak of happiness as a life of
virtue we must, therefore, mean two things; first, that it is a life gen-
erated by activity in accordance with cardinal virtue; second, that it
is a life which includes the highest form of activity in accordance
with intellectual virtue.47

The other problem which remains concerns the acquirement of

virtue in its integrity. The cardinal virtues which generate happiness
must themselves be generated by good acts. How this can be ac-
complished is the central practical problem in ethics. It is solved
only by rules which are inexact and relative to different types of
individuals living under different circumstances. The theoretical
level of moral philosophy goes no further than definition and analy-
sis of ethical principles—happiness and virtue. But moral theory,
even if it be perfectly formulated, is inadequate to direct a man in
the acquirement of virtue and the attainment of happiness. A man
 102
may know, theoretically , all about happiness and virtue, and never-
theless fall far short of being virtuous and happy, because he does
not know how to apply this knowledge in his own life. Moral theory
is practical only in the sense that it underlies the formulation of
rules which are genuinely practical as directive of conduct. On the
one hand, without theoretically true principles, practically wise rules
could not be formulated; for unless the nature of virtue and happi-
ness is understood, how can particular acts be well regulated toward
their achievement? On the other hand, without practical rules, the
truest principles are inadequate in the practical order, and even
sound rules require the work of prudence for their application. But
these practical rules are not capable of universal or exact formula-
tion; and prudential judgments are seldom articulated even by the
man who makes them.48

The man of virtue has practical knowledge of the means and end
of life. Through his moral virtues he has a connatural knowledge of
the good at which these virtues aim; and through his prudence, he
has a living habit of regulating each particular act in accordance
with the demands of virtue. But if having virtue is indispensable for
knowing practically how to live well—if virtue itself mediates be-
tween true theory and sound practice—then how can virtue be ac-
quired? Since theoretical knowledge of principles is insufficient,
since practical knowledge of some sort seems to be both prerequi-
site for the formation of virtue and also consequent thereupon, there
appears to be a vicious circle here, or at least a mystery about how
anyone becomes virtuous. So far as I know, there is only one solu-
tion to this problem, in the domain of purely natural causality. One
man of virtue, thereby having practical knowledge, not only for the
guidance of his own life, but for the direction of others, must be the
efficient cause of the formation of virtue in another man, so that the
latter gradually becomes able, through the practical knowledge
which virtue gives, to regulate his own life. It remains true, of
course, that the principal efficient cause of virtue must be a man’s
own acts, performed voluntarily and directed with prudence by his
own reason. The moral preceptor is, therefore, like any other teach-
er, at best an accessory or cooperative cause: moral training, like
intellectual teaching, is an art of cooperating with nature, aiding the
primary cause which is reason’s own activity. So far as habituation
is concerned, the efficiency of moral training can go no further than
the formation of what Aristotle called “inchoate virtue”—moral
habits formed in one person as the result of the regulation of his acts
 103
by another. The growing person may thus come to possess inchoate
virtue before he knows the principles of moral theory and before he
has practical knowledge whereby to regulate his own conduct. Such
inchoate virtues, nevertheless, operate as the seed of genuine virtue,
for they are a source of good conduct; and as the result of good acts,
though these be initially performed under external regulation, con-
natural knowledge of the good accrues, practical knowledge of how
to act well grows, and prudence itself develops.48a By this route, it
is possible for a man to become virtuous without ever being learned
in moral theory. No one ever becomes virtuous simply by learning
moral theory, though moral theory may confirm a man in virtue, or
remotely facilitate his practice of good acts.

In the light of the foregoing, we understand why Aristotle made

ethics a branch of politics—at least to the extent of viewing the leg-
islator as a moral preceptor, and the just laws of a good society as
causes of virtue and happiness. The government of others, whether
domestic or political, can be best accomplished by those who know
the principles of ethics, and who also, through being themselves
virtuous, have practical knowledge whereby to formulate sound
rules. The theoretical study of ethics is, therefore, primarily for the
government of others, and only secondarily for the enlightenment
and guidance of those who have already begun the life of virtue.49
If moral principles are primarily put into practice by the statesman,
we should be able to find in them the foundations of political phi-
losophy.

NOTES TO CHAPTER VI

15. See Chapter III, p. 39.

16. This issue was briefly considered at one point in the dialectic.
See Chapter III, pp. 43-44: If “the objects we have called goods are
good only because they are desired”; if “there is always a relativity
of the good to actual desire.’’ then “we shall never be able to say
 104
what men should desire, which is central to moral knowledge as
normative or prescriptive. In order to get beyond a mere description
of what men do desire, we must somehow show the student that the
objects men desire, they desire because they judge them to be
good.” But in the discussion itself, this problem was inadequately
solved by proposing an ultimate end, which in fact all men do de-
sire, namely, to live well; and in terms of this end, we argued that
men should desire whatever is necessary for the attainment of this
end. This is a partial solution, since it is true that the goodness of
means may be derived from the goodness of the end they serve; but
it leaves two major questions unanswered: (1) is the end good simp-
ly because all men do desire it, or because it is the ultimate good
which all men should desire? and (2) are any of the means good in
their own right, and apart from being means to the end, so that they,
too, should be desired because they are intrinsically good (i.e., de-
sirable)? The student permitted the discussion to go on at this point,
although he could have stopped it by insisting upon a deeper exami-
nation into the meaning of the good as an object of desire.

17. See St. Thomas Aquinas, Summa Theologica, I, qq. 5, 6. where-

in these traditional maxims are not merely repeated but interpreted
in such a way as to indicate the solution of the problem. It should be
noted that in Q. 5, A. 1, St. Thomas says: “the essence of goodness
consists in this, that it is in some way desirable” (Italics mine). Just
as intelligibility is an aspect of being prior to its being actually un-
derstood, so desirability is an aspect of being prior to its being actu-
ally desired (i.e., by an elicit appetite).

18. St. Thomas makes a fourfold distinction: between being simply

and being relatively, good relatively and good simply. A thing is a
being simply according to its first actuality or essence; a thing has
being relatively according to its second acts, or those accidental de-
terminations which accrue to it from its own operations or its being
acted upon. But a thing is said to be good relatively with respect to
the actuality of its nature (i.e., its essence and powers), whereas it
has goodness or perfection simply according as its nature is com-
pleted in being by second acts. See Summa Theologica I, 5, 1, ad 1.
In the light of these distinctions, the goods which are the objects of
natural desire are the actualities which constitute the relative being
of a thing; the good relatively is never an object of desire, unless it
be on the part of prime matter, which is the potency for being simp-
ly.
 105

The interpretation of natural desire and of being as good in relation

thereto must be differently made for non-living and living substanc-
es, since the former, in all their accidental motions, pass from con-
trary to contrary, and hence never increase in perfection of being;
whereas the latter, having impassible powers and being capable of
immanent activity, can acquire perfections without privation of con-
trary forms, and can therefore, grow in perfection of being. This is
especially true of man whose specifically human powers are capable
of habituation.

19. This account would be untrue only if there were no world of

creatures at all—no imperfect beings subject to natural desire be-
cause of their imperfect natures. Furthermore, the actual separation
between desire and the good which quiets it, can occur only in a
world of temporal beings, beings in motion, for where change is not
possible, there either desire must be forever unrequited or desire
does not exist at all because the good is possessed. This indicates
that our metaphysical understanding of the good as being in relation
to desire is primarily in terms of being in motion—the realm of cor-
poreal and temporal creatures. Goodness in the domain of spiritual
and aeviternal creatures, and the goodness of God, are dimly intelli-
gible to us only by remotion and analogy. The Divine goodness is as
infinitely unlike the goodness of changing things as the pure actuali-
ty of Divine being differs from the actuality of beings composite of
potency and act.

20. It is important to mention, but not necessary to develop here, the

distinction between the two conscious appetites which man possess-
es: the sensitive appetite, the power of tending toward sensible ob-
jects which the sensitive powers estimate to be good or pleasant;
and the intellectual appetite, already defined. Only the latter is spe-
cifically human (i.e., possessed by man alone), whereas the former
the brutes also possess. Animal desires (acts of the sensitive appe-
tite) are distinguished from natural desires by the same two criteria
which distinguish human desire. It should be noted that the sensitive
appetite is not moved by a mere apprehension of the sensible object,
but, as in the case of intellectual appetite, only by an estimation of
the object apprehended as good or bad, pleasant or unpleasant—
whether this estimation be the work of instinct (conditioned or un-
modified) as in the case of brutes, or the work of the cogitative
power in the case of man, i.e., that power of interior sensitivity
 106
which is usually called “the particular reason,” because it operates
with the help of reason in judging sensible things as fitting or harm-
ful. The distinction between intellectual and sensitive appetite is
important, not only because of its bearing on the imperfect voluntar-
iness of brute behavior in contrast to the perfect voluntariness and
freedom of specifically human acts; but also because the conflict of
these two appetites in man explains how man can act contrary to
true rational judgments concerning what should be done or sought.
In the latter connection, the distinction bears on the problem of the
real vs. the apparent good.

21. Natural human desires are, of course, natural in precisely the

same sense as the natural desire of the stone or of the plant. They
are unconscious: the objects of natural human desire are not, as
such, apprehended, judged or estimated. And natural human desires
are determinate as potencies or powers, not as acts. This may help to
explain what we mean when, in the case of man, we speak of his
“unconscious desires.” Such desires are not the acts of his con-
sciously determinable appetitive powers. They are rather identical
with the striving or tendency of his powers themselves—all the
powers vegetative, as well as sensitive and rational, and in the latter
case, apprehensive as well as appetitive. It might be clarifying to use
the word “needs” to name the natural appetite of man, in contrast to
the word “desires” to name his conscious appetitive acts. This
would enable us to deal with obvious phenomena of human behav-
iour—cases in which men are impelled by wants without knowing
what they seek because the want has not yet been elevated to the
level of desire by knowledge of the appetible object.

22. It does not follow that a man who, judging aright, desires what
is really good, will necessarily act accordingly; for in the conflict
between his sensitive and his rational appetites, an apparent good
may dominate his conduct at the moment of action. See Aristotle’s
account of incontinence in the Ethics, Bk. VII, 3. All that is here
being said is that the basic truths which constitute ethical theory—
truths about the variety and order of objects which are really good
for man—rest on metaphysical and psychological knowledge, the
former concerning the foundation of goodness in being, the latter
concerning the foundation of the real human good in the nature of
man. The real human good is the natural human good as apprehend-
ed and thus become an object of conscious desire; the human good,
taken without qualification, is not identical with the natural human
 107
good, for it includes all the apparent goods which are the objects of
mistaken judgments. If the word “good” be restricted in its signifi-
cation to mean “real good,” then the metaphysical statement that the
good is what all desire is true only for the objects of natural desire,
human or otherwise; it certainly is not true for the objects of con-
scious human desire, since what all men desire, consciously, may be
either really or apparently good.

23. This distinction between the object of desire and the object of
knowledge accounts for the radical difference between theoretical
and practical truth: only when the object is an already determinate
nature can truth be in the intellect by conformity to what is; since
the object of desire is also the object of our practical judgments, and
since this object is always a future contingent event, i.e., a change to
be accomplished, the truth of practical judgments cannot be by con-
formity to what is, but must be by conformity of the judgment to
right desire, i.e., desire for a real good, an object of natural desire.

Furthermore, it should be noted that the true is not an object; rather,

being an act of knowledge, it, like the act of desire, has an object.
The good, on the other hand, is an object both of (practical)
knowledge and of desire. The good is not only an object, but an act
of being. The radical difference between the true and the good is
that the true is an act in the order of intentional existence, whereas
the good is an act in the order of real existence.

24. I shall return to this distinction between goods of man and goods
for man in a later discussion of means and ends. This distinction,
made in terms of the axiom that everything seeks its own perfection,
raises profound ethical problems which cannot be discussed here:
for example, the whole problem of altruism and selfishness. Is man
ever obligated to work for the good of any other thing, inanimate or
animate, animal or human, and if so, how is this a real good, i.e., an
object of natural desire, if it is not a good for man as well as a good
of the other thing? This problem becomes particularly acute with
respect to the relation of men in economic and political associations.
In theology, this distinction is exemplified by the consideration of
the Divine being and goodness as the ultimate good for man, where-
as the vision of God is the ultimate good of man’s soul. The theo-
logical case indicates furthermore, that what is good for man may be
superior to him in being, and its efficient causality in bringing about
human perfection may be without human aid, except in the order of
 108
dispositive causality; whereas the material things which are good for
man are not only inferior in being, but usually, if not always, require
efficient causality on the part of man (the work of his art) to render
them useful.

25. This metaphysical meaning of “possession” must be sharply dis-

tinguished from the economic or legal meaning, in terms of which
we have “property rights” with respect to other things.

26. This point profoundly illuminates the distinction between tem-

poral and eternal happiness, for the latter is the good which is bo-
num honestum eminenter. By comparison with eternal happiness, all
partial goods can be called bonum honestum only analogically; and
temporal happiness itself can be called “happiness” only by analogy
with that happiness which can be fully possessed and enjoyed.

27. Ethics. I, 8, 1098b 12-15.

28. Summa Theologica, I, Q. 5, A. 6.

29. Strictly speaking, they are not goods of the “body” in the sense
in which this might be opposed to goods of a living thing, goods it
can possess because it has a soul and is alive. The distinction be-
tween “goods of the body” and “goods of the soul” must be under-
stood, rather, in terms of the distinction between man’s specifically
human powers and those powers which man has because the ration-
al soul virtually includes the sensitive and vegetative souls.

30. See note 28

31. This is not inconsistent with the truth that sensual pleasure, like
every other partial good, is a constitutive means of happiness.

32. The foregoing analysis requires us to make one further discrimi-

nation of aspects of the good in relation to appetite. Appetite may be
in a state of motion toward an unattained object, in which case we
speak of the good as an object of desire, and this is the good in its
primary aspect. Appetite may be in a state of rest through posses-
sion of the object desired, in which case we speak of the good as the
satisfactory, as the object of enjoyment, or the fruition of desire. But
appetite may also be in a state of simple ordination toward an ap-
prehended good, in which case we speak of the good as an object of
 109
love. Unlike the object of desire, the object of love is an existent
perfection—the good which is convertible with the being a thing
already has. Speaking theologically, God in Himself is an object of
love, not of desire; the Beatific Vision is an object of desire and of
enjoyment. Moreover, God, whose perfect being is perfect good-
ness, is the primary object of love: whatever else is loved is loved
secondarily and by relation—ourselves and our neighbors as our-
selves. Since what is enjoyed is a perfection possessed, the objects
of love and of enjoyment may be the same (though different in as-
pect) in the case of loving ourselves, or others as ourselves, or they
may be the objective and subjective aspect of the same, as in the
case of God and the vision of God. And as uti is divided against
frui, so are those objects which can only be used, distinguished from
those which can be loved.

It might be supposed that we would achieve clarity if we defined the

object of love not as the good, but as the beautiful. Love is that
mode of appetitive determination which abstracts from both the
presence and absence of its object, and is thus distinguished from
desire (for the absent object) and enjoyment (of the present object).
If the good is the object of desire and enjoyment, then the good can-
not be, in a strictly univocal sense the object of love: for, since mo-
tion and rest are exhaustive, desire and enjoyment exhaust the
modes of appetite with. respect to the good, when appetite is simply
regarded; love must, therefore, be a mode of appetite relative to ap-
prehension, or a mode of cooperation of apprehension and appetite
(of intellect and will in amor intellectualis). Now the beautiful is not
the object of either intellect or will separately, but of both in ordered
conjunction, for it is that which pleases (is enjoyed) upon being seen
(known), as, in contrast, the good is that which is enjoyed on being
possessed. Hence, the object of love is the good relatively, not simp-
ly, which means it is the substantial perfection of anything as appre-
hended, or, in other words, it is the beautiful. As its object is not the
good simply, neither is love a mode of appetite simply, but rather of
appetite (enjoyment) relative to apprehension of a purely theoretic
sort. This explains the theological truth that it is better to love God
than to know him, and better to know things than to love them, be-
cause the goodness of love is from its object, whereas the goodness
of knowledge is in our mode of knowing. As the object of love, God
is not the transcendent good, but the transcendent beautiful (the
splendor and effulgence of perfect being) so in loving ourselves and
our friends we enjoy what we apprehend as beauty of character (a
 110
work of prudence), just as enjoy the beauty of a work of art beheld.
Love of the beautiful in human beings and human life explains
much that desire for the good (primarily our own perfection) cannot
account for: it is needed to solve the problem of altruism vs. selfish-
ness (see note 24), as charity is needed to supplement justice in so-
cial action.

But the object of love, unlike the beautiful, is not an object of com-
pletely disinterested enjoyment, apart from desire and action. There-
fore, the object of love must be defined as the good of another per-
son; as such it may also be an object of desire comparable to the
desire for one’s own perfection. The truth that everything naturally
desires its own perfection must be understood in such a way that the
desire for the perfection of an alterego, i. e., a loved object, is in-
cluded. Just as we can never possess other things without making
them parts of ourselves, so the perfection of a loved object is a per-
fection we vicariously enjoy as if it were the perfection of our own
personality. Cf. The Theory of Democracy, Part II, Section 2, in The
Thomist, III, 4. Although the object of love is strictly the good and
not the beautiful, there remains a sense in which the loved object is,
apart from desire and as simply apprehended, beautiful. To under-
stand the object of love, either as good or as beautiful, it is neces-
sary to distinguish the “love” which names a passion and the “love”
which names a motion of the will; for in the sphere of sensitive ap-
petite love is strictly selfish, being the origin of desire for posses-
sion, for incorporation of another into one’s own being, whereas in
the sphere of will, love is entirely altruistic, being correlative with
desire for the good of another as other and reaching fruition in the
apprehension of that other’s perfection as beautiful.

33. See Chapter III, pp. 40-42

34. See Chapter VII. One indication that social goods should be
classified primarily as intrinsic and enjoyable goods is that friends
are objects of love—having a goodness to be loved as we love the
goodness of ourselves. Cf. note 32. Another indication is that the
goodness of a community, domestic or political, is a common good
in the sense that it is shared by the members: it is not only a com-
mon object of desire, but also a fruition commonly enjoyed. See The
Theory of Democracy, Part II, Section 2
 111
34a. Cf. St. Thomas’s discussion of the meaning of “good” in the
case of good habits: Summa Theologica, I-II, 55, 4 ad 2.

35. See Chapter IV, p. 48. The second sort were not then called
“functional means.”

36. These last two points are of great importance, for they apply to
the analysis of functional means the deeper principles whereby
goods are distinguished as utile and honestum, and as human and
animal. In one sense, of course, every functional means is useful,
but a good may be useful functionally (in relation to some other
good it serves) without being essentially useful in its own type( i.e.,
not being a good for man, but a good of man). Furthermore, since
happiness is specifically human, only those goods are essentially
constitutive of it, which, as partial goods, are bonum honestum hom-
inis; extrinsic goods, or intrinsically animal goods, are only acci-
dentally constitutive of happiness.

These two points convey Aristotle’s insights about means and ends.
He distinguished between goods which are mere means, goods
which are both means and ends, and the good which is absolutely
and simply the end. A mere means is any bonum utile, whether or
not it also functions as an end; a means which is essentially an end
is any bonum honestum: with the exception of pleasure in the sensi-
ble order, and contemplative activity in the order of intelligible
goods, happiness is the only good which never functions as a means,
and it is the only end simply and absolutely, because even pleasure
and contemplation are constitutive means of human happiness. Aris-
totle also distinguished between antecedent and constitutive means,
and this is the distinction we have made among partial goods (all of
them constitutive in a broad sense) according as they are accidental-
ly or essentially constitutive of happiness. In other words, a mere
means is, as constitutive, accidental to human happiness, whereas a
means which is also genuinely an end is essential to the constitution
of human happiness. These two basic points go much deeper than
the distinction among partial goods as functional means and ends.
See Aristotle, Ethics, I, 7. Cf. John Dewey’s attempt to regard every
good equally as an end, in Reconstruction in Philosophy, New York,
1919: Ch. VII.

37. All the objects of choice are means of this type. We always
choose and act in particular situations, and the good which is the
 112
object of choice is always, therefore, a singular instance of a type of
good—hence always a constitutive means-in-particular. (This classi-
fication of all objects of choice as constitutive means-in-particular
does not prevent the cross-classification of them as functional
means or ends.) This fact about the objects of choice is indispensa-
ble to an understanding of freedom of choice. Only means-in-
particular are contingently constitutive of happiness; means-in-
general, being necessarily constitutive, cannot be objects of free
choice. In willing happiness necessarily, we must will whatever is
necessary to its constitution, as we conceive it; but we need not will
this or that particular instance of a type of good, for such means are
only contingently constitutive of happiness. Failure explicitly to
observe this distinction between constitutive means-in-general and
means-in-particular accounts for some ambiguity in passages in
which St. Thomas discusses the necessitation of the will with re-
spect to means. Vd.

38. Four things should be noted about this classification of means:

first, that the same good may be both a constitutive means and a
functional means, for the cardinal virtues are both constitutive of
and generative of happiness; second, that as a constitutive means a
good may be enjoyable, though as a functional means the same
good is useful—thus, virtues and good acts are both useful and en-
joyable; third, that the same good may be a functional means, both
generally considered and specially considered, for the cardinal vir-
tues are both productive of good acts, and of happiness itself, fourth,
whatever good is a functional means, whether productive of another
partial good or of happiness, must be a means-in-particular, for
functional relationships occur only in the existential order.

It should also be noted that the means-end relationship is causal, and

that constitutive means belong to the order of material causality,
their ends being formal causes—wholes or universals, which can be
concretized or embodied; whereas functional means belong to the
order of efficient causality, their ends being final causes. Hence
functional means partake more of what is essential to the notion of
means than do constitutive means; and cardinal virtue more so than
any other partial good.

Finally, and above all, it should be noted that anything which is a

means to virtue is indirectly productive of happiness (the state or
political common good being, as we shall later see, chief among
 113
such indirect productive means); and the basic distinction in the as-
pects under which the same partial good is regarded as both consti-
tutive and productive, whether of happiness itself or of some other
partial good, is a distinction between that good as enjoyable and as
useful (thus, virtue is both constitutive and productive, both an en-
joyable and a useful good, both an end and a means).

Summa Theologica, I-II, 10, 1; 10, 2, ad 3; 13, 6, ad 1; cf. Ibid., I,

82, 2. It should be added that, although all means-in-particular are
contingent goods, so far as the constitution of happiness is con-
cerned not all are mere means, for some are ends, being instances of
types of good which are essentially constitutive of happiness, be-
cause intrinsic human goods (i.e., bonum honestum hominis, such as
this good act, or this degree of a virtuous habit).

39. A definite “order and variety of goods” is materially what hap-

piness as “all good things”, is formally. Cf. Chapter IV, pp. 55-56.

40. The two theoretical problems of ethics concern, first, the end,
and second, its generative means. The first problem is solved when
happiness is adequately and truly defined; the solution is here
equivalent to knowledge about the order and variety of partial
goods, and their functional relationships to one another. The second
problem is solved when we are able to deduce from the nature of
happiness what must be the nature of its generative means.
Knowledge of the constitutive means is explicitly equivalent to
knowledge of the end; but knowledge of the generative means is
only implicitly contained in knowledge of happiness, and must be
deductively explicated. When we know what the generative means
are, we can proceed to the practical problem of how to possess them
and, through them, how to become happy.

41. By the “integrity of the cardinal virtues” is meant their function-

al interconnection, their co-existence in the same degree of effec-
tiveness as generators of the end. The proof that such integrity is
necessary can be made either from the unity of the end which these
virtues serve as productive means or from the fact that each of the
moral virtues depends upon prudence, and it depends upon all of
them. It is, therefore, true to say that if a man habitually does any
sort of act for the completely right reason. he is a man of perfect
 114
virtue, for he will do every other sort of act in the same way. Cf.
Aristotle, Ethics, VI, 13. Proceeding from happiness as the end and
first principle, we arrive at the conception of virtue as the productive
means and second principle. In this deduction, only the virtue of
prudence is specified; the notion of moral virtue as a good habitua-
tion of the appetites, as their rectification by a right ordering of all
desires, precedes the specification of moral virtue by the definition
of the distinct virtues of temperance, fortitude, and justice. If this
deductive procedure is followed, the integrity of virtue will be un-
derstood before the specific moral virtues are distinguished. Aristo-
tle’s failure to proceed in this way permits readers of his Ethics to
misunderstand the essence of virtue; for only after the integrity or
unity of virtue is understood, can anyone distinguish between ap-
parent virtue (i.e., one “virtue” existing apart from others) and real
virtue.

In his Preamble to Summa Theologica, II-II, St. Thomas says: “We

may reduce the whole of moral matters to the consideration of the
virtues.”

42. Cf. Chapter IV, p. 68.

43. This reasoned truth of moral philosophy must be submitted to

the criticism of the theologian who considers all moral problems in
the light of the dogmatic truth about man’s fallen nature. Cf. note 1,
in which moral philosophy, as a work of reason, was discussed in
relation to moral theology, as based on revelation. Moral philosophy
proceeds on the hypothesis of the natural . If this hypothesis be
false, as revealed religion declares, then all the conclusions of moral
philosophy must either be regarded as hypothetical, or they must be
qualified and transformed by subalternation to theological truth. The
fundamental problem here is whether purely natural virtues are ade-
quate for the achievement of the natural end, the good which is pro-
portionate to the nature; and this turns on whether, in the case of
fallen human nature, the cardinal virtues can be really possessed at
all (i.e., in their integrity) without the help of grace and infused
moral virtues. If not, then temporal happiness is unattainable except
for natures elevated by grace; but, on the other hand, if grace makes
possible the integral possession of the natural virtues, then it not
only enables a man to direct his life toward a supernatural God, but
also enables him to possess natural virtue in such a way that the
temporal happiness, due his nature, can be achieved. With God’s
 115
help a man can live well on earth if, but for the grace of God, he
cannot.

44. If, however, the essential goodness of a virtue be the actualiza-

tion it confers upon a power, then the intellectual are superior to the
moral virtues because they perfect the highest power of man. Cf. St.
Thomas, Summa Theologica, I-II, 66, 3; 66, 5, ad 1; 66, 1.

45. It should be remembered here that happiness must never be

called “the highest good” if that phrase be taken as signifying the
highest type of partial good. Being the whole of goods, happiness
cannot be highest in the order of goods which constitute it, though
in this order of goods itself, one good can be higher than another, or
highest of all. See Chapter IV, p. 55.

46. See St. Thomas, Summa Contra Gentiles, III, 37; cf. Summa
Theologica, I-II, 66, 5 ad 1.

47. Thus the apparent conflict between Book I and Book X of Aris-
totle’s Ethics is resolved. If it was due to any fault on Aristotle’s
part, it can be attributed to his failure to distinguish between two
senses of “the highest good”—(a) the sum of goods, which is happi-
ness as the end, constituted by every type of partial good, and gen-
erated only by cardinal virtue; and (b) the supreme type of partial
good, which is contemplation as the end served by all the other par-
tial goods, though generated directly only by intellectual virtue.
There can be no question that it is the goodness of a whole life, not
speculative activity, which is the ultimate object of natural desire,
and hence is happiness or the last end. Speculative activity is merely
the best aspect of a good life, its most enjoyable phase. This is con-
firmed by Boethius’ definition of happiness as the state of those
who possess in aggregate all good things. Thus conceived, happi-
ness cannot be identified with any single type of good, not even
with speculative activity in this life. I say “in this life’’ because Boe-
thius’ definition reveals the analogy of temporal and eternal happi-
ness (the one constituted by a simultaneous, the other by a succes-
sive, possession of “all good things”), and thereby helps us to un-
derstand how the contemplation of God in the Beatific Vision may
be identified with eternal happiness, whereas no sort of speculative
activity can be rightly identified with temporal happiness. In this
life, the contemplation of God, whether it be the activity of wisdom
as a purely natural virtue or the activity of faith and supernatural
 116
intellectual gifts, is not happiness, but only the highest part thereof.
Hence two things must be said about St. Thomas’ definition of im-
perfect, or temporal, happiness as consisting in contemplative ac-
tivity: first, that this is not an accurate definition of temporal happi-
ness, any more than Aristotle’s definition in Book X is; second, that
it is a better definition of the highest part of temporal happiness,
than could have been given by Aristotle, because St. Thomas con-
ceives contemplation in terms of God as the object, and as supernat-
urally generated by faith and the gift of wisdom, whereas Aristotle’s
meaning for “contemplation” lacks any objective specification, and
even if the Divine be implied as object, merely natural wisdom is
obviously insufficient. Cf. Summa Theologica, I-II pp. 3-5.

48. Aristotle’s remarks about the inexactitude and relativity of prac-

tical knowledge must be regarded as applying only to the rules of
conduct, and not to our knowledge of the principles, which
knowledge is theoretic in mode, though practical in end and object.
It is only knowledge which is practical in mode, as well as in end
and object, that is intrinsically inexact and relative to the individual
practitioner. See Ethics, I, 3; II, 2. Cf. our prior discussion of the
relativity of practical judgments (Chapter II, pp. 27ff), their status as
opinion, and the significance of this in correcting the error of
Socratism and Hedonism, which supposes that theoretic knowledge
of the good commands good acts unfailingly.

48a. What is called “inchoate virtue” is not, properly speaking, vir-

tue at all; yet it is a condition antecedent to the achievement of vir-
tue. To understand the stages on the way to becoming virtuous, we
must first distinguish the antecedent dispositions, which are improp-
erly called “virtues,” from the genuine habits which may be conse-
quently formed, and which are properly virtues. We must further
distinguish, among the antecedent dispositions, those which are nat-
ural and those which are acquired: the natural dispositions toward
virtue are those temperamental inclinations to good acts which Aris-
totle called the “temperamental virtues” (vd. Ethics, VI, 13); in con-
trast, there are acquired inclinations toward good acts, resulting
from extrinsic training in the domestic or the political community:
there are the “inchoate virtues”—not properly virtues because not,
strictly speaking, habits formed by the exercise of prudence on the
part of the agent. Finally, we must distinguish, among those habits
which are genuinely virtues the imperfect from the perfect. Unlike
dispositions, some of which may be natural, all habits are acquired;
 117
the distinction here, between imperfect and perfect virtue, turns on
the disconnection vs. the connection of the several cardinal virtues.
No single virtue is perfectly possessed, unless all are possessed in
some degree. Not only is the acquisition of any single virtue gradu-
al, but so is the achievement of the integrity of virtue itself. In the
temporal order, natural dispositions precede and condition the for-
mation of “inchoate virtues” or acquired tendencies toward good
acts; these in turn precede and condition the formation of genuine
habits, whether these be virtues imperfectly or perfectly. The incho-
ate virtues provide an inclination to the right end, enabling prudence
to begin to operate in the choice of means, and thus initiating the
formation of genuine virtues. Without such dispositions formed by
extrinsic training, there could be no beginning of real virtue. (For St.
Thomas’s discussion of natural inclinations toward virtue, see Sum-
ma Theologica. I-II, 58, 4 ad 3.) Cf. notes 41, 53

49. Herein lies the significance of Aristotle’s statement that it is dif-

ficult, if not impossible, to teach ethics (even the theory) to the
young, in whom virtue is not formed, and who have insufficient
practical experience of “the facts of life.’’ See Ethics, I, 3. Cf. ibid.,
II, 1, 1103b 1-5, X. 9. Cf. St. Thomas’s discussion of the role of law
in training youth: Summa Theologica, I-II, 95, 1.

CHAPTER VII: FROM ETHICS TO POLITICS: THE COMMON

GOOD AND DEMOCRACY

ON THE PRACTICAL LEVEL, ethics is auxiliary to politics in so far as

the knowledge it affords is useful to the legislator or magistrate.
When Aristotle speaks of politics as being the architectonic disci-
pline in the practical order, his criterion is the use of knowledge to
direct action. But, as we know, there is a theoretical level of practi-
 118
cal knowledge, concerned with universal principles, with the defini-
tion and analysis of end and means. On this level, ethics is architec-
tonic and politics subordinate. Political theory is a part of moral phi-
losophy. Though politics has some principles peculiarly its own,
their truth can be verified only by reference to the prior principles of
ethics. as the principles of physics (i.e., the philosophy of nature)
are not first principles, but subalternate to metaphysics, so the
common good, which is the first principle of political theory, cannot
be understood except in the light of virtue and happiness. In other
words, the political common good is an end, but not the end: it is an
end, in a certain order, but also a means to happiness, which is the
end absolutely. Thus, our inquiry in the field of morals, having es-
tablished happiness and virtue as the first and second principles of
ethics, leads us to the foundations of political philosophy, for it ena-
bles us to see the common good both as an end and as a means. All
of political philosophy rests upon this insight because, unless the
common good is the end of political activity, there is no answer to
realpolitik; and unless the common good is itself only a means to
happiness, there is no answer to “totalitarianism”—using that word
for the false political doctrine that the State is the highest good or
the whole good in the temporal order. 50

To establish the foundations of political philosophy (by refuting

realpolitik and totalitarianism), it is necessary to understand the na-
ture of the state, and the precise character of the goodness which
inheres in the political community. The first of these problems con-
cerns the origin of the state and its mode of being. The second re-
quires us to define the common good by locating it under one or
another of the categories in terms of which we have already classi-
fied the types of partial good. By doing this, we shall be answering a
question, previously raised, about the precise character of the social
goods.

(1) The nature of the state. If man were not by nature a social, and
also, through being rational, a political, animal, the state or political
community would not itself be natural. It originates, or comes into
being, in answer to a human need. If man, like other non-gregarious
animals, could subsist in a solitary mode of life, not even the family
would be a natural community, for it would answer to no natural
need. And if man could live well (i.e., live humanly, rather than
merely subsist), apart from a larger community than the family, the
domestic society would suffice, and there would be no natural need
 119
for the state. That which comes into being as indispensable to the
fulfillment of a natural need is itself natural and necessary. This type
of naturalness and necessity is in the order of final causality. To say
that the state is natural in this sense is, therefore, to deny that it is
artificial—i.e., an accidental contrivance which is totally dispensa-
ble, as, for example, any work of useful art that is a mere luxury. At
one extreme, then, there is the error of the social contract which
conceives the state as if it were a mere luxury—not natural at all in
the order of final causality. At the other extreme, there is the errone-
ous notion that the state is entirely natural in the order of efficient
causality—that it flows from instinctive determinations, rather than
is a free work of reason. This latter error confuses human associa-
tion with the instinctively determined communities of the social
brutes. The nature of the state is rightly conceived only when it is
understood both as originating in natural needs, and also as being a
mode of association which is determined by reason rather than in-
stinct.

The needs which the state satisfies can be briefly enumerated by

reference to the partial goods which constitute happiness. Man can-
not live well without a minimum sufficiency of external goods
(wealth) and without health (an intrinsic, though animal, good):
though these can be provided by the domestic economy, the ampler
economic facilities of the political society, through greater division
of labor, can guarantee them more readily. (Hence Aristotle says
that though the family and the village can supply the wants of daily
life, the state is better able to meet these same needs, by extending
and regularizing the economic functions: the state satisfies these
needs on more than a day-by-day basis.51 Wealth and health do not
suffice. Man cannot live well unless he lives virtuously—unless he
has good habits and operates accordingly. But the life of virtue is
itself a social life because man is by nature a social animal: this ap-
plies not only to the moral virtues, as principles of social behavior,
but also to the intellectual virtues, which are formed and operate
through the communication of men with one another. Man is both
rational in his social proclivities, and social in the exercise of his
rationality. Now the structure of the domestic or tribal community
cannot provide an adequate social setting for the acquirement or
exercise of virtue, moral and intellectual. Because it is a more com-
plex and diversified organization, only the political community can
adequately provide the conditions of virtuous living, through the
peaceful unity of a multitude of men, greatly diversified in the
 120
quality of their endowments. (Hence Aristotle says that the state,
which originates in the bare needs of daily life, continues in exist-
ence to provide the conditions of a good, or virtuous, life.52 We see,
therefore, that the social goods, among which the goodness of the
political community is paramount, function as means to all the other
partial good, which constitute happiness.

Everything which is natural exists in some way. To complete our

understanding of the nature of the state, we must ask about its mode
of being. That which exists primarily is a substance, and when we
speak of natures, in the primary sense, we mean the essential natures
of existing substances. The state is certainly not a nature in this pri-
mary sense: it is neither a substance nor the essential nature of one.
Now, whatever else exists, other than substances and their essential
natures, must have some accidental mode of being. Hence, the state
exists accidentally. This is confirmed by its very nature as an orga-
nized multitude, for this reveals it to be a form of composition or
order—an accidental form through which a substantial many is uni-
fied. Though it belongs properly in the sphere of prudence rather
than of art, the state has the same kind of accidental being that a
work of art has (e.g., a house is an accidental unity, realized through
a form of composition and order). But every accident must exist in
some substance: it cannot exist by and of itself. Thus, works of use-
ful art exist as accidental determinations of natural inorganic sub-
stances. Since it is an enduring association of men, a state must exist
as an accidental determination of the natures of its members. Not
being natural in the order of efficient causality, the state cannot exist
in human powers as such. The fact that it is a work of reason indi-
cates that it exists in the habits of men by rational determination of
their powers. In so far as the state is itself a good, it must exist in the
good habits of its members; and since the habits in question must be
habits of social behavior, we reach the conclusion that a state exists
(a) accidentally, (b) in the sphere of habit, (c) through the moral vir-
tues as principles of just conduct, without which (d) the peaceful
association of a multitude acting for a common good is impossible.
Justice is not merely one of the moral virtues, along with fortitude
and temperance; justice, generally considered, is all the moral vir-
tues as integrated and in their social aspect, i.e., as principles of so-
cial behavior. Hence, the state exists through and in the justice of its
members, differing from virtue itself (which also has an accidental
mode of being) in that virtue exists as the accidental determination
 121
of a single human nature, whereas the state exists as the habitual co-
determination of the natures of a multitude of men.53

(2) The goodness of the common good. The preceding analysis of

the nature and mode of being of the state enables us now briefly to
define its characteristic goodness. In the first place, since its very
naturalness and necessity is in the order of final causality, we know
at once that the state is itself a good, i.e., an object of natural desire,
responding to the needs of man’s rational, social nature. In the se-
cond place, since its mode of being is in and through the virtues of
the multitude, we know that the state is identical with the common
good. Just as virtue is an individual good (i.e., a good existing in a
single individual), so the state is a common good (i.e., a good co-
existing in a multitude of individuals). It is, therefore, false to say
that the end of the state is the common good, unless the “common
good,” here mentioned, signifies happiness. The common good,
which is identical with the well-being of the state (i.e., with that uni-
ty of peace which distinguishes the goodness of the state from the
evil of violent organizations), is the end of government, i.e., of po-
litical activity both on the part of rulers and of ruled. In the third
place, just as the virtues are partial goods constitutive of happiness,
so is the common good a partial good: it is the greatest of the social
goods. Moreover, since it is analogous to virtue in its mode of be-
ing, it has an analogous mode of goodness; it is an intrinsically hu-
man good and a bonum honestum. (This is true of all the other social
goods.) In the fourth place, although in the order of partial goods it
is inferior to intellectual virtues and contemplative activity, serving
these goods as do the moral virtues, the state is indirectly productive
of happiness, through being directly productive of moral virtue, i.e.,
forming inchoate virtue, applying sanctions to reinforce virtue,
providing the conditions for virtue’s exercise. Whereas any partial
good may function as a generative means with respect to some other
partial good, the state, like moral virtue, stands in a peculiar relation
to happiness, the whole of goods. Finally, from all this we know that
the political common good, in the sense defined, is not identical
with happiness: the goodness of the state as such is not the supreme
good in the temporal order, for it is neither the highest good in the
order of partial goods, nor is it equivalent to “all good things” —that
which, being possessed, leaves nothing to be desired. On the contra-
ry, the essential goodness of the state, like the essential goodness of
moral virtue, lies in its productive function with respect to happi-
ness.54
 122

Although the goodness of the state, or common good, is not the end
in the temporal order, it may, nevertheless, be an end, in the strict
sense in which every partial good which is a bonum honestum is an
end. Moreover, within the sphere of specifically political activity,
the common good, the welfare of the political community, is the last
end toward which all political institutions, agencies, and acts, are
ordered as means. In the domain of political life, the common good
occupies a place analogous to that of contemplation in the order of
partial goods constitutive of happiness. Furthermore, the common
good is a complex whole of many social goods, whose variety and
order constitute it, as happiness is constituted by the variety and or-
der of all good things. And just as moral virtue is both a constitutive
part of happiness and also its productive means, so good govern-
ment is, as a social good, constitutive of the common good, and also
the one means which is directly productive of the state’s well-being.
(By “government” here I understand all the institutions and agencies
through which both rulers and ruled can act for the common good.)
Just governmental activity not only maintains the common good,
but also facilitates the formation of virtue in the citizens and thus
the state is, indirectly, productive of happiness, or the life of virtue.

This indicates at once the major principles of political theory: the

first principle is the end (the common good), the second principle,
the productive means (good government), analogous to the first and
second principles in ethical theory (happiness and virtue). The ex-
position of political theory consists, therefore, first, in a definition of
the end and an analysis of its constitution; and second, in a defini-
tion and analysis of the productive means. As there are several spe-
cies of virtue, so there are specifically distinct forms of government,
though here there is a basic difference in that, whereas the specific
virtues are conjunctive and must be integrated, the several forms of
government are disjunctive and cannot co-exist. The central prob-
lem in political theory is the classification and evaluation of the
forms of government, and here the crucial question is whether one
form is better than another, on moral grounds, because it is essen-
tially more just, through greater justice effecting a greater common
good and, ultimately, making the state serve the happiness of every
man, rather than that of the many or the few. This question will be
answered in the affirmative, despite traditional opinion to the con-
trary, once it is seen that the common good is the end of govern-
ment, not of the state, and that happiness, which is the end of the
 123
state, is, as a life of virtue, necessarily a political life. When the
forms of government are specified by criteria of justice (the distinct
elements of political justice being separable and cumulatively com-
binable), they can be graded according to their moral goodness and
democracy can be demonstrated to be, on moral grounds, the best
form of government 55 intrinsically, the most just; extrinsically,
productive of the most complete common good and, thereby, of
human happiness.

Thus we have learned that the nature of the state, its mode of being,
and its essential goodness, cannot be understood except in terms of
the nature of man, the perfection of his nature by virtue, and the per-
fection of his life by happiness. Unless, therefore, the principles of
ethics can be ascertained, as objective theoretical truths comprising
wisdom in the practical order, the principles of politics cannot be
established. Unless, as a branch of practical philosophy, it be found-
ed upon moral wisdom, political theory cannot be defended against
the skepticism of realpolitik, or protected against the doctrinal er-
rors of totalitarianism. But no part of practical philosophy (ethics or
politics) can be validated as knowledge of absolute and universal
principles without the prior acknowledgment of ultimate theoretic
truths—about being, goodness, and the nature of man. To overcome
moral skepticism, and realpolitik as its consequence, it is necessary,
first, to eradicate its root in the prevailing positivism which denies
philosophical knowledge—metaphysics, and especially the philoso-
phy of man. This has been illustrated in the very motion of our in-
quiry; the dialectic of morals uncovered the need for a prior dialec-
tic of substance, essence, and man. Until that prior work is done,
most of the truths we seemed to reach, both dialectically and by de-
ductive elaboration, will not convince the student. Lacking convic-
tion with respect to these, the student remains susceptible to all the
“political religions” which rush in to fill the vacuum created by his
doubts and denials. In the world as it is today, a good work the phi-
losopher can do is to labor inductively in the fields of metaphysics
and psychology, for the sake of moral wisdom and, ultimately, for
the foundation of a true political philosophy.

NOTES

50. I say “political common good” because in one meaning of the

phrase “common good” happiness is a common good—a good
common to all men because they are specifically the same in nature.
 124
The political common good is also a good which is common in that
sense, but, in addition, when the phrase “common good” is used in
the restricted political sense, it means the goodness of the political
community as such. In what follows, I shall use the phrase “com-
mon good” only in this restricted sense to mean “the welfare or
well-being of the community itself.”

51. I undertook the dialectic of morals as an effort to overcome

moral skepticism and hedonism on the part of contemporary stu-
dents. See Chapter I. As I pointed out, skepticism about the objec-
tivity of moral truths necessarily leads to the adoption of realpolitik,
which is skepticism about the objectivity or universality of political
principles. See note 9. Furthermore, just as the student who is a
skeptic about moral matters is also usually a hedonist in his explana-
tion of moral phenomena (i.e., the facts of preference), so those who
adopt the position of realpolitik are usually advocates of totalitarian-
ism—even if they would be shocked to discover this. Even those
who think they oppose totalitarianism, because they magnify the
“rights of the individual,” affirm its basic tenets when they claim
that all “moral values” are relative to the mores of the community,
for then there are no independent moral criteria by which the com-
munity itself can be criticized as good or bad. Furthermore, just as
hedonism is the error of converting a partial good into the whole
good (treating pleasure as if it were happiness), so totalitarianism is
the error of similarly converting a partial good—treating the State as
if it were the absolute end. In the dialectic of morals, I did not at-
tempt to criticize every variety of error in ethical theory, but only to
answer the moral skeptic and the hedonist; so here, in considering
the foundations of moral philosophy, I shall try to refute only real-
politik and totalitarianism. Other fallacies in political doctrine can
be readily corrected by anyone who knows the right principles.

52. See Politics, I, 2.

53. See Politics, 1, 2, 1252b 27-30. One point should be stressed,

namely, that the acquirement of the moral virtues depends, as we
have seen, upon good government. See Chapter VI. Wise regula-
tions in the domestic community and just laws in the political com-
munity are indispensable as extrinsic, efficient causes for the pro-
duction of the virtues in an inchoate form; and inchoate virtue is, in
turn, a necessary stage through which the individual must pass in
becoming genuinely a man of virtue, Furthermore, because man is
 125
not simply rational, because he is an animal, a creature of passions,
reason needs external help in enforcing its own rule upon the appe-
tites. The good which reason may truly apprehend exercises authori-
ty over his actions, but the authority is unsupported by enforceable
sanctions. This is the essential defect of ethical eudamonism, for,
considering the individual in isolation, the pursuit of happiness can-
not be enforced: a man cannot impose extrinsic sanctions upon him-
self; he is not forcefully obligated to become happy; he is not duty-
bound by risk of punishment. This defect is, of course, remedied by
considering the individual in relation to his fellow men, with whom
he is associated in the political community. Since other men, as well
as himself, depend upon the common good for the pursuit of happi-
ness, he is obligated to act for the common good; and his social du-
ties, enforced by political sanctions, operate reflexively to support
the rule of reason in his own private life. Purely moral authority be-
ing authority divorced from power, the good as apprehended exer-
cises only moral authority over a man’s actions; authority combined
with power being effective sovereignty, the state exercises sover-
eignty over human life. Sovereignty—and the obligations, duties,
and sanctions which it institutes—is indispensable in the moral or-
der because man is not purely a creature of reason. There are, in
short, two sorts of natural authority: the moral authority of a man’s
own reason, which imposes only intrinsic obligations upon his will
and his acts, and the political authority of a sovereign (i. e., a man’s
community), which obligates him extrinsically, and enforces duties
by sanctions.

Even political sovereignty, the theologian tells us, is not suffi-

cient, when we consider man’s fallen nature. Though virtue is the
intrinsic principle of good acts, two extrinsic principles are required
for the formation of virtue: direction (i.e., law) and help (i.e., grace).
In making this point, St. Thomas means by law, not merely human
law, but Divine law, proceeding from the sovereignty which God
exercises over human life. See note 47.

54. As an unjust law is a law in name only, being really an expres-

sion of force, so a state not founded on the justice of its members is
a state in name only, being really an organization of violence, in
which some men dominate others by force and the others submit
through fear. Though an association of men through force and fear
may endure for some time, it lacks the unity of peace which prevails
only when all the members work for a common good. When men
 126
are organized by violence, those who wield force exercise it for their
private interests, and the rest are enslaved. (Of course, even the just
society requires government with enforceable sanctions: those who
do not act justly out of conscience must be compelled by fear of the
law. See note 53.)

The definition of justice throws light on this point. As temperance is

the habit of forgoing immediate pleasures for the sake of a greater
good, as fortitude is the habit of suffering immediate pains for the
sake of a greater good, so justice, as a special virtue, is the habit of
not willing an excess of good for one’s self at the expense of a di-
minished good for others. And justice, generally, consists of all the
virtues directed simultaneously to the good of others and to one’s
own good through being directed to the common good in the fruits
of which all equally share. “Justice,” said Aristotle, “is the bond of
men in states, for the administration of justice is the principle of
order in political society” (Politics, I, 2, 1253a 37).

55. Though the state appears to be a whole, of which its human

members are parts, the goodness of this whole is not greater than the
goodness of the parts, i.e., the perfection of their being by happi-
ness. On the contrary, the goodness of this whole (i.e., the common
good) is, along with other social goods, a constitutive part of happi-
ness, even though happiness appear to be the good of a part (i.e., the
perfection of a single human life). Though the common good be the
good of a whole, i.e., the community, it is not the whole of goods,
for that is happiness. There is nothing paradoxical about this when
one remembers that the individual man is a substance, whereas the
state is nothing but an accidental being. Just as the perfection of a
man (happiness) is greater than the perfection of any of his powers
(virtues), because the goodness of a substance in being and opera-
tion is greater than the goodness of an accidental being and a princi-
ple of operation; so happiness is greater than the common good—
including it as a partial good, and subordinating it, along with vir-
tue, as an end subordinates its productive means. (On the ambiguity
of “common good,” see Summa Theologica, I-II, 90, 2.)

Traditional statements to the contrary, suggesting that the good of

the state is greater than the good of its members, as the good of the
body is greater than the good of its parts, can be accounted for by
the error of neglecting the fact that the state is not a substance, nor
are its human members substantial parts of a substantial whole.
 127
Though the error is made by Aristotle (see Politics, I, 2, 1253a 19-
24 and Ethics, I, 2, 1094b 7-10), though it is repeated by St. Thomas
in places too numerous to cite, the source of the fallacy is in Plato’s
conception of the state by analogy with the soul, in the Republic. It
is a profound historical misfortune that Aristotle, whose greatest
achievement in metaphysics was the notion of substance, did not
employ this notion to purify his own political thinking of Platonic
reminiscences. This error, combined with a failure to understand the
nature of temporal happiness, has led some scholastics to suppose
that the State, or the common good, is the supreme temporal good,
from complete subordination to which man can be saved only
through his ordination to God and the supernatural good of eternal
happiness. See Father John McCormick, The Individual and the
State, in Proc. Am. Cath. Phil. Assoc., XV, pp. 10-21. Cf. J. Mari-
tain, Freedom in the Modern World: pp. 49-53. Only religion can
save man from totalitarianism. This may be true practically, but it is
certainly not true theoretically, for by principles known through rea-
son alone, we know that temporal happiness is the end which the
state must serve, and that the natural perfection of man’s life, as the
end in the temporal order, subordinates the political common good,
both as a constitutive and a generative means. This truth is not in-
consistent with that other truth, namely, that under certain circum-
stances and in certain respects, the individual’s well-being must be
sacrificed for the welfare of the community, especially when the
existence of the community is threatened by external aggression, the
very fact that such sacrifice can be justified only by extraordinary
circumstances indicates that normally the good of the individual is
paramount.

56. See my paper, The Demonstration of Democracy, in Proc. Am.

Cath. Phil. Assoc., XV, pp. 122-165; my answer to Dr. O’Neil, The
Demonstrability of Democracy in The New Scholasticism, XV, 2,
pp. 162-168; and a series of articles, by Father Walter Farrell, O.P.,
and myself, under the title, The Theory of Democracy, beginning in
The Thomist, III, 3. Many of the points here barely indicated, con-
cerning the nature of the common good and the character of tem-
poral happiness, will be therein expounded with greater analytical
detail than was possible in this brief statement.