| es

i ANDISOGIENY
 $2 DAor-a
ST. NORBERT COLLEGE
320.101 R22p
Reading

35109 0085 aias,

Reading Nozick

DATE DUE

DEMCO NO. 38-298

 PHILOSOPHY AND SOCIETY

General Editor: MARSHALL COHEN

Also in this series:

Lukacs, Marx and the Sources of Critical Theory

Andrew Feenberg
The Reverse Discrimination Controversy
A Moral and Legal Analysis
Robert K. Fullinwider
The Moral Foundations of Professional Ethics
Alan H. Goldman

Evolutionary Explanation in the Social Sciences

An Emerging Paradigm
Philippe Van Parijs
 Reading Nozick
ESSAYS ON
ANARCHY, STATE, AND UTOPIA

EDITED WITH AN INTRODUCTION BY

Jeffrey Paul

ROWMAN & LITTLEFIELD

Totowa, New Jersey
Copyright © in this collection by Rowman and Littlefield, 1981
All rights reserved. No part of this publication may be
reproduced or transmitted in any form or by any means, without the
permission of the Publishers.
First published in the United States 1981 by
Rowman and Littlefield, 81 Adams Drive, Totowa, New Jersey 07512

Library of Congress Cataloging in Publication Data

Main entry under title:

Reading Nozick.

(Philosophy and society)

Bibliography: p.
1. Nozick, Robert. Anarchy state, and Utopia—
Addresses, essays, lectures. 2. State, The—Ad-
dresses, essays, lectures. 3. Civil rights—Ad-
dresses, essays, lectures. 4. Anarchism and
anarchists—Addresses, essays, lectures. 5. Utopias
—Addresses, essays, lectures. I. Paul, Jeffrey.
II. Nozick, Robert. Anarchy, state, and Utopia.
III. Series.
JC571.N683R4 198] 320.1'01 81-12092
ISBN 0-8476-6279-9 AACR2
ISBN 0-8476-6280-2 (pbk.)

Printed in the United States of America

For Ellie
 Digitized by the Internet Archive
in 2022 with funding from
Kahle/Austin Foundation

https://archive.org/details/readingnozickessOO00unse
 Contents

Acknowledgments ix
Introduction 1

PARTI. AN OVERVIEW OF ANARCHY, STATE, AND UTOPIA

1 Bernard Williams: The Minimal State 27

2 Peter Singer: The Right to be Rich or Poor 37

PART II CONTRA ANARCHISM: JUSTIFYING THE

MINIMAL STATE

3 Robert L. Holmes: Nozick on Anarchism 57

4 Jeffrey Paul: The Withering of Nozick’s Minimal State 68
5 Robert Paul Wolff: Robert Nozick’s Derivation of the
Minimal State 77

PART III RIGHTS: OPPOSING THE EXTENSIVE STATE

6 Thomas Scanlon: Nozick on Rights, Liberty, and

Property 107
7 Judith Jarvis Thomson: Some Ruminations on Rights 130
8 Samuel Scheffler: Natural Rights, Equality, and
the Minimal State 148
9 Eric Mack: Nozick on Unproductivity: The Unintended
Consequences 169
10 Thomas Nagel: Libertarianism Without Foundations 191
11 Robert Nozick: On the Randian Argument 206
12 Douglas Den Uy! and Douglas Rasmussen: Nozick on the
Randian Argument 232

vii
viti Contents

13. Ellen Frankel Paul: The Time-Frame Theory of Govern-

mental Legitimacy 270
14. Eric Mack: How to Derive Libertarian Rights 286

PART IV SOCIAL JUSTICE: ENTITLEMENT THEORY

VERSUS DISTRIBUTIVISM
15 Onora O'Neill: Nozick’s Entitlements 305
16 Cheyney C. Ryan: Yours, Mine and Ours: Property Rights
and Individual Liberty 323
17 Lawrence Davis: Nozick’s Entitlement Theory 344
18 David Lyons: The New Indian Claims and Original Rights
to Land 355
19 Hillel Steiner: Justice and Entitlement 380
20 Israel M. Kirzner: Entrepreneurship, Entitlement, and
Economic Justice 383

Selected Bibliography 412

Contributors 417
 Acknowledgments

While the responsibility for the selections and their treatment in

the introduction are entirely my own, the advice of a number of
people was exceedingly helpful in the preparation of this volume. I
owe a special thanks to Robert Nozick whose wise counsel im-
measurably improved this collection and without whose constant
support it might’never have been assembled. Marshall Cohen, the
general editor, made a number of incisive criticisms of the introduc-
tion and content which undoubtedly added to the quality of the
collection. Finally I would like to express my gratitude to my
publisher, Jim Feather, whose Job-like patience and encouragement
saw the book through to completion.

I wish to acknowledge permission to use material contained in

this volume as follows:
1. Bernard William’s ‘The Minimal State,’ Times Literary Supple-
ment, January 17, 1975, is reprinted by permission of the Times
Literary Supplement and Professor Williams.
2. Peter Singer's ‘The Right to be Rich or Poor,’ The New York
Review of Books, March 6, 1975, is reprinted with permission from
The New York Review of Books. Copyright © 1975 Nyrev, Inc.
3. Robert Holmes’s ‘Nozick on Anarchism,’ Political Theory, Vol. 5
No. 2 (May 1977), pp. 247-256, is reprinted by permission of the
Publisher, Sage Publications, Inc. and Professor Holmes.
4. Jeffrey Paul’s ‘The Withering of Nozick’s Minimal State,’
Philosophy Research Archives, 1980, is reprinted by permission of
the Philosophy Documentation Center, Bowling Green State Uni-
versity.
5. Robert Paul Wolff's ‘Robert Nozick’s Derivation of the Minimal

ix
x Acknowledgments

State,’ 19 Arizona Law Review 7 (1977), copyright © 1978 by the

Arizona Board of Regents is reprinted by permission.
6. Thomas Scanlon’s ‘Nozick on Rights, Liberty, and Property,’
Philosophy & Public Affairs, 6, No. 1 (Fall 1976), copyright © 1976
by Princeton University Press is reprinted by permission of
Princeton University Press.
7. Judith Jarvis Thomson’s ‘Some Ruminations on Rights,’ 19
Arizona Law Review 45 (1977), copyright © 1978 by the Arizona
Board of Regents is reprinted by permission.
8. Samuel Scheffler’s “Natural Rights, Equality, and the Minimal
State, Canadian Journal of Philosophy, March 1976, is reprinted by
permission of the Canadian Journal of Philosophy and Professor
Scheffler.
9. Eric Mack’s ‘Nozick on Unproductivity: The Unintended Conse-
quences was originally prepared for this collection.
10. Thomas Nagel’s ‘Libertarianism Without Foundations, The
Yale Law Journal, Vol. 85, pp. 136ff., is reprinted by permission of
The Yale Law Journal Company and Fred B. Rothman & Company
and Professor Nagel.
11. Robert Nozick’s ‘On the Randian Argument, The Personalist,
Spring 1971, is reprinted by permission of The Personalist, Univer-
sity of Southern California and Professor Nozick.
12. Douglas Den Uyl’s and Douglas Rasmussen’s ‘Nozick on the
Randian Argument,’ The Personalist, Spring 1978, is reprinted by
permission of The Personalist, University of Southern California and
Professors Den Uyl and Rasmussen.
13. Ellen Frankel Paul’s ‘The Time-Frame Theory of Governmental
Legitimacy,’ The Personalist, Vol. 60, No. 2 (April 1979), is re-
printed by permission of The Personalist, University of Southern
California.
14. Eric Mack's ‘How to Derive Libertarian Rights,’ is based on
‘How to Derive Ethical Egoism,’ The Personalist, Vol. 52. No. 4
(Autumn 1971) and ‘Egoism and Rights Revisited,’ The Personalist,
Vol. 58, No. 3 (July 1977), and is included by permission of The
Personalist, University of Southern California.
15. Onora O'Neill's ‘Nozick’s Entitlements,’ Inquiry, Winter 1976,
pp. 468-481, is reprinted by permission of Universitetsforlaget,
Oslo.
16. Cheyney C. Ryan’s ‘Yours, Mine, and Ours: Property Rights
 Acknowledgments xi

and Individual Liberty,’ Ethics, January 1977, pp. 126-141, is

reprinted by permission of The University of Chicago Press and
Professor Ryan.
17. Lawrence Davis's “Nozick’s Entitlement Theory, Journal of
Philosophy (73,21) is reprinted by permission of The Journal of
Philosophy and Professor Davis.
18. David Lyons’s ‘The New Indian Claims and Original Rights to
Land,’ Social Theory and Practice, Vol. 4, No. 3 (Fall 1977), pp.
249-272, is reprinted by permission of Social Theory and Practice
and Professor Lyons.
19. Hillel Steiner's ‘Justice and Entitlement, Ethics, January 1977,
pp. 150-152, is reprinted by permission of The University of
Chicago Press and Professor Steiner.
20. Israel Kirzner’s “Entrepreneurship, Entitlement, and
Economic Justice, Eastern Economic Journal, Vol. 4, No. 1
(January 1978), is reprinted by permission of the Eastern Economic
Association and Professor Kirzner.
 A

SCY aa = ‘ia
la Rama daily ePicenlait, OBR f
id ae bp
ak Ra ie
“Ley! wehact epee edit :
2 rer wWh, erie de
~
st eles ‘walla Pe th) eek Met
aeak fetta eta =: we
desis Peay aperyt V lyme?Ay se diaaeey

ee
Ube »

eek a eapeen
= cchinitidbaald edtr chs ere: a A
r 7 - viol iv LU
ob tae Atlas! ‘ vanituenseeetts
2= .
See ee eer aria wo Osi
SO ine beats hese tents
17,
] ~

i —oy é

oe
[ om, Ait

U PA

> af

é
2 .~-

ae -

=
~ ~

=
i 7 ¢
 Introduction

THE IMPACT OF Anarchy, State, and Utopia

To anyone familiar with the contents of Robert Nozick’s Anarchy,
State, and Utopia its reception both in lay and professional circles
could not have been more unexpected. In contrast to the moderate
and conventional ideological stance of John Rawls’s A Theory of
Justice, its distinguished predecessor in political philosophy,
Nozick’s work announced a thesis so out ofjoint with its times that
the critical acclaim accorded it by many of its reviewers must have
astounded its author as much as it baffled its critics. That a treatise
extolling the virtues of eighteenth-century individualism and
nineteenth-century laissez-faire capitalism should not have elicited
either hostility or silence, is both a puzzling and gratifying phenom-
enon; puzzling because its themes run counter to the Zeitgeist and
gratifying because it is a work of considerable philosophical acuity.
In Anarchy, State, and Utopia, Nozick revives the claim long
associated with Locke and Spencer “that a minimal stated limited to
the narrow functions of protection against force, theft, fraud, en-
forcement of contracts, and so on, is justified; and that the minimal
state is inspiring as well as right” (ix). In order to define the contours
of the human rights which circumscribe the intrusive authority of
the state, Nozick develops a conception of justice that he calls the
entitlement theory. According to it, economic goods arise already
encumbered with rightful claims to their ownership. Any social
philosophy which argues that they emerge unaccompanied by such
claims and can be distributed to society without transgressing any
antecedent moral boundaries, is, according to Nozick, therefore
mistaken. Yet, such distributivism (or redistributivism) has charac-
terized the speculations of most twentieth-century political

1
y Introduction

theorists. Moreover, much “objective” contemporary social science

tacitly assumes some variant, usually utilitarian, of distributivism in
both its scientific analyses of society and its prescriptions for social
betterment. In rejecting distributivist notions of social and eco-
nomic justice, Nozick has, then, defined a radically different nor-
mative paradigm—entitlement theory—within which philosophers,
jurists, and social scientists may work in the future.
This suggests that the significance of Nozick’s book is best ap-
preciated when considered against the background of Thomas
Kuhn’s model for theory development in the natural sciences. In his
Structure of Scientific Revolutions, Kuhn suggested that current
scientific knowledge may not represent the accumulated wisdom of
the past but seems to embody the elucidation of a wholly new
conceptual paradigm, one whose principal features are utterly in-
commensurable with its predecessors. If the development of moral
knowledge is analogously structured, then the very incommensura-
bility of Nozick’s theory with its competitors, accounts, in part, for
its impact on social philosophy and science. It explains, too, why
Anarchy, State, and Utopia was able to replicate the enthusiastic
welcome accorded to Rawls’s work only three years earlier. For A
Theory of Justice was written partly to restore the impaired foun-
dations of the current ideological paradigm, distributivist liberalism
—the theoretical support of the western welfare state and mixed
economy; whereas, Nozick’s book struck at the conceptual under-
pinnings of both contemporary western society and its totalitarian
adversaries in the East.
The resultant shift in social perspective has been both disturbing
and illuminating to commentators. Again, the comparison with
Rawls is instructive. The response to Nozick’s book echoed the
response to Rawls’s in intensity (both were widely and thoroughly
reviewed both in popular and technical journals), but in Nozick’s
case there was striking alternation between scarcely qualified praise
and passionate denunciation that was not characteristic of the reac-
tion to Rawls. Thus, Hillel Steiner in Mind could write

Although similar claims have become increasingly common, it would seem

almost churlish not to acknowledge that this book is the best piece of
sustained analytical argument in political philosophy to have appeared in a
 Introduction 3

very long time. Moreover it is, in its way, an extremely moving book, not
the Jeast because of the engaging manner in which it is written.)

While Brian Barry could inveigh in Political Theory,

.. . the intellectual texture is of asort of cuteness that would be wearing in
a graduate student and seems to me quite indecent in someone who, from
the lofty heights of a professorial chair, is proposing to starve or humiliate
ten percent or so of his fellow citizens (if he recognizes the word) by
eliminating all transfer payments through the state, leaving the sick, the
old, the disabled, the mothers with young children and no breadwinner,
and so on, to the tender mercies of private charity, given at the whim and
pleasure of the donors and on any terms they choose to impose.?

The varied responses to the Nozick book in contrast to its predeces-

sor, largely reflect the chasm which separates their respective
visions of the good society.

THE PURPOSE OF THIS COLLECTION

While Anarchy, State, and Utopia’s ideological stance may well

account for the mixed critical reaction to it, the intensity and
analytic sharpness with which it has been examined are principally
due to the uniqueness ofits point of view. Again, in Kuhnian terms,
it has introduced a new paradigm into political discussion. As Nozick
has been eager to point out, only his theory among its competitors in
the philosophy of distributive justice refuses to supply social goals
for economic allocation, substituting for them procedural criteria for
the proper acquisition of income and wealth. Such an “historical”
basis for unqualified rights of property stands in stark contrast to the
familiar distributive formulae of social philosophy which confer
upon their recipients only limited discretion over their property.
Instead of rules of allocation which identify some human characteris-
tic or propensity in accordance with which income and wealth are to
be distributed (to each according to his need, ... merit,
contribution) Nozick merely specifies those activities of appropria-
tion and transfer which confer property rights on people without
reference to the relative condition or standing of their fellows. In
opposition to those distributivist theoreticians who seem to view all
4 Introduction

wealth as appropriately subject to the state’s allocative mechanisms,

Nozick argues that wealth is created with moral encumbrances
which entitle its owners to absolute control over it irrespective of
the social ramifications of such a distribution.
In Anarchy, State, and Utopia, Nozick provides an argument for
his admittedly deviant views which has by now been subject to the
most meticulous scrutiny. This collection is an attempt to sum-
marize the current state of the discussion by selecting from this
outpouring pieces which exemplify the central critical tendencies of
his fellow social theorists.
Now is probably the most appropriate occasion for doing so since
the debates stimulated by the book have begun to shift away from
the particular arguments devised by Nozick to defend his own
unique vision of libertarianism to the question of whether any
plausible demonstration can be supplied for this recent successor of
Lockean liberalism. Since the publication of Anarchy, State, and
Utopia, there has been no shortage of alternative “libertarianisms”
devised to correct perceived shortcomings in Nozick’s version.
Recent additions to the libertarian corpus include Hillel Steiner's
Georgist variant of libertarianism erected upon Kantian foun-
dations? and Tibor Machan’s Aristotelian-Randian scheme of laissez-
faire political economy.*

THE ANTECEDENTS OF NOZICK'S LIBERTARIANISM

While Anarchy, State, and Utopia stands on its own as a work of

political philosophy it cannot be divorced from the tradition of
which it is so obviously a part. Nozick is a twentieth-century
successor to the great classical liberals of the seventeenth
century—Locke, Grotius, and Pufendorf. His strain of liberalism
however has been filtered through the radically individualist social
philosophies of Bastiat, Spencer, Tucker, and Spooner. Moreover,
his immediate predecessors in libertarian social philosophy estab-
lished the theoretical agenda for Nozick’s own work. The American
economist Murray Rothbard, for example, has in the twentieth
century revived Benjamin Tucker's and Lysander Spooner’s argu-
ments for individualist anarchism. This American triumvirate is
Nozick’s invisible collective antagonist as he strives in the first part
of Anarchy, State, and Utopia to establish the propriety of the
 Introduction 5

minimal state and its desirability over its anarchist competitors. And
standing behind the entire work as the gray eminences of
twentieth-century libertarianism are the Austrian economists von
Mises and Hayek, and the Russian-American novelist Ayn Rand.
In a sense Nozick’s book is far less ambitious than the work of his
libertarian predecessors, surprisingly so for a book that has been so
widely read and acclaimed. Its design is less encompassing than that
of Herbert Spencer's Social Statics or F. A. Hayek’s Constitution of
Liberty, for example. But Nozick’s intention for the work was to
carve out a place for a theory of distributive justice that would be the
antithesis of the prevailing views on that subject. Merely to obtain
credibility for an historical theory of distribution would more than
justify the work in his view. Therefore, he was untroubled by the
fact that in the book he merely attempted to make plausible the
entitlement theory of distributive justice without providing a de-
tailed moral foundation for it. He was content to criticize opposing
positions and merely to explicate his own view, leaving the grander
project of providing a positive grounding for it to a later time. The
product of that modest ambition is the subject of the essays that
comprise this volume.

AN OVERVIEW OF Anarchy, State, and Utopia

Part One of this collection contains two useful surveys of Anarchy,
State, and Utopia which supply incisive criticism of its central
claims. Bernard Williams assails Nozick for his wholly formal defini-
tion of rights. It is impossible, Williams argues, to tell when a moral
boundary has been crossed if rights are not supplied with normative
content. Williams quarrels, as well, with the plausibility of Nozick’s
hypothetical narrative which depicts the development of the mini-
mal state. It represents, according to Williams, a bizarre departure
from any common sense account. Finally, he makes the point that
Nozick’s view of distributive justice relies on moral intuitions rather
than on argument and that competing intuitions can always be found
with as great a claim upon our moral sensibilities as those prop-
ounded by Nozick.
Peter Singer, in contrast, finds little to dispute in the book's
internal argument but complains that its antiredistributivist conclu-
sions are morally disreputable. In order to. oppose these, Singer
6 Introduction

recommends a reconsideration of utilitarianism, a position which he

believes Nozick has slighted in his book.

JUSTIFYING THE MINIMAL STATE AGAINST THE

ANARCHISTS

These surveys are followed by a second part devoted to an examina-

tion of Nozick’s claim that, anarchist arguments to the contrary, the
minimal state is justified. In the first third of his book, Nozick
attempts to deflect the claim of individualist anarchists that if rights
to life, liberty, and property are absolute, the very existence of the
state necessarily infringes them. The monopoly of adjudicative and
police powers assumed by any government is maintained by
restricting the abilities of citizens to purchase security from private
sources. Moreover, the taxes collected to afford such protection are
obtained by an invasion of property rights. Hence, argue the
anarchists, the state must violate the very right of property that it is
instituted to protect. The acceptance of the libertarian scheme of
rights, then, requires the rejection of government as their appropri-
ate institutional safeguard. In reply Nozick argues that if we con-
sider a hypothetical state of nature we can trace a possible scenario
in which from a number of protection agencies one emerges which
dominates the rest and does so without violating any of the
aforementioned rights.
Three essays contest Nozick’s argument that a minimal state can
arise from the state of nature without involving itself in transgres-
sions of people's rights. Robert Holmes rejects this argument,
pointing out that Nozick adduces no independent epistemic criteria
on the basis of which the enforcement procedures of the domi-
nant protection agency can be assessed. In the absence of such
criteria, the dominant protection agency's usurpation of all powers
of adjudication and enforcement cannot be justified. For unless its
procedures are morally superior to those of its competitors and can
be shown to be so, their forcible elimination is ethically suspect.
Hence, it cannot justifiably assume the position of minimal state.
Furthermore, Holmes collapses Nozick’s distinction between the
having and exercising of rights, which forms the basis of Nozick’s
contention that even if violations of the latter are required for the
 Introduction ft

assumption of state power by the dominant protection agency such

infractions do not deprive persons of the former. Having dismissed
the basis of this distinction, Holmes accuses Nozick of having
sanctioned the violation of human rights in his attempt to devise a
cognitively tenable foundation for the emergence of the state. In my
own paper, which follows Holmes’s, I delineate the infinitely re-
gressive character of the “procedural rights” deployed by Nozick in
his defense of the dominant protection agency's acquisition of
state-like power. If such rights exist, Nozick argues, a scrupulous
defense of them by a protection agency would morally justify its
accumulation of statelike powers. My reply is that there are logical
grounds for concluding that the existence of such second-order
rights is evanescent.
Robert Paul Wolff focuses his criticism on Nozick’s use of the
notion of a compensated boundary crossing as the foundation of his
favorable analysis of the evolution of aminimal state. Nozick argues
that the riskiness,of the activities of independent protection agen-
cies may morally justify the prohibition of these activities as long as
the clients of the independent agencies are suitably compensated for
their sudden exposure to attack. Wolff questions the measurability
of compensable harms in Nozick’s state of nature, given its fluidity
and lack of structure. For, if the uncertainty which prevails there
unduly circumscribes the possibility of rational calculation, the
compensation owed to the newly vulnerable clientele of indepen-
dent protection agencies can never be accurately calculated. Hence,
a dominant protection agency, Wolff argues, cannot be arithmeti-
cally certain of the moral grounds of its ascension to state-like
dimensions.
Eric Mack challenges the very propriety of Nozick’s introduction
of the compensation principle on the grounds of its inconsistency
with the general framework of deontic rights defened in Anarchy,
State, and Utopia. In his “Nozick on Unproductivity: The Unin-
tended Consequences, Mack identifies the compensation principle
with an incipient utilitarianism which if generalized would erode
Nozick’s libertarianism by condoning rights violations in cir-
cumstances where the subject’s wellbeing is enhanced. Should
Nozick realize his error and abandon the compensation principle he
would, according to Mack, seriously, if not fatally, jeopardize the
8 Introduction

whole anti-anarchist project of Part I. Without the compensation

principle Nozick’s minimal state may be transformed into the wel-
fare state—a Nozickian nightmare.
While the preceding critiques, if successful, would imply a failure
on Nozick’s part to supply a Lockean justification for the evolution of
the minimal state, Ellen Frankel Paul develops an alternative
defense of that institution from a classical liberal perspective. Paul
argues that Nozick’s historical analysis of the state’s emergence is a
superfluous exercise in political casuistry. It is not the moral history
of an agencys ascension to power which reveals the legitimacy of its
current status; rather it is the moral character ofits present activities
by which its legitimacy should be assessed. Any state, says Paul,
that is presently a protector and not a violator of Lockean rights has
thereby justified its existence in the current “time-frame,” whatever
previous moral indiscretions it has committed. Paul, therefore,
rejects the historically-oriented methodology used by Nozick to
evaluate the rectitude of hypothetically emergent states.

RIGHTS: OPPOSING THE EXTENSIVE STATE

Part Three of this collection contains papers which critically analyze

the structure and content of the rights defended by Nozick in
Anarchy, State and Utopia. This aspect of the book is in a
number of respects the most controversial. In the first place, the
rights ascribed by Nozick to persons are by his own admission not
derived from some more primitive moral or factual foundation. He
vaguely alludes to and opaquely describes a possible Kantian basis
for rights (30-34) from which he infers that their logical genesis must
be deontic, not teleological in character. The content of these rights
consists in titles to objects, which titles confer exclusive authority
upon their bearers over the uses to which they are put. Again, the
precise processes by which the bearers rightfully obtain this author-
ity is left undescribed. Nozick merely stipulates that they be
historical in character, in keeping, we may surmise, with the
anti-teleological program on which he has embarked. He does,
however, s .ecifically delineate certain constraints upon the primi-
tive acquisition of such titles which he incorporates into a “Lockean-
Proviso” named after its famous ancestor from the Second Treatise
of Civil Government. Thus, the nature of the system of rights
 Introduction 9

propounded in Anarchy, State, and Utopia is merely adumbrated

and no logical demonstration of their existence is produced. These
omissions have generated consternation among Nozick’s critics, as
well as speculation as to what (and, indeed, whether!) content and
foundation might be plausibly conjoined to the conception of rights
which he espouses.
In addition to the ambiguity of their content and the absence of
any detailed defense of their existence, Nozick’s rights have elicited
criticism on other grounds. Their recognition, it seems, would
preclude the kinds of economic and social arrangements closely
associated with the contemporary welfare state. Any compulsory
transfer of income or wealth, initiated on other than compensatory
grounds, constitutes a transgression of the rights Nozick defends. A
state whose functions include such transgressions has thereby for-
feited its legitimacy. Redistribution and paternalism in its recent
liberal manifestations are from his perspective to be condemned
along with the obviously odious excesses of totalitarian states. The
precise implications of the Nozickian scheme of entitlements for the
distribution of economic goods is the subject of the last part of this
collection. Part Three confines its discussion to the character and
source of the rights which underpin the distributive imperatives of
libertarianism.
In his essay, Thomas Scanlon conjectures about the nature and
extent of the natural property rights that are logically entwined with
the imperative of personal inviolability which Nozick treats as
normatively axiomatic. Scanlon believes that a far narrower concep-
tion of such rights is implied by this moral axiom than Nozick realizes
and, therefore, the extent of unregulated economic activity implied
by it must be vastly attenuated. Moreover, because personal au-
tonomy cannot simply be explicated in terms of the property rights
defended by Nozick, Scanlon raises the possibility ofa teleologically
grounded system of rights whose character would be determined by
their contribution to such autonomy.
Like Scanlon, Judith Jarvis Thomson is skeptical about the possi-
bility of defending the absolutely inviolable character of Nozickian
property rights. This skepticism derives from a number of examples
in which our intuitions incline us toward the view that the property
rights in question may be overridden by other considerations. Thom-
son then asks what is it, at the margin, which sustains the moral
10 Introduction

invincibility of a property claim and conversely, what is it that

justifies the infringement of a property right when we are morally
persuaded to ignore it? Property claims are to be sustained, she
argues, when, in addition to having acquired title to an object in
suitable ways, we value that object highly. Such claims may be
overridden when a life will be lost in the absence of an infringement
of rights. This demonstrates, Thomson argues, that rights are deriv-
able from human interests and needs and this in turn suggests that
the constraints that rights impose upon redistribution are not as
inflexible as Nozick’s deontological conception of them leads him to
believe. With Scanlon, Thomson is attracted to a teleological
justification for property rights which would diminish their moral
insusceptibility to redistributionist activity.
Unlike Scanlon and Thomson, Scheffler argues that the rights
deducible from Nozick’s fundamental moral intuitions are not
merely more narrow in scope and content than those defended in
Anarchy, State, and Utopia but differ radically from the portrait of
them drawn there. Isolating what Nozick characterizes as their
source, Scheffler is able to show that the nature of the rights
emanating from it in fact conflicts with the entitlement view of
them. Nozick conjectures that the feature of human nature from
which absolute rights against redistributionist interference may be
plausibly derived is the capacity of persons to suffuse their lives
with meaning by organizing their activities so as to accord with
some overall purpose (50-51). This feature, when combined with
the fact of the separateness of human lives, implies for Nozick the
sort of impenetrable moral barriers to redistribution and liberty
which he calls side constraints. To seize A’s property in order to
transfer it to B involves for Nozick the sacrifice of A’s opportunity to
impart meaning to his existence to B’s. But this is either to assign a
priority to their interests which flouts the hypothesis of their
separate and equivalent human potential for a meaningful existence,
or it is to combine their interests in a utilitarian calculus which
assumes that they share a center of cognitive response, an assump-
tion antithetical to the ontological distinctness of human. per-
sonalities.
Scheffler rejects these political inferences drawn by Nozick from
the intrinsic value of leading a meaningful life. For Scheffler, if
there is such value then the rights which it sanctions are qualita-
 Introduction 1l

tively different from Nozickian ones. Scheffler argues that if a

meaningful life has moral value then the capacities required to
nurture them are valuable as well. These capacities cannot be
employed unless the material conditions necessary to their support
are met. The provision of those conditions includes that quantity of
distributable goods necessary to ensure a reasonable chance to all of
leading a meaningful life. Hence, the centerpiece of Nozick’s axiol-
ogy leads to the very welfare rights which he set out to oppose,
according to Scheffler.
In his essay, “ Nozick on Unproductivity: The Unintended Conse-
quences,” Eric Mack draws a still more startling conclusion. He
argues that the deontically fixed moral boundaries delineated in
Nozick’s theory of rights are systematically undercut by him through
the introduction of the utilitarian criteria for compensation that he
employs to support his rejection of anarchism in Part I of Anarchy,
State, and Utopia. If Mack is right, then Nozick has eviscerated his
own libertarian-entitlement theory of rights in ways that would
delight many of his critics but are clearly unintended. Given Mack's
interpretation, the damage allegedly inflicted by Scanlon, Thomson,
and Scheffler on Nozick’s libertarianism pales in comparison with
the devastation visited by Nozick upon his own work. Moreover, the
attempts by Scanlon, Thomson, and Scheffler to soften the supposed
rigidity of Nozick’s libertarianism would be superfluous from Mack’s
perspective, for Nozick’s adoption of a utility rule undermines his
libertarian principle.
Mack argues that contrary to what we are led to expect from
Nozick’s depiction of rights as morally inpenetrable boundaries,
Nozick argues that such boundaries may be transgressed if it is
possible to compensate the victims of those transgressions
adequately. Rights violations, then, are only to be prohibited when
compensation is not a feasible alternative, and compensation is not a
viable response to most boundary crossings. Hence, the majority of
border crossings are to be prohibited. However, some border
crossings are permissible if they are done to prevent the success of
activities which seem likely to result in rights violations. Nozick
argues that many such “risky” actions fall into a category of activity
which may be prohibited because it neither benefits the buyer nor is
independently useful to the seller. Such “unproductive” activity
may be prevented even though it has not presently resulted in any
12 Introduction

rights violations. Mack discerns, however, that unproductive activ-

ity as defined by Nozick includes several categories of action which
can never produce nor constitute rights violations but are,
nevertheless, subject to prohibition on Nozickian grounds.
Boycotts, refusal to sell a natural resource to someone who desires
it, and the inheritance of certain types of wealth all may be
prohibited on the grounds of their unproductivity, Mack concludes.
Yet each of these prohibitions would constitute an unmistakable
violation of Nozickian rights. They are permitted, however, ac-
cording to Nozick’s unproductivity doctrine, provided suitable
compensation (the criteria of suitability here are several) is paid to
those who are prevented from exercising their rights because of
them. The extent of that compensation is to be fixed by certain
utilitarian criteria. Rights as deontic imperatives, then, are to be
replaced with rights whose moral boundaries can shift depending
upon how the well-being of the affected parties is altered. As Mack
concludes:

On the resulting outcome conception of rights, a boundary specifies a level

of well-being and the permissibility of others’ actions depends upon the
effect of those actions upon the subject's well-being. Even if what is forcibly
required of the subject is a productive and beneficial activity, no wrongful
boundary crossing will take place as long as the level of the subject's
well-being is fittingly raised. This shift to an outcome oriented conception of
rights should make it difficult for Nozick to sustain his anti-paternalism. For
the subject will have no complaint in terms of his rights if the intervention
does actually maintain or appropriately raise his level of well-being. In
contrast, on the more consistently deontic conception, a boundary is a
frontier which others do wrong to cross and accompanying benefits do not
right such wrongs (p. 187).

According to Mack, Nozick has transgressed the very conception

of inviolability which it is the point of his view to establish. He does
this by adding to his exclusively historical criteria for exercisable
rights the sort of teleological ones which he elsewhere repudiates.
This concession to “teleologism” may well constitute for Nozick the
first step down a slippery slope toward the abyss of patterned
theories of distributive justice. The distributivism which Anarchy,
State, and Utopia seeks to discredit as a valid notion of economic
justice is reinstated by the very work whose major theme is its
 Introduction Hs

inadequacy. Mack seems to have uncovered a serious if not fatal flaw

in Nozick’s libertarianism, tinged as it is by the self-defeating
adoption of two apparently antithetical distributive standards, one
historical, the other teleological. The former gives rise to immutable
entitlements, while the latter confers a protean character on them.
Nozick’s intransigent libertarianism has, according to Mack, been
seriously compromised.
Thomas Nagel compounds this criticism by pointing to the exigu-
ous foundation provided by Nozick for even the uncompromised
remnant of his libertarianism. Nagel insists that the scant foun-
dational remarks made by Nozick are an insufficient defense of the
absolute right to one’s historical entitlements. For example Nozick
argues that the ontological distinctness of persons implies that the
good of one cannot be justifiably advanced by imposing costs on
another because the latter acquires no benefit from any larger social
good in which he can be alleged to participate. Nagel responds that
even in the absence of such a combined good it is evident that the
improvement of the condition of a larger number of people is always
to be preferred to that of asmaller number. The existence of some
social whole need not be assumed in order to affirm such compara-
tive moral advantages. Moreover, Nagel argues, the desirability of
living a meaningful life cannot by itself imply as Nozick suggests it
does, an absolute right against interference by others. For the effect
of such non-interference upon those others who must, according to
Nozick, forbear from intrusive activity has to be weighed against the
interests of persons whose alleged rights are transgressed. Given
the paucity of argumentation for Nozick’s conception of rights and
the counter-intuitive claims embodied in what little argument is
presented in Anarchy, State, and Utopia Nagel is disposed to reject
what he calls a “Libertarianism Without Foundations.”
Nagel is, of course right, in indicting Anarchy, State, and Utopia
for not being the last word on libertarianism, but he is not being
entirely fair in doing so. The book’s significance lies in its power as a
serious work of recent libertarian thought by a professional phil-
sopher. Nozick’s explicit objectives for the book obviously fall short
of providing the sort of “knock-down” argumentation for his position
that we expect of the paradigmatic philosophical work. What he
hopes to do is to lend plausibility to his libertarianism by deflating a
variety of alternative positions, thereby leaving the field open to a
14 Introduction

possible successor. By subsuming all opposing theories of distribu-

tive justice under either or both of two categories (patterned or
end-state) and attacking these two in a way which he hopes will
divest them of all credibility, Nozick aspires to elevate natural rights
libertarianism to serious contention among competing political
philosophies. This aspiration is not ambitious enough to satisfy
Nagel who consequently impugns the book for its neglect of foun-
dational questions.
While Nozick has little to say in response to such questions in
Anarchy, State, and Utopia, he certainly recognizes their ultimate
significance for the credibility of his views and he has not been
unconcerned with them in past philosophical work. During the late
1960s and early 1970s his interest in natural rights libertarianism led
him to test the validity of a purported demonstration of that position
put forth by the novelist, Ayn Rand. Nozick’s examination of it is
interesting in two respects. First, he seems to feel that if Rand's
putative derivation of what she takes to be the ethical basis of
libertarianism is sound, then it would be a trivial task to deduce
from it classical liberalism’s panoply of rights. “Since I share the
view that such a moral foundation [Rand’s] is appropriate and
possible and that laissez-faire capitalism is morally justifiable on
such a basis, I wish to look closely at an actual attempt” (p. 206).
Second, in criticizing Rand’s moral philosophy Nozick implicitly
identifies the differences between himself and other libertarian
political philosophers (e.g., John Hospers, Tibor Machan, Eric
Mack, Douglas Den Uyl, and Douglas Rasmussen) whose views are
more or less informed by Rand’s ethical stance. While Nozick makes
little positive contribution to the exploration of the foundational
questions of libertarian theory some political allies of his have
undertaken to remedy that deficiency through an explication of the
structure and substance of Rand’s moral epistemology. Both
Nozick’s attack and their defense of the “Randian argument” com-
prise most of the remaining essays in this section.
Rand has argued that one can derive from factual premises the
normative conclusion that people ought to act in conformity with the
following injunction: “do that which is required for man’s survival as
a rational being.” To survive as a rational being, she argues, requires
the practice of certain virtues which in turn necessitates the creation
of a social fabric in which such practices are permitted. That social
 Introduction tS

fabric is reducible to an array of rights defined as the entitlement of

human beings to their persons and duly acquired property—the
libertarian rights so enthusiastically defended by Nozick. In his
paper, “On the Randian Argument,” Nozick reformulates Rand's
putative demonstration, putting it into deductive form in order to
assess its validity. Her argument, Nozick alleges, can be understood
in one of two senses. Either she is maintaining that human survival
is a value because it is a necessary condition of the realization of any
values whatever, or she is calling it a value because without it the
very notion of value becomes unintelligible. Against the first in-
terprétation Nozick raises two objections. First, there are a large
number of conditions necessary to the realization of values, yet
Rand selects one, human survival, as having overriding moral
significance. Why? Second, in order to establish the moral value of
that which is instrumental to the realization of values Rand would
first have to show that realizing values is itself valuable. Yet she fails
to do so. The second interpretation of Rand’s derivation, that life is
valuable, is dismissed by Nozick as well, since the very concept of
value is unintelligible when severed from it. Value can be ex-
plained just as meaningfully in terms of its relationship to “death”
or “the greatest happiness of the greatest number of people” or any
of a large number of possible concepts. The synthetic a priori
connection alleged to exist between “life” and “value” by Rand is in
Nozick’s opinion illusory. Now if Rand is unable to establish the
moral foundation of the rights that she ascribes to human beings, the
argument purporting to deduce those rights from that basis is
interesting but not conclusive. Two of her followers have attempted
to save that foundation from the damage done to it by the force of
Nozick’s criticism.
Douglas Den Uyl and Douglas Rasmussen contend that Nozick
has at least partially misconstrued Rand and so his case against her
rests upon a “straw man” argument. They further insist that to the
extent that he accurately rehearses her argument, his criticisms of it
are wide of the mark. Den Uyl and Rasmussen’s reconstruction of
Rand’s argument purports to show that the process of human
valuation and the condition of mortal existence are inextricably
linked so that any axiological scheme which repudiates the latter as
an ultimate criterion for normative judgments is internally inconsis-
tent. Any coherent system of value, they argue, must be ordered in
16 Introduction

terms of the contribution made by its elements to human survival.

They maintain, in addition, that the unique character of human life
implies a certain mode of conduct, i.e., rational and productive
conduct, as the only one appropriate for human survival. Such
conduct can achieve success only where life, liberty, and property
are preserved as its necessary preconditions. This reconstruction of
Rand's argument bears a striking similarity to Ellen Frankel Paul's
elaborated and improved version of Herbert Spencer's defense of
“Lockean” rights in his essay “The Great Political Superstition.”
Her essay which has as its principal focus the replacement of what
she believes to be Nozick’s unsuccessful repudiation of anarchism
with a more convincing successor, concludes by resurrecting and
repairing Spencer's deduction of libertarian rights. Paul begins by
criticising Spencer for failing to anchor his derivation in some
categorical normative principle prescribing the value of life. In-
stead, she says, he argues only that if life is a good then that which
tends to its preservation is good as well. But this means that the
rights which he alleges are instrumental to human survival are of
conditional, not categorical, value. If such rights are the invariant
features of the moral landscape that libertarian theory supposes
them to be, their value cannot fluctuate with our changing attitudes
about the desirability of life. Paul, then, offers an argument similar
to that of Den Uyl and Rasmussen which purports to make clear the
ineluctable relation between valuational processes and mortal exis-
tence, a relation which, it is maintained, makes of human survival a
summum bonum. Paul goes on to explicate the subsequent steps in a
demonstration of the libertarian position on social rights. Like
Spencer she contends that persons in close proximity to one another
need guarantees of non-interference with life-sustaining activities.
The human ability to grasp conceptually the necessity of such
sanctions implies that they are moral imperatives for those of normal
human intellect but do not apply to the conduct of those non-human
creatures incapable of grasping them.
The final piece in this section departs from the kind of transcen-
dental argument for life’s goodness invoked by Randians, Den Uyl
and Rasmussen or the Spencerian, Paul. Eric Mack’s “How to
Derive Libertarian Rights” purports to be an interpretation of Rand,
but differs significantly from the version given by the two authors of
“Nozick on the Randian Argument.” For Mack, Rand’s ethical
 Introduction 1G

theory has a Thomistic dimension which is not captured by the Den

Uyl-Rasmussen exposition of it. Mack’s version of Rand purports to
show that the valuational process has a natural function for human
beings to which it ought to conform if it is to function well. Its
function, according to Mack, is to further the existence of the
valuing human being. Since, he argues, “valuation functioning well”
is analyzed definitionally as “that which is good,” the “good” turns
out to be that which contributes to the survival of the valuing agent.
It is this “good” which serves as the basis for the libertarian rights
derived by Mack in the first part of his essay. Rights for Mack are
the interpersonal obligations required for agents to strive to further
their separate lives. Their obligatory character derives from the
pragmatic contradiction involved in any act that transgresses them;
for any rights-violating act is, for Mack, a statement which con-
tradicts that which is apodictically true—that the natural function of
human valuation and action is the promotion of human survival.
The combined efforts of Mack and his cohorts attest to the fact
that the neglect endured by libertarian theory on foundational
questions at Nozick’s hands is not mirrored in the work of his
philosophical compatriots. In fact, libertarian theory would appear
to be without ethical support in Anarchy, State, and Utopia only.
Elsewhere it seems to have produced a wide variety of foundational
proposals.

NOZICK’S HISTORICAL ENTITLEMENT THEORY OF

DISTRIBUTIVE JUSTICE

In the last part of this collection the problem of how a just distribu-
tion of goods is to be realized according to the theory of historical
entitlements is investigated. According to this theory individuals
obtain absolute control over objects through historical processes.
This means that they may retain or transfer the ownership of them
only at their own discretion. All compulsory transferences, there-
fore, constitute an infringement of rights. Similarly, any coercive
regulation of the uses of owned objects infringes the right of the
owner. That the “redistribution” of owned objects must result only
with and from the consensual agreements of their owners is, then,
an unambiguous implication of Nozick’s theory of distributive jus-
tice. Apparently the initial act of appropriation confers unlimited
18 Introduction

rights of use and disposition upon the acquirer, rights which are
conveyed by voluntary transfer to the subsequent owner of the
originally acquired objects. But what sort of principle governs the
appropriation of resources so that it confers such full “capitalist”
rights of ownership upon its adherents, rights which will accompany
all subsequent changes in ownership? On questions of distribution
generally Nozick is an advocate of what he calls the historical
principle of justice. As applied to original acquisition, this principle
specifies that initial entitlements should arise from the past actions
of persons rather than from some distributive formula. The latter
“end-result” type of principle can never be made compatible with
the moral requirement of personal autonomy according to Nozick.
For it will always require coercive measures in order to enforce
adherence to it. But, while some historical method is the moral
superior of any teleological method of initial acquisition Nozick has
difficulty in specifying precisely which of several possible methods is
to be preferred. Is it through labor, first occupancy, possession,
declaration or some other historical means that one appropriately
secures initial ownership of virginal resources? Nozick is ambivalent
on this issue although he seems at times to suggest that Locke’s
labor theory of original property acquisition might be acceptable, if
it is suitably qualified (174-78). The qualification that he selects is
the previously cited Lockean Proviso. Its original formulation by
Locke requires that a limit be placed upon the amount of a resource
that can be extracted from nature by anyone. That is, it insists that
“enough and as good” be left for others to secure. Nozick recasts this
constraint into an injunction against those acquisitions which de-
press the condition of others below a certain welfare baseline by
depriving them of the use of resources to which they previously had
access.
The original source of Nozick’s entitlements, its justification, and
its appropriate limitations are the central themes of the essays in this
part. These issues are not insignificant, for if the theory of distribu-
tive justice propounded by Nozick is correct, only a laissez-faire
economic order would satisfy its normative requirements. We
would then be morally committed to a radical revision of the type of
economic arrangements to which western society has aspired in this
century.
In her essay, Onora O'Neill addresses the problem of whether
 Introduction 19

Nozick has successfully justified the historical entitlement view of

economic distribution. She aptly recognizes that Nozick provides
little in the way of a positive foundation for his view, preferring
instead to attack opposing theses. This strategy—aimed at leaving
the historical entitlement principle the only unvanquished compe-
titor in a field comprised of all possible theories of distributive
justice—has, in O'Neill's opinion, failed. Its failure, she maintains,
is due to the circuitous character of the argument deployed by
Nozick against alternative conceptions of the distributive ideal.
Nozick argues that such alternatives are universally self-refuting.
The alternative models attempt to realize either a particular dis-
tributive structure (e.g., equality) or a formula which assigns quan-
tities (or to use Nozick’s term “patterns’) of economic goods to
particular individuals based upon some characteristic possessed by
them (e.g., good looks, merit, need). If goods are distributed to
accord with some example of either of these principles the distribu-
tion realized, D1, will in all likelihood shortly be supplanted by
another, D2, by virtue of the voluntary economic transactions that
people will engage in subsequent to D1. Any reversion to D1 will
require an imposed reversal of the decisions made by individuals
pursuant to the initial distribution. This means that on the one hand
persons were given quantities of goods for their use under formula
D1; on the other hand they are not permitted to employ these goods
as they choose, except when by their choice the distributive status
quo, D1, is maintained. To Nozick this allegedly paradoxical result
inevitably afflicts all patterned or end-result rules of distribution.
O'Neill, along with Thomas Nagel and Cheyney Ryan, has
criticized Nozick’s argument against “end-result” and “patterned”
principles of justice for its putative question-begging character. All
of them assert that it assumes that the recipients of goods under Di
are thereby accorded absolute rights of use and disposition. But it is
precisely such rights that have yet to be established. Nor can it be
argued that the liberties of the recipients are infringed by the
continuous reimposition of some distributive pattern, because, ex
hypothesi, their freedom to rescind the desired pattern of distribu-
tion, D1, is a freedom implicitly forbidden by D1. As Nozick has
presented no positive argument in support of the freedom to flout
such distributive formulae, he cannot reject distributionism for
inhibiting it.
20 Introduction

Can the full-blown property rights embraced by Nozick be per-

suasively defended? O'Neill and Ryan investigate this possibility and
conclude that they cannot. O'Neill produces a reconstruction of
Locke’s justification of his labor theory of property acquisition (in
the absence of any such argument by Nozick) only to conlude that it
fails. Ryan, in his essay “Yours, Mine, and Ours,” contends that
Nozick’s principal argument against distributionist theories of jus-
tice rests upon their purported failure to cohere with the ideal of
individual liberty. This criticism, Ryan points out, derives from
Nozick’s particular conception of human freedom which identifies it
with the rightful capacity to control property. But this is merely to
assume the propriety of that which must be demonstrated.
Furthermore, it can be plausibly contended, according to Ryan, that
private property inhibits freedom rather than expands it. The
transition from common ground to enclosed ground in England
rendered large tracts of land inaccessible to those who formerly had
the free use of it. Indeed, Ryan argues against Nozick that the right
to acquire personal property from nature is a source of increasingly
constricted autonomy.
While Ryan and O'Neill complain that unconstrained rights of
property use and disposition are not positively justified in Anarchy,
State, and Utopia, Lawrence Davis and David Lyons insist that
Nozick introduces restrictions upon the applicability of the historical
method of property acquisition which contravene the doctrine of a
pure entitlement to property. If they are correct in their assessment
then we would be forced to conclude that Nozick has not merely
failed to provide a justification for historical entitlement, but that a
more diluted version of this principle is actually espoused by him
than has been generally supposed.
The robust species of historical entitlement doctrine that has been
popularly attributed to Nozick enables individuals to become the
uncontested owners of previously unowned objects through the
proper employment of certain procedures. Those procedures “enti-
tle” their employers to the unlimited use of the formerly unowned
objects to which they have been applied. Such procedures are
invariant with respect to the rights that they confer—they do not
fluctuate to accommodate human needs or conventions. And the
rights that they confer are to the particular objects to which they
have been applied.
 Introduction oT

Davis points out in his piece that such a doctrine, if strictly

adhered to, would have unwelcome consequences for the principle
of rectification of injustice that it must invoke. For, if our entitle-
ments are to particular objects, the destruction of those objects
precludes the restoration of the transgressed rights of their owners.
But this is an unconscionable result for entitlement theory, leaving
it with no remedy for this type of rights infringement. Davis
suggests that Nozick has included in the first part of his work a
doctrine which can extricate the entitlement thesis from this di-
lemma but only at the expense of the purity of the doctrine. Davis
points out that in this first section of the work Nozick argues that
individuals can be compensated for a violated right by raising them
to a level of indifference between their pre-violation and post-
violation condition. This implies that it is their psyches which must
be appeased and that the goods facilitating that appeasement do not
necessarily have to be the ones forcibly seized, damaged, or de-
stroyed by the violator. Even if the removed entitlements no longer
exist the victim can be “compensated” and the injustice done to him
rectified by inducing in him a requisite level of satisfaction through
the return to him of “equivalent” goods. This obviously shifts the
basis of Nozick’s social philosophy away from an entitlement thesis
and toward a doctrine resting upon a utility criterion. Davis's insight
in this regard is reminiscent of Eric Mack’s.
David Lyons makes the point that one does not have to look as far
as Nozick’s rectification theory to discover a dilution of the historical
entitlement thesis. Lyons argues that the thesis is initially weakened
by Nozick’s adoption of his Lockean Proviso as a constraint upon the
unlimited extraction of resources from nature. This proviso defines a
welfare baseline below which no one can be permitted to fall due to
the acquisition of the entire supply of a resource. Lyons maintains
that the proviso makes property acquisition subordinate to human
welfare and thereby divests it of the character of a right. Lyons’s
interpretation of Nozick transforms his social theory into the sort of
teleological distributivism that his book repeatedly denigrates.
Hillel Steiner has drawn a more radical inference still concerning
Nozick’s view of historical entitlements. He contends that—since
the process of property acquisition creates nothing new but rather
involves the extraction of pre-existent resources from nature—
differential entitlements to virginal resources ought to be proscribed
wy, Introduction

by the Nozickian. Moreover, the equal right to liberty to which

Nozick apparently subscribes should commit him to an initial equal
distribution of natural resources. Such an egalitarian constraint
upon original acquisition (not explicitly proposed by Steiner in this
essay but described by him elsewhere®) is surely the most radical
inference that can be drawn from Nozick’s framework, substituting,
as is does, an “end-state” principle of appropriation for the “histori-
cal” Lockean one vigorously propounded in Anarchy, State, and
Utopia.
Steiner inference, however, is implicitly rejected by Israel
Kirzner who contends extraordinarily that most resources are
created ex nihilo by their appropriators. He argues that most
resource acquisition actually involves the discovery of either the
resource itself or of a hitherto unrecognized’ economic use for it. Ina
sense, then, all acquisitions involving such discoveries bring into
existence economic goods and values ex nihilo. If we accept the
ethical principle that the creator of something ought to acquire
exclusive title to it, Nozick’s historical theory of initial appropriation
is rescued from those who would attack it from Steiner's vantage
point. Moreover, Kirzner argues, this theory of the ethical basis of
historical entitlement removes the need for Nozick’s Lockean Pro-
viso. For, if virtually all primitive acquisition is not the mere
extraction of previously existing objects from nature but involves
rather the production of new objects ex nihilo no one can claim to
have been deprived of access to antecedently existing resources.
Extraction multiplies rather than depletes the available pool of
economic values. Instead of subordinating entitlements to the re-
quirements of the proviso as Lyons has proposed, Kirzner argues
that the Lockean Proviso ought to be dispensed with as a constraint
upon property acquisition and transfer. The creation of economic
values involved in most initial acquisition is the source of the
full-blown capitalist rights to property commonly associated with
the historical entitlement doctrine but which O'Neill, Ryan, and
Lyons deny Nozick has established.
The disputes over the precise content of and justification for
historical entitlement which enliven this final section of the book are
abundant testimony to the unsettled controversy which surrounds
Nozick’s libertarianism. No resolution of that controversy will be
found in this volume which is intended to be a prolegomenon
 Introduction a3

for further investigation. But the significance of the disputes stimu-

lated by Anarchy, State, and Utopia should escape no one in-
terested in social philosophy.

NOTES

1. Hillel Steiner, “Anarchy, State, and Utopia: Book Review,” Mind 86

(January 1977):120.
2. Brian Barry, “Review of ASU,” Political Theory 3 (August
1975):331-32.
3. Steiner, “Slavery, Socialism and Private Property,” in Nomos: Private
Property, ed. J. Roland Pennock and John W. Chapman (New York: New
York University Press, forthcoming).
4. Tibor Machan, Human Rights and Human Liberties (Chicago:
Nelson-Hall, 1975).
5. Steiner, “The Natural Right to the Means of Production,’ eC
cal Quarterly 37 (January 1977):41-49.ns
 esse i
- P .
: it , Ve _
<td, 2% A a>
= m ak

s
@ ¢
=<
7 ad =
:

»
— Oe

7 <

; ; 5 >
 I
An Overview of
Anarchy, State, and
Utopia
 : to at es KP,
brs ctate rvs
_ sig”
| 7 a
 1
The Minimal State
BERNARD WILLIAMS

Why is there a state at all? Or, rather, why should there be a state at
all? What is the justification of the state? The sense that these are
real questions has come and gone and come again at various times;
when that sense is present, the questions step in as the basic or first
questions of political philosophy. It is not obvious that they are real
questions, that the demand for ajustification is a sound one. For one
thing, one might be prepared to spend time on the justification only
if one had an idea of some alternative to the state, and it is
reasonable to feel that there are, at least now, no real candidates for
that.
Differently, one may reflect that thoughts about justification
could get a grip only if there were some set of principles or values
which were sufficiently independent of the state (in general) to give
one some leverage on the question. If the ideas by means of which
the state is to be levered into or out of the arena of acceptability
themselves presuppose the state, we shall effect nothing. But some,
with Hegel somewhere behind them, will feel that there is fair
doubt on that score—the moral ideas which supposedly provide the
leverage will seem to them not only to be the historical product of
the state (which is certainly true, but may not be damaging), but
also, in some damaging sense, to get their life from the context of
the state. This last group of doubters have also their Marxist
relatives, who can, dimly, discern a better world without the

2F
28 Bernard Williams

state—it is, after all, what the revolutionary process should eventu-
ally lead to—but who regard present moralizing about the existence
of the state as footling, and ideologically polluted.
Doubters of these last kinds are not going to be reassured by
Robert Nozick’s original, remarkable, and strikingly intelligent
book. He has nothing directly to say to their worries, though they
should have things to learn from various of his arguments, and they
should at least be forced by this energetic and inventive undertaking
to put their claims in a clearer and better-argued form. As should
anyone who wants to think effectively both about the existence of
the state, and about what goes on in the state: for Mr. Nozick not
only revives the exercise of justifying the state, but comes out with
some startlingly unfashionable conclusions about what it should be
up to—or at least, what it should ideally be up to. This qualification,
I shall suggest at the end, is of huge importance: the “ideally” is the
clue to why Mr. Nozick’s book is not what it seems, nor (still less)
what some unsavoury people will, with some encouragement from
the author, undoubtedly take it to be.
Mr. Nozick goes back tothe traditional business of justifying the
state from the ground up, the ground being provided by an imagi-
nary set of circumstances in which there is no state; this is called by
Mr. Nozick, as by the tradition, the State of Nature. This, in his
presentation, helps us to understand what the state is being justified
against: drawing on some particularly American elements in the
anarchist tradition, he spends much more time and ingenuity than
anyone else has ever done in spelling out how things might go in a
partly moralized State of Nature, where various private “protective
associations’ do the job, for a fee, of protecting people’s rights of
life, property, and so on, against force and fraud. It is partly
moralized in the sense that the people in it do, for a good bit of the
time, but not unfailingly, abide by moral considerations, where
these are identified by Mr. Nozick with a hard-core set of notions
about rights, linked rather loosely with Kantian ideas about treating
people as ends and not purely as means. (Utilitarians, of course, will
not accept the moral starting-point: Mr. Nozick assumes, I think
reasonably, that no one with whom it is worth having the argument
of this book will be a Utilitarian (really), and he makes on the way
some excellent remarks to encourage people to realize that they are
not.)
 The Minimal State 29

In taking the partly moralized starting-point, Mr. Nozick is in line

with Locke (an author to whom he closely, and surprisingly often,
refers back). More explicitly free, of course, than Locke from any
historical implications of the model, he differs more deeply by
excluding any idea of a contract: this is State-of-Nature theory
without the social contract. In its place, he aims to derive the state
from the starting-point of the model by a chain of events which
involve no intentional intervention: by what he calls an invisible
hand mechanism, adopting in this the language of classical
economics, which alone of the social sciences is used in the book and
whose methods provide some important parts of its intellectual
structure. The mechanism which eventually, and after some pretty
densely presented speculations, delivers the state without anyone
intending it embodies an important idea of Schelling’s in decision
theory (and has some similarity to David Lewis’s recent work on
convention). It is an elegant idea to apply to classical State-of-Nature
theory a mechanism by which one can arrive at a convention without
(so to speak) holding one.
The state which is delivered is, as once more with Locke, the
minimum “night-watchman” state of classical liberal theory, doing
no more than protect its citizens from force and fraud and such like,
leaving them free to pursue their individual projects. Mr. Nozick
shares Locke’s distaste for taxation, and there is a tough-minded
economic argument to represent it as a form of forced labour
(though I must warn the CBI, before they prematurely rejoice over
this liberating intellectual event, that we shall see that the conse-
quences of all this for things as they are are very far from clear).
Traditional State-of-Nature theorists, in justifying the state, inevita-
bly justified a state of one kind rather than another, with one set
rather than another of powers and restrictions: that was a main point
of the exercise. Mr. Nozick is no exception, and the argument in the
first part of the book, which justifies against the anarchists the
existence of the minimal state, is followed by a second part (as
ingeniously argued, but more relaxedly written) which claims
against socialists, nationalists, and indeed most people that the
minimum state is the most that can be justified, and that more
ambitious moral claims for the role of the state, in particular to
produce justice by redistributive measures, are mistaken. Such
powers, which are of course claimed in varying degrees by all
30 Bernard Williams

modern states, have, according to Mr. Nozick, no moral basis and

offend against people’s individual rights.
The major effort in the second half of the book is the attempt to
argue against conceptions of justice (in particular, but not exclu-
sively, that of John Rawls) that yield redistributive conclusions, and
to give another conception, which does not: we shall come back to it
later. The book ends with an engaging sketch of a pluralistic,
libertarian Utopia, which has the unusual property of really carrying
through the libertarian ideal by not laying it down even that people
should live in a libertarian manner. The libertarian arrangements
exist at the higher-order level of permitting a large number of very
various communities between which people may move—they are all
ordered within the merely Lockean framework, but in themselves
they may be as restrictive or unpermissive as you (or rather they)
like. Some of the difficulties which might spring to mind about these
arrangements are rather disarmingly foreseen.
There is also, it should be mentioned, a bravura short chapter in
which it is argued that the modern state might, after all, be justified
(that is, in Mr. Nozick’s terms, could come into existence without
violating anyone’s rights). However, the squeamish reader should
be warned against pressing this argument against the general tenor
of Mr. Nozick’s conclusions. For the construction proceeds via
everyones selling himself into slavery; and while Mr. Nozick him-
self, more permissive here than Locke, thinks that everyone has the
moral right to do this, and hence that the results of it are not for the
reason impermissible, it is typical of this structurally sophisticated
and self-aware book that the reader should find himself in an ironical
stand-off with Mr. Nozick on this way of getting to the modern state.
The two major parts of the book, the minimally positive and the
ambitiously negative, are connected with each other in more than
one way. The aim of the second part is to show that no larger state is
justified, by removing what Mr. Nozick takes to be the major moral
arguments in favour of such a state, namely arguments from dis-
tributive justice (only the larger state can be in a position to
redistribute, in the interest of what mistaken theories take to be
justice). Other arguments, and indeed other moral arguments,
might be thought of to support the state. But Mr. Nozick is
interested only in a relatively narrow range even of moral con-
siderations: those, roughly, to do with rights, justice and the
 The Minimal State oy

crossing of one person’s “moral boundaries” by another. It is this

which dictates the narrow compass of what he thinks has to be said
about the more ambitious state; equally, it controls the construction
of the minimal state.
Now there is an argument for using the absolute minimum of
hardcore moral notions in the first part of the book in the justifica-
tion of the minimal state. For here one is arguing with someone like
the libertarian anarchist, whose ideas these are, and the shape of
the argument is to say: “Look, even with those (few) moral ideas one
can get to the state” (though of course the rest of us, neither
libertarian anarchists nor very tempted by them, might say right at
the beginning that the anarchists’ bag of moral ideas was too small
anyway, and that we saw no interest in trying to cram that morally
elaborated item, the state, into it). But, even if one is prepared to
defend the state against them on the strength of this moral hard-
tack, one is not bound to say that this is all the argument, even of a
moral character, that can be brought to bear on the state) nor,
correspondingly, are we bound to think that the only defence of the
moral character, that can be brought to bear on the state; nor,
closely related, notions. We might think that there were other
values besides justice which the more elaborated state alone could
advance: and that there was nothing in the intuitions employed in
the first part of the book which will exclude these other values being
weighed in the evaluation of the state. Mr. Nozick’s defence of his
negative claim, that the more elaborate state is not justified, is
inevitably weakened by the restrictions imposed on the range of
arguments which he considers in favour of such a state.
Not just the second part of the book, but the first part as well,
would be weakened if it could be shown that even the hardcore
values, the minimal moral package of notions about rights, derived
some essential support from sources outside the limited repertoire
of the State of Nature (one form of that view, of course, would be
held by those theorists I mentioned at the beginning, who think that
in one way of another the minimal moral notions themselves derive
their life from the state—indeed, from the elaborated state). Locke's
treatment has often been criticized on that score; Mr. Nozick’s
sophisticated reversion to Locke collects sophisticated versions of
the criticisms, and it is surprising he has not done more to head
them off.
a2 Bernard Williams

In particular, he has tried—using, obviously, much ingenuity in

the attempt—to get to his destination while avoiding any general
discussion of a notion central to his views: property. He does have
something to say about Locke’s requirement that, when those in the
State of Nature acquire things, there should be “enough and as good
left over’ for others. But, apart from that (which presents certain
special problems for his theory of entitlement), he does not really
address himself to the issue of what is an originally just holding at
all, or of property as (what he requires) a purely moral notion.
Hence, while there is a great deal in the discussion of the State of
Nature about people’s “boundaries” and how they get crossed, there
is no discussion of where their boundaries are, or of how they get
drawn. There is thus a persistent doubt about whether the State of
Nature can really be got off the ground without taking for granted
conventions and institutions of a kind which the State of Nature does
not itself provide. Certainly we cannot hope to get clear about this
just by using such intuitions as we have as things are about non-legal
and informal ideas of property—they can too readily be seen as
extensions of more institutionalized notions.
Another difficulty with the State-of-Nature argument is what its
rules are. Mr. Nozick reckons to have succeeded in his task against
the libertarian anarchist if an invisible-hand mechanism would
produce the state without violating anyone’s rights, granted that the
individuals are (partly) moralized. But there is an obscurity about
why this thought-experiment operates as it does (why, one might
equally say, this is the thought-experiment): Mr. Nozick presents
us with a set of persons who behave like economic men, but within
the side-constraints, for the most part, of minimal morality; only “for
the most part”, since a lot of the machinery is designed to deal with
persons who do violate others’ boundaries, and not all such viola-
tions are unintentional. Now the steps in the development of the
thought-experiment, though many and complex, are notably un-
realistic, if judged from the position of social or psychological
credibility. Thus, to take just one of many examples, the protection
agencies, which are in economic competition with one another,
show a commendable zeal in establishing the rights and wrongs of
claims against their clients; but even a modest lack of optimism
about human nature would suggest that in fact they would be partial
 The Minimal State 33

towards their clients, hypocritical towards potential clients, and

horrible towards confirmed non-clients.
Now Mr. Nozick is not an optimistic idiot who disbelieves this;
the point is, these considerations do not count. But why not? What
weight is there in the fact that we could, relative to certain wildly
idealized psychological assumptions, reach the state without violat-
ing anyone’s rights? The motivation seems to be, that the (minimal)
state will have been justified if it can be generated by steps each of
which satisfies moral demands; and this is taken to mean that we can
get there without anyone doing anything wrong. But how is this to
be taken? The condition cannot be that we should be able to get
there without anyone doing anything wrongat all, since it is a fact that
some people sometimes do wrong which essentially contributes to
our getting there, and helps to power the invisible hand. By the
same token, the condition cannot be that there is no wrong whose
happening is essential to our getting there. So what exactly is it?
How much wrong goes into the model, and where? Why cannot a
sceptic resist the invisible-hand derivation, on the ground that its
pictured working is too free of wrong to be plausible? As it stands,
Mr. Nozick seems—though I am not sure of this—to have settled for
individuals in the model sometimes doing wrong, but associations
not doing wrong; if that is correct, the model seems arbitrary. In any
case, the derivation as it stands lacks any evident ground for being
precisely as un-Hobbesian as it is.
Still less, of course, does it justify any existing state: for no state
arose in this way, and it is Mr. Nozick’s thought, certainly in his
theory of justice, and I take it here, that how a state of affairs
actually arose is crucial for its acceptability. This is the basic idea of
his theory of justice as entitlement, and of his criticism of Rawls (and
many other conceptions). A holding is just, on this view, if it has
been acquired by a just process from a holding which is itself just: at
the beginning is a notion of just acquisition (on which, as I have
already said, Mr. Nozick has notably little to say). Supplementary to
the processes of just acquisition and just transfer are processes of
rectification for situations which are unjust in one of these respects
(holding, or transfer, or previous rectification): and that is all there is
of the basic theory of justice.
Mr. Nozick makes elegantly clear the difference between such a
34 Bernard Williams

historical theory of just holdings, and an end-state theory, which

concerns itself essentially with the pattern in which holdings end
up, and seeks to adjust the pattern to some desired paradigm:
Utilitarianism, and egalitarianism, and Rawls’s view are all end-state
views. (There is a very nice demonstration that the State-of-Nature
model which Rawls uses, that of the Original Position, is so designed
that it could only yield an end-state conception of justice.) It will be
clear how, at an ideal level of politics at least, Mr. Nozick’s
conception, as against end-state views, favours a strongly conserva-
tive outlook.
What are the intuitive merits of these ideas? There are questions,
right at the beginning, about how to argue these issues. Mr.
Nozick’s basic method, throughout, is to take some everyday,
non-political situation about which we are likely to agree, and apply
our judgment in it to the larger issue of social principle—a method
which, it might be argued, begs the question in his favour, since it
presupposes his view that no new moral principles arise (should
arise?) with the state. But even running the argument by his rules,
his conception of justice does look like an enormous exaggeration of
at best one aspect of our moral ideas. It is hard to know how far this
is so, in fact, because, once more, we lack any theory of original
entitlement. But suppose that, when the Mayflower arrived, some
foresighted fellow, crouching by the gang-plank, jumped off and
bagged a good area of what is now Massachusetts, before his
companions, more cooperative, pious, idle, or enfeebled, got going;
it looks as though Mr. Nozick, if we assume there would no prior
holders, would grant him just title. Do we agree? Would it be unjust
to redistribute in favour of those others (even the idle)? Wasn’t it
unfair of this man to take advantage of the fact that the others did
not spend those crucial moments thinking about property rights?
Would a certain fact about the end-result, namely that the nice guys
came (nearly) last, have no effect at all on our estimate of the justice
of this man’s holdings?
These are questions for Mr. Nozick’s theory of justice (and its
application to this case); but we can notice more broadly that, even if
we eventually agreed that this pushy settler was not to be faulted in
justice, that would only underline the point that we could hope that
the Pilgrims, and ourselves, would have arrived with more virtues
than justice. We are reminded again of that richer range of moral
 The Minimal State 35

resources (of the kinds of character, for instance, that we want to

have in society) which Mr. Nozick’s treatment systematically leaves
out.
There are other ideas and sentiments relevant to justice, which
his treatment also passes over. What advantages, and their rewards,
are candidates for redistribution is a real question, which egalita-
rians should face more honestly than they mostly do; but the fact
that we should agree (most of us) with Mr. Nozick that compulsory
plastic surgery was no just reaction to inequalities in good looks,
need not commit us all that quickly to agreeing with him on the
evidently different matters of money and power. Again, and near
the heart of Mr. Nozick’s view, the very matter of distance (in time,
over successive transfers, or whatever) does in fact affect the
sentiments of many about injustice: the luck of the talented com-
mands more respect, some find, than the luck of those who merely
had a talented father. These are also “our” notions, in as good
standing, at least, as those to which Mr. Nozick appeals, and his
conception of justice merely as a pipe for the rightful delivery of
rights over any distance is not tested against enough notions to be
really persuasive.
This is a book of a very highly theoretical character; indeed its
theories themselves have a tendency to pursue the virtues of formal
elegance rather than of concrete realism, as witnessed by the
presence of much economic theory and virtually no psychology or
sociology. This leaves the conclusions rather high in the air, par-
ticularly above present political realities. These views leave unde-
termined to a high degree what should now, in current political
practice, be done—to a greater degree than most political theory,
including Locke’s; they are in a deep sense Utopian, and the third
element in the title is rightly juxtaposed with the others. This is not
necessarily a failing: but it should be written in larger letters what
the book does not offer. Above all, its theories do not, except in a
very general and associative manner, offer any particular comfort to
contemporary capitalism. For one thing, contemporary capitalism is
a statist enterprise. For another, Mr. Nozick’s derivation theory of
justice does not imply that contemporary property holdings are just;
on the contrary (though it is a matter of unrecoverable fact), it is 99
per cent probable that almost all of them are not. (Mr. Nozick may
well think that much of America rightfully belongs to the Indians.)
36 Bernard Williams

And in a vitally important but unemphatic passage (page 231) he

makes it clear that redistribution by the state may well be, as things
are, necessary for the rectification of past injustice. There is little
comfort in these pages for contemporary friends of business; but Mr.
Nozick hardly makes it as clear as he might that this is so.
Within this abstract, complex, clever, and always stimulating,
structure there is to be found, one suspects, a robust and romanti-
cally creative individualist outlook which, though undoubtedly
tough, is in quite a different street from that of the friends of
business's nastier friends. But it will be partly Mr. Nozick’s own
fault if they, and their enemies, think otherwise.
 2
The Right to be Rich or
Poor
PETER SINGER

When times are hard and governments are looking for ways to
reduce expenditure, a book like Anarchy, State, and Utopia is about
the last thing we need. That will be the reaction of some readers to
this book. It is, of course, an unfair reaction, since a work of
philosophy that consists of rigorous argument and needle-sharp
analysis with absolutely none of the unsupported vague waffle that
characterizes too many philosophy books must be welcomed what-
ever we think of its conclusions. The chances of Gerald Ford
reasoning his way through Nozick’s book to the conviction that he
ought to cut back the activities of the state in fields like welfare,
education, and health are not high. The book will probably do more
good in raising the level of philosophical discussion than it will do
harm in practical politics.
Robert Nozick’s book is a major event in contemporary political
philosophy. There has, in recent years, been no sustained and
competently argued challenge to the prevailing conceptions of social
justice and the role of the state. Political philosophers have tended
to assume without argument that justice demands an extensive
redistribution of wealth in the direction of equality; and that it is a
legitimate function of the state to bring about this redistribution by
coercive means like progressive taxation. These assumptions may be

37
38 Peter Singer

correct; but after Anarchy, State, and Utopia they will need to be
defended and argued for instead of being taken for granted.
Anarchy, State, and Utopia falls into three sections, as its title
indicates. Part I tries to show that a minimal type of state—the
“nightwatchman” state of classical liberal theory, limited to pro-
tecting its citizens against force and fraud—can arise legitimately,
without violating anyone’s rights. In the second part Nozick argues
that the minimal state is the most extensive state that can be
justified and that any more extensive state does violate people's
rights. The book ends with a section contending that the minimal
state is, harsh appearances notwithstanding, an ideal worth fighting
for.
All three sections are well worth reading, although the third is the
slightest. Here Nozick, finding incredible the supposition that there
is one best form of society for everyone, proposes instead a “meta-
utopia’ —a framework for many diverse utopian experiments, all
formed of voluntary communities, so that no one can impose his
version of utopia on others. Within a community people may
voluntarily adopt redistributive measures, and those refusing to
participate may be excluded from the community; but within a
nation, which would include many communities, there should be no
compulsory redistribution. The idea is appealing because it en-
hances individual freedom. But there are serious objections that are
not adequately considered. Could a community that wanted a lot of
redistribution survive the departure of the wealthy members whose
moral principles are weaker than their desire for wealth? Could it
withstand the pressure of applications to join from the down-and-
outs left to starve in neighboring communities run by ruthless
capitalists?
Or, to take a different kind of objection, could a community
maintain its dedication to an austere life of virtue if it were sur-
rounded by the flashy temptations of America capitalism? Nozick
would say that the choice between austere virtue and flashy tempta-
tion must be left to the individual; but doesn’t this assume an ability
to make free rational choices that most people simply do not
possess? Is the free flow of information sufficient to wash away the
encrusted muck of billions of dollars worth of advertising for a style
of life devoted to the acquisition of consumer goods and the elimina-
tion of stains and odors? Nozick’s vision of utopia fails to deal with
 The Right to be Rich or Poor 39

the fundamental Marxist objection to classical liberalism: people

may make choices, but they do so under given historical cir-
cumstances which influence their choices. We do not enable people
to govern their lives by giving them a “free” choice within these
limits while refusing to do anything about the contexts in which
these choices are made.
To say this smacks of paternalism and has unpleasant totalitarian
associations. But what if the choice lies not between paternalism and
freedom, but between making a deliberate attempt to control the
circumstances under which we live and allowing these cir-
cumstances to develop haphazardly, permitting only an illusory
sense of individual liberty? I ask the question seriously, not rhetori-
cally. Perhaps it can be answered, but Nozick passes it by with a
fleeting reference to Tocqueville's idea that being free develops the
capacity for freedom, and this reply does not touch the heart of the
issue.
The arguments of Part I are directed mainly against the anarchist
who objects to any state at all. Nozick does not say that a state is a
good thing and we are all better off with a state than we would be
without one. This obvious procedure for dealing with the anarchist
would be foreign to Nozick’s entire approach and would set a
precedent subversive of his aim in the second part. Instead he
maintains that we can get from a state of nature to a minimal state
without violating anyone’s rights, so that there is no point at which
anyone can claim that the state has assumed authority illegitimately.
Nozick begins his story in a state of nature modeled on that of John
Locke, but he leaves this natural condition by another route,
avoiding the need for the agreement or social contract that has been
a source of so much criticism for Locke and his followers.
Nozick’s minimal state, or “state-like entity” as he sometimes calls
it, is a kind of protection agency to which people in the state of
nature pay a fee for protection from assault, robbery, and so on.
Nozick argues plausibly that clients of the agency would give up to
the agency their rights to punish violations of their rights, and that
one protective association, or federation of protective associations,
would become dominant in each geographical territory. So, without
any express agreements or over-all intention on anyone's part,
people in the state of nature would find themselves with a body that
satisfies two fundamental conditions for being a state: it has a
40 Peter Singer

monopoly of force in its territory, and it protects the rights of

everyone within the territory.
Together with the story of the development of the state in the first
part there are many other interesting subsidiary discussions. There
are sensible answers to such puzzling questions (for laissez-faire
liberals) as why blackmail (payment for the service of silence about
another's affairs) should be prohibited; and why, for that matter, we
should ever prohibit anything, rather than allow violations of rights
provided that the victims are adequately compensated. Although
Nozick admits that the book contains no full-scale presentation of
the moral basis for his views, there is some unorthodox moral
philosophy, including a lengthy discussion of the place of nonhuman
animals in morality. Nozick thereby becomes one of the small but
growing number of contemporary philosophers who have given this
neglected topic genuine consideration, and he joins those who urge
radical changes in our treatment of nonhumans, including the
recommendation that we stop eating them.
Interesting as the first part is, for those of us who have little
difficulty in accepting the moral legitimacy of some minimal kind of
state, the excitement begins only when we enter the second part. A
reader who is sympathetic to government policies designed to
redistribute wealth and who has taken for granted the justice of such
policies will be surprised at the strength of the arguments Nozick
brings against this view.
One book cannot deal with all the reasons that have been urged in
support of extending the functions of the state beyond the protec-
tion of its citizens against force and fraud. Therefore Nozick selects
what he considers the strongest, and most widely accepted, case:
the claim that a more extensive state is justified in order to achieve
justice in the distribution of wealth. It is this claim that receives the
brunt of his attack on the extended state.
Nozick uses the term “holdings” to describe the goods, money,
and property of all kinds that people have. The issue, then, is what
holdings people would have in a just society.
The position Nozick takes is a radical departure from the theories
of distributive justice discussed by most philosophers, especially in
recent years. Nozick characterizes the principles of justice usually
advocated as “patterned.” A patterned distribution is one which (to
put the matter more loosely than Nozick does) can be summed up in
 The Right to be Rich or Poor 4]

some simple formula of the type: “To all according to his—.” The
blank can be filled in by “need,” “labor,” “moral desert,” “IQ,”
“noble blood,” or whatever—the result will always be a patterned
distribution. In any existing society, the distribution of wealth will
presumably not correspond exactly to any preordained pattern, so
that to achieve a just society we shall have to take a bit from here
and give a bit there, until people’s holdings correspond to what we
think is the right pattern.
In contrast to all patterned theories, Nozick proposes the “enti-
tlement theory”: a distribution is just if it arises from a prior just
distribution by legitimate means. Basically, you originally acquire
something justly if you take something that belongs to nobody,
without thereby making worse the position of others no longer able
to use the thing. (For example, I can appropriate land for myself if it
is unowned and there is enough good land left for others.) Here
Nozick again follows Locke, although his account is more precise.
Then there are legitimate ways of transferring things you own,
especially voluntary exchange and gift. As a result there is no
pattern to which a just distribution must conform. People may
choose to retain what they start with, or give some of it, or all of it,
away. They may make profitable investments, or unprofitable ones.
They may live frugally and hoard what they have, or dissipate it in a
wild spree. They may gamble. So long as their original holdings
were justly acquired, and the decisions they made involved neither
force nor fraud, the result will be just no matter how widely people's
holdings vary. The entitlement theory of justice makes the justice of
a given set of holdings depend on the history of those holdings, and
not on the conformity of the outcome to a given pattern.
Both the strengths and the weaknesses of the entitlement theory
are immediately apparent. On the one hand, can it really be just
that one baby should come into the world with a multi-million-dollar
trust fund, the best possible schooling, and family connections with
the nation’s leading politicians and financiers awaiting him, while
another baby faces life in a dingy apartment with no money and
nothing else to help him on his way in the world? Neither baby at
the moment of birth can possibly deserve anything; an equal
division would therefore seem the only just one.
On the other hand, if the father of the first baby acquired his
holdings legitimately, violating no one’s rights in the process,
42 Peter Singer

doesn’t he have the liberty to give whatever is his to his son, if he

should so choose? Isn’t it implied in someone’s owning something
that he has the right to do with it what he will, provided he violates
no one else’s rights? And surely it is far-fetched to hold that the
poorer baby has a right to some of the other baby’s wealth, merely
because his ancestors were less fortunate, less astute, or less frugal
in their handling of their holdings.
Our intuitions lead us in both directions. One must be wrong.
Nozick tries to convince us that it is the former set of intuitions—
those relating to the injustice of inherited wealth and other inher-
ited assets—that we should give up. He does not attempt the
hopeless task of arguing that those born with large fortunes or
valuable natural talents have done anything to deserve these assets.
Nevertheless, he says, people are entitled to their inherited assets,
whether or not they deserve them. In the case of wealth he points
out that orthodox theories of justice overlook the right of the donor
when they consider the worthiness of the recipient of the inheri-
tance. As for natural talents, people do not violate anyone else's
rights by having the natural talents they are born with. An artist has
the right to keep a painting he has done even if his artistic talent was
inherited and he did nothing to deserve it. So why shouldn't a born
entrepreneur have a similar right to the fortune his talents have
brought him through legitimate means?
The legitimacy of redistribution in the direction of equality is, as
Nozick says, more often assumed than argued for. We discover that,
say, the wealthiest 5 percent of the population hold 40 percent of
the national wealth, and then we ask what can be done about it. On
the entitlement view these facts do not in themselves suggest that
we ought to do anything. It all depends on how the present
distribution came about. It might have come about by unjust means,
through force and fraud, or through an unjust original acquisition, in
which case reparations should be paid to those who are now worse
off because of this injustice (though Nozick is unable to explain how
we decide whether a person’s ancestor left sufficient good land for
others when he appropriated his first field five hundred years ago).
But the present distribution might also have come about entirely
legitimately, in which case the compulsory redistribution of wealth
would be a serious violation of people’s rights.
Nozick’s position sounds severe, and so it is. According to Nozick
 The Right to be Rich or Poor 43

we have no obligation to help those worse off than we are. If a

starving man drags himself to our house, where we are entertaining
our friends with a sumptuous banquet, we are perfectly within our
rights in sending him away without a crust. In mitigation, though, it
is important to remember that Nozick has nothing against voluntary
donations from the rich to the poor. The rich are within their rights
to keep everything they have and throw what they cannot use down
the sewer; but they also have the right to give everything away, and
the generous and charitable will no doubt give some away.
Indeed, on the question of voluntary donations Nozick has some
interesting points to make. He argues, I think conclusively, that
those relatively wealthy people who advocate greater government
redistribution (which would take from people like themselves and
give to those poorer) can have no sound reason for not making, while
they wait for the government to act, voluntary donations from their
own pockets of the sum that would be taxed from them under the
scheme they advocate. Presumably this argument applies to those
who advocate greater government foreign aid, as well as to those
who limit themselves to internal redistribution.
An ingenious illustration buttresses the entitlement theory. We
start by supposing that holdings are distributed in accordance with
some patterned conception of justice—let’s say the conception of
equality, so that everyone has exactly equal holdings. Now suppose
that several basketball teams would like to have Wilt Chamberlain
playing for them. He signs a special contract with one, stipulating
that he gets twenty-five cents from the price of every home game
ticket. The fans are happy to pay the surcharge; the excitement of
seeing Chamberlain play is worth it to them. One million people
attend during the season, so that Chamberlain winds up with
$250,000, far more than anyone else in the society.
The transactions between Chamberlain and his fans have upset
the original, hypothetically just, pattern of holdings; but, Nozick
asks, is the new distribution unjust, and if so, why? Can it be a
source of injustice that a million people chose to spend twenty-five
cents on seeing Chamberlain play, rather than on candy bars or
magazines? Since they chose to spend it in this way, knowing that it
would go to Chamberlain, surely they can have no just claim against
the man they have made rich. As for those citizens who did not
attend the games, their holdings are entirely unaffected by the
44 Peter Singer

transactions between Chamberlain and his fans. If these third

parties had no just claim against the holdings of the transacting
parties before the payments took place, how can the transfer give
them a just claim to part of what was transferred? Yet that is
precisely what those who accept taxation for redistributive purposes
must believe.
In general, Nozick says, no patterned principle of justice can
prevail without continuous interference in people’s lives. A socialist
society would, as he puts it, have to “forbid capitalist acts between
consenting adults.”
I have been able to indicate only the main strand of Nozick’s
argument. There are many fascinating sidelines as well. For in-
stance Nozick is able to show that if workers’ control of factories is
desirable, it will be possible to establish it within the framework of
his theory, by voluntary action. Indeed, he points out, the larger
trade unions already have sufficient financial reserves to set up
worker-controlled enterprises; and even smaller groups, or a single
wealthy radical, could do the same, especially since consumers who
favor worker-controlled enterprises could band together and buy
only from these companies. Why, Nozick asks pointedly, has this
not happened?
Nozick also challenges the view that greater equality will produce
an equality of self-esteem and the elimination of envy. Self-esteem,
he claims, is based on criteria that differentiate; if these criteria are
equalized it will need to be based on something else. Trotsky’s
vision of a communist society in which the ordinary man is able to
fulfill his potential to such an extent that he becomes an Aristotle, a
Goethe, or a Marx does not mean that the ordinary man will have
greater self-esteem. New peaks will rise beyond the heights of
Aristotle-Goethe-Marx, the ordinary man will think of himself as
just another Aristotle-like commoner, and envy the new super-
Aristotles.
There is also some hard-headed discussion of the Marxist idea of
exploitation and the labor theory of value. On these side issues
Nozick may not always be right, but he is always stimulating; an
open-minded study of what he has to say could be a healthy tonic for
romantic leftists.
On the main issue, what I have said should be enough to show
 The Right to be Rich or Poor 45

that Nozick’s case against compulsory redistribution is strong. Can it

be met, and if so, how?
The first question here is whether to attempt to meet Nozick on
the ground he has chosen—ground clearly indicated in the very first
sentence of Anarchy, State, and Utopia:

Individuals have rights, and there are things no person or group may do to
them (without violating their rights).

So we must decide whether to try to show that a system of justice

based on individual rights, including a right to property, can
permit—or require—compulsory redistribution; or, on the other
hand, to deny that individuals have the rights that Nozick says they
have, in the strong sense of the term that he intends.
In raising this question we come back to the most basic division
between moral and political philosophers of modern times. For
centuries there have been two lines of thought about justice.
According to utilitarian theory, espoused by David Hume, Jeremy
Bentham, and later utilitarians, principles of justice are rules that
work for the greater good of all. They are governed by the principle
of utility. If we take from the rich and give to the poor we do so not
because the poor are entitled to some of what the rich have but
because the poor will benefit more from this redistribution than the
rich will suffer. The utilitarian who is not trying to hedge will admit
that his account of justice allows property to be confiscated from one
person so that another, or several others, may benefit.
The alternative view of justice associated with John Locke and
Immanuel Kant starts with individual rights and prohibits the use of
one person as a means to another's end. The incorporation of
Lockean rights into the Declaration of Independence and the
Constitution of the United States ensured the dominance of this
tradition in the political rhetoric and in the moral, legal, and
political thinking of this country. There is a certain appropriateness
in the fact that Nozick’s chief opponent within this tradition is the
American philosopher, his Harvard colleague, John Rawls.
In his recent and widely celebrated book, A Theory of Justice,
Rawls tried to develop a conception of justice that would be an
alternative to utilitarianism, taking seriously “the distinction be-
46 Peter Singer

tween persons’ (which he claims utilitarianism, in subordinating

individual rights to the general good, does not do) and ruling out
“even the tendency to regard men as means to one another's
welfare.” The problem Rawls faced, however, was how to square
this with his intuitive conviction that justice requires us to improve
the condition of the poorest members of our society, whose poverty
is not really their own fault.
Rawls attempted to solve this problem by arguing that if people in
what he calls “the original position’—a hypothetical state of nature
in which, to ensure impartial decision-making, people are assumed
to be ignorant of their own talents and socio-economic status—were
to choose the fundamental principles of justice to be followed in a
newly formed society, one of the principles they would choose
would be that inequalities are allowable only in so far as they
improve the position of the worst-off group in the society.
Rawls thinks that people in the original position would make this
principle—which has been called the maximin rule, because it seeks
to maximize the minimum level of welfare existing in the society—
subordinate to another principle guaranteeing maximum equal lib-
erty for all. Whether they would give this priority to liberty need
not concern us here, since we are considering only economic
redistribution.
Rawls’s maximum principle is compatible with considerable in-
equality. If, as some economists argue, steeply progressive taxation
reduces the incentive to work of the most talented members of
society to the point where they contribute less to the society and
everyone, including the worst-off, suffers, then it would be just,
according to Rawls’s principle, to allow these people to keep most of
their wealth, although others may have much less. Doctors, for
example, might be allowed to keep more than others. Nevertheless
the maximin rule is difficult for egalitarians to argue against, because
any attempt to approach closer to equality would necessarily, at the
same time as it narrowed the gap between the worst-off and the
better-off, make the worst-off still worse-off than they were before.
Though strongly protected against the attack from egalitarians
that appeared most likely, the maximum principle was soon shown
to be vulnerable at other spots. Since the appearance of A Theory of
Justice a book and a number of critical reviews? have exposed
fundamental weaknesses in its central arguments, including the
 The Right to be Rich or Poor 47

argument for the maximin principle. The devastating critique of

Rawls in Anarchy, State, and Utopia, directed especially at the case
for redistribution in accordance with the maximin rule, must very
nearly complete the demolition of Rawls’s impressive structure.
In part, the force of Nozick’s criticisms depends on Rawls’s own
desire that his theory account for and systematize the particular
judgments about justice that we ordinarily make.? For Rawls,
finding a plausible general theory that confirms most of our ordinary
judgments of what is just is the aim of any theory ofjustice. But as
Kenneth Arrow has noted in a discussion of Rawls’s theory,4 the
most widely held intuition about distributive justice—which Arrow
and most other teachers find it difficult to dissuade introductory
students from thinking completely self-evident—is the view that an
individual is entitled to what he creates. This view is, of course,
much closer to that of Nozick than to Rawls’s.
I believe that Rawls is mistaken in thinking that the test of a moral
theory is its ability to‘account for the particular moral judgments we
already make. That approach comes too close to making the justifica-
tion of what we already believe the sole task for moral philosophy.
One of the strengths of Nozick’s criticism, however, is that even if
Rawls were to abandon his ideas about how moral theories are to be
tested he would still be unable to defend his position. For Nozick
has shown that Rawls’s case for the maximin principle rests on an
unjustifiable asymmetry between the worst-off and the best-off in a
society. Rawls argues that the worst-off could accept the justice of,
and cooperate in, a society governed in accordance with the maxi-
min principle, but not one governed according to, say, the principle
of utility. This is because in any society governed according to any
principle other than the maximin principle there would always be a
group of people at least as badly off as the worst-off in a maximin-
ruled society.
Provided the maximin rule has been properly applied, this is
necessarily true; but, Nozick insists, Rawls glosses over the equally
important mirror-image question: why should the better-off accept
the justice of and cooperate in the society? Under the maximin rule,
after all, the better off may have to make substantial sacrifices to
help the worst-off, perhaps much greater sacrifices than they would
have to make to satisfy the principle of utility. For instance, to put
the matter in monetary terms, assume that a tax of 75 percent on all
48 Peter Singer

incomes over $15,000 would, after deducting administrative and

other costs, allow welfare payments to the worst-off group to be
increased by only $1 per person per year. The maximum rule would
require that the tax be levied.
So Rawls is able to conclude that the maximin principle would be
the one that people in the original position would agree to only
because he considers the matter from the perspective of those who
fear they will be among the worst-off, rather than from the perspec-
tive of those who hope to be among the better-off. For this reason he
fails in his attempt to derive the maximin principle in a neutral
manner from what reasonable people would agree upon under
conditions requiring impartiality; and in addition Nozick is able to
make the telling point that the fundamental flaw Rawls finds in
utilitarianism—the failure to rule out “even the tendency to regard
men as means to one another's welfare’—can be found in Rawls’s
own principle. The maximin rule treats the better-off as a means to
the welfare of the worst-off. Indeed one could say (though Nozick
does not) that the tendency to treat people as a means to another's
end is greater under the maximin rule than under utilitarianism,
since a utilitarian would give equal consideration to everyone's
interests, whereas the maximin rule forbids giving any considera-
tion to the interests of the better-off, allotting them goods solely in
so far as doing so assists the worst-off.
There remain many interesting and illuminating points in Rawls’s
long book, but its foundations are now seriously undermined. The
question we must face, then, is whether any conception of distribu-
tive justice that accepts individual rights, particularly the right to
property, and prohibits absolutely treating one man as a means to
the welfare of another can withstand the arguments Nozick has
directed primarily against Rawls. If the answer is negative we shall
have to choose between a conception of justice such as Nozick’s and
our conviction that a society does not have to rely on the charity of
its wealthy members for the relief of its poorest members.
The enthusiasm which greeted Rawls’s theory of justice when it
first appeared may in part be explained by the fact that it was the
first fully worked-out alternative to utilitarianism since W. D. Ross’s
intuitionist thoery lost favor in the 1930s.5 If more careful considera-
tion has found Rawls’s theory wanting, opponents of utilitarianism
lack, once again, a developed alternative theory—except, that is, for
 The Right to be Rich or Poor 49

Nozick’s entitlement view. Nonutilitarians not wishing to accept the

conclusion that coercive redistribution of wealth is a serious viola-
tion of rights urgently require an alternative theory of rights.
What else is there? Not much. While, as Nozick points out, there
is no lack of unsupported presumptions in favor of equality, there is
a surprising dearth of arguments for equality. Nozick discusses one
of the few arguments that have been widely discussed—generally
with approval—by philosophers: that put forward by Bernard Wil-
liams in his article “The Idea of Equality.”®
Williams argues that the proper ground of distribution of medical
care is ill-health; and that, therefore, it is irrational for the distribu-
tion of medical care to be governed by the ability to pay. On first
reading many of us will find this argument for some degree of
equality convincing. But, Nozick asks, why should the internal goal
of an activity take precedence over the particular purpose of the
person performing the activity? By a parallel argument it could be
said that the proper ground of distribution of barbering services is
the need to get one’s hair cut; but if we think a barber need cut the
hair only of those able to pay, why should a doctor not do the same?
What Nozick’s facetious counterexample indicates is that the
plausibility of Williams’s argument lies not in any supposed neces-
sary truth about the proper ground of distribution of medical care
but in the claim that a society should provide for the most important
needs of its members. This is a plausible claim, but it is only a claim
and Williams does not argue for it. So we still do not have an
argument for equality.
Wisely, Nozick remarks that his readers will probably feel that
the case for equality all hangs on some other argument, and says he
would like to see that argument set out in detail. That, unfortu-
nately, is where the attempt to refute Nozick on the ground he has
chosen—accepting a doctrine of individual rights that includes a
right to property—rests at the present time. Which is not to say that
it will rest there long. There is tremendous activity in moral and
political philosophy nowadays and if a response to Nozick’s chal-
lenge is not already in preparation it soon will be.
What if we refuse to accept the ground Nozick has chosen? The
natural alternative is then utilitarianism. There are other pos-
sibilities, but none that seems likely to be very attractive to those
who reject Nozick’s position because of its prohibition on coercive
50 Peter Singer

redistribution of wealth. In A Theory of Justice, for instance, Rawls

considers as alternatives to his own theory only utilitarianism and
what he calls “the Principle of Perfection.” Perfectionism, which is
the theory that we should maximize the achievement of excellence,
was most strongly advocated by Nietzsche and is even further from
concerning itself about the worstoff than Nozick’s theory is; and
when more plausible advocates of a perfectionist position talk about
social justice, they tend to water down their perfectionism with a
dose of something that looks like utilitarianism.7
Utilitarianism has no problem in justifying a substantial amount of
compulsory redistribution from the rich to the poor. We all recog-
nize that $1,000 means far less to people earning $100,000 than it
does to people trying to support a family on $6,000. Therefore in
normal circumstances we increase the total happiness when we take
from those with a lot and give to those with little. Therefore that is
what we ought to do. For the utilitarian it is as simple as that. The
result will not absolute equality of wealth. There may be some who
need relatively little to be happy, and others whose expensive tastes
require more to achieve the same level of happiness. If resources
are adequate the utilitarian will give each enough to make him
happy, and that will mean giving some more than others.
A more serious possibility is the one we discussed in connection
with the maximin principle. If it is necessary to give more to those
with talents useful to society, to encourage them to develop these
talents in a way that will benefit others, then utilitarians would have
to do this. Actually the evidence for this commonly accepted
hypothesis is weak; financial incentives may not be as important as
we think. So some inequality would result from the application of
utilitarian principles to a society like ours, but far less than there is
now, and the inequalities that remain would not (in my view) be
objectionable.
Nozick describes Rawls’s view as an “undeniably great advance
over utilitarianism.” From his standpoint that is a reasonable esti-
mate. Rawls’s theory is a half-way house between utilitarianism and
Nozick’s own position. But if having gone half-way with Rawls we
are forced by the logic of our position to go all the way with Nozick,
it could be that we went wrong when we started out. None of the
arguments Nozick uses against Rawls is decisive when invoked
against a utilitarian position. Utilitarianism gives a clear and plausi-
 The Right to be Rich or Poor OL

ble defense not merely of progressive taxation, welfare payments,

and other methods of redistribution, but also of the general right of
the state to perform useful functions beyond the protection of its
citizens from force and fraud. Utilitarianism also provides an argu-
ment in defense of the claim behind Williams’s argument for
equality—that society should, so far as its resources allow, provide
for the most important needs of its members.
Nor do we have to go all the way with the utilitarians to be in a
position to advocate self-directed redistribution of income. The
problem of whether we can accept a utilitarian account of
noneconomic rights like the right to freedom of speech or freedom
of worship need not be raised here, for Nozick’s argument is mainly
addressed to economic rights. We can deal with property in a
utilitarian manner, rejecting the doctrine of an intrinsic right to
property, without necessarily rejecting the idea that there are some
intrinsic rights against the state. For the remainder of this discus-
sion, though I shall talk simply of “utilitarianism,” it will be this
limited economic utilitarianism to which I am referring.
Nozick, aware that utilitarianism is a more fundamental rival to
his position than other conceptions of justice, tries to get it out of the
way in the first part of the book, when discussing the moral
background of his theory. The discussion is sketchy, however, and
falls below the level of the later sections. Nowhere is utilitarianism
fully and systematically confronted. Nozick mentions some well-
known objections but, with one exception, does not pursue the
replies that utilitarians have made when these objections have been
raised in the past.
The exception is interesting. In opposition to the view, which
utilitarians have held, that the only things that are good or bad in
themselves are states of consciousness, or conscious experiences
(pleasant or happy ones being good, painful or miserable ones bad),
Nozick asks us to imagine that we can build an “experience
machine” which would give us the satisfactions of a wonderful
life—any life we'd like—while we float in a tank with electrodes
plugged into our brains.
If we had such a machine, Nozick says, we would choose not to
use it—and this shows that things other than experience matter to
us. In anticipation of the reply that we would not use the machine
because, as good utilitarians, we would be concerned about other
o2 Peter Singer

people’s (and other animals’) experiences as well as our own, Nozick

makes the further assumption that everyone is able to plug into one
of these machines. This means that we cannot give point to our lives
by improving the experiences of other beings; the experience
machine gives everyone who wants them the best possible experi-
ences anyway. Nevertheless, Nozick says, we would not plug in,
and this ,is because in addition to wanting to have certain experi-
ences we want to do certain things and be a certain sort of person.
We desire to live in contact with reality, and this no machine can do
for us.
Perhaps. It is worth noting, though, that it is difficult to know
what, in a world in which everyone could plug into an experience
machine, there would be left to do, other than plug in; and how it
would be possible to “be a certain sort of person.” How could one
be, for example, a kind or courageous person (Nozick’s examples)?
What could I do to anyone else that showed kindness, if everyone
else could have whatever experiences he wanted without my kind-
ness? When would there ever be any point in being courageous?
Maybe it would seem a pointless world, and plugging into the
machine a pointless kind of existence, but that is because we are
used to having the possibility of improving experiences, our own or
those of another, to give point to our normal existence. Take away
the point of trying to improve the experience of ourselves and others
and perhaps we do take away the only thing beyond our own
experience that gives point to our lives. Maybe life as a whole
doesn’t have any point beyond experience itself. Nozick’s example is
bizarre enough to have a bizarre answer, and the bizarreness of the
answer that the orthodox form of utilitarianism gives is an in-
sufficient reason for rejecting that theory.
Even if we find that Nozick’s example does refute the idea that
states of consciousness are the only things intrinsically good or bad,
however, Nozick has refuted only one form of utilitarianism, Several
recent utilitarian writers, including nearly all those writing in the
field of welfare economics, have taken wants or desires, rather than
states of consciousness, as the starting point for utilitarian calcula-
tions. It is intrinsically good, on this view, if someone gets what he
wants, and bad if he does not. This version of utilitarianism is not
threatened by Nozick’s experience machine: if there are things that
 The Right to be Rich or Poor 53

we want other than experiences, well and good, utilitarianism will

try to make it possible for us to get them.
So the utilitarian alternative to a theory of justice based on
individual rights to property remains open; though other theories
will no doubt be put forward, those wishing to avoid the conclusions
of Nozick’s book may find themselves reconsidering one or another
version of utilitarianism, and questioning whether the right to
property must be taken as seriously as American political thought
has taken it for the last 200 years.

NOTES

1. Some other philosophers who have written on this issue are included
in Animals, Men and Morals, edited by S. and R. Godlovitch and J. Harris
(Taplinger, 1973). See my review in The New York Review, April 5, 1973.
2. The book-length study is The Liberal Theory of Justice by Brian Barry
(Oxford University Press, 1973). Among the more notable critical reviews
have been those by Thomas Nagel in the Philosophical Review (April, 1973)
and the two-part critique by R. M. Hare in Philosophical Quarterly, July
and September, 1973.
3. This point was brought to my notice by Gregory Pence.
4. Kenneth J. Arrow, “Some Ordinalist-Utilitarian Notes on Rawls’
Theory of Justice,” Journal of Philosophy, LXX, 9 (1973), p. 248.
5. See W. D. Ross, The Right and the Good (Oxford University Press,
1930).
6. In Philosophy, Politics and Society (Second Series), Laslett and Run-
ciman, eds. (Barnes and Noble, 1962).
7. For example, Bertrand de Jouvenal, The Ethics of Redistribution
(Cambridge University Press, 1951).
8. See, for example, Jan Narveson’s Morality and Utility (Johns Hopkins
Press, 1967). The upshot of R. M. Hare’s ethical theory is also a
utilitarianism of this type: see Freedom and Reason (Oxford University
Press, 1963) and “Wrongness and Harm” in Essays on the Moral Concepts
(University of California Press, 1972).
 pres rere i
pei fe sa Ht tet ag aaaad
4 a ’ QF “i e - 7 71 i of. tg

aT a > vv
i ‘
r oie tO
_

$ uy a

- 7 ¢ he -
1 : : engl
Ps
= wai ce Ve
. 7 1 el -_ ee ie Pee
=" es 4 . = apes
E iz
- ul : “a ® ‘ oa viel a”

. Je t) ae) ae
Oe ™
——*. | (a ates ;
fe Paw

iT 2
 Il
Contra Anarchism:
Justifying the Minimal
State
 3
Nozick on Anarchism
ROBERT L. HOLMES

Those of us who feel that anarchism is the closest approximation to

the truth in political philosophy find it difficult to decide sometimes
whether that theory has suffered most at the hands of its defenders
or at the hands of critics. Since the latter are at least intentionally
bent upon giving anarchism a bad name, and the former are chiefly
an embarrassment to their allies, perhaps it is the critics who most
need to be reckoned with. Among them, in any event, is Robert
Nozick, a large part of whose Anarchy, State, and Utopia‘ is
devoted to attempting to refute anarchism. It is that critique which I
propose to examine here.

The anarchist’s objection to the state, as Nozick represents it, is that

inasmuch as the state (1) monopolizes force and punishes those who
violate this monopoly and (2) provides protection for everyone
within its territory by requiring some to finance protection for
others, it violates certain moral “side constraints” in the form of
natural rights and, hence, is immoral (p. 51).
Nozick may be excused for formulating the anarchist’s position for
him, since with few exceptions anarchists have not done so them-
selves with the exactitude one might wish. We may also, for present
purposes, overlook the fact that the anarchist’s objection is framed
in terms of Nozick’s own particular conception of morality, in which

57
58 Robert L. Holmes

moral side constraints constitute moral prohibitions, and the pursuit

of ends may legitimately be carried on only in ways which do not
violate those prohibitions. This makes the theory a deontological
one and means that it does not reflect the position of any defender of
anarchism on teleological grounds. Finally, it should be noted that
the objection is also framed against the unexamined background of
Nozick’s particular State-of-Nature theory, according to which indi-
viduals have Lockean natural rights. Inasmuch as those rights
include the right to use violence and punishment against others in
relevant circumstances, the unquestioned assumption of such rights
leaves nonviolent anarchists out of the picture from the outset. So,
bearing in mind that anarchism is represented in Nozick’s critique
by at most one species of anarchism, let us see how it fares.

II

Let us first consider the general outline of the argument, after which
we can turn to the relevant details.
Nozick’s contention is that from a State of Nature (SN) in which
people possess natural rights and generally act morally (making this
a “most favored” situation of anarchy), there will, if some persons do
not act morally and violate the rights of others, evolve first Mutual
Protection Associations (MPAs) in which individuals purchase pro-
tection against others like a market-governed economic good, and
eventually a Dominant protective Association (DPA), an MPA of
preeminent power in a territory. The DPA will, in turn, eventually
evolve into an Ultra Minimal State (UMS), a DPA in which protec-
tion is purchased as an economic good but which, in addition, has a
monopoly of force. The latter, finally, will evolve into a Minimal
State (MS) which, in addition to possessing the two features of the
UMS, will make a “redistribution” (though not really a redistribu-
tion, Nozick contends at p. 114) of protection to certain indepen-
dents (nonclients of the UMS) at cost to the clients. Thus, there are
four stages in the process: from SN to MPAs, from MPAs to a DPA,
from a DPA to a UMS, and finally from a UMS to an MS.
Now two questions are central here: (a) whether it is plausible to
suppose that the above process would in fact develop from an SN,
and (b) whether that fact would be of significance if it were. Taking
the latter first, the answer for Nozick is that it would be significant
 Nozick on Anarchism 59

because it would provide a moral justification of the state. Let us see

why. .
To do this we need to introduce some additional concepts. First,
there is the notion of an Invisible Hand Explanation (IHE), which
for Nozick is, in effect, an explanation in terms of an Invisible Hand
Process (IHP). An IHP, in turn, is a process the outcome of which
appears to have been intended—i.e., to be the product of design—
but in fact was not. A Fundamental Potential Explanation (FPE),
finally, is an explanation which would “explain the whole realm
under consideration were it the actual explanation” (p. 8) and would
do so in terms other than those of the realm. Now Nozick’s
contention is that the state is justified if

(1) it can be given an FPE,

(2) that explanation is (or contains, or makes use of) an IHE, and
(3) no morally impermissible steps are involved in the IHP by reference to
which the IHE_is given.?

And, indeed, he believes that the account of the evolution of the

state (MS) implied by the preceding meets these conditions. To
show that an MS would eventually evolve from an SN without
presupposing that it was the deliberate intention of persons to
produce a DPA, a UMS, or an MS, and without presupposing any
morally illegitimate steps, is precisely to give an IHE of the State of
the required sort. Thus, in answer to question b, an MS can be
shown to be morally justified if it can be shown that this process
would occur under the above conditions.

Ill

This leaves us with question a, whether it is plausible to suppose

that the process by which an MS would evolve from an SN would in
fact take place. I want to focus principally upon the transition from a
DPA to a UMS which is effected by the acquisition of amonopoly of
force by the DPA. Nozick’s burden is two fold here. He must show
not only that it is plausible to suppose that the monopoly of force
would in fact arise within the DPA, but also that it would be morally
legitimate that it do so. The anarchist’s objection is not that a DPA
would acquire a monopoly of force (indeed, most anarchists would
60 Robert L. Holmes

agree that it would), but that its doing so would violate the rights of
some individuals.
Here the argument breaks down into two parts. One proceeds
with the “facilitating assumption” of so-called procedural rights, the
other without it. Let us take these in turn.
Though the notion of a procedural right is not explained by
Nozick, the following presumably constitutes such a right: the
everyone has a right to resist, in self-defense, “if others try to apply
to him an unreliable or unfair procedure of justice” (p. 102; or, as he
sometimes adds, a procedure of unknown reliability of fairness).*
Since anyone can transfer this right to the DPA (presumably by
purchasing the appropriate protection policy), the DPA can acquire
the right to intervene and prohibit independents from privately
enforcing justice against its clients. And if people do in fact so
empower the DPA, then the DPA has this right, and so acts
legitimately in monopolizing force, and so, in the process, comes to
constitute a UMS.4
We can now see the ground of the final transition to the MS. For
Nozick points out that the independents in a given territory will
obviously be disadvantaged by being prohibited from defending
themselves or enforcing their right to punish wrongdoing. This fact
calls into play a (presumably moral) principle which requires that
when one is disadvantaged by being prohibited from performing a
risky activity, as the independent’s “procedure of justice” sup-
posedly is by virtue of being unreliable, he must be compensated for
that disadvantage. Through reasoning which we need not detail
here, and with some qualifications concerning ability to pay which
we may ignore, Nozick concludes that it is morally required that
independents be compensated for this disadvantage, either in
money or in kind. Whereas self-interest has been the chief factor in
the IHP up to this point, this final trasition to the MS thus makes
use of amoral motivation. It is only ifthe UMS is motivated by the
desire to compensate independents for the disadvantage its
monopoly of force imposes upon them that it will transform itself
into an MS.
The problem in all of this centers about the fact that although the
process leading to the emergence of the MS must take place without
the presence of any morally impermissible steps, the argument for
the final stage in the process presupposes a morally impermissible
 Nozick on Anarchism 61

step. The disadvantage to independents resulting from the

monopoly of force is morally wrong. It is for this reason that the
independents deserve compensation. And that the monopoly is
exercised without their being compensated is the distinguishing
feature of the UMS, which means that there is an essential stage in
the overall process which is without moral justification. This gives
rise, therefore, to the following dilemma: either the UMS is legiti-
mate or it is not. If it is, then there is no reason to suppose that
people will move beyond it to the MS, in which case the overall
process does not result in a state. If it is not, then the whole process
contains an obviously impermissible stage, in which case the ac-
count fails to meet Nozick’s own criterion for the legitimacy of a
state as well as that part of the anarchist’s objection which holds that
the monopoly of force by the state is immoral. In short, either it is
wrong to prohibit independents from trying to enforce their rights
and to punish them for doing so, in which case the UMS acts
immorally in doing these things, or it is not wrong, in which case
there is no reason to transcend the UMS.
But let us leave this aside and return to the second part of the
argument to legitimize the monopoly of force leading to the UMS,
that which proceeds without the assumption of procedural rights, an
assumption which Nozick concedes might seem to make his argu-
ment for the monopolization of force “too easy” (p. 103). The
reasoning here is almost impenetrably obscure until the final stages,
but, as it is the latter with which we shall be concerned, I shall
merely sketch the rest.
First, the chief problem confronting the argument: if the client is
actually guilty of violating the rights of an independent, then not
only does the independent have the right which everyone possesses
to punish the wrongdoing, but the client himself has no right (of the
procedural sort in the first part of the argument) which he can
transfer to the state to warrant its intervening on his behalf. The
burden of Nozick’s argument, then, is to justify the DPA’s pro-
hibiting (before the fact) of presumably innocent victims of wrong-
doing at the hands of its clients from exercising their rights of
self-defense and proportionate punishment; and to justify its
punishment of them (after the fact) for having exercised that right
against a client.> (Nozick, it should be noted, shifts back and forth
between speaking of punishment and speaking of porhibition. It is
62 Robert L. Holmes

prohibition which he needs to justify for purposes of the final stage

of his argument which makes use of the notion of compensation. If
one were to suppose that, instead of prohibiting private enforce-
ment of justice, the DPA merely punishes independents for such
enforcement, the notion of compensation would become pointless.
If one is going to compensate someone for having punished him, it
would be simpler not to punish him in the first place. And if
compensation is morally required, it is difficult to see how the
punishment for which it is offered could be morally justified.)
Now Nozick contends that no one has a right to use an unreliable
procedure (or one of unproven reliability) to decide whether to
punish another, and hence no one has a right to punish another as
the result of a determination made by such a procedure. But this
appears to contradict the ascription to everyone of the right to
punish the guilty. For if the client is guilty, then the victim has the
right to punish him. So, we might ask, what difference does it make
whether the procedure by which the victim determines the latter’s
guilt is one of proven reliability? He is doing what he has a right to
do in any case and, hence, one would have thought, cannot justifi-
ably be prohibited by anyone—least of all, the anarchist will add, by
the protostate.
Nozick attempts to deal with this objection by saying that it
matters not whether we say (a) that a person does not have a right to
do x in the absence of appropriate knowledge of wrongdoing on the
part of the person to be affected, or (b) that he has the right but does
wrong to exercise it in the absence of such knowledge (p. 106). (It is
strange that he says this, since given the background theory of
natural rights with which he is operating, alternative a would imply
that people do not have the various “natural rights” antecedently to
coming to possess the appropriate knowledge; a condition which,
given even moderately stringent conditions of knowledge, would
suggest that we have few if any rights in the SN.) This represents a
“terminological fork,” and Nozick opts for alternative b which opens
the way to the formidable sounding Epistemic Principle of Border
Crossing (EPBC);:

If someone knows that doing act A would violate Q’s rights unless condition
C obtained, he may not do A if he has not ascertained that C obtains
through being in the best feasible position for ascertaining this. [106f.]
 Nozick on Anarchism 63

Now Nozick contends that anyone (presumably including the DPA)

may punish violators of this principle, so long as he himself does not
violate the principle in the process. Since our victimized indepen-
dent is, by hypothesis, about to enforce his rights by means of a
procedure of unproven reliability, this means that condition C in the
EPBC (the determination of the client’s guilt by means of a proce-
dure of proven reliability and fairness) has not been met, and hence
that he is in violation of the principle. The DPA is, therefore, fully
justified in prohibiting his action or punishing him after the fact of
its performance (so long, of course, as it does not violate the
principle itself)—even though its client is guilty and the victim has a
right to punish him.
But recall that this stage of the argument to justify the monopoly of
force is supposed to proceed without the assumption of procedural
rights (specifically, one presumes, without the right enunciated in
the key premise of the first stage of the argument, to the effect that
everyone has a right to resist in self-defense if anyone attempts to
use an unfair or unreliable procedure of justice against him). And
notice that the claim that anyone has the right to punish violators of
the EPBC implies that anyone has the right to punish a person who
attempts to punish someone without first having ascertained their
guilt by a procedure of justice of proven reliability. If everyone has
the right to punish violations of this principle, then anyone whose
rights are violated by it has that right. And if one has the right to
punish violators of the principle against himself, it seems unprob-
lematic (by whatever reasoning Nozick himself uses to shift from the
right to prohibit to the right to punish in his own argument) that he
has the right to resist before the fact of having suffered the violation.
But this is precisely the right to resist the use against oneself of an
unreliable procedure of justice, which means that the very “pro-
cedural right” which was alleged to make the first part of the
argument seem too easy has reappeared in the second part.
If the DPA can punish violators of the EPBC only on condition
that it not itself violate the principle, this implies, moreover, that its
own procedures are of proven reliability and fairness. Yet there is no
independent criterion by which to establish this to be found in
Nozick. He simply reiterates that the DPA will apply its own criteria
according to its lights and will be the one entity in a position to
enforce their acceptance (p. 118). If reliability and fairness are
64 Robert L. Holmes

defined in terms of what the DPA decides, this of course is

question-begging and yields a theory of “might makes right.” If they
are not, some independent criterion must be forthcoming before
one can judge whether the DPA itself is in violation of the EPBC.
For, just as Nozick can hypothesize that independents are likely to
employ unreliable procedures. States are made up of individual
persons and derive their rights and conception of rights from them,
so that whatever measure of unreliability is thought to attach to the
procedures of independents—and presumably individuals
generally—is likely to carry over to the state itself. The very
reasoning by which the state judges independents’ procedures to be
unreliable should therefore lead it to question its own procedures. If
so, the DPA acts wrongly in prohibiting independents from self-
help enforcement, in which case that particular feature of the
monopolization of force cannot be justified. One might counter this,
of course, by arguing that if the state uses only fair and reliable
procedures, then it is justified in its actions against independents.
But an anarchist can agree with this. His objection is not to the truth
of that conditional, but the claim that the antecedent is, or is ever
likely to be, fulfilled. By the same token one can way that if the
independents’ procedures are fair and reliable, then there is no
need of the state. And Nozick can agree with this. The relevant
question concerns the likelihood of the procedures being fair and
reliable in the two cases, and one side cannot refute the other by
assuming simpliciter that they will be fair in the one and not in the
other.
Let us suppose, however, that these problems can be surmounted
and concern ourselves finally with the key assumption in Nozick’s
argument. That assumption, implied by option b in the terminologi-
cal fork, is that we can distinguish between the having of a right and
the having of a right to exercise that right. It is this which enables
Nozick to avoid the contradiction cited earlier. For if all that the
DPA does is to prevent the victimized independent from exercising
his right to private enforcement of justice, and does not in the
process violate his right to private enforcement, there is no con-
tradiction. The independent, then, still has the right like everyone
else to defend himself and to punish those who violate his rights.
What he does not have is the additional right to exercise the first
 Nozick on Anarchism 65

right. Therefore, when the state prohibits him from exercising that
right, it does not violate any of his rights, and hence its monopoly of
force involves no morally impermissible steps.
Now this is a strange notion—that before I can do whatever it is
that I have a right to do, I must first have a right to exercise that
right. This would suggest that before I can do what the second right
authorizes, I must first have a third right, namely, the right to
exercise the second right, and so an ad infinitum; either that, or that
there is an ad hoc criterion by which to distinguish rights whose
effective implementation presupposes a second-order right from
those which do not. But can one even make sense of requiring as a
condition of being justified in doing what one has a right to do that
one have a further right to exercise that right? I think not. A right do
x is, if one may be pardoned, a right to do x. To do x is to exercise
the right. The exercise of the right is not another act over and above
that which, in the possession of the first right, one has a right to do.
It may be that what Nozick has in mind here is that one might
actually have a right to do x, but have insufficient evidence that he
has this right to be warranted in his belief that he has it. His
reference to “subjective” and “objective” rightness would suggest
this (p. 107). That this is possible is a fair and uncontroversial point.
But if there are right and wrong actions determined by natural
rights, and the latter exist independently of social conventions and
of their being thought to exist by individuals—as Nozick seems
committed to holding by virtue of his conception of the SN—then if
a person has a right to do x, as by hypothesis the victimized
independent does, it follows that it is actually right for him to do x,
whatever his grounds for thinking so. And if to have a right do x
means that it would be wrong for anyone to prohibit you from doing
x, then it is in fact wrong for the DPA to prohibit you from doing x.
If the DPA does not know this, then it is the DPA, and not the
hapless victim of its client who is about to be victimized a second
time, which acts wrongly. Even for the DPA correctly to perceive
that the independent has not proven that he is not violating its
client’s rights in punishing him does not alter the fact that it violates
his rights when it intervenes. The independent acts rightly in doing
what he believes he has a right to do, even if the ground for his
belief is inadequate, whereas the DPA does not.
66 Robert L. Holmes

If this is correct, then the argument to show how the state would
emerge from an SN fails, and the anarchist will regard this latest
effort to justify the state as simply revealing the lengths to which one
must go—whether under libertarian or totalitarian banner—to try to
justify the violation of individual rights that is inherently associated
with the state. If one holds that rights are only a part but not the
whole of morality, then one can argue that sometimes it is right ot
even obligatory to violate individual rights. One can do so on the
grounds that it would minimize the overall violation of rights or
promote a greater good. But these are options which Nozick ex-
pressly rejects. And one cannot have it both ways. If individual
rights are inviolable, and if individuals, as Nozick insists, may not be
sacrificed for the good of others, then they may not be sacrificed for
the good of, or in the interest of, the state, and even less for the sake
of enabling the state to exercise a monopoly of force.

NOTES
1. (New York: Basic Books, Inc., 1974) Page references will be included
in the text. I have benefited in what follows from suggestions by Peter
Hamblin and Ted Lockhart.
2. This is a reconstruction of Nozick’s argument. Although it diverges at a
few points from what he actually says, it is, I believe, the most coherent
rendering of his overall criticism.
3. By “procedure of justice” Nozick sometimes means a procedure by
which to enforce one’s rights, sometimes a procedure by which to deter-
mine guilt or innocence, and sometimes a procedure by which to punish.
Inasmuch as the first of these seems to be the appropriate sense in the
context of this portion of his discussion, and inasmuch as, in that context, it
could be taken to imply the other two, we shall take all of these to be bound
up in the notion ofa “procedure” as he conceives of it, with different aspects
being emphasized at different times.
4. This right to monopolize force, it should be noted, is considered by
Nozick to constitute only a de facto, not a de jure, monopoly (p. 108f). By
this he means only that the right which the DPA comes to possess is not a
unique right of which it is the sole possessor. Everyone has this right.
5. There is an overlay of qualifications here which include the following:
(a) that the client on whose behalf the DPA is intervening has purchased the
appropriate policy to protect him in cases of this sort; (b) that the indepen-
 Nozick on Anarchism 67

dent’s “procedure” of enforcing justice is either unreliable or unfair or of

unknown reliability or fairness (presumably by comparison with other
practices of enforcing justice); (c) that prior to the infliction of punishment
by the independent the DPA may, if it deems his procedure (in this case for
determining guilt or innocence) unrealiable, itself investigate and presuma-
bly prohibit punishment only if it determines that its client is innocent; (d)
that the independent’s action is a “risky” one which would be feared by
others if expected. It is difficult to be certain that these exhaust the
qualifications Nozick intends throughout chapter 5, but they do not alter
the basic features of the situation highlighted above.
 4
The Withering of Nozick's
Minimal State
JEFFREY PAUL

In Part I of his Anarchy, State, and Utopia Robert Nozick proposes

a new solution to the old problem of political legitimacy. The
problem of political obligation within a natural rights framework
received its classic treatment in John Locke's Second Treatise of
Civil Government. There, Locke adduced two necessary and,
cumulatively, sufficient conditions of political obligation: first, that
civil society be created by the unanimous agreement of the gov-
erned and second, that the public instrument so created be granted
only the substantive power of protecting natural rights. As a natural
rights libertarian Nozick finds the social contract an insufficient basis
for establishing political legitimacy. In the first place, the state
created by it monopolizes protective services and, thereby, restricts
the property rights of those succeeding generations who were not
parties to the contract. In the second place, it transfers to the
majority the power to tax the citizenry, binding subsequent genera-
tions to that power, again, an apparent contravention of Lockean
property rights. These weaknesses lead Nozick to seek a new
instrument of political legitimacy in an “invisible hand” process
which can transform the state of nature into civil society without
depriving the governed of their Lockean natural rights.
In what follows I will contend that this attempt to replace the

68
 The Withering of the Minimal State 69

social contract with an invisible hand mechanism is a failure, so that

Nozick’s minimal state never receives the justification that would
politically obligate its citizenry. In attempting to construct that
justification Nozick tries to be responsive to the libertarian critique
that the contractarian created minimal state is rights violating in the
aforementioned ways. This critique would, if sustained, imply that
any state, no matter how limited in authority, is not justified. It is
the libertarian anarchist, then, that Nozick feels compelled to
answer in justifying his new instrumentality of political obligation.
The libertarian anarchist, according to Nozick, opposes on two
counts even the minimal state which confines its activities to the
protection of libertarian rights. First, the state, says Nozick’s
anarchist opposition, is a coercive monopoly in the provision of
certain services (legislation, adjudication, and punishment) and so,
arbitrarily and forcibly prevents the free use of personal property to
hire or finance such services. This type of restriction upon the use of
property is an obvious violation of Nozick’s principle of justice in
transfer and so, is apparently anti-libertarian in nature. The state’s
coercive monopoly is, for this reason, rejected by Nozick’s liberta-
rian anarchist.
But the anarchist has another reason for condemning the state
according to Nozick. Its geographical monopoly on legal violence
implies that it assumes responsibility for the protection of all those
within its domain whom it has forcibly deprived of the means of
self-protection, including those who will not voluntarily remunerate
it for such protective services. This obligation seemingly requires
the redistribution of property from those who are willing (and able?)
to pay the state for its services to those who are neither willing nor
able to make such payments. Hence, the state, of necessity, appears
to violate libertarian principle by being redistributive.
Nozick’s method of rebutting these anarchist complaints includes
an explication of the steps by which a state of nature becomes
transformed into a minimally governed society and an alleged
demonstration that each of these steps is morally permissible, i.e., is
rights preserving. In order to assure that the transition from anarchy
to minarchy violates no one’s Lockean rights, Nozick undertakes to
show that the state’s monopolistic and redistributive functions were
acquired precisely in order to defend rights. I will contend that
Nozick’s arguments in this regard fail.
 70 Jeffrey Paul

Nozick’s narrative of transition begins with a state of nature in

which a number of mutual protection associations have arisen, each
engaged in the defense against and punishment of violations of
libertarian rights. How, Nozick inquires, does one of these become
a government without transgressing libertarian principle? That is,
on what grounds can one agency violently restrict the rights pro-
tecting activities of other agencies without infringing upon the
property rights of those agencies and their clients? On the grounds,
answers Nozick, that the means being employed by those other
agencies are inappropriate to the end of rights protection; that they,
in fact, systematically risk rights violation. And their very riskiness
provides the justification for their forcible prevention, according to
Nozick.
Now, how can this be? Risky behavior is certainly not the moral
equivalent of rights violating behavior. And on the Nozickian view it
is only examples of the latter that may be coercively prohibited.
How can agencies whose behavior is risky but not rights violating
have their operations forcibly, but justifiably, curtailed?
Nozick’s view, here, seems to be that a procedure of rights
protection which risks a rights violation can be forcibly prevented,
provided that the individual whose rights were being defended via
the risky procedure is suitably compensated. But, how does the
compensation “erase” what ought to be considered on Nozick’s own
_ principles a straightforward violation of rights? Either the risky
activity of the agency constitutes a violation of rights, or it does not.
If it does, then its prohibition is merely a legitimate defense of rights
and no compensation ought to be due the perpetrator of the
violation who would be a criminal on Nozickian grounds. If the risky
activity was no violation of rights, then, its forcible prevention is a
wrongful (rights violating) act which deserves at least punishment
and, perhaps, compensation. But, to say that an act of “risk preven-
tion” is not wrongful and, yet, requires the payment of compensa-
tion to its victim would appear to be, within Nozick’s libertarian
context, a contradiction. For, either a principle of entitlement has
been contravened or not. If so, compensation is due, if not, no
restitution is indicated. Nozick’s alleged third category of threats to
entitlement would seem to be vacuous from an entitlement perspec-
tive.
Further, if a particular risky activity is neither actually rights
 The Withering of the Minimal State 71

violating nor intended to be rights violating, then someone who

finds the activity threatening is free to negotiate its cessation with its
perpetrator, or to threaten him with prosecution should the activity
result in injury. More than this he cannot do on entitlement
grounds.!
Nozick has objected to this kind of argument because it is
sometimes too difficult or costly to negotiate with the perpetrator.
But, so what? Difficulties and costs are not, according to Nozick’s
historical principles of justice, entitlements.
In order to buttress his case for an alleged third category of
activity, neither criminal nor tortious, but requiring prohibition
nevertheless, Nozick unfurls the notion of a procedural right. If the
requisite degree of riskiness can be identified with a failure to abide
by certain procedural proprieties and, if those proprieties can be
elevated to the status of a right, then occasions for prohibition
become, at once, objectively evident and morally justifiable. They
are objectively evident because what Nozick calls “relatively reli-
able procedures” are codified by various protective agencies and,
therefore, the question of whether they have been adhered to can
be independently assessed. Morally, an individual “. . . may resist,
in self-defense, if others try to apply to him an unreliable or unfair
procedure of justice. In applying this principle, an individual will
resist those systems which after all conscientious consideration he
finds to be unfair or unreliable. An individual may empower his
protective agency to exercise for him his rights [emphasis my own]
to resist the imposition of any procedure which has not made its
reliability and fairness known, and to resist any procedure that is
unfair or unreliable.”? If procedural reliability has attained the
status of a moral right, then it may be robustly defended.
Although everyone possesses this right, according to Nozick, and,
therefore, “Everyone may defend himself against unknown or unre-
liable procedures and may punish those who use or attempt to use
such procedures against him’4 only the dominant protective agency
will be able to enforce its clients’ procedural rights:

. its strength leads it to be the unique agent acting across the board to
enforce a particular right. It is not merely that it happens to be only the
exerciser of a right it grants that all possess: the nature of the right is such
that once a dominant power emerges, it alone will actually exercise that
te Jeffrey Paul

right. For the right includes the right to stop others from wrongfully
exercising the right, and only the dominant power will be able to exercise
this right against all others. Here, if anywhere, is the place for applying
some notion of a de facto monopoly: a monopoly that is not de jure because
it is not the result of some unique grant of exclusive right while others are
excluded from exercising a similar privilege. Other protective agencies, to
be sure, can enter the market and attempt to wean customers away from the
dominant protective agency. They can attempt to replace it as the dominant
one. But being the already dominant protective agency gives an agency a
significant market advantage in the competition for clients. The dominant
agency can offer its customers a guarantee that no other agencies can match:
“Only those procedures we deem appropriate will be used on our custom-
ers.°

Therefore, by morally permissible means we have, according to

Nozick, the morally permissible state. Or do we?
The argument appears markedly flawed at several stages. First,
there is the notion of procedural rights itself. This is the notion of a
right to be judged by a certain type of procedure which implies the
rightful authority to punish the wielder of unreliable procedures.
But, whence comes this right? Clearly, it is not an entitlement
principle. It does not specify how certain “holdings” come to be
justifiably acquired. Nor do substantive procedures seem deducible
from the manifestly formal Lockean rights. How could specific sets
of procedures be implied by entitlement principles? Such proce-
dures are obviously strategies adopted by legislators and jurists, not
entailments of the natural rights. Typically, a procedural right is a
convention defined within and granted by a particular legal code. It
is not an abstract, pre-legal natural right, but a right created by and
recognized within a specific body of law. Its authority does not
extend beyond that particular legal code. To speak of procedural
rights which transcend and may be defined apart from any particular
code is at least odd, if not obscurantist. Yet, Nozick analogizes his
conception of procedural rights to Locke’s extra-legal notion of
natural rights, thereby attempting to impart to it the political force
of a Nozickian entitlement.
Nozick’s commitment to this construal of procedural rights would
seem to involve him in a regression problem. For, if everyone
accused of violating Lockean rights has a right to be judged accord-
ing to reliable procedures, then, suppose one’s procedural rights are
 The Withering of the Minimal State fe

themselves transgressed. Must not this alleged transgression be

evaluated according to a set of relatively reliable second-order
procedures? And those second-order procedures will require a
third-order set so that putative invasions of their second-order
predecessors might be adjudicated reliably. And so on. Each rights
violation is potentially encumbered with an infinite set of decision
procedures. And no agency will ever be able to dominate its
competitors for it will never be able to complete its deliberations
which may require an infinite length of time should there be
continuous allegations of procedural impropriety.
At this point Nozick might wish to respond that the dominant
agency will simply terminate its deliberations at any point in time
that it becomes convinced of the reliability of its procedures. But
this move would fail on two counts. First, a putative property rights
violation can only be established through a reliable procedure.
Similarly, an alleged procedural rights violation can only be prop-
erly identified by another reliable procedure. As long as procedural
improprieties are alleged by someone to have occurred, that allega-
tion can only be evaluated by a review which employs some further
set of procedures. Once the spectre of procedural impropriety has
been raised the issue can be resolved either by further procedural
review or through the exercise of force majeure by the dominant
protective agency in order to arbitrarily terminate the reviewing
process. In the former case, the procedural review is a potentially
infinite one and, consequently, one which will not culminate in the
emergence of an ultraminimal state. In the latter situation, the
dominant protection agency after failing to establish its claims of
procedural propriety through further procedural review may well
transform itself into the minimal state, but it will have done so by
trampling the rights of others and, so, contra Nozick, the state will
have emerged through the use of immoral means. In either case, a
moral minimal state will not have arisen and Nozick’s argument for
its evolution, therefore, fails.
In fact, even if Nozick’s commitment to procedural rights did not
imply an infinite set of such rights, the transition from dominant
protection agency to minimal state might still not be morally
effected. For suppose X is the dominant protection agency but
applies suboptimal procedures to the clients of others while believ-
ing its procedures to be more reliable than those its competitors. Y,
74 Jeffrey Paul

on the other hand, is an independent whose procedures are, in fact,

more reliable than X’s. X’s erroneous beliefs concerning its own
procedures lead it to inhibit the use of the more reliable procedures
of Y on its clients. X, thereby, becomes a minimal state, but by
morally impermissible moves.
Furthermore, I would contend that Nozick’s attachment to pro-
cedural rights implies a commitment to what I would call “stale-
mated anarchism.” Nozick concedes that “Everyone may defend
himself against unknown or unreliable procedures and may punish
those who use or attempt to use such procedures against him.”®
This, of course, includes the clients of the dominant protection
agency. Yet, in the process of prohibiting procedures which it
believes to be defective the dominant agency may deprive others of
their rights to reliable procedures as they understand them. The
problem, here, is that there is no specific set of procedures common
to all protection agencies that would enable public or independent
verification of alleged procedural rights violations. Procedural rights
in the state of nature refer to no body of rules other than those
devised by anyone who cares to think them up. With many persons
believing their procedures to be best and with no independent
means of assessing the comparative merits of those beliefs, why is
anyone entitled to impose his own views to the exclusion of others?
Of course, Nozick has demonstrated that the dominant agency has
the strength to do so, but does it have the right to do so? If everyone
has the right to impose procedures which are in fact ideal, but, if,
because of the emptiness of the Nozickian concept of procedural
rights they are never able to identify those procedures with cer-
titude, then everyone is entitled to impose what they believe to be
procedurally best. But, this means that the dominant agency may
not act in contravention of the beliefs of others. Hence, it morally
may not act at all (although, of course, it is fully able to act) when its
views of procedural propriety are in dispute. And so long as there is
a plurality of views about what constitutes procedural justice, it is
morally impermissible to move from anarchy to statism. And so we
are left in the limbo of “stalemated anarchy.”
Recently, Nozick appears to have modified his views on pro-
cedural rights somewhat.? His views, now, appear to be that one
may prohibit behavior which risks a rights violation if one compen-
sates the potential violator for any disadvantage incurred by him.
 The Withering of the Minimal State 75

This argument does not utilize the procedural rights notion. But,
the problem, here, is that no right has been actually violated. On
Nozick’s own entitlement criteria, the use of coercion in order to
prohibit a risky but non-rights violating action would be inappropri-
ate.
One might negotiate with the risk taker, attempting to induce
him to desist from his risky activity. Nozick rejects such prior
negotiations as not conducive to what he calls “productive ex-
changes.”® Such exchanges are defined in terms of the benefits
accruing to the parties as a result of the exchange, as well as the
motivations of the parties, and not in terms of coercion and owner-
ship. If X and Y both justly acquired what they exchanged and did so
voluntarily, then the propriety of the exchange is, on entitlement
grounds, established whether or not that exchange is “productive.”
Hence, a negotiation motivated by the desire to avert risky but
non-coercive action is a perfectly plausible and morally superior
alternative to the forcible prohibition of risky activity. And such
prohibitions, if not clearly defensive in nature are violations of
rights. If such prohibitions are in fact defensive then no compensat-
ory payment is due the perpetrator of the offensive and invasive act.
But, if the dominant agency does not have to compensate those
whom it justifiably defends its clients against, no basis exists for
“redistributing” its services to the clients of independent agencies.
As a monopolization of such services is a sine qua non of the minimal
state, no such entity can possibly arise. For without procedural
rights, agencies can merely defend their clients and punish crimi-
nals. They cannot forcibly restrict the enforcement power of others,
nor are they obliged to replace the prohibited enforcement services
of others with their own as a form of compensation.
And so, we must conclude that with or without the conception of
procedural rights Nozick’s minimal state cannot arise by morally
permissible means. The “invisible hand” mechanism, designed by
Nozick to replace Locke’s social contract fails as a justificatory
lubricant for political authority: hence, the withering of Nozick’s
minimal state.
76 Jeffrey Paul

NOTES

il. Nozick considers this sort of criticism and dismisses it, cryptically, as
“too short.” If he is referring to the parsimony of this kind of argument then
his dismissal is logically groundless. Brevity is neither a formal nor informal
fallacy of reasoning. See Robert Nozick, Anarchy, State, and Utopia (New
York: Basic Books, Inc., 1974), p. 83.
2. Nozick, pp. 71-73.
. Nozick, p. 102.
Nozick, p. 108.
. Nozick, p. 109.
. Nozick, p. 108.
w . Remarks made by Robert Nozick at the Liberty Fund Conference on
TMP
-~URD
Tiberty: Ethics, and the Economy held by the Center for Study of Public
Choice, Virginia Polytechnic Institute and State University, Blacksburg,
Virginia, July 11-22, 1977.
8. Nozick, pp. 84-88.
 o
Robert Nozick’s Derivation
of
the Minimal State
ROBERT PAUL WOLFF

In Part I of Anarchy, State, and Utopia,! Robert Nozick undertakes

to demonstrate, on the basis of what would ordinarily be considered
libertarian anarchist moral and metaphysical assumptions, that a de
jure legitimate state could come into existence by a sequence of
steps, no one of which violated any person’s rights; that such a state
would satisfy a plausible definition of the state of the sort Max
Weber enunciated; that it could function as a state without violating
anyone's rights; and that such a state would be a genuine minimal,
or nightwatchman, state. In Part II, Nozick goes on to argue that a
state so conceived could be no more than a minimal state without
violating someone’s rights.
In this Article, I propose to subject the argument of Part I of
Anarchy, State, and Utopia to examination and criticism. After a
brief summary of Nozick’s argument, intended to bring into view
the elements of it which are especially important for my analysis, I
shall develop my critique in three stages, beginning with purely
internal considerations of the consistency of Nozick’s argument,
given his premises, and proceeding to more and more “external”

77
78 Robert Paul Wolff

considerations. My conclusions will be that Nozick’s argument is

internally unsuccessful; that a number of the background assump-
tions of his argument are wrong, in ways which vitiate his theory;
that his entire mode, or style, of doing political philosophy is
inappropriate to its subject matter; and finally, that the peculiar
tone of Anarchy, State, and Utopia serves as a clue to what is awry
with it philosophically, as a piece of political theory.

NOZICK SARGUMENT
Nozick begins with a group of individuals in a Lockean state of
nature.2 He simply assumes that there is a clear, objective, ration-
ally knowable moral law which determines the absolute and inviola-
ble rights possessed by those individuals and the duties each owes to
others. The individuals, on the whole, are not so righteous as to
ensure that they will always act as the moral law commands, but
they are sufficiently righteous so that rights-violations, while a
genuine social problem are nonetheless a marginal rather than a
central fact of life in the state of nature. The individuals have
conflicting interests, but they can benefit from far-reaching, system-
atic exchange, interaction, contract, and cooperation. What is more,
it makes coherent sense to speak of them as individuals, in abstrac-
tion from or independently of their social origins and inheritance.
The moral law, as Nozick invokes it, has rather blurry outlines,
although the author appears to have a penetratingly clear intuition
of it. However, certain of its key provisions emerge in his discus-
sion. Rights are inviolable; hence they function as absolute, not
merely as prima facie, constraints on the actions of others. Political
philosophy “is concerned only with certain ways that persons may
not use others; primarily, physically aggressing against them.”
Oddly, but not surprisingly, Nozick construes the attachment of
one’s property as an act of physical aggression, and hence as fit
subject matter for political philosophy. Each individual has the right
to punish others for their aggressions against him, although he does
not have the right to punish them unless they have aggressed
against him, nor may he punish them inappropriately.4 Most
important of all, any person, A, has a right to punish any other
person, B, for B’s violation of the rights of a third person, C. This
claim is merely asserted by Nozick without proof, but it is the
 The Derivation of the Minimal State 79

foundation stone of the entire edifice (however minimal) of the

legitimate state.
With this set of assumptions, Nozick proceeds to develop his
argument fairly rapidly in four steps:
1. Individuals in a state of nature have a right to band together,
through contractual agreement, for purposes of mutual protection.®
2. Those individuals have the right, collectively, to assign to
employees or agents such rights of self protection, punishment, and
so forth as they possess individually and have pooled contractually.®
3. Market forces, strategy calculations, and the like may lead to
the emergence of a dominant protective association in a territory.
Such an association will possess a de facto monopoly of physical
force, which it has acquired by a series of totally permissible acts.”
4. The monopoly protective association, or “ultra-minimal state,”
will have an obligation to compensate nonclients, if there are any,
for the disadvantage they suffer in their dealings with clients backed
by so powerful a protective association. Hence it will have a right,
indeed, it will have a duty, to “tax” its clients for the money to buy
some sort of protection for the disadvantaged non-clients. This
apparent “re-distribution” constitutes it a nightwatchman state, in
the traditional sense.®
In Part II, Nozick elaborates a neo-Lockean theory of property,
the “entitlement” theory, for the purpose of denying any further
claims that may be made against the nightwatchman state—claims of
the sort that usually go under the banner of “social justice.” Since
the rights dealt with in Part I are all property rights, given Locke’s
view that each of us owns, or has sole property in, his own body; and
inasmuch as the argument of Part I is couched entirely in terms of
boundary crossings, disadvantages, compensations, and the like; it
is clear that the theory of property is really presupposed from the
beginning, or at least it is presupposed that this theory must be
developed as a part of the derivation of the minimal state.
Nevertheless, I shall follow Nozick’s lead, and ignore the entitle-
ment theory in this analysis of the arguments for a de jure legitimate
state.
Before we subject Nozick’s argument to analysis and criticism,
there are several questions of a general sort that might be worth
raising about the logical status of that argument, and its precise
80 Robert Paul Wolff

purpose. One might imagine, from the way Nozick talks, that the
argument is intended as a straightforward deduction, or derivation,
from a set of assumed premises taken over from moral philosophy.
However, so much of the argument depends, at crucial points, on
specific interpretations and elaborations of that moral theory, with
little or no proof of the interpretations adduced, that after a while it
seems that Nozick is providing us with nothing more than a recon-
struction or systematization of a set of moral intuitions. Roughly
speaking, we might say that his argument is a rational reconstruction
of a libertarian moral consciousness. If I disagree with one of
Nozick’s claims about morality, for example with regard to when,
where, to whom, and to what extent I am obligated to pay compen-
sation, what sorts of arguments would he consider it relevant for me
to offer? I confess that I cannot tell.
Nozick talks repeatedly of developing a “theory” of this or a
“theory” of that. Does he mean a rational reconstruction of our
moral intuitions? Whose intuitions? Does he mean, rather, a deriva-
tion of normative principles from a set of premises? In this regard,
his opening methodological remarks about types of explanation are
seriously misleading. Nozick’s task is to show that under certain
circumstances, a state of a certain sort can be justified, not that it, or
its appearance, can be explained. For this purpose, fact-, law-, and
process-defects are irrelevant.
Finally, it should be noted that despite the contrary impression
created by some of his language, Nozick is attempting to show that a
de jure legitimate minimal state could come into existence by a
series of morally permissible steps, not that it would come into
existence under any particular set of social circumstances. In short,
the purpose of Part I of Anarchy, State, and Utopia is to establish
the possibility of a de jure legitimate state. Let us turn now to an
examination of Nozick’s argument for that claim.

AN INTERNAL CRITIQUE OF THE ARGUMENT

The first difficulty we encounter when we examine Part I is that

Nozick, by his own admission, has not proved what he set out to
prove, even if his argument is sound. The “state-like entity” whose
generation by morally permissible or morally obligatory steps
Nozick sketches is not, in his own words, the “sole authorizer of
 The Derivation of the Minimal State 81

violence.”!° It has a right to interfere in disputes between two

nonclients, but no special right beyond that which any person
possesses. This entity has no right to stop one or another of those
nonclients from forcibly but rightfully exacting compensation from
the other nonclient for a wrong suffered. In other words, Nozick’s
“state-like entity” has no right to prevent nonclients from taking the
moral law into their own hands in their dealings with other
nonclients.
Nozick deprecates the importance of this inferential shortfall,
quoting at length from an anthropological account of the state to
support his claim that the state-like entity is near to a full-fledged
state.!! Now, for the pure theory of the state, the gap between state
and state-like entity is exceedingly important. One might as plausi-
bly respond to Kenneth Arrow’s General Possibility Theorem,!? as
an acquaintance of mine once did, by pointing out that majority rule
only rarely produces an inconsistent social preference order!
Leaving pure theory to one side, the significance of the limita-
tions on Nozick’s “state” will depend on certain matters of fact about
which he is silent. His language encourages us to imagine a society
in which no more than a handful of individuals choose not to sign up
with the dominant protective association. But suppose as many as a
sixth or a fifth of the residents of a territory are nonclients. Suppose,
further, that they are geographically scattered and not easily
identifiable by dress, manner, or occupation.1? How will the min-
ions of the state-like entity be able to tell who is and who is not a
client, as they walk street patrol or rush into a barroom to break up a
brawl? Will the state-like entity be forced to conclude that, for all
practical purposes, it must claim to be the sole authorizer of violence
(with due compensations paid, of course)?
Will the state-like entity claim for its employees—its private
police—special rights in their role as representatives of the total
clientele of the association, over and above their rights as individu-
als? Will the state require its employees to be clients as well?
(Premiums could be conveniently withheld from one’s paycheck.) If
a nonclient resists a private policeman who mistakenly interferes in
what he thought was a fight between clients, will the issue between
him and the nonclient be an issue between two individuals, or an
issue between an individual and the state-like protective association
as represented by its agent, the private policeman? If n individuals
82 Robert Paul Wolff

can, through n acts of contractual agreement, transfer their indi-

vidual rights to a single protective association, can that association,
by a single contractual agreement, transfer those aggregated rights
to its agent, the policeman? If so—and Nozick can hardly say
no—then will that policeman, in his personal on-duty encounters
with others, whether clients or nonclients, have rights quite differ-
ent from those possessed by the individuals he encounters? Also will
that effectively deprive ordinary nonclients of their rights vis-a-vis
the policemen?
In short, Nozick’s argument will not do as a justification of the
state. The plain fact is that states claim de jure legitimacy, and it is
such claims, not assemblages or aggregations of transferred indi-
vidual rights, that ground the further claims made by the state on
behalf of its agents, its policemen, its courts, and its executioners.
The second difficulty with Nozick’s argument is that it does not, in
its own terms, establish its intended conclusion. Since he begins
from the libertarian side of the debate, Nozick feels very little need
to argue the claim that a dominant protective association can
legitimately come into existence without violating anyone’s rights.
For him, as for all libertarians, the real problem is how to show that
the protective association has a right (or indeed, a duty) to tax its
clients in order to “redistribute” income to those who cannot or will
not buy protection contracts and thereby become clients. In short,
for Nozick the real nub of the issue is: What obligation have the rich
to buy protection for the poor? His answer—and the linchpin of the
entire construction—is the principle of compensation. The se-
quence of Nozick’s exposition of the principle, somewhat obscured
by his tendency to follow up interesting side-issues, is as follows:
a. After sketching the notion of an area in moral space around an
individual that contains his rights, Nozick asks: “Are others forbid-
den to perform actions that transgress the boundary or encroach
upon the circumscribed area, or are they permitted to perform such
actions provided that they compensate the person whose boundary
has been crossed?” 4
b. A principle of compensation is suggested: “[T]hose who are
disadvantaged by being forbidden to do actions that only might
harm others must be compensated for these disadvantages foisted
upon them in order to provide security for the others.”15
 The Derivation of the Minimal State 83

c. The principle is reiterated with a charmingly ingenuous ac-

knowledgment of its shaky logical status.16
d. Finally, the principle of compensation is flatly invoked as the
justification for “redistribution,” in the form of supplying protection
for nonclients, which traditional libertarians decry as an invasion of
the rights of those taxed.!7 Three pages later, Nozick announces that
his argument is complete.1®
As this survey indicates, Nozick nowhere argues for his principle,
but even if we grant it to him, we must still raise objections to his
employment of it. The clients of the dominant protective association
are obliged to compensate nonclients for their loss of the ability to
enforce their rights against clients. But presumably, only the loss of
the ability to enforce their rights properly need be compensated.
The associated clients are under no obligation to compensate
nonclients for the loss of their ability to enforce their rights impro-
perly, or to enforce false rights claims.19 If Imay, I will employ a
Nozick-style pair of examples: First, if some madman proposes to
enforce his property rights by going out into the world, when he has
suffered a robbery, and randomly torturing people to death until he
obtains a believable confession, I am not required, should I stop him
from doing so, to compensate him for depriving him of the method
of rights-enforcement he has chosen; nor am I obliged to compen-
sate a different madman, should I deprive him of the ability of
enforce the absurd claim that he, as the first-born of God, has a right
to all the movable goods in the human world.
Thus, the clients of the dominant protective association are only
obliged to compensate nonclients for depriving them of their ability
to enforce their true rights properly against clients. By hypothesis,
however, the protective agency employs methods that it considers
proper, and only prohibits methods it considers improper. So from
its point of view, no disadvantage has been suffered by the
nonclients. Hence, again from its point of view, it has no obligation
that it can see to pay compensation, and consequently no right to tax
its clients in order to pay for such compensation. If this suggests
Catch-22 or Big Brother, I can only reply that it does indeed.
Nozick’s dominant protective association looks very much like the
traditional state, with the velvet glove of legitimacy removed from
the iron fist of enforcement.
84 Robert Paul Wolff

Let us examine the notion of compensation more closely, since

Nozick rests so much of his argument upon it. It will be seen that in
this topic, many of the most important lines of his argument come
together. Nozick starts with a very strong version of the classical
liberal conception of the individual. This conception assumes a
sharp and clear distinction between what is inner, internal, private,
or one’s own, and what is outer, external, public, or someone else's.
Nozick captures this conception quite nicely in his metaphor of the
“moral space” of each individual, circumscribed by “a line (or
hyper-plane).”2° After introducing the image of a boundary of one’s
moral space, he thereafter frequently refers to rights-violations as
boundary-crossings. When speaking of actions that threaten to
violate the rights of others, or that run the risk of violating the rights
of others, he speaks of individuals who come dangerously close to
the boundaries of others, and so forth. We are encouraged by such
language to conjure up either of two images whenever Nozick
speaks of actual or threatened rights-violations. The first is the
image of the body. Its surface is the “boundary,” and an invasion of
its surface is a violation of a person’s rights in his own person. It is in
light of this image that we can understand the remark that political
philosophy is primarily concerned with physical aggression. The
second image is that of a piece of “real property,” of land, whose
boundaries may be crossed only with the permission of the owner.
Nozick’s examples as often suggest this image of “boundary-
crossings.”
There is a very considerable difference between the physical
invasion of my body (or its forcible manipulation by others) and an
act of trespass on my property. The frequent talk about “fear,” and
so on, which we will attend to presently, makes sense only in
respect to the threat of physical aggression. The examples of rights
violations, however, are clearly couched in terms of the rights in
real property. What is more, many violations of property rights
cannot plausibly be understood either on the model of physical
aggression or on the model of trespass. When I infringe your
copyright, or steal your car from the public street in front of your
house, I am neither invading your body space nor trespassing on
your land.
Speaking somewhat more abstractly, Nozick’s metaphor assumes
that in moral space, my rights constitute, topologically, a compact
 The Derivation of the Minimal State 85

closed set, the boundary of which is contained in the set. It follows

that there cannot be any points in my rights space entirely sur-
rounded by points in your rights space, although—whatever this
means—my rights space might be entirely surrounded by yours. As
Marx says, each of us, in this liberal model, finds in each other the
barrier, not the realization, of his liberty.21 But suppose that our
rights are not so neatly partitioned into compact subspaces of the
moral space. Suppose, indeed, that in moving from right to right in
the interior, and not just at the boundary, of my rights space, I must
cross the rights of innumerable other persons. In that event, the
notion of a boundary-crossing will dissolve. Nothing in Nozick’s
discussion provides any support for his account of the structure of
the moral space of individual rights.
But this talk of moral topology has about it the air of a jeu d esprit.
There are other problems with the account of disadvantage and
compensation that cut more deeply into the heart of Nozick’s
argument. The moral presuppositions of Anarchy, State, and
Utopia, as we have observed, are radically individualist. They
depend on a very sharp distinction between an inner sphere, where
society in general and other persons in particular have not even a
legitimate concern in general and other persons in particular have
not even a legitimate concern and an outer or public arena of
interpersonal interactions, in which alone the question of the claims
of others against me can appropriately arise.
Nozick, in the language and style of his argumentation, leans
heavily on such notions as utility maximization, compensation pay-
ments, indifference curves, and the like, which presupposes the
abandonment of that sharp public versus private distinction. The
fear or anxiety I may suffer, on account of my anticipation of a
possible violation of my rights, is fair game for an expected utility
calculation. But a Locke-Mill theory of the private and the public
would rule out such considerations as irrelevant to any moral
deliberation concerning rights and duties. If Christian proselytizers
set out, sincerely, to convert Jews to their faith, they may thereby
generate anxiety in the Jewish community over the survival of
Judaism. Would this anxiety count as a disadvantage to the remain-
ing Jews, for which—with suitable adjustments and qualifications—
they would have a right to be compensated? On Nozick’s view, the
answer is presumably yes. However, neither. Locke, nor Mill, not
86 Robert Paul Wolff

any of the classic theorists of rights and border-crossings would

agree. Ironically, Nozick has adopted a model that was developed as
a theoretical elaboration of utilitarianism, and a moral theory an-
tithetical to the intrusive paternalism of utilitarianism.
In general, the argumentation of Part I assumes a situation of
choice under risk rather than choice under either certainty or
uncertainty. All the talk about increased and lowered probabilities,
the explicit assumption of measurement of utility on an interval
scale,22 and the calculations of expected utility, assume choice
under risk. Now, in itself, this assumption need not be fatal; it is an
idealization of reality, and all theoretical analysis requires some such
idealizations. However, there are at least three serious difficulties
with this assumption in the context of Nozick’s argument.
First, the underlying assumption of the derivation of the minimal
state is that individuals, fearful for their lives and property, will
band together into protective associations for security. Leaving to
one side Nozick’s bizarre example of the arm-breaking machine, it is
clear that the fear that fuels the drive for security is a product of
uncertainty, not of risk. Even if we ignore the inner versus outer
problem raised above, and admit this fear into our moral calculus,
we must recognize that it is a product precisely of situations which
lack the structure required for the probability estimates and ex-
pected utility calculations on which the theory of compensation
rests.
Second, as Hobbes and many others have noted, fearful, isolated,
uncertain individuals in a state of nature band together precisely to
achieve that security and predictibility that will, for the first time,
make rational calculation possible.23 One might say, anachronisti-
cally and somewhat facetiously, that one of the purposes of a social
contract is to transform situations of choice under uncertainty into
situations of choice under risk. One of the arguments that can be
advanced in support of a state-enforced system of laws—especially
in the area of property law—is that it reduces uncertainty and
thereby facilitates rationally self-interested economic activity. Such
a formulation makes Nozick’s analysis of the formation, growth, and
stabilization of a dominant protective association circular, for it
assumes the prior existence of the very state of affairs it is supposed
to produce.
Finally—a point to which we shall return—the elaborate calcula-
 The Derivation of the Minimal State 87

tions implied by Nozick’s theory of compensation presuppose an

extremely advanced stage of social, economic, and political integra-
tion. To take a relatively simple example, consider the degree of
bureaucratization of medicine that must come into being in order to
generate usable statistics on the relationship between heart disease
and air pollution. One cannot even raise the question of “compen-
sating” someone for having inflicted on him an increased risk of
heart disease unless one has data of this sort, and the collection of
such data requires a very advanced stage of social integration. The
invasion of privacy, de jure or de facto required by that stage of
social integration, is precisely the evil that Nozick seeks to rule out.
Merely in order to calculate what it owed in compensation to
nonclients, the dominant protective association would have to do
most of the snooping and prying and standardizing and regulating
that is now carried on by the modern welfare state. The only
difference between the two, so far as I can see, is that after inflicting
itself on all of us, as the state now does, Nozick’s state-like entity
would be uncommonly niggardly when it came to distributing
benefits.
In short, Nozick’s real problem is that given his extremely strong
theory of individual rights, side constraints, and so forth, he ought
in all consistency to come to the conclusion that no unconsented-to
boundary-crossings (i.e., rights violations) are permissible, regard-
less of compensation. But that is a crazy conclusion, as he realizes. If
accepted, it would immobilize us all, making us much like a bizarre
gathering of morally musclebound rights freaks, lovely to look at,
but unable to lift a finger for fear of encroaching on one another's
moral space. So Nozick compromises. Of course, once he starts,
only his intuition, or the degree of his moral finickiness, tells him
when and where to stop, and how much to pay in order to achieve
what a mathematically sophisticated Anglo-Saxon of the eighth
century might have called a discounted wergelt raising the tribe to
its previous indifference curve.
Before concluding this first, internal, stage of my critique, I
should like to raise an additional question, and also correct an error
in Nozick’s one explicit use of Game Theory. The error is of no great
importance to his argument, but the question, I believe, goes to the
very heart of his theory, and indicates one of the ways in which it is
inadequate. To put the question as succinctly as possible, what price
88 Robert Paul Wolff

will a dominant protective association charge for its services? A

protective association is merely a private individual or group of
individuals who go into business to sell a service. When such
associations first spring up, price is determined by market con-
siderations. Once either oligopoly or monopoly develops, however,
the dominant protective association can raise the price. For obvious
reasons, there will be rather severe inflexibilities restricting entry of
new firms into the market. If Nozick is correct, and a dominant
protective association emerges with a de facto monopoly, the price
will soar. The owners will charge as much as the market can bear,
which will, in the nature of their service, be a good deal. What is
more, like other monopolistic firms, the dominant protective asso-
ciation will not maximize output, which is to say that its maximum
profit will probably result from a rather lower level of social stability
and security than it could provide or than its customers would like.
The customary laissez-faire safeguard against the dangers of
monopoly is to assign to the minimal state the job of preserving the
conditions of competition, but quite obviously it cannot perform
such a function in this case!
We may therefore conclude that the protective association, once
it acquires a monopoly, ought to charge only the fair market price,
not the monopoly price. However, there is no such thing as a fair
market price for the service sold by the dominant protective associa-
tion! There is not even such a thing as what the fair market price
would be. There could not be a “market” for what the association
sells, because what it sells is the guarantee of a monopoly. Since
there are no substitutes for law and order, consumers cannot even
set limits to the monopoly price by switching commodities. Need-
less to say, the owners of the dominant protective association,
inasmuch as they are merely businessmen out for a profit, will not
be restrained by any of the traditional, irrational constraints on the
exercise of political power, such as patriotism, public spirit, or a
concern for the general welfare.
Finally, a few words about the payoff matrices, and accompanying
analysis.24 The hypothetical matrix is wrong, and the dominance
arguments based on it do not go through. The problem lies in the
figures postulated for payoffs DA’, DB’, D'A, and D'B. Nozick
proposes the following matrix:
 The Derivation of the Minimal State 89

Matrix IT

Person II

A’ B' Gr iDh
A easy 4,6 10, 0 10, 0

6, 4 5, 5 10, 0 10, 0
Person I
0, 10 0, 10 Kix Bo 69
D\ 0, 10 0, 10 Peo ae

This matrix is wrong. An effort to block one’s opponent from joining

any protective association must have some nonzero probability of
success, p. If it succeeds, and if one does not oneself attempt to join
a protective association, then the outcome (ignoring costs of the
effort) will be a mutual state of nature. Hence the true payoff matrix,
including the expected utility calculations thus generated, should
be:?5

Matrix IT’

Person II

AC B' Cc D'

A 5.5
> 456
> 10 On Ee)
> p'x

B Ga
> ais
> 10s08 4ee
> p'x

PersonI
0, 10 O10 FO 90 eG

D P%s PR, Cas ink

(px+(1—p)10) (px+(1—p)10)

Various assumptions about the values of x, p, and p’ will yield

matrices with differing solutions. Consider, very briefly, the fol-
lowing three cases, concocted entirely out of imagination:
i, Letx = 9
p =.l
p'=.8
90 Robert Paul Wolff

In other words, the state of nature is really quite pleasant,

individual I has very little chance of stopping individual IT from
joining a protective association, but II has a very good chance of
stopping I. If one substitutes the numbers into Matrix I’ and
computes the values, it is found that strategy B dominates for I. II,
recognizing this, chooses strategy D'. The payoff is (9.2, 7.2), and
the strategy pair is that I attempts to join a protective association,
while II does not, but II attempts to stop I from joining.
ii. Letx = 8
p =.9
p' = .2
The state of nature is a trifle less attractive. I has a very good
chance of blocking I, while II has a slender chance of blocking I.
Under these assumptions, I has no dominant strategy. II's dominant
strategy is B’. I, recognizing this, chooses D. The payoffis (7.2, 8.2),
and the strategy pair is that II attempts to join a protective associa-
tion while I tries to stop him.
iii. Letx = 6
p=.4
p'=.7
The state of nature is not so hot. I has a fair chance of stopping I,
and II has a good chance of stopping I. I’s dominant strategy is B;
II's dominant strategy is B’; the outcome is (5.5), and the strategy
pair is that each attempts to join and attempts to prevent the other
from joining a protective association. This is a “prisoners dilemma,”
since any of the state of nature outcomes is mutually preferable.
However, all of this is utterly irrelevant to questions of serious
political philosophy! It is also very odd-sounding to anyone who has
been brought up, theoretically speaking, on the great traditions of
Western political writing. In the last section of this Article, I shall
try to come to terms with the deeper meaning of that oddness. Now
let us turn to some considerations of greater moment. In the next
section of my discussion, I shall stand off a bit from the detail of
Nozick’s argument, and raise objections to certain of the assump-
tions that seem to underlie his approach to political philosophy.2®

AN EXTERNAL CRITIQUE OF THE ARGUMENT

Perhaps the most irritating weakness of Nozick’s book is its com-
plete failure to take account of the most obvious and well-known
 The Derivation of the Minimal State 91

facts of human motivation and social experience. For example,

much of his discussion of the workings of a protective association
seems to presuppose that the serious rights-violations against which
one needs protection, are committed, by and large, by the sorts of
solid citizens who will have joined a competing association, will be
paid up on their premiums, and will have known addresses where
they can be found. This may indeed be so in a small, rural
society—one in which everyone knows everyone else, and in which
an act of barn-burning or cattle-rustling can pretty certainly be laid
at the door of those no-account Finkelstein brothers. But in the
context of big-city street crime, Nozick’s model is simply irrelevant.
To put the point more generally, Nozick presupposes a society so
settled, so orderly, that one might never feel the need for a
protective association at all, let alone a state!
Nozick seems to me equally insensitive to the psychological,
social, and institutional problems involved in creating and staffing a
responsible, controllable police force, whether “public” or “private.”
The problem begins as soon as one introduces the notion of an
agent. An agent is a private individual who adopts a social role. As
an occupant of that role, he has rights, powers, responsibilities, and
duties which he would not have were he not occupying the role, and
which he puts aside when he steps out of the role. Thence—given
the limits of the power of reason—comes the function of uniforms,
titles, oaths of office, and similar accoutrement. They serve both to
inform others of the role one is playing and to strengthen one’s
identification with the role. From this follows also the importance of
internalizing the norms associated with a role, as opposed merely to
making the appropriate adjustments in one’s expected utility calcu-
lations. Nozick knows all of this, of course. He simply ignores it in
the construction of his model of the rational individual and his
analysis of the moral relationships between individuals.
Perhaps we can develop the philosophical underpinnings of these
observations more systematically by examining the protective asso-
ciation on which Nozick erects his justification of the state. Follow-
ing the standard libertarian account, Nozick represents such asso-
ciations as companies that offer a service in the market, advertise for
customers, promote sales by such devices as 13 weeks free protec-
tion with a 2-year subscription, money-back guarantees, and so
forth. As he repeatedly insists, these companies are groups of
92 Robert Paul Wolff

individuals, and they have only individual rights and aggregates of

individual rights which they, as individuals, exercise either directly
or through their agents. There are no emergent rights, attaching
only to corporate bodies and incapable of being decomposed into
component individual rights.
The possibility of a protective association (that is to say, of a
morally legitimate protective association) rests on four supposed
moral facts, asserted (but not shown) to be facts by Nozick:
1. Each person in the state of nature has the right to enforce his
(other) rights in a morally proper manner, and to exact suitable
compensation in an appropriate manner from those who have
violated his rights.
2. Each person has the right—suitably hedged around—to
punish rights violations against third parties.
3. Several persons may, through free and mutual agreement, do
collectively in the way of rights enforcement and infraction punish-
ment.whatever they may do severally and singly.
4. An individual, and hence a group of individuals, may assign
the tasks of enforcement, punishment, and so forth, to other persons
as their agents (perhaps, but not necessarily, as their employees).
These agents will act not in their own right as persons, but in their role
as the authorized representatives of others. Rights are transferable
in such manner that one person might, through a number of such
transfers, come to be the bearer of many rights, just as one
representative might bear many proxies in a committee election, or
one lawyer represent the property interests of many clients in a
Sule."

The operative assumption is clearly assumption 4, which under-

lies the moral legitimacy of protective associations as opposed to
mere mutual aid societies. Let us assume that I can assign my rights
to an agent, hire him to represent me, to do in my name what I have
a right to do but what he, merely as an individual, might not have
the right to do. Even granting all that, it must be obvious that I
would stand under an obligation to monitor the actions of my
representative, to ascertain that he has done only what I have
authorized him to do, and that only in permissible ways. This
obligation follows from the fact that I have the same obligation when
I act as my own agent. If my agent violates the rights of others, I as
well as he can be held responsible.28
 The Derivation of the Minimal State 93

Although it may be a relatively simple matter to monitor the

behavior of my personal bodyguard, my personal lawyer, or the
holder of my personal proxy, it very quickly becomes impossible in
practice for me to exercise effective oversight as the protective
association grows. Bureaucratic rationalization and institutionaliza-
tion take over. It is not I who hire the association’s enforcers (or
private policemen); bureaucrats in the association's employment
office do. I merely write out a monthly check to pay the premium on
my comprehensive insurance policy. Since the protective associa-
tion is, we may suppose, a mutual benefit insurance company, I
receive in the mail each year a notice of the annual shareholders’
meeting, together with a request from the management for my
proxy. I have roughly the same sort of control over the actions taken
by the protective association in my name as I do now over the
actions of the telephone company—with one exception: Now, if I get
mad enough at the telephone company, I can write to my Con-
gressman and ask,that the government pass a law regulating the
telephone company. In Nozick’s model, however, the dominant
protective association is the government! As a device for guarante-
eing individual liberties and enforcing absolute side constraints, this
is, to put it gently, a trifle feckless.
Nozick, we must recall, is not an anarchist. His purpose is to
prove that the just state is possible, not that it is impossible. Perhaps
it not de facto tyranny to which he objects, only income redistribu-
tion. Therefore, we cannot defeat his argument merely by observing
that it is an ideological rationalization for AT&T. Let us therefore
take a closer look at assumption 4, with which we began this line of
analysis, and at the argument that depends upon it.
The key to the assumption is the claim that person A can transfer a
right in toto to person B. In Nozick’s view, the full right passes, by
means of a contractual agreement. (His theory of justice in transfers,
which is part of his theory of entitlement, is merely a special case of
this general claim.) Hence the entrepreneurs who own the protec-
tive association accumulate a stack of rights from their clients. They
can in turn transfer those rights, in aggregation, to employees of the
firm who walk the streets, staff the jails, ran the courts, and collect
the fines, all of them living bearers of those aggregated rights.
If total transfers of that sort are in fact permissible and possible,
then Nozick might be able to carry his argument through (leaving to
one side such objections as have already been raised earlier in this
94 Robert Paul Wolff

article). However, Nozick is guilty here of an error that we might

label “the fallacy of the transitivity of rights transfers.” It bears a
resemblance to the notion that indifference is transitive, although
not too much weight can be placed on that comparison. When an
individual is called upon to order a set of elements by means of the
relation “preferred or indifferent to,” he may judge himself to be
indifferent between x and y, and indifferent between x and w. This
would be explained by the fact that the differences between the
members of each pair were too small to affect his preference
judgments, too small to be noticed, whereas the aggregated differ-
ences, as revealed in the comparison of x with w, might exceed his
threshold of indifference.
By analogy, in a simple rights transfer, as when I hire a lawyer to
close a real estate sale for me, there is a minute slippage or blockage
in the rights transfer, due to the fact that my agent is also an
independent human being. Because he is a person as well as an
agent, there is a small but nonzero probability that he will exceed
his authority, or get his instructions confused, or interpret a situa-
tion in a manner that I would not approve. There is also a nonzero
probability that I will be unaware of the breakdown of agency, or
will be unable to rectify it. Because Nozick focuses his attention on
simple rights-transfers, where the probability of slippage falls below
the minimal threshold of moral awareness, he fails to see that as the
protective association grows, as the rights collected are transferred
and retransferred, as my relationship to my so-called agent grows
ever more attenuated, I will become less and less able to see my
own will, my own moral agency, in the actions of the association’s
owners and employees. As the imperfection of the transfer
magnifies, my right to consider the transfer as having taken place
diminishes. Eventually, I must recognize that for all practical moral
purposes, I cannot exercise the oversight that is a necessary compo-
nent of any permissible rights-transfer. I must therefore withdraw
my authorization from the association. The net result is an unstable
fluctuation in the size of the clienteles of the protective associations,
with the mean size oscillating between limits the higher of which is
no where near large enough to permit even a momentary pretense
of dominance.
Obviously, this point could be expanded upon at great length, but
inasmuch as others have done so,?® there is no need to elaborate on
the subject here. Suffice it to say that Nozick appears to have no
 The Derivation of the Minimal State 95

appreciation of the staggering problems of controlling a protective

association, or monitoring those actually entrusted with the tasks of
enforcement. Since he must assume some level of rights violation,
and hence some tendency of individuals to commit such violations,
in order to get his argument going (otherwise, who needs a state?),
he cannot pass this off as a practical detail from which his model
abstracts.
The real problem—indeed, the underlying problem with all of
Anarchy, State, and Utopia—is Nozick’s persistent failure to take
account of the nature of social reality. Nozick’s models, methods,
and arguments all treat social relationships as transparent rather
than as opaque. He portrays social interactions as marginal to the
existence, integrity, and coherent identity of the individuals who
participate in them, rather than as central and constitutive. It
follows that he can have no usable notions of false consciousness, of
self-deception, of alienation, and of the objectification of subjective
categories. The démystification of social reality, which ought to be
set as a major task for social theory and social practice, is simply
assumed by Nozick as a given presupposition of his analysis.*°
Nozick frequently assumes, for purposes of assessment ofliability
and payment of compensation, a degree of transparency of social
relationships sufficient to permit plausible and usable ascriptions of
responsibility, disadvantage, gain, or loss, to specific, identifiable
individuals. When I speak of “the transparency of social relation-
ships,” I mean the possibility of tracing out at some length the
filiations, the links in a chain of causes and effects, before the
complex interconnections are obscured by the mist of social sur-
faces. To get some sense of what such a tracing out might mean, one
can contrast a title search for a piece of land, which is carried out
through ten or fifteen changes of ownership over more than a
century, with the attempt—almost sure to be frustrated after two or
three points of exchange—to trace back the lineage of a quantum of
money as it is exchanged for a commodity, divided and exchanged
again for several commodities, combined with other monies and
exchanged once more for a service, and so forth.
Even a little reflection will reveal how problematic such ascrip-
tions of disadvantage or responsibility are in a society like ours. Any
intelligent and thoughtful person is perpetually troubled both by
the legal attributions of specific liability to persons whose role in
some rights-infringement seems a social accident, and by the con-
96 Robert Paul Wolff

sequent dilution of the notion of individual responsibility, with the

inevitable conclusion that no one in our society ever does anything,
and that some nonbeing called “society” bears all the blame. Nozick
excuses himself from the burden of this contradiction by simply
denying one-half of it and doggedly asserting the other half. For
example, in order to distinguish a legitimate protective association
from a protection racket, Nozick invokes the distinction between
productive and unproductive exchange.*! The distinction, however,
as he elaborates it, is internally contradictory. It requires that we be
able to trace out the effects, actual or possible, of one person's
actions on others. This, in turn, presupposes a level of knowledge
possible only at a very advanced stage of socioeconomic integration.
However, as Nozick acknowledges in his discussion of property
rights and the theory of entitlement, once we have reached that
advanced level of socioeconomic development, no one any longer is
in a position to assert, with confidence, the sorts of property rights
on which the theory of productive exchange relies.
Somewhat more generally, the degree of development of social
interdependence and economic integration necessary to provide the
institutional base for the sorts of calculations Nozick posits is so great
as to undermine the privateness of property rights. This is especially
evident in the lengthy aside on pollution. Nozick treats pollution
as marginal to the operations of a modern industrial economy
(although not therefore unimportant, of course). More precisely, he
treats external economies and diseconomies as marginal to the
operations of firms or of individuals. But “pollution,” taken broadly,
is a metaphor for modern society! That is what Marx meant by the
socialization of production.
It is an odd historical fact that Nozick’s principal methodological
tool—the concepts and models of game theory—came on the intel-
lectual scene at a time when it was peculiarly inappropriate to social
reality. Historically, classical economic theory emerged at a time
when economic relationships were still relatively more transparent
than they are today. Nevertheless, the theory assumed a total
opacity of economic relationships. It posited a system of producers
and consumers in which the size of the contribution of each firm,
consumer, or worker to the market was vanishingly small. This
permitted the theory to treat wages, prices, and the various other
indices of economic life as given objective facts, to which the
 The Derivation of the Minimal State 97

individual or firm adjusted on the basis of a calculation of relative

advantage. As the theory developed, it took account of the facts of
oligopoly and monopoly in its formal models. In short, it acknowl-
edged, and made theoretical room for, the fact that individual firms
might be able to calculate the effects of their actions on supply or
demand curves. But the calculations still assumed an opaque social
fabric. The identities and characters of the other economic actors
were obscured from view. During the forties, when advanced
industrial society had achieved so high a degree of functional
integration that individual robber-barons had given way to anonym-
ous board chairmen, and family firms had given way to multinational
corporations, along came a new, mathematically sophisticated
model of economic analysis—game theory. For the first time an
economic theory took formal account of the existence, identity,
values, and reasoning processes of the other economic actors whose
rationally chosen policies, in ongoing interplay with one’s own,
produced those price levels, wage levels, and supply and demand
curves that conventional economic theory had initially treated as
given.
Nozick’s extensive use of the game-theoretic model of rational
choice is systematically inappropriate to his subject. He assumes
throughout that the formal criteria of rational decision can be
abstracted from the concrete social reality which is their matter or
content, as when he rather irrelevantly brushes aside a century and
a half of sustained criticism of the classical and neo-classical
rationalization for industrial capitalist wealth with a fantasy about
Wilt Chamberlain.#* As we have seen, the Nozick model implicitly
makes assumptions—risk rather than uncertainty, transparency
rather than opacity—that presuppose specific stages of
socioeconomic development. The net effect is to beg most of the
important questions of social philosophy in a manner that provides
ideological comfort for policies and doctrines which have never been
established by argument.

ON THE WEIRDNESS OF ANARCHY, STATE, AND

UTOPIA

In recent years, a number of philosophers, political scientists, and

economists have adopted the style of language and mode of analysis
98 Robert Paul Wolff

that one finds in Anarchy, State, and Utopia. The rhetoric of game
theory, if I may characterize it in that way, first appeared in the late
fifties and early sixties in discussions of nuclear deterrence theory. I
have in mind such books as Schelling’s The Strategy of Conflict,**
which was probably the most distinguished intellectual contribution
to that debate. More recently, it has appeared in the writings of
such theorists as James Buchanen and Gordon Tullock,*° and—in a
rather subdued manner—in John Rawls’ work. When I read books
of this sort, I have two initial reactions. The first is that they are
clever, witty, iconoclastic, that they look at old questions in remark-
able new ways. The second is that they are creepy, that there is
something fundamentally awry in the language and reasoning of the
work. When I read Anarchy, State, and Utopia, I have both of these
reactions. The first is easy to explain; Nozick is easily the brightest,
most imaginative, most ebullient political philosopher to appear on
the American philosophical scene for some time. The second reac-
tion, however, is somewhat more difficult to explain, and it is only
after some considerable reflection that I think I am able to get at its
roots. In this last section of my discussion, I shall try to account for
the curious impact of Nozick’s style of political argument on myself
and, I suspect, on other readers as well.
The growth of capitalism transformed certain spheres of human
activity—the productive, the economic—by rationalizing them (in
Max Weber's sense of that term). It came to be accepted, even
praiseworthy, to apply rational principles of cost, profit, and benefit,
to activities that had previously been dominated by customary,
religious, or other norms. But broad though the scope of the
economic is in social life, there remained a great deal of life that was
very much less considerably affected by the change, notably relig-
ion, politics, family life, and personal relationships.
Utility theory, game theory, and their associated models of
rational choice, seek to extend the methods of calculation, the
presuppositions and rhetoric of rationalized economic activity into
spheres of life hitherto shaped or governed by quite different sorts
of considerations. One can make a joke of this move, as when one
asks whether love is a zero-sum game, a bargaining game, or a game
of perfect coordination. One can use the rhetoric and methodology
for covertly ideological purposes, as the deterrence theorists did in
 The Derivation of the Minimal State 99

the late fifties.87 One can also seriously undertake to explore

political and moral life with these models, as Nozick does. However
one deploys the models, the sense of surprise comes from the
incongruity of applying a terminology drawn from one field to
phenomena usually considered in an entirely different field. An-
thropologists achieve this surprise when they apply a terminology
that we associate with primitive societies to the urban life of an
advanced industrial society.
Nozick continually employs this rhetorical trick. He will consider,
for example, forcibly restraining someone from defending his own
interests according to his own view of them and paying him off for
thus restraining him. Nozick describes this as compensating him
sufficiently to raise him to his previous indifference curve,#° a form
of speech that we expect to find in a formal treatment of problems in
welfare economics. The notion of an indifference curve presupposes
the rationalization of a sphere of human experience. It presupposes
that notions such as homogeneity, continuity, and substitutivity can
find plausible application. What makes talk of this sort creepy (if I
may repeat my rather unphilosophical word) is the assumption
thereby insinuated that a hitherto uninvaded sphere of human
activity should be similarly rationalized—and thus made ready for
the extension into it of these models and methods.
Now perhaps we can see why Nozick’s book is so strange. Nozick’s
decision to write about questions of morals and politics in the
manner he does constitutes a covert proposal to transform into
quasi-market-rationalized form important areas of human experi-
ence that have until now not been so treated. Such a proposal is
inhuman; that is to say, it is a proposal to dehumanize much of our
experience. To see that this is so, one need only reflect on the effect
of such rationalization on the world of production and exchange.
Nozick may reply that I take too simple-minded a view of the
matter, that I impute to him a vision of amechanized, computerized
life that bears no relation to his discussion. Just as there is room in
economic calculations, he might point out, for some workers’ pref-
erence for leisure over higher wages, or for a consumer's
“noneconomic” pleasure in doing business where he is personally
known, so there is room in Anarchy, State, and Utopia for obsessive
fears of bodily harm, for soul-deep commitments to home and
100 Robert Paul Wolff

family, or for dogmatic religious convictions (all of which might, on

some inadequate construal of the term, be stigmatized as “irra-
tional”).
However, such a reply would be too quick, and fundamentally
wrong, in my judgment. The methodology infects the reasoning.
The root problem is not at all that the method is too precise for the
data (as Aristotle might have objected), but quite the reverse. If
Nozick’s inferences were tight, then we would be obliged to live
with them, no matter how counterintuitive his conclusions might
be. But of course we are not presented with inferences at all. We are
offered a flood of rapidly sketched situations—scenarios—in which
there are either no actual figures cited or in which the figures are
“for illustration only.” The real burden of the argument is not on the
reasonings themselves (for without more elaborate sophistication or
more stringent simplification, we could never judge their validity),
but on the plausibility of looking at matters in the manner implied
by the language and methodology. Nothing is ever said to suggest a
reason for accepting that new and peculiar way of looking at things.
Consider simply the notion of compensating someone for a
“boundary-crossing.” Some compensation involves, among other
things, paying him for the indignity of the infraction. Now, it is one
thing to pay a man damages for an affront to his honor. It is quite
another to say that his honor has a price—that the payment, in fact,
has determined the market price of his honor! Indeed, once it has
been established that a person’s honor has a price, he may plausibly
be said to have lost his honor, in which case its market value is nil.

CONCLUSION

Despite its brilliance, its imaginativeness, and its sheer air of

intellectual high spirits, Nozick’s book cannot, in my opinion, be
judged a success. Its central argument will not stand up, although it
shares that failing with most of the truly distinguished works in the
corpus of Western political theory. More seriously, its treatment of
politics abstracts from the essential character of social life, and
thereby merely fails to come to terms with the most complex and
intractable problems of political theory and practice. Most seriously
of all, its language and methodology encourage us to treat as already
rationalized those spheres of human experience that have not yet
 The Derivation of the Minimal State 101

been subordinated to the dehumanization of quasi-economic

rationalization, and that ought to be protected at all cost from such
subordination.

NOTES

1. R. Nozick, Anarchy, State, and Utopia (1974).

2. Id. at 9; see J. Locke, Two Treatises of Government 309-23 (Laslett
ed. 1963) (Second Treatise).
R. Nozick, supra note 1, at 32.
. Id. at 10-11. Moral intuition appears to be the guide here.
pidsatl2—15,
Id.
Sidvat 15-22,
wDWARDMA
A midsatoA—25:
9. I cannot resist calling attention to one rather curious historicological
point concerning Nozick’s theory of property, however, particularly since it
reinforces the general conclusion of this Article. Nozick presents a recursive
theory of entitlement, according to which repeated acts of just acquisition or
transfer of property necessarily result in a just set of individual holdings,
regardless of its pattern. Although Nozick never enunciates a principle of
justice in acquisition (a fatal flaw, one might have thought, in a recursive
theory), and explicitly rejects Locke's attempt to ground just acquisition in
the notion of mixing one’s labor with a bit of unheld property, he does
invoke, as an essential element ofhis “theory,” a well-known qualification in
the chapter on property in the Second Treatise that Nozick labels “the
Lockean Proviso.” The passage reads:

“Whatsoever, then [man] removes out of the state that nature hath
provided, and left it in, he hath mixed his labour with, and joined to it
something that is his own, and thereby makes it his property. . . . For this
labour being the unquestionable property of the labourer, no man but he
can have a right to what that is once joined to, at least where there is
enough, and as good left in common for others.”

J. Locke, supra note 2, at 329 (emphasis added). The italicised words are
the “Proviso.” Nozick construes this as claiming that property is initially
private and individual, so that society, or the state, can assert no claim to
the holdings of an individual that he does not freely warrant—so long as the
Proviso has been satisfied. But a careful look at the remainder of Locke's
discussion of property makes it clear that his view is the very opposite of
Nozick’s! According to Locke, God acquires title to the universe by creating
102 Robert Paul Wolff

it. Since he creates it ex nihilo, its entire value is value added—there being
no raw materials. He pays nothing to primary producers, and since by that
creative act He also creates the space which the universe occupies, he need
not even pay rent. God, out of His infinite Goodness, then gives (i.e., in
Nozick’s terms, transfers) the earth to mankind in common. At the same
time, He lays down the conditions under which an individual may rightfully
remove a piece of property from the common holding, and appropriate it for
himself. If we secularize this theory, it is not difficult to see that it is really
based on the supposition that property is originally social or collective, and
that individual rights to property are granted by—and hence can be limited
or taken away by—society. The opposite view, that property is originally
individual, is completely contrary to Locke’s orientation, and also, to the
facts of history and society. When Nozick points out, in his attack on Rawls,
that commodities come into the world already loaded down with individual
entitlements, he forgets that by his own theory, such entitlements arise in
the first instance only from just acts of transfer (or labor-power and the other
factors of production), and hence presuppose some adequate grounding in
just acts of acquisition.
10. R. Nozick, supra note 1, at 117.
11. Id. at 116-17, quoting L. Krader, Formation of the State 21-22
(1968).
12. See K. Arrow, Social Choice and Individual Values (2d ed. 1963).
13. They are, we miay imagine, all that remains of the great anarchist
movement that overthrew the more than minimal state and created the
conditions out of which the dominant protective association emerged.
14. R. Nozick, supra note 1, at 57.
oy Wiel Sis Vey
16) Td-at'87,
17. See id. at 110-11. The phrase “[a]ccording to our principle of
compensation given in Chapter 4,” id. at 110, makes it clear that by this
point in the text, Nozick is taking the principle as having been established,
not merely suggested. See id. at 115.
182 Id, at 114:
19. Since Nozick seems to interpret his own principles ad hoc, to suit his
argumentative purposes, I am at a disadvantage in attempting to determine
what inferences can and cannot be drawn from them, a disadvantage for
which, presumably, I ought to be compensated by being held to a some-
what less stringent standard of proof.
20. R. Nozick, supra note 1, at 57.
21. See K. Marx, On the Jewish Question 163, in 3 K. Marx & F. Engels,
Collected Works 146-74 (1975).
22. R. Nozick, supra note 1, at 58.
 The Derivation of the Minimal State 103

23. See T. Hobbes, Leviathan 189-201 (MacPherson ed. 1968).

24. R. Nozick, supra note 1, at 121-25.
25) Where p = the probability that I will stop II from joining an associa-
tion.
(1-p) = the probability that II will succeed despite I's efforts.
p' = the probability that II will stop I.
(1-p') = the probability that I will succeed despite II's efforts.
px + (1-p)0 = px = the expected value to I of the effort to stop II.
px + (1-p)10 = the expected value to II of an attempt to join an
association in the face of I’s opposition.
26. For the purpose of organizing my remarks in this Article, I have
distinguished in the titles of my sections between “internal” and “external”
criticisms of Anarchy, State, and Utopia. The distinction, however, is
scarcely fixed, and certainly not of any philosophical importance, so readers
who do not find it intuitively clear are urged to ignore it.
27. To see the force of this assumption, we need only observe that even
though A, in a state of nature, has a right to punish B’s violation of C’s
rights, he may not have the same right that C does to punish B. C may have
the right to use risker methods of defense or of compensation; he may have
a right, that A does not have, to forgive B for the infraction, or to offer B
alternative modes of compensation. Should A become the agent of a
protective association to which C has transferred his rights of retaliation and
enforcement, however, he would then acquire in his role as C’s agent the
rights that C, but not he, possessed in the state of nature.
28. This is a point on which Nozick’s mentor, Locke, lays heavy em-
phasis. See J. Locke, supra note 2, at 365-66.
29. For a brilliant exposition of this point, see Rousseau, The Social
Contract 88-96 (Cranston ed. 1970).
30. IfI had been trained on the continent, in the dialectical mode, rather
than in America by analytic philosophers, I might be tempted to suggest
that there is, in the history of modern social theory, a dialectical progres-
sion:
From the classical liberal assumption that social relationships are
transparent, so that rational individuals already possess an
adequate understanding of the true nature of society;
which is the first thesis:
To the conservative, irrationalist view that society is mysteri-
ous, nonrational, incomprehensible, so that human reason
cannot fathom it; which is the first antithesis:
To the higher claim that society is now opaque, mysterious,
incomprehensible, but that reason can, by developing or
perfecting itself, arrive finally at the realization that soci-
104 Robert Paul Wolff

ety is truly rational, and hence that social relationships can

be grasped by reason; which is the first, or Hegelian,
synthesis:
From the Hegelian synthesis, which becomes the new, or second
thesis:
To _ the utopian socialist doctrine that society is now irrational,
and must be changed immediately by action to make it
conform to reason’s dictates; which is the second antithesis:
To the recognition that the achievement of collective, or so-
cial, rationality is a collective human project, requiring the
union of thought and action, and requiring both a trans-
formation of social institutions and a transformation of our
thought about social institutions, each transformation both
assisting and drawing assistance from the other; which is
the final, or Marxian, synthesis.
31 . See R. Nozick, supra note 1, at 84-87.
32 . Id. at 79-81.
33 . R. Nozick, supra note 1, at 161-63. In the nineteenth century,
apologists of capitalism, such as Nassau W. Senior, spoke the religious
language of “abstinence.” See N. Senior, An Outline of the Science of
Political Economy 58-59 (1836). “Outstanding athletic ability” is not much of
an improvement as an explanation for the existence of massive accumulations
of capital!
34. T. Schelling, The Strategy of Conflict (1960).
35. J. Buchanen & G. Tullock, The Calculus of Consent (1962).
36 . See J. Rawls, A Theory of Justice 153-58 (1971).
37 . See, for example, H. Kahn, On Thermonuclear War (1960).
38 . R. Nozick, supra note 1, at 57.
 III
Rights: Opposing the
Extensive State
 pee

if

(6b 1) AAAS D necteagl S ~

7 ; _ me @ Saw G 6 ae oy
se
ce ge ‘etl me ~ ° ee eh ee : \
‘ban PElRy. tly me mre
@ PALLY Gow emut
pai Pe quert ——! mae Re Gtis aaa :
 6
Nozick on Rights,
Liberty, and Property
THOMAS SCANLON*

In Anarchy, State and Utopia’ Robert Nozick approaches political

philosophy within a framework which at first sight seems both
familiar and congenial to contemporary liberal thought. It is a
framework which emphasizes individual rights and the derivation of
political obligation from consent. The conclusions of the book,
however, are liberal in the nineteenth-century sense of the term.
Nozick holds that the only legitimate state is the minimal state,
whose activities are confined to the protection of individuals and
their property and to the enforcement of contracts. This state is
unique among social organizations in having the right to force
residents to pay for its services whether or not they have consented
to do so. Citizens may band together for whatever other purposes
they may desire—to provide education, to aid the needy, to or-
ganize social insurance schemes—but such schemes must be purely
voluntary, and the state must enforce anyone's right not to be
compelled to contribute to them.
Nozick reaches these conclusions by adhering as closely as possi-
ble to the idea that, in economic life as in politics, all valid
obligations derive from consent. Of course, consent alone cannot be
theoretically basic. Something must determine the conditions under

107
108 Thomas Scanlon

which acquiescence counts as morally binding consent. In addition,

the obligations and entitlements one person acquires through volun-
tary agreements can affect the alternatives open to others who have
not been parties to these agreements. Something must determine
when such side effects make an agreement void. In Nozick’s theory
these conditions and limits are set by a skeletal framework of rights
derived from Locke. The minimal role allowed to the state and the
great scope left to voluntary agreement and consent in his theory are
direct consequences of the particular character of these rights. This
system of rights is not argued for directly in the book, however, and
Nozick does not claim to have given these rights a foundation (p. 9).
The impact of the book and the support it offers to Nozick’s view
derive mainly from a series of challenging questions, engaging
examples, and theoretical devices designed to make his conceptions
of rights and justice intuitively appealing and to make alternative
views appear untenable. I will therefore begin by considering a
number of these examples, returning later to the framework of
rights and its Lockean pedigree.
The central theoretical device of the book is the classification of
principles of justice as “historical,” “end-state,” and “patterned.”
Nozick classifies a principle of justice as historical if that principle
makes the justice of a distribution depend on how it came about (p.
153). It will follow from a historical principle of justice that “past
circumstances or actions of people can create differential entitle-
ments or differential deserts to things” (p. 155). By contrast, under
what Nozick calls an end-state principle, the justice of a distribution
will depend only on certain structural features of the situation it
represents, for example, on the amount of utility produced or on the
degree of equality obtaining. Of course, the structural features of a
distribution that are deemed relevant by a principle of justice might
make reference to historical events. A principle might require, for
example, that people's holdings should be proportional to their
moral worth as determined by their past actions. Such a principle is
historical in Nozick’s sense, but it clearly has a great deal in common
with end-state principles. Nozick calls such a principle, one which
specifies that holdings are to “vary along with some natural dimen-
sion” or some combination of such dimensions, a patterned princi-
ple. Nozick’s own theory of justice is based on unpatterned histori-
cal principles. This theory is an entitlement conception of justice.
 Rights, Liberty, and Property 109

Such a conception is specified by three components: a principle of

(initial) acquisition, a principle of transfer, and a principle of rec-
tification. Its central tenet is that any configuration of holdings that
results from the legitimate transfer of legitimately acquired holdings
is itself just. There is no reason to expect (or to require) that such
holdings conform to any natural pattern. (The principle of rectifica-
tion comes into play to explain how, if holdings are affected by
violations of the principles of just acquisition and transfer, this
situation is to be remedied.)
Many theories of justice, almost any theory perhaps, will give
some role to considerations of entitlement; that is, they will recog-
nize some processes as conferring legitimacy on their outcomes.
What is special about Nozick’s view is that it makes entitlement
principles the beginning and end of distributive justice. While his
principles are not described in detail, it appears that his theory
differs from other pure entitlement conceptions chiefly in admitting
fewer restrictions on the acquisition and exchange of property. He
mentions only one such restriction, called “the Lockean Proviso,”
which provides that any acquisition, transfer, or combination of
transfers is void if it leaves third parties worse off then they were in
the state of nature.2 Such a worsening might occur, for example, if
someone were to buy, in simultaneous secret transactions, rights to
all the available sources of water. This restriction could be substan-
tial were it not for the fact that the baseline for its application is set
by conditions in the state of nature. According to Nozick the
productivity of the capitalist system in improving our material
condition makes it unlikely that (in a competitive economy) anyone
could acquire holdings that would leave others below this standard.
Nozick clearly feels that the distinction between historical (unpat-
terned) principles of justice and patterned or end-state principles is
of fundamental importance. He emphasizes that almost all of the
principles of justice commonly offered are end-state or patterned
principles and, as such, are clearly mistaken. If this were correct it
would indeed be important. Certainly Nozick’s distinction does
capture something intuitively appealing. It has often been said as a
criticism of utilitarianism that it ignores morally significant relations
resulting from past actions.? Nozick’s distinction is of great interest
if it gives this criticism a more abstract form and shows it to apply
not only to utilitarianism but also to Rawls’ theory and to virtually
110 Thomas Scanlon

every other theory commonly offered. But I do not think that the
distinction has the importance claimed for it. To see why, let me
consider the reasons Nozick offers for holding that all patterned
principles are clearly wrong.
These reasons can be quickly seen in Nozick’s frequent remark
that, since gift giving can upset a pattern of distribution, supporters
of patterned principles of justice would have to forbid this form of
“loving behavior” (p. 167). More generally, let D1 be the distribu-
tion of goods obtaining in a given society and suppose that this
distribution is in accord with our favorite pattern (for example, strict
equality). This distribution can be changed into another distribution
De, not in accord with this pattern, by any one of a variety of means:
by gifts, by someone's starting a very successful business in his spare
time using only resources to which he was already entitled under
Di, or, as Nozick suggests, by all one million of us willingly paying
Wilt Chamberlain 25¢ apiece for the privilege of watching him play
basketball. To maintain D1 one would have to restrict these ac-
tivities. Such “continuous interference” is, Nozick says, obviously
unacceptable. Therefore one must conclude that no patterned con-
ception of justice can be correct.
One immediate response to this argument is to doubt whether
anyone ever held a “patterned” conception of justice in the sense
that is here refuted. A person who objects to the inequality in the
world is unlikely to be concerned with those who have less as a
result of their giving away or trading part of what was once an equal
share. What offends an egalitarian primarily is the great inequality
in the initial resources people have as a result of the social positions
in to which they are born. But here Nozick can respond that this
does not escape his point. Arbitrarily great inequalities in the
starting places of members of one generation can result from gifts
and voluntary exchanges by members of previous generations.
Thus, maintaining even this looser kind of equality can require
restricting these activities.
So put, this is not such a startling conclusion; certainly it does not
make egalitarianism look as foolish as first appeared. This is so, first,
because there is no longer the appearance of unanimous consent. It
is no longer plausible to respond, “Well, if the fans are all happy to
pay [everyone now living in the society is a fan] and Wilt is willing to
play at that price, how can a meddling egalitarian object?” Second,
 Rights, Liberty, and Property o17

this way of looking at the example changes our picture of the

liberties that are likely to be infringed. The liberties involved in the
example seem to be these: the liberty of the fans to pay an extra
quarter to see Wilt play, his liberty to keep any amount he may
receive through such transactions, his liberty to decide whether or
not he wants to play for the amount remaining after taxes from what
the fans and promoters offer him, and, finally, the liberty of his heirs
to keep any amount of money he wishes to pass on to them. It does
not seem likely that egalitarians, if their objectives are as I have
described them, will want to keep watch over everyone’s quarters or
to conscript basketball stars. What is at issue, then, is the right of a
person to keep as much as others are willing to pay him for his
services and the right of heirs to receive unlimited bequests. But
there is no strong intuitive ground for thinking that these rights are
absolute, and little ground for surprise at the suggestion that the
pursuit of equality might call for their infringement.
Nozick tries to make such measures seem more alarming to us by
tying them to moré extreme forms of intervention. Thus he says that
“Taxation of earnings from labor is on a par with forced labor’ (p.
169), and elsewhere he asks why, if we are going to set a limit on
how long a person can control goods and transfer them to others, we
do not have immediate confiscation (p. 163). But there seems to be
no reason to disregard such obvious differences in the degree of
regulation of a person’s life. It may be true, as Nozick claims, that
there is a continuum of interferences extending from taxation to
forced labor, each foreclosing a few more options than the preceed-
ing. But the fact that there is such a continuum is no reason why we
must be indifferent between any two points along it. Even if Nozick
does not convince us that restrictions on earnings or inheritance in
order to maintain equality are unacceptable, however, his examples
do raise the question why any interference at all should be justified
in order to preserve a pattern. As he says, what is so great about a
pattern?
There are many different concerns which lead people to call for
greater equality, and not all of these involve a pattern in a funda-
mental way. For example, a person’s primary political goal may be
to alleviate the terrible conditions under which many are forced to
live. The fact that others are at the same time much better off shows
that it would be possible to eliminate this suffering and, one might
hb Thomas Scanlon

add, to do so without reducing anyone else to this low a level. The

resources are there; they just need to be redistributed. For a person
taking this position, a humanitarian, equalization is merely a means
to the improvement of the lot of those currently worst off. It is
possible that a person who is intensely concerned with this cause
today might be quite satisfied if the living standard of everyone in
the world were significantly improved, even if the gap between rich
and poor was left unchanged.
A second position would take pattern more seriously but still
assign it a purely instrumental role. A person taking this position is
concerned by the fact that where there are great differences be-
tween rich and poor, especially where wealth is concentrated in a
few hands, the wealthy come to have an unacceptable degree of
control over what jobs there are, over what is to be produced and
over political processes as well. For this reason, the growth of
inequality can turn acceptable institutions into unacceptable ones
even when this inequality is generated through what otherwise
appear to be innocent means. These considerations seem to me
powerful where they apply, but they argue only for the elimination
of the more extreme forms of inequality. A more rigorously egalita-
rian position might hold that even where neither of the preceding
evils arises (no one is in want, and there is no threat of domination),
inequalities are still objectionable because they are incompatible
with healthy social relations and the development of genuine com-
munity. Putting the matter in terms of the pursuit of a social ideal
seems to rob the demand for equality of some ofits force. It needs to
be explained why this particular ideal is morally important.
This explanation might be sought in a fourth egalitarian position.
A person holding this position would object to inequalities in
life-prospects flowing from differences in family wealth by arguing
that all differences in treatment require justification, and these
differences are undeserved and arbitrary. It is worth noting that
Nozick, while being generally hard on egalitarian claims, allows that
the demand for a justification of inequalities in initial resources
would be valid if these were the result of some centralized
mechanism of distribution (p. 223). He rejects this demand on the
ground that such inequalities do not result from “state action” but
instead flow from the independent actions of many individuals all
 Rights, Liberty, and Property 118

acting within their rights. The results of such a process, he claims,

need no independent justification. I shall return to this point later.
A supporter of this fourth position needs to say something about
why unequal distributions, and not equal ones, require special
justification.4 One reason might be that we recognize it is a distinct
kind of bad thing for a person to be made worse off than others in his
society are. The evil in question here is essentially comparative. It is
not just that it is a bad thing to be at a low level of well-being, nor is
it just that anyone would prefer to be at a certain higher level (the
one where, as it happens, others are). What is bad is being at a lower
level when others around are much better off. (It is worst when the
level others have attained is the norm in your society.) If this kind of
relative disadvantage is a bad thing, then institutions which inflict it
on people require a defense. Such a defense can be given. The
better circumstances of others may be somehow earned, or it may
be impossible to eliminate such differences or too expensive in
terms of other benefits to do so. What is special about equal
distributions is just that they require no defense of this particular
kind.
If the evil of being relatively disadvantaged justifies eliminating
inequalities by redistribution, however, it may be asked whether it
does not provide an equally strong reason for simply worsening the
position of the better off when redistribution is not possible. This
may sound irrational, but in the case of many social inequalities, for
example, distinctions of rank or social caste, egalitarian demands for
the elimination of nonredistributable advantages are not implausi-
ble. In other cases, where we think that nonredistributable advan-
tages should not be eliminated, this is not because these advantages
are consistent with pure egalitarianism but because we temper the
demands of equality with other considerations.> Equality is not our
only concern.
Similarly, it is open to the supporters of any of these egalitarian
positions to recognize that powers to dispose of one’s
possessions—to give them away, to exchange them for others, to
determine what will happen to them after one’s death—are very
important. Indeed in speaking of “distribution” they have always
assumed that to distribute a good to a person is to give him some
powers of this kind over it. Nozick’s examples show that the
114 Thomas Scanlon

interests served by these powers are among those things which must
be weighed against the various considerations supporting equality.
This is something that a realistic egalitarian can accept.
Certainly a theory cannot talk sensibly of patterns of holdings
without considering how these patterns are to be produced and
maintained. If this were all he was claiming in saying that no purely
end-state or patterned theory is tenable, then Nozick would cer-
tainly be right. This would not do much to clear the field, however.
As Nozick rightly points out, philosophical theories of distributive
justice have often neglected the problems of how patterns of dis-
tribution can be established and preserved, but a theory can incor-
porate such considerations, and so avoid being an end-state theory
in this narrow sense of the term without coming close to the position
Nozick favors.
It seems, however, that Nozick’s rejection of end-state theories
encompasses more than the claims I have just endorsed. For he
wishes to reject Rawls’ theory as an unacceptable end-state theory,
despite the fact that it incorporates considerations of entitlement
through the notion of pure procedural justice.® If the basic institu-
tions of a society are just, according to Rawls, then the holdings
people acquire through the operation of those institutions are
legitimate, whatever these holdings may be, and people have rights
over these holdings as the rules of the institution provide. The basic
structure itself is just, according to Rawls, if itconforms to his Two
Principles, namely the principle of maximum equal basic liberties
and the principle that institutions generating unequal holdings are
just only insofar as these inequalities are to the benefit of the worse
off, and only if the positions of greater reward are open to all under
conditions of fair equality of opportunity. Nozick objects to this
theory on the ground that the entitlements it supports have only a
derived status; its fundamental principles, he says, are end-state,
and it is therefore to be rejected. This rejection would also apply to
the modified egalitarian positions I have described.
What is the basis for this strong claim, that any acceptable theory
must make entitlements fundamental? In arguing against Rawls,
Nozick maintains (pp. 199-202) that a theory which brings in
entitlement principles as derived principles to be defended by
appeal to more fundamental moral notions together with empirical
facts will strike us as wrong for the same reason that act-utilitarian
 Rights, Liberty, and Property 15

attempts to account for rights seem so obviously mistaken. What is

derived in such theories will be only approximations of the princi-
ples we intuitively want, and even where they support the same
conclusions as these principles do, they do so for what seem to be
the wrong reasons.
Whether this objection is persuasive against an end-state theory
will depend on the character of the end-states with which that
theory is concerned. Consider, for example, a theory concerned
solely with the production of certain valued states of consciousness,
or one concerned with securing equality in what people physically
possess. It might be claimed within such a theory that certain rights
to dispose of one’s holdings are justified because they are a good
means for producing an end-state of the required kind. Such an
argument would indeed strike us as mistaken for reasons of the kind
Nozick mentions, reasons strictly analogous to those that plague an
act-utilitarian account of the obligation to keep one’s promises or of
the prohibition against paternalism. But a modified egalitarian
theory of the kind I suggested above would not have this problem.
In such a theory, control over various aspects of one’s life is
something which has an independent value. This provides a direct
basis for arguments in support of the personal rights that secure and
protect such control, removing any need to appeal to whatever
tendency these rights may have to promote other, intuitively unre-
lated effects. It is on such a basis that Rawls’ theory recognizes
powers and liberties, including the right to hold personal property,
as primary social goods.7
The value attached to the ability to exercise control over a certain
aspect of one’s life is not the same thing as a right assigning a person
a particular form of such control. (For example, the value attached
to being able to be unobserved is not the same thing as a right which
is designed to secure a certain form of privacy.) But this value is the
natural reason for having such a right, and it is, I think, the element
often missing from utilitarian accounts that seek the value of a right
in its tendency to promote some further unrelated effects. To
recognize a particular interest as meriting protection in rights is not
to say that it is to have absolute protection. It may be a difficult
question how a right can be designed to protect that interest and
how much protection can be given at tolerable cost. Rights of
privacy, for example, represent a strategic attempt to protect our
116 Thomas Scanlon

interest in being free from unwanted observation while not making

life too difficult in other ways.
This general point about the relation between workable rights and
the human interests that make them important has a further rele-
vance to Nozick’s argument. In defending his particular system of
rights, Nozick often seems to assume that any alternative rights
would be wholesale in character. One such right which he con-
siders is “the right to have a say over what affects you” (p. 268).
Nozick properly points out that a right literally to have a say in all
decisions that affect you, or even in all those that affect you deeply
and intimately, would be impossibly broad. It is essential to distin-
guish between different ways in which something can affect a
person. As Nozick puts it, what decision affects me more deeply
than the decision the person I love makes in deciding whom to
marry? But this does not mean that I should have a right to a role in
making that decision. Nozick’s conclusion is that it is my “Lockean”
rights that determine which things, among all those that affect me, I
have the right to a say over.
I would agree that, while the importance of rights largely flows
from the importance of having control over things that affect one,
the function of a system of rights is to distinguish between the
various ways that things can affect people and to apportion out
particular forms of control. It follows that if we are agreed how this is
to be done, then we will refer to people’s rights in order to
determine what they are entitled to a say over. But it does not follow
that Nozick’s “Lockean” rights are the correct ones, and this is just
what is at issue.
If a supposed right turns out to give the person holding it an
obviously unacceptable degree of control over other people's lives
then that is ground for saying that there is no such right. The
proposed “right to have say over what affects you” fails this test. But
what the objection formulated in terms of this right is really claiming
is that unrestricted property rights of the sort favored by Nozick
must also be rejected on the same grounds.
This issue is also raised by an objection which Nozick takes up in
the section entitled “Voluntary Exchange.” This objection maintains
that a situation in which workers accept employment at very low
wages cannot adequately be defended by saying that this is a
voluntary agreement. For the alternatives faced by the workers may
 Rights, Liberty, and Property 17

be so bad that they have no choice but to accept the terms offered
them. Responding to this, Nozick claims that whether limitations on
one’s alternatives undermine the voluntariness of one’s action de-
pends on what these limitations are. First, they must be human
actions, and, second, they must be actions that the agents lacked the
right to perform. He cites the example of someone who marries the
only available person (all the more attractive partners having already
chosen others) as a case of an action that is voluntary despite
removal of all but the least attractive alternative through the legiti-
mate actions of others.
To begin with, voluntariness does not seem to be the relevant
notion here. A person’s action could remain voluntary even if
illegitimate intervention removed the more attractive alternatives.
Perhaps we would say in such a case that he was forced to choose the
lesser of the remaining evils, but the moral significance of this
remark is not clear. Sometimes it is all right to force someone to do
something by making the alternatives unacceptable, and being so
forced does not always invalidate agreements made. It may depend
on who does the forcing. Thus, even if the notions of forcing and
voluntariness sometimes incorporate notions of rights, as Nozick’s
analysis of voluntariness suggests, they certainly do not always do
so. Where they do not, it seems unlikely that we can settle the
question of the moral acceptability of a form of treatment by appeal
to intuitions about forcing and voluntariness. If these notions do
incorporate moral principles, then such appeals to intuition are
going to be suspect when these principles are themselves in dis-
pute, as they are in the present case. Disagreements about these
principles will be translated into conflicting judgments about the
voluntariness of actions and into disagreements in particular cases
over whether “voluntary” is being used in a morally charged way.
The real question at issue in the case at hand is whether it is
justifiable to allow wages to be determined by bargaining under the
conditions here envisaged. It is the connection with justification that
makes plausible Nozick’s restriction of attention to limitations on
alternatives that are brought about by human action. Even though
acts of nature may limit our alternatives, they are not subject to
demands for justification. But individual human actions are not the
only things subject to such demands; we are also concerned with
social institutions that make it possible for agents to do what they
118 Thomas Scanlon

do. The objection that Nozick considers challenges the assignment

of rights which determines the bargaining positions of employers
and workers. It raises the question whether this system of rights
does not protect the liberties of some people in a way which gives
them an unacceptable degree of power over others. This question
cannot be met merely by reaffirming the rights in question. A
further level of argument is invoked in the analogy with marriage.
The suggestion is that any interference with the rights of employers
would be an intolerable intrusion—as a forcible reassignment of
marriage partners would. Serious consideration of such a claim
would bring Nozick’s argument onto the same plane as the objection
he is confronting. To settle the question between them, one would
need an assessment of the relative importance of the various forms
of liberty that are at stake and an account of how these and other
values would be affected by alternative assignments of rights.
It is worth pointing out that the standards of importance
employed in such an argument will be socially relative. Whether
assigning one person the right unilaterally to deny others access to a
certain good (say, university education) gives him a morally
significant or questionable degree of control over them depends on
the role that this good plays in the lives that people lead and aspire
to in that society. This question is not settled by asking whether
people could have done without the good in the state of nature or by
asking how much it is valued by the particular individuals who are
involved in a given case. Something between these two is required;
something less subjective than the latter but more historically
variable than the former. Nozick describes such a standard as the
relevant one for deciding whether a person must be compensated if
others prohibit him from a course of action which creates risk that
their rights will be violated. Compensation is due, he says, if the
actions are of a type that “are generally done, play an important role
in people's lives and are not forbidden to a person without seriously
disadvantaging him” (p. 81). It would seem that this standard should
also be appealed to in defending claims about how far people’s rights
extend, but Nozick seems not to allow such appeals.
In Nozick’s conception, the primary threat to liberty is the
imposition of obligations to which one has not consented. Liberty is
to be safeguarded by keeping such obligations to a minimum,
leaving the greatest: possible scope for voluntary agreements and
 Rights, Liberty, and Property 119

exchange. This concern is evident in Nozick’s rejection of the

Hart-Rawls Principle of Fairness. As stated by Rawls, this principle
holds as follows:

A person is required to do his part as defined by the rules of an institution

when two conditions are met: first the institution is just (or fair), that is, it
satisfies the two principles of justice; and second, one has voluntarily
accepted the benefits of the arrangement or taken advantage of the oppor-
tunities it offers to further one’s interests.?
Against this principle Nozick urges the following example. Suppose
that some of the 364 other people in your neighborhood decide that
it would be nice to have regular programming over the public
address system already installed on your block. They initiate the
practice and post a notice assigning each person a day in the year on
which he is to man the microphone, to play music, discuss
philosophy, tell jokes, or whatever else he wishes to do. The
practice goes on for some time and although you never discuss the
system with your neighbors, you enjoy it greatly. But then your day
comes and you would rather go fishing. Are you obligated to
perform instead? Nozick says clearly not, and in this he may be
right. He also claims that the Hart-Rawls principle would require
you to perform and is therefore mistaken. He goes on to hold that
there is no way to patch up the principle so that it would avoid this
consequence while still generating a nonconsensual obligation to
obey the commands of the state.
It should be said that insofar as this criticism is addressed to
Rawls, it is slightly off the mark. In A Theory of Justice, Rawls
rejects the idea of deriving the obligation to obey the law from
citizen’s receipt of the benefits of government. The Principle of
Fairness is to apply only to cases where the receipt of benefits is
voluntary, and in the case of the state it is usually not. Where there
is a moral requirement to comply with nonvoluntary institutions
such as the state, this requirement derives (via what Rawls calls the
Natural Duty of Justice) merely from the fact that those institutions
are just.
Whichever way we look at it, however, the justice of an institu-
tion is a crucial factor in assessing our obligations to it. Is the
neighborhood broadcasting network described by Nozick a just
institution in the relevant sense? Nozick’s positive answer seems to
120 Thomas Scanlon

be based on a narrow construal of the appropriate standards of

justice. He takes it to be just because the benefits are evenly
distributed and the burdens (days on duty) shared equally. But
there are other considerations than these. For example, there are
considerations of liberty: the power to determine the schedule
(about which nothing is said) and the power to determine who is to
participate (“the organizers” just draw up a list). My intuitive
judgment as to whether there is an obligation to perform is quite
sensitive on both these issues. If we take the organizers’ action as an
official fiat then the obligation seems nonexistent. But if we take
their list merely as a suggestion to which the person in question
might have objected but did not, then the obligation is less prepos-
terous. Similarly, in general, in assessing the justice of an institu-
tion, we must consider the restrictions it imposes on people’s
liberty. We ask whether these restrictions are rationally related to
the justifying purposes of the institution (how much would be lost if
participation were made fully voluntary?) and whether these pur-
poses (and this incremental contribution toward them) are
sufficiently important to justify these restrictions. Arguably the
neighborhood network fails on both counts. Arguably some ac-
tivities of the state beyond those countenanced by Nozick pass.
The contrast between Nozick’s and Rawls’ views on political
obligation illustrates the important difference between two types of
consent theory. In theories of the first type, actual consent has a
fundamental role as the source of legitimacy of social institutions.
Theories of the second type start from the assumption that the
institutions with which political philosophy is concerned are funda-
mentally nonvoluntary. These institutions are held to be legitimate
if they satisfy appropriate conditions, and the idea of hypothetical
consent enters as a metaphorical device used in the formulation and
defense of these conditions. Questions of actual consent arise only as
internal questions ofliberty, that is, as questions about what options
acceptable institutions must leave open to those living under them.
The difference between these two theories is magnified by the fact
that the idea of consent involves choice against some background of
alternatives. If what is at issue is initial consent to institutions from
without, then the relevant background is that of this pre-
institutional condition. It is only this viewpoint that makes the
“baseline” of conditions in the state of nature seem relevant. By
 Rights, Liberty, and Property 121

contrast, since the questions raised through the device of the

hypothetical contract are questions about the justifiability of social
institutions to people who find themselves living under them, the
relevant background is given by the alternatives actually available to
people in societies and the values that such people attach to these
alternatives. This is not to say that the values assigned to various
choices and prospects by people in a given society are always
morally determinative. They may be set aside if they can be shown
to be artifactual in a way that makes them morally suspect. But there
is no temptation, on this view, to take the standards of some earlier
(for example, pretechnological) age as relevant to the acceptability of
contemporary institutions.
The idea that respect for individual liberty requires that consent
be a necessary condition for all obligations beyond the requirements
of a minimal framework of rights arises in the same way as the idea
that makes subjective preference seem the only acceptable basis for
ethically significant judgments of relative well being. Further, the
two views involve similar mistakes. Welfare economists and those
who support subjective versions of utilitarianism are moved by the
belief that the interests of the affected parties are the bases on which
social policies should be appraised and by the belief that it is
unacceptable to “impose” on these parties, as the relevant account
of their interests, a system of values that they do not share. The
response to these beliefs is generally to bring individual preferences
into a theory at the foundational level, making them the basis for all
judgments of relative value. A few restrictions on what can count as
admissible preferences may be allowed in the form of requirements
of consistency, transitivity, and so on, but anything beyond such
purely formal restrictions is seen as a threat. When a theory is
constructed in this way, so that it treats almost all preferences at face
value regardless of their origins or content, its conclusions can be
substantially affected by the social conditions which influence pre-
vailing preferences and their relative strengths. This robs the theory
of an important kind of critical power and, in addition, makes it an
uncertain guardian of even those values of individual autonomy
which it set out to protect. Many different conditions are important
for the development of autonomous preferences, and the ability of
individuals to give effect to their preferences in their own lives and
in the determination of social policy depends on a variety of powers
2 Thomas Scanlon

and liberties. To give appropriate recognition to the value of indi-

vidual autonomy a theory must assign appropriate weights to all
these factors in balancing them against each other and against other
competing considerations. Autonomy is not adequately recognized
simply by letting these weights and all others be determined by
whatever constellation of individual preferences happens to prevail
at the point at which the theory is applied.
Similar problems arise for a view which, acting out of a desire to
safeguard individual liberty, brings consent in at the foundational
level as the basis of almost all obligations, and allows it to be
restricted by only a minimum of “imposed” moral requirements.
The consequences such a theory can endorse are unacceptably open
to determination by factors affecting the relative bargaining strength
of various individuals, for example, variations in the demand for and
scarcity of particular talents and resources. In particular, the ability
of individuals to exercise the kind of control over their lives that
freedom from imposed obligations is supposed to secure will be to
an unacceptable degree merely a function of their bargaining
strength. As in the previous case, the conclusion to be drawn here is
that individual liberty is not adequately protected simply by bring-
ing consent in as the foundation of obligation. An adequate theory
must take into account the various ways—other than merely by
being morally free to withhold one’s services—in which individuals
may be enabled to exercise control over their own lives and their
common institutions (or disabled from doing so.)
Preference-based theories of social welfare and consent-based
theories of obligation can be seen as, respectively, teleological and
deontological responses to similar intuitive ideas. The two are
brought together when utility is taken as the basis of arguments for
the efficiency of the free market. Each also derives support from a
form of skepticism about the existence of an ethically significant,
objective basis for the comparison and balancing of the interests of
different individuals. 1°
The two forms of consent theory correspond to two differing views
of rights. Either view may recognize rights as a basis for individual
claims against social institutions. Thus both see some rights as
“natural” in the sense of having validity that does not derive from
positive law or social institutions. On the first view, however, the
rights that are the basis for moral criticism and defense of social
 Rights, Liberty, and Property 223

institutions are seen as “natural rights” in the stronger sense that

they are the very same rights which individuals possess and can
claim against one another in a state of nature. On the second view,
rights represent general judgments about the conditions of legiti-
macy of social institutions, for example, judgments of the form “Any
institutions granting that power are morally unacceptable.” Exactly
which such generalizations seem true and important—what things
are rights and what these rights encompass—are matters that will
change as social conditions change. Some of these rights concern
things that would be of no relevance, or only a very different and
more limited kind of relevance, in a state of nature. (Rights to
freedom of expression, due process of law and political participation
seem to have this character.)
It is central to Nozick’s argument that the rights with which he is
concerned are claimed to be natural rights in the stronger sense.
The objections I have raised to his examples almost all demand that
he consider the consequences of enforcement of absolute property
and contract rights and that he explain why the loss of liberty this
involves for some people is not worse than that which is involved in
the alternative systems which he deplores. Such objections suppose
that the property rights enforced by the minimal state and those
embodied in socialist institutions are two alternative social systems
open to the same kind of objections and needing the same kind of
defense. Nozick rejects this symmetrical picture. In his view, the
particular property rights protected by the minimal state are not
licensed or created by it and consequently do not need to be
defended as part of its justification. These rights are ones that
individuals have quite independently of the social institutions in
which they live. In enforcing these rights the minimal state is only
doing for them what they were already entitled to do for them-
selves. Consequently it is not doing anything that could be held to
infringe anyone's liberty.
How plausible is the claim that the rights appealed to in Nozick’s
examples are ones that individuals would have in a state of nature?
This claim has greatest initial plausibility with respect to the right of
nonaggression. An unprovoked attack occurring today on the streets
of New York seems to be wrong for the same reasons that would
apply to a similar attack in the state of nature. But the right of
nonaggression as Nozick interprets it covers more than this. It
124 Thomas Scanlon

prohibits generally “sacrificing one person to benefit another” (p.

34). I take it that what Nozick wants to rule out here is any use of
force or the threat of force to make one person contribute to the
welfare of another who has no right to this contribution. This last
qualification reduces the right considerably, but without it the right
would be absurd. This shows that the right of nonaggression cannot
be interpreted in isolation from other rights. Its invariance between
the state of nature and other conditions will consequently depend
on that of these other rights.
Chief among these is the right to one’s property. A system of
property is a set of rules defining the conditions under which a
person owns an object and specifying the extent and character of the
rights of owners. What a person’s property rights are will normally
depend not only on what systems of property could be validly
enforced under the conditions in which he lives but also on what
system is actually in use. To the extent that this system is morally
legitimate, its provisions determine his rights. But the provisions of
this system may also be wrong. They may claim for him rights that
no one could really have or they may fail to protect claims that any
valid system would have to recognize. Surely we can imagine an
incident, occurring in a state of nature, which strikes us intuitively
as a violation of property rights. Imagine that a family is living in the
wilderness when a group of strangers comes along and drives them
off part of their land and takes their crops. This strikes us as a clear
wrong. I take it that the point of saying that this happens “in a state
of nature” is just that the wrongness involved does not seem to
depend on any system of law or social convention. But it is open to
question whether what we feel to be violated in such examples is
really a natural right to property.!! For these cases strikes us as clear
wrongs only if we suppose, first, that what is taken is of use to the
person who loses it (that is, that the taking actually constitutes an
interference with his life and activities) and, second, that his appro-
priation and use of the thing did not already constitute an interfer-
ence with others. (The notion of what constitutes an “interference”
will depend on, but perhaps not be exhausted by, a historically
varying notion of “normal appetites.”) When these conditions are
satisfied, the taking infringes upon what might be called the natural
right of noninterference. A system of property rights goes beyond
this primitive right by specifying formal criteria of ownership. If a
 Rights, Liberty, and Property 125

person is deprived of something to which he has acquired title in the

specified way, then his property right has been violated whether the
taking makes any difference to his life at all. Different systems of
property carry out this extension in different ways, each specifying
its own criteria of ownership and defining and limiting the rights of
owners in its own way. These extended rights require justification
since, as one’s person’s claims to forebearance cease to be limited by
the requirements of a normal life, the justification for these claims
become more attenuated and the threat they present to others
grows more serious.
To support the claim that some property rights are natural rights
we need to think ofa state of nature example involving a clear wrong
which seems to violate one of these rights without violating the
primitive right of noninterference. But if we imagine such a case we
may be open to the question of why we should imagine a state of
nature containing that particular system of property, rather than
some other system which would not be violated by the act in question.
This objection could be avoided if we could show that the
primitive right of noninterference does not exhaust the common
core of systems of property rights. Perhaps there are certain provi-
sions falling outside this right which would be incorporated in any
system of property rights that could plausibly be held to be valid in a
state of nature. It could then be argued that the provisions which
Nozick’s examples turn on fall in this class, for example, that an
unrestricted right of inheritance does so. But this is far from clear.
Suppose the grandfather of the family we previously imagined lived
on land a short distance away and that when he died he said, “Now
this is yours.” But they had all they could do to take care of their own
place, and one day they noticed that someone else had moved onto
their grandfather's old farm. Are they entitled (in the state of nature)
to throw the people off or to demand payment? It is not obvious to
me that they are. Even if there is “as much land and as good” not
far off, their claim to demand that the new people move to it is quite
debatable. Furthermore, even if we were to be convinced by such
examples that any system of property valid in a state of nature would
have to include unrestricted inheritance, there would remain the
question of how much this judgment is dependent upon our assess-
ment of the consequences that this provision would have in a
“natural state.” These consequences are apt to be quite different
126 Thomas Scanlon

from those that would result from the same provision under other
social conditions.
It is of interest here that Locke clearly distinguishes between the
natural property rights that he sees as holding in a state of nature
antecedent to law or social convention and the systems of property
that arise later with the introduction of money and the creation of
government.!2 The system of natural property rights under which
men can acquire title to things by laboring on them is held by Locke
to be valid without consent. It is crucial to his argument for the
validity of these rights that, under the conditions of the state of
nature, the holdings to which people can be expected to acquire
title will not extend beyond “the conveniences of life.”1% They will
not do so because the right itself is restricted by the proviso that
things not be held if they will just go to waste and because the things
men are interested in acquiring in a primitive state are generally “of
short duration.’14 This limit on the extent of holdings is important to
the positive case for the natural right of property since it means that
the things the right protects are needed for a normal life. It also
forestalls objections to the right by providing an important part of
the reason for believing that acquisition under it will not allow one
person to “entrench upon” others but will leave them with “enough
and as good.”!5 Thus, under the conditions Locke believes to hold
in the state of nature, his natural right of property will not
significantly extend what I have called the right of noninterference,
and Locke’s argument for the validity of his right depends upon this
fact.
Once the introduction of money gives men the means to store up,
without spoilage, more than they can use, and commerce gives
them a reason for doing so, there is no longer any reason to expect
holdings to be limited to the conveniences of life. When this
happens, the original moral foundation for property rights is no
longer valid, and a new foundation is required. Locke takes consent
to be this foundation. The “disproportionate and unequal possession
of the earth” which may obtain after money comes into use is
legitimated, according to Locke, by the “tacit and voluntary con-
sent” which men give to the use of money and without which it
would not work.!® Later systems of property founded by positive
legislation derive their authority from the consent men have given
to their governments.
 Rights, Liberty, and Property 127,

Nozick appears to reject both of these latter foundations for

property rights. He avoids invoking a social contract, and he denies
Locke’s empirical claim that a functioning system of money requires
consent, suggesting that it could arise instead through an “invisible
hand process” (p. 18).!7 He faces the problem, then, of deriving an
extended system of property rights involving money, commerce,
and extensive holdings from something like Locke’s original
“natural” foundation. This derivation faces two problems. The first
is that the lack of natural bounds on acquisition means that others
are likely to be threatened—there may not be enough and as good
left for them. As I have already mentioned, Nozick’s response here
is that the increase in the stock of goods due to increasing pro-
ductivity will keep pace with increased acquisition, making it unlikely
that anyone will be made worse off relative to the baseline of
expectations in the state of nature. The second problem is that, with
holdings extending far beyond “the conveniences of life” (certainly
far beyond what these included in the state of nature), the case for
absolute protection of these holdings becomes weaker. This makes
even more controversial the choice of an extremely low baseline for
determining whether the condition of others is worsened.
In closing, let me mention two important problems to which
Nozick’s book calls attention. It is a virtue of the book that it forces
us to consider economic institutions not merely as mechanisms for
the distribution of goods but also, like political institutions, as
placing restrictions and demands on us which raise questions of
obligation. When things are seen in this way it becomes apparent
that questions of economic liberty must be considered, along with
political and civil liberty and fair distribution, as conditions for the
legitimacy of social institutions. I hope that this will have an impact
on contemporary moral and political philosophy, where economic
rights and liberties have generally been neglected in favor of
political philosophy, where economic rights and liberties have
generally been neglected in favor of political and civil liberties and
rights and liberties have generally been neglected in favor of
political and civil liberties and rights of other sorts. I have argued
that the particular framework of property and contract rights which
Nozick proposes does not constitute an adequate account of the
claims of economic liberty. In opposition to this framework I have
appealed vaguely to the value of having control over various aspects
128 Thomas Scanlon

of one’s life, and I have asserted that a workable system of rights

could be developed that would secure this value, and others, more
adequately than the rights Nozick advocates. What is needed,
however, is a systematic account of the relevant forms of liberty
and control and of the values associated with them.
The idea that rights constitute constraints on the pursuit of the
maximum social good (p. 166) has great appeal as a way of avoiding
the objectionable consequences of utilitarianism. I have not been
willing to accept the particular system of rights which Nozick
opposes to utilitarianism, however, and I have complained that
these rights require defense. But this demand for justification, a
demand that any alternative conception of rights must also satisfy,
seems to lead inevitably back to an appeal to consequences, and to
the balancing of individual benefits and burdens. I have certainly
made heavy use of such balancing in my objections to Nozick’s
views. The problem remains, then, of giving a satisfactory account of
what this balancing involves and how it is relevant. This account
must avoid falling back into an objectionable utilitarianism without
simply invoking another set of a priori rights.

NOTES

*T am grateful to Dennis Thompson for helpful comments. NEH Fellow-

ship support is also gratefully acknowledged.
1. (New York, 1974). Page numbers in the text refer to this book.
2. Worse off, that is, in what they are able to use; it is not enough that
they be worse off with respect to what remains available for initial appropri-
ation. See p. 178.
3. Thus, for example, W. D. Ross says: “The essential defect of
the ‘ideal
utilitarian’ theory is that it ignores, or at least does not do full justice to, the
highly personal character of duty. If the only duty is to produce the
maximum of good, the question who is to have the good—whether it is
myself, or my benefactor, or a person to whom I have made a promise to
confer that good on him, or a mere fellow man to whom I stand in no such
special relation—should make no difference to my having a duty to produce
that good. But we are all in fact sure that it makes a vast difference.” The
Right and the Good (Oxford, 1930), p. 22.
4. This question is raised by Thomas Nagel in his review of Nozick’s
book. See p. 148 of his “Libertarianism Without Foundations,” The Yale
Law Journal 85 (1975). The hypothesis I go on to discuss is consistent with
 Rights, Liberty, and Property 129

Nozick’s remark (p. 210) that an egalitarian sees inequalities as involving a

cost.
5. For such a view of equality see Christopher Ake, “Justice as Equal-
ity,” Philosophy & Public Affairs 5, no. 1 (Fall 1975): 69-89.
6. See John Rawls, A Theory of Justice (Cambridge, Mass., 1971),
pp. 83-88, 304.
7. Ibid., pp. 61, 62, 92.
8. For a general form of this objection, see Nozick, p. 238.
9. A Theory of Justice, pp. 111-112.
10. In Nozick’s case the skepticism concerns the ethical significance of
such balancing, not its epistemological basis. For a thoroughgoing rejection
of balancing see pp. 32-33. This extreme position is criticised by Nagel in
his review cited in fn. 4 above. A more moderate position, according to
which rights set the limits of permissible balancing, is suggested by Nozick
on p. 166.
11. The following argument is drawn from my article, “Liberty, Contract
and Contribution,” to appear in G. Dworkin, G. Bermant and P. Brown,
eds., Markets and Morals.
12. Second Treatise of Government, §50.
13. Ibid., §36.
14. Ibid., §46.
15. Ibid., §36.
16. Ibid., §50.
17. Nozick does not discuss the normative purpose of Locke’s claim, and
it is not clear that Locke’s argument is affected by his objection. Locke uses
the word “agreement,” but he does not clearly assert, and does not need to
claim, that an explicit agreement is needed to establish a system of money.
All he needs to claim is that the continued functioning of such a system
involves everyone’s tacit consent.
 i
Some Ruminations on
Rights
JUDITH JARVIS THOMSON

In Anarchy, State, and Utopia, Robert Nozick says that a govern-

ment which imposes taxes for the purpose of redistribution violates
the rights of its citizens.1 The word “imposes” perhaps needs no
stress: Nozick could hardly object to a government's withholding a
percentage of income for this purpose if its citizens had unanimously
requested it to do so. What he objects to—on the ground of its
constituting a violation of rights—is forcing payment for this pur-
pose on those who do not wish to pay. What we might expect Nozick
to give us, then, is a theory of rights, or at least a clear picture of
why this should be so. In fact, we get neither.
Nozick makes two quite general points about rights, both of them
important. He says, first, that the fact that if we bring about that
such and such is the case there will be more good in the world than
there otherwise would be does not by itself justify our bringing
about that it is the case, and this on the ground that to bring it about
may be to violate a right. This seems to me to be wholly right.
Suppose, for example, that if we bring about that Alfred takes a
certain aspirin tablet there will be more good in the world than there
otherwise would be. This does not by itself justify our bringing
about that Alfred takes it, for it might be that to do so would be to

130
 Some Ruminations on Rights 131

violate a right. For example, it might be that Bert owns that aspirin
tablet and does not wish Alfred to take it; in that case, to bring about
that Alfred takes it would be to violate a right of Bert’s. Indeed, it
might be that Alfred himself owns it but does not wish to take it; in
that case, to bring about that he does would be to violate, paternalis-
tically, a right of Alfred’s.
This point, though important, is familiar enough. What is perhaps
less familiar is Nozick’s second point: That the fact that if we bring
about that such and such is the case there will be more good in the
world than there otherwise would be does not by itself justify our
bringing about that it is the case—even if we require that in
assessing how much good there will be in the world account be
taken of which rights, if any, will be infringed and of how “stringent”
those rights are. This point too seems to me to be wholly right. If we
do opt for this requirement on an assessment of how much good
there will be in the world, then it seems to me we may suppose that
if we bring about that Alfred takes a certain aspirin tablet there will
not be more good in the world than there otherwise would be,
however bad Alfred’s headache may be: for there would have to be
considered in arriving at the assessment, not merely the fact that if
we bring about that Alfred takes the aspirin his headache will go
away, but also (as it might be) the fact that a right of Bert’s will be
infringed, or (as it might be) the fact that a right of Alfred’s will be
infringed. If so, this is not really a case in which, even though there
will be more good in the world if we act than there otherwise would
be, it is not morally permissible for us to act. But there are other
cases. Suppose that a villain threatens to kill five people if you will
not kill Charles. Even prima facie it seems that if you act, there will
be more good in the world than there otherwise would be since five
lives are four more than one life. And now let us include in our
assessment infringements of rights. If you act, fewer rights will be
infringed than if you do not, for five violations of the right to not be
killed are four more than one violation of the right to not be killed.
Therefore, if we require that in assessing how much good there will
be in the world account be taken of which rights, if any, will be
infringed and of how stringent those rights are, this is a case in
which there will be more good in the world if you act than if you do
not. Yet you surely cannot act, since you surely cannot kill in
response to such a threat.
132 Judith Jarvis Thomson

This kind of case has been appearing fairly often in recent

literature.2 The kind of case I mean is this: For the agent to act
would require him to infringe a right, but he is under threat that if
he does not act, others will infringe more, equally stringent rights.
Most people agree that the agent in such a case cannot act. What is
particularly good in Nozick’s treatment of these matters is the
connection he makes between cases of this kind on the one hand,
and the case of Alfred on the other hand. Nozick’s discussion brings
out that if a utilitarian saves his theory in face of putative counter-
cases such as that of Alfred by claiming that right-infringements
themselves have disvalue, which disvalue must be counted in
assessing how much good there will be in the world if the agent acts,
he thereby ensures that cases where the agent must infringe a right
to avoid greater right infringement on the part of others will be
counter-cases.
As I say, I think these points are wholly correct. Nozick does not
argue for them; nor shall I. But to have arrived here is to be miles
away from Nozick’s thesis about government and taxation for the
purpose of redistribution. What we have so far is that the fact that if
we bring about that such and such is the case there will be more
good in the world than there otherwise would be does not by itself
justify our bringing about that it is the case. Thus suppose redis-
tribution is, in one way or another, a good, and that if we make a
certain redistributive move there will in fact be more good in the
world than there otherwise would be. What we have is that that fact
does not by itself justify our making that redistributive move.
However, this leaves it wide open that something which
includes—or even something entirely other than—that fact does
justify our making it.
Let us begin with a point of terminology. Suppose that someone
has a right that such and such shall not be the case. I shall say that
we infringe a right of his if and only if we bring about that it is the
case.° J shall say that we violate a right of his if and only if both we
bring about that it is the case and we act wrongly in so doing. The
difference I have in mind comes out in the following case, which I
shall call A:

(A) There is a child who will die if he is not given some drug in
the near future. The only bit of that drug which can be ob-
 Some Ruminations on Rights 133

tained for him in the near future is yours. You are out of town,
and hence cannot be asked for consent within the available
time. You keep your supply of the drug in a locked box on
your back porch.
In this case the box is yours, you have a right that it not be broken
into without your consent; since the drug is yours, you have a right
that it shall not be removed and given to someone without your
consent. So if we break into the box, remove the drug, and feed it to
the child, we thereby infringe a number ofrights of yours. But I take
it that a child’s life being at stake, we do not act wrongly if we go
ahead; that is, though we infringe a number of your rights, we
violate none of them.
It might be said that we do violate one or more ofyour rights if we
go ahead, but that our act, though wrongful, is excusable. In other
words, although we act wrongly if we go ahead, we are not to be
blamed for doing so. It is true that for clarity about rights we need,
and do not have, “a general account of when one should say “a
nonwrongful infringement of a right” and when one should instead
say ‘a wrongful, but excusable, infringement ofa right.” I think (but
without great confidence) that the difference lies in this: The former
may not be said where, and the latter may only be said where the
agent ought not act or ought not have acted. If so, then the proposal
we are considering is false: For it surely is plain that a third party
would not speak truly if he said to us, given we are in (A): “You
ought not go ahead.”
In any case, the proposal in a certain sense hangs in mid-air. What
I have in mind is this. It is presumably agreed universally that if we
go ahead in(A), we are not to be blamed, punished, scolded, or the
like, for doing so. Now the question is: Why? One possible answer
is: If we go ahead in (A) we do not act wrongly, and that is why we
are not to be blamed for doing so. That this is my answer shows itself
in the paragraph in which I first set out (A). But how is a proponent
of the proposal we are now considering to answer? On his view, we
act wrongly if we go ahead; what, on his view, is the reason why we
are not to be blamed for doing so? There are cases in which there is
an answer to an analogous question. Thus if I break your box in a
rage which you provoked, then I acted wrongly, but perhaps
excusably, and the reason why I am not to be blamed (if I am not) is
134 Judith Jarvis Thomson

at hand: you yourself provoked the rage out of which I acted. Again,
a reason why I am not to be blamed in another case might be: I was
not fully aware of what I was doing; or I was so frantic with worry I
could not think clearly; or I was so frantic with worry, nothing else
seemed to matter. If (A) had read: “Our child will die if he is not
given ...,” then there might have been a toehold for an answer of
the kind just pointed to. But (A) says: “There is a child who will die
ifhe is not given . . .”; and it is possible to suppose that we go ahead
in (A)—break the box, and give the drug to the child—calmly,
coolly, carefully weighing all the relevant considerations. If so, just
what is a proponent of the proposal we are now considering to give
as an answer to the question of why we are not to be blamed for
doing so?
So I shall simply assume that this proposal is false, and I shall take
it, then, that while we infringe some of your rights if we go ahead,
we do not violate them.4
A second way of responding to what I said of our act if we go ahead
in (A) is this: True, we violate no rights if we go ahead, but we also
infringe no rights if we go ahead. What I have in mind is the
possibility of saying that you do not have either of the rights it might
have been thought you had—that you do not have a right that your
box not be broken into without your consent, and that you do not
have a right that your drug not be removed and given to someone
without consent—on the ground that it is morally permissible for us
to go ahead in (A). What rights do you have over your box and drug
on this view? Well, I suppose it would be said that what you have is
at most a right that your-box-not-be-broken-into-and-your-drug-
not-taken-without-your-consent-when-there-is-no-child-who-needs-
that-drug-for-life. The inclination to take, everywhere, either the
view discussed just above, or the view indicated here, is the inclina-
tion to take it that if a man has a right that such and such shall
not be the case, then if we bring about that it is the case, we act
wrongly in so doing. As the point might be put, every infringing of
a right is a violation of a right. So if a man really does have a right
that such and such shall not be the case (as it might be, that his
drug not be removed from his box), then we act at best excusably
if we bring about that it is the case—as in the view discussed
just above. If we do not act wrongly in bringing it about, then he
did not really have a right that it not come about, but at most a right
 Some Ruminations on Rights 135

that it-not-come-about-when-the-circumstances-are-so-and-so—as
in the view indicated here.
It seems to me, however, that you do have a right that your box
not be broken into without your consent and a right that your drug
not be removed and given to someone without your consent, and
that what shows this is the fact that if we go ahead in (A)—break into
your box and give some of the drug to the child—we shall have later
to pay you some, if not all, of the cost we imposed on you by doing
so. We shall have to pay some, if not all, of the cost of repairing or
replacing the box and of replacing the drug we removed.*® You may
reject payment: you may say, on your return, that, the cir-
cumstances having been what they were, all is well, and that you do
not mind bearing the costs yourself. But we must at least offer. If
you had no right that we not do these things without your consent,
why would we have to pay you some of the costs we imposed on you
by doing them?
It is sometimes said® that if we go ahead in (A) we shall have to
compensate you for the costs we imposed on you by doing so, and
that that is what shows that we infringed some of your rights by
going ahead—for compensation is repayment for a wrong. But I
think that this is not a good way to put the point, and will bring out
my reason for thinking so below.7
In any case, it seems to me we do well to agree that rights are not
all absolute: There are rights which can be infringed without being
violated. In particular, it seems to me that if we go ahead in (A), we
infringe some of your property rights, but do not violate any of
them.
What people who would agree with me on this matter would say
is this: If we go ahead in (A), we will infringe your property rights,
but we would not violate them, since those rights are “overridden”
by the fact that the child will die if we do not go ahead.
A more stringent right than your property-rights over your box
and drug might not have been overridden by this fact. For example,
if it had been necessary for the saving of the child’s life that we kill
you, then it would not have been morally permissible that we go
ahead. Your right to not be killed is considerably more stringent
than any of your property rights, and would not have been overrid-
den by the child’s need.
The question just how stringent our several rights are is obviously
136 Judith Jarvis Thomson

a difficult one. It does not even seem to be obvious that there is any
such thing as the degree of stringency of any given right. Perhaps a
right may be more or less stringent, as the rightholder’s cir-
cumstances vary, and also, in the case of special rights, as the means
by which he acquired the right vary. One thing only is plain: Only
an absolute right is infinitely stringent. For only an absolute right is
such that every possible infringement of it is a violation of it.
Indeed, we may re-express the thesis that all rights are absolute as
follows: all rights are infinitely stringent.
There are passages in Anarchy, State and Utopia which suggest
that Nozick thinks all rights are infinitely stringent. He says: “[O]ne
might place [rights] as side constraints upon the actions to be done:
don’t violate constraints C. The rights of others determine the
constraints upon your actions. . . . The side-constraint view forbids
you to violate these moral constraints in the pursuit of your
goals... .”® If you use “violate” in the way I suggested we should
use it, this “side-constraint view” does not amount to much—under
that reading of the term, all Nozick says is that we may not wrongly
infringe a right. Of course we may not. But I think he does not mean
so to use the term “violate”, in this passage at any rate: I think that
in this passage all he means by it is “infringe.” Thus I think that we
are to take this “side-constraint view” to say that we may not ever
infringe a right. Accordingly, every infringing of a right is wrong.
Compare what Nozick says a few pages on:

A specific side constraint upon action toward others expresses the fact that
others may not be used in the specific ways the side constraint excludes.
Side constraints express the inviolability of others, in the ways they specify.
These modes of inviolability are expressed by the following injunction:
“Don't use people in specified ways.”®

Now Nozick does not in fact say that his view is the “side-constraint
view,” so interpreted, but he implies that it is. Certainly his thesis
about redistribution suggests it: for according to that thesis it is not
morally permissible to tax people for the purpose of redistribution,
however dire the human need which makes redistribution seem
called for, and if dire human need does not override a right, what on
earth would?
There are also passages which suggest that Nozick thinks that
rights may be overrideable, and thus not infinitely stringent, though
 Some Ruminations on Rights 137

very stringent all the same. He says that it is an open question

“whether these side constraints are absolute, or whether they may
be violated in order to avoid catastrophic moral horror.”!° Catas-
trophic moral horror is pretty horrible moral horror; so even if rights
are overrideable, as the passage suggests is possible, it is likely to be
a rare occasion on which they are overridden. Unfortunately,
Nozick leaves the question unanswered; he says it “is one I hope
largely to avoid.”
There are also passages which suggest that Nozick thinks that
some rights at least are overrideable even where catastrophic moral
horror is not in the offing. In the course of a discussion of what may
be done to animals, he asks: “Can’t one save 10,000 animals from
excruciating suffering by inflicting some slight discomfort on a
person who did not cause the animals’ suffering?”!2 And he adds:
“One may feel the side constraint is not absolute when it is people
who can be saved from excruciating suffering. So perhaps the side
constraint also relaxes, though not as much, when animals’ suffering
is at stake.”18 Of course Nozick does not say the side constraint
relaxes when animals’ suffering is at stake, but he seems to think so,
and it would surely be mad to think it did not. Well, perhaps 10,000
animals suffering excruciating pain counts as catastrophic moral
horror. But does it require 10,000 of them, in excruciating pain, to
override your right to not be caused some slight discomfort? I take it
you have a right to not be pinched without your consent. But surely
we can pinch you without your consent, if doing so is required to
save even one cow from excruciating suffering. Indeed I should have
thought we could do so if doing so is required to save just one cow
from suffering which is considerably less than excruciating.
This wobbling about the degree of stringency of rights makes a
reader feel nervous. It also makes it very unclear just how Nozick is
to get from his starting point, which is that we have rights, to his
thesis that a government which imposes taxes for the’ purpose of
redistribution violates the rights of its citizens. I am inclined to
think that what happens is this: At the outset, he is unclear what
degree of stringency should be assigned to rights (and hopes to avoid
having to take a stand on the matter), but by the time he gets to
government, all is forgotten, and rights—at any rate, property
rights—are infinitely stringent. It is my impression that his argu-
ment for his thesis rests entirely on the supposition that they are.
138 Judith Jarvis Thomson

But surely it is plain as day that property rights are not infinitely
stringent. I suppose it hardly needs argument to show they are not.
In any case, the fact that it is morally permissible for us to go ahead
in (A) would show—if it needed showing—that they are not.
Consider, now, a case (B) which is in an interesting way different
from (A):
(B) There is a child who will die if he is not given some drugs in
the future. The only bit of that drug which can be obtained for
him in the near future is yours. You are out of town, so we
telephone you to ask. You refuse consent. You keep your
supply of the drug in a locked box on your back porch.
“They did it without Jones’ consent” covers two interestingly differ-
ent kinds of cases: In the one kind, they were unable to get Jones’
consent because he was not available to be asked for his consent; in
the other kind, they were unable to get Jones’ consent because he
refused to give it. In the latter kind of case they acted, not merely
without Jones’ consent, but against his wishes. (A) is a case of the
first kind; we cannot reach you to ask for consent. (B) is a case of the
second kind; if we go ahead in (B) we act, not merely without your
consent, but against your wishes. I said it is morally permissible for
us to go ahead in (A); Is it morally permissible for us to go ahead in
(B)P
The fact is that our going ahead in (B)—our breaking into the box
and removing the drug to give it to the child—seems morally
suspect in a way in which our going ahead in (A) does not. Why?
And should it?
Anyone who thinks that it is morally permissible for us to go
ahead in (A) but not in (B) must think that there is at least a good
chance that in (A), you would give consent if we were able to reach
you to ask for consent. Surely if it were known that if we were to ask
for consent in (A) you would refuse to give it, then it would be no
better to go ahead in (A) than it is to go ahead in (B). For then (A)
too would be a case in which going ahead would be acting against
your wishes—though not against any wish that was in fact given
expression.
Anyone who thinks that it is not morally permissible for us to go
ahead in (B) must think that the box and the drug in it are in some
way very important to you—that you place a very high value on the
box not being broken into, and on the drug not being taken away
 Some Ruminations on Rights 139

from you. Suppose, however, that there is a toothpick on your desk,

and it is in no way special to you. By virtue of some peculiarity in
nature, we can save a life if we snap it in two. We ask if we can, but
you are feeling refractory and say “No.” Can we not go ahead and
snap it in two, despite your expressed wish that we not do so? By
contrast, suppose what is on your desk is the last remaining photo-
graph of your dead mother, and what we need to do to save the life
is to burn it. Well, some people would say we can go ahead all the
same. Suppose that what we need to do is to destroy all the now
existing beautiful works of art, and that their owners (individuals,
museums, governments) say, “Alas no, we are very sorry, but no.”
Could we go ahead all the same?
If (X), “The box and drug are, at most, of little value to you,” is
true, then we may surely go ahead in both (A) and (B). If(X) is true
and we are in (A), then in the absence of information to the
contrary, we shall rightly assume you would consent if we were able
to ask. But even if we have information to the effect that you would
not consent—even if we were in (B) instead of (A)—it is morally
permissible for us to go ahead all the same. Why? Because if (X) is
true, then it would be indecent for you to refuse consent in (A), and
it is indecent for you to refuse consent in (B). I said you might be
feeling refractory; alternatively, you might think: “What is that child
to me?” There are other possible sources of refusal, but none of
them bears looking at.
What if, instead, (Y), “The box and drug are of immense value to
you, is true? Some would say we can go ahead all the same. I feel
considerable sympathy for this view, but I do not hold it myself. It
seems to me that if (Y) is true, we may not go ahead in (B), and in
the absence of reason to think you would consent despite the truth
of (Y), we may not go ahead in (A) either. I hope that when I first
produced (A) above, your intuition agreed with mine; if so, I think
that was because you were assuming that nothing so strong as (Y)
was true. Why may we not go ahead if (Y) is true? It is not morally
splendid to value bits of property more than human lives; but if
there are some which you do—and this for no morally suspect
reason—then it seems to me that there are cases, and that this is one
of them, in which we must withdraw.
There are all manner ofpossibilities between (X) and (Y), but it is
not necessary for our purposes that we attend to them.
It is also not necessary for our purposes that we attend to a very
140 Judith Jarvis Thomson

interesting question which is raised by consideration of the differ-

ence (which I take there to be) between what we may do if (X) is
true and what we may do if(Y) is true, but I suggest we have a brief
look at it all the same. What I have in mind is the question in
precisely what way the difference between (X) and (Y) makes such a
difference. One way of explaining it is this: If (Y) is true, then your
rights that your box not be broken and drug not be taken are more
stringent than they would be if(X) were true. More generally, that
(T) The stringency ofA’s right that x not be broken and y not
taken away from him varies with the degree to which he
values x’s not being broken and y’s not being taken away from
him.
If so, then more is required to override your rights over your box
and drug if(Y) is true than is required to override them if(X) is true.
In particular, the fact that a human life may be saved by going ahead
overrides your rights if (X) is true, but not if(Y) is true.
I think, myself, that this is how we should explain the difference
(which I take there to be) between what we may do if (X) is true and
what we may do if(Y) is true. Indeed, I think we should adopt (T). 14
But the question whether or not (T) is true is very important for the
logic of rights; and so it should be noticed that there is yet another
way of explaining the difference even if (T) is rejected. What might
be said is this: The stringency of your rights that your box not be
broken and your drug not be taken is no greater whether (Y) is true
or (X) is true; and these rights are overridden by the fact that a
human life may be saved by going ahead. But if (Y) is true, then it is
less likely, perhaps even impossible, that we are going to be able to
reimburse you for all of the costs we impose on you by going ahead;
and if we take “immense” very seriously, it is less likely, perhaps
even impossible, that we are even going to be able to pay you a
meaningful part of those costs. Now it will be remembered that I
said earlier that if we go ahead in (A) we are going to have to pay you
some, if not all, of the costs we impose on you by going ahead. This
means that you have a right, not merely that your-box-not-be-
broken-and-drug-taken-without-consent, but also that your-box-not-
be-broken-and-drug-taken-without-consent-without-reimburse-
ment-for-some-if-not-all-of-the-costs-imposed-by-the-breaking-and-
taking. The former, simpler right is overridden by the fact that a
 Some Ruminations on Rights 141

human life may be saved by going ahead; the latter, more complex,
right is more stringent, and is not overridden by this fact—indeed,
it would be violated if we went ahead without reimbursing you. If
(X) is true we can easily make the required payment; but if (Y) is
true we cannot. So if (X) is true we may go ahead without violating
any right of yours, for we can pay later; but if (Y) is true, then if we
go ahead we shall violate, not the simpler right, but the more
complex one, for we cannot pay later. And that is why we may go
ahead if (X) is true, but not if (Y) is.
I have no objection to the supposition that you do have this more
complex right as well as the simpler one. And I imagine that it is
more stringent than the simpler one.!> But, as I said, I think we
should adopt (T), and if we do, we can explain the difference in the
simpler manner I pointed to earlier. If (T) is not true, then the
stringency of a right is independent of the value the rightholder
places on its being accorded to him, and that makes the source of
rights very dark indeed. If(T) is true, then we can understand why
one’s right to life is more stringent than one’s right to not have, for
example, an arm broken, and why one’s right to not have an arm
broken is more stringent than most of one’s property rights; if(T) is
not true, it is obscure why this should be so. The truth of(T) is just
what you would expect if rights issue from interests in some way or
another. And if they do not issue from interests, what on earth do
they issue from?
However, this is no argument for (T). Fortunately it is not
necessary for our purposes that we decide on the truth or falsity of
(T). It is enough for our purposes that if (X) is true, then we may go
ahead in (B) as well as in (A). For with that in hand we are in a
position to return to Nozick’s thesis that a government which
imposes taxes for the purpose of redistribution violates the rights of
its citizens.
I said that it is my impression that Nozick’s argument for this
thesis rests entirely on the supposition that property rights are
infinitely stringent, and I said also that it is plain as day that they are
not. Well, setting aside Nozick’s argument for the thesis, what about
the thesis itself?
The rights which Nozick thinks would be violated by a redistribu-
tive move are property rights. I shall make no criticism here of his
account of the source and content of those rights. However, it is
142 Judith Jarvis Thomson

perhaps worth just drawing attention to the fact that Nozick allows
that title to property is clouded in existing states: He grants that
injustices lie behind their current property distributions.'* This
means, then, that a redistributive move in an existing state may very
well not really conflict with property rights, and in fact there is no
practical moral lesson about redistributive moves in existing state to
be learned from Nozick’s book.
In light of that fact we had better take Nozick to be speaking only
of governments in “ideal” states—states in which property rights are
not clouded; more precisely, states in which the distribution of
property satisfies Nozick’s principles of distributive justice.
One thing we know is that there are circumstances in which it is
morally permissible, and hence no violation of any right, to take
from Smith—even against his wishes—to give to Jones. Any case in
which Jones needs something, and he needs it for life, and the only
way of providing him with it is by taking it from Smith, and Smith
places at most little value on it, is such a case. Suppose we live in an
ideal state. Then there are circumstances in which agents of gov-
ernment can arrange this redistribution. Would that count as im-
posing a tax for the purpose of redistribution? It is hard to see why
not.
Something of great interest comes out if we consider, now, a
second kind of case. Suppose there is an “ideal” state of only eleven
people. One person will die if he is not provided with a certain
amount of a particular drug. Eight of the remaining ten people
would very much like for him to get that amount of that drug. (I
make it a large majority, though I have no very clear idea how its
being a majority matters. I also made the sick one be a citizen of the
state, though I have no very clear idea how his being so matters.)
The eight can scrape together the needed amount of the drug from
among their own supplies, but to do so would require each of them
to deplete his supply drastically—not to the point at which any of
their lives is at risk, but to the point at which they would have a bare
sufficiency. By contrast, the remaining two people have ample
supplies; each of them could, himself, easily supply the needed
amount. But these two individuals refuse to contribute.
This case is different from (A) and (B): In this case, by contrast
with those we have been looking at, the agents do not have to take
anything from anyone else in order to meet the need of the
 Some Ruminations on Rights 143

eleventh. They can meet his need themselves. Is it the case that if
they want his need met, they must meet it themselves? On Nozick’s
view they must. On Nozick’s view, the meeting of human needs is a
consumer good like any other. Or rather like any other expensive
consumer good.17 If you want a color television set, and buying one
will deplete your assets to the point at which you have a bare
sufficiency to live on, well, so be it, it is up to you whether or not a
color television is worth that much to you. You certainly cannot take
from anyone else in order to be able to buy one without having to
deplete your assets! Similarly for the meeting of human needs.
It is plain enough, however, that the meeting of human needs is
not a consumer good like any other. I hasten to say I have no
account of what marks needs off from mere wants. But certainly if a
man will die unless he gets something, then that thing is something
he needs. And we know that if we cannot provide him with that
thing which he needs for life without taking from Smith, then—at
least in such cases as Smith places at most little value on it—it is
permissible for us to take from Smith. This marks a difference. For
even if you cannot get a color television at all unless you take from
someone else, then all the same you cannot take from him in order
to buy one, even if he has plenty of money.
But is this difference relevant to the case at hand? Suppose
Nozick were to grant it, and say: “Very well, the meeting of human
needs is not a consumer good like any other—it differs from color
televisions in the way you indicate. [He would thus acknowledge
that property rights are not infinitely stringent.] Still, if the eight
can meet the need of the eleventh by themselves, how can they
presume to take from the two who do not care if the need is met?”
Nozick might go on: “In those cases you have been describing in
which Jones needs something for life, and it is permissible to take it
from Smith and give it to Jones, what overrides Smith’s right that
the thing not be taken from him is not the mere fact that Jones needs
it for life, but the complex fact that Jones needs it for life and we can
provide it in no other way than by taking it from Smith. Suppose
what Jones needs for life is a drug which you have ample supplies of
and Smith has only a little of; surely you cannot say: ‘How nice! The
fact that Jones needs that drug for life overrides Smith’s right that
his drug not be taken away from him, so I do not have to provide for
Jones myself—I can take from Smith to provide for Jones.’ Surely
144 Judith Jarvis Thomson

you cannot take from Smith if you have plenty yourself! But if it is
the complex fact I pointed to which is doing the real work in the
cases you describe—if it is that fact which really does the
overriding—then those cases have no bearing at all on the case now
at hand. So I repeat: Given the eight can meet the need of the
eleventh by themselves, how can they presume to take from the two
who do not care if the need is met?”
Nozick might go on: “And wouldn't it be like that in ideal states
generally? In other words, that those who refuse to contribute
would be few enough so that those willing to contribute could, by
themselves, meet such needs as they wanted met?!® If so, nothing
so far said counts against my thesis that a government of an ideal
state which imposes taxes for the purpose of redistribution violates
the rights of its citizens.”
It is hard to know what to say about people who would live in
“ideal” states if there were any. What would they be like? But I join
the many other readers of Anarchy, State, and Utopia who have
their doubts.!%
Moreover, the instability of the situation I invited you to imagine
is obvious. Suppose that if only one of the eight ceased to be willing
to contribute, then the remaining seven could no longer meet the
need of the eleventh by themselves, so that the case would then
collapse into a case of the kind we were looking at earlier. Would it
not pay them to draw straws to choose one among them to volunteer
to say he has changed his mind? Then, instead of the eight having to
deplete their own supplies of the drug, the remaining seven could
take from the two who are rich in it. Would they even need to draw
straws to choose a liar? If the eight were given the information that if
there were only seven, the seven could take from the two, would
there not be at least one who would really change his mind? It
would be an odd moral theory that yielded either the conclusion
that the eight must not be given that information, or the conclusion
that the eight must meet the need of the eleventh by themselves
unless they are lucky enough to get that information, in which case
they do not have to.
All the same, the question my hypothetical Nozick raises is a hard
one. If the eight can meet the need of the eleventh by themselves,
how can they presume to take from the two who do not care if the
 Some Ruminations on Rights 145

need is met? I am sure that the instability I pointed to should figure

in the answer, but I do not see clearly how.
One's intuition, I think, is that it just is not fair that the eight
should have to deplete their supplies so drastically in order to meet
the need of the eleventh. The source of that intuition is, I think,
this: One thinks of the need as having to be met by the citizens of
that state, and therefore thinks that the burden of meeting it should
be shared, as in the case of any other project which the citizens have
to carry out.
Why does the need have to be met by the citizens of the state? By
hypothesis, the need is one which can be met by them at little cost
to any of them, for each of the two with ample supplies could easily
meet the need by himself. But if a need can be met at little
cost—remembering that it is a need for something to sustain life
itself—then it is indecent that the need not be met. (I here say
something of a community which would be true of an individual.) So
it has to be met. So, as in the case of any other project which the
citizens of a state have to carry out, it is only fair that the burden of
doing so be shared. But if the two with ample supplies give nothing
at all, the entire burden falls on the remaining eight, who can least
afford to share it. Hence it is not fair that it should fall on them
alone.
If the two with ample supplies can each meet the need at very
little cost, then it makes little difference whether or not one takes
the whole amount needed from one, or takes half the needed
amount from each, or imposes a proportional tax on all ten of them,
under which the two pay the lion’s share, and the remaining eight
pay a grain or two each. Another possibility is that each of the two
might be ordered to provide half, and the remaining eight suffer a
comparable loss by having to pay the two, or the community at
large, in some commodity other than the drug.
There are cases, however, in which it will make a difference. Let
us look back again at case (A). I said that if we go ahead, and break
into the box and give the drug to the child, we shall have later to pay
you some, if not all, of the cost we imposed on you by doing so.
Kindhearted students sometimes look askance at this proposal—for
ifwe go ahead, we do so to save the life ofa child, after all. But the
idea that the burdens must be fairly shared cuts both ways. If we go
146 Judith Jarvis Thomson

ahead, we must share, with you, the burden of meeting that child’s
need: We must not impose the entire burden of meeting its need on
you. If I am right, it follows that we need not reimburse you for the
entire cost of repairing or replacing the box and replacing the drug,
but only such part of that cost as leaves you to pay the same amount
as each of the rest of us. It is for this reason that I preferred not to
speak of that payment as compensation: its point is not so much to
compensate for a loss as to reduce that loss to the point at which it is
no greater than ours.
I should stress, however, that the cases I have drawn attention to
are all cases in which the redistribution aimed at is aimed at in order
to meet human needs. None of them is a case in which the
redistribution aimed at is aimed at simply in order that there be less
inequality. Taxation for redistribution for that purpose is a wholly
different matter.

NOTES

1. R. Nozick, Anarchy, State, and Utopia 171-74 (1974).

2. A typical example is the following: You are a sheriff in a small southern
town. A murder has been committed, and you do not have the least idea
who committed it, but a lynch mob will hang five others ifyou do not fasten
the crime on one individual.
3. This is a simplified account of what I mean by “infringe a right.” For
example, someone might have a right that such and such shall be the case,
and we might bring about that it is not the case, but our act might at one and
the same time bring about both that it is not the case and that he no longer
has a right that it is the case. It is possible that in some cases (that is, those
in which we infringe no other right of his in bringing about that he no longer
has that right), no right of his is “infringed,” in the sense I mean this word to
have. But the difficulties I point to here are of no interest for present
purposes, so I ignore them.
4. It is worth noticing, in passing, that for present purposes it would not
matter if Iwere wrong to make this assumption. There are acts which
Nozick says are violations of rights. I shall say that some of them, anway, are
nonwrongful infringements of rights. Suppose I am mistaken in this way:
That what I should have said is that they are wrongful, though excusable,
infringements of rights. Since Nozick plainly thinks those acts are not
merely wrongful, but inexcusable, what I shall say would still conflict with
what he thinks.
 Some Ruminations on Rights 147

5. It is of the greatest interest whether or not we have to pay all this

back, a question to which I shall return later. See text & note 19 infra.
6. See J. Thompson, Self-Defense and Rights (1976).
7. See discussion at conclusion of text.
8. R. Nozick, supra note 1, at 29.
SE WGh ESP
10. Id. at 30n.
Li Td:
12. Id. at 41.
13. Id. (emphasis in original).
14. Or something like (T), for of course we shall want to allow for
irrationality, preferences immorally inculcated, and so forth. I do not for a
moment want to suggest that I think the proper spelling out of the thesis
would be easy; it is merely that the difficulties are irrelevant for present
purposes.
15. Surely, however, it is not infinitely stringent: I should imagine it is
overrideable, even if not overridden in the cases at which we are looking.
16. See R. Nozick supra note 1, at 152-53.
17. See id. at 160-64, 168-72.
18. See id. at 182, 265-68.
19. Thomas Nagel, in his review of Nozick’s book, makes the interesting
suggestion that insisting that contributions be voluntary is “an excessively
demanding moral position” and that “excessive demands on the will. . . can
be more irksome than automatic demands on the purse.” Nagel, Book
Review, 85 Yale L.J. 136, 14546 (1975).
 S
Natural Rights, Equality,
and the Minimal State
SAMUEL SCHEFFLER!

INTRODUCTION

The idea of equality exerts considerable influence on our moral

imaginations, yet it has remained philosophically elusive. Although
men and women have thought equality worth dying for,
philosophers have been largely unable to give any systematic ac-
count for its importance as a moral ideal, or of its function in moral
and political theory. Seizing on this fact, Robert Nozick maintains
that while “there is no lack of unsupported statements of apresump-
tion in favor of equality,”? there is “a surprising dearth of argu-
ments’ (p. 233) capable of supporting such presumption. According
to the “entitlement conception of justice” which Nozick presents in
Anarchy, State, and Utopia, the non-voluntary redistribution of
income to achieve “equality of material condition” (p. 232) is morally
impermissible. In fact, Nozick’s argument is even more far-
reaching. He believes that any non-voluntary redistribution of
income is morally impermissible. In particular, he believes that it is
morally illegitimate for any government to tax some of its citizens in
order to provide food, shelter, medical care or social services for
other, less fortunate citizens.
Nozick presents his argument from the standpoint of natural
rights theory. In this discussion, I want to examine the structure and
content of Nozick’s argument. I will argue that within natural rights

148
 Natural Rights, Equality, and the Minimal State 149

theory, there is indeed a strong argument to be made for the

extensive redistribution of income, and even for the goal of equality
or near-equality of material condition. More specifically, I will
examine the way in which Nozick’s political theory depends upon
his moral theory, and the way in which his moral theory incorpo-
rates a particular theory of personal rights. I will argue that by
adopting a different theory of personal rights, we may arrive at a
different moral theory, and ultimately a different theory of the state.
Moreover, I will try to show that this alternative theory of rights,
with its moral and political consequences, is both independently
plausible and more consistent (than Nozick’s theory of rights) with
Nozick’s own remarks about the moral basis of rights. Finally, I will
examine the extent of the redistribution which is warranted under
the alternative theory, and I will introduce some further con-
siderations bearing on the legitimacy of egalitarian redistribution.

THE ROLE OF MORAL THEORY IN NOZICK’S

ARGUMENT

Non-voluntary redistribution of income is impermissible, on

Nozick’s view, because it violates the rights of those individuals
whose income is redistributed. Other coercive governmental
mechanisms designed to achieve egalitarian goals, such as the
imposition of wage ceilings or minima, are also said to be morally
impermissible. For, it is claimed, they violate the rights of those
people who would either have received more for their services, or
paid less for the services of others, in the absence of governmental
regulation. On Nozick’s view, the activity of the state is limited in
such cases by the moral rights of individuals; political design is
constrained by moral theory. Thus the burden of supporting
Nozick’s conclusion that the redistributive state is unjustified, falls
squarely on his moral theory in general, and on his theory of
individual rights in particular. Given the structural importance of
the moral theory for Nozick’s argument, it is presented with sur-
prising haste, and in little detail. He concedes this point at the
outset:

The completely accurate statement of the moral background, including the

precise statement of the moral theory and its underlying basis, would
150 Samuel Scheffler

require a full-scale presentation and is a task for another time. (A lifetime?)

That task is so crucial, and the gap left without its accomplishment is so
yawning, that it is only a minor comfort to note that we here are following
the respectable tradition of Locke, who does not provide anything remotely
resembling a satisfactory explanation of the status and basis of the law of
nature in his Second Treatise. (p. 9)

Nozick is of course quite right not to take much comfort from the
fact that Locke’s moral theory is similarly inadequate. Nozick is
himself working in the Lockean tradition, and he can hardly hope to
defend a weakness in his own theory by pointing out that the very
same weakness infects the entire tradition on which his work is
based. Still, Nozick does say some things about his moral theory,
and we should examine those things in order to gain insight into the
moral assumptions upon which his political structure is built.
According to Nozick, each individual has a right to be free from
force, fraud, and physical aggression. In addition, Nozick seems to
follow Locke in holding that each person has a right to exact
compensation from someone who violates his rights. Also, individu-
als acquire rights to whatever property they come to hold in
accordance with the rules specified by the entitlement conception of
justice. A property right in X is the right to dispose of X as ones sees
fit (provided that such disposal does not itself violate anyone else’s
rights), and interference with such rights through the use of force,
fraud, or physical aggression is prohibited. In short, the Nozickian
rights are (roughly) the Lockean rights: to be free from force or fraud
directed against one’s “life, health, liberty, or possessions.” (quoted
in Nozick, p. 10)
Nozick says that an individual's rights define a moral boundary
around the person. These boundaries constrain the permissible
behavior of every agent. To put it another way, the rights of other
people determine moral side constraints upon the activity of each
individual. The choice of the expression “side constraints” is made
to emphasize the point that, when a person acts, the non-violation of
other people’s rights must take priority over the pursuit of any goal:
even the goal of minimizing the total number of rights violated in
the society. Thus Nozick writes:
This view differs from one that tries to build the side constraints into the
goal G. The side-constraint view forbids you to violate these moral con-
 Natural Rights, Equality, and the Minimal State 151

straints in the pursuit of your goals; whereas the view whose objective is to
minimize the violation of these rights allows you to violate the rights (the
constraints) in order to lessen their total violation in the society. (p. 29)

Given this conception of the structure and content of moral

theory, it is not difficult to understand why, on Nozick’s view, the
non-voluntary redistribution of income is morally impermissible.?
By using the threat of force to coerce some people into contributing
to the welfare of others, a government violates the rights of the
people who are coerced. On the side-constraint view, such violation
is morally impermissible even if it would reduce the total number of
rights violated in the society. It is impermissible, in particular, even
if the failure of the government to coerce A will have the conse-
quence that the unaided citizen B will eventually have many more
of his rights violated (by third parties), than the number of rights
which the government would violate by taxing A to support B. It is
an important claimof Nozick’s moral theory that “your being forced
to contribute to anothers welfare violates your rights, whereas
someone else’s not providing you with things you need greatly,
including things essential to the protection of your rights, does not
itself violate your rights, even though it avoids making it more
difficult for someone else to violate them.” (p. 30) It is an extraordi-
nary but apparent consequence of this view that for a government to
tax each of its able-bodied citizens five dollars a year to support
cripples and orphans would violate the rights of the able-bodied,
and would be morally impermissible, whereas to refrain from
taxation even if it meant allowing the cripples and orphans to starve
to death would be the morally required governmental policy.4
Since the minimal state is built on the foundation of this theory of
individual rights, we should look more closely at Nozick’s concep-
tion of rights. Contemporary philosophers who ,choose to work
within the natural rights tradition need to explicate their use of
natural rights terminology. Given the metaphysical assocations of
the tradition, such philosophers must explain what they mean by
assigning rights to people. They must, further, say something about
the source of these rights, and they must deal with a variety of
epistemic questions. How do we know what rights people have?
What sorts of evidence justify us in believing that people do have
certain rights but do not have certain others?
152 Samuel Scheffler

Nozick has painfully little to say about most of these issues. His
book, he admits, “does not present a precise theory of the moral
basis of individual rights.” (p. xiv) Without being mean-spirited, it
bears noting that the omission is serious. It is serious not only
because Nozick builds his own political theory on his conception of
individual rights, but also because competing political theories can
be built on competing conceptions of individual rights. Unless we
have some answers to questions about the nature and source of
those rights, we can have no way of deciding among the rival
conceptions and theories. Actually, Nozick does say some things
about the basis of natural rights, and I will want to examine what he
says very carefully. First, however, I want to sketch an alternative
conception of individual rights. Although my presentation of the
alternative conception will not be complete, I hope to establish the
prima facie defensibility of the view. I will then return to Nozick’s
remarks about the moral basis of rights, and see whether his
conception of individual rights or the alternative conception rests
more comfortably on the moral basis as he describes it.

AN ALTERNATIVE CONCEPTION OF INDIVIDUAL

RIGHTS

In presenting this conception, I shall understand the assertion that a

person has a natural right to X to mean that that person has a natural
prima facie justified claim to X. The term “prima facie” is used to
indicate that such claims are overridable, but I understand them to
be overridable only in extreme circumstances. The force of the word
“natural” is to express the idea that in order to have such a right, it is
sufficient for a person to possess certain natural attributes. The
possession of natural rights is thus in no way dependent upon the
existence of a legal system or any other social artifact. I do not here
specify which natural attributes are sufficient conditions for the
possession of natural rights. Familiar suggestions are rationality, the
capacity for self-consciousness, and the capacity for forming a plan of
life. Alternatively, Rawls suggests “the capacity for moral personal-
ity.”> I assume that some such attribute or set of attributes is
sufficient, and I do not discuss the question whether the same
attributes are necessary. This interpretation of natural rights is in
accord with the interpretation articulated by Rawls:
 Natural Rights, Equality, and the Minimal State 153

[Natural rights are] claims [which] depend solely on certain natural attrib-
utes the presence of which can be ascertained by natural reason pursuing
common sense methods of inquiry. The existence of these attributes and the
claims based upon them is established independently from social conven-
tions and legal norms. The propriety of the term “natural” is that it suggests
the contrast between the rights identified by the theory of justice and the
rights defined by law and custom. But more than this, the concept ofnatural
rights includes the idea that these rights are assigned in the first instance to
persons, and that they are given a special weight. Claims easily overridden
for other values are not natural rights.®

According to the alternative conception, every person has a

natural right to a sufficient share of every distributable good whose
enjoyment is a necessary condition of the person’s having a reason-
able chance of living a decent and fulfilling life, subject only to the
following qualification. No person has a natural right to any good
which can only be obtained by preventing someone else from having
a reasonable chance of living a decent and fulfilling life. A sufficient
share of a necessary good is defined as a share of that good which is
large enough to satisfy the necessary condition. In other words, A
has a sufficient share of a necessary good G if A has at least the
minimum amount of G which is necessary in order to have a
reasonable chance of living the kind of life in question. The restric-
tion to distributable goods has the consequence that people are not
said to have a natural right to good health or a good life, for these are
not distributable goods. Rather, each person is said to have a right to
adequate food, clothing, shelter, and medical care.” In addition,
each has a natural right to a substantial degree of personal liberty.®
Precisely what amount of each of these goods constitutes a sufficient
share, as defined above, is a matter of empirical investigation. The
stipulation that each person has a right to a sufficient share of all
those goods whose enjoyment is a necessary condition of having a
reasonable chance of living a decent and fulfilling life means that a
good is not shown to be unnecessary, and a share is not shown to be
too large, by the mere fact that some extraordinary person may have
managed, against the odds, to live a decent and fulfilling life without
the good in question, or with a lesser share. Conversely, neither
individual shares nor the total package of goods to which a person
has rights is shown to be too meager by the mere fact that some
people may enjoy all of these goods and still fail to achieve decent
154 Samuel Scheffler

and fulfilling lives. For, according to this conception of rights,

people have rights to those goods which are necessary (not
sufficient) for having a reasonable chance (not an absolute guarantee)
of living such lives. (In any case, there is no known distributable
good whose enjoyment constitutes a guaranteed sufficient condition
for living a decent and fulfilling life.)
Several features of this conception of natural rights require com-
ment. First, it is clear that this conception takes a dimmer view of
property rights than does the Lockean conception. On the alterna-
tive conception, each person has a natural right to such material
goods as are necessary to insure him a reasonable chance of living a
certain kind of life. Beyond that, his property rights derive from his
right to liberty, but since the latter is limited, so is the former. This
point will be treated with greater thoroughness in the course of the
discussion that follows.
Second, assume for the moment that the alternative conception is
correct, and that people really do have the natural rights which it
assigns to them. That assumption, if true, constitutes a prima facie
justification for the establishment of governmental mechanisms
which would insure the satisfaction of those rights which the alterna-
tive conception specifies. It constitutes only a prima facie justifica-
tion because it might not follow simply from the fact that people had
such rights that it was possible or permissible for a government to
enforce them all. Thus a full-fledged argument would have to
address itself to this question. Nevertheless, we can safely say that
the correctness of the alternative conception would provide the
foundation for a natural rights argument to the effect that gov-
ernmental mechanisms should be established to insure that no
citizen lacks adequate food, clothing, shelter, or medical care, as
well as to insure that citizens enjoy substantial personal liberty.
Third, it appears on the face of it that a government could insure
the rights of all its citizens without violating the rights of any.
Although it is true that the mechanisms necessary to guarantee the
satisfaction of the “welfare rights” would require some restrictions
on personal liberty, it is not the case that any restriction on personal
liberty constitutes a violation of rights. For according to the alterna-
tive conception of rights, individuals have a right to whatever liberty
is necessary to insure a reasonable chance of living a decent and
fulfilling life. While this entails the right to a substantial degree of
 Natural Rights, Equality, and the Minimal State 155

personal liberty, it does not entail a right to unlimited liberty, for

unlimited liberty is not necessary to insure a reasonable chance of
living such a life. In particular, the kind of mechanisms necessary to
guarantee the satisfaction of the welfare rights of all would not
violate the degree of personal liberty which the alternative concep-
tion assigns to individuals by right. To see this we have only to
observe that the taxation of wealthy North Americans can hardly be
said to deprive them of that degree of personal liberty which is
necessary in order to have a reasonable chance of living decent lives.
Fourth, the last two points suggest that if we incorporate the
alternative conception of rights into a side-constraint structure like
Nozick’s, the redistribution of income emerges as morally permissi-
ble. Such a structure, as we have seen, forbids the violation of any
person’s rights as a means to maximizing some desired end. If
indeed a government could enforce all the rights assigned by the
alternative conception without violating any, that shows that by
joining our moral content (the alternative conception) to Nozick’s
moral structure (the side-constraint), we arrive at a side-constraint
theory which judges redistribution not to violate anyone’s rights.
Thus the alternative conception embodies a rejection of Nozick’s
moral content, but it is compatible with his moral structure. We will
have occasion later to consider and reject an argument by Nozick
that the side-constraint structure entails his moral content.
Fifth, it appears that from the standpoint of the alternative
conception, the minimal state described by Nozick is seen not only
as protecting the rights of its citizens to liberty, but also as guarantee-
ing them a greater degree of liberty than they can claim by natural
right.? On the other hand, the minimal state permits the welfare
rights of its citizens to go unsatisfied, provided only that their
non-satisfaction does not result from some violation of the victims’
Lockean rights. Thus, if the claim that a “welfare state” could
guarantee all the rights of its citizens without violating any is
correct, then there is a prima facie case for asserting that, from the
standpoint of the alternative conception, the welfare state (which
protects all the rights of its citizens) is morally superior to the
minimal state (which protects only some of those rights).?°
Sixth, the alternative conception is not embarassed by Nozick-
style objections to the effect that the non-voluntary redistribution of
welfare goods is morally continuous with the forcible redistribution
156 Samuel Scheffler

of bodily parts or sexual partners. It is instructive to see why this is

so; consider for a moment the question of sexual experience. On the
alternative conception, there is no natural right to sexual experience
(or to partners needed to achieve sexual experience), for at least
three reasons. First, insofar as it is plausible to speak of some kind of
sexual experience as a distributable good (a government could
coerce some people into serving as sexual partners for others), it is
in no way plausible to claim that such experience is necessary to
insure a reasonable chance of living a decent and fulfilling life.
Coerced sexual activity could only degrade its participants, and
would more likely reduce than promote their chances of living well.
The bonds of intimacy and trust which unite real lovers are in no
sense distributable, and this suffices to show that the kind of sexual
experience which we value most dearly lies beyond the scope of
justice. Second, even this kind of sexual experience cannot truly be
said to be necessary for having a reasonable chance of living the sort
of life in question. While experience of this kind may enhance one’s
chances of living well, it is hardly on a par with food or medical care.
Moreover, if sexual experience of any kind were necessary—if, for
example, people who didn't have sexual intercourse at least three
times a day would die—we might well have different intuitions
about the permissibility of coerced sexual partnership. But also if
this were the case, it is worth wondering whether mankind as a
species would ever have begun to use sexual activity as a medium
for communicating affection and need and trust to the extent that it
actually does.!! The third reason why there is no natural right to
sexual experience on the alternative conception is that this concep-
tion denies people rights to any goods which can only be obtained
by preventing someone else from having a reasonable chance to live
a decent and fulfilling life. Coerced sexual activity would violate this
provision, both by the extent of the restrictions on personal liberty
which it would require, and by the violence it would do to people’s
perceptions of themselves and their own personalities.
Seventh, the alternative conception of rights provides, as we have
seen, a prima facie justification for the establishment of governmen-
tal mechanisms designed to insure people’s welfare rights, as well as
their rights to liberty. This conception does not rule out the
possibility, however, that additional arguments could establish that
governments should guarantee even more liberty and/ more mate-
 Natural Rights, Equality, and the Minimal State 157

rial goods to each of its citizens, if that could be done without

violating rights. In other words, additional arguments might be
brought to bear to urge that governments should protect certain
kinds of liberty more extensively, or that governments should
enforce a still greater equality of material condition, than would
result just from the guaranteeing by the state of each individual's
natural rights. The alternative conception does not rule out argu-
ments to this effect; it just specifies that they will not be arguments
from people's natural rights. For on the alternative conception,
people’s natural rights are satisfied and exhausted once their
minimum welfare and liberty needs are met.!” I will return to the
question of whether there are additional good arguments in favor of
material equality, and whether such equality could be achieved
without violating natural rights. First, however, it is time to exam-
ine Nozick’s remarks about the moral basis of natural rights, and to
see whether his conception of rights or the alternative conception is
a more accurate reflection of that basis. Then I will consider briefly
some of the major objections to the alternative conception of rights;
I will conclude by returning to the question whether there is a case
to be made for material equality beyond that degree of equality
which could be justified by the alternative conception alone.

THE MORAL BASIS OF NATURAL RIGHTS

The alternative conception of rights, as it has been described, does

not explain the moral basis of the rights it assigns. While it identifies
(roughly) the natural attributes whose possession is a sufficient
condition of having rights, and while it provides a fairly clear
criterion for deciding which rights people have, it does not explain
why it is that individuals who possess the attributes in question
should be said to have those rights. As we have seen, Nozick
concedes that his book does not include a detailed account of the
moral basis of those rights which his conception assigns. Yet those
comments which he does make about the moral basis are suggestive,
and they provide sufficient ground for making a tentative choice
between the Lockean conception of rights and the alternative
conception. In other words, proceeding on the assumption that the
moral source of individual rights is in fact something like Nozick
describes it as being, we can see whether the Lockean conception or
158 Samuel Scheffler

the alternative conception provides a more accurate specification of

the rights which would flow from such a source. Of course, a
complete presentation of the alternative conception would have to
include a positive account of its own of the moral basis.
Nozick approaches the problem of identifying the moral basis of
rights by considering the question of which natural attributes are
necessary and/or sufficient for possessing rights. He argues that
there must be a connection between these characteristics and the
moral basis of rights:

. in virtue of precisely what characteristics of persons are there moral

constraints on how they may treat each other or be treated? We also want to
understand why these characteristics connect with these constraints . . . It
would appear that a person’s characteristics, by virtue of which others are
constrained in their treatment of him, must themselves be valuable charac-
teristics. How else are we to understand why something so valuable
emerges from them? (p. 48)

Nozick rejects various common suggestions about what the relev-

ant characteristic(s) might be, and concludes with a proposal of his
own:

I conjecture that the answer is connected with that elusive and difficult
notion: the meaning of life. A person’s shaping his life in accordance with
some overall plan is his way of giving meaning to his life; only a being with
the capacity to so shape his life can have or strive for meaningful life . . .
This notion, we should note, has the right “feel” as something that might
help to bridge an “is-ought” gap; it appropriately seems to straddle the two.
(p. 50)

Nozick concedes that there are difficulties with this answer, and
that the notion of “the meaning of life” requires fuller elaboration.
For our purposes, what is important is that Nozick believes that the
moral basis of rights has to do with the capacity to live a meaningful
life. For then we may ask, what rights would such a basis be the
basis of? In other words, if the capacity to live a meaningful life is a
uniquely valuable characteristic, and if we say that beings with this
characteristic have rights, in virtue of which there are constraints on
the way others must behave, then presumably the function of the
rights is to safeguard the ability of beings with this valuable charac-
 Natural Rights, Equality, and the Minimal State 159

teristic to develop it. To say that the valuable capacity to live a

meaningful life is the basis of rights, is presumably to suggest that
the moral protections and guarantees which rights assign to people
may be understood as jealous of people’s ability to actually live
meaningful lives. But then it seems clear that the alternative
conception of rights is a much more accurate specification than the
Lockean conception of the rights which people actually have. For
the alternative conception assigns to each individual the right to a
sufficient share of all distributable goods whose enjoyment is neces-
sary to have a reasonable chance of living a decent and fulfilling life.
Now whatever a meaningful life may turn out to be, it seems
reasonable enough to suggest that any distributable good necessary
for living a decent and fulfilling life (such as food) will also be
necessary for living a meaningful life. In fact, it seems reasonable to
suggest that the distributable goods necessary to have a reasonable
chance of living a meaningful life are just the same as the goods
necessary to live a decent and fulfilling life. If that is so, then the
alternative conception of rights is clearly the correct one: for it alone
insures that all the necessary material conditions for having a
reasonable chance of living a meaningful life will be met.
Even if, as seems unlikely, the goods necessary for living a
meaningful life are not the same as the goods necessary to live a
decent and fulfilling life (if, for example, there are some additional
goods necessary for the former), it is still the case that the alterna-
tive conception is clearly superior to the Lockean conception, which
denies people rights to goods like food and medical care which are
certainly necessary for living a meaningful life. On Nozick’s own
view of the moral basis of rights, then, we see that the alternative
conception emerges as superior to the Nozickian conception of
which rights people actually have. In fact, Nozick’s overall view
begins to appear irrational: how can one hold both that rights are
necessary to protect and guarantee the valuable capacity to live a
meaningful life, and that people only have rights to some of the
distributable goods which are necessary in order to have any chance
of living meaningful lives? Of course, it would not be irrational to
hold these two views in conjunction if one believed, as Nozick
presumably does, that one could not succeed in providing sufficient
shares of all necessary distributable goods to all people—that one
could only provide sufficient shares of the welfare goods to some by
160 Samuel Scheffler

denying sufficient shares of liberty to the others. But, as I have

argued, it is simply false that the restrictions on liberty necessary to
guarantee the welfare rights of all would be so severe as to prevent
the victims of, say, taxation, from living decent and fulfilling lives.
Nor is it any less false that such taxation would prevent them from
living meaningful lives. If the meaning of life is our concern, then
starvation, not taxation, is our worthy foe.
In light of these considerations, we need not be impressed by
Nozick’s “argument from moral form to moral content.” (p. 34) He
says that “the form of morality includes F (moral side constraints),”
(p. 34) that “the best explanation of morality’s being F” (including
side constraints) is “a strong statement of the distinctness of indi-
viduals” (p. 34), and that given this explanation of moral form, there
follows a particular moral content: namely “the libertarian con-
straint” which is determined by the Lockean conception of rights.
For, he believes, only the Lockean conception of rights reflects the
distinctness of persons. The redistributive state, he says, ignores the
fact that “there are only individual people, different individual
people, with their own individual lives,” (p. 33) and that “there is
no justified sacrifice of some of us for others.” (p. 33) We reject this
argument, for we believe that even if one adopts a side-constraint
structure for moral theory, and even if one does so out of a belief
that a side-constraint structure best expresses the distinctness of
every person, then the rational moral content to adopt will be that
set of side constraints which is determined by the alternative
conception of rights, and not the Lockean conception. We believe,
in other words, that it is the alternative conception, with its
protections and guarantees of welfare as well as liberty, which truly
cherishes each and every individual human life. We deny that the
political structures necessary to implement the alternative concep-
tion involve the sacrifice of some people for the sake of others. We
maintain, on the contrary, that only the alternative conception
guarantees to each person the chance to live a life that can be looked
back on without bitterness or regret.1%

SOME CRITICISMS OF THE ALTERNATIVE

CONCEPTION
Thus far I have presented the alternative conception without con-
sidering the criticisms which can be directed against it. Although I
 Natural Rights, Equality, and the Minimal State 161

will not now try to give a full defense of the alternative conception,
it is important to at least indicate the sort of replies which could be
offered to some major objections. One such objection focuses on the
fact that the distributable goods to which the alternative conception
assigns people rights are not, in most cases, naturally obtainable.
Medical care does not grow on trees, and while some foods do, in
practice most of the welfare goods which people actually obtain have
been produced or prepared through the labor of others. Thus, it
may be said, the alternative conception assigns people rights to
goods which, in practice, will frequently be the fruit of other
people’s labor. And in so doing, it surely infringes the liberty of the
laborers.
This objection misunderstands the moral picture painted by the
alternative conception. The alternative conception does not divide
the world into laborers and consumers. The fact that the welfare
goods are obtainable in practice only through the mediation of a
complex society. of working men and women is certainly of
significance for the alternative conception. But its significance is not
to show that people do not, after all, have the rights which the
alternative conception assigns to them. Rather, its significance is to
show that individuals with rights also have duties: duties, in par-
ticular, to contribute their labor, according to their talents and
abilities, to the enhanced functioning of the society in which they
live. Let us say that a society is ordered in part by the alternative
conception if its government guarantees the satisfaction of all the
rights which the alternative conception assigns. In such a society,
the government would not force one group of people to work so that
another group could live idle and secure. Rather, such a govern-
ment would guarantee, through the establishment of institutions
like taxation, that neither helpless persons, nor those ablebodied
persons who were willing to work, would ever be allowed to die for
lack of food, clothing, or medical care. (The refusal of an ablebodied
person to do any kind of work might suffice to override that person's
natural welfare rights.) In such a society, as we have already
insisted, it would indeed be necessary to restrict the liberty of
citizens. For example, no citizen would be at liberty to enjoy
unlimited material gain while others were starving. But, as we have
argued, this kind of restriction on liberty does not violate anyone's
natural rights.
A second objection to the alternative conception focuses on the
162 Samuel Scheffler

structure of the rights assigned. This objection notices that the

welfare rights are structured differently from the rights to liberty.
While the latter prohibit certain kinds of direct aggression against
the individual, and correlate fairly neatly with the moral duties of
others, the welfare rights determine constraints on our behavior
that are notably more complex. That is, they seem to require that no
person consume more than he is entitled to by natural right so long
as other members of his society have some of their natural rights as
yet unsatisfied. Moreover, the alternative conception seems to
suggest that a person sitting alone in his room, apparently minding
his own business, can, indirectly at least, violate the rights of people
he has never met just by consuming more than a certain amount.
The precise degree of indirection involved in such violations can be
emphasized by pointing out the difficulty in specifying the particular
people whose rights are violated by this solitary overconsumer (all of
his fellow citizens whose rights are not yet satisfied, perhaps), and
by pointing out that, in contrast to the libertarian rights, an offender
against welfare rights might cease his offensive behavior without any
benefit necessarily accruing to the victim.
The alternative conception concedes these distinctions between
the rights to liberty and the welfare rights, but does not view them
as objections. Liberty and food are different sorts of good, and so it
is not suprising that the rights to each should be differently struc-
tured.14 Despite these differences, both kinds of rights are rights in
good standing; both conform to the precise definition provided by
the alternative conception, and both constrain our behavior and
limit the area within which we may permissibly do as we please. The
fact that the welfare rights constrain our behavior in a way that is
more subtle and complex than the libertarian rights does not make
them any less significant, from a moral point of view.
A third objection maintains that the alternative conception viol-
ates moral intuition by allowing minor infringements of liberty
provided that they do not prevent the victims from having a
reasonable chance of living decent and fulfilling lives. If Iwalk up to
you on the street and punch you in the nose, you may still walk away
with a reasonable chance of living a decent and fulfilling life. Thus,
this objection urges, the alternative conception is too weak in its
protection of personal liberty, for it would not rule out such offenses
as a random punch in the nose.
 Natural Rights, Equality, and the Minimal State 163

This objection makes the mistake of assuming that the alternative

conception of rights constitutes a complete moral theory. This is not
true; any conception of rights is just one component of such a
theory. Such a conception might be combined with other general
moral injunctions in a variety of ways, as Nozick himself notes. (p.
29) Within a side-constraint structure, for example, a moral theory
might include such rules as: so long as you do not violate any of the
rights specified by the alternative conception, act so as to minimize
suffering, or, so long as you do not violate any rights, treat all people
with respect—and so on. Even when incorporated into a side-
constraint structure, what the rights specified by the alternative
conception do is to prohibit certain kinds of behavior, and thus to
protect certain central areas of human activity. This obviously does
not entail, however, that all other kinds of behavior are morally
permissible, or that the noncentral areas of human activity are fair
game for intruders. It is the task of the complete moral theory to
deal with this question thoroughly.
Before proceeding to the final section of this paper, one further
comment about Nozick’s argument is called for. In the discussion so
far, I have avoided direct consideration of Nozick’s “entitlement
theory” of distributive justice. It should be obvious, however, that a
rejection of the entitlement theory as stated is implicit in the
argument of this paper. By suggesting that the rights specified by
the alternative conception set limits to the kinds of voluntary
behavior that are morally permissible, the argument of this paper
circumscribes the area within which voluntary exchanges may take
place in a way that is clearly incompatible with the entitlement
theory as Nozick presents it. The alternative conception does not
deny that entitlement considerations are ever morally relevant, but
it stipulates that if they are, it is only after the welfare rights of all
have been satisfied. The relative weight of entitlement con-
siderations as compared to egalitarian considerations in the post-
welfare stage is discussed in the final section.

BEYOND WELFARE: THE LEGITIMACY OF

EGALITARIAN REDISTRIBUTION

In a society blessed by abundance, it may be possible to satisfy the

welfare rights of all people while permitting the continued existence
164 Samuel Scheffler

of vast inequalities of material condition. Earlier, I suggested that

while no arguments from natural rights could be brought to bear in
favor of still further egalitarian redistribution, there remained the
possibility that other moral arguments toward this end might be
brought forward. I should like at this point to briefly advance two
such considerations. Since further redistribution would require still
greater restrictions on certain kinds of liberty, the considerations I
shall discuss would need to be carefully weighed against con-
siderations of liberty in order to arrive at a final judgment about the
legitimacy of such further redistribution. In the space available, I
shall simply mention the two egalitarian considerations, and in-
dicate my inclination about the direction in which a final balancing
would go. The argument will not be exhaustive, and people with
contrary inclinations will find wide room for disagreement.
The first consideration pertains to the stability of an inegalitarian
society from the standpoint of rights. We have considerable evi-
dence that in most societies there is a high correlation between
wealth and political power. In light of this high correlation, we have
reason to wonder whether a highly inegalitarian society would be
stable with respect to the rights of all. In other words, since some
people in such a society would be much more powerful than others,
we have some reason to worry whether such people would find it
tempting to exploit the less powerful and deprive them of their
natural rights. There is some evidence that a radically inegalitarian
welfare society would be unstable in this sense. If this is in fact the
case, then the desire to satisfy the rights of all might argue for
further redistribution beyond the minimal amount which would be
technically necessary to satisfy those rights.
The second consideration pertains to the well-known claim that
the self-respect of the less fortunate suffers in inegalitarian societies.
Societies which are sharply divided with regard to material condi-
tion tend to foster feelings of haughtiness and contempt on the part
of some, deference and resentment on the part of others. Clearly
such feelings benefit no one: neither the less nor the more fortu-
nate. That they do not benefit the less fortunate seems obvious.
That they may cause suffering among the more fortunate as well is a
point that is expressed with painful clarity in Margaret Drabble’s
novel, The Needle’s Eye. Witness her description of a visit by Rose
 Natural Rights, Equality, and the Minimal State 165

Vassiliou, the unwilling heiress, to the home of her parents, and of

her feelings when required to give instructions to their servants:

Asking about dinner had made Rose feel so miserable, in so familiar a way:
she knew that it made no difference to the people in the kitchen, to be
asked to provide an extra two meals, to be asked to make up a couple more
beds, she knew that they were paid to do such things, that they could leave
if they wanted to, that they were often asked to do much worse things with
less notice and in less polite a manner, and yet nevertheless such a wash of
embarrassment had poured over her, as she stood there clumsily and
nervously, ill at ease, that it had transported her back twenty years, to the
humiliation of being half-employer, half-servile child, treated by the staff
with a mocking deference, and yet at the same time privileged to hear their
complaints, their moans about her parents and the employing class in
general, a tenant of both worlds, belonging to neither, recipient of the
confidences of each about the other, and therefore all too painfully aware of
the mutual contempt that reigned between them. It was these years,
perhaps, that had made her so neurotically incapable of relying on. the
services of others: She recalled the relief that had filled her when she
discovered that it was possible to get through life cleaning one’s own shoes,
cooking one’s own meals, washing one’s own pants, that it was not a law of
nature that decreed her to suffer for ever the humiliation of having these
things done for her by people who despised her.1*

Nozick suggests “one possible explanation of why certain in-

equalities in income . . . rankle so: not due to the feeling that this
superior position is undeserved, but to the feeling that itis deserved
and earned.” (p. 241) It is highly implausible, however, to suggest
that the richest people in societies like our own do in fact possess, or
are perceived to possess, some attribute which makes them the
people who most deserve to be rich. It is only this assumption which
enables Nozick to argue the unlikely elitist position that resentment
by the poor of the rich is really just envy by the deservedly poor of
the deservedly rich. But the rich of our society are not the hardest
working people, or the most dedicated, or the most talented; there
is no plausible sense of ‘desert’ in which the richest people can be
said to be the most deserving. It is precisely this fact which explains
why the inequalities of material condition in societies like our own
can be so destructive of self-respect. For they distort human re-
lationships by rewarding people who do not deserve it, and by then
166 Samuel Scheffler

permitting them to command the deference and self-abasement of

other equally worthy people. They thus create a far-ranging social
evil which can undercut the ability of large numbers of people to
live fulfilling lives. Since it is also an evil which could plausibly be
eliminated by egalitarian redistribution, and by arranging such
inequalities as remain in accordance with a reasonable principle of
desert or incentive, it provides us with a substantial reason for
taking the possibility of such further redistribution seriously.
As I have indicated, the two considerations just mentioned mili-
tate in favor of still greater egalitarian redistribution, which in tum
would entail further restrictions on certain kinds of liberty, and a
further narrowing of the legitimate range of voluntary business
exchanges. I believe that further redistribution is justified, and,
moreover, that it could be accomplished without violating anyone's
libertarian rights. For there seems no reason to believe that such
measures as highly progressive taxation and limitations on inheri-
tance need deprive anybody of the liberty necessary to live a decent
and fulfilling life. Though I think them justified, I recognize that
these further restrictions on the good of liberty are unwelcome. This
only teaches us what we must surely have already known: that in
political theory as in life one can't always enjoy all good things
together. All too often we must sacrifice one to save another. That is
one reason why in political theory as in life there is cause for
heartache as well as hope.

NOTES

1. I would like to thank Thomas Nagel for his comments on an earlier

version of this paper. I have also been helped by discussions with Janet
Broughton, John Campbell, Peter Railton, and Cornel West, and I wish to
express my gratitude to them.
2. Anarchy, State, and Utopia, Basic Books, New York, 1974, p. 233, All
page references included in the text of the article refer to this book.
3. It is important to see that this impermissibility depends on conjoining
Nozick’s moral structure with his moral content. One could, as we will later
demonstrate, take his moral structure (the side-constraint structure), con-
join it with a different moral content (an alternative conception of rights),
and conclude that redistribution is permissible. Alternatively, one might
reject his moral structure in favor of a maximizing structure, retain his
 Natural Rights, Equality, and the Minimal State 167

moral content (the Lockean conception of rights), and again conclude that
redistribution is permissible.
4. Nozick would of course be all in favor of voluntary aid for such people.
Failing that, however, he would apparently judge it a morally superior
outcome if the cripples and orphans died than if the government taxed its
citizens to support them. Nozick says that he hopes to avoid “the question of
whether . . . side constraints are absolute, or whether they may be violated
to avoid catastrophic moral horror.” (p. 30n.) How many cripples and
orphans would have to die in order to constitute a “catastrophic moral
horror’?
5. A Theory of Justice, Harvard University Press, Cambridge, 1971, p.
505.
6. Rawls, pp. 505n.—506n.
7. Is there a natural right to education, on the alternative conception? In
most modern societies, the answer is certainly yes, for education is neces-
sary in order to have a reasonable chance ofliving a decent and fulfilling life
in most such societies. But in thinking about this question, we see that the
goods to which people have natural rights can vary across time and different
societies, on this conception.
8. Some people will balk at the suggestion that liberty is a distributable
good. We could satisfy such people by substituting a right to the distributa-
ble good of police protection for the right to liberty; only terminological,
non-substantive changes would result. We will continue to treat liberty
itself as a distributable good, however, for reasons of convenience. While it
is not transferable, liberty is distributable in the weaker sense that govern-
ment policy can directly regulate the amount which people enjoy, and that
will be sufficient for our purposes.
9. But note Thomas Scanlon’s argument that “establishment of a system
of property rights based on free contract means that some people, in order
to gain the means to life, have to devote virtually all their productive
energies to whatever tasks and pursuits are desired by those who control
the goods necessary for life in their society.” (“Markets, Liberty, and the
Obligation to Contribute,” unpublished xerox, p. 28. This paper was
presented at the Battelle Seattle Research Center conference on Markets
and Morals in May, 1974. It is scheduled to appear in the proceedings of
that conference, which will be edited by Gerald Dworkin and Gordon
Bermant.) He concludes that a libertarian society, contrary to appearance,
may indeed involve restrictions on the liberty of some of its citizens. (See
Scanlon, pp. 28-29)
10. Scanlon arrives at a similar conclusion from a different direction. He
argues that we could justify the restrictions on liberty required by a welfare
168 Samuel Scheffler

state to those people whose liberty was restricted, whereas we would be

unable to produce a satisfactory justification of market institutions to those
whose liberty they restricted. On our alternative conception of rights, the
disparity between the relative strengths of the two sorts of justification can
be explained by the fact that even those people whose liberty was restricted
by the welfare state would not be having their rights violated, while the
least fortunate citizens in a libertarian society might well suffer such an
outcome. (See Scanlon, pp. 29-31)
11. People do not ordinarily communicate their deepest feelings about
each other by eating together. Might the situation have been reversed if sex
were necessary for individual survival but food were not?
12. For a different view, see Gregory Vlastos, “Justice and Equality,” in
R. Brandt ed., Social Justice, Prentice-Hall, Englewood Cliffs, New Jersey,
1962, pp. 31-72.
13. I wish to emphasize that the argument presented in this paper is
intended to show that the alternative conception is compatible with a
side-constraint structure. This is of course the moral structure which Nozick
favors, and I have tried to suggest that even when the competing concep-
tions of rights are both incorporated into this kind of structure, the
alternative conception emerges both as more consistent with Nozick’s own
remarks about the moral basis of rights, and as independently plausible. It
should be obvious, however, that the alternative conception could be
incorporated into other structures as well (for example, it could be incorpo-
rated into a maximizing structure). Nozick aside, the question of which
structure is best suited to the alternative conception is a separate one, and
one which I shall not try to answer here.
14. Notice that a right to police protection would also be liable to indirect
violation by overconsumers.
15. The Needle’s Eye, Popular Library, New York, 1972, p. 329.
 S,
Nozick on Unproductivity:
The Unintended
Consequences
ERIC MACK*

By far the most difficult chapter of Robert Nozick’s Anarchy, State

and Utopia is the one entitled, “Prohibition, Compensation and
Risk.” Accordingly, this central chapter has been almost entirely
ignored by commentators friendly and hostile alike. Such caution is
probably a mark of wisdom. Nevertheless, in this essay I aspire both
to give an account of the basic structure and structuring motivations
of “Prohibition, Compensation and Risk” and, more specifically, to
provide an explication and critique of Nozick’s crucial notion of
unproductivity and of his use of this notion. The discussion also
serves to indicate an ambivalence on Nozick’s part about the charac-
ter of rights. The format is roughly as follows: Section I deals with
the structure of “Prohibition, Compensation and Risk.” It focuses on
the central role of the Principle of Compensation in support of
which Nozick introduces the notion of unproductivity. Section II
explicates the notion of unproductivity. Section III traces the impli-
cations of Nozick’s use of this notion; and Section IV concludes with
some general observations on the profitability, for Nozick’s enter-
prise, of his focus on unproductivity. We shall see that Nozick’s use

169
170 Eric Mack

of the notion of unproductivity has a variety of unforeseen implica-

tions which violate the spirit of his libertarianism. If the integrity of
his libertarianism is to be preserved, the use of this notion must be
abandoned. A consequence of such an abandonment is that the
Principle of Compensation which, for Nozick, plays a crucial role in
legitimating the state against individualist anarchist objections is left
unsupported. Furthermore, and perhaps more fundamentally, we
shall see that in the process of developing his argument in “Prohibi-
tion, Compensation and Risk” important unacknowledged shifts
occur within Nozick’s conception of rights. Essentially this involves
a shift from a fully deontic conception of rights as indicative of what
“no person or group may do’ (ix) to something approaching a
utilitarian calculus. Such a shift, it will be argued, largely undercuts
Nozick’s historical entitlement theory of distributive justice, while
lending support to a kind of end-state theory explicitly opposed by
him.

For Nozick, the central explicit task in “Prohibition, Compensation

and Risk” is the development of an account of how one may treat
agents who engage in or are about to engage in risky actions, i.e.,
actions which risk eventuating in violations of rights. Within the
overall plan of Anarchy, State and Utopia, Nozick’s purpose is to
indicate why a protective association could legitimately prohibit the
risky protective activities of other associations and thereby move
toward statelike status. Risky acts, here, are acts which have a
significant likelihood of eventuating in significant enough violations
of rights so that, intuitively, it seems that one can do more to
discourage them than merely negotiating with the would-be risk-
imposers in order to purchase their abstentions from these acts.
Risky acts stand, somehow, between those actions whose prohibi-
tion is fully unproblematic and those actions which are sufficiently
non-risky as to count as fully innocent. Nozick’s central explicit
hypothesis in “Prohibition, Compensation and Risk” is that there is
a special intermediate, and legitimate mode of reaction to the actions
in the intermediate, risky category. Thus he claims that, “the
dilemma, ‘either you have a right to forbid it so you needn't
compensate, or you don't have a right to forbid it so you should
 Unproductivity: The Unintended Consequences Wye

stop, is too short.” (83) Specifically, agents may be prohibited from

performing risky acts if they also receive compensation for having
suffered these prohibitions. That is, we may deal with risk-imposers
in accordance with the Principle of Compensation which declares
that, “those who are disadvantaged by being forbidden to do actions
that only might harm others must be compensated for these disad-
vantages foisted upon them in order to provide security for the
others.” (82-83)?
The Principle of Compensation opts for the prohibition of risky
acts in contrast to requiring that those who disvalue risky activities
bargain with would-be risk-imposers in order to purchase their
abstentions from these risky endeavors. Yet Nozick offers arguments
which purport to show that in general one may not proceed to
require a desired action (or abstention) from another even if one
compensates this party. In general one must secure the other party's
action (or abstention) by means of prior negotiation and agreement.
The question, then, is why do these general arguments not apply to
the securing of others’ abstentions from risky activities? This ques-
tion is to be answered by indicating what these general arguments
favoring prohibition are and why they are thought not to apply to
the securing of people’s abstentions from risky actions.
To get at the general arguments we begin with the most general
question posed by Nozick in “Prohibition, Compensation and Risk.”
“Why ever prohibit?” (58) Why prohibit even clearcut boundary
crossings—rather than allowing them but then requiring the trans-
gressors to compensate their victims?? Given Nozick’s overall insis-
tence on rights (boundaries) and their importance, it may seem odd
that Nozick raises this question at all. It would seem that we
prohibit because there are rights to be respected and protected. But
Nozick denies that X’s having a right to G entails X’s being justified
in forcibly forbidding Y’s taking G. (91-92) Thus, the justification of
such a prohibition, if there is one, must be found at least partially
elsewhere. For Nozick, the justification of prohibition is to be found
in a comparative evaluation of alternative modes of response to the
act whose prohibition is under consideration. Nozick seems to hold
that if an action would violate rights it follows that this action may
permissibly be nullified. Nullification, however, may take the form
of forbidding the offending action or it may take the form of
requiring subsequent compensation to the victim. Whether nullifi-
WHS, Eric Mack

cation should take the form of prohibition depends upon the

viability of the compensation alternative. The case is made for
prohibition if and only if this compensation alternative is shown
to be not viable.
That this is the way a case is to be made for prohibition is
illustrated in Nozick’s arguments for the prohibition of (ordinary)
boundary crossings. Nozick contends that posterior compensation
for boundary crossings simply is not a viable alternative to the
prohibition of boundary crossing. His most general argument
pointing to the non-viability of posterior compensation is the divi-
sion of the benefits of exchange argument. According to this argu-
ment there are two natural candidates for appropriate posterior
compensation: full compensation, i.e., the return of the victim of
the crossing to his previous indifference curve, and market compen-
sation, i.e., the provision of the victim with what he would have
received as the result of freely negotiated agreement to be subject to
what, in the absence of such an agreement, was this boundary
crossing. The central problem with the first candidate for posterior
compensation is that,

. a system that allows all boundary impingements provided that full

compensation is paid is equivalent to a system requiring that all prior
agreements about the right to cross a border be reached at that point on the
contract curve most favorable to the buyer of the right. (63)

Under such a system all the benefits of an exchange forced by a

boundary crossing would be distributed to the boundary crosser.
Yet there is no reason why all the benefits from an exchange
(especially a forced exchange) should go to one of the parties to that
exchange (especially the party who forces it). “Allowing boundary
crossing provided only that full compensation is paid ‘solves’ the
problem of distributing the benefits of voluntary exchange in an
unfair and arbitrary manner.” (64)
There is, however, a different sort of problem with the second
candidate for posterior compensation, viz., market compensation. If
the appropriate compensation is what would have been agreed to
had free negotiations been carried out, then at least in those cases in
which negotiation is possible it is absurd to allow a person to
 Unproductivity: The Unintended Consequences 173

boundary cross and then seek to determine what price would have
been agreed upon if a prior agreement had been reached. One
avoids the tangle of counterfactuals simply by requiring prior
negotiation. (And once such negotiations are completed what is
done in accord with them are not boundary crossings.) That parallel
problems exist for a policy of allowing risky acts while requiring
posterior compensation for the harm that eventuates from them
suggests that risky acts also may be forbidden.
There are further and more utilitarian-sounding considerations
which Nozick also brings against allowing certain boundary
crossings and acts which significantly risk boundary crossings. These
considerations involve the fear which would be produced by allow-
ing such crossings or acts risking them—a fear that would afflict
even those whose boundaries turn out never to be crossed and who,
therefore, are not likely candidates for compensation. Even these
utilitarian-sounding considerations can be construed as pointing to
the absence of any adequate theory of compensation for victims of
boundary crossings.» However, we should note that Nozick takes
the arguments which turn on the fear provoking character of certain
boundary crossings and the fear inducing nature of acts which risk
certain boundary crossings to be separate from the division of
benefits argument. When contemplating the possibility of failure for
the argument from fearfulness for prohibiting boundary crossings,
Nozick says, “There would remain the argument about the division
of benefits of exchange.” (69) And, presumably, the force of the
division of benefits of exchange argument with respect to prohibit-
ing risky acts is equally independent of the fearfulness argument for
forbidding risky acts. Furthermore, even if all the fearfulness
arguments go through, Nozick still needs the division of benefits
arguments. For only by these arguments does Nozick establish, if he
establishes at all, the legitimacy of prohibiting non-feared violations
of rights and acts that risk non-feared violations of rights. The most
general argument, then, for prohibiting boundary impinging acts
and suggesting the prohibition of acts which risk the impingement of
boundaries is the division of the benefits of exchange argument. The
arguments from fearfulness will favor prohibiting an activity over
allowing it (while requiring that compensation be paid to its victims)
when that activity is fear provoking.
174 Eric Mack

The Principle of Compensation itself, however, expresses the

legitimacy of certain impinging actions, viz., prohibitions directed
against would-be risk-imposers. Are such impingements really to be
allowed (with compensation paid to those impinged upon) rather
than to be prohibited? It is not as though impingements against
would-be risk-imposers have an obvious or well-established place
within the tradition of Lockean state of nature theory. If such
impingements are allowable, justified by posterior compensation, it
must be that, in these cases, determining due posterior compensa-
tion is not subject to the problems which makes nonviable the policy
of allowing rights violating actions provided due posterior compen-
sation is paid for them. Spetifically, the division of the benefits of
exchange argument must not apply to “exchanges” in which absten-
tions from risky actions are secured. And fear-related arguments will
not apply against allowing these impingements because, presuma-
bly, these prohibitions are not fear provoking.
How might it be that the division of benefits argument does not
apply? If it did apply, the argument against allowing impingements
on would-be risk-imposers would run as follows: The impinging
party benefits from his exchange with the would-be risk-imposer.
Even returning the would-be imposer to his previous indifference
curve would distribute all the benefits of exchange to the impinging
party. Attempting, instead, to determine after the fact what the
would-be imposer and the impinger would have settled upon as the
price for the former's abstention is inferior to prohibiting the
impingement until and unless a bargain is struck. So, the argument
would conclude, if there are benefits for the impinging party in the
would-be imposer’s not imposing risks, then impinging on the
imposer should be prohibited—just as rights violating actions are
prohibited.
The crucial element, then, for applying the division of the
benefits of exchange argument would be the claim that the would-be
risk-imposer’s abstention benefits the party who would impinge to
force that abstention. The way, then, to maintain that the benefits
argument does not apply to such impingements is to deny that
abstentions from risky actions benefit those who seek these absten-
tions. The strategy described here is in fact the one adopted by
Nozick at the end of the penultimate section of “Prohibition,
Compensation and Risk” where he argues that,
 Unproductivity: The Unintended Consequences 175

In our earlier discussion of border crossing we noted the absence of any

compelling reason why all of the benefits of voluntary exchange should go to
one of the parties. Which of the admissible points on the contract curve was
to be selected, we said, was a question appropriately left to the parties
involved. This consideration favored prior negotiation over posterior pay-
ment offull compensation. In the present subclass of cases, however, it does
seem appropriate uniformly to select one extremity of the contract curve.
Unlike exchanges in which both parties benefit and it is unclear how these
benefits are to be divided, in negotiations over one party's abstaining from
an action that will or might endanger another person, all the first party need
receive is full compensation. (84)®

There is, of course, a puzzle to be overcome here. If abstentions

from risky actions do not benefit those who (non-altruistically) seek
them, why are they sought? Solving this puzzle is the task of the
final section of “Prohibition, Compensation and Risk” in which the
notions of unproductive exchange and unproductive activity are
introduced.

II

In this section we are asked, e.g., to imagine a case in which a

neighbor plans to erect an ugly (to you) but pleasing (to him)
structure on his own land. Displeasing as it is to you, he has every
right to do this and if you were to secure his abstention from
erecting this structure you would benefit. You would get what
otherwise you would not get—the non-agitation of your aesthetic
sensibilities. It is true, however, that you would be left no better off
than if your neighbor did not exist at all or had nothing whatsoever
to do with you.” For this reason Nozick says that the exchange to get
the non-construction satisfies a necessary condition for being an
unproductive exchange. The abstention from construction satisfies a
necessary condition for being an unproductive activity.
Imagine instead that the neighbor himself finds nothing lovely in
his planned monstrosity (and finds nothing lovely in your distaste for
it). Rather his only motive for planning the structure is to sell to you
his abstention from building it. In this case the second necessary
condition for unproductivity is also satisfied, viz., your securing this
abstention “merely gives you relief from something that would not
threaten if not for the possibility of an exchange to get relief from it.”
176 Eric Mack

(85)8 A sign that this crucial second condition for unproductivity is

satisfied is that if an exchange securing this non-construction “were
impossible or forcibly prohibited so that everyone knew [it] couldn't
be done, one of the parties to the potential exchange would be no
worse off.” (85) Indeed, the purchaser of non-construction would
have been better off had purchase of this activity been impossible or
forcibly prohibited. He would have gotten the non-construction at
no cost. For the neighbor, by hypothesis, has no motive in planning
to construct it except that of selling his abstention.® This neighbor's
abstention from construction satisfies both conditions for being an
unproductive activity.
Since unproductive activities do not benefit those who seek them,
the division of the benefits of exchange argument cannot be urged
against prohibitions which require unproductive activities. Notice
that the taint of unproductivity does not function (as one might
expect) to justify forbidding unproductive activities. Rather, an
activity’s unproductivity functions to legitimate requiring that ac-
tivity, e.g., to legitimate requiring the neighbor's non-construction
by prohibiting his construction. Due to its unproductivity, those
who require the activity (counting abstentions as activities) do not
benefit. Hence, the problems of dividing benefits do not arise.?° If,
then, there is no other reason against requiring an unproductive
activity, prohibitions may be issued which require that activity.
Specifically, if those activities requiring prohibitions do not them-
selves involve feared interventions, they may go forward. Thus it is a
startling and insufficiently noticed feature of Nozick’s position that
there are two quite separate strands of justification for prohibition.
One, of course, ties the prohibition of an act to its rights violating
character. The other ties the prohibition of an act to the unprod-
uctivity of abstention from that act.
Thus, after specifying the conditions for unproductivity, Nozick
endorses the prohibition of blackmail. And the argument turns not
at all on any purported rights violation by the blackmailer. Rather
we are told that,

Though people value a blackmailer’s silence, and pay for it, his being silent
is not a productive activity. His victims would be as well off if the
blackmailer did not exist at all, and so wasn’t threatening them. And they
 Unproductivity: The Unintended Consequences 177

would be no worse off if the exchange were known to be absolutely

impossible. (85)

Of course, different motives might lie behind my planning to

publish your secrets. I may want you, disturbed at the prospect of
my revelations, to buy me off. Such an exchange would uncom-
plicatedly be unproductive. There is no benefit at all to you in such
an exchange being possible. But, alternatively, I may plan to
publish because I want the royalties which will accrue to me
because others want to read about your secret vices. In such a case,
Nozick talks about what I may legitimately charge you for my
silence. I may charge, he says, “an amount of money equal to [the]
expected difference in royalties between the book containing this
information and the book without it.” (85) Yet this is not simply what
I may legitimately charge you if I choose to sell my silence to you. It
is the payment to me which, according to Nozick, justifies your
forbidding me to publish those secrets.
Now it is not clear why acquiring the silence of the more puntiely
oriented scandal-monger counts as an unproductive exchange.
Hence it is not clear why this scandal-mongering may be prohibited.
or, after all, if I am independently pecuniarily motivated to
publish your secrets, then when I am silent you do get relief from
something which threatened and not merely because of the possi-
bility of exchange with you to get relief from it. So this looks like the
case in which you get your neighbor not to erect a structure pleasing
to him. It looks like a case in which you benefit. Nozick’s point,
however, may be that once I am offered the difference between the
book royalties with the secrets included and the royalties with them
concealed, I could have no motive for planning to reveal your
secrets except pure blackmail. At this point you are threatened with
revelation only because of the possibility of exchange to get relief
from it.1}
How should we speak of such cases of layered motivation for
planning to do what another disvalues? Consider the possessor of
your secrets who is motivated to publish them by the expectation of
a $1,000 return over costs but who, knowing your fears, could
successfully insist upon more than $1,000 from you for his silence.
Suppose that, as Nozick proposes, he is prohibited from publishing
178 Eric Mack

those secrets and paid $1,000. That $1,000 buys you relief from
publication which threatens because of the interest of others in your
secrets. That aspect or part of the exchange is not unproductive. It’s
like paying your neighbor to forego his cherished edifice. So that
aspect or part of the secret-holder’s silence is not unproductive
activity. However any payment beyond the $1,000 would be un-
productive. Such payment would only buy relief from something
which would not threaten except for the possibility of such an
exchange to get relief from it. Notice that compared to an uncon-
strained bargaining situation you are better off if payments of over
$1,000 are forbidden. But you are not better off ifpayment of $1,000
is forbidden. For, were this payment forbidden also, the secret-
holder would turn to his scandal-loving readership. Thus, if a
portion of the freely negotiated price for a person's activity would be
paid solely because the seller is free to sell at that price or higher,
i.e., if less would suffice to elicit his activity were less all he could
demand, then an exchange for that activity (action or abstention) is
in part unproductive and the purchased activity is in part unprod-
uctive. We can, then, state Nozick’s view as: If an activity is all or in
part unproductive, it may be required as long as the party subject to
the prohibiting requirement is compensated for the productive part
or aspect of his activity. Freely negotiated sale of the activity is not
the preferred alternative. This account of the conditions of unprod-
uctivity and of the role for claims about unproductivity sets the stage
for a consideration of what activities fulfill these conditions and for
an evaluation of the doctrine that unproductive activities may be
required.

III

It is not hard to see that the second condition for unproductivity

viz., giving relief somewhat would not threaten except for the
possibility of exchange to get relief from it, is satisfied—at least in
the partial way which Nozick himself takes to be adequate to
satisfaction—in all typical free market exchanges. For in all typical
free market exchanges the purchase price of some activity, of some
service or transfer of goods, is as high as it is partially because of the
possibility of sale at that or higher prices. Characteristically, the
seller gets more than some price m which would have motivated
 Unproductivity: The Unintended Consequences 179

him to sell had exchange for more than m been forbidden or

impossible (and this was known to the seller). Once the seller has
been offered m—which characteristically is less than what uncon-
strained bargaining will settle upon—he withholds his activity solely
because of the possibility of eliciting a higher payment. If exchange
at above m were forbidden or impossible the buyer would be better
off.12 Just as you are better off when payments of more than $1,000
to your blackmailer are forbidden or impossible. Anything that the
buyer pays over m only purchases relief from what would not have
threatened given the offer of m except for the possibility of exchange
(purchase of abstention from the threatened action or abstention) for
more than m.
We can also see that in typical free market exchanges the activity
of the buyer also satisfies, in the same partial way, the second
condition for unproductivity. For “buyers” are just sellers of money
and “sellers” are just buyers of money. So the previous argument
applies symmetrically. Alternatively, we can think about the price
(in money) n where n > m and n is the price at which our original
buyer is minimally motivated to purchase what our original seller is
offering. The buyer will purchase for n if and only if exchange at less
than n is forbidden or impossible. (And he will not purchase for
more than n.) Typically free negotiation will settle upon a purchase
price of less than n. Hence, the seller would benefit from the
prohibition or impossibility of exchange for less than n. Whatever
the seller receives less than n is the cost to him of relief from the
withholding of the purchaser’s act—a witholding which would not
threaten given the offer to sell at n except for the possibility of
exchange at less than n. What the market is all about is discovering
what things you have or have access to or can produce which others
value more than you and providing these others with these goods or
services for more than you value them—where this value is the
value that you place on these goods or services independently of the
possibilities of exchange. This is why it should be no surprise that
typical free market exchanges satisfy, in the partial way indicated,
Nozick’s second condition for unproductivity.
If I am correct about this satisfaction of the second condition, then
adding this condition to the first can hardly serve to narrow
significantly the instances of unproductivity. The fact is that if
typical free market exchanges are not to count as unproductive, it
180 Eric Mack

must be because of the failure of the first condition, viz., the buyer
gains nothing from the seller that he would not possess if either the
seller did not exist at all or existed without having anything to do
with the buyer. Now it is reasonably clear that in very standard
cases of a seller's producing (or importantly contributing to the
production or delivery of) what the buyer chooses to purchase, this
condition for unproductivity is not satisfied. If B acquires a book
produced by S, then B gains something that he would not possess if
either S did not exist at all or existed without having anything to do
with B. There are, however, many interesting cases in which one
party does provide a good or service to another, this provision seems
to be part of a free market exchange, and yet this provision also seems
to satisfy both the first and second conditions for unproductivity.
Let us consider initially the complex case of boycotts (including
threats of boycott). Boycotts are also interesting because indepen-
dently they are evidence that there is no significant borderline
between hard market bargaining and blackmail. When X boycotts Y
he offers Y his future patronage solely on condition of Y's changing
his ways in some manner pleasing to X where this change is external
to the normal conception of what Y sells. Thus, X boycotts Y if he
indicates that he will not patronize retailer Y unless Y ends racial
discrimination in his hiring practices or unless Y begins racial
discrimination or unless Y supports this or that political campaign,
etc. Since what is demanded of Y is external to the normal concep-
tion of what Y offers on the market, Y will see the financial and
psychic costs of securing relief from the boycott as purchasing
something which he would have gotten anyway in the normal course
of events. He will be tempted to describe the boycott as “blackmail”
even more than we are all so tempted whenever the parties we are
bargaining with strongly resist our favored terms.
To consider boycotts in more detail imagine that a wicked retailer
racially discriminates in hiring and that there is no recourse in law
against him. The good guys among those who have been patronizing
this retailer pledge themselves to a boycott. They pledge to with-
hold their business from this retailer unless he agrees to abandon
discrimination. Their threat is based, let us suppose, on the expecta-
tion that the retailer will accede to their demands. Their hope is to
stop the discrimination. If they were to see that they could not
 Unproductivity: The Unintended Consequences 181

secure the retailer's non-discrimination, they would regretfully

continue their patronage. (But the retailer does not know this.) If
these boycotters and this retailer exchange continued patronage for
non-discrimination, their exchange clearly satisfies the second con-
dition for unproductivity. The retailer merely gets relief from
something which would not threaten him (the boycott) except for
the possibility of exchange to get relief from it. Also, the retailer
would be no worse off if the exchange of continued business from
the good guys for his non-discrimination were forbidden or impossi-
ble. The lifting of the boycott by the good guys satisfies the second
condition for being an unproductive activity.
Is the first condition for unproductivity also satisfied? Would the
retailer be better off if these patrons did not exist at all or had
nothing to do with him? At first glance, the answer is, no. The
retailer prefers the package of patronage and non-bigotry to having
those patrons go their separate way. Given his investment in this
operation at this site, he is better off with the crusading customers
than without them. But he would be better off still if these custom-
ers had never existed so that he would have invested elsewhere or
had they never had anything to do with him because he had invested
elsewhere. He, at least, would be no worse off. There is a clear
sense, then, in which unless the potential customers at alternative
sites would also have developed heightened racial sensibilities, the
retailer is worse off because of the existence of or because of his
contact with his actual customers. He may even be worse off than he
would have been had he been forbidden from ever becoming a
retailer. In the absence of Nozickean arguments to the contrary, it
seems plausible to claim that the lifting of such a boycott satisfies the
first as well as the second condition for unproductivity. Thus, it
seems that the lifting of the boycott does not benefit the retailer in a
way which makes applicable the division of the benefits of exchange
argument against the practice of forbidding the boycott while pro-
viding posterior compensation. If forbidding such a boycott is itself
not a fear provoking act, then no arguments remain in favor of prior
negotiation. Such a retailer, then, need not negotiate with the
boycotters. He (or his agents) may forbid the boycott as long as due
compensation is paid to the good guys if any is due. And it looks as
though no compensation would be due because the payment which
182 Eric Mack

the good guys would have negotiated for lifting the boycott, were
they allowed to boycott, is not part of their loss due to the prohibi-
tion of their boycott.!*
Let us turn to even more clearcut instances of free market
activities and exchanges which satisfy both conditions for unprod-
uctivity. Suppose that B is a roadbuilder who wants a right-of-way
through S’s cultivated field. S, I am supposing, has a Nozickean
entitlement to the field and hence to the right-of-way in virtue of the
transformations which he or others have wrought. But what B needs
is not this transformed land but merely the right-of-way and this
would have existed even if S (or earlier laborers upon this land) had
not. Surely if B were to purchase this land from S he would gain
nothing which he would not have possessed if S had not existed at all
or existed without having anything to do with B. (If someone else
would hold the land if S didn’t, then B’s purchase from that party
would be unproductive.) Of course, characteristically such an S will
have some basis for insisting upon payment from B other than the
possibility of exchange with B. Let us suppose, e.g., that S’s income
from his land is the same as the income available to S from $1,000.
In such a case, insofar as S insists on a $1,000 payment, he is
proposing a productive exchange!4—just as was the secret-holder
who charges $1,000 for his not including your secrets in his
scandal-mongering book. In both of these cases an exchange at
$1,000 provides the buyer with relief from something which would
have threatened (S keeping his land, your secrets being revealed)
even without the possibility of exchange between the relevant seller
and buyer. In the land case, it is $’s desire for the income from his
land which threatens B with S’s continuing to hold that land. But
any payment which S might receive from B in excess of $1,000 will
merely buy B relief from something which would not threaten
except for the possibility of exchange at over $1,000. B would not be
worse off if payments of over $1,000 for this land were impossible or
forbidden.
Thus, in such cases of exchange both conditions for unprod-
uctivity are satisfied—the second condition being satisfied in the
partial way which Nozick has allowed. How, then, may rightful
holders of such property be treated? They may be prohibited from
continuing to hold their property, i.e., they may be required to
 Unproductivity: The Unintended Consequences 183

surrender their property, as long as they are paid what would have
minimally motivated them to surrender these parcels of land had all
more generous offers been impossible or forbidden.'> In short,
Nozick’s views regarding unproductivity imply the legitimacy of the
practice of eminent domain, i.e., the legitimacy of taking desired
land from its rightful possessor with the payment of compensation
which does not cover the loss due to not being allowed freely to
negotiate a sale of that property. Indeed, such taking of what is the
rightful property of others seems to be permissible not only on the
part of some special “public authorities” but also on the part of
anyone.!®
What is central to his land seizure case is that had X who holds the
relevant good G which is desired by Y not existed, the G still would
exist and would be available to Y. In light of this central feature we
can construct a parallel Nozickean argument for the seizure of some
inherited property. Suppose that X holds some G as an inheritance,
but G would have been produced even if X had not existed (it was
not produced in order to be bequeathed) and G would be available
to Y if X did not exist. In such a case, Y's purchase ofG from X would
be an unproductive exchange. Y would be better off if X had not
existed at all and whatever Y pays to X over what would minimally
motivate X to sell merely provides Y with relief from something
which would not threaten (viz., X’s retaining G) except for the
possibility of exchange to get relief from it. In such a case, then, Y
may seize G—as long as he pays X what would minimally motivate X
to sell G were no more generous offers possible.*7
Clearly, for Nozick, these have to be discomforting results. For
surely it is in the spirit of his overall enterprise to condemn all
prohibitions of peaceful boycotts, takings by eminent domain, and
seizures of rightful inheritances. Interestingly, these results seem to
flow in part from a Marx-like aversion (built into the second condi-
tion for unproductivity) to persons’ perceiving and valuing the goods
which they might provide to others as objects for economic ex-
change, i.e., as commodities. The second condition for unprod-
uctivity seems to reflect the view that the production and distribu-
tion of goods is better, less alienating, and so on when directed by a
social perception of their usefulness rather than independent indi-
viduals’ perception of the capacity of goods to bring goods in
184 Eric Mack

exchange. And such a view is, of course, naturally associated with a

preference for command or customary economies over market
economies.

Li

At the outset of this essay I said that the central explicit task of the
chapter on “Prohibition, Compensation and Risk” was the develop-
ment of an account of how one may treat agents who engage in or are
about to engage in risky action. Within the overall plan of Anarchy,
State and Utopia the justification of the prohibition of risky acts
allows Nozick to sketch a process whereby one protective associa-
tion legitimately journeys toward statehood by prohibiting even the
risky protective acts of other associations and of independent defen-
ders and retaliators.18 Now clearly B would be no worse off if S, who
acts in some risky way toward B, did not exist or had nothing to do
with B. For this reason B’s purchase of S’s abstention from the risky
activity satisfies the first necessary condition for unproductivity in
exchange. The problem for Nozick is that the risk-imposers he is
concerned about, especially risky self-defenders and independent
protective associations, characteristically are not motivated to set
themselves to perform risky actions in order to sell their abstentions
from these acts. Rather, they have all sorts of independent motives
for pursuing their risky courses. For this reason, Nozick thinks that
he must acknowledge that in these cases the second condition for
unproductivity is not satisfied. Nevertheless, Nozick goes on to
insist on the propriety of forbidding these risky acts.

What of those cases where only the first condition of unproductive exchange
is satisfied, not the second: X is not better off as a result of the exchange
than if Y didn’t exist at all, but Y does have some motive other than selling
abstention. If from Y’s abstention from an activity X gains only a lessened
probability of having his own border crossed (a crossing whose intentional
performance is prohibited), then Y need be compensated only for the
disadvantages imposed upon him by the prohibition of only those activities
whose risk is serious enough to justify prohibition in this manner. (86)

Nozick is asserting, then, that if S’s act is a risky one, then even if
S’s abstention from that act fails to satisfy the second condition for
unproductivity in activity, that abstention may be still required of
 Unproductivity: The Unintended Consequences 185

S—just as if the second condition were satisfied. Yet how can this
be? For the failure of the second condition signifies that B benefits
from an exchange to secure S’s abstention. And since B benefits
from the exchange, it would seem that the division of benefits of
exchange argument would require that B negotiate with S to secure
S’s abstention. Granted, S is set to perform a risky action. But it
remains obscure how the riskiness of S’s action combines with the
abstention from that action’s satisfying (only) the first condition for
unproductivity to yield the legitimacy of forbidding S’s act. Since S’s
act is risky it comes close to being subject to the violation of rights
justification for prohibition. And since, as Nozick sees it, abstention
from S’s risky act satisfies one condition for unproductivity, this act
comes close to being subject to the unproductivity justification for
prohibition. But two such near misses do not seem to fit together so
as to constitute a hit.
Notice, however, that if I am correct about the much greater
extent to which the second condition for unproductivity is satisfied,
Nozick is in a better position regarding risky acts than he realizes.
For given the ease with which, in fact, the second condition is
realized, it is the satisfaction of the first condition which is pivotal for
the applicability of the unproductivity justification. And this first
condition is satisfied by (exchanges securing) abstentions from risky
acts. That the would-be imposer of risk has some independent
motive for his risky action simply makes him like the secret-holder
who has lucrative publishing prospects. A negotiated exchange
securing S’s abstention from a risky act will satisfy the second
condition in the partial manner discussed above if S is paid more
than would minimally motivate him to abstain from that act if freely
negotiated exchange were impossible or forbidden. What B pur-
chases with this incremental payment is relief from what would not
threaten once the minimally motivating offer was made except for
the possibility of exchange at a higher price. So, as in the case of the
would-be revealer of secrets, the would-be imposer of risk may be
prohibited from his actions provided he is offered this minimally
motivating amount. Thus, we can provide a Nozickean rationale for
prohibiting independently motivated risky actions provided com-
pensation is paid. However, full compensation seems to be re-
quired, not just compensation for disadvantages. Perhaps, some-
how, the riskiness of the prohibited acts plausibly works to reduce
186 Eric Mack

the required compensation.!9 But it is as hard to see how riskiness

cooperates with unproductivity to yield this result as it is to see how
riskiness cooperates with the first condition for unproductivity. The
two argumentative strands justifying prohibitions remain unwoven.
In any case, the cost to Nozick of having this rationale for something
at least close to his conclusion about risky actions is the endorse-
ment of the unlibertarian implications developed in Section III.
Beyond these problems, Nozick’s views with respect to unprod-
uctivity threaten to subvert the strongly deontic conception of rights
with which Anarchy, State and Utopia begins. As noted earlier, the
question which establishes the overall program for “Prohibition,
Compensation and Risk,” viz., Why ever prohibit? is surprising. It
is surprising because in the context of a strongly deontic conception
of rights an answer comes so readily to mind. That answer is, to
secure persons rights by preventing persons from doing what
morally they may not do. The question, Why ever prohibit? foretells
the appearance of a different conception of rights. Indeed, the
deontic conception is indicative of what “no person or group may
do” (ix) retreats, in “Prohibition, Compensation and Risk” before
another, more economic outcome oriented conception. According
to this outcome oriented conception, X’s right to G consists in X’s
having a claim against others’ taking G (or preventing his G-ing)
without compensating X appropriately. If X’s surrender of G is
unproductive, then appropriate compensation will at most be full
compensation. If X’s surrender of G is productive, then appropriate
compensation will, it seems, be market compensation. On this
conception, although not on the deontic conception, a system in
which others’ taking G from X is always accompanied by X’s being
compensated duly would fully protect X’s rights. If only we could
insure that X is always compensated fittingly for any (apparent)
boundary crossing, we could insure the protection of X’s rights
without ever instituting prohibitions (except, perhaps, for prohibit-
ing the non-payment of due compensation). The shift to this out-
come orientation is a product of making the wrong of boundary
crossing rest upon the inadequacies of schemes of posterior com-
pensation for boundary crossing. And, in turn, the wrong of bound-
ary crossing is given this construction so that the prohibitions which
are justified in the name of unproductivity, e.g., the prohibitions of
blackmail, will not count as wrongful boundary crossings. Yet we
 Unproductivity: The Unintended Consequences 187

have noted that the unproductivity justification for prohibitions is

distinct and separate from the rights violation justification for pro-
hibitions. It should not be surprising, then, that accommodating
prohibitions in the name of unproductivity involves an alteration in
the conception of boundaries.
On the resulting outcome oriented conception of rights, a bound-
ary specifies a level of well-being and the permissibility of others’
actions depends upon the effect of those actions upon the subject's
wellbeing. Even if what is forcibly required of the subject is a
productive and beneficial activity, no wrongful boundary crossing
will take place as long as the level of the subject's wellbeing is
fittingly raised. This shift to an outcome oriented conception of
rights should make it difficult for Nozick to sustain his anti-
paternalism. For the subject of paternalistic intervention will have
no complaint in terms of his rights if the intervention does actually
maintain or appropriately raise his level of wellbeing. In contrast,
on the more consistently deontic conception, a boundary is a
frontier which others do wrong to cross and accompanying benefits
do not right such wrongs. Certainly in his intuitive appeal to
“the Kantian principle that individuals are ends and not merely
means; they may not be sacrificed or used for the achieving of
other ends without their consent,” (31-21) Nozick is invoking
this more consistently deontic conception. But his deontic con-
ception of rights is lost when even having a morally unambiguous
boundary is understood as being on a certain indifference curve
from which others may only move one outward, towards higher
indifference curves, at least a fair distance, if there is a fair distance.

NOTES

*For their helpful discussion or listening, I thank Robert Mackay, Robert

Nozick, and Mary Sirridge. Such thanks in no way attributes agreement or
responsibility.
1. Robert Nozick, Anarchy, State and Utopia (New York: Basic Books,
1974).
2. The Principle of Compensation is expressed in terms of compensation
for disadvantages imposed. Nozick holds that one can be made worse off by
not being permitted some activity without being disadvantaged. Thus,
compensating a person for disadvantages imposed may not fully compensate
him. Nozick makes little attempt to clarify or develop these distinctions.
188 Eric Mack

3. In fact, prohibiting is defined as requiring more of a transgressor than

compensation for damages inflicted. (57)
4. The term “Nullification” is my own, not Nozick’s. But see footnote 8 of
chapter seven. (345)
5. We should note the exceptions to Nozick’s rejection of boundary
crossing accompanied by compensation. Non-feared crossings with com-
pensation are to be permitted when “prior consent iS impossible or very
costly to negotiate” (72) and the benefits of the crossings are high. The
appropriate compensation for such a crossing would seem to be market
compensation. But Nozick seems inclined toward mere full compensation.
At least on p. 86 he indicates that in an “ordinary border crossing” the
victim should be raised “to the position he was in before he was interferred
with.” Question: Does Nozick’s stand on these exceptions to prohibitions on
boundary crossings indicate the stand he would take regarding coercive
taxation to finance the provision of public goods, i.e., goods which purpor-
tedly would not otherwise be financed because sufficiently widespread
consent to contribute to their production “is impossible or very costly to
negotiate?”
6. Also see the restatement of this strategy on 145. While it seems to me
that Nozick is correct to say that “all the first party need receive is full
compensation,” this is not Nozick’s own considered conclusion. For more
commonly (e.g., on 87 and 145 and in the very expression of the Principle of
Compensation) Nozick holds that this party need only receive compensation
for disadvantages inflicted. I try to say something later in this essay about
where compensation only for disadvantages may come from.
7. Nozick notes ambiguities and complexities with this formulation.
Suppose that on other occasions this neighbor benefits you. Suppose that
were this neighbor not there worse neighbors would be.
8. Although Nozick might be read as attempting to replace the earlier
condition which has been discovered to be merely necessary with a new
condition which is necessary and sufficient the more generous and plausible
reading is that there are two necessary conditions which are supposed to be
jointly sufficient. Cf. 145.
9. How can we know that the second condition is satisfied? Perhaps by
simply refusing to pay for the desired activity, e.g., the non-construction.
That is, perhaps we discover that the second condition is satisfied by calling
what we hope is the other party’s bluff. This suggests that any scheme of
justification for prohibiting, e.g., the monstrous construction, which waits
upon the satisfaction of this second condition is pointless. For, it seems, we
will know that the second condition is satisfied only after successfully calling
the other party’s bluff, e.g., seeing him not engage in construction. But if
the bluff has been successfully called, there is no need for prohibition.
 Unproductivity: The Unintended Consequences 189

However, there might be cases in which the neighbor sets himself to erect
the offending building solely in order to bargain for his abstention but,
having so set himself (or having, thereby, his reputation on the line) he will
proceed with construction if his price is not met.
10. I do not challenge here the truth of the claim that one party does not
benefit from an unproductive exchange (i.e., from securing the other's
unproductive activity). One is tempted, though, to answer Nozick’s remark,
“A strange kind of productive exchange it would be whose forbidding leaves
one party not worse off!” (85) with “A strange kind of voluntary exchange it
would be whose forbidding did not leave all the parties better off ex ante!”
Alternatively, one might emphasize that the issue is whether a given party
is better or worse off with respect to his rights. Neither party to an
unproductive exchange has his rights violated. Each surrenders a right and
each acquires a new right (e.g., the right to the neighbor’s not building the
displeasing structure). And, presumably, each prefers the alteration of
rights to the status quo.
11. Nozick also notes the case in which I delight in revealing your
secrets. (86, note) According to Nozick, in this case I “may charge differ-
ently” than the party who “may not charge the best price he could get from
the purchaser of his silence.” (85-86) Does this mean that taking this
delight, I may charge whatever the market will bear?
12. Iam not suggesting that in particular cases this m can be known. On
the contrary, it will characteristically not be known—even to the seller.
This suggests a severe practical limit to the application of any principle
which requires that we know, for compensation purposes, just what m is.
Nozick might welcome such a practical limitation. But these issues cannot
be discussed further here.
13. This sentence follows the phrasing of Nozick’s final sentence in the
section “The Principle of Compensation.” (84) Although in general Nozick is
discussing risky actions, his final sentences in this section are about what
can be said when the division of the benefits of exchange argument does not
apply and require that an activity not be forbidden.
14. However, this income itself might be the result of unproductive
exchanges. It might, for instance, be the result of S’s sale of the wild berries
which grow on this land and which would be more available to the buyers
were S not to exist.
15. For the sake of economy, I have identified what leaves S indifferent
and what minimally motivates S. This is a common way of speaking. But if
one wants to hold that S would not be motivated to sell by the prospect of an
exchange moving him to a different point on the indifference curve he
already occupies, then one will want to say that C may legitimately charge
$1,000.01
190 Eric Mack

16. Presumably, holding everyone’s preferences set, the land will tend
to be held by the party with the strongest independent desire for it. For it
would not pay anyone else to take the land from and then compensate this
party.
17. Suppose that X has inherited money (representing wealth which
would have existed even if he hadn't existed). X’s compensation for having
all this money taken would be the smallest basket of goods and services
which he would have bought with that money if purchase of larger baskets
were impossible or forbidden.
18. In “Nozick’s Anarchism” in Anarchism (NOMOS XIX) edited by
J. R. Pennock and J. W. Chapman (New York: New York University Press,
1978), I argue this journey does not bring the protective association to
statehood.
19. On pp. 86-87 Nozick seems to anticipate something like my recon-
struction and warns against it by saying that “. . . this does not explain why
all are not returned to the indifference curve they would occupy were it not
for the prohibition.” But how does the riskiness of their proposed actions
show that they should not be so compensated. Nozick suggests that to
require full compensation from the prohibitors of risky actions is to treat
them like ordinary boundary-crossers. But it seems that ordinary crossers
should have to provide market compensation, not mere full compensation.
If neither the cooperation between riskiness and the first condition of
unproductivity nor the cooperation between riskiness and unproductivity
can be spelled out, prohibitions of risky acts must depend entirely upon
their riskiness or entirely upon the unproductivity of the activities thereby
required. I have shown how Nozick has easy access to the latter option. At
times (cf. p. 82) Nozick seems far less hesitant about the prohibition of
actions on the basis of their riskiness to rights. At these times the only worry
seems to be about individuals whose well-being is intimately tied up, and in
unusual ways, with their performance of risky actions.
 10
Libertarianism Without
Foundations
THOMAS NAGEL

Liberalism is the conjunction of two ideals. The first is that of

individual liberty: liberty of thought, speech, religion, and political
action; freedom from government interference with privacy, per-
sonal life, and the exercise of individual inclination. The second
ideal is that of a democratic society controlled by its citizens and
serving their needs, in which inequalities of political and economic
power and social position are not excessive. Means of promoting the
second ideal include progressive taxation, public provision of a
social minimum, and insulation of political affairs from the excessive
influence of private wealth. To approach either of these ideals is
very difficult. To pursue both of them inevitably results in serious
dilemmas. In such cases liberalism tends to give priority to the
respect for certain personal rights, even at substantial cost in the
realization of other goods such as efficiency, equality, and social
stability.
The most formidable challenge to liberalism, both intellectually
and politically, is from the left. It is argued that strong safeguards of
individual liberty are too great a hindrance to the achievement of
economic and social equality, rapid economic progress from under-
development, and political stability. A majority of the people in the

191
192 Thomas Nagel

world are governed on this assumption. Perhaps the most difficult

issue is posed by economic power and the political inequality it can
create. The criticism from the left is that harmful concentrations of
economic power cannot be attacked—or prevented from
forming—unless individual actions are more closely restricted than
is permitted by the liberal ideal of personal freedom. Radical
redistribution is unlikely in a liberal democracy where private
wealth controls the political process. A defense against this criticism
must either challenge the factual claim or argue that the importance
of freedom outweighs these disadvantages.
Liberalism is also under attack from the right. The most con-
spicuous attacks are not theoretical: the right in its more prominent
political manifestations is not particularly attached to individual
liberty when that liberty threatens the unequal distribution of
wealth and power. But there is also a theoretical challenge from the
right, called libertarianism, and while it does not present as serious
a moral issue for liberals as does the attack from the left, the two are
in some ways symmetrical. Libertarianism, like leftism, fastens on
one of the two elements of the liberal ideal and asks why its
realization should be inhibited by the demands of the other. Instead
of embracing the ideal of equality and the general welfare, liber-
tarianism exalts the claim of individual freedom of action, and asks
why state power should be permitted even the interference repre-
sented by progressive taxation and public provision of health care,
education, and a minimum standard ofliving.
In Anarchy, State, and Utopia,! Robert Nozick attempts to set
forth the libertarian position in a way that will persuade some of
those who do not already accept it. Despite its ingenuity of detail,
the effort is entirely unsuccessful as an attempt to convince, and far
less successful than it might be as an attempt to explain to someone
who does not hold the position why anyone else does hold it. The
book may come to occupy the position of an official text of libertarian
political theory, but it is unlikely to add to the ranks of believers in
that view unless it converts a few unwary philosophical anarchists by
persuading them that the minimal state need not after all violate
their austere moral requirements.
To present a serious challenge to other views, a discussion of
libertarianism would have to explore the foundations of individual
rights and the reasons for and against different conceptions of the
 Libertarianism Without Foundations 193

relation between those rights and other values that the state may be
in a position to promote. But Nozick’s book is theoretically insub-
stantial: it does not take up the main problems, and therefore fails to
make the kind of contribution to political theory that might have
been hoped for from someone of his philosophical attainments.? In
the preface he announces that he was converted to libertarianism by
the decisive force of the arguments,? but no such arguments appear
in the book. He has left the establishment of the moral foundations
to another occasion, and his brief indication of how the basic views
might be defended is disappointing. I shall explain below why it is
unlikely to survive further development.
Nozick starts from the unargued premise that individuals have
certain inviolable rights which may not be intentionally transgressed
by other individuals or the state for any purpose. They are the rights
not to be killed or assaulted if one is doing no harm, not to be
coerced or imprisoned, not to have one’s property taken or de-
stroyed, and not to be limited in the use of one’s property so long as
one does not violate the rights of others. He concludes that the only
morally permissible state would be the minimal nightwatchman
state, a state limited to protecting people against murder, assault,
theft, fraud, and breach of contract. The argument is not one which
derives a surprising conclusion from plausible premises. No one
(except perhaps an anarchist) who did not already accept the conclu-
sion would accept the premise, and the implausibility of each can
only serve to reinforce a conviction of the implausibility of the other.
Naturally any opposition to the power of governments will meet
with a certain sympathy from observers of the contemporary scene,
and Nozick emphasizes the connection between his view and the
fight against legal regulation of sexual behavior, drug use, and
individual life styles. It is easy to develop an aversion to state power
by looking at how actual states wield it. Their activities often include
murder, torture, political imprisonment, censorship, conscription
for aggressive war, and overthrowing the governments of other
countries—not to mention tapping the phones, reading the mail, or
regulating the sexual behavior of their own citizens.
The objection to these abuses, however, is not that state power
exists, but that it is used to do evil rather than good. Opposition to
these evils cannot be translated into an objection to welfare, public
education, or the graduated income tax. A reasonably persuasive
194 Thomas Nagel

can
practical argument for reducing the power of governments
perhaps be based on the unhapp y results of that power. But it is
doubtful that a government limited to the functions of police,
be-
courts, prisons, and national defense would be conspicuously
nign, or that it would be especial ly protect ive of individu al rights.*
In practice, it would probably include the worst parts of what we
have now, without much of the best. That is why those concerned
with individual liberty are usually not opposed to strong govern-
ment with power to promote desirable ends, so long as the exercise
of that power is limited by strong safeguards. Governments should
promote what is good and prevent evils, as well as protecting rights.
How could anyone disagree?
If there is an answer to this question, it must come from the
ethical foundation of political theory. Nozick states:

Moral philosophy sets the background for, and the boundaries of, political
philosophy. What persons may and may not do to one another limits what
they may do through the apparatus of a state, or do to establish such an
apparatus. The moral prohibitions it is permissible to enforce are the source
of whatever legitimacy the state’s fundamental coercive power has.°

I believe that this principle is correct and important. The exercise of

state power is not the action of a separate entity with moral rights
greater than those of individual persons, rights to use force against
persons for reasons that would not justify the use of force by
individuals or groups of individuals per se. If governments have the
right to coerce, it must be a right possessed by the people who
establish and sustain governmental institutions, and those who act
through them.
There is a problem about stating this position in a way that avoids
triviality. For someone who believes that governments have much
larger rights than individuals could always add that the existence of
such rights implies a corresponding individual right to combine with
others to institute a government and act through it to exercise those
larger rights of coercion and control. But in such a view, these
individual rights would be derivative from the rights of the state,
and not the other way around. Nozick’s position, which seems
correct, is that individual rights and duties are the basis of what
governments may and should do.
But he appears to infer from this ethical principle a strong
 Libertarianism Without Foundations 195

epistemological consequence which it does not have: that it is

possible to determine what governments may and should do by first
asking what individuals, taken a few at a time in isolation from
large-scale society, may do, and then applying the resultant princi-
ples to all possible circumstances, including those which involve
billions of people, complicated political and economic institutions,
and thousands of years of history. What is more surprising, he
discovers in himself intuitions about the moral requirements on
men in a state of nature which he is willing to endorse as universal
principles unmodified in their cumulative effects when applied in
any circumstances whatever.
Abstractly described, this procedure sounds hopelessly mis-
guided.® It is hard to see how anyone could seriously arrive at firm
moral opinions about the universal principles of human conduct
without considering what it would be like if they were universally
applied, in iterations which might create complex effects of scale.
When we pass from an abstract to a more substantive description,
the implausibility of the view increases. For the intuition that
Nozick discovers in himself is that everyone has an absolute right to
be free from coercion, and an absolute right to acquire and dispose
of his property—so long as he is not violating the same rights of
others and so long as his acquisition of property does not, for
example, give him sole title to the formerly public water supply of a
desert community.”
Nozick’s intuition is that each person is entitled to his talents and
abilities, and to whatever he can make, get, or buy with his own
efforts, with the help of others, or with plain luck. He is entitled to
keep it or do anything he wants with it, and whomever he gives it to
is thereby equally entitled to it. Moreover, anyone is entitled to
whatever he ends up with as a result of the indefinite repetition of
this process, over however many generations. I assume that most
readers of Nozick’s book will find no echo of this intuition in
themselves, and will feel instead that they can develop no opinion
on the universal principles of entitlement, acquisiton and transfer of
property, or indeed whether there are any such universal princi-
ples, without considering the significance of such principles in their
universal application. One might even agree in part with Nozick’s
views about what people should do in the limited circumstances that
define interpersonal relations in the state of nature, but not agree
196 Thomas Nagel

that the proper generalization of those judgments is their un-

modified application to all cases no matter how complex or ex-
tended. They might be based instead on principles which give these
results for small-scale individual transactions but rather different
results for the specification of general conditions of entitlement to
be applied on an indefinitely large scale.§
The fact is, however, that Nozick’s moral intuitions seem wrong
even on a small scale. He denies that any of the rights he detects
may be overridden merely to do good or prevent evil. But even if it
is not permissible to murder or maim an innocent person to promote
some highly desirable result, the protected rights do not all have the
same degree of importance. The things one is supposed to be
protected against are, in order of gravity; killing, injury, pain,
physical force, deprivation of liberty of many different kinds
(movement, assocation, and activity), destruction of one’s property,
taking of one’s property; or the threat of any of the above (with all
their variations in gravity). It is far less plausible to maintain that
taking some of an innocent man’s property is an impermissible
means for the prevention of a serious evil, than it is to maintain that
killing him is impermissible. These rights vary in importance and
some are not absolute even in the state of nature.
The sources of morality are not simple but multiple; therefore its
development in political theory will reflect that multiplicity. Rights
limit the pursuit of worthwhile ends, but they can also sometimes be
overridden if the ends are sufficiently important. The only way to
make progress in understanding the nature of individual rights is to
investigate their sources and their relations to each other and to the
values on whose pursuit they set limits. Nozick says little about the
basis of the inviolability of persons, but the following remark
indicates where he would be inclined to look:

[W]hy may not one violate persons for the greater social good? Individually,
we each sometimes choose to undergo some pain or sacrifice for a greater
benefit or to avoid a greater harm: we go to the dentist to avoid worse
suffering later; we do some unpleasant work for its results: some persons
diet to improve their health or looks; some save money to support them-
selves when they are older. In each case, some cost is borne for the sake of
the greater overall good. Why not, similarly, hold that some persons have
to bear some costs that benefit other persons more, for the sake of the
overall social good? But there is no social entity with a good that undergoes
some sacrifice for its own good. There are only individual people, different
 Libertarianism Without Foundations 197

individual people, with their own individual lives. Using one of these
people for the benefit of others, uses him and benefits the others. Nothing
more. What happens is that something is done to him for the sake of others.
Talk of an overall social good covers this up. (Intentionally?) To use a person
in this way does not sufficiently respect and take account of the fact that he
is a separate person, that his is the only life he has. He does not get some
overbalancing good from his sacrifice, and no one is entitled to force this
upon him—least of all a state or government that claims his allegiance (as
other individuals do not) and that therefore scrupulously must be neutral
between its citizens.9

It is not clear how Nozick thinks individual rights derive from the
fact that each person’s life is the only one he has. He appears to draw
the implication that a benefit to one or more persons can never
outweigh a cost borne by someone else. This, however, is far too
broad a claim for Nozick’s purposes. It is both obviously false and
unsuitable as a basis for constraints on the treatment of individuals.
To make sense of interpersonal compensation it is not necessary
to invoke the silly idea ofa social entity, thus establishing an analogy
with intrapersonal compensation. All one needs is the belief, shared
by most people, that it is better for each of 10 people to receive a
benefit than for one person to receive it, worse for 10 people to be
harmed than for one person to be similarly harmed, better for one
person to benefit greatly than for another to benefit slightly, and so
forth. The fact that each person’s life is the only one he has does not
render us incapable of making these judgments, and if a choice
among such alternatives does not involve the violation of any rights
or entitlements, but only the allocation of limited time or resources,
then we regard those comparisons as excellent reasons for picking
one alternative rather than the other. If we can help either 10
people or one person, not included in the 10, and we help the 10,
then we can say that rescue of the 10 outweighs the loss of the one,
despite the fact that he does not get some overbalancing good from
his sacrifice, and his is the only life he has.
So for the purpose of comparing possible outcomes of action,
where the violation of rights is not in question, it is clear that the
distinctness of individuals does not prevent balancing of benefits
and harms across persons. If special constraints enter in when a
sacrifice is to be imposed on someone as a means to the achievement
of a desirable outcome, their source must lie elsewhere. Such
198 Thomas Nagel

constraints should not derive from a principle which also has the
consequence that practically nothing can be said about the relative
desirability of situations involving numbers of different people.
Furthermore, the source of rights of the general kind Nozick
advocates cannot be discovered by concentrating, as he suggests we
should, on the meaning of individual human lives and the value of
shaping one’s own life and forming a general conception of it. Vague
as his suggestions are,!° they all suffer from an error of focus, for
they concentrate solely on features of persons that make it bad for
certain things to happen to them, and good for them to have the
opportunity to do certain things. But rights of the kind that interest
Nozick are not rights that certain things not happen to you, or rights
to be provided with certain opportunities. Rather they are rights not
to be deliberately treated or used in certain ways, and not to be
deliberately interfered with in certain activities. They give rise to
claims not against the world at large, but only against someone who
contemplates deliberately violating them. The relation between the
possessor of the right and the actor, rather than just the intrinsic
nature of the possessor and of his life, must enter into the analysis of
the right and the explication of its basis.
Any theory of rights must explain this structural feature, even if it
does not follow Nozick in elevating the unimpeded exercise of the
will into the supreme principle of morality. It is of the first im-
portance that your right not to be assaulted is not a right that
everyone do what is required to ensure that you are not assaulted. It
is merely a right not to be assaulted, and it is correlated with other
people's duty not to assault you. This cannot be explained simply by
the fact that it is bad to be assaulted, which is merely an item in the
catalogue of values by which the desirability or undesirability of
occurrences or sets of occurrences is to be weighed. That assault is
disagreeable or bad does not explain why the prohibition of it should
serve as a constraint on the pursuit of other values or the avoidance
of other harms, even if those other values outweigh the badness of
assault in a pure calculation of the relative desirability of possible
outcomes. Sometimes one is required to choose the less desirable
alternative because to achieve the more desirable one would have to
violate a right.
As Nozick points out,! the constraints on action represented by
rights cannot be equivalent to an assignment of large disvalue to
 Libertarianism Without Foundations 199

their violation, for that would make it permissible to violate such a

right if by doing so one could prevent more numerous or more
serious violations of the same right by others. This is not in general
true. It is not permissible in Nozick’s view (or mine) to kill an
innocent person even to prevent the deliberate killing of three other
innocent persons. A general feature of anything worthy of being
called a right is that it is not translatable into a mere assignment of
disvalue to its violation.
An explanation of the basis of rights would therefore have to
concentrate on the actor and his relation to the person he is
constrained not to treat in certain ways, even to achieve very
desirable ends. And it would have to explore the interaction be-
tween those constraints, and the goals whose pursuit they constrain.
There is no reason to think that either in personal life or in society
the force of every right will be absolute or nearly absolute, i.e.,
never capable of being overridden by consequential considerations.
Rights not to be deliberately killed, injured, tormented, or impris-
oned are very powerful and limit the pursuit of any goal. More
limited restrictions of liberty of action, restrictions on the use of
property, restrictions on contracts, are simply less serious and
therefore provide less powerful constraints.12
Moreover, there is a big difference between suddenly exprop-
riating half of someone’s savings and attaching monetary conditions
in advance to activities, expenditures, and earnings—the usual form
of taxation. The latter is a much less brutal assault upon the
person.1? Whether this kind of limitation of individual liberty
should be permitted, to acquire resources for the promotion of
desirable ends, is a function of the gravity of the violation and the
desirability of the ends. (And as I have observed, this does not mean
that it is justified whenever the result is a maximal social balance of
benefits and costs.)
Nozick would reply that such ends can be achieved by voluntary
donations rather than by compulsion, and that people who are
well-off and who deplore the existence of poverty should donate
significant portions of their assets to help those who are unfortu-
nate.14 But this is no more plausible coming from Nozick than it was
coming from Barry Goldwater. Most people are not generous when
asked to give voluntarily, and it is unreasonable to ask that they
should be. Admittedly there are cases in which a person should do
200 Thomas Nagel

something although it would not be right to force him to do it. But

here I believe the reverse is true. Sometimes it is proper to force
people to do something even though it is not true that they should
do it without being forced. It is acceptable to compel people to
contribute to the support of the indigent by automatic taxation, but
unreasonable to insist that in the absence of such a system they
ought to contribute voluntarily. The latter is an excessively de-
manding moral position because it requires voluntary decisions that
are quite difficult to make. Most people will tolerate a universal
system of compulsory taxation without feeling entitled to complain,
whereas they would feel justified in refusing an appeal that they
contribute the same amount voluntarily. This is partly due to lack of
assurance that others would do likewise and fear of relative disad-
vantage; but it is also a sensible rejection of excessive demands on
the will, which can be more irksome than automatic demands on the
purse.
A political theory that reflected these moral complexities would
assign society the function of promoting certain goods and prevent-
ing certain evils, within limits set by the differing constraints of
different individual rights. It would not judge processes and proce-
dures solely by their tendency to produce certain outcomes, nor
would it judge outcomes solely by the processes that had produced
them. Social institutions and the procedures defining them would
be assessed by reference both to their respect for individual rights
and liberty, and to their tendency to promote desirable ends like the
general welfare.
Nozick offers a classification of principles of distributive justice
into which such a theory does not fit.45 After defining a historical
principle as one which asserts that the justice of a distribution
depends on how it came about,!® and an end-result principle as one
which denies this,!7 he defines a patterned principle as one which
specifies “that a distribution is to vary along with some natural
dimension, weighted sum of natural dimensions, or lexicographic
ordering of natural dimensions.”!8 His own theory, the entitlement
theory, is easy to describe in these terms. It is a nonpatterned
historical principle, for it specifies that any distribution is just if it
was’ arrived at by a series of individual transactions among people
entitled, by natural rights of acquisition and transfer, to make them.
But suppose a theory says that a distribution is just if it results
 Libertarianism Without Foundations 201

from a process governed by rules that reflect (a) the suitability of

certain patterns, (b) the desirability of increasing certain good
results and decreasing certain evils independently of any pattern,
and (c) a respect for individual rights of differing importance. Such a
theory will be at bottom neither purely historical nor purely pat-
terned. It will be formally historical, but the “historical” or process
criterion will be partially determined by considerations of pattern
and considerations of total outcome. Therefore Nozick’s concen-
trated attack on patterned principles and nonhistorical principles
provides no reason to think that his alternative is correct.!9
Apart from this defect, the attack is still unsuccessful. Nozick asks
us to imagine some patterned principle realized, and then argues
that its preservation would require interference with individual
liberty: peopie would have to be prevented from using their alloca-
tions under the pattern as they wish. For example, preservation of a
reasonably equal distribution would require that individuals not be
permitted to pay Wilt Chamberlain 25 cents for each basketball
game they see him play with the understanding that he can keep it
all, even if it amounts to $250,000 a year. This is perfectly obvious,
and it is part of what would be meant by a patterned principle of
distribution: the adoption of a general system of acquisition, taxation
and exchange that tends to preserve a certain pattern.
It only seems a problem to Nozick, and a further violation of
liberty, because he erroneously interprets the notion of a patterned
principle as specifying a distribution of absolute entitlements (like
those he believes in) to the wealth or property distributed. But
absolute entitlement to property is not what would be allocated to
people under a partially egalitarian distribution. Possession would
confer the kind of qualified entitlement that exists in a system under
which taxes and other conditions are arranged to preserve certain
features of the distribution, while permitting choice, use, and
exchange of property compatible with it. What someone holds under
such a system will not be his property in the unqualified sense of
Nozick’s system of entitlement. To suppose otherwise is to beg the
question, and that is exactly what Nozick does when he says:

There is no question about whether each of the people was entitled to the
control over the resources they held in D1 [the original patterned dis-
tribution]; because that was the distribution (your favorite) that (for the
202 Thomas Nagel

purposes of argument) we assumed was acceptable. . . . If the people were

entitled to dispose of the resources to which they were entitled (under D1),
didn’t this include their being entitled to give it to, or exchange it with, Wilt
Chamberlain ?2°

This mistake drains the argument of its force.

Let me turn briefly to the difficult subject of equality. While the
elimination of misery, poverty, and disease are probably more
important social goals than the achievement of economic equality,
the latter is one of the ends some people regard as legitimate.
Nozick’s view is predictable. If inequality arises as a result of the
free exercise of entitlements, it cannot be objected to on grounds of
injustice, and liberty may not be in any way infringed to reduce it.
Since people are entitled not only to the wealth they inherit but to
their natural assets,?! further inequalities resulting from the
employment of these assets are just.
But there is no reason to believe in an absolute natural right to
profit from one’s natural assets, even if a less than absolute right to
their free exercise is acknowledged as a limitation on the pursuit of
equality or other social goals. Someone who regards equality as a
good will assume that its achievement does not take absolute
precedence over efficiency, liberty, and other values. Nevertheless,
more than this is required to answer Nozick, for it is not clear what
makes equality of distribution a good thing at all. Nozick does not
acknowledge the right of the state to limit liberty to produce any
merely desirable outcome. But why should someone with a more
standard view about individual rights be in favor of a goal of social
and economic equality? Perhaps he can argue that the average level
of well-being—both in material terms and in terms of contentment
and self-esteem—is likely to be higher in a relatively equal society
than in an unequal one of comparable total wealth. Perhaps he will
argue that the political effects of economic inequality are harmful to
individual liberty and general welfare. But these considerations,
though very important, are not reasons for regarding equality as a
good in itself; yet that is a common moral view.
It cannot be defended by claiming that inequalities are arbitrary
unless based on morally relevant differences among people. Arbit-
rariness is a moral defect only if it can be contrasted with an
alternative that is selected on the basis of morally relevant factors.
 Libertarianism Without Foundations 203

Unless there is independent justification for equality, an equal

distribution is just as arbitrary from a moral point of view as any
other. To defend equality as a good in itself, one would have to
argue that improvements in the lot of people lower on the scale of
well-being took priority over greater improvements to those higher
on the scale, even if the latter improvements also affected more
people. While I am sympathetic to such a view, I do not believe it
has ever been successfully defended.
I have dwelt on the book's theoretical shortcomings; there is
much in it that I have not discussed, including a final chapter which
describes a pluralistic libertarian utopia,2? and interesting analyses
of such diverse topics as Marx’s labor theory of value2? and the
treatment of animals.?4 Unlike most works of philosophy, it is funny,
fast-paced and a pleasure to read. Nozick’s writing, though inele-
gant, has great energy, and meets a high standard of clarity and
definiteness. One is rarely in doubt about what Nozick is claiming,
or about what one denies in rejecting his views.

NOTES

1. R. Nozick, Anarchy, State, and Utopia (1974) [hereinafter cited to

page number only].
2. Nozick is the author of three important articles: “Coercion,” in
Philosophy, Science, and Method 440 (S. Morgenbesser, P. Suppes & M.
White eds. 1969); “Newcomb’s Problem and Two Principles of Choice,” in
Essays in Honor of Carl G. Hempel 114 (N. Rescher ed. 1970); “Moral
Complications and Moral Structures,” 13 Nat. L. Forum 1 (1968). The book
reaches their level of trenchancy only in Chapter 4, “Prohibition, Compen-
sation, and Risk,” a brilliant discussion of the choice among various methods
of dealing with injurious or dangerous behavior: when to prohibit, when to
punish, when to require compensation, when to compensate someone who
is inconvenienced by a prohibition. It is also the chapter with the greatest
importance for legal theory. Pp. 54-88.
Bo LP Tbk
4. This helps to account for the romantic appeal of anarchism. Nozick’s
attempt to refute the anarchist view that even a minimal state will violate
individual rights is not, I think, successful. He argues at length that a
minimal state could arise by an invisible process from a state of nature
without the process violating anyone's rights: people could voluntarily join
private protective associations, one of which would naturally achieve domi-
nance over a territory even if not everyone had agreed to join. It could then
204 Thomas Nagel

exercise limited control without violating anyone’s rights. This is supposed

to show that a minimal state is morally permissible. But why should the
mere conceivability of such a process persuade an anarchist of that conclu-
sion? He would already have been prepared to admit that a minimal state
established by unanimous agreement of the participants would be allowa-
ble. He just believes no actual state will be of this sort. Similarly, he may
credit Nozick with having imagined another way in which a minimal state
“could” arise which violated no one’s rights, even though based on less than
unanimous agreement. But the likelihood of any actual state meeting these
moral conditions will be almost as low. The rejection of anarchism requires
the rejection of its moral premises.
on Py 6:
6. Nozick defends the procedure in a section entitled “Macro and
Micro.” He says:
[C]omplex wholes are not easily scanned; we cannot easily keep track of everything
that is relevant. The justice of awhole society may depend on its satisfying a number
of
distinct principles. These principles, though individually compelling (witness their
application to a wide range of particular microcases), may yield surprising results
when combined together. . . . [O]ne should not depend upon judgments about the
whole as providing the only or even the major body of data against which to check
one’s principles. One major path to changing one’s intuitive judgments about some
complex whole is through seeing the larger and often surprising implications of
principles solidly founded at the micro level.

Pp. 205-06. Obviously; but another way to change one’s intuitive judg-
ments about the scope or truth of principles at the micro level is by seeing
their larger implications. The fact that the rights of governments derive
from the rights of individuals does not imply that we can come to know the
rights of individuals without thinking about governments; just as the fact
that the properties of molecules derive from the properties of atoms does
not imply that we can come to know the properties of atoms without
investigating molecules. The logical and the epistemological connections
need not go in the same direction: even if political philosophy is logically
dependent on ethics, our knowledge of some aspects of ethics may derive
from an investigation of political philosophy.
7. The latter is the familiar proviso in Locke’s theory of property
acquisition, but according to Nozick it will not operate as a serious restric-
tion in a free market system. P. 182.
8. The example of entitlement that he offers (p. 206) as a decisive retort
to such skepticism—a natural right not to be deprived of one’s vital organs
for the benefit of others—is plausible partly because of the extreme
character of such an assault and partly because there is no possibility that
protection of this right will lead to the accumulation of vast hereditary
wealth or inequalities of social and political power.
 Libertarianism Without Foundations 205

9. Pp. 32-33 (emphasis in original; footnote omitted).

10. See pp. 49-50.
ive bee:
12. The fact that a right can be overridden to avoid sufficiently serious
consequences does not mean that its violation can be assigned a disvalue
comparable to the disvalue of those consequences. For that would give the
occurrence of such a violation greater weight in a calculation of outcomes
(e.g., when the question is what may be done to prevent such violations by
others) than it in fact has. Therefore, although rights may on occasion be
overridden, the violation of some people’s rights cannot automatically be
justified because it leads to a reduction in the more serious violation of other
people’s rights. This issue arises in connection with preventive detention,
wiretapping, and search and seizure, all of which might be useful in the
prevention of robbery, murder, assault, and rape.
13. It may be objected that taxation must be backed up by the threat of
stronger force and even violence. But this arises only if there is resistance at
the first level. If the original, nonviolent demand is legitimate, escalation
may occur if the subject resists it and uses stronger methods to resist each
succeeding method for enforcing the previous one.
14. Pp. 265-68.
15. Of course distribution is only one of the things covered in any
political theory, but we may leave others aside for the moment.
1Gy P53;
Tif; Loo.
18. P. 156. “To each according to his need” would be an example.
19. More specifically, his arguments against Rawls are seriously
weakened by a Procrustean attempt to portray Rawls’s principle of distribu-
tive justice as a nonhistorical end-result principle. Rawls does not maintain
that the justice of adistribution can be determined independently of how it
was produced. He believes that its justice depends on the justice of the
institutions, including legal institutions defining entitlement, which were
involved in its production. These are assessed only partly on the basis of
their tendency to promote a certain distributive end-state. Rawls, for
example, gives priority to the preservation of individual liberty, and while
he does not mean by this what Nozick means, it certainly restricts the
procedures by which a distribution can be justly arrived at. See generally J.
Rawls, A Theory of Justice (1971).
20. P. 161 (emphasis in original).
ONL. Ves, PPABy,
22. Pp. 297-333.
23. Pp. 252-62.
24. Pp. 35-41.
 1]
On the Randian Argument
ROBERT NOZICK

The title I have been given is: the moral foundations of capitalism.
Many supporters of capitalism, especially among the very young,
think that these foundations have already been provided; indeed
that we already possess in the writings of Ayn Rand a demonstra-
tion, a proof, a cogent argument, an establishment of a moral view
from which capitalism can easily be justified.
I have two reasons for wanting closely to examine the argument:
(1) Some persons are not devoting thought to fundamental issues
about morality, thinking that the essence of the job has already been
done.
(2) The argument itself is an attempt to provide a non-
utilitarian-non-social-contract-natural rights ethics.
Since I share the view that such a moral foundation is appropriate
and possible and that laissez-faire capitalism is morally justifiable on
such a basis, I wish to look closely at an actual attempt.
I would most like to set out the argument as a deductive argument
and then examine the premises. Unfortunately, it is not clear (to me)
exactly what the argument is.1 So we shall have to do some
speculating about how steps might be filled in, and look at these
ways. It may be, of course, that I have overlooked some other ways,
which would make the argument work. If so, I presume someone
206
 On the Randian Argument 207
else, who claims to possess and understand the demonstration, will
supply the missing material.
As I see it, there are four parts to the argument. (I use the roman
numerals ambiguously to refer both to stages of the argument, and
to the conclusions of the stages.)
I. to the conclusion that only living beings have values with a
point.
II. from I, to the conclusion that life itself is a value to a living
being which has it.
III. from II, to the conclusion that life, as a rational person, isa
value to the person whose life it is.
IV. from III, to some principle about interpersonal behavior and
rights and purposes.
I shall examine each of these in turn.

(1) Only a living being is capable of choosing among alternative

actions, or,
(2) Only for a living being could there be any point to choosing
among alternative actions, for
(3) Only a living being can be injured, damaged, have its welfare
diminished, etc., and
(4) Any rational preference pattern will be connected with the
things mentioned in (3), and since
(5) Values establish a (rational) preference ordering among alter-
native action,
it follows that
(6) Only a living being can have values, with some point to them.
Values have a purpose only for living things.

To make this point fully clear, try to imagine an immortal, indestructible

robot, an entity which moves and acts, but which cannot be affected by
anything, which cannot be changed in any respect, which cannot be dam-
aged, injured or destroyed. Such an entity would not be able to have any
values; it would have nothing to gain or lose; it could not regard anything
as for or against it, as serving or threatening its welfare, as fulfilling or
frustrating its interests. It could have no interests and no goals. (“The
Objectivist Ethics”, p. 16.)
208 Robert Nozick

I do not wish here to enter into complicated issues about what

particular cognitive and choice functions (if any) machines could be
capable of performing. Does coming to know that some situation has
been realized count as a machine’s being “changed”, in the requisite
respect? Note the assumption that each of our values concerns how
things affect us, a stronger assumption than that our values concern
how some affectable being or other is affected. (Some views would
claim to hold values unconnected with how any being is affected,
e.g., religious views which hold that it’s better if God is praised and
worshipped, not because God is affected by this, but because it’s a
fitting or suitable response to him. Or views which maintain that it’s
better that talented and able people be respected and admired,
even if they never learn of it and even if their knowing of it doesn't
(seriously) affect them; and better not because of the effects on us of
doing so, but because this is a fitting and appropriate response to
other people's achievements.) Presumably then I can’t value, or, in
a valuing fashion, act to achieve some state of affairs, in a far off
place, knowing that I shall never know whether my act has actually
succeeded or not, and knowing that in either case its consequences
will not affect me. If there were an island somewhere where we
couldn't go but could send things, and from which we would not be
affected in any way, then, it seems, we could not value their being
moral people and having just social arrangements, sending them
copies of some book whose reading we think is morally instructive.
Perhaps it is being claimed, not that each value in each instance
must be connected with oneself being affected, but that, for a being
to have values, some of them must be so connected. So that a being
which changed so that it became immortal, indestructible, unable to
be affected and so on could not evalute its alternatives and act so as
to make probable the realization of some value it had previously
held (e.g., the lessening of injustice in the world) knowing that
(because of isolation) it would not know whether its act had suc-
ceeded. Or perhaps the claim is that though a being can continue to
hold values in this state, if it was always in this state, and always had
been, and always realized that it was, it couldn’t hold values. It
couldn't read, for example, Atlas Shrugged and become convinced
by the arguments (we are not placing in doubt the machine’s having
certain cognitive capacity: so which part of the argument depends,
for its persuasiveness, upon the reader's being vulnerable and
 On the Randian Argument 209

having an affectable welfare?) and come to value and act so as to

realize a world in which the ideals of that book are instantiated. But
why couldn’t it? Must it say, “What's all that to me?”
Issues about the centeredness of one’s goals would repay further
study (I take up some connected issues with some of the examples in
section V), but I propose to proceed to see whether particular goals
and values can be gotten out of (6) and a self-centered view of goals.?

II

From the consideration of the preceding section a conclusion is to

be reached to the effect that the prolonging and maintaining of life is
itselfa value. Note, first, that it is unlikely that such a conclusion can
be reached without the introduction of significant additional mate-
rial. For suppose that death was a great value. (Suppose.) Only
living beings could achieve it, strive for it, choose to reach it,
striving for it would establish a preference ordering among alterna-
tive actions, and guide the choices of a living being. It seems that
death’s being a value is compatible with all that has been said in
section I, and hence that the considerations of section I do not rule
this out. One cannot reach the conclusion that life itself is a value
merely by conjoining together many sentences containing the word
“value” and “life” or “alive”, and hoping that by some process of
association and mixture, this new connection will arise. There may
be many other connections between life and value; the question is:
what is the argument for the particular connection embodied in
“Life itself is a value”?
Consider the following argument:
(1) Having values is itselfavalue
(2) A necessary condition for a value is a value
(3) Life is a necessary condition for having values.
Therefore,
(4) Life itself is a value.
But is (2) true? Are all necessary conditions for values, values
themselves? If getting cured of cancer is a value, is getting cancer
(which is a necessary condition for getting cured of it), or having
(say) a particular virus act on one, a value? And why is having values
itself a value? We might have the following argument for (1):
210 Robert Nozick

Achieving values is valuable. We are not pre-programmed to

achieve values. Therefore, if we are to achieve values more often
than accidentally, we will have to choose actions which will do so,
with doing so as their goal. But intentionally choosing actions to
achieve things is itself having and acting on the basis of values, and
therefore having and acting on the basis of values is itself valuable.
Note that this argument, in reaching its conclusion, itself seems to
use (2) as a principle of inference. The argument also begs the
question against death’s being a value, for if it is, we are preprog-
rammed to achieve it. Perhaps, instead of (2) we would have (2') If
something is a necessary condition for (achieving) all other values
(any other value) then it itself is a value.
Ignoring the issue of whether the question of death’s being a
value isn't thereby begged, is (2’) true? (And let us pass over, for
now, whether (2’) is meant to rule out there being a situation in
which dying is necessary in order to achieve one’s legitimate
values.) Because “not having yet achieved all values” is a necessary
condition for achieving any given value, for achieving each value, is
it itselfavalue? [We could also ask this question in extension, using
a list of all values, if we had such a list.] Similarly, if on Miss Rand’s
account being vulnerable, destructible, mortal, is a necessary condi-
tion for achieving (and having) values, does it follow that this
condition itselfis a value?
We do best, in view of these difficulties, to consider another line
of argument, which we might introduce by considering an objection
to our earlier contemplation of the possibility of death’s being a
value. “How can death be a value? Value is judged against a
standard of injury, harm, etc., and the greatest of these is death”.
(But on what basis, which we have been given, do we know that the
greatest harm isn’t the extension of life’s experiences?) In an essay
by someone who had been closely associated with Miss Rand, which
may shed light on her intentions, we find “ ‘should’ is a concept that
can have no intelligible meaning, if divorced from the concept and
value oflife”.? Here we have an example of some form of transcen-
dental argument, so beloved by philosophers and so tricky to handle
and get clear about. Before proceeding, we should have before us
another aid, John Hosper’s sympathetic presentation in his Intro-
duction to Philosophical Analysis (Second Edition)
 On the Randian Argument S|

Suppose someone said, “Prove to me that life is valuable.’ Rand would hold
that his request contains an inconsistency. It is, she holds, the existence and
nature oflife that sets the conditions for what is valuable: it is the distinctive
nature of life that gives rise to the need for values. In saying this, one is
saying much more than simply that man must be alive in order to pursue
values: one is saying that man must pursue values in order to remain
alive—and that this is the base of ethics and of all questions of moral value.
Just as (Rand would say) it is only the concept of life that gives rise to such
concepts and health and disease—just as it would be meaningless to talk of
health and disease except with reference to the standard and goal of life, and
talk about health is meaingful only in that context—so it is meaningless to
speak of values, of good and evil, except with reference to the needs of a
living organism. The concept of value, Rand maintains, is genetically and
epistemologically dependent on the concept of life, just as the concept of
health and disease are genetically and epistemologically dependent upon
the concept of life. Thus to say ‘Prove that it is morally obligatory to value
life’ is similar to saying “Prove that it is medically obligatory (that is,
necessary for health) to value life.’

Here we have the claim that apart from a background where life is
assumed as a value, no content can be given to “should’—
statements and if life is assumed as a value, then content is given to
“should”, namely (roughly) one should do those things which main-
tain and enhance one’s life. This account is of the form: you should
do an act if it leads to the greatest realization of X (where X is the
greatest value). Even if one were to accept this form of account,#
why must we substitute something about life for X? Cannot content
be given to should-statements, by substituting “death” for X, or “the
greatest happiness of the greatest number’, or any one of a vast
number of other dimensions of possible goals? And wouldn't such
content enable “should” statements to guide choices, apply only to
living beings etc.? Given this, it is puzzling why it is claimed that
only against a background in which life is (assumed to be) a value,
can “should”-statements be given a sense. It might of course, be
argued that only against this background can “should’-statements
be given their correct sense, but we have seen no argument for this
claim. Others who find a goal-directed type of account of “should”
illuminating, but who do not know what in particular to substitute
for X, may have “should” with the content “leads to the greatest
value”, leaving it as an open question what is the greatest value.
O12 Robert Nozick

These people, in asking “Should I do A?” would be asking “Does A

lead to the greatest value (whatever that is)?” For these people, and
those who substitute for X something other than life, an argument is
needed to demonstrate that life is the greatest (or a) value. I do not
see that such an argument has yet been offered.

Iil

Ignoring the difficulties discussed in Section II, suppose that we

have gotten, somehow, to the conclusion that for each individual, his
life and the prolongation of it is a value for him. How do we get from
here to:
III For each man, the preservation and prolongation of his life,
qua man, as a person, is a value for him.

Possibilities:
(1) What man’ is, qua man, is completely determined by what's
special to man (which is rationality). And what's special to beings
should be preserved and should flourish. Why? Ignoring the dodo
argument (viz., it’s a tragedy if any bit of diversity becomes extinct),
we focus on the idea that what is special to a thing marks its function
and from this we can get its peculiarly appropriate form of behavior.
But no conclusion could depend, in this way, on some property P
being special to man. For suppose it turned out that dolphins or
some other being somewhere also had property P; would this stop
the conclusion? It might be said that here we would discover a new
kind of thing (viz. man or dolphin) and P would be special to it. But
there might be nothing special to man, for all his nice properties
might be had by other things which have further super-duper
properties Q also, which man doesn’t have. (Call these things,
merely for a label, angels.) It might now be said that man (or
dolphin) does have something special, namely P and not-Q. But how
could one think that anything depended on its being special; that
something of moral interest did not follow from

Man has P

in conjunction with other premises, but did follow from these

premisses conjoined with
 On the Randian Argument 213

Man has P, and nothing else has P?

Could discoveries on other planets show us that our fundamental

moral conclusions don’t follow? Surely, it’s in virtue of man’s having
P that conclusions follow, and not because other beings don’t have
P. All this is not to deny that it is heuristically useful to focus on
what is special to man. For if we don’t apply moral standards and
principles to the other beings we actually have encountered, and we
do to man, then we can ask what properties man has in virtue of
which he is moral agent and a subject of moral judgments. Something
fundamental to ethics does lie in those properties which do distin-
guish man from other things we have actually encountered, but
nothing morally fundamental depends on the fact that these proper-
ties are distinguishing ones.
(2) What man is, qua man, is determined by his essence, which is
rationality. He should act so as to continue his essence.
(a) If the essence mentioned in this argument is real essence, it’s
a dubious theory. Also, it (e.g., a man) would no longer exist if its
essence changed (but another thing would), so that if its continued
existence is a value, so is the continued existence of its essence.
However, in the case of this argument, its essence doesn’t change,
it's just not exercised. The being continues to exist. And the
conclusion of II is too frail a reed to hang such an argument on. For
if it stops existing, another kind of value pursuer, intimately con-
nected with it, will exist. And why should it care which does?
(b) If, in this argument, essence is: those properties, relative to
current knowledge, which underlie, systematize, account for etc.
the rest of our general knowledge of such entities (those properties
mentioned in our most fundamental, at the time, reducing unre-
duced theory about the entities, or if two different such theories,
both sets of properties) then the injunction to preserve and exercise
such properties requires reasons. Why shouldn't one change?
Surely, it is not merely a conservative injunction. Two answers
suggest themselves. One, it’s change to a lower form of life, and so
shouldn’t be done. But one needs a theory to back this up, which
isn’t given, and secondly it implies that it would be all right, if
possible, to change to a higher essence. Even though if one could
and did change to an angel-like being, one wouldn't be preserving
one’s life qua man. The second, and more interesting answer, is that
214 Robert Nozick

one won't survive (period), as alive, if one doesn't live as a person.

This deserves consideration as a separate argument for III.
(3) Ifa person does not prolong his life as a rational being, he will
not survive for long afterwards as any kind of living being. For, the
argument runs, we have no automatic way of knowing what will
prolong our lives. We have to figure this out, using our conceptual
apparatus. And then we have to devise ways to do this. Otherwise,
we will not survive at all, unless by accident, or unless some other
rational being cares for us. And, if the goal is to prolong one’s life (in
years) and minimize, constantly, the probability of its ending, then,
much knowledge, invention, etc. will be useful.
There remains the possibility of being a “ward”, making one
decision to let another care for you, give you orders, etc. To this it
might be objected that this might not work out (and you wouldn't
know when to desert the ward-ship before it begins to sink), and
that such a life is parasitic and depends upon another person's not
living as you do.
There are two forms to the parasite argument, a consequential
one and a formal one. The consequential argument is that being a
parasite won't work in the long run. Parasites will eventually run out
of hosts, out of those to live off, imitate, steal from. (The novel Atlas
Shrugged, argues this view.) But in the short run, one can be a
parasite and survive; even over a whole lifetime and many genera-
tions. And new hosts come along. So, if one is in a position to
survive as a parasite, what reasons have been offered against it?
The formal argument is difficult to make precise, and difficult to
fit into a Randian view. It holds that moral rules are applicable to
everyone, so that if following certain rules and values can work only
if others follow different ones, and can work only because others
follow different ones, then the rules and values in question cannot
be the correct ones. [“What if everybody did that?”]
But it is difficult to find the appropriate level to speak at, using
this argument. My being a teacher succeeds only because other
people do other things, e.g., grow food, make clothing. Similarly for
the activities of each of these others. The question “What if
everyone did that?” shouldn't apply here, but how is the case to be
marked off?
Intuitively, there is some description of what I’m doing [accepting
 On the Randian Argument 215

a job to fill need, etc.] which is o0.k.; that is, it is permissible for
everyone to do that.
So, there being some description D1 which fits what I do, where
it would be disastrous if everyone did D1, doesn’t show I shouldn't
do it. For there may be another description De which also fits what I
do, and it would be all right if everyone did things of sort De.
But we should not convert this fact into a condition which says
that if there is some description which fits my activity which is such
that it’s o.k. if everyone instantiates that description, then my
activity is permissible. For this is too weak a sufficient condition for
moral permissibility. Presumably some general descriptions of this
sort would also fit the thief's activity, e.g., “doing things which lead
to their being fed”.
Here it might be suggested that the appropriate description is
that which presents the essence of an action. But even if such a
notion could be clarified, as in (2b) above, its application here would
beg the question, for especially important among the relevant
general facts to be accounted for by the essence will be moral facts.
Hence we must first agree about these moral facts before agreeing
about the essence of an act, so that considerations about the essence
of acts cannot come first as a basis on which to ground their moral
nature.
My purpose here is not to pursue the details of how such
generalization arguments might best be stated (there is a growing
literature on the subject), but to note that if such arguments can be
made to work, they will involve extra principles (and not merely the
claim that a particular case of parasitism is doomed to fail). Some
philosophers view providing the foundation for such principles, and
stating them precisely, as one of the central tasks of moral
philosophy, but it is one that has not even been attempted by Miss
Rand.
One final way to get a formal principle should be mentioned. It
might be said that a rational person follows principles, general
policies, and so we must consider those principles of action which
make man’s survival possible. But it has not been shown why each
person must follow the same principles, and why I may not, as a
rational being, have a clause in mine which recommends parasitism
under certain conditions.
216 Robert Nozick

IV

Supposing that it is granted that living as a rational being is, for each
person, a value, how do we get to some social conclusion about
people's rights?

The basic social principle of the objectivist ethics is that just as life is an end
in itself, so each living human being is an end in himself, not the means to
the ends or the welfare of others—and, therefore, that man must live for his
own sake, neither sacrificing himself to others nor sacrificing others to
himself. To live for his own sake means that the achievement of his own
happiness is man’s highest moral purpose.®

Starting with:
(1) For each person, the living and prolongation of his own life is
a value for him
to get to
(2) No person should sacrifice his life for another
we would need, in addition to (1)
(1') For each person, the living and prolongation of his own life
(as a rational being) is the greatest value for him.
[and: Each person ought to pursue his greatest value. ]No argument
has been offered, yet, for (1). Miss Rand has some things to say
about life being an ultimate value, which might lead her to accept
(1’), but these things, in view of our earlier discussion in Section II
of life as a value, aren’t sufficiently clear and forceful to establish
(Lens
But suppose we have (1’), and have gotten to (2). How do we go
on to argue for the important social conclusion:
(3) No person should sacrifice another person (’s life) to himself
(his own.)
Why shouldn't he? There is the parasite argument already con-
sidered: sacrificing another's life to your own is not in your own long
run interests. But this is no argument to convince (or apply to)
someone living at a time before the victims have run out, e.g., the
present. A more promising approach focuses on the notion of rights.
Consider the following argument.
(4) Each person has a right to his own life, i.e., to be free to take
the actions required by the nature of a rational being for the
 On the Randian Argument 217

support, the furtherance, the fulfillment, and the enjoyment

of his life.
(5) Since each person has this right, to force a person to sacrifice
his interests to your own violates this right.
(6) One should not violate another’s rights.
Therefore
(7) One should not force another to sacrifice his interests for your
own or that of yet another person.
But why does each person have a right to his life, to be free to
perform those actions? If we grant, for the purposes ofthis argu-
ment, (1’) above, and we add
(8) Each person has a right to be free to pursue his greatest and
highest value,
then we get (4) above. But (8) is surely too strong: did Hitler have a
right to pursue his highest value? (But, it will be said, his highest
value wasn’t a rational value. So let us focus instead on “should”.)
Perhaps instead the argument for (4) is as follows:
(9) a) For each person, he should pursue the maintenance of
his life as a rational being.
b) The “should” in a) is and should be of more weight, for
him, than all other “shoulds”.
(10) a) Each person has a right to do what he should do.
b) Each person has a right, which is non-overrideable, to do
what he should (where this “should” has the most weight)
do.
(9) is meant to follow from other considerations which we have
examined in section I, and found inconclusive. What of(10)?
The force of “right” here is that others shouldn't intervene, using
force, to prevent one from exercising it. The question is: why, given
that you should do something, shouldn't I intervene to stop you?”
Perhaps one has a vision of a morally harmonious universe in
which there are no irreconcileable conflicts of duty, of shoulds, and
in which if you should do something, I shouldn’t forcibly prevent
you from doing it. But no conclusive arguments have been offered
for such a vision. If one believes that ethics involves (something like)
one dimension or weighted set of dimensions which is to be used to
judge us and the world, so that all of our moral activity (the moral
activity of each of us) is directed towards improving (maximizing) the
218 Robert Nozick

world’s score on this dimension, then it will be natural to fall into

such a vision. But if we legitimately have separate goals, and there
are independent sources of moral commitment, then there is the
possibility of an objective conflict of shoulds. So that perhaps, with
some slight modification of Sophocles’ characters, Antigone should
bury her brother and Creon should forbid and prevent this burial.*
But Miss Rand needs something even stronger, for her argument,
than objective harmony of “shoulds”; she needs an objective har-
mony of interests.?
What I shall call the optimistic tradition holds that there are no
objective conflicts of interest among persons. Plato, in the Republic,
being the most notable early exponent of this view, we might
appropriately call it the Platonic tradition in ethics.1° Miss Rand
falls in this optimistic or Platonic tradition in ethics, believing that
there are no objective conflicts of interest among persons, and that
there is no situation in which it is in one person’s interests to force
another to do something which it is not in his interests to do; no
situation where one person’s forcibly stopping another from ad-
vancing his interests, advances his own interests. No knock-down
argument has been offered for this thesis,1! and Miss Rand has not
produced (or tried to produce) any reason why, if such conflicts are
possible, I, in following my interests, should limit myself so as not to
interfere forcibly in your pursuit of your life as a rational being.
Miss Rand's position is a constrained egoism; egoism subject to
the constraint of not violating certain conditions (which are other
persons’ rights).!2 One way to argue for constrained egoism is to
argue that it is identical with egoism to argue that the constraints
add nothing because they are in your rightly understood interests
(the optimistic tradition). If such arguments equating the egoism
and the moral constraints fail, then a holder of this position will have
to choose.

We have until now considered only one part of the social non-
sacrifice principle (don't sacrifice another to yourself), and found the
arguments for it inconclusive. We turn now to the other part of the
principle: don't sacrifice yourself to another: live for your own sake.
 On the Randian Argument 219

“To live for his own sake means that the achievement of his own
happiness is man’s highest moral purpose”.
Is it? We have action, endorsed by Miss Rand, in the novel Atlas
Shrugged, which appears incompatible with this. In the novel, John
Galt risks his life to save that of Dagny Taggart, whom he loves, and
he says that he will kill himself if she is tortured to make him talk.
How can he do this? He says to Dagny Taggart:

“It won't be an act of self-sacrifice. I do not care to live on their terms. I do

not care to obey them and I do not care to see you enduring a drawn-out
murder. There will be no values for me to seek after that—and I do not care
to exist without values.”

But this is quite incredible. For it seems from this that, were Dagny
Taggart tragically to suffer and die of some disease, then Galt would
commit suicide. It would be a terrible loss, but does Galt, “the
perfect man’, have so little moral fiber and resources, that life
would be intolerable for him ever afterwards [and would the agony
of the time immediately after her death outweigh the life which
could be led after time has done its work]? Would he come from
Galt’s Gulch to attempt to save her life, if he had the option of
staying there, obeying no one else, and not being present to observe
her murder?
Would Galt save Dagny Taggart’s life, knowing that this will be at
the cost of his own? Is this inconsistent with his principles? Would
doing this mean he wasn't an egoist? A fruitless path to follow? is to
say that it is the short period of happiness while he is sacrificing his
life that justifies the whole thing. So that, in thinking about a
situation in which both are unconscious and only one can be saved
by a third party, he would prefer it were himself because in that
situation he wouldn't get to feel the happiness of saving her life? Or
are we to imagine that in this thinking about this situation he does
prefer that her life be saved, and this because of the happiness he
now gets in thinking about it?

But (a) we can imagine he’s answering, quickly, a large number of

questions on some psychological test, so that there's no time to
pause for a glow of happiness. Is it now for the happiness he knows
220 Robert Nozick

he'll feel after the test is over (mightn’t he forget and not recall the
question?) that he answers as he does?
(b) Why doesn’t he feel sad in contemplating his dying and her
surviving? To be sure, he may think it’s better than their both
dying, or her dying and his surviving, but why should the existence
of some worse alternatives than alternative A make one happy in
contemplating A? In fact, it doesn’t, and we don't have this easy
path to happiness.
(c) Most importantly, this answer gets things backwards. For one
would be made happy by placing one’s family in the only places in
the raft!4 only because one values their survival above one’s own,
their happiness above one’s own.

It is not that one (chooses to) have these values, because knowing
one has fulfilled them will make one happy. And it is not that I do it
because not doing it will make me afterwards feel guilty. For this
would be so only if there were some other moral reason to save their
lives over one’s own; it cannot be that the primary reason is to avoid
later guilt. And also, we could imagine cases where the knowledge is
expunged via a chemical producing selective forgetting.
Such science fiction possibilities cause difficulties. If one were
concerned only with one’s child’s happiness, and one had the
capability, one would implant a device to get the child to act on
principles P (the correct moral principles) except in situations S
(where he knows that deviating from them will be in his interests,
e.g., by murdering someone and taking his fortune) where he will
deviate from them, afterwards forgetting that he’s done so. Such a
person would be happier than one only following principles P, and
his life will be identical to one with only principles P, except at a few
selected times. Furthermore, he will think he always, with great
integrity follows principles P, and he will have great self-esteem.
And if someone were concerned only with his own happiness, he
would wish that he himself had been so preprogrammed. If one
doesn't wish this for oneself, then one isn’t concerned only with
one’s own happiness. Saying that one is so solely concerned, but
such a preprogrammed person can’t be happy because he fakes
reality begs the question (ignoring the fact that he doesn’t fake it: it’s
faked for him); it seems as though one can describe a case where
“rationality” (and rational self awareness) and happiness diverge;
 On the Randian Argument 221

where someone less rationally self aware will be more happy. If in

contemplating this case you would choose rational self awareness
and moral rectitude, rather than happiness, then the former have
independent value, and are not justified in your eyes only because
they lead to happiness.
Driving the point further, suppose we read the biography of a
man who felt happy, took pride in his work, family life, etc. But we
also read that his children, secretly, despised him; his wife, secretly,
scorned him having innumerable affairs; his work was a subject of
ridicule among all others, who kept their opinion from him; every
source of satisfaction in this man’s life was built upon a falsehood, a
deception. Do you, in reading about this man’s life think: “what a
wonderful life. I wish I, or my children, could lead it”? And don't
say that you wouldn't want to lead the life because all the deceptions
and falsehoods might come out making the man unhappy. They
didn't. Of course, it is difficult to imagine the others behaving
appropriately, and the person himself not being nagged by doubts.
But is this the ground of one’s reaction? Was it a good life? Does it
lack happiness?
This man lived a lie, though not one that he told. We can imagine
other cases. You have what you believe is a private relationship with
someone. However, unbeknowst to you, I am filming it with my
super-duper camera and sound equipment, and distributing the film
to people whom you will never encounter. Nothing in your life is
changed by the fact that people are packing the pornographic movie
theaters in Outer Mongolia to keep up with the latest serial install-
ment in your life. So, should anyone care? And is the only ground on
which my action can be criticized, the nature of all viewers’ experi-
ence?!5
I have listed all these examples, not only to bring the reader to
feel the inadequacy of Miss Rand’s view,‘® but for another purpose
as well. For the examples we have discussed count against a more
general view as well, which we might call experiential ethics. Put
briefly, experiential ethical theories hold that the only facts relevant
to moral assessments of actions are how these actions do, or are
intended to affect the experiences of various persons. The only
morally relevant information (though other information may be
relevant via being evidence for this kind) is that about the distribu-
tion of experiences in society. Theories will differ about which
PPP) Robert Nozick

experiences they pick out, or about the criterion of optimal distribu-

tion of experiences, but they will agree that all of the considerations
have to do with such experiences, and how they feel from the inside.
Indeed, it may seem, how could anything else matter other than the
experience people have, how things feel from the inside. What else
could there be that’s of any importance? It is a task of some interest
to explain what else does matter, and why, and to account for the
pull of the experiential picture on us.+7
Let me, in closing, reiterate that my purpose has been to examine
Miss Rand's arguments for her conclusions. It has not been to argue
that death is a value, or that we should sacrifice others to ourselves,
or that people don’t have rights to our non-interference in their
lives, or to demean the virtues of rationality, honesty, integrity,
productiveness, pride independence, justice. It has been to see
whether, in her published work, Miss Rand indeed objectively
establishes her conclusions. She doesn't.

NOTES

1. The main sources are Atlas Shrugged, especially Galt’s long speech,
and the essay “The Objectivist Ethics” in her book of essays The Virtue of
Selfishness. The other essays in this book are helpful also, (later references
to some of her other essays will be to essays in this book) as are her other
books, including Introduction to Objectivist Epistemology, in understand-
ing her views.
Since I shall be quite critical of Miss Rand’s argument in the remainder of
this essay, I should here note (especially since she has been given a largely
vituperative and abusive hearing in print) that I have found her two major
novels exciting, powerful, illuminating, and thought-provoking. These vir-
tues, even combined with a “sense of life” that is worthy of man do not, of
course, guarantee that her conclusions are true, and even if we suppose
they are true, all this does not, of course, guarantee that the actual
arguments offered will be cogent, that they will prove their conclusions.
Nothing I say in this essay is meant to deny that Miss Rand is an interesting
thinker, worthy of attention.
2. Note in passing that persons may have innate preferences uncon-
nected with their own welfare, and survival. There is a natural selection
argument that, in the evolutionary process, preferences that tend to help
keep you alive to the reproductive age will be selected for, but also there
are natural selection arguments that innate preferences for behavior which
 On the Randian Argument 223

is reproductive behavior, and for behavior after reproduction which en-

hances the chances of the progeny’s survival, even to destroying oneself to
guard one’s young, will be selected for in the evolutionary process. So beings
may be preprogrammed or predisposed to do things which lessen their
individual chances of survival.
3. Nathaniel Branden, “Rational Egoism: a Reply to Professor Emmons’,
The Personalist, Spring 1970, p. 201.
4. It may seem a necessary truth that “right”, “ought”, “should”, etc. are
to be explained in terms of what is (intended to be) productive of the
greatest good. So, it is often thought, that what is wrong with utilitarianism
(which is of this form) is that its conception of good is too narrow. It doesn’t,
for example, take rights and their non-violation into account in the proper
way, but leaves them a derivative status. (Hence many of the counterexam-
ple cases to utilitarianism; punishing an innocent man, to save a neighbor-
hood from a vengeful rampage, etc.) But even if one includes the non-
violation of rights into a theory in a primary way (and those of us for whom
there is some desirable society in which we would choose to live, even
though in it some ofour rights are sometimes violated, rather than move to
an island on which we could survive, alone, do not think that it is the sole
greatest good), one may include it in the wrong place, in the wrong manner.
For suppose we build into the desirable endstate to be achieved, some
condition about minimizing the amount and significance of the violation of
rights which takes place in the society. We then have something like a
utilitarianism of rights, which could still require us to violate someone's
rights if this act of violation leads to a minimizing of the amount of violation
of rights in society. It might do this perhaps by deflecting others from their
intended action of gravely violating people’s rights, removing their motive
for doing it, diverting their attention, and so on. (A mob rampaging through
a part of town, killing and burning, will be violating the rights of those living
there, and so one might try to justify one’s act of punishing someone you
know to be innocent of the crime which enraged the mob, on the grounds
that this action will lead to a minimizing of the weight of the violations of
rights in the society.)
As against this conception, which builds rights into the end-state to be
achieved, one might say that their proper place is as a constraint upon the
actions to be done. So the structure of the view is: Among those acts
available to you which don’t violate constraints C, act so as to maximize goal
G. This differs from one which says: act so as to maximize G, and which tries
to build the side constraints C into the goal G. For the view with the side
constraints forbids your violating having the minimizing of the violation of
these constraints in the goal, allows you to violate the constraints in order to
lessen the total amount of violation of such constraints in the society. I
224 Robert Nozick

ignore questions about finding a way of putting the side-constraint view in

the form of the goal—without—side-constraints view, e.g., perhaps by
having a distinction made, in the goal, between your violating the con-
straint, and someone else’s doing it, where the former is given infinite
weight in the goal, so that no amount of stopping someone else’s doing it
can outweigh your violating the constraints. Even with this possibility,
notice that indexical expressions (“my doing something’) appear in the
goal. I also ignore questions about whether sometimes you are allowed to
violate rights of innocents in order to prevent monstrous deeds by others.
For example, suppose that the only way to have stopped the Nazis from
conquering all, was to have used some weapon on them which would have
also killed innocent people in the city attacked who couldn't leave and
werent choosing to stay there. Perhaps avoiding great moral horror swamps
people’s rights, so that one would be justified in doing something one knew
would kill innocent people, in order to stop the horror. (See Michael
Walzer’s paper “World War II: Why Was This War Different?”, delivered
before the American Political Science Association, Sept. 1970 meeting.)
The possibility of “swamping” of rights may lead one to think that it is the
goal structure, with a-»weighting of constraints inside the goals, which is
appropriate, rather than the structure with absolute side constraints. But
other structures are possible, e.g., a side-constraint structure with princi-
ples governing the setting aside of the whole structure. (But since many of
the set aside structure's features will have to reappear again in its replace-
ment, is this the proper way to view things?) These issues, connecting with
some discussed in my “Moral Complications and Moral Structures”,
Natural Law Forum, Vol. 13, 1968, pp. 1-50, are very complex, and I hope
to discuss them in detail on another occasion.
These considerations enable us to clarify a difficulty with the classical
conception of the night-watchman state, which limits the legitimate func-
tions of the state to the protection of its citizens against violence, theft,
fraud, and to the enforcement of contracts, ete., but specifically excludes
redistributive tasks. The difficulty (which I first presented in a paper, “A
Framework For Utopia,” delivered at the APA Eastern Division Meeting in
1969, with Sidney Hook as commentator) is that the protection of people in
this way costs money (for detectives, police to bring criminals into custody,
courts, prisons). How is to be paid for, and, in particular, is there some
non-redistributive way offinancing it? We might imagine a system in which
the state provides these protective services only to those who specifically
pay for them, with perhaps different packages of protective services offered
for different fees. People who don’t buy a protection contract don’t get
protected. Now consider this system of an individual's purchasing protection
from the state, combined with a voucher system (cf. Milton Friedman’s
 On the Randian Argument 225

school vouchers, which, of course, allow the people a choice of schools)

funded by taxation, under which all people, or those in need etc., are given
vouchers, backed by the funds collected by taxation, which can be used only
for the purchase of a protection policy from the state. It is clear that the
latter system is redistributive (compare it to the former), yet this latter
system is equivalent to the system of the night-watchman state. So, unless
some other nonredistributive method of financing protection for everyone
can be found (and, on the face of it, it is difficult to see how it can be), the
night-watchman state itself performs some redistributive functions (viz.,
having some people pay for the protection of others). And, it might well be
asked, if this redistributive function of the state (redistribution in order to
protect everyone) is legitimate, why is redistribution for other attractive
and desireable purposes not legitimate as well? On the other hand, if the
non-redistributionist accepts something like the first system described,
then his position will seem to many to be inconsistent, as well as monstrous.
I wish here to focus only on the first of these charges. The proponent of
the night-watchman state is greatly concerned with the violation of rights,
and with protecting these rights; so that he makes the protection the only
legitimate function of the state, protesting that other functions are illegiti-
mate because their performance will itself involve the violation of rights.
How can he then, given his paramount placing of the protection and
non-violation of rights, move to the first position which would, it seems,
leave some persons’ rights unprotected or ill-protected; how can he do this
in the name of the non-violation of rights— This objection assumes that the
proponent of the first system is a “utilitarian of rights,” whose goal is, say,
the minimization of the amount and significance of the violation of rights in
the society, and who will pursue this goal, even through using means which
themselves violate people’s rights. But if he places the non-violation of
rights as a constraint upon action, rather than building it into the endstate to
be realized (it may, let me hasten to add, occur in both places), and his
conception of rights is such that your being forced to contribute to another's
welfare violates your rights, and is such that someone else’s not providing
you with something you need greatly, essential to the protection of your
rights, does not itself violate your rights (though it may facilitate, or avoid
making more difficult, someone, else’s doing so) then this theorist will be
led to something like the first system of individuals purchasing protection
for themselves (and those others they voluntarily choose to do so for). And
this will be a consistent position. (That it is a consistent position does not, of
course, show that it is an acceptable position.)
Given the subject of this essay, we do well here to mention Ayn Rand’s
proposal, in her essay entitled “Government Financing in a Free Society.”
Briefly put, her suggestion is that the government enforce the contracts of
226 Robert Nozick

only those who have paid a special fee (others may sign contracts backed by
trust, desire to continue their good reputation etc., but not by govern-
ment’s powers of force) and that the government's other legitimate func-
tions be financed out of the funds so collected from those who voluntarily
choose to make use of (and pay for) the government's contract enforcing
powers. But it is difficult to see why she finds this a legitimate solution to
the problem. For the government has a legal monopoly on the use of force,
and the proposal involves the government charging extra to enforce con-
tracts, in order to cover the costs ofits other protective functions. (Compare
what it would charge for enforcing contracts if it did not take,on these other
protective functions.) Why is this not illegitimate forcible redistribution,
and isn’t talk of people voluntarily dealing with the legal monopoly (and
voluntarily paying the higher fees to cover the legal monopoly’s other
functions?) too quick. If in the United States today, the post office, with a
legal monopoly on the right to carry the mails, charged one dollar per letter,
in order to cover the costs of other redistributive activities, (and thus
charged more than even a monopolistic post office, covering its costs, which
was singlemindedly devoted to delivering the mail) would Miss Rand think
that a legitimate means had been found to, finance redistributive activities,
say the educational costs of college students. Could anyone who otherwise
thought that there was a problem with governmental redistributive ac-
tivities, feel that problem is avoided by tying redistributive activities to the
provision of some service protected by a legal monopoly? To be sure, there
are powerful arguments for the state’s having a legal monopoly on the use of
violence (rather than allowing e.g. private contract-enforcing firms to use
force to enforce contracts they have been hired to enforce), as there are not
for the post office’s having a legal monopoly to deliver the mails. But since
the arguments for the state’s having a monopoly on the use of violence do
not involve redistributive considerations, it is difficult to see why it is
legitimate to allow the monopoly thus established to exploit its monopolistic
position in order to pursue redistributive aims. These considerations are not
addressed to and will have, perhaps, little interest for those who find no
problem with the state’s forcible redistributive activities; they are meant
merely to point out that those who do find a problem cannot avoid it by
slipping the state's redistributive activities in through the (only) back door,
in the fashion of Miss Rand.
5. Rand, “The Objectivist Ethics”, p. 27
6. See also our discussion in section V, which considers a case where
John Galt seems not to act on 1.
7. If rights are explained as “conditions of existence required by man’s
nature for his proper survival”, then we can reconstruct the Randian
argument for a right to life, but this argument will not answer the question:
 On the Randian Argument 227

why should I not violate another’s right to life; why should I not intervene
by force to eliminate one of the conditions of another man’s existence
required by his nature for his proper survival? That is, with this explanation
of rights, argument is needed for (6) above. If we assume that rights are not
to be violated, and others should not forcibly intervene in the exercise of
someone's rights, then argument is needed to the conclusion that a person
does have a right to his own life, that is, that others shouldn’t forcibly
intervene in it, even granting that its maintenance is his highest value.
Taking either approach, we face the question of why one person shouldn't
intervene by force to thwart the conditions of another man’s existence.
A similar question is raised by Mortimer Adler's argument in his The
Time of Our Lives: The Ethics of Common Sense (1970) where there is a
transition (Chapter 14, Section 2) from an individual's moral obligation to
pursue his own real good to an individual’s moral right (entailing obligations
of others not to interfere) to pursue his own real good. He attempts to
bridge this gap (Chapter 16) by grounding our moral obligation not to
violate the rights of others on our obligation to make a good life for
ourselves. He argues as follows:
1) Each of us needs civil peace as a means of making a good life for
ourselves. Every universal act which injures other men or the community is
a breach of civil peace. “Hence when I act unjustly toward others or act in
any way that is contrary to the good of the community, I am injuring myself.
It may not appear to be so in the short run; I may gain apparent goods by my
injustice toward others or by criminal activities that injure the community
itself. But in the long run, I may have gained these apparent goods only at
the loss of a real good that I need—the civil peace of the community in
which I live. It is only in the short run that injustice can appear to be
expedient. In the long run, which is the omen of my whole life, the just
tends to be the expedient”. (Page 173, my italics)
Note that my individual conduct breaches the peace; it does not destroy
it. And it is an undestroyed peace (rather than an unbreached one) that it is
said I need.
2) Under ideal conditions “when an individual seeks only those things which
are really good for him, he does not infringe on or interfere with the pusuit
of happiness on the part of others through their seeking the same real goods
for themselves” (Page 174). For Adler says, consider that one man’s pursuit
of real self-improvement (which is the greatest good) cannot interfere with
another’s similar pursuit. But if there are scarce (in the economist’s sense,
which does not entail that the situation is non-ideal, in Adler's sense (cf.
Page 177-178) resources of self-improvement (books, musicial instruments)
then such interference is possible; similarly, with teachers, with the added
possibility of forcing one of the few able people in a subject (who would
228 Robert Nozick

rather be improving himself) to instruct you. It is even easier to see how this
might happen with the goods other than self-improvement that Adler
discusses.
3) Under non-ideal conditions, men may not obtain the privileges which
enable them to lead good lives by force or fraud. (footnote 3, Pages
309-310). No reason connecting this constraint to an individual’s obligation
to pursue a good life for himself, is offered by Adler. (Nor is any other
reason offered.) It would be interesting to hear what reasons he would
produce, in view of his statement (pages 171-172):
If it could be shown, as I think it cannot be shown, that the individual’s obligations to
others and to the community are independent of his obligation to make a good life for
himself, then the discharge of such obligations would impose a burden on him that
might interfere with or even frustrate the pursuit of his own happiness. I would like
to add, in passing, that those who regard their duties to others or to the community as
independent of their obligation to make a good life for themselves are either
sentimentalists or thoughtless do-gooders!!
8. Or, to go to a case which is merely meant to caution one about how
one formulates the harmony condition, each of two boxers, who have
promised to win the fight, and who each have contracted with different
outside parties to win, should win, and furthermore, it is permissible for
one to thwart the others’s attempt to do what he should do.
9. I say “stronger”, realizing that she would deny that it is a stronger
claim, because she thinks that there is no additional step needed to get from
something being in one’s interests to its being something one should and
ought to do.
10. We should notice here a slide many writers in and readers of this
tradition make, from X is a (morally) better act than Y, to one is (morally)
better doing X than Y, to one is better off doing X than Y. An argument that
right conduct and self-interest don’t diverge requires independent ways of
identifying each, plus an argument that these two independently identified
things always go together; at any rate, so it must go if the ordinary view that
self-interest and rightness are distinct notions is to be adhered to. Alterna-
tively, it might be argued that they don’t diverge because there are not two
independent notions, but rather one of them is primary and the other one
can be explained in a coherent fashion only in terms of the first (in such a
manner as to yield their non-divergence). So, one might explain “morally
right” or “morally ought” in terms of “self-interest,” or against such a
background (Rand). Or, one might attempt to argue that the notion of
self-interest itself cannot be coherently explained except in terms of morally
right (with a particular form, and perhaps content) in such a fashion that
they don’t diverge. This latter possibility is an exciting and tantalizing one,
which should be pursued.
 On the Randian Argument 229

Il. See her essay, “The Conflicts of Men’s Interests”, in The Virtue of
Selfishness.
12. It is sometimes assumed that the unconstrained (constrained) egoist
can give no weight to the interests of others except as they are reflected in
his own interests (and the constraints). But a theoretically possible egoist
position could require: Maximize self-interest, (subject to constraints) and
among those actions which tie so far as (constrained) self-interest is con-
cerned, pick that one which best satisfies condition C (where C may be any
condition, e.g., the utilitarian one.). Here we would have a lexicographic
ordering, with the (constrained) egoistic score occupying the first place in
the vector, so that nothing can override considerations of self-interest, and
other things are given some weight later on in the vector. I use “constraint”
here as it is used in linear programming, to mean bound on another
maximizing or goal-directed process.
13. And not a path that Miss Rand would follow, I think. (See “The
Objectivist Ethics”, pp. 28-30). But one cannot merely say, to use Miss
Rand’s terminology, that happiness is the purpose of ethics, but not the
standard. For the problem here is that it is known that the action will not
achieve this purposes when it is guided by Galt’s valuing Dagny Taggart’s
life above his own. And in the science fiction case below, the purpose of
happiness is achieved by his not following the standard, without his making
whims or irrational desires his guide to conduct.
What the egoist condition on desires is which allows non-self-centered
valuing of another's life above one’s own or, the valuing of the triumph of
justice and right above one’s own life, but rules out other, altruistic, desires
as incompatible with rational egoism, is a puzzle.
I discuss the position in the text because when the components of Miss
Rand's view, assumed to be indissolubly intertwined, part company, it is
one path that can be taken to reconstruct the position (and has been taken in
conversation with me by persons who term themselves followers of Miss
Rand, for whose intellectual meanderings she is, of course, not responsi-
ble). As I have said, it is not a path that the preponderant thrust of her view
(the components of her view which, I assume, have major weight for her)
would lead her to take, I think. But in that case, it is unclear what
alternative teleological argument she would offer for a life lived in accord-
ance with the virtues of rationality, honesty, integrity, pride, productive-
ness, justice and independence.
14. If “Happy” is what it is. More likely terribly sad, hoping that they'll
reach safety.
15. Privacy invasions raise an interesting problem for libertarian theory,
for it seems that we can imagine such invasions without any of the particular
sorts of physical invasions which libertarians tend to focus upon. For
230 Robert Nozick

example, suppose that there is one telepath who picks up emanations from
you which can’t be screened in. He knows exactly what you are doing, and
thinking, at any given moment. Perhaps he broadcasts you on a particular
T.V. wavelength. Anyone can tune in, any time, to see what you are doing
and thinking at that moment. On Saturday night, after you are asleep, there
are the highlights of your week. No thought or action of yours is private, yet
the telepath has not invaded you or your property in any way (as libertarians
speak of invasion). Are there any grounds to legally forbid the telepath from
so operating, which the libertarian can formulate? Would we wish to forbid
it in this case?
It might be said that such cases are impossible, and it is not a difficulty
with a view that it does not handle some impossible case as we would wish.
But it is, I think, an objection to a view if it does not handle this kind of case
correctly (as it would be to show that a consequence of a moral view was that if
there were anyone who could travel faster than the speed of light, then it
would be morally permissible for him to murder whomever he wished),
even though the task of marking off which kind of impossibilities cause
trouble for a moral theory still requires doing.
16. By inadequacy, I mean that the view does not handle some particular
cases as the reader, in his considered judgment, would wish to see them
handled. Ofcourse, ifthe reader insists on seeing each example through the
principles he accepts [“Well, since it doesn’t violate principle P, it’s o.k.”]
then it will be impossible to produce what he will accept as a counterexam-
ple to the principles. Issues about how non-dogmatically to hold principles
so that they're open to counterexamples (as well as to counterarguments:
but even here, one sufficiently attached to a consistent principle P could
deny any statement Q from which not-P follows), but still to hold and accept
them rather than merely contemplating them, are of great interest and
merit extended discussion.
Followers of Miss Rand should not scorn holding principles in this
fashion, if 1am right in thinking that
(a) They do not possess a knock down deductive demonstration of their
principles
(b) a large part of the attraction of the Randian view for people is the way
it handles particular cases, the kind of considerations it brings to bear, its
“sense of life”. For many, the first time they encounter a libertarian view
saying that a rational life (with individual rights) is possible and justified is in
the writings of Miss Rand, and their finding such a view attractive, right,
etc., can easily lead them to think that the particular arguments Miss Rand
offers for the view are conclusive or adequate. Here it is not the argument
which has led them to accept the view, but rather the way the view codifies,
integrates, unifies, extends many of the judgments they want to make, feel
 On the Randian Argument 231

are right, and supports their aspirations. If this is so, then one should hold
the view so that it is open to challenge from just that sort of data that has
provided its main support.
Here we do well to mention a problem which has received little discus-
sion in the literature on ethics. Some writers on ethics have viewed their
task as offering moral principles which would unify and account for the
particular moral judgments we make, often adding that the reciprocal
process of formulating such principles, and modifying particular judgments
itself is the process of moral justification. We may ask the philosopher of
science’s question: Does the data uniquely determine one theory which
accounts for it, or are there alternative theories which equally well account
for all the data we have or could have? Asking this question forces us to
clarify the notion of all possible moral data: is it, for each act in each
situation, a pairing of the act in the situation with a judgment of moral
permissibility? Here it seems plausible to think that alternative different
theories will equally well account for all this data (as plausible as to think
that alternative physical theories can account for all possible observational
data). Perhaps adding into the data, other types of particular judgments
(e.g., of persons’ characters) will help, but one suspects, not much. More
promising is the claim that in the data, we have not only particular
judgments ofparticular actions, but also, often, some (partial) reason offered
for the judgment. And so the theory must not only yield the particular
judgment but also, when we have them, certain sorts of reasons for the
judgment. [I say “certain sorts” to mark a problem: if we offered as our
reason for a judgment, “because it follows from T” where T is a complete
fundamental moral theory, then any theory which accounted for our
judgments would have to be, contain, or yield T. So it is a more delimited
sort of reason which is needed. ] Placing this into the data, is it still the case
that there are alternative theories which equally well account for the data,
where no one theory accounts for and reduces all the others. If so, the
various well-known positions in the philosophy of science as to the content
of theories become available as options, and one may pick an option in
ethics different from the one one holds with regard to a particular theory of
physics. The whole area is open, and would repay detailed investigation.
17. Independently, Thomas Nagel argues against the principle of expe-
riential ethics in his paper “Death”, Nous, 1970.
 12
Nozick on the Randian
Argument
DOUGLAS DEN UYL and DOUGLAS RASMUSSEN*

Ever since the publication of Anarchy, State and Utopia Robert

Nozick has become, in the philosophic community at large, the
leading exponent of libertarian political philosophy. Nozick’s work
has become the subject of much interest among_ political
philosophers. Yet before Nozick’s rise to prominence contemporary
libertarian political philosophy had been in existence for well over a
decade. Before Nozick’s rise most academic libertarian political
philosophers looked to another figure as the philosophic head of
libertarian thought. This figure was the non-academic novelist/
philosopher Ayn Rand.! One would expect therefore that Nozick
would have some thoughts on the figure that was so dominant in
libertarian thought for so many years. Such is indeed the case, for in
the Spring 1971 issue of the Personalist (Vol. LII, No. 2) Nozick
devotes an entire article to an examination of Rand entitled “On The
Randian Argument.”
Nozick’s article seems to us to be of interest for a number of
reasons. First of all, Nozick and Rand share the same basic political
conclusions. Sharing similar conclusions is not in itself interesting,
but it can become so when one of the two parties claims the other is
fundamentally mistaken in the way his or her conclusions were
drawn. Such is Nozick’s claim about Rand. Secondly, when one

232
 Nozick on the Randian Argument 233

person attacks the position of another a good deal can be gathered

about the attacker's philosophic style and methodology. Such infor-
mation is always useful when that person is in a position of intellec-
tual prominence. Finally, whether the criticisms of one of the
parties succeeds is always of interest if for no other reason than the
bearing success or failure will have in determining the intellectual
scope of coming to similar conclusions.
In this paper we shall show that Nozick’s criticisms of Rand fail
completely, and that they fail largely because Nozick has not
understood or appreciated either the content or philosophic
methodology of Rand’s thinking and thought. Nozick’s own paper is
divided into five sections. We shall proceed by examining each of
these sections in turn. Nozick’s method in his paper was to present
the “Randian argument” and then to criticize the argument. We
shall operate by presenting Nozick’s construction of the Randian
argument, criticizing Nozick’s construction, and, where necessary,
we shall outline Rand’s own position on a given matter (where
Nozick has failed to accurately represent that position). Our paper is
meant only to apply to Nozick’s discussion of Rand. Whether what is
said here in any way affects or is relevant to Nozick’s own
philosophizing elsewhere is not our concern.
Before turning to the particular sections of Nozick’s paper we
must note that Nozick begins by saying that he will be basically
concerned with four central Randian conclusions. We quote these
conclusions below to help orient the reader to Nozick’s overall
project.

I. “only living beings have values with a point.”

Il. “life itself is a value to a living being which has it.”
Ill. “life, as a rational person, is a value to the person whose life it
is.”
IV. “some principles about interpersonal behavior and rights and
purposes.”
These four basic areas of concern given above shall be referred to in
the paper as Conclusion I, II, etc.
I
In order to examine Conclusion I, Nozick constructs what he
believes Rand’s basic argument for that conclusion to be. The
following is that argument:
234 Douglas Den Uyl and Douglas Rasmussen

(1) “Only a living being is capable of choosing among alternative

actions, or,
(2) “Only for a living being could there be any point to choosing
among alternative actions, for
(3) “Only a living being can be injured, damaged, have its welfare
diminished, etc., and
(4) “Any rational preference pattern will be connected with the
things mentioned in (3), and since
(5) “Values establish a (rational) preference ordering among alter-
native action, it follows that
(6) “Only a living being can have values, with some point to them.
Values have a purpose only for living beings.”

We shall examine these premises closely in order to see if Nozick

has constructed an argument truly representative of Rand's position.
We shall also consider, as we go along, the criticism Nozick offers
regarding this argument when it concerns a position that Rand does
indeed hold.
Before this can be done, however, we must note that the conclu-
sion Nozick sets up for consideration seems not to be one that Rand
holds! Nozick’s conclusion that “only living things have values with a
point” seems to mean that values have a purpose or serve a function
only for living beings, but Rand’s contention is that only living
beings can have values, not that only living beings have values with
a purpose. Rand holds that “value is that which one acts to gain
and/or keep.? In other words, value, when most generally consid-
ered, is an object of an action, an end, a goal or purpose. Thus,
when Rand states that “only a living entity can have goals or can
originate them,’* she is contending that values period exist only for
living things. She is contending that if there were no living things,
there would be no ends, no goals, no purposes, and no values in the
world. The very existence of goal-directed behavior is metaphysi-
cally dependent on the existence of living things for Rand, and
Conclusion I does not seem to mean this. Conclusion I merely notes
that values have a purpose only for living things and says nothing as
to whether there could be values (values without a purpose), if there
were no living beings. The qualifying prepositional phrase “with a
point” which Nozick applies to “value” is very important, because it
seems to leave open the question of there being values without
 Nozick on the Randian Argument 235

living beings. Rand’s position notes the metaphysical conditions

required for the existence of values and rules out the existence of
values save in relation to living things. Thus, there seems to be
quite a difference between Conclusion I and Rand’s central thesis.
Conclusion I does not seem to indicate the metaphysical nature of
Rand's approach to value, and because of this, Nozick seems to miss
an absolutely crucial element of Rand’s theory of value. Missing this
crucial element obscures what is philosophically significant about
Rand's approach to ethics.
It is quite puzzling that Nozick fails to get this most basic
contention right. Possibly he is not using “value” in the same way as
Rand; or maybe he wants to use the concept differently.> If such is
the case Nozick should have said so. On the other hand, maybe
Nozick means by Conclusion I, nothing more or less than, “there
would be no reason, no basis, no purpose for values if there were no
living beings.” In other words, maybe Conclusion I means exactly
what we said it should mean. Though this is not what Conclusion I
seems to say, we will take it as Nozick’s meaning so that we can
consider the premises he constructs for it.
(1) Premise (1) is not quite what Rand contends. According to
Rand, only human beings choose among alternative actions. Purpo-
sive behavior does not apply to insentient living things. Such living
things can act for one alternative as opposed to another, but only in
the sense that an organism’s actions result in the achievement of
some end or in failing to achieve some end. So, premise (1) should
be either “only a human being is capable of choosing among
alternative actions’ or “only a living being is capable of acting for
one alternative as opposed to another.” Premise (1) blurs an im-
portant distinction between facing alternatives and choosing be-
tween them. Living beings qua living beings face the alternative of
life or death, i.e., a living thing’s actions either result in ends that
are conducive to its life or not. Human beings qua living beings face
this alternative too, but as a human being they can choose what end
to achieve. This distinction is the basis for the further distinction
between “value” and “moral value” in Rand’s thought. Living things
as organisms which, as a result of automatic (or determined) func-
tions achieve ends (values) do not have the category of moral value
applied to their actions, while human beings which choose what
values to achieve with their actions do. This distinction allows there
236 Douglas Den Uyl and Douglas Rasmussen

to be values which are unchosen and does not let the will of man be
creative of all values. Metaphysically speaking, “will” is not the
fundamental factor behind the existence of values; rather it is life.
Yet, this distinction does let only those values which are chosen be
the ones subject to moral appraisal. By blurring the distinction
between facing alternatives and choosing between them, Nozick
gives the impression that for Rand values exist only as an object of
choice or will, i.e., only as a result of an individual’s aspirations, and
not as a result of possessing a characteristic he shares with other
entities—life. The blurring of this distinction is in part responsible
for Nozick’s belief that for Rand values could not exist for a person if
they were not an object of personal concern. For example, our not
knowing the consequences of an action we took and our inability to
be affected by it supposedly renders such an action incapable of
being a valuing one. Here the supposition is that values exist only
because they are an object of choice and because they affect us
personally. According to Rand, however, we have values because
we are alive; this is why there are choices and why we are affected
(or not) by them. Nozick interprets Rand’s theory of value in this
metaphysically egocentric manner in part because the distinction
between facing alternatives and choosing between them is blurred
in premise (1). We shall have more to say on this point later when
we consider premise (3).
(2) The key to understanding premise (2) is how we interpret that
troublesome term “point.” Yet no matter how we interpret this
term, this premise is subject to the same objection as (1). As we said,
Rand believes that only human beings choose among alternatives.
Let us put this difficulty aside, however, and consider what Nozick
means by “point.” He seems to be using the term to mean “reason
or basis.” In other words, (2) contends that only for a living being
could there be a reason or basis to choosing among alternative
actions. If so, this is something that Rand holds, for she argues that
the very phenomenon of choice is dependent on entities faced with
an alternative—namely living things. Only for living beings could
there be choice. It should be noted that this use of “point” seems to
be different than the use the term has in Conclusion I. As we have
already discussed Conclusion I, it is not necessary for us to repeat
that discussion here. Nevertheless, if there is a variation of the use
of the term “point” Nozick could be charged with building an
 Nozick on the Randian Argument oor

equivocation into the argument. Yet, setting this problem aside,

premise (2) does seem to say what Rand contends, i.e., that only
living beings are capable of choice, or more exactly, only human
beings can choose.
Nozick finds this contention problematic. He wonders if there
could not be machines created which could perform cognitive or
choice functions. He wonders also if there could be machines so
designed as to be capable of valuation. Supposedly, these specula-
tions indicate the shakey ground upon which the contentions “only
human beings choose” and “only living beings have values” rest.
Actually, this is not the case at all. If it is argued that computers,
robots or whatever can “choose” and man is not the only entity that
can choose, it can be replied that man creates or makes such
machines, and whatever capacity for choice a machine might have is
a result of man’s capacity for choice. Such machine choice is nothing
more than an extension of human choice. Metaphysically speaking,
choice exists only;for human beings. Man’s ability to expand the
domain of his choice is not a challenge to this contention; rather it is
an affirmation ofit. Also, ifa machine could be created with “values”
and “concerns,” this would be an extension of the human ability to
choose and value. If there were no human valuation, there would be
no machine valuation. Values exist only for living beings; more
specifically, they exist only in relation to them. Man’s ability to
lengthen the relationship between himself and his values, i.e., by
programming or “giving” machines ends and rules for action, only
underscores the very fact that values exist only in relation to a living
being.
(3) Premise (3) is a position that Rand holds, but it is not her basic
reason for (2) or (1) as Nozick indicates. The capacity to be harmed,
damaged, or have one’s welfare diminished does not constitute the
primary reason for claiming that either only human beings choose or
only living things act for one alternative as opposed to another.
According to Rand, the basic reason for either of these is the
phenomenon of life. It is the existence of life that makes alternatives
possible which includes the acting for or choosing of alternatives.
The capacity to be harmed, damaged or have one’s welfare di-
minished is merely a result of the phenomenon of life. Premise (3)
could just as well say that only a living being can be healthy and
have its welfare increased. There is nothing metaphysically primary
238 Douglas Den Uyl and Douglas Rasmussen

about this premise. Life is a way of being that requires constant action
for its continued existence. An entity which possesses the charac-
teristic of life must take actions whose results are either conducive
to its life or not. Being harmed, damaged, having one’s welfare
diminished or being healthy, whole, having one’s welfare increased
are manifestations of and dependent upon this basic fact. They
indicate that a living thing’s actions are conducive or not conducive
to its life. Yet such indications are important only because the entity
is a living thing. They are not the basic reason for (1) and (2).
Nozick’s failure to note the primary reason for (1) or (2) allows him
to interpret Rand as saying that a value can exist only insofar as it has
a concrete, particular effect upon the valuer. If there is no particular
result from one’s valuing X, then X is not a value. But Rand’s
position is more subtle than this. Rand holds that something is a
value because it is an object of an action, because it is an end.
She does not hold that an end is valuable only when it affects the one
who acts for it. Such a view would allow values to exist only through
their relation to the particular effect they produce on a particular
actor. This is what we have called the metaphysically egocentric
interpretation of Rand's theory of value. Yet, such an interpretation
fails to note that Rand is asking the basic metaphysical question of
“What are the conditions for the possibility of value?” which re-
quires a most basic answer. Something is a value not because it
produces particular effects on the one who acts for it. In fact, it
might be the case that none of the particular ends a particular actor
seeks ever actually affects him. Yet these ends would still be values
according to Rand, because such a consideration as what particular
effect an end has on the one who acts for it is not the reason why one
pursues ends, why an end is valuable. Values come into existence
because life is a way of being that requires that a certain type of
entity (a living one) be an actor for ends, a valuer of ends. Living
things are the only type of beings that can be affected by their
actions, and if such actions could not ultimately make a difference to
the life of a living being, there would be no values. Yet, none of
these necessitates that something is valuable only when some
particular effect results to the actor from the end being sought. Rand
is only arguing that if there were not this kind of being in the
world—living being—there would be no ends valued, no values.
 Nozick on the Randian Argument 239

What ends or values are conducive to the existence of a living being

(a valuer) is determined by the nature of the thing, and for a
choosing entity these considerations become moral considerations.
For these reasons Rand does not argue that the way an object of
an action affects one be fully determinate in advance before it can be
a value. Rather she argues that when a person is determining
whether or not a projected object of an action is to be of value, the
reasons behind such an action should be in accord with those
principles which determine the successful maintenance of one’s
(i.e., qua human being) life or not. This must be the case since it is
not always possible to determine what the affects of an action will
be. Moreover, some actions (e.g., exercise for the sake of health, to
use Aristotle’s example) may seem to adversely affect one when in
fact they lead to our long run advantage. This could only be
determined by looking to principles as guides to action. Sending
copies of Atlas Shrugged to beings in a distant place (to use Nozick’s
example) can be perfectly proper for Rand (assuming we have
nothing better to do) either as an act of good will (the desire to see
the other rational beings succeed in their lives) or as a prospective
means to better enhance the environment or community in which
one lives. Neither alternative (or explanation) conflicts with the
requirements of living a successful life. Indeed both are a staple of
the Randian ethics.®
(4) Premise (4) seems to imply that the standard for rational
conduct is the avoidance of harm, damage or diminished welfare.
Though it is not altogether clear what Nozick means by “rational,”
we assume that in this context it means that which is correct or
morally good. If so, then this is not the standard according to Rand.
Rand’s standard of goodness is man’s life: “that which is proper to
the life of a rational being is the good.”7 This standard is not
equivalent with momentary or merely physical survival, and it does
not mean “survival at any price.”® Rather, man’s survival as a
rational animal means the “terms, methods, conditions and goals
required for the survival of a rational being through the whole of his
lifespan—in all those aspects of existence which are open to his
choice.”? Yet, this premise (premise (4)) does not suggest such a
standard: on the contrary, it could be taken to imply the “survival at
any price” viewpoint. As such, premise (4) cannot be taken as
240 Douglas Den Uyl and Douglas Rasmussen

representing Rand’s position. Once again Nozick has failed to

reconstruct the argument in a way that expresses Rand's real
doctrine.
(5) Premise (5) is not Rand’s either. Values, goals, and ends do not
in themselves establish a rational (correct, right) ordering of alterna-
tive actions. Strictly speaking, a value is an object of an action;
whether such an action aims at a rational end (that which is proper to
the life of a rational being) cannot be determined from the fact that
something is an end of an action (a value). Nozick seems to be
confusing value with moral good, but values qua value for Rand do
not necessarily give, or even indicate, a rational ordering among
alternative actions. Rand states: “What is open to his (man’s) choice
is only . . . whether he will choose the right goals and values or
not.’1° Thus, premise (5) cannot be taken as expressing Rand’s
viewpoint either.
(6) Statement (6) is a virtual restatement of Conclusion I. We shall
consider our criticisms of Conclusion I to be equally applicable to
statement (6).
In this section we have seen that Nozick has not properly under-
stood the Randian position. In other words, his construction of the
“Randian” argument was not in fact Randian. Thus many of the
criticisms Nozick makes against the “Randian” argument necessarily
miss their mark. We have tried also in this section to offer as much
as possible just what Rand’s basic argument actually looks like.
Hopefully we have been able to indicate adequately just where
Nozick’s misunderstandings lie.

II

We have argued that Nozick’s reconstruction of Rand’s argument in

Section I is incorrect and that Nozick has failed to appreciate Rand’s
central thesis—namely, that only living beings have values. In
Section II Nozick continues his reconstruction attempts. Nozick
tries to find an argument that will justify the conclusion that life
itself is a value to a being which has it. He does this in a three step
argument:

(1) “Having values is itselfa value.

(2) “A necessary condition for a value is a value.
 Nozick on the Randian Argument 241

(3) “Life is a necessary condition for having values.

“Therefore, life itself is a value”
Nozick finds this argument unsatisfactory and moves on to what he
calls the “transcendental argument.” The “transcendental argument”
claims that the request for a demonstration of the valuable nature of
life is ultimately an inconsistent one, for it is only the phenomenon
of life that gives rise to values. Nozick finds this argument uncon-
vincing and finds no reason why “death” or the “greatest happiness
of the greatest number’ could not be the ultimate value. Thus
Nozick sees no convincing argument for the conclusion that life is a
value to a being which has it.
Some initial comments must be made about this section of
Nozick’s article. First, the conclusion to be established—namely
that life is a value to a being which has it—is not quite correct. It is
not that life is just a value for Rand; rather it is an ultimate value.
Second, premises (1) and (2) of the argument Nozick constructs are
not positions that Rand would hold. We know of nowhere in Rand’s
work where she holds that having values is itself a value or that a
necessary condition for a value is a value. Rand does contend that a
necessary condition for there being values is life, but this is premise
(3) of the argument and so cannot be taken to be reflected in premise
(2). Possibly, Nozick can show how premises (1) and (2) are premises
that Rand uses, but we do not see how, and thus we also see no
reason to consider them at this time. Third, premise (3) of the
argument Nozick constructs is definitely one that Rand holds. In fact
it is her central ethical thesis. Nozick, however, does not say
anything about this premise. His comments are confined to showing
the inadequacy of premises (1) and (2). This is regrettable, because a
careful consideration of premise (3) and all its implication is central
to any appreciation of the so-called “transcendental argument.”
We shall examine this “transcendental argument” and show how
it does establish the position that life is the ultimate value. In so
doing, we shall indicate why “death,” “the greatest happiness for
the greatest number” or any other such thing could not be the
ultimate value.
Rand’s approach to ethics begins with the question: what are
values and why are there such things? “Value” for Rand is a morally
neutral term in its most basic sense. A value is the object of an
242 Douglas Den Uyl and Douglas Rasmussen

action. A value can be the object of a purposive action or it can be an

object of a non-purposive action. Insentient living things attain
values in the sense that the automatic functions of the organism
result in the attainment of an end. So, a value is a goal or end of an
action— ‘that which one acts to gain and/or keep.” Given that there
is such goal-directed behavior, why is there such behavior? What
are the conditions that make such behavior possible? We have
already indicated the answers to these questions in section one, but
let us continue by providing a more complete explanation.
(1)Goal-directed behavior by its very nature implies that there is
an alternative present. If there were no question of achieving the
goal then there would be no reason to act to gain it. Goal-directed
behavior demands the existence of an alternative.
(2) Goal-directed behavior requires by its very nature the exis-
tence of an entity faced with an alternative, i.e., an entity to which
its actions could achieve or fail to achieve a goal. If success for failure
with respect to some goal were not conditional on the entity itself,
then there would be no reason or basis for that entity to act to
achieve the goal—it could have no goals. Goal-directed behavior
requires the existence of an entity faced with an alternative.
(3)Goal-directed behavior requires by its very nature that the
alternative faced by an entity make a difference, have an affect, or
have a consequence upon the entity. If the consequence of suc-
ceeding at achieving some goal were no different to an entity than
the consequence of failing to achieve some goal, then there would
be nothing to differentiate achieving some goal from not achieving
some goal. Hence no alternative would be faced by the entity. An
alternative must make a difference to the entity which faces it or
there can be no goal-directed behavior. So, in order for goal-
directed behavior to exist there must be: 1.) an alternative, 2.) an
entity whose actions are capable of succeeding or failing with
respect to some goal, and 3.) an alternative that makes a difference
to the entity which faces it.
All three of these conditions are present in a single class of
entities—living things. As Rand argues,

There is only one fundamental alternative in the universe: existence or

non-existence—and it pertains to a single class of entities: to living or-
ganisms. The existence of inanimate matter is unconditional, the existence
 Nozick on the Randian Argument 243

oflife is not: it depends on a specific course of action. Matter is indestructa-

ble, it changes its forms, but it cannot cease to exist. It is only a living
organism that faces a constant alternative: the issue of life or death. Life is a
process of self-sustaining and self-generated action. If an organism fails in
that action, it dies; its chemical elements remain, but its life goes out of
existence.!! (emphasis added)

Non-living things face no alternative of existence or non-existence.

A non-living thing, e.g., a sofa or boulder, may be open to the
possibility of non-existence. The sofa could be reconstructed into
something else—say a bed; a boulder could be smashed into a
million pieces so that it became gravel or sand, but neither of these
is an alternative that the sofa or boulder faces. They may be
respective possibilities open to the sofa or boulder but they are not
alternatives faced by them. The sofa or boulder does not achieve its
existence or fail to achieve it as a result of its actions. Its existence is
not an object, a result, or an end of its actions. The basic “stuff” of
the world may change or evolve toward increasing complexity or
simplicity, but it cannot cease to be—its existence is conditional on
nothing. Thus non-living being faces no alternative and hence
cannot perform goal-directed actions. Only living beings can do so.
Living beings necessitate or make possible the existence of goal-
directed behavior, and they are the only type of beings which do so.
Living beings may face many alternatives and may have many goals,
and they may achieve a goal or fail to, but they have no basis or
reason for acting to achieve a goal if there is ultimately no difference
between achieving or failing to achieve a goal. We can see this if we
will but consider the following question: What determines ifa goal is
achieved or not achieved? The only answer possible is the differ-
ence, the effect, it makes on the part of the entity who acted to
achieve the goal. There is a result that comes from achieving a goal
and a result that comes from failing to achieve a goal (even if there is
a lack ofa positive result). The difference in the result to the being
who acted to achieve the goal determines, tells us, if the goal is
achieved. What differentiates the results of goal-directed behavior?
The most fundamental difference possible is the difference between
existence and non-existence. If such a difference did not exist, if
some being were not conditional, if an action could not result in the
existence or non-existence of the entity that acted to achieve a goal,
244 Douglas Den Uyl and Douglas Rasmussen

then there would be no differences in the results of achieving a goal

or failing to. If there were no difference in result with respect to an
entity existing or not existing, then what other differences could
there be? What could make results differ if there were not this basic
difference? None. Thus, it is the difference between a living being
existing or not existing that creates all the other alternatives a living
being faces, and it is because life is something that must be
maintained that there are goals in the first place.
Rand holds that an ultimate value is, “the final goal or end to
which all lesser goals are the means—and it sets the standard by
which all lesser goals are evaluated.”!? Is there anything that
constitutes a final goal or end? From all that has been said, we see
that life is the ultimate or final goal of all goal-directed behavior.
Otherwise, there would not be such behavior, and this is Rand’s
very point when she states:

Without an ultimate goal or end, there can be no lesser goals or means: a

series of means going off into an infinite progression toward a non-existent
end is a metaphysical and epistemological impossibility. It is only an
ultimate end, an end in itself, that makes the existence of values possible.
Metaphysically, life is the only phenomenon that is an end in itself: a value
gained and kept by a constant process of action. Epistemologically, the
concept of “value” is genetically dependent upon and derived from the
antecedent concept of “life.”13

What determines whether the life of an entity is achieved? To be

a living thing and not be a particular sort of living thing is impossible
and, thus, we cannot speak of life as an ultimate end without also
adding the words “as the sort of thing it is.” It is the nature of the
living entity, the kind of thing it is, that determines whether the life
of the entity is achieved.
Rand has spoken of the ulimate end as the standard by which all
other ends are evaluated. When the ends to be evaluated are chosen
ones the ultimate end is the standard for moral evaluation. Life as
the sort of thing a living entity is, then, is the ultimate standard of
value; and since only human beings are capable of choosing their
ends, it is the life as a human being—man’s life qua man—that is the
standard for moral evaluation.
Why should this be the standard for moral evaluation? Why must
 Nozick on the Randian Argument 245

this be the ultimate moral value? Why not “death” or “the greatest
happiness for the greatest number’? Man’s life must be the standard
for judging moral value because this is the end toward which all
goal-directed action (in this case purposive action) is directed, and we
have already shown why goal-directed behavior depends on life. In-
deed, one cannot make a choice without implicitly choosing life as
the end. Let’s consider the following argument:
1) X is an object of choice
2) Y is a necessary condition for the existence of X as a
value.14 Y makes X as a value possible.
3) If P chooses (values) X, P must choose (value) the necessary
condition for P’s valuation of X.
4 P chooses (values) X.
Thus 5 P chooses (values) Y.
6 Y is man’s life.
Thus 7|gealglee
aRNa P chooses (values) man’s life in choosing (valuing) X.
In so far as one chooses, regardless of the choice, one must choose
(value) man’s life. It makes no sense to value some X without also
valuing that which makes the valuing of X possible (notice that this is
different from saying “that which makes X possible”). If one lets X
be equivalent to “death” or “the greatest happiness for the greatest
number,” one is able to have such a valuation only because of the
precondition of being a living being. Given that life is a necessary
condition for valuation, there is no other way we can value some-
thing without also (implicitly at least) valuing that which makes
valuation possible. Paradoxically perhaps, we could value not living
any longer, but in making such a value we must nevertheless value
life. Death, a living thing not-being, does not require any actions for
its maintenance. Death is not a positive way of being. Rather, it is a
negation—the absence of being a living thing. It has no required
actions; it has no needs. Death cannot be an ultimate value, then,
simply because it does not require any actions and thus cannot be
the reason or cause of goal-directed behavior.1° Therefore, we
cannot “suppose” death or anything else (other than life) as the
ultimate value, for the very activity of “holding something as a
value,” let alone as an ultimate one, depends on life being the
ultimate value in the sense of “ultimate” discussed earlier. Thus
246 Douglas Den Uyl and Douglas Rasmussen

there is an inconsistency in the request, “Prove life is valuable.” The

very meaning of “valuable” presupposes the value of life.
Man’s life is the ultimate moral value not because it is a precondi-
tion for other moral values as Nozick suggests. Rather, man’s life is a
precondition for moral values because it is the ultimate moral value.
Man’s life is the ultimate moral value because man’s life is the only
thing that fulfills those conditions we noted earlier for the existence
of goal directed behavior. Neither “death” nor “the greatest happi-
ness... qualifies in this regard. Incidently, more will be said on
valuing death in section five of this paper.
Finally, it should be noted that premise (3) of our characterization
of the “transcendental argument” given above might be considered
as equivalent to premise (2) of Nozick’s three-step argument, but
only under certain important conditions. (2) states that a necessary
condition for a value is a value. (3) states that a necessary condition
for someone’s ability to value is a value. These premises can be best
examined if we will but consider Nozick’s example of cancer being
the necessary condition for the existence of the value “being cured
of cancer.” Strictly speaking, cancer is a necessary condition for the
existence of the state or condition of “being cured of cancer,” but it
is not a necessary condition for the existence of the value “being
cured of cancer.” According to Rand, values do not exist without
valuers, and “being cured of cancer” is a value only in relation to
a living being which values that state or condition. As Nozick under-
stands premise (2) of this three-step argument, cancer would have to
be a value if one valued being cured of cancer. Here, Nozick views
values in a manner Rand would call intrinsic.1® Rand, however,
could view premise (2) as not requiring that cancer be a value,
because something is a value only in relation to someone’s ability to
value, and cancer is not a necessary condition for that. Premise (3) of
our argument has the advantage of not confusing necessary condi-
tions for the existence of something with the necessary condition for
the existence of something qua value. Premise (3) of our argument
emphasizes that something is a value only in relation to a living
thing and as such is much closer to Rand’s position than is Nozick’s
premise.
We have seen once again that Nozick’s criticisms of Rand do not
have the force they might otherwise have had because Nozick has
failed to come to grips with the full Randian position. This is
 Nozick on the Randian Argument 247

unfortunate because these first two sections lay the groundwork for
much of what comes in later sections. Had Nozick been more careful
in these two sections perhaps some of the errors he makes in later
sections could have been avoided.

II

In Section III of his paper Nozick considers the possibilities that

would allow Rand to conclude that life as a rational being is a value
to the person whose life it is. We shall not examine all the pos-
sibilities Nozick considers. In fact, we shall ignore most of them,17
for we believe that Nozick mentions the approach that Rand uses,
though he does not examine that approach. We are referring to the
last paragraph of Section III where Nozick states:

It might be said that a rational person follows principles, general policies,

and so we must consider those principles of action which make man’s
survival possible.

This we think is Rand’s basic approach to establishing the conclusion

that life as a rational animal is the moral good for man.
Nozick fails to examine this approach because he has not under-
stood the metaphysical level of Rand’s arguments. Values exist for a
human being not in virtue of the particular effect they produce for
his life but as a result of his nature as a particular kind of living
being. Because a living being is a creature for which the results of its
actions make a difference, a living being acts to attain certain ends.
What ends these are are determined by the nature of the entity. A
human being has the ability to choose what ends he will seek, but
the ends that are conducive to his life are set by his nature. As stated
in Section II, the ultimate end is not just life. It is impossible for a
living being to be and not be a living being of some kind—there is
no such thing as unspecified life, i.e., life existing in some abstract
way. Thus, the nature of a living thing—the kind of thing it
is—determines whether the life of the entity is achieved. Life as the
sort of thing it is, then, is the ultimate end.
This ultimate end consists of two distinct aspects—the life of the
entity and that entity's nature. Though distinct, these are never
separate in a living thing. Life, as such, explains the necessity for
248 Douglas Den Uyl and Douglas Rasmussen

goal directed action. (This is what Nozick considers the “consequen-

tialist” side of the argument.) A thing’s nature, as such, provides the
basis for evaluating the living thing's development or actions. It
provides the principles or rules that are used as standards for action.
(This is what Nozick considers the “formalist” side of the argument. )
These two aspects, however, cannot be considered separately when
speaking of an ultimate end. Living beings do not exist without
being of a certain kind, and natures do not exist save “in” the
entities which possess them—that which is separate in thought need
not be separate in reality. (Moreover, since “nature” is not a special
metaphysical constituent for Rand as it appears to be for Aristotle,
there is no bifurcation of an entity. Rand would argue that her
entities are through and through “one” or “whole.” Thus any
attempt to separate “life” from a thing’s “nature” or a thing's
“nature” from “life” would not be possible in a Randian system.)
Because of this basic unity between the aspects, a given “nature”
and “life,” it is not possible to separate consequential considerations
from formal considerations when arguing for man’s life qua man as
the standard of value. It is because life requires certain things for its
maintenance that a human being, with no automatic form of knowl-
edge, must use the only tool available to him—his conceptual
ability. The principle of rationality is dictated by the nature of the
human being and the life of ahuman being necessitates the adoption
of such a principle. Both the consequences and formal con-
siderations are part of one principle for Rand, and cannot be
considered separately. It is the grossest misunderstanding of “man’s
life qua rational animal” to attempt to consider the question, “is life
as a rational being necessary for the life of ahuman being?”, in ways
that separate “life” from the kind of life it is.
In the last paragraph, Nozick notes that it has not been shown (by
Rand) “why each person must follow the same principles, and why I
may not, as a rational being, have a clause in mine [my principle]
which recommends parasitism under certain conditions.” But the
answer to this has already been implied. Each person has the same
human nature and because of this the principles of action that guide
that person's actions must be the same principles that apply to others
as well. There cannot be a special clause allowing parasitism because
then we would not be talking about principles. Why must we be
talking in terms of principles? The kind of thing we are sets the goals
 Nozick on the Randian Argument 249

and dictates the type of actions that will achieve them. This level of
generality is inherent in our very being. We can, of course, ignore
all these things and act in any manner we feel like. We may or may
not live very long, but the fact remains that we would not be living
the life of ahuman being; we would be a metaphysical misfit living
by sheer luck and/or by the moral behavior of someone else. Rand
would say that such an existence would not be a happy one, but this
is an issue we cannot examine fully here. Suffice it to say that Rand’s
view of happiness is not one of a momentary feeling but something
like a feeling of rightness with the world—a sense of efficacy and
self-worth (see section four of this paper).
The point is that life as a rational being is the kind of life that is
proper or suited to the kind of being man is. As we have said those
principles derived from man’s nature are meant to guide him in
living a successful life as a rational being. Exclusion clauses are not
possible unless one wants to exclude himself from the human realm
altogether. In the Randian scheme of things one must argue (for a
moral principle in terms of whether or not the principle applies to
living a successful human life (which is what it means to argue in
terms of principles in this area). Thus if one were to argue for
parasitism, one would have to do so in general principled terms and
not in terms of exclusion clauses.
Of course, a human being can act in a manner inconsistent with
certain standards and not be literally dead, but such “non-death”
cannot be considered successful human life. The standards Rand
sets out are meant to guide men’s lives, which implies that not every
mistake or evasion will result in immediate and literal obliteration.
Possibly, this is not a sufficiently strong form of obligation for
Nozick. He seems to believe that for a moral principle to be really
obligatory immediate and devastating consequences must reign
down upon you if a moral principle is violated. Maybe Nozick
believes that this kind of moral catastrophism is necessary to keep
people “in line;” but it is unrealistic to expect it to do so, and it is not
necessary to abandon teleological ethics if such consequences are
not forthcoming. The consequences of an immoral action are seldom
as immediate or ostensible as moralists often want, but this is no
reason to say that the principles are any less obligatory. Moreover,
this in no way implies that Rand is somehow less meaningful when
talking in terms of following principles than is any other moralist.
250 Douglas Den Uyl and Douglas Rasmussen

IV

The central project of Section IV of Nozick’s paper is to criticize

Rand’s attempt to go from “living one’s life as a rational being is a
value” to a social conclusion about people’s rights.1® As he has done
in all previous sections of his paper Nozick sets out what he
conceives to be Rand’s argument. He provides us with a ten step
“Randian” argument. At certain points in this ten step argument
Nozick pauses to draw a conclusion and to criticize the argument up
to that point. Our procedure here will be to deal with the argument
just as Nozick does, i.e., we shall present his characterization of the
Randian argument and we shall pause for discussion where Nozick
pauses for discussion.19
The first sub-argument (of the ten step argument) is a three step
argument. The argument runs as follows:
1) “For each person the living and prolongation of his own life
is a value for him.
1’) “For each person, the living and prolongation of his own
life (as a rational being) is the greatest value for him.
Thus 2) “No person should sacrifice his life for another.”
After providing us with this sub-argument Nozick pauses to make
two comments: a.) Rand gives no arguments for 1’, and b.) however,
Rand might accept 1’ because she does say certain things about life
being an “ultimate” value. In view of the criticisms we shall make
shortly, we need not comment here on Nozick’s statement about 1’
(i.e., “a” above). This implies that we need not say anything about
“b” either since “b” depends on “a.” However, some comment on
“b” does seem to be in order here. First of all, it is rather curious
that, given the fact that Nozick is supposed to be constructing a
Randian argument, Nozick chose the term “greatest value” for 1’
when he goes on to tell us that Rand uses “ultimate value.” It would
seem more appropriate to stick to Rand’s own terminology. Perhaps
Nozick attaches some importance to his choice of words in this
context. If so, Nozick has not brought out that significance, and
there is no way we can see to derive that significance from reading
the text. We must therefore assume that the two phrases mean
essentially the same thing for Nozick. However, from our point of
view there does seem to be a difference, at least by implication,
 Nozick on the Randian Argument 251

between the phrases “ultimate” and “greatest” when applied to

“value.” When speaking of “greatest” one seems to be speaking of a
value that is greatest with reference to some standard. When
speaking of “ultimate” (at least in a Randian context) one is not only
referring to “life” the value but also “life” the standard of value. In
short, “ultimate” is metaphysically more potent in a Randian context
than is “greatest.” Nozick’s apparent assumption that the two terms
are the same reflects again his failure to fully appreciate the Randian
position.
Returning now to the three step argument given above, a number
of comments can be made. First of all, premise one is not un-
qualifiedly true as it stands. Prolongation of life is definitely not a
value at any cost for Rand. That this is so is evident from the John
Galt suicide example in Atlas Shrugged which Nozick himself
references in Section V of his paper (and which we shall discuss at
the appropriate time). Since premise one is not unqualifiedly true
neither is 1’. Nozick is thus initially on shakey ground in presenting
these premises as Randian.
Yet Nozick makes more fundamental errors in his three step
argument. Let us examine the argument more carefully. Apparently
the difference between 1 and 1’ is meant to be only one of degree,
i.e., one of going from “a value” to “the greatest value.” However,
as we argued above, “living and prolongation” is not unequivocally
the greatest value because it is not unequivocally a_ value.
Moreover, 1’ cannot logically serve as the connective premise to the
conclusion because the connective phrase “living and prolongation
of his own life” is used equivocally between 1 and 1’. The crucial
phrase “as a rational being’?° (which Nozick relegates to mere
parentheses) is present in 1’ but not in 1. As such the phrase “the
living . . .” is used equivocally in the argument. Thus not only are
the premises of questionable Randian character, but the argument
itself is faulty logically. Such an argument can hardly be a fair
representation of Rand's position.
Furthermore, it is not even clear why Nozick bothers to add the
phrase “as a rational being” in 1’. Given the way Nozick has set up
the argument the phrase is not necessary for getting to the conclu-
sion (i.e., 2) that Nozick draws. From the way Nozick takes the
argument, the conclusion could just as well have been drawn if the
phrase had been left out (though the fact that it is included still
252 Douglas Den Uyl and Douglas Rasmussen

provides evidence for our charge of equivocation above). This brings

us to the way Nozick does understand Rand as indicated by this
argument. Obviously the phrase “as a rational being” does not seem
to make much difference to Nozick. However, the qualification this
phrase provides is crucial to Rand, and the phrase should even have
been included in premise 1. Because Nozick ignores this phrase he
takes Rand to be saying that everyone’s concern ought to be with
their own “prolongation” or mere physical survival. This is clearly
not what Rand argues for as indicated by the following passage
dealing with man’s survival qua man.
It [man’s survival] does not mean a momentary or a merely physical survival.
It does not mean the momentary physical survival of a mindless brute,
waiting for another brute to crush his skull. It does not mean the
momentary physical survival of a crawling aggregate of muscles who is
willing to accept any terms, obey any thug and surrender any values, for the
sake of what is known as “survival at any price,” which may or may not last a
week or a year.?}
One can see from the above that Nozick has completely failed to
accurately represent Rand's position. The non-sacrificial principle in
Rand’s philosophy is justified not in terms of keeping oneself alive,
but rather in terms of following those standards appropriate to living
the life of a rational being. More will be said later about how the
Randian argument might look.
That Nozick takes Rand’s meaning in the way we have suggested
above is implied by step 3 which goes: “no person should sacrifice
another person (’s life) to himself (his own).” Nozick argues that
there is no reason given in the argument he constructed (i.e., 1, 1’,
2 and 3) that can answer the question of why one should not sacrifice
another to himself. Nozick takes this to be a forceful question
because he takes Rand’s argument to be on the level of physical
survival. Indeed, given the way Nozick seems to take the meaning
of the argument he is quite right. If mere prolongation of life were
the greatest value for men there would be no argument against
saying one should sacrifice others to oneself—in fact the argument
Nozick gives could very well imply that one should if the situation
arose. Along with the question of “why not,” Nozick should have
asked, “is such a way of acting in accord with the type of being man
is?” This latter question would have forced him to focus on Rand’s
 Nozick on the Randian Argument 253

metaphysic of man’s nature and thereby to the significance of the

phrase “as a rational being.” It is Rand’s view of human nature that
is behind the conclusions Nozick is examining and to which Nozick
should have paid more attention. It is certainly possible that a man
can keep himself alive by sacrificing others to himself. Yet the real
question (the one Rand is especially concerned with and the one
Nozick never seems to recognize) is, is this the kind of activity that
accords with the way man’s nature means him to behave22—is this
the kind of principle that can serve as a principle for guiding men
in general? We have already seen that Rand’s answer to this question
must be “no.”
Nozick argues that 3 of his argument is not a satisfactory social
conclusion so he goes on to consider a more “promising” approach
by focusing on rights. He offers the following argument as another
attempt to arrive at a Randian conclusion from supposedly Randian
premises:

4) “Each person has a right to his own life, i.e., to be free to

take the actions required by the nature of a rational being
for the support, the furtherance, the fulfillment and en-
joyment of his life.
5) “Since each person has this right, to force a person to
sacrifice his interests to your own violates this right.
6) “One should not violate another’s rights.
Thus 7) “One should not force another to sacrifice his interests for
your own or that of another person.”
We might begin our discussion of this second sub-argument by
asking why it is that Nozick thinks the above is a more promising
approach? Nozick never provides any clue as to what the answer to
this question might be. However, Nozick does seem to believe that,
whatever the reason, rights-talk is more promising than value-talk
and that the two represent two different and distinct areas of
inquiry. Whether or not the two areas really are distinct is beside
the point. Since Nozick is trying to construct a Randian argument he
should realize that for Rand the two areas are definitely not distinct.
For Rand it makes no sense to jump into rights-talk from out of
nowhere (as Nozick does here) because the notion of rights is for
Rand, a moral concept. Thus one’s ethical and thereby value theory
254 Douglas Den Uyl and Douglas Rasmussen

must be intact before one can legitimately move to rights-talk. The

movement for Rand is from a general ethical theory to a theory of
human rights (with the latter being dependent on the former). Rand
is quite clear on this matter as we see from the following passages.

Rights are a moral concept—the concept that provides a logical transition

from the principle guiding an individual’s actions to the principles guiding
his relationship with others . . . the link between the moral code of a man
and the legal code of society, between ethics and politics.
The principle of man’s individual rights represent[s] the extension of
morality into the social system. . . .2° (emphasis added)

Thus Nozick’s “more promising approach” is not promising at all

because it does not represent accurately (or even at all) the Randian
attempt to derive a theory of rights from her ethical theory.
But Nozick does seem to recognize that something more than just
starting with “rights” is needed, because he does ask the question of
why a person has a right to his own life (i.e., with reference to 4). He
shows some indication of grasping the Randian argument by con-
structing a third sub-argument to prove 4. This third sub-argument
has 4 as the conclusion, 1’ as the major premise and 8 as the minor
premise. 8 reads: “Each person has a right to be free to pursue his
greatest and highest value.” Except for 1’, which we have already
discussed, this sub-argument (1', 8, 4) would not be a bad charac-
terization of Rand’s position provided the premises were understood
in light of Rand’s metaphysical positions and argument. Unfortu-
nately Nozick does not decide to see the argument in this way. He
takes “greatest and highest value” in 8 to be a wholly subjectivist
affair (“greatest” and “highest” values meaning whatever anyone
decides is “greatest” and “highest”). Thus Nozick concludes that 8 is
too strong because Hitler did not have a right to pursue his greatest
and highest values. That a subjectivist position is not Randian is
clear from even the most superficial reading of Rand. Nozick
recognizes this and rebutts his own conclusion about Hitler in his
next sentence. Yet instead of trying to come to grips with what
might have been wrong with the way he has understood the
argument Nozick simply drops discussion of this sub-argument and
moves to another one to establish 4.
 Nozick on the Randian Argument 255

This last sub-argument centers around steps 9 and 10 and runs as

follows:
9) a) “For each person, he should pursue the maintenance of his
life as a rational being.
b) The “should” in (a) is and should be of more weight, for
him, than all other “shoulds.”
10) a) “Each person has a right to do what he should do.
b) Each person has a right, which is non-overrideable, to do
what he should (where this should has the most weight) do.”
Thus: Premise 4
Nozick’s first comment on the above is that he will not deal with
step 9 because that is meant to follow from things he has said in
Section II of his paper. In other words, Nozick claims step 9 does
not hold because of criticisms he made in Section II. By the same
sort of reasons, we shall not discuss step 9 here, since we have
handled the criticisms Nozick gave in his Section II earlier on in this
paper. '
If we look at the sub-argument (i.e., 9, 10, 4) as a whole we see
that it suffers from the same kind of problem as the previous
sub-arguments (i.e., 4, 5, 6, 7; and 1’, 8, 4). In all cases (here in 10a)
the notion of a “right” is simply introduced with no indication of
what might legitimize its use. As we have already mentioned, the
notion of rights for Rand is meant to follow from and be an aspect of
her general moral doctrine. Thus it is is not a Randian argument to
merely introduce the notion unexplained as Nozick does. If Nozick
wants to reply that in fact Rand does not succeed in deriving her
notion of rights from her general ethical views, then this is certainly
a proper thesis to pursue. However, if that thesis were true then
there would be no way to construct a Randian argument in the way
Nozick has tried to do here. In other words, given the way Rand
sees her own philosophic system, if she is unsuccessful in deriving
rights from her general ethical view, then it is completely pointless
to start talking in terms of rights. This causes one to wonder whether
Nozick has understood the relationship between rights and Rand's
overall ethical view, since in fact he provides no arguments to show
that Rand does not succeed in deriving her notion of “rights” from
her ethical view. This is one of the most serious difficulties in
256 Douglas Den Uyl and Douglas Rasmussen

Section IV of Nozick’s paper. One cannot attempt to derive Rand's

social conclusion (as Nozick set out to do in this section) without
saying something about the derivation of rights from Rand's ethical
view.24 This lack on Nozick’s part indicates a lack of serious appreci-
ation of Rand’s philosophy in this area.?°
With reference to step 10, Nozick, both in the body of the text
and in his footnote 7, asks the question of why, even granted that
one has a right (as in 10a and 10b), one should not interfere with that
right. This is certainly a curious question, for it seems to us that the
concept of a right includes the exercise of that right. Whether or not
one is actually allowed to exercise the right in practice is beside the
point. In principle, having a right entails being allowed to exercise
it—otherwise there is no meaning to “right” (as it is a concept
concerning, at least in part, implementation in action).2° Thus on
the level of principle, it would be meaningless to say that one both
has a right and also that there is no reason why one should not
interfere with exercising it. Thus if Nozick grants to Rand “rights-
talk” his question is meaningless. Of course, what has just been said
could be argued against if Nozick holds the view that rights are
merely prima facie, but nowhere does he say this. It is certainly the
case that Rand does not hold that rights are prime facie.
In conclusion, one can now see from what we have said above that
Nozick has completely failed to construct a Randian argument. Thus
any effective criticisms he does make are effective only against his
own constructions and not against Rand.
Nozick ends Section IV of his paper by making two further points.
First of all he says that Rand needs both a harmony of “shoulds” and
a harmony of “interests.” This radical distinction between “shoulds”
and “interests” strikes us as a purely Kantian move, and nowhere
does Nozick argue for the validity of the distinction. If Nozick is
coming out of a Kantian tradition he should make it clear to us that
this is where he is coming from. Our impression was, however, that
Nozick was trying to criticize Rand from within her own framework
which is explicitly not Kantian.
Secondly, Nozick places Rand in the “optimistic tradition” of
ethics. This, so far as it goes, strikes us as an accurate assessment.
However, Nozick shows no appreciation for the metaphysical nature
of the tradition and Rand's place in it. There is nothing problematic,
as Nozick seems to imply there is, about men having different
 Nozick on the Randian Argument 257

interests within this “optimistic” framework, since all a “harmony”

of interests and shoulds means is that every man ought to act in
accord with the demands of his nature as a human being. This leaves
a lot of room for divergent activities and interests as one can see
from the variety of interests, careers, and life-styles presented by
Rand in her novels. It may be true in general to say that the
“optimistic tradition” in ethics (e.g., Plato) has not been very
tolerant of diversity. Such is not the case, however, with Rand as
she has broadened the scope of what it means to live a rational life.27
Of course, we should quickly add also that Rand is not one who
believes in radical tolerance.
It would not be quite fair to conclude our discussion of Section IV
of Nozick’s paper without offering something that positively reflects
Rand's position. In what follows we shall briefly try to outline some
of the steps involved in moving from “life as a rational being” to the
social conclusion of liberty as a primary value. The statements made
here should be taken in light of what we have said in previous
sections of this papér. Moreover, what follows should be regarded as
an outline and not as an exhaustive and definitive statement of the
Randian derivation.?® Since we are not quoting directly from Rand
other such derivations (certainly more complete ones) might be
possible. Nevertheless what follows should help the reader see
more clearly what the Randian position looks like and thus where
Nozick has failed to appreciate the mode of philosophizing Rand
engages in.

1) Life is an ultimate end, and end in itself, for any living thing.
2) To be a living thing and not be a living thing of a particular
kind is impossible.
3) The particular kind of living thing an entity is determines
what one must mean when talking of“life” with respect to a
given entity.
Ci Thus, life as the kind of thing it is is the ultimate value for
each living thing. (c.f., Sec. II of this paper.)
4) A human being is that kind of living being which can be
designated as a rational animal.
Ce Thus, life as a rational animal is the ultimate value for each
person.
5) A rational animal is an animal whose mode of consciousness is
258 Douglas Den Uyl and Douglas Rasmussen

characterized by the use of concepts, i.e., conceptual aware-

ness is the basic differentiating form of awareness.
Cs Thus, conceptual awareness must characterize one who lives
as a rational animal, and one only lives as a rational animal in
so far as one engages in conceptual activity. Conceptual
awareness is a peculiarly human form of existence.
6) The conceptual mode of cognitive contact with reality is
man’s only means of determining how to deal with reality so
as to sustain his own existence.
7) Actions taken in accord with judgments of how to deal with
reality are man’s only means of dealing with reality so as to
sustain his own existence.
Ca Thus, living as a rational being means, minimally, acting in
accordance with conceptual judgments.
8) Conceptual awareness is not automatic. It must be initiated
and sustained by a constant act of choice or volition on the
part of the person. Conceptual awareness cannot exist save
through the person's choice to engage in such a mode of
awareness.
9) C4
Cs Thus, a precondition for living the life of a rational animal is
that within any given context one be free from interference
upon acting according to one’s judgment.
10) Since C2
11) And since Cs
Ce Anything which threatens the precondition for living the life
of a rational animal is of ultimate disvalue.
12) In a social context, the initiation of physical force (and by
extension the threat thereof) by one man against another
serves to destroy the precondition for living the life of a
rational animal, since acting upon one’s judgment becomes
impossible.
Cz Thus, the initiation of physical force is an ultimate disvalue.
13) Ca and C7 have moral significance, i.e., one ought to act as
described in C4 and ought not act as described in C7.29
Cs Thus, one human being ought not initiate the use of physical
force against another man.
14) “Rights” is a moral concept determining the limits or bound-
aries of human interaction.
 Nozick on the Randian Argument 259

15) Cs represents a boundary condition of human action.

Ca ‘Thus, one has no right to initiate the use of physical force
against another human being.
16) Because of Cz, Co is the ultimate social principle.
Cio Thus, no other social principle or other rights can include the
initiation of physical force as their means of being exer-
cised.3°
As we have already said, the above provides an outline of steps.
To any philosopher there would indeed be a number of controver-
sial points and a number of areas that need further elaboration. We
have already noted one such place at step 13. Many such steps could
be filled in by reading Rand’s own work. The filling in of other steps
could not be gotten from Rand’s work alone but would have to be
filled in by drawing out the implications of what Rand does say.
Whatever the case may be, the above scheme at least better
indicates the Randian argument than does Nozick’s own construc-
tion. Granted that Rand has not done the complete job in this area,
nevertheless she has provided sufficient guideposts for determining
her position. Such guideposts were passed over in the fourth section
of Nozick’s paper.

Vv

Nozick begins his discussion in Section V of this paper by saying that

he intends to deal with the other side of the social principle
question. The first side of this question was dealt with in Section IV
and was “don't sacrifice another to yourself.” The side to be dealt
with here is “don't sacrifice yourself to another.” In this connection
Nozick quotes Rand's statement of, “To live for his own sake means
the achievement of his own happiness is man’s highest moral pur-
pose.” Before examining Nozick’s arguments an initial point should
be made. It is not at all clear that Nozick dealt just with the “don’t
sacrifice another to oneself” principle in Section IV. Indeed, if we
go back to Section IV and review the argument there one wonders
whether Nozick has simply just forgotten what he had said earlier.
The first sub-argument here (1, 1’, and 2) explicitly deals with the
topic that is supposedly being taken up anew here in Section V.
Whatever the reasons may be, let us nevertheless examine what
Nozick does say in Section V.
260 Douglas Den Uyl and Douglas Rasmussen

As a counter example to the non-self-sacrifice principle Nozick

brings up the suicide example from Atlas Shrugged. The case
involves Galt threatening to commit suicide if Dagny, whom he
loves, is caught and tortured. Apart from this case looking like it
violates the non-self-sacrifice principle, Nozick goes on to argue that
Galt would also be committed to suicide if Dagny were to die of
some disease. This indicates, Nozick claims, little “moral fiber” on
Galt’s part.
Nozick’s analysis, however, suffers from the fatal flaw of context
dropping. In the first case Galt contemplates suicide because it
seems to him that the possibility of living in a world of rational
values is being closed off for him. Moreover, Galt is in a very unique
position here. It is he that his enemies are after (not Dagny). The
dramatic action of the novel indicates that were Galt to succumb to
his enemies (i.e., by meeting their demands) the chances for there
being a world of rational values at all would be non-existent. For the
same reason that many of the businessmen in the novel destroyed
their own businesses rather than give in, Galt is contemplating
suicide. In other words, rather than sell out Galt will make it
impossible for such values (the rational ones) to be expropriated.
The disease example is a far cry from such an extraordinary cir-
cumstance. Disease is part of the normal risk of living. Death by
disease hardly places Galt in the pivotal position he takes on in the
novel. The interesting question here is whether under normal
circumstances it is possible in Rand’s system for one to rationally
contemplate suicide at simply the death of one’s lover. So far as we
know a full answer to this question is not forthcoming from Rand.
Yet one can say that risking one’s life for one’s lover is a part of
Rand's doctrine in that taking risks for one’s highest value?! would
be a normal part of what one would be willing to do to try and keep
such a value. Nozick once again seems to think that living for one’s
own sake means merely striving to keep oneself alive. In any case,
we must recognize the somewhat unrealistic and dramatic charac-
terization of John Galt and the Galt/Taggart love affair. The picture
Rand paints is such that the only person meant for Galt (or worthy of
him) was Dagny and vice versa. Given such a condition perhaps the
possibility of suicide seems more intelligible. One should also
recognize that part of the reason Rand paints the picture she does is
to dramatize her view of love in human interaction. The partner in a
 Nozick on the Randian Argument 261

love relationship is the best embodiment of one’s own value judg-

ments. The partner best expresses those values one wants to see
most in others. Rand has, in effect, dramatized what the destruction
of that can mean to an individual. In short, Nozick’s lack of apprecia-
tion of differences in context do not allow him a means of dealing
fairly with the Randian position. It is ultimately, in our opinion,
context which must decide any question of suicide for Rand (such as
the question we posed above); and in all her novels the question of
risking one’s life and contemplating suicide are set in extraordinary
circumstances. The normal man seldom faces such circumstances
and usually has other ways of overcoming his grief in cases of
tragedy. Where all such channels are closed, suicide may not be an
irrational course of action. In any case, one cannot examine suicide
in abstracto. Context must always play a part in any analysis, which
it did not in Nozick’s account.
Nozick continues his discussion by asking a series of questions and
posing three kinds of problematic cases (given as (a), (b), and (c) ).
We cannot go into an explanation and analysis of each and every
question and problem Nozick poses (nor is there any need to).
However, the following quoted passages are a fair representation of
Nozick’s line of attack and should sufficiently indicate to the reader
what Nozick’s approach is.

A fruitless path to follow is to say that it is the short period of happiness

while he is sacrificing his life that justifies the whole thing. So that, in
thinking about a situation in which both are unconscious and only one can
be saved by a third party, he would prefer it were himself because in that
situation he wouldn't get to feel the happiness of saving her life? . . .
But (a) we can imagine he’s answering, quickly, a large number of
questions on some psychological test, so that there’s no time to pause for a
glow of happiness. Is it now for the happiness he knows he'll feel after the
test is over (mightn’t he forget and not recall the question?) that he answers
as he does?
(b) Why doesn’t he feel sad in contemplating his dying and her surviving?
To be sure, he may think it’s better than their both dying, or her dying and
his surviving, but why should the existence of some worse alternative . . .
make one happy in contemplating A [(a)?]? In fact, it doesn’t, and we don't
have this easy path to happiness.
(c) Most importantly, this answer gets things backwards. For one would
be made happy by placing one’s family in the only places in the raft only
262 Douglas Den Uyl and Douglas Rasmussen

because one values their survival above one’s own, their happiness above
one’s own.

The basic problem with the arguments quoted above (and with all
of what Nozick says here) is that he takes “happiness” for Rand to
be some kind of mometary “inner glow” or pleasure (e.g., as in (a)
and (b) above). Like many moderns, Nozick seems to boil all feelings
down to the same level, making no differentiation between an itch
on one’s nose and an aesthetic response to a great symphony. Rand
is very specific about what happiness means. She defines it as a state
“of non-contradictory joy.22 The implications of this definition,
which Nozick has missed, are that, while indeed happiness is a
psychic state, it is a state which is not a momentary pleasure or
isolated feeling but rather an attitude resulting from a well inte-
grated set of rational values utilized over a long period of making
choices and taking actions. Happiness is not a momentary “inner
glow’ or pleasure. The following passages from Rand indicate this
quite clearly.

Happiness is that state of consciousness which proceeds from the achieve-

ment of one’s values. If a man values productive work, his happiness is the
measure of his success in the service of his life. But if a man values . . .
mindless “kicks” . . . his alleged happiness is the measure of his success in
the service of his own destruction.
Neither life nor happiness can be achieved by the pursuit of irrational
whims. Just as man is free to attempt to survive by any random means, as a
parasite, a moocher, or a looter, but not free to succeed at it beyond the
range of the moment—-so he is free to seek his happiness in any irrational
fraud, any whim, any delusion, any mindless escape from reality, but not
free to succeed at it beyond the range of the moment nor to escape the
consequences.
. . And when one experiences the kind of pure happiness that is an end
in itself—the kind that makes one think: “This is worth living for’—what
one is greeting and affirming in emotional terms is the metaphysical fact
that life is an end in itself.38

Happiness in Rand's philosophy functions in much the same way as

in Plato’s and Aristotle’s. It is something that takes time and training
to achieve. If one is to call it a pleasure it is a pleasure of a vastly
different order and significance than the way we use the term in
 Nozick on the Randian Argument 263

everyday speech. Happiness is a lasting and continuous sense of

rightness with the world. The sense of this was meant to be
conveyed by Rand’s choice of the word “joy.” A more classical
synonym might be “felicity.”"°4 Because Nozick has failed to ap-
preciate Rand’s view of happiness (or to give any signs of such
appreciation), his criticisms are simply not directed at anything even
approaching a Randian doctrine.
Nozick’s mode of attack changes somewhat in the second half of
Section V. Here Nozick constructs two examples to show that one
can indeed be happy even though moral rules are violated.*®
The first case is where a father implants in his child a chemical
that makes him act according to the correct moral principles P at all
times except for certain situations S where “he knows that deviating
from them will be in his interests,” e.g., “by murdering someone
and taking his fortune.” The child afterwards forgets what he has
done in situation S and lives a life of bliss. Thus, so Nozick seems to
reason, we have a case where one is happy while violating moral
rules. Nozick also claims that one concerned with his own happiness
would want to have such a device implanted in himself.
The example falls wide of its mark on a number of counts: a) a
necessary part of what it means to live by moral principles is that the
one who lives by them be responsible for his actions. There is no
reason here to believe the child is acting as a morally responsible
agent either in situation S or in following principles P. To respond
by saying that whether the child was responsible or not makes no
difference because the point is that the child took the actions and
was happy, is not an intelligible response. In fact, the child did not
take the actions in any morally relevant way (though his body did
perform certain motions). Nozick might just as well have said that it
is possible to be happy by being put in a state of drugged stupor for
the rest of one’s life, for this is all Nozick’s example comes down to.
b) Nozick seems to be in the Kantian camp by regarding moral
activity as a mere obedience to moral law. His example comes down
to acontemporary facsimile of Kant’s arguments for the separation of
moral law from interest. Nozick, of course, never lets the reader
know explicitly just where he is coming from. c) Nozick seems to
think that happiness consists in merely having a fortune or the
like.3 This is certainly not obviously the way to happiness, and
there is absolutely nothing in Rand to indicate that the mere
264 Douglas Den Uyl and Douglas Rasmussen

possession of objects constitutes happiness. d) The case itself is

simply absurd. Nozick is going to have to assume away, the ques-
tions the child will ask about the fortune and where it came from,
the law and investigations into the case, rumors about the way the
fortune was gotten that might reach the child, doctors who might
come upon the device or chemical implanted in the child during an
examination, etc. In short, Nozick has to assume a worldwide
conspiracy. We might as well take this Cartesian path even further
and suppose that there is an evil genius who has deceived everyone
about moral principles! Such examples are not really worth analyz-
ing as they assume away too many relevant considerations; yet since
we have come this far we might as well finish the job. If we suppose
that there was this iron clad world conspiracy, that the child knows
absolutely nothing about it and also was fully convinced that all his
actions were done according to P and that the consequences of his
actions could not affect him in any way other than a “good” way,
etc., then we have reached a point where we might legitimately
respond in the following way: “show us how the ‘possible’ case you
have constructed precludes the possibility of the child actually being
the one who followed P while everyone else was deceived into
thinking they had deceived the child.” If Nozick cannot preclude
this possibility then either the case is a subjective fantasy and not a
real possibility (i.e., if Nozick responds by merely saying that he did
not set up the case that way), or there is no way to decide between
the two alternatives (i.e., between Nozick’s case or ours) and thus
what Nozick is offering us is no alternative at all (since an alternative
presupposes at least some way of deciding upon it, which means
enough information must at least be theoretically present to exclude
one of the alternatives).
Nozick’s second example is another case of world conspiracy.
Here we have a man who thinks he has lived properly but everyone
secretly despises him. Nozick’s basic error here is to suppose that
one’s happiness is derived from the opinions of others—something
that Rand explicitly denies.?7 Thus the example does not touch her
doctrine. But Nozick never tells us why the others despise him.
There are two basic options: 1) the people who despise the man are
evil and the man is good. This is essentially Plato’s example of the
just man suffering injustice completely alone, which is presented in
the Republic. Here Rand’s answer is similar to Plato’ s—happiness
 Nozick on the Randian Argument 265

necessarily results from right action whether others recognize the

correctness of those actions or not.38 Or b) the man was evil and the
good people despised him for it. If this is the case then Nozick begs
the question because he was going to show how one could be happy
and immoral. In short, both examples simply fail to deal with Rand.
A more promising approach to attacking Rand would be to use
similar arguments to those that have been advanced against tradi-
tional teleological ethical theories. This, however, is something
Nozick does not do.
Nozick concludes the final section of his paper by placing Rand in
a tradition which he calls “experiential ethics.” Experiential ethical
theories hold that the only facts relevant to moral assessments of
actions are how these actions do, or are intended to affect the
experiences of various persons.” Now if membership in the camp of
experiential ethical theorists is determined by Nozick’s under-
standing of Rand in this section, then clearly Rand does not belong
to that group of theorists. However, if membership in the camp
would include any theorist who does not divorce interest from
morality (e.g., Plato, Aristotle, Thomas Aquinas, etc.) then surely
Rand falls into the tradition. Nozick does not say enough to let us
know which of these alternatives he is referring to. Yet if Nozick
means to be attacking the whole tradition of interest theories then
the question is well beyond the Randian position (which is only one
of many). If such is the case, one paragraph (with a footnote to
Nagel) is hardly sufficient to combat that tradition.
We have seen throughout this paper that Nozick’s attacks on Rand
have failed largely because he has misunderstood or avoided under-
standing Rand’s doctrine. There are certainly places where this
misunderstanding is more excusable than others. Moreover, there is
no question that often Rand’s style of philosophizing is literary,
hyperbolic, and emotional. This always makes interpreting her
difficult and troublesome. Nevertheless, an implicit purpose of this
paper has been to show that there is a meaningful philosophic
doctrine underlying Rand’s seemingly merely ideological positions,
and that patience and sensitivity can bring that doctrine out. We
regard Nozick’s failure to understand Rand as probably due to the
difficulty of reading her properly. We also do not want to appear to
be claiming that Rand’s doctrine is beyond responsible criticism.
Yet for such criticism to be meaningful understanding must first be
266 Douglas Den Uyl and Douglas Rasmussen

present. Perhaps some of Nozick’s points could be forceful criticisms

if restated in light of what has been said here. Such criticisms would
certainly be welcomed by us.

NOTES

*This paper was presented at the annual meeting of the American

Association for the Philosophic Study of Society at Bowling Green State
University in Bowling Green, Ohio on October 9, 1976.
1. Though Rand has not engaged in any discussion of her philosophical
theories in any journal which we know of, aspects of her thought have been
considered by others in various journals and books. In The Personalist the
following articles deal with or use aspects of Rand’s thought: Nathaniel
Branden, “Rational Egoism: A Reply to Professor Emmons’ (Spring 1970);
Douglas Den Uyl, “Ethical Egoism and Gewirth’s PCC” (Autumn 1975);
William Dwyer, “The Argument Against an Objective Standard of Value”
(Spring 1974); Donald C. Emmons, “Rational Egoism: Random Observa-
tions” (Winter 1971); John Hospers, “Rule Egoism” (Autumn 1973); Dale
Lugenbehl: “The Argument for an Objective Standard of Value” (Spring
1974); Tibor Machan, “Rationale for Human Rights” (Spring 1971); Eric
Mack, “How to Derive Ethical Egoism” (Autumn 1971) and “Egoism and
Rights” (Winter 1973); George Mavrodes, “Property” (Summer 1972): and
Douglas Rasmussen, “A Critique of Rawls’ Theory of Justice” (Summer
1974). The following works by Tibor Machan utilize Rand’s philosophical
position either explicitly or implicitly: “Human Rights: Some Points of
Clarification,” The Journal of Critical Analysis (July/October 1973);
“Selfishness and Capitalism,” Inquiry (Autumn 1974); “Ayn Rand: A Con-
temporary Heretic?,” The Occasional Review (Winter 1976), and “Prima
Facie versus Natural (Human) Rights,” The Journal of Value Inquiry
(forthcoming). Relevant books by Machan include The Pseudo-Science of B.
F. Skinner (Arlington House, 1974) and Human Rights and Human Liber-
ties (Nelson Hall, 1975). Other discussions relevant here are: Edward Regis
Jr.’s review of John Hosper’s Libertarianism in the Journal of Critical
Analysis (January 1972); Kenneth J. Smith, “Ayn Rand: Objectivism or
Existentialism,” Religious Humanism (Winter 1970); Stephen E. Taylor, “Is
Ayn Rand Really Selfish . . . Or Only Confused?,” Journal of Thought
(January 1969); Austin Fagothey, Right and Reason, 6th edition (C. V.
Mosby, 1976); Vincent C. Punzo, Reflective Naturalism (Macmillan, 1969):
Hazel Barnes, An Existentialist Ethic (Vintage, 1971); John Hospers,
Human Conduct (Harcourt Brace Jovanovich, 1972), An Introduction to
Philosophical Analysis, 2nd ed. (Prentice-Hall, 1967), and Libertarianism
(Nash, 1971).
 Nozick on the Randian Argument 267

2. Ayn Rand, “The Objectivist Ethics,” The Virtue of Selfishness (New

York, The New American Library, Signet Edition), p. 15.
3. Rand does not believe that “purposive” is applicable to the behavior of
all living things. She judges it as inappropraite for insentiate nature.
“Goal-directed” is applicable to the behavior of insentiate living things only
in the sense that the automatic function of the organism results in the
achievement of some end. Ibid., p. 16.
4. Ibid., p. 16.
5. Since Conclusion I seems to say nothing as to whether there could be
values (values without a purpose) if there were no living beings, it may be
the case that Nozick views “value” as something which does not require a
valuer. In other words, it may for Nozick be that something could be a value
simply in virtue of what it was and not because it fulfilled some need or
purpose. Nozick may, most assuredly, speak this way, but he should not do
so when trying to present Rand’s position. c.f., note 16.
6. Nathaniel Branden, “Benevolence versus Altruism,” The Objectivist
Newsletter, Vol. I, No. 7 (July 1962), pp. 27-28.
7. Rand, “The Objectivist Ethics,” op. cit., p. 23.
B,. Ubid., ps 24.
On thd ps2.
10. Ibid., p. 22.
11. Ibid., p. 15.
12. Ibid, p. 17.
abies 1
14. As will become evident soon, if not already, Rand’s conception of
value is one that considers value as neither belonging to the subject alone
(as for example, an attribute of its action) nor to the object of the subject's
action alone (as for example, an attribute of the object itself). Rather, a value
is the relationship an aspect of reality holds to a living thing in virtue of the
characteristics both possess—namely, the characteristics of the aspect of
reality that cause it to be an object of an action, an end, and the characteris-
tics of the living thing that cause it to act for an end. It would be a mistake,
then, to attempt to understand Rand’s approach to value as either making
value a subjective phenomenon only or an intrinsic one only. We believe
Tibor R. Machan in his book Human Rights and Human Liberties (op. cit.),
has expressed Rand’s view of value very adequately when he wrote:
“[T]here are no intrinsically beautiful or good or right things, only things
that are good, right, or beautiful in relation to living entities for which
things can be good, right, and beautiful in terms of purposes or goals.” (p.
66)
15. Further, death is not even a value since it is not something one can
act to gain and/or keep. Death is “achieved” only in the sense that the life
268 Douglas Den Uyl and Douglas Rasmussen

sustaining actions of a living thing fail to maintain the entity's life, and thus
result in its not-being, death. In short, it makes no sense to speak of
“gaining a complete cessation.
16. c.f., “What is Capitalism,” Capitalism: The Unknown Ideal (New
York, New American Library, Signet Edition, 1967), pp. 21-22, for a
discussion of the “intrinsic” view of the good.
17. We also believe that our explanation of Rand’s approach shows how
these possibilities and their respective problems are not ones that concern
Rand's argument.
18. The phrase in quotes is Nozick’s way of speaking and not necessarily
Rand's.
19. Our numbering of the steps of the argument will of course follow
Nozick’s own.
20. One should note that even Nozick’s use of “as a rational being” can be
misleading. A more correct phraseology would be “as a rational animal.”
21. Rand, “Objectivist Ethics,” op. cit., p. 24.
22. This is the kind of question that indicates Rand’s place in the
Aristotelian/Thomistic tradition—a tradition that in some ways Rand her-
self acknowledges being in. Yet Nozick shows no indication of seeing that
Rand might fit in this tradition.
23. Rand, Capitalism: The Unknown Ideal; “Man’s Rights,” op. cit., p.
320; 321,
24. c.f., Tibor Machan, Human Rights and Human Liberties, op. cit.,
chapters 2-4.
25. Nozick does appear to be starting to correct this problem in footnote
7 but never really comes to the issue here either.
26. Indeed, Nozick even uses the term “non-overrideable” in 10b, but in
light of what he says afterwards this seems to be another case of a term
Nozick has merely chosen to ignore in his criticism.
27. For example, consider, “it is not the degree of aman’s ability nor the
scale of his work that is ethically relevant here, but the fullest and most
purposeful use of his mind.” “Objectivist Ethics,” op. cit., p. 26-27.
28. c.f., 24 above for a more complete account.
29. Admittedly this is a large step even given what we have said earlier in
the paper. To help clarify the rationale behind this step we refer the reader
to Eric Mack, “How to Derive Ethical Egoism,” op. cit.
30. The full implications of this principle can be obtained from both
Rand's fictional and non-fictional work.
31. It should be noted here that one’s highest value—in this context,
one’s lover—is not equivalent with the ultimate value which is man’s life
qua man.
32. “Objectivist Ethics,” op. cit., p. 29.
 Nozick on the Randian Argument 269

33. Ibid., p. 28, 29.

34. The term “felicity” is derived from the latin “feliciter” which expres-
ses a sense of “fruitfulness,” “abundance,” “advantage,” “favorableness,”
and “successfulness.” No sense of momentary “glows” or “pleasures” are
included.
35. Rand, of course, claims that, “Happiness is possible only to a rational
man, the man who desires nothing but rational goals, seeks nothing but
rational values, and finds his joy in nothing but rational actions.” Ibid., p.
29.
36. For a discussion of determining what is in one’s self-interest c.f. Den
Uyl, “Ethical Egoism and Gewirth’s PCC,” op. cit., Sec. I.
37. For a fuller discussion of this whole issue c.f., “The Psychological
Meaning of Man’s ‘Need’ of Approval,” in The Objectivist Newsletter,
March 1962, and “Social Metaphysics,” Nov. 1962, Vol. I, No. 3, 11.
38. This is not to say that one could not be happier if others also
recognized those actions as right.
 13
The Time-Frame Theory of
Governmental Legitimacy
ELLEN FRANKEL PAUL

Political theories grounded upon a natural rights moral foundation

have presented varied alternatives to the question of how people
come to be obligated to political regimes. In the early tradition of
natural rights argumentation, men like Locke! and Spencer? at-
tempted to combine a natural rights moral underpinning with
consent theory through the device of the social contract. A contem-
porary political philosopher, Professor Robert Nozick, attempts to
elucidate a new approach which would, presumably, replace the
social compact as an explanation of how men can become obligated
to government without their natural rights being violated. It will be
the contention of the first section of the ensuing paper that both
solutions to the problem of political obligation—the Lockian-
Spencerian and the Nozickian—fail, and that a new theory of
political legitimacy, the time-frame theory, can solve the problem
and still remain compatible with a natural rights moral framework.
In the final section, I will attempt to give an ontological deduction of
natural rights based upon a revision of the argument advanced by
Herbert Spencer in his essay, “The Great Political Superstition,”
which appeared in his work The Man Versus the State.

270
 The Time-Frame Theory 271

What can we make of a theory which attempts to combine both a

theory of consent (whether actual, as in Locke, or hypothetical, as in
Spencer) and one of natural rights to determine the nature and
extent of governmental legitimacy? Both Locke’s and Spencer's
arguments take the following form: 1) men have, by nature, the
natural rights of life, liberty, and property, for the protection of
which governments may be established; 2) a government may be
legitimately established by the unanimous consent of its future
citizens, and that government's authority has just such bounds as
they choose to impose, with the proviso that men cannot (Locke) or
will not (Spencer) consent to absolute, arbitrary rule; 3) when
government exceeds its legitimate function—i.e., legislating the
protection of natural rights—then it acts as an aggressor and may be
legitimately resisted by its citizens.
Is this account of governmental legitimacy internally consistent?
The answer is, clearly, in the negative. If natural rights are proven,
then the consent of individuals to government is purely extraneous
to the question of legitimacy and contradictory to a natural rights
position. It is contradictory to assert, it will be argued, that a rights
protecting state needs some sanction other than its performance as a
rights protector to render it legitimate.
What Locke and Spencer have done by arguing that consent,
whether hypothetical or actual, is the legitimator of government is
to pose an alternative theory to natural rights, without realizing
what they have done. If consent is the vehicle oflegitimacy, then it
is perfectly conceivable that individuals could voluntarily consent to
a government which systematically violated natural rights. Whims
of individuals being unpredictable, it is merely an act of faith on
their part to suppose that universal consent would yield the rights
protecting minimal state which they desire. Even the device of a
hypothetical contract is a tenuous gambit, for it is quite evident that
a Rawlsian? would derive a redistributive state from such a con-
tract, much to their distaste.
Thus, point (2), that government authority is derived from indi-
vidual consent, and (1), that governments are legitimate only when
they uphold natural rights, are not coextensive. To say that indi-
viduals by consent establish and set the limits to governmental
272 Ellen Frankel Paul

action, and then to assert that there is some other ethical

sanction—natural rights—by which laws are to be judged, is to
conflate two entirely different theories.
Finally, point (3) appears to direct citizens to break with moral
impunity laws which are not consistent with natural rights. The
possible effects of such an injunction might be that an individual
may break laws unanimously agreed upon by the contractees (or by
a majority), as long as he thinks his natural rights are being
abrogated. Obviously, Spencer and Locke cannot have it both ways;
either natural rights provide ethical sanction for government, or
individual consent establishes governmental legitimacy. By at-
tempting to combine the two theories they have succeeded only in
promulgating a theory which leads to contradictory evaluations in
specific cases; e.g., government X is legitimate because it was
established by universal consent, and government X is illegitimate
because its laws violate natural rights.
They were able to contain these two theories in alliance by the
belief that individuals would voluntarily and universally consent
only to the non-rights violating minimal state. But to make such an
assumption is, quite simply, to engage in an act of faith. It is
far more consistent to follow the Buchanan-Tullock* position,
agreeing that any government which emerges from an act of unani-
mous consent is legitimate even if it might establish a rights
violating state. If such a concession were made, as a consistent
contractarian would be compelled to admit, then it becomes appa-
rent that one must either be a contractarian on the question of the
origins of governmental legitimacy or a natural rights proponent.
We have discovered from the previous discussion that the at-
tempt to combine natural rights and contract-consent theory fails.
What I propose to accomplish in the ensuing argument is to
demonstrate that a consistent theory of governmental legitimacy can
be formulated on a natural rights moral underpinning alone, without
recourse to any element of consent theory. A defense of natural
rights will, once again, be assumed for the sake of this argument.
Robert Nozick, in his book Anarchy, State, and Utopia,® proposes
an historical account of how a minimal state might emerge from a
state of nature without violating anyone’s natural rights. While such
an account may be intrinsically interesting, it is entirely irrelevant, I
 The Time-Frame Theory 973

shall argue, to the question of determining whether government X,

Y, or Z is, in fact, legitimate.
What I propose in place of such an historical theory is an entirely
a-historical theory, one which examines the constitutional
framework, laws, acts of enforcement, and method of adjudication of
a given government over a designated period and asks only the
following question—is government X acting to preserve and protect
the natural rights of its subjects? If the answer is unequivocally (very
unlikely in the real world) or predominantly in the affirmative, then
it is a legitimate government and its citizens would be acting in
contravention of natural rights if they attempted to overthrow it; if
the answer is unequivocally or predominantly in the negative, then
it is illegitimate and may be overthrown with moral impunity and as
prudential calculations indicate an opportune moment. I have re-
frained from using the conventional terminiology—i.e., that when
government X is legitimate its citizens are obligated to obey it—
because of the statist connotations of such phraseology. It is not that
citizen A owes allegiance to legitimate state X, but that legitimate
state X owes citizen A the protection of his rights. The only thing
citizen A must do is to refrain from violating the rights of citizens B,
C,D... etc. He owes the government nothing, for its function is
simply, in Locke’s formulation, to give a univocal and universal
interpretation to the natural law, that is, the law which designates
exactly what categories of entities and territory qualify as property,
what constitutes a violation of an individual's right to life, and what
actions of one person limit the liberty of another.
I shall designate this theory of governmental legitimacy the
time-frame theory, the injunctions of which are the following:

1. Examine government X at time-frame tyt,, or tg-tz3 ... or

th-i-th;
2. Determine whether the government's constitutional structure
(written or traditional), body of laws, and method of enforcement
and adjudication are consistent with and protective of the natural
rights of the individuals who live under its jurisdiction;
3. If, on balance, the government is acting in a non-rights
violating manner (if it is not systematically violating rights by
conscripting, taxing, depriving people of their property, or not
274 Ellen Frankel Paul

protecting the lives or property of individuals, and furthermore, if

violations of rights occur as mere accidents, non-systematically and
are justly compensated), then it is legitimate;
4. Such a government, which systematically violates no one’s
rights, and protects the rights of all subjects from force, fraud, and
violence ought not to be acted against; anyone who overthrows such
a government would be acting in an immoral fashion, he would be
acting to endanger the sanctity and security of the natural rights of
his fellow citizens.

What this theory alleges, then, is that while Nozickian “entitle-

ment theory’® is the proper determinant of the legitimacy of private
property holdings and transactions, such an historical theory is
extraneous to a determination of the legitimacy of any particular
government. Under the time-frame theory, a government is legiti-
mate only if it is a non-rights violating government and it guarantees
the rights of its citizens from invasion by other individuals or states.
In other words, it must be a minimal state, limited in its functions to
internal and external defense, promulgating the natural law, and
establishing a judicial system with a final arbiter of disputes. Hence,
it makes no difference to the question of legitimacy how such a
government came to power, whether by violence, conquest, fraud,
tradition, election, laws of succession, or whatever.
What about the cases in which a government has come to power
through such unsavory, rights violating acts as mass slaughter or the
forcible expropriation of peoples’ property? Would such a state be
legitimate under the time-frame theory if it were now, in tj-to,
non-rights violating. The answer would be yes, it is legitimate. But
what about the individuals who now occupy positions of power in
this government and who were massive violators of the natural
rights of individuals prior to their acquisition of power? The Nozic-
kian entitlement theory would apply to them as individuals, not as
current officials of a government. The charges would be that they
attained the property of individuals A, B, C, D by illegitimate
means (not by either first occupancy, purchase, transfer, bequest, or
inheritance from the legitimate title holder) and they deprived
individuals E, F, G, H of their lives. As rights violators the present
minimal state would try and punish them. If it failed to do so it
would lose its status as a legitimate state.
 The Time-Frame Theory 275

Can this theory successfully handle such extreme cases as the

following: A government comes to power through extensive rights
violations and, now, all property is in the hands of or under the
direct control of a dictator, and yet the government was strictly
non-rights violating in t)-tz. Would it be a legitimate government?
Well, we might say it is illegitimate because the government goes
beyond the minimal state in being an owner of property and an
intervener in the economic realm which ought to be strictly separate
from government. But what about the case where the dictator as a
private individual owned all the wealth in the country and everyone
else was forced to do his bidding in order to remain alive. In this
instance, the government is not exercising any illegitimate
economic function and the dictator is not using the power of the
state to coerce anyone because the government owns no property
and simply defends the property rights of citizen Dictator under the
minimal state. The answer to the question of legitimacy would be
precisely this—that the government is legitimate because it upholds
rights and is a minimal state. However, the dictator who gained his
wealth by violating property rights before he acquired political
power is a criminal under the property “entitlement theory,” and he
holds no legitimate title to the property he forcibly confiscated.
Thus, the original title holders could bring their cases before the
judiciary of the dictator's minimal state. If their claims were hon-
ored and their property restored, an unlikely occurrence, then the
government would retain its legitimacy. If, as is far more likely, the
dictator's judges dismissed the legitimate claims of the title holders,
then it would be acting in a rights violating manner and the
government would no longer be legitimate. In Locke’s felicitous
phrase, one would then legitimately (i.e., morally) make an “appeal
to heaven.”
The previous example helps clarify the way in which the time-
frame theory handles the question of governmental legitimacy: the
theory applies to already existing governments, it is present
oriented, and it is an a-historical principle. Thus, it stands in
contrast to the theory of governmental legitimacy which is past
oriented and historical.
What course can the individual pursue against a government
which is a gross violator of rights under the time-frame theory? Ifa
government is a massive transgressor against individual rights, then,
276 Ellen Frankel Paul

obviously, there is no immorality entailed by acts which violate its

laws that are themselves rights-violating. And, in more extreme
cases of gross and systematic incursions by government on indi-
vidual rights, revolutionary measures would be justified; i.e., a
revolution that would seek to establish a minimal state. Whether
one obeys the edicts of such an “outlaw” government is simply a
prudential, and not a moral, question to be decided on conse-
quentialist grounds; i.e., will my resistance be successful, am I sacri-
ficing myself and my own self-interest without even a remote chance
of securing the minimal state, is my chance of successful resistance
worth to me the likely penalty if I fail, is the time ripe, is it likely
that the present government will be replaced by one even worse
than the present one rather than the minimal state I seek. . . .?
Is this theory unduly destabilizing and destructive of existing
governments as Jeremy Bentham, for one, feared from all a priori,
natural rights theories? The answer is yes and no. Yes, since in theory
it undercuts any supposed “moral obligations” to obey the edicts of
almost all’existing governments, and some of the rights violating
laws of even the best governments. But probably no, on prudential
grounds, because in all but the most grievous cases of governmental
oppression men will bear quite a lot from government without
resorting to arms, as Locke foresaw.7 And as the proliferation and
apparent permenance of contemporary totalitarian states indicate,
where rights transgressions are the most egregious and systematic,
the opportunities, and indeed the inclination, for resistance are the
most restricted. Rather than being destructive of all governments,
the time-frame theory provides a much needed antidote to the
twentieth century fascination with the all encompassing, over-
weaning, rights-violating, bureaucratic state.
One further point of clarification ought to be made, although of
necessity rather briefly, and that is in refutation of the natural rights
anarchists’ claim that even the minimal state is by its very nature an
offender against natural rights. The force of this claim is somewhat
blunted when the minimal state is denied the powers of conscription
and compulsory taxation, and when it protects only those individu-
als who voluntarily contract with it to defend their rights. However,
the appeal of the anarchists’ claim lies in their contentions that a) the
minimal state limits your rights by denying to you the discretion to
decide what in actuality constitutes a violation of your rights, and b)
 The Time-Frame Theory 277

the state acts as a rights violator when it imposes its laws upon an
innocent person and comes, mistakenly, to arrest him.
In regard to contention (a), I would claim that it is not an inherent
component of natural rights theory that each individual has discre-
tion over the interpretation of what laws will most felicitously give
expression to the natural rights of life, liberty, and property. As
Locke contended,® in the state of nature where no common power
exists to promulgate the natural law (the law which lays out the
limits of the natural rights of persons living in groups) each indi-
vidual can be a legislator and an enforcer of his own natural rights, (I
am not, here, adopting an historical argument from a state of nature,
but simply illustrating what is and what is not an inherent part of
natural rights). That when men live under a government this power
they had when government did not exist is taken from them, cannot
be gainsaid. Yet, this power to legislate for themselves in a state of
nature is the result of there being no more efficient mechanism for
universally (that is, throughout a given territory) giving expression
to the natural law. It is not a moral question, this question of who
shall interpret the natural law, as would be the question of whether
men do in fact have natural rights, but purely a prudential question.
That is, in whose hands will the power to give written expression to
the natural law provide the most efficient framework for the protec-
tion of natural rights. On Lockian grounds? of efficiency—i.e., that a
state of nature provides no common judge, no readily ascertainable
laws, no means of surmounting partiality, etc.—I would contend
that the minimal state provides a more efficient means of giving
definition to natural rights than would the state of nature scenario in
which each individual acts as legislator. By the state’s promulgating
laws in accordance with the natural rights of the individual, one is
not being deprived of an inherent capacity given to one irrevocably
under the natural rights theory. It is, nevertheless, true that the
individual is not as free as he was in a state of nature. A prudential
trade-off has occurred: a tangential, circumstantial freedom for each
to legislate for himself has been abrogated in order for natural rights
to be systematically protected.
As regards the anarchists’ second point, it is true that the state
violates an individual’s rights when he is unjustly arrested. How-
ever, the libertarian anarchists’ protection agencies would be placed
in the same position of rights violators if they came to arrest an
278 Ellen Frankel Paul

innocent person. Under anarchism one would have the right to

resist such an unjust enforcer, while under the minimal state one
would have a right to compensation and damages for false arrest.
Again, it is purely a prudential question—which system protects
rights more efficiently: one that sanctions a multiplicity of protection
agencies, each legislating their own interpretations of the natural
law, arresting and punishing offenders, making their own share of
mistakes in arrests, and being resisted by innocent (or even guilty)
individuals whom they come to arrest—the anarchist model; or one
that sanctions a single legislator, who enforces and makes its share of
mistakes, and compensates the unjustly accused in a peaceful
transaction in accordance with the rule of law—the minimal state
model?

II

Throughout the previous discussion of the problem of governmental

legitimacy, I have simply assumed that natural rights is the proper
moral foundation for a political system. However, leaving matters as
they presently stand, with natural rights being nothing more than a
hypothesis, seems philosophically incomplete. It is my hope and
expectation that we can do better than that. Herbert Spencer’s
deduction of natural rights will present an interesting, albeit ulti-
mately unsatisfactory, beginning.
Spencer begins by transferring the question of natural rights from
the “court of politics” to the “court of science;” he proposes to trace
men’s rights back to the “laws oflife;” i.e., the general conditions of
individual life, and only then, the general conditions of social life.
The verdict will, he claims, be the same in both cases. In Spencer's
words:

Animal life involves waste; waste must be met by repair; repair implies
nutrition. Again, nutrition presupposes obtainment of food; food cannot be
gotten without powers of prehension and, usually, of locomotion; and that
these powers may achieve their ends there must be freedom to move about.
If you shut up a mammal in a small space, or tie its limbs together, or take
from it the food it has procured, you eventually, by persistence in one or
other of these courses, cause its death. Passing a certain point, hindrance to
the fulfillment of these requirements is fatal. And all this, which holds of the
higher animals at large, of course holds of man.?°
 The Time-Frame Theory 279

Given that such are the conditions of life, is there any ethical
warrant to be found for these actions, by which life is sustained?
Well, says Spencer, if we adopt philosophical pessimism with the
implication that life in general is evil and should be speedily
terminated, then there is no moral sanction for these life maintain-
ing activities, and the whole question becomes moot. However, if
we adopt the optimistic or melioristic claim that life is a value, that
life is good (i.e., that, on balance, life yields more pleasure than pain
or that it is tending toward that pleasurable advantage), then these
actions by which life is maintained are justifiable, and so is the
freedom which enables us to perform them:

Those who hold that life is valuable, hold, by implication, that men ought
not to be prevented from carrying on life-sustaining activities. In other
words, if it is said to be “right” that they should carry them on, then, by
permutation, we get the assertion that they “have a right” to carry them on.
Clearly the conception of “natural rights” originates in recognition of the
truth that if life is justifiable, there must be a justification for the perform-
ance of acts essential to its preservation; and, therefore, a justification for
those liberties and claims which make such acts possible.

We have not, however, reached the desired ethical warrant as

yet, because these conditions are true of the lower creatures as well
as of man. “Ethical character,” Spencer contends, “arises only with
the distinction between what the individual may do in carrying on
his life-sustaining activities, and what he may not do.”!? The
distinction arises as a result of the presence of other men. And men
in close proximity are likely to interfere with one another; so, “in the
absence of proof that some may do what they will without limit,
while others may not, mutual limitation is necessitated. The non-
ethical form of the right to pursue ends, passes into the ethical form
when there is recognized the difference between acts which can be
performed without transgressing the limits, and others which can-
not be so performed.18
As regards the question of the general conditions of social life, it
will be apparent that while the positive element of the natural life to
carry on life-sustaining activities originates from the laws of life, the
negative elements which proscribes limitations and, hence, the
ethical constituent, arises from the conditions produced by social
aggregation. Social life is prompted by the advantages which men
280 Ellen Frankel Paul

derive from it; the condition of social cooperation, then, is that those
who join together gain as individuals from that union, i.e., that all
shall share in the benefits.
In order to facilitate our investigation of Spencer's proof of natural
rights, I have laid out his argument in a more clearly deductive
fashion.

Herbert Spencer's Defense of Natural Rights

I. From the “laws oflife,” i.e., physical laws:
1. Animal life involves waste;
2. Waste must be met by repair;
3. Repair implies nutrition. Nutrition presupposes obtainment offood;
food cannot be gotten without powers of prehension and usually
locomotion;
4. That these powers (prehension and locomotion) may achieve their
ends, there must be freedom to move about. If you shut up a
mammal in a small space, or tie its limbs together, or take from it
the food it has procured, you eventually, by persistence on one or
other of these courses, cause its death. Passing a certain point,
hindrance to the fulfillment of these requirements is fatal. All this
holds of higher animals and of men.
Such are the conditions of life.
II. Is there any ethical warrant to be found for these actions by which life
is sustained? No ethical warrant if we adopt philosophical pessimism—
that life in general is evil and should be terminated. If we adopt the
optimistic claim that life is a value, that life is good (balance of pleasure
over pain, or tending towards it), then these actions by which life is
maintained are justifiable, and so is the freedom which enables us to
perform them. So:
1. Those who hold that life is valuable, hold by implication;
2. That man ought not to be prevented from carrying out life-
sustaining activities. If it is said to be “right” that they should carry
them on, then, by permutation we get the assertion that they “have
aright” to carry them on. Thus—f life is justifiable there must be a
justification for the performance of acts essential to its preservation
and, therefore, a justification for those liberties and claims which
make such acts possible.
III. We still need an ethical warrant because these conditions are true for
lower animals as well as man. Ethical character arises only with the
distinction between what the individual may do in carrying on his
life-sustaining activities, and what he may not do.
 The Time-Frame Theory 281

_ . This dictinction arises as a result of the presence of other men.

bo . Men in close proximity are likely to interfere with one another.
3. In absence of proof that some may do what they will without limit,
while others may not, mutual limitation is necessitated.
4. The non-ethical form of the right to pursue ends, passes into the
ethical form when there is recognized the difference between acts
which can be performed without transgressing the limits, and
others which cannot be so performed.
IV. The conditions of social life—social life is prompted by the advantages
which men derive from it; the condition of social co-operation is that
those who join together gain as individuals from that union. The
conditions for the maintenance of both individual life and social life are
the same—the recognition and maintenance of individual rights.
1. The positive element of the natural right to carry on life-sustaining
activities originates from the laws of life;
2. The negative element which proscribes limitations, the ethical
constituent, arises from the conditions produced by social aggrega-
tion. :

Now, there are several glaring weaknesses in this argument which

must be isolated and rectified before a refined Spencerian deduction
can be proposed. 1) The principal difficulty with the argument arises
when Spencer comes to the point of transition between statements
of fact (that is, the conditions of man’s life) and ethical warrants. He
is left in the position of saying nothing stronger than that those who
hold life as a value (the optimists or meliorists) hold by implication
that life-sustaining activities are justified and ought not to be
prevented. But he has not provided us with an argument to show
why life is an ultimate value. In other words, as his argument stands
in this incomplete form, it is a matter of individual whim whether
one adopts the optimistic or pessimistic position. And, in conse-
quence, his argument would have nothing at all to say to those who
adopt the latter position. What is required, then, is an argument to
show that life is an ultimate value, independent of whether men
think it is or not. 2) There is nothing explicitly offered in this
argument to distinguish animals and, indeed, plants from human
beings, and to demonstrate why the latter have a right to life and the
two former do not. 3) The permutation in [II.2] from saying “It is
right that I do X” to “I have a right to do X” seems to be little more
than a facile verbal play on two meanings of ‘right.’ If ‘good,’ for
example, is plugged into the first statement, then the parallelism
282 Ellen Frankel Paul

crumbles. 4) And, finally, the ethical component of the argument

enters at step [II] rather than [III], as Spencer claimed, because
point [II.2] definitely makes an “ought” claim. If Spencer's deduc-
tion of natural rights is to be salvaged, then these errors and
omissions must be rectified.
(1) Spencer’s hypothetical statement of life as the ultimate value is
unsatisfactory for an ontological defense of natural rights. However,
it can be demonstrated that his choice of optimism over pessimism is
necessary, rather than contingent upon the whim and temperamen-
tal proclivities of individuals. The argument would take the follow-
ing form: if individual X values anything, he must necessarily value
life. Life, then, is the precondition for holding all other values. For
person X to say “I value money, but I do not value my life” would be
to utter a contradiction. Without life, X can value nothing.!4 Even
the suicidal individual, as he values his own death, must value life
until the instant that he terminates it by the selected means. As long
as he is an agent, that is, as long as he is alive and not comatose, life
must be his ultimate value.
(2) There is nothing in Spencer's argument to distinguish men
from beasts and plants, and to determine why men have rights and
these entities do not. This problem can be surmounted by two
steps. The first would be to place in section [I] of Spencer's
argument the following statement: It is a requirement of life-
sustaining action that repair involves the acquisition of food; food for
humans requires the sacrifice of lower orders of beings. This would
be a “condition of life” statement, to employ Spencer's terminology,
and would involve no ethical component. It is indisputable that man
must eat something, whether plants or animals is debatable, in
order to remain alive. The ethical content—that is, what we should
eat—would enter at a later stage in the argument. The second move
would occur at stage [III] where the fact of the existence of other
men is brought into the argument and the ethical consequences
elicited. The claim would have to be made that there is some
capacity within man that separates him from the lower forms of
existence, e.g., reason, capacity for suffering, ability to laugh,
featherless biped, etc. None of these possibilities appear overly
promising, for animals can reason and generalize (e.g., the recent
experiments in teaching language to monkeys), albeit over a much
 The Time-Frame Theory 283

narrower range than can the human mind. What I propose, instead,
is the human capacity to recognize the transposition of the argument
for my individual right to life (from section II) to a similar right for all
beings possessing the same mature. This would eliminate very
intelligent apes and dogs from the natural rights framework because
they have no demonstrated capacity to recognize that if they have a
right to life derived from their “conditions of life,” so does every
other creature of their kind. But a human being can make such a
transposition. By denying a similar right to life to other human
beings of the same nature as mine, I have undercut my own claim to
a right to life. Now, what about the case of Ardbankous from the
planet Margenfufel who can a) understand the derivation of their
individual rights to life, and b) make the transposition and see the
claims of other Ardbankous and human beings to their rights to life.
Quite apparently, then, they fall under the natural rights nexus and
we must not deprive them of their life, liberty, or property no
matter how alien or repugnant their physiognomy might be.
Objections (3) and (4) will be met by the revised form in which the
Spencerian deduction will be advanced. The attempt at an ontologi-
cal defense of natural rights takes the following form:
I. From the “laws of life,” i.e., physical laws:
1. Animal life involves waste.
2. Waste must be met by repair.
3. Repair implies nutrition. Nutrition presupposes the obtainment of
food; food cannot be gotten without powers of prehension and,
usually, locomotion.
4. It is a fact of reality that food can be obtained only by killing and
preying upon other creatures, by the strong or the cunning van-
quishing the weak or stupid.
5. That these powers of prehension and locomotion may attain their
ends, there must be freedom to move about. Beyond a certain point
hindrance to such powers by one creature over another brings about
the latter's death.
II. The Ethical Component.
1. Life is the ultimate value for each being because it is the precondi-
tion for the holding of all other values. Without life no valuation
process is possible.
2. Life being the ultimate good for each existent being, each agent, he
has a right to seek those life-sustaining means which are requisite to
284 Ellen Frankel Paul

his existence. Hence, a right to life. (To deny this would be to

become ensnared in a contradiction—that is to say that life is the
ultimate value but it is wrong to attempt to sustain it.)
. Life is justifiable (it is the ultimate good for him who has it); hence,
there is a justification for the performance of acts essential to its
preservation, and consequently, a justification for those liberties
which make such acts possible. Hence, a right to liberty.
. From the rights to life and liberty (that is, from the essential
motility requirements to sustain life, from [I], comes the right to
property; i.e., the right to that with which one has a) mixed one’s
labor by removing it from a state of nonownership, or b) acquired by
means of a voluntary exchange, bequest, gift or inheritance. With-
out property in that which one has labored to attain, one could not
sustain one’s life. Life is the good for any individual and property is
a necessary requirement for preserving that good.
Ii. Conditions imposed on the individual's pursuit of life-sustaining ac-
tivities (the rights to life, liberty, and property) imposed on him by the
existence of other living entities.
1. From [II] (that life is the good for each existent being) follows the
claim that life is a good for all beings. By proving that X has a right
to life from X’s nature and the “conditions of life,” one has proved
the same for all beings of the same nature subject to the same
conditions.
. Only beings who have the type of rational faculty that can enable
them to recognize [III.1] have a right to life, liberty and property,
because only they have a capacity to uphold the natural rights
framework.
. Men in close proximity are likely to interfere with one another,
therefore, mutual recognition of rights is necessitated.
. The fundamental condition of social life (i.e., of men living together)
is that individual rights be recognized and maintained.

It is apparent from this argument that men owe nothing to each

other from their nature as separate entities except the recognition of
the same rights to life, liberty, and property in others as they claim
for themselves as a result of their nature and the “conditions of life
and social life.” Thus, there are no “rights to:” to a good
education—to be provided by the labor of others; to a balanced
diet—to be financed by the sacrifice of others; etc. There is, quite
simply, the right to sustain your life by means which do not violate
the same rights of others.
Locke’s and Spencer's conflation of consent and natural rights
 The Time-Frame Theory 285

theory has been shown to be contradictory and the Nozickian

formation has proved irrelevant, a new theory of government
legitimacy—the time-frame theory—has been advanced, and an
attempt has been made to give an ontological defense of natural
rights based on the Spencerian argument.

NOTES

1. John Locke, The Second Treatise of Government (Indianpolis: The

Bobbs-Merrill Company, Inc., 1952).
2. Herbert Spencer, “The Great Political Superstitution,” The Man
Versus the State (Caldwell, Idaho: The Caxton Printers, Ltd., 1940).
3. John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard Univer-
sity Press, 1971).
4. James Buchanan and Gordon Tullock, The Calculus of Consent (Ann
Arbor: University of Michigan Press, 1962).
5. Robert Nozick, Anarchy, State and Utopia (New York: Basic Books,
1974), Part I.
6. Ibid, Part II.
7. Locke, Ch. XIX.
8. Ibid., Ch. II.
OP ibid. Cha itt:
10. Spencer, p. 195.
11. Ibid., pp. 196-97.
12. Ibid., p. 196.
13. Ibid.
14. Jeffrey Paul, “The Ontological Basic of Morality,” an unpublished
paper delivered at the 1976 conference of the American Association for the
Philosophic Study of Society at Bowling Green State University.
 14
How to Derive Libertarian
Rights
Eric MACK*

The basis for each person's possessing natural (i.e., non-contractual,

non-special) rights against all other persons is the moral truth that
for each person his living (his life) well is what is of ultimate value.
In the first section of this essay I will work toward a statement of
what I shall call “eudaemonistic egoism” and indicate how it pro-
vides the basis for natural human rights of the sort advocated by
Robert Nozick in Anarchy, State and Utopia. In the second section I
will state the argument for eudaemonistic egoism. And in the third
section I will defend this argument against likely objections.
The central notion of living well will be the main obscure notion
in this sketch. It should be unnecessary (but is not) to say that living
well involves interaction with and beneficence toward other per-
sons. It may even involve having, psychologically, a direct interest
in the well-being of others. To say that each person's ultimate value
is his living well is not to say that each person should be some sort of
a psychological egotist directly interested only in his own states. But
it is to say that nothing can be of positive value to a person if it does
not contribute to his living well, to his well-being.

286
 How to Derive Libertarian Rights 287

Reading in the history of philosophy I find a sustained, though

ill-defined, tendency to link the notion of each person’s life being his
ultimate value and the notion of individual (natural) human rights.
For Locke, persons should not be subordinated to others because
they are not made for one another's purposes. A secularized Locke
might say that each person’s life is that person’s purpose. Classical
natural rights theory, as in Locke, affirms the propriety of each
individual’s pursuit of his own self-preservation and self-interest.
Rights sanctify this pursuit by establishing a sphere for each person
within which this pursuit is (morally) immune from interference. In
more Kantian language, it is because each person has a dignity or
value in himself that each person should be treated as an end-in-
himself, not as an object at the disposal of other persons. In “Two
Concepts of Liberty,” Isaiah Berlin maintains that the view that the
individual must not be manipulated even for his own benefitis
founded on the view that “. . . there is no value higher than the
- individual.”!
Often the underlying importance for human rights of this moral
individualism is expressed obliquely in terms of the importance of
the fact that persons are separate beings one from another. For
instance, in Anarchy, State and Utopia? Robert Nozick cites, as the
notions and contentions that motivate the natural rights point of
view, the “principle that individuals are ends and not merely means;
they may not be sacrificed or used for the achieving of other ends
without their consent,” and the views that, “To use a person...
does not sufficiently respect and take account of the fact that he is a
separate person, that his is the only life he has,” and “that there are
different individuals with separate lives and so no one may be
sacrificed for others. . . .” And in characterizing the ideal of nega-
tive liberty, Berlin says,

To threaten a man with persecution unless he submits to a life in which he

exercises no choices of his goals; to block before him every door but one, no
matter how noble the prospect upon which it opens, or how benevolent the
motives of those who arrange this, is to sin against the truth that he is a
man, a being with a life of his own to live.?

But clearly it cannot be the fact of the separateness of persons or the

288 Eric Mack

fact that each has only his own life that underlies rights. For these,
alterable, facts might be seen as calling for a program for unifying all
persons into a great social organism and ridding the world of
bourgeois talk about individual rights. What underlies rights must,
at least in part, be the value, desirability, or rightness of persons
being separate beings, each having his own life and living it. One
could only sin against a person's being a being with a life of his own
to live if his having and living his own life is what ought to be.
It seems that the only thing which could constitute a ground for
each person’s having (natural) rights is the value or rightness of each
person’s separate individual life and well-being. Why will nothing
else serve? If the reason B has for not inflicting some treatment t on
A is the incongruity of that treatment with some goal of B’s or some
goal common to B and A, then the objection to t is its ineffective-
ness. But ineffectiveness of effort (or misdirection of resources) is
not the sort of objection made in claiming that B’s doing t would
violate A’s rights. Rights-correlative obligations are, to use Nozick’s
phrase, moral side-constraints. They are claims against action which
are independent of the effectiveness of those actions in generating
consequences that are either desirable for the agent or desirable for
some group which includes the agent and the subject of his actions.
Perhaps, alternatively, rights exist without being grounded in any-
thing. There simply are these deontic claims which are accessible to
the morally insightful. I take it, however, that any coherent account
of rights will be preferable to this non-account. Another possibility
is that rights rest essentially on formal considerations, e.g., persons
(perhaps inevitably) make certain claims for themselves and are,
thereby, committed by generalization to acknowledging like claims
on behalf of others. This sort of account, however, does not touch on
what is crucial, viz., the rationality of the initial claim. If I claim that
there is a dog on my roof, then I would be irrational not to
acknowledge also that there is a mammal on my roof. But to note
this is not to establish the rationality of my believing that there is a
mammal on my roof. Nor would the rationality of this belief be
established if we also knew that I could not but affirm that there was
a dog on my roof. We are left with the hypothesis that if there are
certain (deontic) moral side-constraints on how individuals may be
treated, these must flow from or be based upon the value of
rightness of each person’s separate lives and well-beings.
 How to Derive Libertarian Rights 289

Of course we have not yet said how its being the case that the
ultimate value for each person is his living well grounds rights.
What difference could this make to B in his moral relationship to A?
What is the connection, if any, between each person’s being an
end-in-himself in the sense that his well-being is the ultimate value
for him and each person’s being an end-in-himself in the sense ofnot
being a (morally permissible) means to others’ goals? What is the
connection, if any, between the rightness of rational self-interested
pursuits and the possession of rights? What is the connection, if any,
between there being no value higher than the individual and the
moral demand that individuals not be subordinated? Here is where
the speculation comes in.
The character of the claims of the right-holder is well-expressed
by means of the notions of treating a person as a means, using a
person, and (in contrast) treating a person as an end-in-himself. Yet
how can we explicate claims such as, “a person is not to be used,” so
as to arrive at something more interesting than, “a person ought not
to be treated in ways in which a person ought not to be treated?” We
can make sense of talk about misusing persons if for each person
there is some “objective” proper end, an end the satisfaction of
which constitutes the “objective” proper function of his life, ac-
tivities, capacities, faculties, etc. Only if there is such an end can
particular employments of a person, his life, activities, capacities,
faculties, etc. constitute objective misuses. Since the “objective”
proper end of a thing is that the attainment of which constitutes
successful use of that thing, the view that living well is success in life
and action (is successful use of one’s life) provides us with a claim
about the “objective” proper end for each person, viz., that the
“objective” proper end of each person’s life is his living well. The
proper function of his activities, capacities, faculties, etc. is to be
utilized in his living well.
Furthermore, if we are to be explicating rights claims, the end
specified as “objective” and proper for any person, say A, must not
be (numerically) identical with the end specified as “objective” and
proper for any other individual, say B. For if it were identical, then,
e.g., all failures of B to employ A, A’s activities, A’s capacities, etc.,
in accord with their specified end would be instrumental
failures—failures (on B’s part) to utilize A, A’s activities, etc. effec-
tively for B’s (proper) goal. If A’s “objective” proper end were
290 Eric Mack

identical with B’s, then the objections A could relevantly bring

against B’s utilizations of A, his activities, etc. would be objections
to the effectiveness of these utilizations. But the wrong done in
misusing a person, the wrong objected to in standing on one’s
rights, is not merely an instrumental error. On the view that for
each person his living well is his “objective” proper end, each
person’s proper end is (numerically) distinct. Thus, we can avoid the
implication that, e.g., when B fails to treat A in accord with A’s
“objective” proper end, when B misuses A, B merely commits an
instrumental error.
Still, just how does the claim that the “objective” proper function
of each person’s activities, etc., is to be utilized in that person’s
living well play a role in, e.g., certain treatments of A by B being
deontically unjustified? I construe well-being as an active, ongoing
process. Self-directedness or autonomy is an essential aspect of
living well. Unless a person’s activity is his own it will not contribute
to or be constitutive of his living well (or his living well). A person
whose activities are not his own does not live successfully. How can
the centrality of the autonomy of A in A’s living well be the ground
of A’s rights against others? If B personally values A’s autonomy and
respects it only on this basis, then B’s respect for it is merely
instrumental and not a matter of recognizing a moral side-constraint
on his actions. Similarly, if B disinterestedly commits himself to
maximizing autonomy in the world and his only basis for respecting
A’s autonomy is that in doing so he furthers this utilitarianism of
autonomy, then B is not recognizing a right of A’s. Rather, the
centrality of autonomy, as a property necessary to any activity’s
being constitutive ofliving well, allows us to be more specific about
the (proper) function of a person’s activity, capacities, etc. It is the
(proper) function of a person’s activities, capacities, etc. to be
employed by that person in (toward the end of) his living well. The
function of aperson’s activities, etc. is individualized not only with
regard to whose well-being it is the end of the activity (capacity,
etc.) to serve but also with regard to who must employ the activity
(capacity, ete.) for it to fulfill its function. The activity (capacities,
etc.) of A must be employed by A if it is to fulfill its function of
contributing to the active, ongoing, process of A’s living well. (And
A’s activities, capacities, ete. have no “higher” end.)
A person misuses his own life, capacities, ete. when he employs
them in ways that do not contribute to his active well-being. Here
 How to Derive Libertarian Rights 291

only certain specific uses will constitute misuses. But any use of A,
his capacities, ete. by another (which use prevents A from using
himself, his capacities, etc.) is a misuse. While to use oneself is not
as such to misuse oneself, to use another is to misuse him. Any
action which prevents a person from utilizing his life, capacities,
etc. misuses that person. At least when done intentionally, such
prevention treats a person as a means, as a resource at another
party's disposal, as an entity without a proper end of its own.
The moral propriety of B’s employment of his activities, capacities
and so on is to be determined by whether these utilizations accord
with the function of these activities, capacities and so on. Similarly,
the moral propriety of B’s employment of A’s activities, capacities,
and so on is to be determined by whether these utilizations accord
with the function of these activities, capacities and so on. So the
same proposition which a person must invoke (and which it is
rational for him to invoke) to justify his pursuit of his own well-being
in contrast, say, to his pursuit of the greatest good for the greatest
number, viz., that the function of each person’s activity, capacities,
and so on is to be employed by that person in his living well, shows
that a person would be unjustified in bringing it about that another's
activity is not directed by that other person. Two types of
consideration—both based on what I have called eudaemonistic
egoism—can play a role in determining the justification of B’s action.
There is the consideration of whether B employs his actions in accord
with their proper end and there is the consideration of whether B
prevents A’s activity, capacities, and so on from being employed in
accord with their proper end. If B’s action is unjustified on the basis
of the latter consideration, it is unjustified on the basis of the charac-
ter of his treatment of A and not on the basis of the effective-
ness with which his (proper) goals are pursued. Such action would be
deontically unjustified and A has a claim against such action in virtue
of his being a moral end-in-himself. A similar claim can be made on
behalf of (at least) each human person. And for each of them this
claim constitutes a natural (non-special) human right.

II

But can the eudaemonistic egoism which underlies natural rights

and obligations itself be made plausible? Providing an underlying
rationale for this ethical egoism involves nothing less than the
292 Eric Mack

derivation of a fundamental normative claim from non-normative

premises. What follows in this section is the outline of such an
argument.
Prop. I. If there is some need or requirement which explains
(plays a role in explaining) the existence of some thing
(object, activity, process, etc.), then that thing func-
tions well if and only if its use or enactment satisfies
the need or requirement which explains the existence
of that thing.
Glossuls The nature of the function of some thing can be
determined by the requirement which accounts for
the existence of that thing. The requirement accounts
for the existence of some thing when the existence of
that thing is necessary to the satisfaction of the re-
quirement. Tools are, for instance, required for cer-
tain human goals, and their value as this or that
particular type of tool is determined by how well their
use satisfies the respective goals for which they are
designed. A heart beats well if it satisfies the re-
quirement, namely the need of the cells for nutrients
and oxygen, which necessitates (and thereby explains)
the existence of the heart.
Prop. II. With respect to each living thing, it is the fact that
remaining in existence as a living thing (not merely as
a collection of dead cells) requires the successful
completion of numerous processes that explains the
existence of valuation.
Gloss II. Valuation is the process of pursuing and maintaining
goals. This is a process that exists within a living entity
just in virtue of its being a living thing, i.e., just in
virtue of it facing alternatives, being faced with alter-
natives. That there are objects of value for any given
living entity is necessitated (not merely made possi-
ble) by the fact of that entity’s living existence. The
requirement which explains valuation is the require-
ment of acting successfully in order to remain a living
thing.
Theorem I. Valuation functions well if and only if its use or
enactment satisfies the requirement of the valuing
 How to Derive Libertarian Rights 293

organism to complete processes successfully if it is to

remain a living thing. (from Prop. I and II)
2h Valuation is a process that is carried on by living
things; it is goal-directed action. (from Prop. II, cf.
Gloss I)
Te Sih, The functioning of valuation is the performance of
goal-directed actions. (from Th. II)
Th. IV. Goal-directed actions are performed well if they
satisfy the requirement for, the need of, acting suc-
cessfully in order to remain a living thing. The stan-
dard for goal-directed actions is the satisfaction of this
need. (from Th. I and III)
Th. V. The satisfaction of this need is good with respect to
the acting organism. (from Th. IV) That is, the result
of valuation functioning well with respect to any living
entity is simply that which is good for that entity. This
is obvious by the very meaning of the concept valua-
tionfunctioning well.
Th. VI. Performing successfully the actions that sustain its life
is that which is good with respect to any given or-
ganism. (from Th. IV and V)
Prop. III. If astandard for goal-directed action is complied with
as a result of choices made by the acting entity, then
the (normally) resulting good is a moral good, and the
actions (specifically the choices) of the agent are
morally good.
Gloss III. Under these conditions the agent is responsible for
his actions and their (normal) consequences. Simi-
larly, if the standard is not complied with, and this
non-compliance is the result of choices, then the
(normally) resulting bad is moral evil and the actions
of the agent are morally deficient.
Prop. IV. Whether or not this standard (or any alleged standard)
for goal-directed actions is complied with is a matter
of choice for human beings.
Th. VII. For human beings, that which is good in the sense of
theorem IV is morally good. That which is bad in the
sense of theorem IV is morally deficient. (from Prop.
III and IV)
294 Eric Mack

Th. VIII. The morally good, with respect to each human being,
is the successful performance, and the results of the
successful performance, of those actions that sustain
his existence as a living thing. (from Th. VI and VII)

Iil

Obviously the key premises in this argument are Proposition I and

Proposition II. Now Proposition I itself follows from two prior
propositions. The first is: Anything which in functioning satisfies its
(or its users) goal is functioning well. This is surely acceptable
simply on the basis of the meanings of the terms. Note that we have
not yet specified how the function of a thing is determined. This is
done by the second premise, namely: The function of a thing is its
use or enactment in regard to the satisfaction of the requirement
which explains the existence of that thing, if there is such a
requirement. This premise has been briefly explained and illus-
trated in Gloss I.
This is the crucial claim in two senses. First, it defines function in
such a way that there can be functions which are objective features
of things, features that cannot be changed merely by some alteration
in persons’ attitudes or intentions. The existence of such a function
is independent of persons’ actions, hopes, wants, etc. This indepen-
dence is what allows us to make claims about things satisfying their
objective functions well and, thus, being objectively good things. In
effect, this second premise for Proposition I introduces something
like a doctrine of natural ends. This position might be expressed as
the view that the natural end of a man’s valuation is the sustenance
of his life. But “natural end” here does not mean “actual end” or
“likely end.” Rather, “natural end of aman’s valuation” means “the
end of the objective function of a man’s valuation.” In summary
form, the view is that the end of the objective function of a man’s
valuation is the sustenance of his life.
The second sense in which the definition of function is crucial is
that it is just at this point that an apparently strong objection to the
argument as a whole can be raised. There are really two related
objections that are possible here. Both are based upon the notion
that functions of things are determined by the use to which they are
put. In what I shall call the moderate version of the objection it is
 How to Derive Libertarian Rights 295

not explicitly denied that a thing may have an objective function in

the sense of Proposition I, but it is asserted that things do have
functions in virtue of the use to which they are put. At least some of
the functions of athing are determined by the present goal(s) of the
user(s) of it. For instance, someone might use a heart for the sake of
keeping time. To the extent that this goal is satisfied the heart would
be functioning well. For this heart, the keeping of time becomes a
standard of value. Similarly, it might be argued that valuation can be
used to satisfy goals other than the goal or need which explains the
existence of valuation. For instance, someone might employ valua-
tion for the sake of slowly and painfully killing himself. Could one
not then say that for this peculiar individual a function of valuation
was the production of slow and painful death, and thus acts of
valuation that satisfy this function are good acts? It seems that one
could. Thus, for this individual, good acts include acts that bring
painful death, not the sustenance of life. Hence any doctrine which
asserts that for each man the good acts are those that sustain his. life
must be false. In short, this objection states that if one accepts both
the claim that anything which in functioning satisfies its (or its
users) goal is functioning well and the claim that things do have
functions in virtue of the use to which they are put, then one cannot
also, consequently, accept ethical egoism. This statement is mista-
ken. The argument in support of it, the argument of the advocate of
the moderate objection, fails to distinguish different senses in which
things may have functions and thereby be good or bad things.
Certainly a heart may be used as a time-keeper, as a musical
instrument, and or as an example of a certain type of muscle tissue.
On the basis ofits time-keeping, its use as a musical instrument, ete.
we may well make judgments about the heart. Yet in each of these
cases we judge how well the heart functions as a time-keeper, as a
musical instrument, as an example. There is no inconsistency in
asserting that the heart is good as an example of a type of muscle
tissue but bad as a musical instrument. In general, in virtue of uses
any X may have a function as a Y. The X may be judged good or bad
as a Y or as an X-functioning-as-a-Y. The X may also have a function
as a Z and be judged good or bad as a Z or as a X-functioning-as-a-Z.
Yet there is no ground for inconsistency here. Similarly, X may have
a function as a Y, be judged good or bad as a Y or as an X-
functioning-as-a-Y, yet this entails nothing about how good or bad X
296 Eric Mack

(or X-as-an-X) is. The issue of how good a YX is, or how good an
X-functioning-as-a-YX is, is completely unrelated to the issue of how
good X is. To return to the familiar example, whether a heart is good
or bad as a clock, or whether a heart is good or bad as a heart-
functioning-as-a-clock, is completely unrelated to whether the heart
is a good heart. The fact that a heart can have functions other than its
objective function qua heart, functions in addition to the function it
is determined to have on the basis of what need explains its
existence, in no way interfers with our judging the value of the heart
qua heart solely in terms of this objective function.
Insofar as we judge the heart qua heart, not the heart merely as a
time-keeper or as a musical instrument, we judge it solely on the
basis of how well it satisfies the requirement which explains its
existence. That is, we judge it qua heart solely on the basis of how
well it supplies the cells with nutrients and oxygen. Similarly, it may
be that valuation may function well as a means to a slow and painful
death. But this does not mean that valuation qua valuation is good
insofar as it produces (for the agent) slow and painful death. Con-
sidering the valuation in question qua its suicide producing role, it
may be judged good—but merely in the sense that it is effective in
its current role as a suicide producing process. We may judge the
valuation-as-suicide-producing-process good, but this is completely
unrelated to any judgment about the individual’s valuation itself.
The judgment on the suicide producing process is completely
unrelated to any judgment about the individual’s valuation as valua-
tion. The judgment of what is good valuation for any given indi-
vidual is made on the basis of the objective function of valuation.
That is, the judgment is made on the basis of how well the valuation
satisfies the need which explains its existence. Thus, it is perfectly
consistent to assert that only valuation that sustains life is good
valuation and that some acts of valuation are to be judged good as
means toward slow and painful death, if the agent seeks this end,
and by his actions, attains it. There is no inconsistency between
ethical egoism and the fact that some of the functions of things can
be determined by the use to which those things are put.
With the moderate objection out of the way, we are now in a
position to consider what I shall call the extreme objection. In this
objection to Proposition I, it is maintained that the functions of
things are determined solely by their actual uses or goals. In other
 How to Derive Libertarian Rights 297

words, there are no objective functions. If there is no objective

function of valuation, then all valuation is simply valuation for the
sake of so-and-so current ends. Acts can be evaluated only in terms
of how well they satisfy the actual aims of the acting individual
(individual relativism), or how well they produce the results which
are endorsed by some type of consensus (some type of cultural
relativism). In either case there are no objective standards, only
subjective and relative ones.
The first thing to notice is that the view that functions are
determined by present use cannot be strictly maintained. A man
does not use his heart well or poorly in order to supply nutrients and
oxygen to the cells of his body. Does this mean that (assuming no
time-keeping use, etc.) the heart has no function? Of course not.
Any goal-directed process, or object which carries out such a
process, has some function. The critic, then, must formulate his
objection in this way, “Certainly the function of hearts is to supply
nutrients and oxygen, even though the heart is not used for this end.
But the heart has this function solely because this is the current,
actual, goal of its activity, not because this goal explains its exis-
tence. This implies that the heart might just as well have some other
goal, anda fortiori, some other function. Thus, there is no objective
function of hearts, and similarly for all other things.”
The reply to this is that a heart could not just as well act toward
some other goal—though it is certainly possible for it to act toward
some other goal. That is, in replying to the extreme objection we
must distinguish between a goal in the sense of that which will
comply with some standard and a goal in the sense of an actual
current aim. Now, of course, the advocate of the extreme objection
denies the validity of this distinction. This is just the point at issue.
And so our criticism of the extreme objection takes, for the moment,
the form of emphasizing the plausibility of this distinction and, at
the same time, elaborating upon the rationale for Proposition I.
Certain objects and activities, by their very nature, do seem to
have, respectively, certain objective functions—or so to speak
natural ends. Consider again the heart example. Hearts do not
acquire the goal (and thereby the function) of supplying nutrients
and oxygen to the cells. They simply have this function, and it is part
of our understanding of the notion heart that we recognize this
function. However to say this is not equivalent to saying that the
298 Eric Mack

satisfaction of this function is that toward which (or even among

those things toward which) the heart is always directed. The posses-
sion of a certain objective function by some thing is perfectly
consistent with the actual current goal of that thing being something
other than the fulfillment of that function. This is a little difficult to
see in the heart example. For we may be inclined to say that
something is not an actual heart (though it may be a dissection
specimen) unless it be directed toward supplying nutrients and
oxygen. In other words, it seems that all cases of bad hearts (not
fulfilling their function) would be cases where the current goal is the
supplying of nutrients and oxygen but the heart does not effectively
pursue this end. Thus, it might seem that we can, at most, talk
about the objective function of an X when being directed toward the
fulfillment of that function is a necessary condition for being an X. If
this were so, then all replies to the extreme objection would be
irrelevant to the defense of ethical egoism. For the cases of valuation
that the egoist wishes to morally condemn are essentially those
where the valuation is (according to the egoist) misdirected, not
merely ineffective. For the egoist’s argument to work it must be
possible that a thing function poorly not merely by being ineffective,
but also by not being directed toward the fulfillment ofits objective
function. Fortunately, even a heart can malfunction in the sense of
being misdirected.
Consider the case of a Yogi who is able to will the cessation of the
supplying of nutrients and oxygen to his body’s cells. Certainly we
ought not to say that to the extent that he so wills he ceases to have a
heart. Rather, during these moments, qua heart, his heart is not
functioning well (though as a device to help induce certain peculiar
mental states it may be functioning well). Note that it is the
extension of choice to the determination of the current goal of the
heart that makes it possible for the heart to function poorly in the
sense of being misdirected. Now let us consider an activity the
current goals of which are normally determinable by choice. It is
typically within the power of medical practitioners to choose what
will be the actual current goals of their medical activities. Hence we
ought to expect that there will be some variation in their respective
goals. But this does not show that the function of medical acts varies
from act to act. These acts gua medical acts do not acquire func-
tions. Rather, in virtue of being medical acts, they have a specific
 How to Derive Libertarian Rights 299

function—namely, producing and/or sustaining health. Medical

activity has this function not in the sense of its fulfillment being the
actual goal of that activity, but rather in the sense that the standard
for evaluating this activity as medical activity is the production
and/or sustenance of health. Certain acts are simply medical acts
and as such there exists a standard for judging them. This standard is
not affected by how many medical maniacs there are. That is, it is
not affected by what percentage of the white-robed, medical
school-graduated, instrument-bearing and/or drug-prescribing per-
sons are guided by diabolical schemes leading to the destruction of
health. It is possible for every medical practitioner to be practicing
bad medicine not simply in the sense of unsuccessfully aiming to
satisfy health needs, but in the sense of not having that aim at all.
If the standard for, and the function of, medical practice are not
determined by the actual goals of medical practitioners, then by
what are they determined? There must be some need the satisfac-
tion of which is inseparable from medical activity in the sense that
an understanding of medical activity involves an understanding that
good medical activity must satisfy this need—though the actual
satisfaction of this need may not be what medical activity is actually
seeking. A comparable need must exist with respect to hearts, and
with respect to any thing which can function poorly by being
misdirected. For both hearts and medical activity there is some
need or requirement for the satisfaction of which the respective
existence of hearts and medical activity is necessary. There is the
need that cells be supplied with nutrients and oxygen, and there is
the cluster of requirements for the production and sustenance of
health. These needs give rise to hearts and medical activity. One
could not understand either of these phenomena without reference
to their generating needs. These needs explain (necessarily play a
role in explaining) the means (hearts and medical activity) to the
respective satisfaction of these needs. And in both cases we see that,
being necessitated by a certain need, the satisfaction of that need is
the function of, and sets the standard for, evaluating that object or
activity. This correlation of necessitating need on the one hand, and
function and standard on the other hand, is just the way the world
is. Proposition II tells us that we can apply this truth to the
phenomenon of valuation.
It may be best, at this point, to take note of certain problems that
300 Eric Mack

will be encountered by the advocate of the extreme objection to

Proposition I. No doubt the advocate of this objection will want to
distinguish between the function of X and the function of X-as-a-Y.
For he certainly wants to be able to speak of hearts being good
whether or not they satisfy their simultaneous use as musical
instruments, clocks, etc. Yet he cannot draw this distinction on the
basis of the objective function versus merely supplementary use.
The advocate of the extreme objection must insist that the correct
interpretation of the phrase, “the function of X” (or “the function of
X as an X) is “the typical purpose or role of Xs.”> If there is no
statistically normal role (as in the case of valuation), then no sense at
all can be given to the phrase “objective function.” But from this
insistence, the advocate of the extreme objection is lead to one of
two equally peculiar positions. If he accepts (as he is unlikely to do)
the view that having the normal role of an X is a necessary condition
of being an X, then he must accept that there is no such thing as an
X functioning as a Y where functioning as a Y is inconsistent with
having the normal role. That is, he would be committed to the view
that the Yogi's heart cannot be a heart—at least not during the times
that it assumes its unusual role. The advocate of the extreme
objection may even be committed to the even odder view that if all
but one heart takes the Yogi role, then the remaining entity ceases
to be a heart. However, what the advocate of this objection is more
likely to say is that having the typical role of an X is not a necessary
condition of being an X. But satisfying the typical role is a condition
for being a good X. Xs that do not play the typical role will be poor
Xs. But suppose that the typical role of a certain type of thing
changes. For instance, suppose that medical activity comes to be
typically aimed at instilling disease. At the point that this goal
becomes the typical goal medical activity that does instill disease
becomes good medical activity and medical activity that prevents
disease becomes poor medical activity. If there is a return to earlier
aims, evaluations must again be reversed. Two instances of medical
practice may be qualitatively identical and differ only in date of
occurrence, yet one could be good and the other bad. Or it might be
suggested that “typical” not be interpreted as “typical at this time,”
but rather as “typical of all acts of this type—past, present, and
future.” However then one would have to say, “It seems to me that
this health-saving medical act was a good medical act, but I cannot
 How to Derive Libertarian Rights 301

be sure for I do not know how medical practitioners will act in the
future.” Of course none of these consequences will seem terribly
troublesome to someone who holds to medical relativism. But who
could hold to such a view? For medical relativism is not the view
that the moral value of health is a non-objective value, it is the view
that health is not an objective medical value. Medical relativism
holds that medical activity that produces health is not objectively
better as medical activity than medical activity that fails to promote
health. This view, I submit, is simply false to the facts. Thus, we can
see that, through his insistence that “the function of X” means
merely “the typical role of Xs,” the advocate of the extreme objec-
tion against Proposition I is led to one or another of two untenable
positions. Either hearts disappear as Yogis increase or health-
destroying medical activities become medically good as medical
maniacs increase.
I have considered what I take to be the two fundamental objec-
tions that might be raised against Proposition I and have found them
wanting. In the process I have attempted to display the plausibility
of this premise of the egoist’s argument. Given Proposition I we can
formulate the argument for ethical egoism as I have done in Part I of
this discussion. We have there a formulation in which the normative
conclusion follows quite strictly and soundly from the non-
normative premises. Furthermore the resulting eudaemonistic
egoism is expressive of the value and importance of individuals and
their separate lives—a value and importance which underlies indi-
vidual human rights.

NOTES

*This essay partially reproduces material from two previously published

essays “Egoism and Rights Revisited,” The Personalist, July 1977 and “How
to Derive Ethical Egoism,” The Personalist, Autumn 1971.
1. Isiah Berlin, Four Essays on Liberty (London: Oxford University
Press, 1969), p. 137.
2. Robert Nozick, Anarchy, State and Utopia (New York: Basic Books,
1974), p. 30f.
3. Berlin, p. 127. Emphasis added.
4. To fail to provide a person with an opportunity to sos is not to prevent
him from doing s even if he will not otherwise have an opportunity to do s.
5. If the advocate of the extreme objection interpreted “the function of
302 Eric Mack

X” in terms of what was typical for this or that individual rather than what
was typical throughout a group of individuals fundamentally the same
objections would apply, in addition, this interpretation would mean that
one could not say of some X, “It is functioning well as X” until one had
extensively investigated its particular history.
 IV
Social Justice:
Entitlement Theory
Versus Distributivism
 a
Maal , 5
TS wnt 9 nae te
i Wreks nh a els “Ws gre 49
vent ‘ ratty bo =
Sacee

iat “patent Inin0e —

pth (food ft eons in| FE
snetontucientaidl eens
—~
7 -

bg = aa

ae
y= i
~. Ne tine 7
>

_-& —
: =

=
—

>

-
 15
Nozick s Entitlements
ONORA O'NEILL

This essay examines Nozick’s claim (in Anarchy, State and

Utopia)
to have shown that a commitment to individual liberties
requires
acceptance of full capitalist property rights. The main gap in
Nozick’s
argument is that he fails to show how individuals can become
entitled
to full control over previously unheld resources. Nozick draws on
Locke’s view that title is acquired by ‘mixing one’s labour’. But he
excises certain (dubious) premisses on which Locke’s theory relies
and provides no alternative grounds for thinking that the labourer is
entitled to full control over his product.

INTRODUCTION!
In Anarchy, State and Utopia? Robert Nozick revives, clarifies and
in many ways strengthens Locke’s defence of individual property
rights. Like Locke he eschews utilitarian arguments in favour of
individual property rights and starts from the position that individu-
als have certain basic (natural) rights and duties, in particular the
rights not to be harmed in life, health, and liberty. If his argument is
successful, Nozick will have shown that it follows from these basic
rights that individuals may justly acquire complete control over
unlimited resources without needing the consent of all whose
liberty will be restricted by these property rights. He will also have
shown that other proposed principles of distributive justice cannot

305
306 Onora O'Neill

be grounded on a commitment to individual rights. If the Lockean

argument can be made convincing we face a clear dilemma: either
we cannot be fully committed to individual (basic) rights, or we must
accept that it is unjust to limit or in any way to regulate the holdings
of individuals. The dilemma is not trivial. Justice, in Nozick’s view,
precludes any restriction of individual property rights, and so is not
merely compatible with but requires a capitalist economic order.
I shall accept Nozick’s starting point here, but query his conclu-
sions. I start therefore by assuming that individuals have at least the
rights not to be harmed in life, health, or liberty, but that their
property rights—rights to control resources—have yet to be estab-
lished. I also accept that Nozick has shown that a minimal state can
(though undoubtedly it did not) arise without violation of individual
basic rights, but that extensions of state activity beyond enforcing
the law against harms and threatened harms to life, health and
liberty have not been established.

THEORIES OF DISTRIBUTIVE JUSTICE

Nozick starts his argument in favour of a Lockean principle of

distributive justice by considering rival theories and their defects.
He classifies theories of distributive justice so that whole ranges of
theories can be levelled with one objection. Theories of distributive
justice can be divided in two ways. They can be patterned or
unpatterned, and they can be historical or unhistorical.
Patterned theories of distributive justice require that resources
should be distributed “along with some natural dimension, weighted
sum of natural dimensions or lexicographic ordering of natural
dimensions’ (p. 156). When a distribution is patterned there is some
dimension such that each individual's ranking on that dimension
corresponds to the ranking of his share of resources. Patterned
principles of distribution are historical if they distribute on a dimen-
sion which summarizes past actions (e.g. moral merit or hours
worked); they are unhistorical if they distribute on a dimension
which does not refer to past actions (e.g. I.Q. or race). The most
celebrated patterned principles of distributive justice include ‘to
each according to his merit’, ‘to each according to his need’, and ‘to
each according to his contribution’.
Unpatterned principles ofdistributive justice do not assign shares
 Nozick’s Entitlements 307

of resources in proportion to individuals’ ranking on some dimen-

sion. Such principles may also be either historical or unhistorical.
Principles of distribution which are neither historical nor patterned
specify the structure or profile of just distributions, but do not assign
particular individuals to particular positions within these distribu-
tions. Nozick calls such principles ‘current time-slice principles’ or
‘end-result principles’ (p. 154). End-result principles may vary
greatly in strength. A principle which states that no distribution is
just in which the average income of the top quintile is more than
four times the average of the bottom quintile is exemplified by an
enormous range of distributions. A utilitarian principle which holds
only those distributions just which maximize utility can be exem-
plified by more than one distribution. A strictly egalitarian principle
requiring the same amount for each person is met only by distribu-
tions in which the size of each individual’s share is determined, and
so is not only an end-result principle but incidentally patterned. It
determines not only the profile of distributions but the size of
individual shares for’any given total of resources, so is extensionally
equivalent to a patterned principle requiring distribution in propor-
tion to individuals’ (equal) humanity.
This continuity is revealing. End-result principles and patterned
principles differ in that the former usually do not determine the size
of individuals’ shares, and the latter do. But their similarity is more
important. Both are principles for determining the (relative) size of
shares. Neither sort of principle is concerned with the composition
of individuals’ shares, but only with their size. Neither sort of
principle can regard confiscation-with-replacement as unjust since
they determine only how much people should have, and not which
things they should have.
Nozick’s theory of just distribution, the entitlement theory, is
neither an end-result nor a patterned theory. It is unpatterned and
historical. It specifies just distribution not by giving a rule for the
size of individual shares, nor by stating a mandatory profile or range
of profiles for the pattern made by the sizes of individual shares, but
by listing procedures by which individuals may justly acquire title to
particular resources. Hence the name, ‘entitlement theory’. Ac-
cording to entitlement theory the size of just holdings is always
determinate, but this does not determine the holdings’ justice,
which depends solely on the methods by which an individual
308 Onora O'Neill

acquires holdings. Nozick therefore calls entitlement theory an

historical principle of justice, in that it too holds that ‘past cir-
cumstances or actions of people can create differential entitlements
(p. 155). But it is an historical principle of justice in a very different
sense from that in which certain patterned principles are historical.
For it requires not that the size of shares be proportioned to past
action, but that particular items be acquired by particular proce-
dures. According to entitlement theory ‘an injustice can be worked
by moving from one distribution to another structurally identical
one (p. 155). This statement is too weak. Entitlement can also be
disregarded by changing the composition of individual holdings
without altering their size. What is distinctive about entitlement
theory is that it connects individuals with particular holdings ac-
quired by particular procedures.

NOZICK §ARGUMENT AGAINST REDISTRIBUTIVE

THEORIES

Nozick advances a single argument to counter end-result and pat-

terned principles of distributive justice. The argument leans heavily
on one attractive and distracting example. Suppose some distribu-
tion of holdings which is just either according to a pattern or
according to an end-result theory of distribution. Then transfers are
made, for example, premium payments to Wilt Chamberlain by
thousands of basketball fans who want to watch him play. The result
is a new distribution, distinct from the first, which may not conform
to the pattern or end-result in terms of which the first distribution
was justified. More generally,

any favored pattern would be transformed into one unfavored by the

principle, by people choosing to act in various ways ... to maintain a
pattern one must continually interfere to stop people from transferring
resources as they wish to. (p. 163)

So also to maintain a desired end-result. Yet intuitively there seems

to be nothing very wrong about premium payments to Wilt Cham-
berlain.
But this will hardly do as an argument against patterned and
end-state principles ofjustice or for entitlement theory. The argu-
 Nozick’s Entitlements 309

ment presupposes, so does not demonstrate, that it is wrong to

interfere to restore disturbed patterns or end-states, and that such
restorations are always redistributive and violate individuals’ prop-
erty rights. But it is just these property rights which have yet to be
established. All the theories of distributive justice Nozick classifies
assign individuals some sorts of property rights—some claim to
exercise some control over some resources. They differ in the extent
of these rights. For example, theories which hold that pattern-
restoring interference by the state maintains rather than violates the
rights of individuals regard property rights as limited in certain
ways—perhaps by an upper limit on individual holdings, or on what
individuals may do with what they hold, or on how or to whom they
may transfer it. Nozick comments at one point that we lack a theory
of property (p. 171). We do indeed, but the lack cannot warrant the
assumption (cf. pp. 281-2) that individual property rights are rights
to control resources in all ways, to dispose of them however and to
whomever the owner wishes, or to accumulate them without limit.
This interpretation of property rights must be established before the
restoration of patterns or end-states by state action can be rejected
as unjustified interference which violates individuals’ rights. Nozick
is quite right when he points out that ‘socialist society would have to
forbid capitalist acts between consenting adults’ (p. 163). So it
would; but this restriction of liberty might be compatible with or
required by justice if the property rights of individuals are less than
full capitalist property rights, i.e. are not rights to control any
amount of material resources in all ways, subject only to the
condition of not violating others’ basic rights not to be harmed in
life, health, or liberty.
Until Nozick provides an argument to show that property rights
must be full capitalist property rights the Wilt Chamberlain exam-
ple shows nothing. We can infer little from an intuition that there is
nothing very wrong about making premium payments to Chamber-
lain, particularly when other intuitions suggest nothing very wrong
in a bit of pattern-restoring redistribution by the method of taxing
Chamberlain’s now enlarged earnings. And even if Nozick had
shown such restoration of patterns unjust, he would not have shown
that it must involve unpredictable state depredations of individuals’
holdings. It can be done by familiar devices such as income tax and
welfare payments, and can allow many (though not all) capitalist acts
310 Onora O'Neill

to continue untrammelled. Nozick therefore exaggerates when he

suggests that restoration requires ‘continuous interference with
individuals’ actions and choices’ (p. 166). The restoration of patterns
(whether or not it is just) falls short of continuous interference.

ENTITLEMENT THEORY

If Nozick’s refutation of patterned and end-result theories ofjustice

is inconclusive, then entitlement theory must be judged on the
arguments presented for it without the support of finding itself
the only contender in a vanquished field. Nozick presents an
inductive definition of entitlement theory: (p. 151)

(1) A person who acquires a holding in accordance with the principle of

justice in acquisition is entitled to that holding.
(2) A person who acquires a holding in accordance with the principle of
justice in transfer, from someone entitled to the holding, is entitled to
the holding.
(3) No one is entitled to a holding except by (repeated) applications of (1)
and (2).

As it stands this definition is extremely general. It can be applied

only when the two further principles mentioned—the principles of
just acquisition and just transfer—are specified. To these a third, a
principle of rectification, would have to be added if we were to know
what should be done about breaches of the first two principles. If we
are to assess entitlement theory then we should look for precise
formulations of at least the first two principles and arguments in
favour of these particular formulations. Amazingly, Nozick writes ‘I
shall not attempt that task here’ (p. 153). However, there is more to
be said, since Nozick relents and sketches a principle of just
acquisition.
I shall now consider this sketch and the possibilities of making it
more detailed. I fail to give the same consideration to the principle
of just transfer not because it is unproblematic,4 but because it is
less fundamental. If there are no entitlements no problem arises
about transferring them. If Nozick is to justify a principle of just
acquisition which will show patterned and end-result principles of
just distribution untenable, he must both show that property rights
 Nozick’s Entitlements Odd

must be full capitalist property rights and explain how particular

individuals acquire such title to particular resources.

JUST ACQUISITION
Just transfers lead to just acquisitions only when the previous
holdings were just. Without a principle of just acquisition entitle-
ment theory can view holdings only as conditionally just. Hence the
importance of the principle of just acquisition. Nozick’s principle is
Lockean: title is acquired by mixing labour with unowned objects,
provided that ‘enough and as good is left for others. This principle
has a long history, and seems to have the advantage of a certain
nonarbitrariness over other principles of just acquisition. Although
it would be difficult to give reasons for thinking that the holdings ofa
specific date—say 1900, or, if you like, 1500—were just, it is worth
considering what would be lost by formulating such a principle of
justice in acquisition:

(1) A person who has held a holding since 1900 is entitled to that holding.

The arbitrariness of the date is perhaps compensated for by the

ascertainability of holdings at that date. Nozick even provides an
argument for thinking that such arbitrariness might be laundered
out over the years:

_ it is held that the operation of the system over time washes out any
significant effects from the initial set of holdings. As an example of the
latter, if almost anyone would have bought a car from Henry Ford, the
supposition that it was an arbitrary matter who held the money then (and so
bought) would not place Henry Ford’s earnings under a cloud. (p. 158)

But this is too quick. IfA and others would buy from C, but only A
has the resources to do so, andA has no just title to these resources,
then when A buys an x, A’s holding of x is still not just (it was
acquired with ill-gotten funds) and equally C’s holding of A’s
payment is just only if C had title to the x he sold. Henry Ford's
holdings after sales will be tainted if his holdings of the means of
production of Ford motor cars was tainted; his customers’ holding of
their cars will be tainted if the purchase money was tainted. Lack of
ANY Onora O'Neill

entitlement is not generally transferred in sales, but transfers do

nothing to reduce the total of unjust holdings. Entitlement theory
cannot avoid the liabilities of being an historical theory, and Nozick
has good reason, despite his misplaced faith in laundry, to look for a
non-arbitrary principle of acquisition.®
He claims to find this principle sketched in Locke's chapter on
Property in the Second Treatise. Yet his attitude to Lockean scrip-
ture is ambivalent. The questions he raises about Locke's theory
(pp. 174-5) are a hilarious—and penetrating—piece of philosophical
writing. But the fun Nozick gets from exploring Locke’s problems is
merely fun; it is not demolition. For if entitlement theory is to work,
some principle of just acquisition must be provided. There must be
some way by which persons can come to hold previously unheld
things, and this method must yield full capitalist property rights if
Nozick is to sustain his claims against principles of distributive
justice whose maintenance needs state intervention. If Nozick is
right there must be something which a person can do to a thing
which will make it his own to use, neglect, alienate, or destroy as he
chooses.
In arguing towards these conditions, Nozick does two things to
Locke's principle of just acquisition: he streamlines it and extends it.
I shall argue that his extention is plausible and powerful, but that
the streamlining unfortunately smoothes away an indispensable part
of Locke's theory, the part which tries to explain why the mixer of
labour is entitled to the product. If this part is restored property
rights can be explained, at the cost of some extra assumptions, but
the property rights so established are not full capitalist property
rights.

LOCKE AND NOZICK ON MIXING LABOUR

Locke and Nozick both see it as their task to ‘show how men might
come to have a property . . . without any express compact’ (§ 24).§
They will do this by showing that private property rights do not
violate others’ rights, who therefore need not be consulted. If their
argument succeeds they will show that private property rights are
part of the (basic) right of liberty and so not merely compatible with
but required by justice.
Certain sorts of acquisition of property clearly violate others’ basic
 Nozick’s Entitlements Bs

rights. For example, slavery is incompatible with the right to

liberty. But according to Locke and Nozick control over things need
not violate others’ liberty if there remains ‘enough and as good left
in common for others’ (§ 26). IfA appropriates some natural object,
such as acorns, grass or water (Locke) or grains of sand (Nozick)
(§§ 27-29: p. 175), but there remains enough unappropriated for
others to do anything they might have done before A’s appropria-
tion, then the appropriator has not violated any rights. Each person
has as large a liberty as before, “for he that leaves as much as another
can make use of does as good as take nothing at all’ (§ 31).
The problem is to be sure that the ‘enough and as good’ proviso is
not violated. On this point Nozick and Locke diverge. Locke is
worried:

Nobody could think himself injured by the drinking of another man, though
he took a good draught, who had a whole river of the same water left him to
quench his thirst. And the case of land and water, where there is enough of
both, is perfectly the same.({§ 32)

This suggests that there may be many places where there is not
enough and as good remaining once a certain amount of appropria-
tion has happened. Locke points backwards to underpopulated days
when:

it was impossible for any man, this way, to entrench upon the right of
another or to acquire to himself a property to the prejudice of his
neighbour, who would still have room for as good and as large a possession
(after the other had taken out his) as before it was appropriated. (§ 35)

He points to ‘the inland vacant places of America’ (§ 36) and other

waste lands (§ 45), though he admits that unappropriated land will
vanish with the use of money (§ 47).
Nozick overcomes the worry by pointing out that it is possible to
compensate others for their lost liberty. If appropriation and enclo-
sure leave no further land or resources to appropriate, the proper-
tyless have lost the right to use the resources appropriated but this
will be justifiable if their overall position is not worsened (pp.
175-9), if the opportunities gained through others’ appropriations at
least compensate for the liberties lost. There is no need to point out
vacant places or unappropriated resources to show present holdings
314 Onora O'Neill

justifiable, no need indeed to use acorns and grains of sand as

examples: one has only to demonstrate that the present situation
leaves the propertyless no worse off than they would be in an
(admittedly hard to describe) baseline situation (p. 177). This exten-
sion of the ‘enough and as good’ proviso shows that Locke’s theory
need not reckon unjust all situations in which some sorts of re-
sources have been entirely appropriated, leaving some people
without property of this sort or the chance to appropriate any. It is
enough (and as good) to compensate others with equivalent re-
sources and opportunities, and not necessary for there to be either
the same or the same sorts of items as existed before appropriation
began. For it is only after appropriation that there are entitlements
to particular resources. No rights are violated because the Stone
Age way of life has been destroyed by appropriation, provided there
is a compensating bundle of opportunities and resources.7 Liberty of
the person does not require any particular set of resources or
opportunities.
So far this account of Locke’s and Nozick’s principles of just
acquisition has shown how a system of private property could exist
without violation of rights. But no argument yet given explains how
one rather than another individual acquires a particular holding.
Nor does consideration of the “enough and as good’ proviso show
that there would be anything unjust about reassigning holdings,
provided that the ‘general post’ of holdings does not violate the
proviso.
This gap, of course, is supposed to be filled by the claim that
entitlement is produced by mixing one’s labour. However, the
transition from

A mixed his labour with x

to

A is entitled to (has the right to control) x

requires justification. Nozick believes that we should ‘hold onto the

notions of earning, producing, entitlement, desert and so forth’ (p.
155). But, of course, holding on is not enough; we need an argument
too. And this argument proves perennially elusive.
 Nozick’s Entitlements Sp

Locke pointed out one line of argument when he said of the

original use of nature's products:

The fruit or venison which nourishes the wild Indian, who knows no
enclosure, and is still a tenant in common, must be his and so his—i.e. a
part of him, that another can no longer have any right to it before it can do
him any good for the support of his life (§ 25)

In this case entitlement requires incorporation and the right to

control follows from the liberty of the person. But most appropria-
tion is not literally incorporation. The point of property rights is
gone if we can't have things unless we eat them: for even without
property rights we can still use what nature provides. Locke there-
fore tries to push appropriation back from the moment of consump-
tion to the moment when an item is removed ‘from that common
state Nature placed it in’ (§ 26) and claims that ‘by this labour
something is annexed to it that excludes the common right of men’
(§ 26). When we appropriate we mix our labour, and the final
product is therefore not only nature’s but ours, since it incorporates
something of us, our work’ or ‘labour. We come to own things not
because we incorporate them, but because they incorporate (part of)
us:

. every man has a ‘property’ in his own ‘person’. This nobody has any
right to but himself. The ‘labour’ of his body and the ‘work’ of his hands, we
may say, are properly his. Whatsoever, then, he removes out of the state
that nature hath provided and left it in, he hath mixed his labour with it, and
joined to it something that is his own, and thereby makes it his property.
(§ 26).

The transition in this argument from ‘joined to it something that is

his own’ to ‘makes it his property has been given various ingenious
glosses by Locke’s commentators. One view suggests that he made
an illegitimate transition from ‘work’ or ‘labour’ in the sense of
activity (whose control is justified by liberty of the person) to ‘work’
or ‘labour in the sense of product (whose control therefore remains
unjustified).8 Another engaging suggestion is that Locke was misled
by thinking of production as reproduction. Aristotelianly conceived,
in which ‘something’ is mixed with Nature, ‘the common mother of
all’, and the result is the begetter’s own.® This is not a plausible
316 Onora O'Neill

attribution to a writer who believes that mothers and fathers have

equal and limited rights (§ 52). Nozick himself raises only questions.
He points out the many difficulties in the way of assigning complete
ownership to the contributor of one factor of production (pp. 174-5),
and the more fundamental problem of understanding why any rights
should accrue from mixing one’s labour. Why should not labouring
be a way of losing one’s labour, of improving what is ‘in the common
state’? Why, at best, should the labourer acquire more than a share
in the final product in proportion to his contribution?

LOCKE'S SOLUTION
I want to suggest that Locke provides an explanation of “how a
property begins’, which can be dug from the text without much
exegetical ingenuity, but requires a retreat from Nozick’s stream-
lined Locke. This explanation too rest on a dubious premiss.
Locke’s starting point is that the Earth is common property:

God, who hath given the world to men in common, hath also given them
reason to make use of it to the best advantage oflife and convenience. (§ 25)

He has therefore to explain how common property can give way to

private property. Nozick, by contrast, sets out to explain how what
is unowned can become private property (p. 174). But this sim-
plification does more than eliminate a redundant theological
framework. There is a substantial difference between arguing from
an initial equal liberty to use natural resources and arguing from an
initial equal liberty to use them ‘to the best advantage of life and
convenience’. Again and again Locke states that the purpose of
property is that men should not only use, but make the best possible
use of resources.

‘God has given us all things richly’. Is the voice of reason confirmed by
inspiration? But how far has He given it us—to enjoy’? As much as any one
can make use of to any advantage of life before it spoils, so much he may by
his labour fix a property in. (§ 30)

This ‘non-waste condition’ Nozick sees as merely a means to ensure

that the ‘enough and as good’ proviso is met for as long as possible
 Nozick’s Entitlements aby

(p. 176). But if property is linked to making the best possible use of
resources, then we must avoid waste even when there is no danger
of violating the ‘enough and as good’ proviso. For we do not improve
or use to advantage what we waste even when our waste does not
deprive others of any opportunities.
As Locke sees it, the non-waste condition limits appropriations
because our productive powers and our capacities to consume are
both limited. He belabours this point in §§ 35-35:

As much land as a man tills, plants, improves, cultivates and can use the
product of, so much is his property. (§ 31)
God, when He gave the world in common to all mankind, commanded man
also to labour, and the penury of his condition required it of him. God and
his reason commanded him to subdue the earth—i.e. improve it for the
benefit of life and therein lay out something upon it that was his own, his
labour. (§ 31)
God gave the world to men in common, but since He gave it them for their
benefit and the greatest conveniences oflife they were capable to draw from
it it cannot be supposed He meant it should always remain common and
uncultivated. He gave it to the use of the industrious and rational . . . (§ 33)

And finally:
subduing or cultivating the earth and having dominion, we see, are joined
together. The one gave title to the other. (§ 34)

which leads straight on to an explicit claim that production requires

individual property:

God, by commanding to subdue gave authority to appropriate. And the

condition of human life, which requires labour and materials to work on,
necessarily introduces private possessions. (§ 34)

and to the claim that individual property maximizes production:

he who appropriates land to himself does not lessen but increases the
common stock of mankind. (§ 37)

From these passages it is easy to reconstruct a valid argument

showing how title is acquired by the mixer of labour. It might run:
318 Onora O'Neill

(1) A mixes his labour with x.

(2) A improves x.
(3) A can improve only what he has the right to control.
(4) A has the right to control x.

Locke would argue more strongly that we ought to (try to) control
some x's, since we have a duty to improve the Earth. But the above
argument leaves it open whether we have duties to seek property
rights.
Line (2) of this argument is merely an interpretation of line (1): to
mix one’s labour with something is not merely to cause a change in it
but to improve it in some way (perhaps to make it more desirable or
more accessible). Both Nozick and Locke would presumably accept
this restriction on the sense of ‘mixing one’s labour’. They would not
think that Locke’s acorn gatherers appropriate the forest floor by
leaving their footprints there; they only appropriate the now
gathered acorns.
But line (3) of the argument is implausible. Locke is arguing that
the conditions of production demand the entitlement of producers.
Property rights are rights to use nature productively (improve it),
not just to use it. Such improvements, Locke alleges, can only be
made when producers have property rights. This beliefisuntenable
on two counts. Production can take place when producers do not
have complete control of resources; and it can take place when they
have no right to control resources. Much productive labour is
collective; much has been produced without individual holdings;
even more without full capitalist property rights. On the other hand
complete (or even partial) but unjustified control is also often
sufficient for production. Even thieves or squatters may improve
their ‘holdings’. Locke’s argument can justify individual property
rights only when they are a necessary condition of production,
hence in many cases could justify no sort of property right.
Even in those cases where he could justify individual property
rights, these would not be full capitalist property rights. Since
control is justified as a requirement of improving things, it could
never extend to a right to destroy or fail to make productive use of
some resource. The Parable of the Talents is not far from Locke’s
mind.
It seems therefore that Locke’s argument to show why the mixer
 Nozick’s Entitlements 319

of labour should have title has two defects from Nozick’s point of
view. First, it rests on a false premiss about the requirements of
production; secondly it leads to too weak a conclusion. If entitle-
ment theory is to be rescued both defects must be overcome.
Line 3 of the argument might perhaps be replaced by the weaker

(3') A will improve best what he has the right to control.

In this case the justification of private property rights would be that

they provide better incentives and lead to more efficient production
than can be achieved either with noncapitalist property rights or by
mere possessors. Nozick alludes to these considerations as ‘the
familiar social considerations favoring private property (p. 177). If
we accept (3’) and disregard the familiar social considerations
favouring different sorts of property rights, the argument will not
hold up without an additional premiss mandating maximal produc-
tion, which would suffuse the justification of property rights with a
utilitarian tinge. Even were this accepted, the property rights so
justified would fall short of rights to destroy or to idle productive
resources, and so would not be full capitalist property rights.
Even though it might not be possible for Nozick to adapt Locke’s
justification of property rights, there might be some quite separate
line of argument from (1) to (4). But the available lines of thought do
not seem likely to lead to the desired conclusion.
Nozick might, for example, argue that labour always costs the
labourer something (he suggests this approach on page 175), and so
that the labourer is entitled to some sort of recompense. But this
recompense is unlikely to be complete control of the product, and in
any case cannot be any sort of entitlement since Nozick holds that
nothing is owed in virtue of unsolicited benefits conferred (p. 95). In
the absence of specific contracts we do not have any right that others
should compensate us for effort we voluntarily expend from which
they benefit.
Alternatively Nozick might try to argue that any abridgement of
full capitalist property rights abridges the liberty of the person,
which includes capitalist acts between consenting adults. However,
we could know that such acts were permitted by the right to liberty
only on the basis of some other argument, such as the ‘enough and
as good’ proviso, to show that such acts are not themselves rights-
320 Onora O'Neill

violating. And even were such an argument given, we would need a

way of linking particular holdings to particular individuals. Without
a separate argument to show how entitlement arises, capitalist acts
would include any taking of others’ possessions (to which after all
they can demonstrate no entitlement), and no accumulation of such
acts could establish title. Before capitalist acts between consenting
adults can take place in an orderly way, there must be solitary
capitalist acts. Acquisition precedes transfer.
These negative conclusions are not altered by considering Locke's
supplementary argument that full capitalist property rights arise
only after the introduction of money (§§ 46-50), which permits
unlimited accumulation without the non-waste condition being
violated. It is true that an imperishable medium of exchange allows
more extensive holdings, but it will still not confer rights to hold
unlimited quantities of perishable resources or to hold productive
resources idle or to destroy resources. Nor, of course, will the
supplementary argument give any further explanation of why par-
ticular individuals should have title to particular holdings.

PROCEDURAL AND STRUCTURAL THEORIES OF

DISTRIBUTION

These comments on entitlement theory have been largely adverse.

However, I believe that Nozick has enormously clarified the discus-
sion of theories of just distribution by emphasizing the differences
between entitlement theory and other theories. Entitlement theory
is about procedures for acquiring and transferring property justly;
patterned and end-result theories of distribution are about the
structure of property rights ofa society. There is no reason why both
a structural (patterned or end-result) account and a procedural (e.g.
entitlement) account of a single set of holdings should not be given.
For, contrary to the implications of the Wilt Chamberlain example,
many patterns and end-results are remarkably stable in the face of a
system of voluntary transfers. Of course, in justifying a distribution
of resources one can point one way or the other. Either one may
hold that a particular structure determines whether a distribution is
just, or that following certain procedures shows that it is just. But it
may then turn out that the best route to a just structure is by
controlling (or not controlling) procedures, or that the best way to
 Nozick’s Entitlements Bol

guarantee that certain procedures are followed is to specify a range

of acceptable structures for their outcomes. (Ifthe latter point seems
implausible, consider maximizing the frequency of transfers [gifts
and bequests] to non-related individuals not by specifying rules
facilitating such transfers, but by setting an upper limit on indi-
vidual holdings.)
Nozick is alive to these possibilities, for he praises invisible-hand
explanations which show ‘what looks to be the product of someone’s
intentional design as not being brought about by anyone’s inten-
tions (p. 19). If an end-result or pattern of distribution is sustained
by following certain precedures this would be an example of an
invisible-hand mechanism; similarly if certain procedures are fos-
tered by requiring or ruling out certain end-results. Nozick himself
(p. 202) points to Hayek's claim that capitalist procedures are the
best route to implement Rawls’s difference principle (maximizing
the resources of the worst off) as a case of maintaining an end-state
by adhering to procedures.
But Nozick is not so alive to the vast range of possible procedural
theories of just distribution other than entitlement theory. There
are many possible principles of acquisition which are weaker or less
individualist than Nozick’s streamlined Lockean principle, and
there are many possible principles of just transfer. Such principles
might justify less than full entitlement of individuals to unlimited
amounts of resources of any sort. They might restrict the uses to
which certain (or all) sorts of resources are put, or the persons or
groups who could hold or transfer resources.
Since Nozick’s argument fails to show how individuals can acquire
full entitlement in the first place, he has not given reasons for
rejecting other procedural theories ofjustice. Since the Wilt Cham-
berlain argument presupposes full capitalist entitlements, he has
not shown why one must reject non-procedural theories of justice.
Even if we share with Locke and Nozick the view that individuals
have rights not to be harmed in life, health, or liberty, we have so
far no reason to accept entitlement theory.

NOTES

1. I would like to thank G. A. Cohen for helpful comments and conversa-

tion.
829 Onora O'Neill

2. Robert Nozick, Anarchy, State and Utopia, Basic Books, New York
1974. Parenthetical page references are to this book.
3. Nozick defines the minimal state as ‘protecting all its citizens against
violence, theft and fraud’ (p. 26). This definition jumps the gun since the
latter two protections presuppose property rights which are argued for in
Part II of the book.
4. The problem lies in explaining what is required of just transfers over
and above their being voluntary. Involuntary transfers are clear violations of
entitlement; but so are some voluntary transfers. Since Nozick thinks that
the state should protect individuals against fraud, he presumably places
conditions over and above mere consent on just transfers. Only informed
consent will justify. Once we say this much the specification ofa principle or
just transfer becomes extremely intricate.
5. Later (p. 293) he suggests that historical injustices would be washed
away if one could show that a given situation could have been reached by
some ‘hypothetical just history’ not too different from the actual story. If we
may rewrite history, there will be little we cannot justify.
6. John Locke, Of Civil Government, Second Treatise, Everyman’s
Library, London 1924. Parenthetical paragraph numbers refer to this work.
7. Cf. Anarchy, State and Utopia, p. 178, n. MacPherson holds that
Locke himself extended the “enough and as good’ proviso in this way. C. B.
MacPherson, The Political Theory of Possessive Individualism, Clarendon
Press, Oxford 1962, pp. 212-14.
8. J. P. Day, “Locke on Property,” Philosophical Quarterly, Vol. 16
(1966), attributes this (and kindred) confusion to Locke.
9. R. Brandt, “Zu Locke’s Lehre von Privateigentum’, Kant-Studien,
Vol. 63 (1972).
 16
Yours, Mine, and Ours:
Property Rights and
Individual Liberty
CHEYNEY C. RYAN*

A people averse to the institution of private property is without

the first element of freedom. [LoRD ACTON The History of
Freedom]

Defenders of private property have traditionally linked the rights of

ownership with individual liberty. Adam Smith considered private
acquisition of goods, and the propensity to “truck and barter” them
for others, to be part of the “range of natural liberty” which men
occupied in their prepolitical state; attempts by the government to
interfere with the “free market” resulted not only in inefficient
economies but in intolerable restrictions on man’s natural freedom.
This outlook reflects Locke’s earlier views on property and its
relation to the civil order, as expounded in his Second Treatise.
Unlike Hobbes, Locke believed that the rights of private property
and free exchange logically preceded the establishment of the
political order, arising instead from man’s natural propensities—
particularly for Locke the need for self-preservation. Thus he writes

323
324 Cheyney C. Ryan

(at the beginning of the Treatise) that man’s natural condition is “a

State of perfect Freedom to order their actions, and Persons as they
think fit, within the bounds of the Law of Nature, without asking
leave, or depending upon the Will of any other Man.” When the
rights of property were secured in the civil constitution, the state
merely agreed to respect those activities men freely and “naturally”
engaged in; to do otherwise would be coercion. A century and a half
later, and from a radically different philosophical perspective, Hegel
would voice similar sentiments on the ties between private property
and liberty. “From the standpoint of freedom,” he writes in the
Philosophy of Right, “property is the first embodiment and so is in
itself asubstantive end.”?
In Robert Nozick’s much publicized Anarchy, State, and Utopia,?
the proposed connections between private property and personal
liberty play a key role in his defense of property rights against the
incursions of the welfare state and the schemes of distributive
justice which motivate such welfare programs. Nozick’s defense of
the “laissez-faire” position is a fresh one, though its themes are
traditional. In this paper I will examine some of Nozick’s arguments,
with the hope of illuminating the links between property and liberty
which have so impressed others.
Sections I and II discuss Nozick’s claim that enforcing “patterned”
principles of distributive justice invariably leads to restrictions on
personal liberty, an intolerable interference with people’s lives. The
problem with this argument is that it assumes prior commitment to
the rights of private ownership (and the freedoms those rights
insure); where those rights are not taken to apply, the “restrictions”
necessary to maintain patterns seem of little consequence. In Sec-
tion HI I show that the claim that free exchange “naturally” gives
rise to private ownership of the means of production also holds true
only if we assume extensive private property rights. Since the
debate between “socialist” and “capitalist” conceptions of justice
turns largely on the existence of private property rights, an argu-
ment like Nozick’s against alternative views which presumes such
rights appears somewhat off the mark. It seems that Nozick takes
the value’ of private property rights to be self-evident because of
their link with individual freedom. But in Section IV I present an
argument to show that the institution of private ownership itself
involves continuous interference with the freedom of the majority.
 Property Rights and Individual Liberty 325

Nozick’s response to this argument—strangely enough— involves

dropping the great importance he elsewhere attributes to individual
freedom. In the Conclusion I discuss some points raised by discus-
sion concerning the general relation of liberty to property rights.

LIBERTY AND PATTERNS

I want to consider first Nozick’s critique of patterned conceptions of

justice and the defense of his own entitlement theory. The entitle-
ment theory is designed to bolster the case for Nozick’s updated
version of the night-watchman state, for Nozick recognizes that a
chief objection to this minimal state will be that, without a more
extensive state apparatus, the claims of distributive justice will go
unmet. Earlier proponents of laissez-faire policies (such as Frank
Knight and F. A. Hayek) met this charge by arguing that the “free
market” mechanism, left to itself, would reward individuals ac-
cording to their productive contribution, thus generating a naturally
just distribution. Nozick is aware of some of the problems with the
theory of income distribution which underlies this argument. To
begin with, marginal products determine income only with a dis-
tribution of holdings already given (so the problem of justice is
merely shunted back a step); even more seriously, the “Cambridge
Controversy” has raised questions about the logical coherence of the
theory of marginal products for a nonstationary economy.? So
Nozick wisely avoids pegging the justice of capitalism to the way the
price mechanism distributes wealth and attempts instead to disarm
the claims of distributive justice at the outset by attacking the very
notion that a just society is one whose goods and services are
distributed according to some preconceived pattern. A just distribu-
tion, Nozick will argue, is whatever distribution is brought about by
the process of “free” exchange.
Nozick’s defense of his own theory rests primarily on his critique
of its competitors. “It is not clear,” he writes, “how those holding
alternative conceptions of distributive justice can reject the entitle-
ment conception ofjustice in holdings.”4 Against patterned concep-
tions of justice he argues that personal liberty and such patterned
principles will always be at odds with one another: liberty will tend
to “upset” any given pattern once instituted and, conversely, the
attempt to maintain such patterns (by the more than minimal state)
326 Cheyney C. Ryan

will always restrict individual liberty. “No distributional patterned

principle of justice,” he argues, “can be continually realized without
continuous interference with people's lives.” The chief point in
favor of the entitlement theory is, in Nozick’s opinion, its compati-
bility with the individual's liberty.
(I would agree with those who see the appeal to personal liberty
as the key argument in his account of distributive justice; as James
Dick writes in a perceptive article on distributive theories, the
“fundamental reason” for Nozick’s rejection of the “patterned”
approach—and for his adoption of a “process” criterion— ‘is that the
value ofliberty requires it.”® In my opinion, it is the most important
argument of the whole book. Since many of us have not felt Nozick’s
attraction for anarchism—or if we have, it has been for different
reasons—his argument that a state could exist which did not violate
anyone’s rights is an interesting exercise in dialectic, but little more.
What is striking is his contention that the more-than-minimal state,
which is seemingly required by “patterned” conceptions ofjustice,
must infringe on personal liberty. If this is true, we are in a position,
it seems, of accepting Nozick’s general views, or resigning ourselves
to an inevitable conflict between the claims of justice and those of
liberty.)
“Let us call a principle of distribution patterned if it specifies that
a distribution is to vary along with some natural dimension, weigh-
ted sum of natural dimensions, or lexicographic ordering of natural
dimensions.” As examples of patterned principles, Nozick suggests
that we can turn to “almost every suggested principle of distributive
justive”—those that would distribute holdings according to moral
merit, or need, or marginal products, etc. Suppose that an initial
patterned principle determining a distribution of holdings is
realized. Nozick does not specify what sort of goods, or “holdings,”
are being distributed; we may presume, then, that his argument is
to hold true for all types of holdings. “Let us suppose it is your
favorite one and let us call this distribution D1; perhaps everyone
has an equal share, perhaps shares vary in accordance with some
natural dimension you treasure.”? Given this initial position, it is
easy to imagine circumstances where individuals might agree among
themselves to transfer their holdings, exchange them for others, or
just give them away, and in so doing upset the favored pattern. As
an illustration, Nozick chooses a case in which individuals exchange
 Property Rights and Individual Liberty 427

parts of their personal income for the services of others. Basketball

fans, for example, may willingly pay a special twenty-five cents to
see Wilt Chamberlain play basketball (in his off-hours, say), thus
providing Chamberlain with an income far greater than his initially
determined share; or—less realistically—you and I in our off-hours
may agree to give philosophy lectures for remuneration from our
neighbors.
Any such actions will involve moving to a new distribution of
personal income (D2), one which does not correspond to the initial
patterned allocation of income shares. But: “If Di was a just
distribution, and people voluntarily moved from it to D2, transfer-
ring parts of their shares they were given under D1. . . isn’t D2 also
just?” “There is no question about whether each of the people was
entitled to the control over the resources they held in D1; because
that was the distribution (your favorite) that (for the purposes of
argument) we assumed was acceptable.”® To prevent the transition
to Dz in the name of maintaining the favored pattern distribution
seems to involve preventing people from doing what they choose to
do, and this is Nozick’s point. People’s liberty—as embodied in
their desires to transfer their holdings to one another—grates
against the maintenance of a pattern of distribution. “To maintain a
pattern one must either continually interfere to stop people from
transferring their resources as they wish to, or continually (or
periodically) interfere to take from some persons resources that
others for some reason chose to transfer to them.”® Sustaining such
patterned distributions would seem to involve, then, continuous
interference with people’s lives on the part of the state.
Nozick feels that he has a strong argument here—it is designed to
prove, after all, that proponents of any of the competing conceptions
of justice must ultimately be drawn to the entitlement theory. Let
us be clear on what it purports to show.
Traditionally critics of the more-than-minimal (welfare) state have
charged that “meddling” by the government to bring about or
maintain a fairer distribution of resources involves tampering with
the rights of private property. At a time when rights of private
ownership were accorded a supreme value, this was regarded as a
pretty good argument against distributive programs—but the days
when private property rights were held sacrosanct are past. An
argument which pegged everything on a prior commitment to the
328 Cheyney C. Ryan

rights of private property would, on its own merits, not (I think) be

very telling. Certainly such an argument would not accomplish what
Nozick feels his argument does—the demonstration that all other
conceptions of justice must be rejected in favor of the entitlement
theory. For not all other conceptions ofjustice share a commitment
to private property rights; a critique of the socialist conceptions of
justice which presupposed the rights of private property would
stand accused of missing the thrust of the socialist case.
So what Nozick tries to show is that personal liberty upsets
patterns: not private property rights, but personal liberty requires
that we adopt an entitlement conception of justice. His claim would
seem to be that, regardless of the strength of our commitment to
such property rights, our shared commitment to personal liberty
requires that we forsake whatever principle of justice we now hold
for that which Nozick proposes. This claim, if made to stick, would
have the required force to show patterned principles unacceptable,
for it is certainly legitimate to assume a consensus concerning the
value of liberty (what Feinberg calls a “presumptive case for lib-
erty )!° and the undesirability of any social order which continually
infringed on the liberty of its citizens. Hence it is on the commit-
ment to personal liberty that Nozick hopes to build his case against
patterned principles ofjustice.

HOLDINGS AND OWNERSHIP

Is Nozick’s strategy successful? I begin this section by describing a

simple case of a patterned distribution of holdings in which indi-
viduals are not “free” to give away or exchange their holdings, yet it
is clear that no coercion is involved. What is unique about the
distribution I describe is that individuals are not presumed to have
rights of ownership in their holdings; the point it illustrates is that
restricting someone's freedom to exchange a holding constitutes
coercion (infringement on personal liberty) only if the holding falls
under the rights of private ownership (only if the holding is private
property.) If Nozick’s appeal to the restrictions of free exchange
needed to sustain patterns is to be sufficient for rejecting patterned
distributions of holdings in general, he must assume that all hold-
ings come under the rights of private ownership—that all holdings
are private property. But such an assumption of extensive private
 Property Rights and Individual Liberty 329

property rights, I argue, undercuts his general case against pat-

terned principles ofjustice.
In our public university systems there are a finite number of
teaching positions in philosophy. These positions are held by men
and women of various ages, having been awarded them presumably
on the basis of merit and ability. An educational researcher might
investigate how the available positions were alloted to the available
Ph. D.’s ina given state in a given year; looking at the overall pattern
of who held what jobs, he might reach some conclusion as to the
equity or fairness of the distribution of available positions (in one
state, for example, the tenure system was criticized because it led to
the unfair situation of older, but less qualified teachers holding
positions over younger, more competent Ph.D.’s). There is of
course little occasion to assess the overall distribution of jobs in
philosophy among the available philosophers, but most of us (I
would assume) hold some patterned conception of what a fair
allotment of teaching positions would look like: merit, teaching
ability, and the needs of the department would undoubtedly be our
main criteria.
Assume that some (patterned) principle for determining who
holds what position is realized. Clearly, freedom of a certain sort
would continually disrupt this “pattern” —the freedom of those who
held positions to give their jobs away, exchange them for the
services of others, or will them to a loved one. If individual
jobholders were free to do such things, it would be impossible to
insure that those who held jobs “deserved” them by whatever
criterion of desert we have chosen. But of course individuals do not
have such freedoms, for such freedoms do not follow from the rights
of holders of teaching jobs (that is why they hold their positions,
rather than own them). Would we then conclude that the personal
liberty of those who hold positions is infringed upon, or restricted,
because they cannot sell their jobs? That maintaining a pattern of
jobs held by merit involves “interference” with people’s lives? This
would be absurd.
This example shows that where the holding in question is some-
thing like a teaching position—something which the holder does not
own—maintaining a pattern of distribution by “preventing” indi-
viduals from freely exchanging their holdings (jobs) does not seem to
involve violating their personal liberty. It is certainly true that we
330 Cheyney C. Ryan

are not “free” to exchange our jobs as we wish to, but whether or not
this lack of freedom constitutes an infringement on personal liberty
depends on the rights we have over the holding in question. Where
my rights in something I have come to hold do not include the right
to sell or exchange it, then preventing me from doing so hardly
constitutes a restriction of my liberty. If a book belonging to one of
my colleagues inadvertently falls into my hands, I am not “free” to
sell it (or give it away) to one of my students, but this lack of freedom
does not amount to a restriction of my liberty since I have not the
right to sell it; similarly, if the university I work for provides me with
a car for a trip required by my professional duties, it is not a
restriction of my personal liberty that I cannot immediately sell it to
the nearest car dealer (though I do have rights to, say, the use of the
car). In the case of a teaching position, one certainly acquires some
rights over a job by signing a contract or being given tenure: one’s
right to a job and right to benefit from the job are infringed, for
example, when one is dismissed for improper reasons, and the
possession of such rights seems to imply that a holder has a
“property interest” in his job in some weak sense. But if the rights of
ownership are taken to include (as they usually are) not only the
right to use and the right to appropriate returns from a holding but
also the right to dispose of aholding as one chooses, then holders
of teaching positions cannot be said to own them and hence cannot
be coerced when prevented from selling what they do not own. (Of
course, teaching positions could be private property, just as
positions in the army were at one time things that could be bought
and sold. Our difficulty with thinking of such positions as private
property is attributable, I think, to the belief that they should not be
bought and sold. If rights of ownership were extended to teaching
positions, then they would come under the class of “intangible”
property that now includes such things as seats on the stock
exchange and copyrights of tunes. )!2
How does this example bear on Nozick’s general critique of
patterned distributions of holdings?
The case of teaching positions shows that for any set of holdings,
sustaining a pattern implies coercive restrictions (restrictions on
personal liberty) only if the “holdings” are private property—
holders have full rights of ownership in them. In Nozick’s illus-
trations involving shares of personal income, we may assume that
 Property Rights and Individual Liberty 331

individuals own what they exchange. But to generalize from this sort
of case to all economic holdings (as Nozick does) we must assume
that all holdings are, like personal income shares, privately owned.
Without the assumption that private property rights extend to all
present and potential economic holdings, Nozick’s general conten-
tion that sustaining patterns of distribution implies the restriction of
liberty simply will not hold: if each individual's holdings are not
assumed to be his private property, then there is no reason to
conclude that restrictions on the “free exchange” of holdings consti-
tutes coercion.
That Nozick equates “holdings” with private property, and hence
assumes an extensive set of private property rights, is clear from his
own remarks. After asking the hypothetical opponent to choose his
“favorite” patterned principle and consider holdings to be distri-
buted accordingly, Nozick remarks that there can be “no question”
that each person is entitled to “control” over the resources he holds,
since the distribution is one which the opponent has chosen (by
“control” Nozick means the right to sell or exchange).!* But there is
no question only so long as we assume that, once the holdings have
been distributed, individuals are accorded rights of ownership in
the things they hold. But why should we assume this? Different
conceptions of justice differ not only in how they would apportion
society's holdings but in what rights individuals have over their
holdings once they have been apportioned. The Marxist conception
of justice, for example, sees the rights of ownership as extending
only to personal items, while the vast bulk of society's holdings (its
means of production, natural resources, etc.) falls under the rights of
public property; individuals are accorded the right to use or exploit
this latter sort of property for specific purposes, but they do not
possess the right to give it away or sell it. Hence whether or not an
individual's coming to hold a good entitles him to sell it depends, for
the Marxist, on what sort of holding it is. If Nozick’s hypothetical
opponent is a Marxist, there is no reason to conclude (as Nozick
does) that simply because an individual has received a holding
according to the Marxist’s “pattern” he has the right to dispose of it
as he wishes.
We see, then, that Nozick’s is not really a case against patterned
distributions of “holdings” in general—it is at best a case against
patterned distributions of privately owned property. As such, the
332 Cheyney C. Ryan

argument cannot accomplish what Nozick hoped it would. First of

all, it does not show why those who hold alternative conceptions of
justice must come to hold the entitlement theory. If Nozick’s appeal
to the coercion involved in sustaining patterns rests (as I have
argued) on the assumption that “holdings” are equatable with
“private property,” then his argument does not even address those
conceptions of justice which reject extensive private property
rights. In the second place, Nozick’s argument is unsuccessful in
suspending the question of property rights and appealing directly to
considerations of personal liberty. It is not clear how Nozick’s
argument could have succeeded here. To claim that certain
restrictions on what people may do with their holdings constitutes
coercion presupposes (I have suggested) prior determination of the
rights they have over their holdings. Nozick’s appeal to personal
liberty presumes commitment to the rights of private ownership,
and it would seem that any appeal to personal liberty in the matter
of how people may employ their holdings must rest on a conception
of what rights individuals have over their holdings in the first
place.14

THE SLIPPERY SLOPE INTO CAPITALISM

A number of Nozick’s remarks suggest a possible response to my

claim that he has failed to appreciate the distincitons which alterna-
tive conceptions ofjustice draw between types of holdings and the
rights applicable thereto. It might be argued that if rights of
ownership are allowed in any set of holdings (such as personal items)
liberty will require that those rights be extended to all other types of
holdings, such as means of production and natural resources; hence
any restriction of rights of ownership implies restrictions of personal
liberty. Nozick specifically addresses restrictions on the ownership
of means of production, and so I will deal with the argument as it
applies to those goods in particular.
If we allow ownership of personal items, the argument goes, what
is to stop someone from working in his off hours (performing
services for others, say) to accumulate a far greater share of personal
goods than anyone else, and then—by exchanging them for the
labor of others—starting his own “capitalist enterprise?” Through
such activities, Nozick remarks, factories would spring up in
 Property Rights and Individual Liberty oe

socialist society unless forbidden: “I melt down some of my pos-

sessions (under D1) and build a machine out of the material. I offer
you, and others, a philosophy lecture once a week, in exchange for
your cranking the handle of my machine, whose products I ex-
change for yet other things, and so on.” From the possibility of such
activites we may see ~. . . how private property even in means of
production would occur in a socialist society that did not forbid
people to use as they wished some of the resources they are given
under socialist distribution D1.”15 The point of this parable is that if
socialist society allows any privately owned goods, then free ex-
change of those goods and their use may generate privately owned
means of production. To prevent ownership in the means of produc-
tion from occurring in this way (by passing a law against it) means
restricting people’s use of the resources they have been allotted.
But this—we are to conclude—amounts to an intolerable restriction
on their personal liberty (the particular liberty at issue here would
seem to be the freedom to use one’s goods as one wished, along with
the freedom to exchange them).
It seems undeniable that, by working in one’s off-hours and with
the cooperation of others, one could come to possess a factory in a
way described by the parable. But does this show that private
property in the means of production “naturally” emerges from such
a process? The distinction between possession and ownership be-
comes relevant here. Possession ofa good is a physical relationship,
while ownership is a normative one involving the rights an indi-
vidual has over a thing; to say that a good is someone's private
property is to say not merely (or not even) that he possesses it but
that he owns it. The mere fact that one has come to possess a good
does not imply that one has acquired the rights of ownership in it.
One does not own a good if it has been acquired through theft. What
must be shown, then, is that people’s free use of their resources can
give rise not only to the acquisition of means of production but to
their ownership.
Now the force of Nozick’s parable derives from the fact that the
factory the individual possesses was not stolen or acquired through
some illicit means but was gained through his diligence and with the
cooperation of others. And implicit in the argument is the notion
that if a good has been so acquired, then the holder should quite
properly have rights of ownership in it, and to deny him those rights
334 Cheyney C. Ryan

is to infringe on his personal liberty. This latter notion may be

questioned, though, for the fact that a good has been acquired
through free exchange and exploitation of personal resources is not
by itself sufficient ground for granting ownership of the good.
Consider a case which Nozick himself provides: an individual could
conceivably through a clever series of purchases appropriate all the
drinkable water in a community (or conceivably the world). Does
this imply that he should be granted permanent property rights in
the water? Nozick, concurring with Locke on this point, denies that
ownership of an entire community s water supply is something that
any individual should own (just think of the power that such
ownership would entail). Similar considerations weigh against own-
ership of a single water supply if it is the only supply available to a
community. For holdings such as this, which no individual may
own, it matters little that someone might acquire it through dili-
gence or free market channels.!® Certainly the possibility of such
acquisition is not sufficient ground for extending rights of private
ownership to such a holding. Does restricting private rights in this
way infringe on personal liberty? Allowing an individual to
monopolize the drinkable water would certainly threaten the free-
dom of access of a great many to that water, and would ultimately
threaten their survival.
Nozick seems to grant, in this case, that the manner in which a
holding is acquired is not definitive in determining whether that
holding may be privately owned. Certain considerations (for Nozick,
those contained in the Lockean Proviso) simply exclude certain
types of holdings from private ownership. If the freedom to use
one’s resources fails to generate private property rights in a commu-
nity’s water supply, will it generate such rights in the means of
production? The answer to this question depends on the principles
one accepts for excluding and including different types of holdings
from private ownership. A Lockean will answer yes, because his
proviso does not exclude such holdings from private control; but a
Marxist, whose principles do not commit him to such extensive
ownership rights, will answer no. Nozick’s contention that free
acquisition of means of production will “naturally” give rise to
ownership of such goods thus presumes a view of property rights
which Marxists and others will contest. But does not personal
liberty provide a presumptive case for private rights, here and
 Property Rights and Individual Liberty 335

elsewhere? The water-supply example has already suggested that

the freedom to use and exploit personal resources may not be the
only freedom at issue in determining whether a holding may be
privately owned. If the appeal to liberty is to be sincere, then it
seems that one should consider all the freedoms affected by such an
extension of rights (this point will be taken up again in the next
section).
I do not intend to discuss where the socialist’s arguments against
private ownership of capital; as in the water supply case, they
involve the power over others which ownership accords the hol-
der.‘7 But clearly if one agrees with the socialist that a community's
means of production, like its water supply, should not be privately
owned, then it is hard to see why the mere possibility of free
acquisition should commit one to granting rights of private owner-
ship in the former case any more than the latter. In both cases, the
restrictions on the use of individual resources implied by public
ownership will seem intolerable only if one assumes that the holding
in question properly belongs in private hands. Nozick assumes this
for means of production, hence his concern for restricting private
ownership; but his appeal to liberty fails to show, I think, why we
must share this concern.

ENTITLEMENTS AND LIBERTY

Nozick’s argument that liberty will upset any patterned distribution

of holdings and his argument that “free” use of resources will
naturally give rise to private property in the means of production
both rest on prior commitment to extensive private property rights.
Nozick’s case thus hinges on thejustification he can provide for such
rights: if that justification is a sound one his attack on other social
ideals may work, but if it is not his arguments, as I have tried to
show, cannot possibly succeed. What case does Nozick present for
his framework of rights?
At first glance, it does not seem he presents any case! He seems to
say as much in his preface: “This book does not present a precise
theory of the moral basis of the individual rights.”!® Moreover some
of the traditional justifications of private property rights are
explicitly (or implicitly) rejected. He eschews the utilitarian
justification of private property which points to its efficiency; he
336 Cheyney C. Ryan

criticizes the Lockean view that rights emerge when an individual

mixes his labor with unowned objects; and his methodology pre-
cludes that such rights could be the product of a social contract
(Hobbes,) because he rejects the social contract approach for “in-
visible hand” descriptions.
An important structural difference between Nozick’s theory and
that of John Rawls with regard to the “grounding” of a set of rights is
worth noting here. Nozick assumes—it seems—a set of rights (or
asssumes that the link between personal liberty and these rights is
obvious) and then examines the political structures which could
arise within the constraints set by these rights—thus the invisible
hand approach. In Rawls’s theory we do not begin with a set of
rights, but rights are themselves subject to debate in the contract
situation. Rawls himself feels that his theory is neutral with regard
to economic systems, but certainly a case might be made—on the
grounds of justice as fairness—against extensive private property
rights. In this respect Rawls’s approach gives us a lever for evaluat-
ing such rights from the point of view of justice that is simply not
present in Nozick’s approach, which requires acceptance of his
fundamental rights merely to get off the ground. In the all-
important controversies about which rights to respect, then, Rawls’s
theory proves considerably more powerful.'9
How are we to interpret Nozick’s seeming omission of a justifica-
tion for private property rights? Nozick may feel that he can simply
appeal to our “moral intuitions” on this matter; recourse to such
intuitions plays an important role in the justification of any moral
theory, but such a recourse in the matter of individual property
rights would seem to involve resolving by fiat what is the major
point of controversy between “capitalist” and “socialist” conceptions
of justice. Nozick’s commitment to private property rights is instead
based (I think) on his belief that such rights are demanded by the
values of personal liberty, or autonomy. In arguing against those
conceptions of justice which would restrict such rights Nozick, as we
have seen, persistently appeals to personal liberty; in linking private
property with personal liberty he seems to ally himself with that
tradition of defenders of laissez faire mentioned at the outset of this
paper. If the link were obvious, it would constitute a powerful
endorsement of the rights of ownership, since our (assumed) shared
commitment to liberty would then carry over to those rights as well.
I will now argue that the link between private property and
 Property Rights and Individual Liberty ney E

liberty is not obvious, for a case can be made that the institution of
private property itself brought about (and sustains) a substantial
restriction on individual freedom.?° The ways in which “entitle-
ments’ restrict freedom present serious difficulties for anyone who
would link private property with individual liberty, and Nozick does
not adequately deal with these difficulties. More importantly,
though, the case against ownership rights (along with Nozick’s
arguments against patterned principles) itself suggests a general
difficulty with justifying rights in terms of the liberties they restrict.
I will touch on this difficulty in the Conclusion.
In precapitalist forms of property most communities had large
tracts of common ground (“commons”) which were available to all
for the grazing of sheep, growing of crops, even habitation. It was
the right of every townsperson to use this land, and the freedom of
access to it was taken to be one of the citizen’s basic liberties. The
enclosure movement subsequently transferred this land from the
authority of common ownership to one of private ownership—it
involved, in other words, a shift in the rights applicable to the land.
For the vast majority of people living in England during this period
of enclosure, the extension of (modern) rights of ownership meant
the abolition of rights they had previously enjoyed, a significant
diminishing of their freedom of access to the land, the freedom to
use its soil, water, etc. This restriction was, of course, permanent.
And it was viewed at the time as a significant attack on the rights and
liberties of Englishmen (of the lower classes, of course). Here is the
opinion of a Norfolk laborer, addressed to the proponents of enclo-
sure: “You do as you like, you rob the poor of their Commons right,
plough the grass up that God sends to grow, that a poor man may
feed a Cow, Pig, Horse, or Ass; lay muck and stones on the road to
prevent the grass growing. ... There is 5 or 6 of you that have
gotten all the whole of the Land of the parish in your own hands and
you would wish to be rich and starve all the others of the poor.”??
Because it involved a transition from rights and liberties held by
many to private rights held by a relative few, one could sensibly
argue that the extension of private ownership brought about (then,
and for subsequent centuries) a diminishing of liberites (think of all
those places that commonfolk were no longer free to go, free to
use—whose access was denied them by the policies of the state,
respecting rights of private ownership).
From this familiar bit of history one can easily construct a case
338 Cheyney C. Ryan

against the rights of individual ownership by pointing to the overall

restrictions of liberty they brought about. It is worth noting that the
freedoms involved here are not the same ones Nozick has in mind
when he appeals to “liberty” in general. Nozick’s argument against
patterns is concerned specifically with the freedom to dispose of
property, while the freedoms restricted by private rights are those
to use and exploit common property. If our concern is with freedom
in the aggregate, then all freedoms with respect to holdings should
be considered; hence the case against extensions of private property
rights.
Nozick touches on this problem, though not in the form I have
put it, in his discussion of Locke’s theory of acquisition. But
surprisingly Nozick’s response is to play down the importance of
individual freedom, which he elsewhere treasures so highly.
Aware of the controversies surrounding enclosure at the time he
was writing, Locke asserted the right to appropriate property with
the proviso that there be “enough and as good left in common for
others.” If we take this to mean that individual appropriation cannot
diminish the freedom of others to use land and goods they now
exploit in common, it is obvious that the proviso puts an effective
throttle on almost any extensions of private acquisition and the
extension of private property rights. It may even imply, Nozick
comments, that “no natural right to private property can arise by the
Lockean process.” It certainly raises the question of how the
extension of exclusive property rights can be compatible with the
general freedom, as embodied in the freedom to use and exploit
nature's resources.
Nozick’s response to this difficulty is confusing. His attitude is
stated most clearly in a footnote to his discussion of the proviso:
“Fourier held that since the process of civilization had deprived
members of society of certain liberties (to gather, pasture, engage in
the chase) a socially guaranteed minimum provision for persons was
justified as compensation for the loss . . . But this puts the point too
strongly. This compensation would be due those persons, if any, for
whom the process of civilization was net loss. for whom the benefits
of civilization did not counterbalance being deprived of these par-
ticular liberties.”?® Nozick recognizes that certain liberties are very
definitely restricted by the extension of private ownership; it is good
to keep in mind that these restrictions are enforced by the modern
capitalist state, defending the rights of private acquisition (and
 Property Rights and Individual Liberty 339

interfering, perhaps, in people’s lives?), Yet Nozick invites us to

look to the “benefits” of (capitalist) civilization. “Is the situation of
persons who are unable to appropriate (there being no more acces-
sible and useful unowned objects) worsened by a system allowing
appropriation and permanent property?” To suggest that it is not,
Nozick has introduced the familiar arguments for how private
property is the source of long run material gain (by increasing the
social product, encouraging efficiency, etc.).22 The soundness of
these arguments is not so important as their intent: to show that
private property benefits the physical welfare of the people.
Now previously we have seen Nozick invoke personal liberty as
the decisive ground for rejecting patterned principles of justice and
restrictions on the ownership of capital (arguments here that such
policies may maximize general well-being he considers irrelevant).
But where the rights of private property admittedly restrict the
liberties of the average person, he seems perfectly willing to trade
off such liberties against material gain for the society as a whole.
Why is liberty accorded a primary importance in one case and not in
the other? Why should the freedom to dispose of one’s property be
of such great concern, while the freedom to use, walk on, look at
common property can be dispensed with in the interest of the
general welfare? Quite possibly Nozick feels that person autonomy
requires not liberty in general, but the particular freedoms which
the rights of private ownership ensure. Such a view seems to be
suggested elsewhere in Nozick’s discussion; it would explain why he
so prizes the freedoms to sell and exchange and places such little
value on others.24 But the arguments of Nozick’s which I have
discussed, particularly that against patterns, purport to appeal to
liberty in general, and not merely to those liberties which presup-
pose private ownership. More important, though, is the fact that
Nozick has yet to make clear (to me, at least) the direct connection
between personal autonomy, which we all treasure, and the exten-
sive private property rights which he treasures. Until he does, the
priority he gives to some freedoms over others will remain unex-
plained.

CONCLUSION

The argument against exclusive property rights raises (at the very
least) some difficulties for those who directly link private property
340 Cheyney C. Ryan

rights with personal liberty. Taken together with Nozick’s argu-

ments, it also raises some general problems about the relation
between property, liberty, and rights—problems which are worth
noting.
1. Though we speak of “ownership” as a right one has in a
holding, it is in fact a bundle of rights which (ordinarily) includes the
right to use a holding, to benefit from (or appropriate) returns from
it, and to dispose of (or alienate) it. Each of these rights secures a
particular freedom the owner has with respect to his holding, and it
is sometimes important to distinguish these particular freedoms. As
I have already noted, when Nozick appeals to “liberty” in his
argument against patterns, his appeal is not to liberty in general but
to the particular freedom an owner has to sell his property. The
argument against private property appeals to “liberty” as well: but
the particular freedoms it appeals to are the freedoms to (nonexclu-
sive) use of, and benefit from, a set of holdings, and the freedom to
exchange property is not an issue here. Once we appreciate the
multiplicity of rights one may have in a holding, and the different
sorts of freedoms these rights secure, we may come to be skeptical
of claims that liberty per se demands a particular set of (property)
rights. Any given set of rights (such as private property rights)
insures some freedoms (such as the freedom to exchange) but also
thwarts others (the freedom of others to use property now privately
owned). “Freedom” in the abstract does not determine, one way or
another, whether private property rights should be adopted; if we
adopt such rights, it is because we treasure certain freedoms more
than others, we value the freedoms which private property rights
secure more than those secured by other sets of rights. And this
valuation must itself, of course, be justified.
2. In delineating the links between property rights and indi-
vidual liberty, one need not focus solely on the freedoms a set of
rights secures with respect to holdings. And in suggesting that the
choice between private rights and “public” ones is a choice between
the types of freedoms these rights secure over holdings, I do not
mean to ignore the links some may draw between a particular set of
property rights and freedoms ofa“noneconomic” sort which may be
promoted by those rights. Milton Friedman, for example, argues
persuasively (though, I believe, unsuccessfully) that private prop-
erty is necessary for a truly pluralistic society, in which the freedoms
 Property Rights and Individual Liberty 341

of speech, thought, and political participation are secured.2> His

argument is a forceful one precisely because the “liberties” which he
links to private property are the sort which no one (in our political
tradition at least) would want to question. But where the liberties
appealed to directly involve what one may (or may not) do with
holdings, it is a different matter: for a dispute over property rights
simply is a dispute over the importance of the liberties such rights
establish.
3. Finally, I should note an important limitation shared by
Nozick’s argument against patterns and the argument of Section IV
against private rights. We saw in Nozick’s case that to argue that a
particular restriction on what people may do with their holdings
constitutes coercion one had to assume a set of rights which entitled
individuals to do with their holdings what they were (presumably)
restricted from doing. The need to assume some background set of
rights holds true for the case against private property as well:
restrictions on the use of (heretofore) common property will consti-
tute restrictions on liberty only if we assume the rights of earlier
forms of property, by which individuals were entitled to use of
common lands. The problem with arguments which must assume a
set of rights to make their appeals to personal liberty stick is that
they will be oflittle consequence when confronted with views which
reject their particular set of rights. The failure of Nozick’s argument
to even address conceptions of justice which reject private property
rights is an illustration, I think, of this more general point.

NOTES

*For their comments and criticisms of this paper I would like to thank
Antonia Ristorcelli, Hillel Steiner, Michael Gorr, and the editor of Ethics.
A portion of this paper was read at the Pacific Division of the American
Philosophical Association, March 1976. I am indebted to the remarks of my
commentator, Carl Cranor.
1. See Smith’s The Wealth of Nations, book 1. Locke believed that our
natural rights (including the right to private property) required little
justification because they followed from the axiomatic proposition that men
are naturally free and equal (see the discussion of Locke’s theory of property
rights in C. B. MacPherson, The Political Theory of Possessive Indi-
342 Cheyney C. Ryan

vidualism [Oxford: Oxford University Press, 1962], pp. 197-221). Quote

from Locke is from the Second Treatise, sec. 4.
2. Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books,
1974); hereafter cited as ASU.
3. On the “Cambridge Controversy,” see Capital and Growth, ed. G. C.
Harcourt and N. F. Laing (New York: Penguin Books, 1971); G. C.
Harcourt, Some Cambridge Controversies in the Theory of Capital (Cam-
bridge: Cambridge University Press, 1972). It would appear from his
comments that Nozick does not grasp the full import of the Cambridge
critique of the traditional theory of distribution (see ASU, pp. 188-89).
4. ASU, p. 60.
5. Ibid., p. 163.
6. James Dick, “How to Justify a Distribution of Earnings,” Philosophy
and Public Affairs 4.
7. ASU, pp. 160-61.
8. Ibid., p. 161.
9. Ibid., p. 163.
10. Joel Feinberg, Social Philosophy (Englewood Cliffs, N.J.:
Prentice-Hall, Inc., 1973), chap. 3.
11. See E. Furubotn and S. Pejovich, The Economics of Property Rights
(Cambridge: Ballinger Publishing Co., 1974), introduction.
12. On tangible property, see C. Reino!ld Noyes, The Institution of
Property (New York: Longmans, Green & Co., 1936), pp. 382-87.
13. The relevant passage is cited above—it appears in ASU, p. 168; see
also ASU, p. 166.
14. This point will be taken up again in the Conclusion.
15. ASU, pp. 162-63.
16. Ibid., p. 179. Nozick’s discussion alludes to the interesting views of
Hastings Rashdall, “The Philosophical Theory of Property,” in Property, Its
Duties and Rights (London: Macmillan Co., 1915).
17. Since much of the force of Nozick’s parable derives from his
identification of means of production with small machines, the sort one
could easily manufacture privately, it is worth noting that Marxist argu-
ments against private ownership of capital do not bear on such items.
Nozick has some interesting things to say about these arguments in chap. 8,
“Equality, Envy, Exploitation, Etc.”
18h ASU, py xiv.
19. See John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard
University Press, Belknap Press, 1971). Rights are not recognized in the
original position, but are created by the individuals contracting to form
society. Theoretically, it is possible for the contractors to deny rights which
to us seem natural (this point is raised by John Chapman in his article,
 Property Rights and Individual Liberty 343

“Justice and Fairness,” in Justice: Nomos VI, ed. Carl Friedrich and John
Chapman [Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1963]).
20. This same point is touched on in Henry Sidgwick, The Methods of
Ethics (New York: Dover Publications, 1966), pp. 276-78; and in H. J.
McCloskey, “A Critique of the Ideals of Liberty,” Mind 74 (1965): 486-87.
21. F. O. O'Connor, The Employer and the Employed (1844), cited in E.
P. Thompson, The Making of the English Working Class (New York:
Vintage Books, 1966), pp. 230-31. Thompson’s book gives an excellent
account of how the enclosure movement was perceived.
22. ASU, pp. 178-79; my italics
23. Ibid., p. 177.
24. See, for example, Nozick’s discussion of moral constraints in chap. 3.
This line of thought was suggested to me in private communications from
Hillel Steiner and Michael Gorr; my remarks here are a brief (though
inadequate) response to their criticisms.
25. See Friedman’s arguments in his Capitalism and Freedom (Chicago:
University of Chicago Press, 1962).
 7,
Nozick’s Entitlement
Theory
LAWRENCE DaAvIs*

Robert Nozick’s Anarchy, State, and Utopia is an investigation of

the moral state—its size, function, and possible origin. In the course
of the investigation, Nozick deals boldly with an astonishing range of
topics, including animals’ rights, how to get to utopia, Wilt Cham-
berlain’s rights, an analysis of envy, and what constitutes a just
distribution of goods. In treating the last topic Nozick sketches a
theory which may well turn out to be the most important part of the
book. He calls it the entitlement theory, and it is likely that social
philosophers to come will rank it with Marxism, various forms of
utilitarianism, and other classical views having to do with the just
distribution of goods in a state. Thus it may be of interest here to
note some problems with the entitlement theory as it stands,
propose some solutions, and examine the effect of such proposals on
the central argument in Nozick’s book.

THE ENTITLEMENT THEORY

Nozick presents the entitlement theory as a complex constructed

out of three more basic theories. These three theories, the theory of
just acquisition of holdings, the theory of just transfer of holdings,
and the theory of rectification, are not given us. At first blush this
344
 Nozick’s Entitlement Theory a45

would appear bizarre, but Nozick says enough about them to give us
an idea of the way they would function and the way the entitlement
theory would employ them, were they worked out. In fact, we are
able to make some interesting observations and objections with
respect to the hints we are given, and that is enough to ask of the
first version of any high-level philosophical theory.
To begin with, let us consider descriptions of the three theories
from which the entitlement theory is constructed:
The theory of just acquisition of holdings specifies under which
conditions a person is entitled to a holding taken from a state of
nature. Locke had some remarks to make on this topic. Nozick
refutes them, considers some revisions, and rejects these also. The
exact nature of this theory is left for future work.
The theory of just transfer of holdings specifies those means by
which individual’s entitlements to holdings may legitimately be
changed. Presumably, theft, extortion, swindle, and taxation would
be disallowed by thisstheory, whereas market transactions, charity,
and barter would be allowed.
The theory of rectification specifies a procedure through which
the effects of past injustices may be wiped out. More will be said
about this theory later.
The entitlement theory itself has two parts. The first part is called
the theory of justice in holdings. It specifies those cases in which an
individual is entitled to a holding. For the most part, it consists of
the following definition (taken from page 151. Clause 3, which was
withheld until later for expository purposes, is included):

(1) A person who acquires a holding in accordance with the principle of

justice in acquisition is entitled to that holding.
(2) A person who acquires a holding in accordance with the principle of
justice in transfer, from someone else entitled to the holding, is entitled
to the holding.
(3) A person who acquires a holding in accordance with the principle of
rectification of holdings is entitled to that holding.
(4) No one is entitled to a holding except by (repeated) applications of
(1)-(3).

This definition is elegant and appealing. In form it is an induction

with clauses 1 and 3 as base clauses and clause 2 as the induction
clause. But it seems to require modification.
Let us consider an incident taking place on a South Sea island. All
346 Lawrence Davis

of us on the island are entitled to our holdings until the night I go

out and steal my neighbor's pig. According to the definition above,
the rest of us are entitled to their holdings, but I am not entitled to
the pig (since theft is presumably not a legitimate means of transfer
and clause 2 is therefore unsatisfied). This conclusion seems unob-
jectionable. Suppose, however, that I trade the pig to someone in
another village for a chicken and that the trade is a fair one.
According to the definition above, I am entitled to the chicken,
since I have satisfied clause 2 by getting it in a fair trade from
someone who was entitled to it. The holder of the pig, however, is
not entitled to the pig, since I wasn't entitled to it before our trade
and clause 2 thereby goes unsatisfied. This result seems to be
counterintuitive. If one of us is to lack entitlement, it seems he
should be me. More likely both of us ought to lack entitlement to
our holdings. At least, a rectification would probably adjust the
holdings of both of us.
This problem seems to be most naturally solved by modifying the
second clause of the definition to read:

(2)') A person who acquires a holding in accordance with the

principle of justice in transfer, where all parties to the transfer
are entitled to their holdings, is entitled to the holding.
Of course, such a modification may result in the nonentitlement of
individuals to goods spreading plaguelike after an injustice. Such
outbreaks will require doctoring with heavy doses of the rectifica-
tion principle.
Now we are ready for the second part of the entitlement theory,
the principle of distributive justice. Nozick puts it in this way (151):

The complete principle of distributive justice would say simply that a

distribution is just if everyone is entitled to the holdings they possess under
the distribution.

But this definition seems unsatisfactory. Suppose, for example,

that I, a believer in the voleano god, became convinced that
propitiation of the god is required in order that our island remain
untroubled by eruptions of lava and ash. Suppose further that the
distribution of goods among us is just. One night I go through our
 Noxzick’s Entitlement Theory 347

village stealing all the goods I can without arousing my fellow

villagers. I load the goods in my cart and drag them to the crater’s
rim. At this moment the theory of justice in holdings tell us that
everyone except me is entitled to the holdings he has. The principle
of distributive justice tells us nothing about such a case, since the
principle is an implication rather than an equivalence, but it seems
likely that Nozick would want it to say that the distribution is unjust.
Now I tip the cart into the crater in a frenzy of vulcanistic
devotion and all is incinerated. The theory of justice in holdings tells
us that everyone is entitled to the holdings he possesses. Therefore
the principle of distributive justice tells us that the present distribu-
tion is just. This conclusion is not likely to appeal to those of my
fellow villagers whose property I purloined in the night. Nor is it
likely to appeal to the reader that, according to the theory, one
might change an unjust distribution into a just one merely by
destroying the holdings to which one was not entitled.
Nozick says something on page 151 which suggests a way out of
this difficulty:

A distribution is just if it arises from another just distribution by legitimate

means. . . whatever arises from a just situation by just steps is itself just.

What Nozick is (accurately) describing here is the transitivity of

justice under his recursive definition. But the vulcano-votive exam-
ple is problematic because the principle of distributive justice says
the wrong thing about “just” distributions arrived at in unjust ways.
It seems that the best solution to this difficulty is to recast the
principle of just distribution along the lines of the passage quoted
above, in the form of an inductive definition:
(1) The distribution in which nobody has anything and nobody
has ever had anything is a just distribution.
(2) A distribution arrived at in accordance with the principle of
rectification is a just distribution.
(3) A distribution derived from a just distribution in such a way
that nowhere in the derivation did people acquire goods to which
they were not entitled is a just distribution.
(4) No distribution is just except those derived through (re-
peated) applications of (1)-(3).
348 Lawrence Davis

Here clauses 1 and 2 are the base clauses and clause 3 is the
induction clause. The theory of justice in holdings is employed in
clause 3.
A principle of distributive justice based on this definition seems to
do what Nozick wants done. Now that we have it in a more workable
form, we need to look closely at one of its most important parts, the
principle of rectification. Here is what Nozick says about it:

Idealizing greatly, let us suppose theoretical investigation will produce a

principle of rectification. This principle uses historical information about
previous situations and injustices done in them. . . and information about
the actual course of events that flowed from these injustices, until the
present, and it yields a description (or descriptions) of holdings in the
society. The principle of rectification presumably will make use of its best
estimate of subjunctive information about what would have occurred (or a
probability distribution over what might have occurred, using the expected
value) if the injustice had not taken place. If the actual description of
holdings turns out not to be one ofthe descriptions yielded by the principle,
then one of the descriptions yielded must be realized (152/3).

Aside from this passage we are not told a great deal about the
rectification principle, and this is unfortunate because it seems to be
the most problematic part of the entitlement theory. It is certainly
an essential part; for, without it, owing to the inductive nature of the
definition of entitlement, if there has been a single injustice in the
history of a state, no matter how far back, the state will not be able
to achieve ajust distribution of goods in the present. Let us consider
the principle’s weaknesses and attempt a reformulation.
The principle of rectification tells us to follow a procedure some-
thing like this: Given an unjust distribution, go back to the last just
distribution. Pretend that our lives are taking place in a movie
projected on a screen with many, many projectors pointed at it.
Project the movie we are in until a frame containing an injustice is
projected. At that frame, turn off the present projector and turn on
the projector containing the most likely movie of our lives up to this
point, with the injustice written out of the script. Run that pro-
jector, starting from the frame at which the first movie was stopped.
Continue projecting until a frame containing an injustice is pro-
jected, or until a frame depicting the present is projected. Ifaframe
containing an injustice is projected, stop the projector and turn on
 Nozick’s Entitlement Theory 349

the projector containing the most likely movie of our lives up to that
point, with the injustice written out of the script. Run this projector
from the frame at which the other was stopped. Continue in this way
until a frame depicting the present is projected. Stop the projector at
that frame, make a list of who has what, and realize the distribution
of goods in that list. Now a just distribution has been arrived at and
the effects of all previous injustices have been wiped out.
This theory, with its promise of being able to wipe the slate clean
of past injustices, is appealing to those of us who feel that our
present distribution of goods may not be completely just. But, as the
theory is stated, the possibility of correcting past injustices is ruled
out for us, and a good many everyday situations in which we would
normally say that injustices have been corrected will not count as
rectifications. This result comes about because, according to the
theory, injustices may be rectified only by realizing the distribution
of holdings in the present frame of the rectified movie of our lives. It
follows from this that if some member of the set of goods in that
distribution does not exist, no rectification is possible.
A trivial example of this would be my stealing a banana from a
neighbor and eating it. As the theory stands, rectification cannot be
brought about, for the rectified movie of our lives would include the
(eaten) banana as a part of my neighbor's holdings, and we cannot
realize that distribution. I could give my neighbor a thousand
bananas, but the theory would not rule such restitution rectification.
This problem will become acute in cases concerning an injustice
that has drastically altered the subsequent flow of events. For
example, suppose that my clan becomes disgruntled at the unen-
lightened agricultural practices employed on our island and the
measly benefits we gain from them. One day we take over the
communal farmlands by force and begin to raise bananas. After
several years the banana boats come and we make a fortune for
everybody by trading bananas for medical supplies, qualified
teachers, books, etc. We then turn the fields back to our fellow
islanders and everybody is happy raising bananas.
It seems that Nozick would approve of our act on balance. We
have displayed the sort of initiative and intelligent employment of
resources that he applauds (except possibly for our returning of the
fields). But as the theory stands, there is no way we may rectify the
injustice of our appropriation of the farmlands, for without the
350 Lawrence Davis

injustice we would have continued to raise our crops of spindly

coffee bushes, trading what wretched harvests we made for whiskey
and beads from the coffee boat. Because of our agronomical savoir
faire everybody is much happier than before, and everybody prefers
the way things are to the way they would have been without the
injustice. But there is very little of the rectification set of holdings
on the island with which to bring about a rectification, and the
entitlement theory tells us that the situation cannot be righted.
Nozick might leave the rectification principle as it is and swallow
such consequences, but this does not seem to be consistent with his
theory of compensation given in Part I of his book. There a person
who has been transgressed against is raised to at least as high an
indifference curve as he would have been on had the transgression
not taken place, and this constitutes compensation for the transgres-
sion. Note that this theory says nothing about the identical goods
which the person would have had if untransgressed against; instead
the relevant notion is that of being indifferent between the state
with no transgression occurring and a compensated present state. If
Nozick is to be consistent it appears that he will use the notion of
indifference here too.
These considerations suggest the following restatement of the
rectification principle:

To achieve a rectification, go back to the time of the last just distribution

and project events forward in the most likely way without the injustices.
Figure out what the present set of holdings would be in the projection so
derived. Create a rectification distribution by giving everybody at least
enough goods to make them indifferent between their holdings in the
projected distribution and the rectification distribution.

This principle seems to avoid the difficulties of the last, and it is

faithful to the spirit of earlier passages in the book. But before going
on we ought to note three things about it. First, it will not be
applicable in any case in which there are not enough goods available
at the time of the rectification to put everyone on the appropriate
indifference curve, but that may be all right. Perhaps, if we can’t get
things to be as good as they would have been without injustice, then
there really is no way to rectify affairs.
Second, the rectification principle in this form allows for many
 Nozick’s Entitlement Theory 351

different rectification distributions whenever there are more than

enough goods available to raise everyone to the appropriate indif-
ference curve. Let us assume that the theory includes some way to
rank these distributions so that only one of the possibilities is the
rectification distribution.
Finally, we ought to note that, in both its original and its
amended forms, the principle of rectification will probably not help
us if we attempt a full-scale rectification of the injustices in our
societys past, for were we to project the 200 years of our country’s
history in a rectified movie, the cast of characters would surely differ
significantly from the existing cast. Had our ancestors lived and
moved in a rectified version of our history, quite likely many of us
would not be alive today.
The theory, then, seems to be satisfactory only for cases in which
the existing cast of characters is identical with the cast of characters
in the present frame of a rectified movie. If the existing cast contains
a member who does not appear in the rectified movie, the theory
would seem to assign nothing to that member. This is surely an
unfair consequence of the theory. And of course, if the cast contains
members who do not appear in actuality, we are not told how to
distribute their goods.
We assume for the sake of subsequent discussion that this sort of
problem will be cleared up, and we now move to more general
comments on the theory.

THE ENTITLEMENT THEORY AND ITS RIVALS

We have discussed two versions of the entitlement theory: Nozick’s

and an amended form. Nozick’s version of the theory works well in
the context of his book, but the amended version proposed here
does not do nearly so well. The reader may know that in Part I of
Anarchy, State, and Utopia Nozick argues that an ultraminimal
state may arise in a morally acceptable way. Part IH is devoted to
upholding the claim that a more-than-minimal state is not morally
acceptable. The entitlement theory is produced in order to counter
the claim that a more-than-minimal state is justified in order to
achieve and maintain a just distribution of goods. Nozick’s counter
consists of the argument that the other theories of just distribution
ooo Lawrence Davis

of goods are incorrect but the entitlement theory is not, and he

implies that an ultraminimal state will be able to implement the
entitlement theory.
The dominant theme in Nozick’s attack on opposing theories of
just distribution of goods is that the implementation of the other
theories (to each according to his need, for example, or various
versions of utilitarianism) will require wresting property from
people entitled to it in order to give it to people who have done
nothing to earn it. Nozick stresses the point that only the entitle-
ment theory recognizes the fact that goods come encumbered with
entitlements which ought to be preserved.
This observation is correct, given Nozick’s formulation of the
entitlement theory. But, when the theory is amended as suggested
above, the difference in this respect between it and its rivals is not
so pronounced. For example, let us suppose that we discover two
sailors who have been shipwrecked on our island for the last twenty
years. At the time of the shipwreck Jones tucked his machete into
his belt before leaping into the sea, swam to shore, and passed out.
Smith took nothing with him when he swam to shore, but he
compensated for his lack of foresight by appropriating Jones's
machete while Jones slept. For twenty years Smith has used the
machete to construct a large hut, irrigate gardens, build traps and
enclosures for wild animals, and furnish his hut in luxurious style. If
he had not stolen the machete he would have done little, but the
machete has become a symbol of civilization for him, spurring him
into hewing a stylish existence from the jungle.
Jones, however, awakened and assumed his machete had been
lost in the surf. For twenty years, without the tool which Smith has
used to such effect, Jones has lived an idle life. As time has passed
Jones has degenerated into a brutish drinker of palm wine, although
he would have achieved as much as Smith in the way of material
comfort had he kept his machete.
There is an injustice to be rectified here, for theft is presumably
not a just means of transfer of holdings. According to the entitle-
ment theory, we rectify affairs by projecting events forward without
the injustice. We see that the roles of the two castaways would have
been reversed had the original theft not occurred. Hence we rectify
the injustice by giving all of Smith’s goods to the drunkard Jones and
 Nozick's Entitlement Theory Bo0

giving Jones's few possessions to Smith. Smith’s protests that his

goods were rightfully taken from a state of nature at great expense of
time and labor will count for nothing, nor will his protests that Jones
will allow the gardens to become overrun, the herds to run away,
and the hut to fall to ruin.
It may come about, then, that employment of the rectification
principle, even in its original form, will result in goods being taken
from individuals who appear to be entitled to them. Further, when
we consider the operation of the theory as amended in this paper,
we see that it becomes as much an indifference as an entitlement
theory; for, on the amended theory, it is on indifference curves
rather than on historical entitlements that rectification distributions
are based. The effect of the amendments suggested here, then, is to
eliminate a good deal of the entitlement from the entitlement
theory, thereby weakening Nozick’s arguments against the opposi-
tion.
It appears that Nozick is presented with a dilemma. If the
entitlement theory remains as he has presented it, then a good
many small injustices will be declared unrectifiable, merely because
some member of the set of goods in the rectification distribution
happens not to exist. On the other hand, if the tack proposed here is
taken and the rectification principle is formulated in terms of
indifference curves, then our intuitions about correcting injustices
will be upheld, but the theory will cease to preserve entitlements to
particular objects.
Nozick has given us a new and ingenious theory ofjust distribu-
tions. His attempt to clear away the opposition appears to have
failed, but he has articulated another point of view from which to
assess the justness of the patterns of holdings around us. Let us
conclude by noting the striking differences between the entitlement
theory and two of its rivals. In the example of the shipwrecked
sailors, a Marxist, distributing to each of the two members of the
society according to his needs, would divide the goods evenly
between Smith and Jones and declare the result a just distribution.
Some utilitarians, noting that Smith will derive much more utility
from the goods than Jones, would leave things much as they are and
declare the result a just distribution. An entitlement theorist would
reverse the holdings of the two castaways and declare the result a
354 Lawrence Davis

just distribution. These three theories, one looking to the present,

one to the future, and one to the past, give us quite different
versions of what the correct apportionment of holdings ought to be.
Which of them, if any, is correct? In the wake of Nozick’s book, this
question is likely to attract a good deal of attention.
*This paper owes a good deal to the comments and criticisms of Professor
Robert Paul Wolff.
 18
The New Indian Claims
and Original Rights to
Land
DAvID LYONS

THE NEW INDIAN CLAIMS

Most Americans take this country’s possession of its territory for

granted, even though we all know that a great deal of its land was
wrested by force or fraud from those who occupied it before the
Europeans came—from Native Americans, who were dispossessed
and either massacred or subjugated, the survivors displaced from
their homelands and in large part consigned to live on shrinking
reservations. The monumental theft of land that was involved in the
European conquest of America is regarded as a neutral fact about
the past with little, if any, practical bearing on the present.
It is that attitude that I ask you now to suspend, if you have not
done so already. I assume here not only that the socially weak and
disadvantaged condition of Native Americans in our society repre-
sents a wrong it is incumbent on us to right, but also that their
dispossession may call for significant rectification. I shall concen-
trate on the latter issue because the serious possibility of radical
social surgery to correct it must now be contemplated. The unthink-
able idea of giving the land back to the Indians has suddenly become
thinkable.

355
356 David Lyons

Early in its constitutional career, in 1790, Congress passed the

Indian Nonintercourse Act, which requires that all transfers of lands
from Indians to others be approved by the federal government. The
Act was modified from time to time over the next forty-odd years,
but it was not changed in any relevant respect, and it remains in
effect today. Its purpose is clear. It was meant to guarantee security
to Native Americans against fraudulent acquisition by others of the
Indians’ allotments of land. Such guarantees were plainly needed.
By 1790, expropriation had been practiced by Europeans for nearly
two centuries. Fraudulent land acquisitions by colonists had been a
source of friction between them and the British government, which
occasionally leaned towards protecting Native Americans. Security
for Indian land was an important bargaining point during the
Revolutionary War, when Indian support or at least neutrality was
desparately needed by the rebellious colonists. The Nonintercourse
Act of 1790 pledged federal security for Indian land holdings. Under
it, the federal government is bound to act as guardian or trustee,
overseeing all transfers of Indian lands, including those to states and
other branches of government as well as to private parties.
Several suits that have recently been initiated by American
Indian tribes for recovery of lands held by them when the Noninter-
course Act took effect in 1790 invoke this law. It is alleged that
certain transactions by which lands were subsequently lost to them
are invalid because federal approval was neither sought nor ob-
tained in those cases. Those historical facts have not been contested.
A great deal is at stake. Some of the suits concern hundreds,
others thousands, of acres. The largest tract, claimed by the
Penobscot and Passamaquoddy tribes, amounts to twelve and a half
million acres in Maine, comprises more than half the state, and has a
value estimated at twenty-five billion dollars.?
In some cases, recovery of the lands is being effected smoothly, as
in Gay Head, on Martha’s Vineyard, off Cape Cod, where voters
have approved the transfer of about two hundred fifty acres back to
the Wampanoag. But that is an atypical case, since the land is
undeveloped and has always been reserved for public use, which is
what the Gay Head branch of the Wampanoag wish to secure it for,
and about half the voting residents are Wampanoag descendants.
More typical, perhaps, is the claim in Mashpee, on Cape Cod,
where another branch of the Wampanoag is seeking to recover
 New Indian Claims and Rights to Land 3O7

jurisdiction over about seventeen thousand acres, comprising most

of the town of Mashpee, in an area that is currently undergoing
rapid commercial development. That suit has thrown a cloud over
land titles, freezing real estate transactions and development. Since
Indian lands are not subject to local taxation, it has also disrupted thé
sale of municipal bonds. The professed aim of the claimants is
reportedly not to dispossess current homeowners or active busi-
nesses. A prime objective is to regain lost hunting and fishing rights;
another is to reassert control over the portion of the land that
remains undeveloped, in order to inhibit such development. Almost
all of the undeveloped land at issue in these suits, however, has
passed into private hands. Consider the situation in Maine. Al-
though the claims there embrace populated areas, including whole
cities and towns, they chiefly concern, and the tribes appear chiefly
interested in, vast tracts that are not only undeveloped but are,
unlike the Mashpee claim, unlikely to undergo any ordinary com-
mercial development. For these are mainly huge forest reserves
that are owned by paper companies and related interests.
These lands will not be returned to the Indians without a bitter
struggle. At first the claims were not taken seriously. But once they
began to receive favorable attention in the federal courts, current
owners—especially the large landed interests in Maine—began to
mobilize a political campaign against recovery by the Indians.
Their concern and the subsequent political maneuvering is un-
derstandable. Federal court decisions have affirmed federal respon-
sibility in such cases, whether or not the federal government has
officially recognized the tribes in question,? and they have dismis-
sed as inapplicable the various standard defenses, such as adverse
possession (which invokes a statute of limitations on claims), laches
(which invokes a similar doctrine in the law of equity), and estoppel
by sale (which would use the prior transactions as a bar to re-
covery).4 The law seems clear: any title to Indian land that has been
obtained without explicit federal approval is null and void.
It does not follow—either legally or morally—that all of the land
in question must be returned to the Indians. But an observer might
well suppose that some, at least, of the lands should be restored to
them. I wish to examine that idea, not only to help us in determin-
ing what justice requires, but also to evaluate some lines of reason-
ing in support of and in opposition to it. For the most natural
358 David Lyons

arguments that might initially be advanced on both sides of the

issue—arguments that appear to be implicit in the rhetoric already
surrounding these cases—center on what we, following Robert
Nozick,® might call “historical” considerations affecting social jus-
tice. These cases give us an opportunity to scrutinize Nozick’s
conception of justice and, more generally, the idea of a right to
property.

THE HISTORICAL ENTITLEMENT ARGUMENTS

Suppose that one is aware of the current plight of Native Americans

and its background. This might understandably make one sym-
pathetic to the new land claims. But it is incumbent on one to ask:
What are the moral foundations of such claims? How can they be
defended, not so much in a court of law as in the court of con-
science?
One natural (one might almost say naive) way of reasoning about
the claims is this. Native Americans were the first human occupants
of this land. Before the European invasion of America, the land
belonged to them. In the course of that invasion and its aftermath,
the land was illicitly taken from them. The rightful owners of the
land were dispossessed. The current owners lack a well-founded
right to the land, which now lies illicitly in their hands. Ideally, the
land should be restored to its rightful owners. This may be impracti-
cal; compromises might have to be made. But the original wrong can
most easily be righted by returning the land to them—or by
returning it wherever that is feasible.
This sort of argument turns upon the idea of original acquisition
and, somewhat less directly, on the idea of legitimate transfer.
Without original acquisition by the Indians, they might have had no
rights to the land that dispossession was capable of violating. The
argument concerns legitimate transfer by claiming that the transfers
by which the Indians lost the lands were illegitimate.
The argument also assumes that rectification in these cases is, at
least in principle, most straightforward. Injustice is corrected,
justice is done, by restoring the land to its original and still rightful
owners. That is a most important feature of the argument. For, if
correct, it means that we can avoid getting bogged down in the
 New Indian Claims and Rights to Land 359

uncharted territory of compensatory or reparative justice. Without

that assumption, the problem has no easy solution, even in theory.
A frequent reply to the current claims is that one cannot simply
ignore two hundred years of history. Those who now possess the
land did not, in fact, secure it illicitly. They obtained it from others,
by purchase, gift, and inheritance. To find illicit land transfers, we
would have to trace the chain of transactions back several genera-
tions. But all of that is, by now, dead history. Somewhere along the
line, custom and settled expectation generated new rights to the
land. And it is these rights, not some long extinguished “original”
right, that must now be enforced.
This line of reasoning has much in common with the usual
defenses against claims whose foundations go far back into the past.
The defense of adverse possession, for example, rests on the rule
that one cannot validly reclaim property after a certain period of
time has passed, during which one has registered no complaint
about another's misappropriation. Such a defense has been ruled
inapplicable to the current cases, as we have seen. But that is not
decisive here, when we are concerned with the moral foundations
for such claims; it might simply amount to a legal technicality, which
has only limited implications about what is right and what is wrong.
If adverse possession is in general a reasonable defense against such
claims, it might be legally inapplicable here only because of a
strange quirk in the law. But I shall not pursue that complex matter
now. Any defense against such claims is also likely to rely upon
inheritance, and this issue will be more central to our concerns in
this paper.
In what follows, my purpose will be to challenge these naive
arguments, on both sides of the issue, by throwing doubt upon
property rights as we usually think of them. I shall suggest that
property rights, including rights to land, are thinner and much more
flexible, or variable with circumstances, than these arguments
allow. If that is so, our whole way of looking at such matters may
require radical revision.
Let me relate this now to Nozick’s theory of justice. We are
dealing here with property, on which Nozick concentrates, and the
particular issues in the case correspond to major aspects of his
theory—acquisition and transfer. The two main elements of
360 David Lyons

Nozick’s theory are what he calls “the principle of justice in acquisi-

tion” and “the principle ofjustice in transfer.” The former concerns
the circumstances in which one can acquire rights to things by
appropriating them. The latter concerns the ways in which one can
receive rights from others, such as gifts, inheritance, and exchange.
Many textbook theories of justice ignore historical factors, espe-
cially those involving voluntary transfer. Some theories imply, for
example, that I possess a thing unjustly unless I can be said to merit
or deserve it. Nozick reminds us, however, that justice does not
frown upon gifts and favors, charity and generosity, and fair bar-
gains. One can obtain things by such means without meriting or
deserving them—and yet without any injustice being done.
Historical factors are thus relevant to justice, and any adequate
theory must accommodate them. It does not follow (nor is it true)
that Nozick’s theory alone is capable of accommodating them. In
fact, Nozick goes to the opposite extreme, exaggerating the role of
historical considerations.
Nozick defends the following thesis: to establish the moral foun-
dations of one’s ownership of a thing, it may not be necessary to
show that one’s ownership fits into some preferred social pattern,
such as equality. It may suffice to show that one obtained the thing
in accordance with the principle of justice in acquisition or the
principle of justice in transfer. This is significant because voluntary
transfers can upset preexisting patterns of distribution, such as
equality. If the results of such transfers are unobjectionable, then
patterns cannot exhaust the important truths about social justice.
Nozick’s examples tend to show that historical factors are relevant
to matters of justice, independently of other factors. That is an
important point; but Nozick tries to stretch it further. From the
claim that historical factors are relevant he seems to infer that they
are the only factors relevant to justice, that all other considerations
are irrelevant, such as merit and desert and the relative distribution
of benefits and burdens in society. But the latter simply does not
follow from the former, and it may very well be false.
In this paper I shall argue that Nozick’s theory incorporates
another exaggeration—the notion that property rights, once legiti-
mately acquired, are virtually unaffected by circumstances. I shall
then show how this undermines the historical entitlement argu-
 New Indian Claims and Rights to Land 361

ments concerning Indian land claims. But I argue, finally, that it

does not defeat the current claims.

ORIGINAL ACQUISITION

Let us first consider the idea of original rights to land. How are such
rights to be understood? Locke is one of the few writers to discuss
the subject, so it seems reasonable to begin with his view of it.®
Locke says that one acquires property, originally, by “mixing
one’s labor” with an unowned thing, or something that belongs to all
humanity in common. ($27) Locke clearly means us to take this
notion of “mixing one’s labor” with a thing very loosely—to cover,
for example, one’s picking up an acorn with a view to eating it. ($28)
But, as Nozick observes,’ there are problems with this notion. The
limits of what I can acquire in this way are radically indeterminate.
If no one yet owns them, can I make the oceans my own property by
simply stirring water at the shore? More fundamentally, it is not
clear why mixing my labor with a thing that I do not own is a way of
acquiring that thing rather than a way of losing my labor.
When Locke applies his general theory to the acquisition of land
he obtains a doctrine that is at least much clearer. He says that one
must cultivate the soil, make it productive agriculturally, and be
able to consume its products. ($32) Mixing one’s labor with a parcel
of land in this way removes it from the common stock of land that
has been provided for all humanity and gives one original title to it.
This cultivation test seems natural enough—so long as we assume
that cultivation is the only proper way of using land. But a moment's
reflection reminds us that, even for the purpose of obtaining food,
land can effectively be used in other ways—hunting, gathering, and
herding, for example. And, of course, land can be used in ways
unrelated to food production. Locke was aware of this. How, then,
did he justify his narrow cultivation test?
His reasoning is suggested by the following passage which Locke
had added to the collected edition of his works:

[H]e who appropriates land to himself by his labor does not lessen but
increase the common stock of mankind; for the provisions serving to the
support of human life produced by one acre of enclosed and cultivated land
are—to speak much within compass—ten times more than those which are
362 David Lyons

yielded by an acre of land of equal richness lying waste in common. And

therefore he that encloses land, and has a greater plenty of the conveni-
ences of life from. ten acres than he could have from a hundred left to
nature, may truly be said to give ninety acres to mankind; for his labor now
supplies him with provisions out of ten acres which were by the product ofa
hundred lying in common. ($37)

Locke maintains, moreover, that “there is land enough in the world

to suffice double the inhabitants,” provided it be cultivated. (§36)
By his reckoning, there is not a fifth enough land for all the people
of the world if none of it is cultivated. The argument therefore
seems to be that, by enclosing and cultivating land, one actually
performs a service to others. If I appropriate and cultivate only so
much land as I require for my own subsistence, I thereby release
land to others that I would have needed to support myself by hunt-
ing on it or gathering food from it. Cultivation involves the far more
efficient use of land, from which all benefit. Most important, the
failure to cultivate suitable land results in privation for a corres-
ponding number of people.
Nozick appears to reject Locke’s theory about the original ac-
quisition of land. The argument suggests a utilitarian rationale for
property rights, which Nozick regards as unacceptable. And, even
within its utilitarian framework, the argument is not very promising
as a way of showing that cultivation is the only legitimate basis for
acquiring unowned land. The empirical premises of the argument
are dubious, and alternative grounds of appropriation seem possi-
ble. More importantly, Nozick, as we have seen, raises skeptical
doubts about Locke's general theory of property acquisition.
Nozick’s alternative account appears to be that appropriation
alone is sufficient for the legitimate acquisition of property, pro-
vided that a certain condition is satisfied. Nozick calls this the
“Lockean proviso”; this is a requirement Locke included in his own
theory, that “enough and as good” of whatever is being appropriated
be left for others. (§27) If enough and as good land is left for others
when one appropriates land, the appropriation is legitimate, justice
is done, and one acquires a right to the land. Otherwise the
appropriation is illegitimate and one acquires no right to the land.
Let us suppose, for the sake of argument, that some sort of right
to land can be established, and also that Native Americans estab-
lished original rights to land in the Americas before the European
 New Indian Claims and Rights to Land 363

invasion, land that was later taken illicitly by some of the invaders.
The question we must face is, what difference that can make today.
The argument invoking original rights to land in support of the
current Indian land claims assumes that original rights are very
stable. They are unaffected by changes in circumstances, because
they are still valid today despite the passage of history. They are also
largely independent of the institutions that are internal to a society.
That is to say, they do not merely regulate relations within a
community, but also relations between the community and the
outside world. For these original rights are supposed to set limits on
the conduct of persons outside Indian society; they are supposed to
be valid claims relative to nonassimilating Europeans.
It may be useful here to distinguish between two different kinds
of moral rights to property. There may be morally defensible
property rights within a given social system, taking into account the
laws and other social rules governing property and the general
circumstances of the community. The moral justification of claims
couched in terms of such rights makes essential reference to social
rules and circumstances. For the reasons just given, it is doubtful
that original rights to land can be of this type. At least, the
arguments invoking them ignore radical shifts in circumstances and
fail to explain how ancient Indian institutions have a direct bearing
on current claims to land. The original rights to land that are
invoked would seem to be strictly nonconventional and inherently
stable rights, which are not relative to social rules or circumstances.
I shall argue here that moral rights to property are not so stable.

INHERITANCE

I can best begin by considering the inheritability of rights. This is an

appropriate point of departure because Nozick takes for granted that
property rights are inheritable, and inheritance would seem to be
involved in the argument concerning the current claims. After
discussing inheritance, I shall go on to consider other common
features of property rights.
Inheritance may appear to be a factor in the naive argument
supporting the Indian claims because the current members of the
tribes that are suing for recovery of lands once held by them are not
the same individuals who belonged to those tribes when the land
364 David Lyons

was acquired or when the contested transfers were made. The

current members are descendents of those individuals. If the cur-
rent members have valid claims to the land, claims deriving from
their ancestors’ original rights, it would seem that those rights must
have been passed down through the generations by means of
inheritance.
That is, interestingly, not the case. The current claims are being
made, not on behalf of individual Native Americans who are sup-
posed to have inherited ordinary property rights in land (or shares in
a land-owning company) from their ancestors, but rather on behalf
of entire tribes collectively. The land was originally held in common
by the tribes, and that is how the land would be recovered. A tribe
is a continuously existing entity, like a nation, that spans the
relevant generations of human beings. Its ownership of land is like
the possession by a nation of its territory. Its ownership need not be
thought of as involving transfer from one individual to another by
inheritance or any other means.
So inheritance would not seem to be an element in the Indian
claims based on original rights. That gives us one way of distin-
guishing between the opposing arguments—for inheritance is al-
most certainly an element in the arguments on the other side. Most
of the land in question has long been in private hands. It is virtually
certain that inheritance has been involved in transfers from one
generation to another since the contested transfers took place. It is,
admittedly, conceivable that land (or shares in a land-owning com-
pany) should be transferred from one individual to another over an
indefinite period without inheritance entering the picture. But it is
extremely unlikely in the current cases. So, ifinheritance is suspect,
then some of the objections to the current claims—those based upon
inheritance—are suspect too.
I wish to throw some moral doubt on claims based upon inheri-
tance; but that is a secondary purpose. What I mainly wish to show
is that inheritability cannot be an essential feature of moral rights to
property. This is because the moral acceptability of inheritance is
relative to circumstances. And I wish to do this, not by rejecting
historical factors affecting justice, but rather by developing them
beyond the point at which Nozick stopped. My argument will not be
that inheritance is morally objectionable from a nonhistorical point
 New Indian Claims and Rights to Land 365

of view (that may well be true, but will not be considered here). My
argument will be that inheritance can undermine justice in transfer,
and thus can be objectionable from an historical point of view.
Nozick’s idea is that transfers are legitimate and their outcomes
are consequently morally unobjectionable when they are voluntary
and do not violate the Lockean proviso. The theoretical model for
this idea has been provided by John Rawls, who distinguishes
between “perfect” and “imperfect” procedural justice, on the one
hand, and “pure” procedural justice on the other.® Perfect and
imperfect procedural justice are virtues of transactions based upon
the real or likely outcomes of the transactions. Pure procedural
justice is a virtue of transactions that derives from the character of
the processes themselves. Thus, the results of a lottery are morally
unobjectionable when the lottery itself has certain characteristics
and is consequently fair. Bargains and agreements can be judged in
this way too. Their outcomes are morally unobjectionable when the
bargains and agreements themselves are fair. Iam not sure whether
the applicable notion of fairness is captured by Nozick’s require-
ment that transfers be voluntary and not violate the Lockean
proviso; but I doubt it. At any rate, bargains and agreements are not
fair unless fraud as well as force is absent and the parties are
(roughly speaking) equal and informed as well as free.
Now, one trouble with inheritance is that it often promotes
concentrations of wealth and power. This is not an egalitarian
objection. My point is, rather, that concentrated wealth and power
is able to impose its will on smaller and weaker parties, thus creating
bargains, agreements, exchanges, and social arrangements generally
that are unfair. Extremes of power undermine the legitimacy of
social processes, and the outcomes cannot be assumed to be morally
unobjectionable. (If they are unobjectionable that will be so by
virtue of nonhistorical considerations.) In such circumstances, in-
heritance promotes injustice in transfer. Embedding inheritance
into property rights would therefore create internal difficulties for
historical principles.
I do not mean to suggest that inheritance is intrinsically objec-
tionable. The effects to which I refer are clearly relative to social
conditions. Inheritance will have such consequences in some cir-
cumstances and not in others. Specific rules governing inheritance
366 David Lyons

may thus be justified in some circumstances. But it is implausible to

suppose that inheritance is morally fundamental. Specific rules
governing inheritance could not be incorporated into a basic princi-
ple ofjustice in transfer.
Nozick appears to assume that inheritance is an indispensable
feature of private property. He discusses original acquisition in such
terms, as if one could not acquire property that was not permanent
and inheritable. The alternative possibility is never considered. But
that is surely a mistake. The idea of a right to property does not
entail that it be inheritable. That is simply one possible form that
property rights can take.
And, of course, inheritance as we know it is a straightforwardly
conventional arrangement, a certain type of economic institution.
Its moral justification is the justification of an institution, which
must take social circumstances into account. An heir can morally
defend his claim to some (conventionally) inherited thing only by
appealing to the rules of such an institution. His claim is morally
successful, so to speak, only if the institution itself is morally
defensible. Nozick does not seem to look at inheritance in this way.
He seems to assume that an heir could defend his claim to some
inherited thing without making any reference to laws and other
social rules. He appears to assume, more generally, that the morally
supportable property rights that we have correspond precisely to
moral rights that do not presuppose any laws or other social rules.
That is another mistake, based perhaps on a failure to distinguish
between morally defensible property rights within a given social
system, the justification of which is relative to social rules and
circumstances, and moral rights to property that are not relative to
social rules or circumstances.
We <an conclude, then, that moral rights to property are not
necessarily imheritable. We should accordingly observe (although
this seems to have no direct bearing on the current Indian claims,
for reasons already noted) that original rights to land or other
property cannot be assumed to be inheritable. Furthermore, the
given defenses against the current Indian claims are suspect so far as
they rely upon inheritance. For our society contains concentrated
wealth and power, which does impose its will on others, and
inheritance contributes to those conditions.® Property rights af-
fected by inheritance are thus subject to further moral scrutiny.
 New Indian Claims and Rights to Land 367

THE LOCKEAN PROVISO

What I wish to suggest now is that other typical features of property

rights as we know them are morally defensible only relative to
circumstances and therefore cannot be assumed parts of stable
rights to land or other property, and cannot be morally fundamen-
tal.
Nozick himself, surprisingly, appears to admit as much. The
contents, at least, of aproperty right are relative to circumstances.
How I may transfer or use, or deal with others’ use of, a thing that I
have previously acquired depends on how my behavior would affect
others. Nozick develops this point in terms of the Lockean proviso,
which was introduced to regulate original acquisition and now is
extended to regulate property rights much more extensively.
It will be recalled that Locke’s proviso was that “enough and as
good” of whatever is being appropriated be left for others. Nozick
does not regard this formulation as satisfactory, perhaps because. it
does not cover certain sorts of cases. “The crucial point,” he says, “is
whether appropriation of an unowned object worsens the situation
of others.”1° That is the sufficient condition, Nozick thinks, to place
upon legitimate appropriation. If one does not worsen the situation
of others, one acquires the right to a thing one has appropriated.
But, in Nozick’s view, the Lockean proviso has an “historical
shadow,”!! making arguments like the following possible. Suppose
that I am landless in a world with much land that has already been
appropriated but little that remains unowned. Suppose further that
I cannot appropriate land for my own use and enjoyment without
leaving enough and as good for others, that is, without worsening
others’ situation, because there is simply not that much to go
around. If that were so, then it might be argued that the last persons
to appropriate land worsened the situation of others, me in particu-
lar, and by similar reasoning that all prior acts of land appropriation
are now illicit because they worsened the situation of others. For
scarce resources, such reasoning might seem to show that private
acquisition is simply illegitimate. 1”
Nozick wishes to meet this objection apparently because he
wishes to defend private acquisition, even of scarce resources. He
seems so preoccupied with that objective that he fails to draw
attention to the odd, retrospective character of these lines of
368 David Lyons

reasoning. The objection is supposed to show that past acts of

appropriation are now to be regarded as illicit because acts of
appropriation now would violate the Lockean proviso. But the
problem now might result entirely from changes in circumstances,
such as a decrease in the amount of usable land or an increase in the
population. In such a case, the prior acts of appropriation could not
plausibly be said to have violated the Lockean proviso.
Nozick’s treatment of the “historical shadow” of the Lockean
proviso, here and elsewhere, makes clear that his concern is not just
with acquisition but more generally with the effects of continuing
private ownership on others. It is not that acquiring the land
violated the Lockean proviso but that keeping it appears to do so.
Even if the original acquisition was perfectly legitimate, retention of
the property might worsen others’ situation in the same kind of way
that the Lockean proviso is intended to prevent. The underlying
idea is that property arrangements must accommodate the basic
needs and interests (Nozick would probably say the rights) of others.
Nozick them seems to reason as follows. The argument against
original acquisition or retention of scarce resources goes through
only if we assume that property rights entail the right to exclusive
use and enjoyment. But others’ situation is not worsened by appro-
priation or retention if that assumption is rejected. In order to
protect his assumption that property rights should be permanent
and bequeathable, Nozick is prepared to modify the contents of
such rights, or rather to make their contents variable with cir-
cumstances. This comes out most clearly in the following passage:

Each owner's title to his holding includes the historical shadow of the
Lockean proviso in appropriation. This excludes his transferring it into an
agglomeration that does violate the Lockean proviso, and excludes his using
it in a way, in coordination with others or independently of them, so as to
violate the proviso by making the situation of others worse than their
baseline situation. Once it is known that someone’s ownership runs afoul of
the Lockean proviso, there are stringent limits on what he may do with
(what it is difficult any longer unreservedly to call) “his property.” Thus a
person may not appropriate the only water hole in a desert and charge what
he will. Nor may he charge what he will if he possesses one, and unfortu-
nately it chances that all the water holes in the desert dry up, except for his.
This unfortunate circumstance, admittedly no fault of his, brings into
operation the Lockean proviso and limits his property rights. Similarly, an
 New Indian Claims and Rights to Land 369

owners property right in the only island in the area does not allow him to
order a castaway from a shipwreck off his island as a trespasser, for this
would violate the Lockean proviso.!%

But my objection to inheritance was more far-reaching. Nozick

wishes to retain rights and let them be passed down through
inheritance by eliminating exclusive use and restricting other forms
of transfer. This segregation of inheritance from other forms of
transfer seems arbitrary: if the other forms of transfer can be
restricted, then so can inheritance. Inheritance is not an essential
feature of property rights. Similar reasoning shows the same thing
about other normal features of property rights, such as transfer by
gift or exchange and exclusive use and enjoyment: these are not
essential features of property rights. Or, if they are essential features
of property rights, then property rights are inherently unstable.
One cannot have it both ways—not, at least, if one thinks that
justice must accommodate the basic needs and interests of others.

THE INSTABILITY OF PROPERTY RIGHTS

Now I wish to suggest that property rights may be even more

unstable than has so far been argued. Let us expand on an example
that Nozick uses. Suppose that we are occupants of an isolated
island. We have arranged to use the land and all its other resources
among ourselves, and we live comfortably, with some less perisha-
ble goods set aside for rainless seasons. One day, a party of
castaways from a shipwreck are washed up on our shore. They are
uninvited but also involuntary guests. There is no prospect for their
safe removal, and they have no resources beyond their capacity to
work. But they are also unaggressive. What are we to do? Nozick
would agree that we may not drive them back into the sea just
because they come with no rights to anything on our island. Nor
may we merely allow them to stay without sharing our resources
with them. It is incumbent on us to share with them—whether we
like the idea or not—even if that means that we enjoy a lowered
standard of living as a consequence.
Do the new islanders acquire any rights to things on the island?
Nozick seems to imply that they acquire only such rights as is
required to meet their minimal needs.!4 And they acquire no
370 David Lyons

property rights—except, say, what they might acquire through our

charity or in exchange for their labor. We retain all our original
rights, though our exercise of them, or their contents, has become
restricted,
But this is not entirely satisfactory, for reasons we have consid-
ered before in connection with inheritance. How are the new
islanders supposed to live for the indefinite future on our island?
Unless they have equal access to its resources, they may well be
condemned to an economically and socially subordinate position. In
some circumstances, at least, justice would not smile gladly on such
a prospect—even justice within an historical theory like Nozick’s.
For they might then be cheated and exploited because of their poor
bargaining position and lack of social power. Justice requires the
establishment and maintenance of background conditions for fair
bargains and agreements and for fair social arrangements generally.
This may well require that we share with them more radically than
Nozick might envisage.
One cannot say, a priori, what form such sharing would have to
take. One possibility is this. Suppose that we original islanders held
our resources in private parcels. If we all agreed to this arrange-
ment, and it served us well, no one suffering as a consequence, then
it may have been beyond repraoch. But, once the newcomers are
present, economic arrangements must be adjusted to accommodate
the increased demand upon our resources, the complications arising
from a changed population, and so forth. It is not beyond the realm
of possibility that, under the new conditions, a system of private
property will serve us very badly, while shared property, with
carefully managed use and enjoyment, would serve us well. Such a
change might be accomplished voluntarily, in which case Nozick
would presumably have no objection. But it is conceivable that a
private property holdout among the original islanders would prop-
erly be obliged to cooperate, against his will, and be required to
place his resources, along with everyone else’s, in a common pool.
This suggests that property rights themselves, and not just their
exercise or contents, are relative to circumstances.
But that inference is not irresistible, and for all practical purposes
it might make no difference which way we describe the
situation—as a modification of our original rights or as a redistribu-
 New Indian Claims and Rights to Land ere

tion of property. The question is whether the original islanders

retain some latent, prior claim to their original holdings, which
entitles them to recover their original property (so far as that might
be possible) when conditions change again.
To explore this possibility, suppose that our island is volcanic and
occasionally rises from the sea with the consequence that a greater
land surface becomes available to its inhabitants. Suppose that, after
the arrival of the new islanders, we eliminated private property in
land and managed its exploitation collectively. Then, one day,
during our current generation, the land rises due to volcanic
activity, providing a new doughnut-shaped area available for settle-
ment and exploitation. It might be the case that under these
circumstances private ownership of land would once again be satis-
factory and that everyone elects to adopt that system. The question
then is whether we original islanders have a superior claim to
property within the portion of the island that we originally oc-
cupied. If so, it looks as if there is a definite point in Nozick’s
suggestion that we continue ascribing the original property rights to
their original holders, so long as they have not relinquished them of
their own free will.
Now, I do not wish to deny that our original occupation of the
center of the island might provide good and sufficient reasons for
returning that part to us when redistribution is effected once again.
It may be assumed that we grew up in that part of the island and
regard it as our home, that we would be less happy elsewhere,
which is not true of the later arrivals on the islands. Such factors are
undoubtedly relevant to a humane as well as fair redistribution of
the land. But it is not clear that they amount to rights.
To see this, consider a case in which such sentimental attach-
ments are missing and in which the idea of a prior claim con-
sequently appears pointless and absurd. Suppose (expanding on
another example of Nozick’s) that the sole source of fresh water on
the island is a set of virtually identical water holes at some distance
from our dwellings. Through custom each family had exclusive use
of its own water hole. One day all of the water holes but one dry up.
As a consequence, the water from this hole must be shared by all the
islanders. If this condition continued indefinitely, the water hole
might amount, in effect, to common property. But is that mere
372 David Lyons

appearance? Suppose that after a while the other water holes are
unexpectedly replenished and become good natural sources of water
once again. It is reasonably concluded that each family should once
again have exclusive use of its own water hole. Suppose, however,
that the water holes are literally indistinguishable without some
conventional signs indicating their respective family assignments.
When all the holes but one dried up, the signs were not maintained,
and now that all the water holes are flourishing once more, we find
that they are almost indistinguishable. Since all the water holes are
equidistant to each person’s home, all are equally usable, and they
have no distinguishing features, there is no point in any family
invoking a prior claim to recover its original water hole. No one,
presumably, has sentimental attachment to a water hole. If one
somehow acquired such an attachment,then our humanity might
require that the person's feelings be respected. But that seems a far
cry from a claim of right.
Examples of this sort suggest that property rights are not stable
even within a single generation. They can be extinguished without
being voluntarily transferred. The very persistence of a right to
property such as land, and not just its content, is relative to
circumstances.
If we wish to identify a right to property such as land that is not
relative to circumstances, then we must make the right itself
inherently more flexible and responsive to circumstances. Locke’s
proviso suggests a possible model for beginning to construct such a
right. Its core would be conduct that is not harmful or dangerous to
others (or, as Nozick might prefer, conduct that does not encroach
upon or violate others’ rights). To this we add an obligation upon
others not to interfere with such conduct. This yields a full right of
action, composed of what jurists have called “liberty-rights,” to do
certain things, which are protected by “claim-rights,” not to be
treated by others in certain ways, the latter correlating with others’
obligations not to interfere. A minimal right to land may be seen,
initially at least, as a special case of such a right of action, so long as
Locke's proviso is satisfied. For, if enough and as good is left for
others—or, more generally, others’ situation is not worsened by
one’s appropriation of a parcel for, say, use and occupation—then
one may be said to have the liberty-right to use and occupy the land
and others the obligation not to interfere with such use and occupa-
 New Indian Claims and Rights to Land Sys)

tion. As conditions change, of course, the concrete implications of

such a right can vary. Most important, such a right would not imply
a permanent title to the parcel of land.
I am uncertain whether one can have any fuller right to property,
such as land, independently of laws and other social rules. Within a
given social setting, one might acquire morally defensible rights to
land and other property, but only so long as the institutions involved
are themselves morally defensible.

APPLICATIONS

Let us now consider a variant of our original example. Suppose that

the castaways Who arrive upon our shore are not friendly and
cooperative but aggressive and domineering. We try to be hospita-
ble but they do not reciprocate. They cheat us, kill many of us, and
force the survivors to reside in a small area of the island, away from
our homes, while they appropriate a disproportionately large part,
including the most desirable sectors, for themselves.
What does justice call for in such a case? It cannot require less for
us than it would have done in our original example, when it
required that we share with the newcomers. We too have a right to a
fair share of the island’s resources. If justice requires more, then it
may well include compensation from the piratical invaders for the
wrongs we have suffered at their hands. We may be too weak to
secure our rights; but that does not invalidate our claims.
Suppose that we are too weak and that we pass from the scene
without justice being done. Once we are dead, it is impossible to
compensate us for the wrongs we have suffered. Likewise, once the
invaders die away, the wrongdoers cannot contribute to any re-
ctification that justice may require.
Consider now the claims of our descendents, and for this purpose
imagine two alternative (or possibly successive) historical develop-
ments. In the first continuation of our island’s story we imagine that
our descendants continue to be subjugated, cheated, and denied a
fair share of the island’s resources, and continue to reside in that
portion of the island that was earlier assigned to us, their departed
ancestors. They too have valid claims, analogous to those we had
that were never respected. For justice requires that they receive
not only a fair share of the island’s resources but also, we may
374 David Lyons

assume, compensation for the wrongs they themselves have suffered

in being deprived during their lifetime of that fair share.
In the second continuation of our island's history, we imagine that
enlightenment finally spreads across the island. The descendants of
the piratical invaders come to live in harmony with our own
descendants, so that no one is deprived of a fair share of the island's
resources. Can we assume that any of our descendants, in this happy
sequel to our unhappy history, have additional claims against the
others on the island, the descendants of the piratical invaders? I do
not see how we can. If the generation in question has been deprived
of no part of its own fair share of the island’s resources, if they suffer
no continuing disadvantage owing to the legacy of the former system
on the island, what relevant matter might have been overlooked?
The wrong that was done to us, the wrong that was never rectified,
cannot now be corrected. That part of history is irrelevant to their
current claims.
It is important to see now that similar considerations apply to the
former case, the first and less happy continuation of our current
example. Our subjugated descendants have claims to a fair share of
the island’s resources and to compensation for wrongs done them by
a system on the island that deprives them of that fair share. That
system and thus their deprivation and their claims are rooted,
causally and historically, in the wrongs that we, their ancestors,
suffered at the hands of the invaders. But this is not to say that their
claims are normatively derived from ours, that they inherited our
original rights, or that their claims for compensation are claims for
correction of wrongs that were done to us, as distinct from wrongs
that have been done to them.
My metaphor and its moral may by now be obvious. Let the
island be America and the original islanders Native Americans, to
whom all the land may be said initially to belong. If those who had
landed on these shores had been impoverished outcasts from
Europe, unaggressive and cooperative, with no resources save their
labor power and no place else to go, it would have been incumbent
on their hosts not only to share their resources with them but also to
reshape their social arrangements to accommodate the new mem-
bers of their universe. For the purpose of this general point, it
makes no difference how the original occupants of the land had used
 New Indian Claims and Rights to Land 375

it, how they had divided it up, how they conceived of property
rights, whether they held it individually or collectively, and so on.
That is not, of course, the way things happened, and so history
developed much more like the unhappy history in the example of
this section and its first, unhappy continuation. Native Americans
by and large tried to be hospitable to their uninvited and unex-
pected guests, but the guests did not generally reciprocate. To be
sure, some of the guests were impoverished, some were outcasts,
some were unable to leave once they had arrived, and some,
perhaps, would have been prepared to form an integrated society or
to settle contentedly on limited tracts set aside for them by their
hosts. But too many acted rather as invaders, slavers, and con-
querors, who proceeded by force and by fraud to appropriate the
land and to eliminate or drive out the people living here.
I do not wish to deny any of this or to minimize the wrongs that
were done. I most especially do not mean to deny or to minimize
the valid claims of Native Americans living today. My point is that
their claims are unlikely’ derive normatively from their ancestors’
original rights. The original rights of Native Americans were no
more sacrosanct than anyone else’s. From the fact that they had
morally defensible claims two hundred or four hundred years ago it
cannot be inferred that those claims persist. But the initial argument
assumes just that; it assumes that circumstances had no effect on
those rights.

THE NEW INDIAN CLAIMS RECONSIDERED

Native Americans have systematically been discriminated against in

our society. They have a valid claim to a fair share ofits resources as
well as to social and economic opportunities. They also have a valid
claim to compensation for unjust deprivation that the current
generation has suffered from past injustices. But it is highly doubtful
that they have any special claims based upon their distant ancestors’
original occupation of the land. For circumstances have significantly
changed. After the European dispossession of the Indians, waves of
impoverished immigrants arrived on these shores in little better
shape than castaways from a shipwreck. Most of the occupants of
America today have had little, if anything, to do with dispossession
376 David Lyons

of Native Americans. This does not mean that they have no com-
plicity in a pattern of unjust deprivation of current Native Ameri-
cans, for which compensation is required. But that is another matter
entirely, and a much more complex matter too.
I suggest, therefore, that the current Indian land claims be
viewed, not as invoking an original right to the land, a right that has
been passed down to current Native Americans and that now needs
to be enforced, but rather as an occasion for rectifying current
inequities (some of which, of course, may trace back causally to the
dispossession of Native Americans and the aftermath).
Now that I have made my major points, I must try to note some
complications.
One set of complications turns upon the fact that the current
Indian claims are being made on behalf of tribes rather than private
persons. Tribes originally held the land, and a tribe, like a nation,
can hold a right over generations. This has some bearing on the
current claims. It does not affect my main point, which was not just
that inheritance is suspect but more generally that moral rights to
land are inherently unstable or variable with circumstances. We
cannot assume that rights held generations ago, even if they were
held by tribes, have persisted to this day. But this aspect of the cases
is relevant to claims invoking the notion of compensation for wrongs
done. Some past wrongs can no longer be corrected, but some can.
It may be impossible to compensate the ancestors of current Native
Americans for wrongs that they suffered long ago, but it may be
possible to compensate tribes for past wrongs done them. If the
tribes were wronged, those wrongs may well have involved viola-
tions of original rights, even if those rights did not survive the
changing circumstances and did not persist into the current genera-
tion. If tribes can indeed be wronged, and such wrongs are subject
to compensation, then the current claims can be supported by
related considerations: this sort of argument transcends the valid
claims of current Native Americans for compensation in view of
wrongs done to them as individual human beings. I do not wish to
deny such possibilities here. They require careful and systematic
examination.
The tribal character of the current claims is relevant in other
respects too, which raise complex and difficult issues. I have noted,
for example, that one aim of the current suits appears to be not mere
 New Indian Claims and Rights to Land 377

ownership of the land but control over its development. There is the
prospect of conflict between the interests of Native Americans in
preserving undeveloped land and the others who wish to develop it,
build on it, live and work on it. This is not like the conflict between
conservationists and developers. For the Native Americans involved
are seeking to rebuild a way of life that turns upon certain ways of
dealing with the land, and an issue here is the right to inhibit
development (which may involve sorely needed jobs, and not just
profits) based on the right to secure a culture.
That brings us to a central argument favoring the current claims.
And it is important to support the current claims, since radical steps
have been threatened to undermine them, including retrospective
legislation.
One thing that makes the claims under the Nonintercourse Act so
important is that they appear to be legally well-founded. Unlike past
calls for reparations for black Americans, in view of the legacy of
slavery and discrimination, the current claims under the Noninter-
course Act turn upon existing law. Radical new legislation or
executive action is not needed to sustain them.
But it may reasonably be urged that these cases test the sincerity
of our historical commitments. The federal government long ago
assumed “fiduciary” responsibility for securing Indian lands and
protecting Native American interests. It has however adhered to the
law chiefly when that worked to the Indians’ disadvantage. Now,
when at last Native Americans have marshalled the legal resources
to secure some lost benefits, the threat is that the law will not be
followed. Even handed fairness would seem to require that the
federal government live up to its past commitments and not retroac-
tively change the rules just when it would undermine Indian
interests to do so.
Beyond this, it may dutifully be observed that justice would not
be done by simply returning all the lands in question to the tribes
now claiming them. This would impose enormous burdens on small
home owners and small businesses without sufficient reason. It
seems, in any case, that undeveloped land is the primary target of
the tribes, the other land being unavoidably blanketed in under the
legal claims. The federal government should work to negotiate a
satisfactory settlement. This is what the tribes have been seeking for
some time.
378 David Lyons

Ifa settlement results in burdens on individuals or states, it would

seem reasonable for the federal government to assume responsibil-
ity for compensation too. For the federal government not only has
greater resources than the several states, some of which are hardly
affluent; it was also negligent, under its own law, in failing to
oversee the transfers of land and in thus failing to discharge its legal
responsibilities as trustee. The responsibility for righting such
wrongs and for paying what it costs to do so should not be allowed to
fall on the nearest and perhaps most vulnerable parties, but should
be shared by society at large.
These costs will be of two types. First, cash payments are being
sought, in addition to the lands, for back rents and damages. (These
claims, incidentally, appear immune to attack by retroactive legisla-
tion.) Second, cash settlements will undoubtedly be made in lieu of
some land that might otherwise be recovered. That seems a sensible
solution for much developed land within the tracts in question, and
a solution that the Indian tribes are quite willing to achieve. Such
costs should be borne by the federal rather than by individual state
governments.
Claims under the Indian Nonintercourse Act are different from
some other claims that Native Americans may make for recovery of
land, since the former turn upon plainly illegal transactions while
the latter may involve marginally legal but unjustifiable acts by the
federal government. The rhetoric that I have anatomized in this
paper does not distinguish between the cases. I do not mean to
suggest that the claims are unsupportable because the rhetoric is
unilluminating. The point is rather than the claims are stronger then
the rhetoric may suggest. My purpose here has been to challenge
certain ways of thinking about moral rights to property—ways that
are typically invoked to secure unjust holdings. Property rights are
not sacrosanct. They must bend to the needs and interests of human
beings.!5

NOTES
Ie Zor UK Sh GRAR tae
2. On the Maine cases in particular see Robert McLaughlin. “Giving it
back to the Indians,” Atlantic Monthly (February 1977): 70-85: more
generally see Akwesasne Notes 9 (early Spring 1977): 18-21.
 New Indian Claims and Rights to Land 379

3. Joint Tribal Council of the Passamoquoddy Tribe v. Rogers C. B.

Morton, 528 F 2d 370 (1975).
4. Narragansett Tribe of Indians v. Southern Rhode Island Development
Corporation, 418 F. Supp. 798 (1976).
5. See his “Distributive Justice,” Philosophy and Public Affairs 3 (Fall
1973): 45-126, especially Section I, 46-78. This essay was later published as
Chapter 7 of Nozick’s Anarchy, State, and Utopia (New York: Basic Books,
1974).
6. John Locke, The Second Treatise on Government, Chapter V (“Of
Property). (Citations in the text are to numbered sections of the work.)
7. On Locke’s theory, see Nozick, “Distributive Justice,” 7Off.
8. See A Theory of Justice (Cambridge, Mass.: Harvard University
Press, 1971), 85-87.
9. For example, by helping to fix vast wealth within a very limited
number of families, and thus helping to establish permanent power elites.
10.“ Distributive Justice,” 72.
11. “Distributive Justice,” 76.
12. “Distributive Justice,” 72.
13. “Distributive Justice,” 76.
14. “Distributive Justice,” 75-77.
15. Earlier versions of this paper were read at The Catholic University of
America, Hamilton College, and the University of Vermont. I would like to
thank those who commented on those occasions for their helpful criticisms
and suggestions. I would also like to thank John Fischer, Stephen Massey,
Robert Summers, and William Wilcox for comments, as well as the readers
for Society Theory and Practice, and Matthew Lyons for research on the
historical background of the cases under discussion here.
 19
Justice and Entitlement
HILLEL STEINER

Robert Nozick and, more recently, Eric Mack have presented

compelling arguments as to why, if justice assigns inviolable enti-
tlements to individuals, “patterned” or “distributionist” allocative
principles cannot be principles of justice.! But although both au-
thors may be said to have demonstrated the injustice of redistribut-
ing holdings among the members of a group whose membership is
constant (in identity and, therefore, in number), their arguments
cannot be readily extended to one whose membership varies.
Human society is a group whose membership varies. If the object of
justice is to prescribe the inviolability of every member of a group,
any theory of justice in human society must incorporate principles
which provide this coverage to new members as well as to continu-
ing ones.
One way of construing the point historically at issue between
exponents of the Nozick-Mack conception ofjustice and exponents
of (some) redistributively inclined conceptions of justice is to see it
as consisting in the question of whether a successfully enforced set
of rules prohibiting only interpersonal coercion is sufficient to
ensure that individual holdings are just. Most exponents of the
former conception are traditionally credited with answering this
question in the affirmative, while their opponents usually return a
negative verdict. Interestingly enough, however, Nozick himself

380
 Justice and Entitlement 381

answers it in the negative, and there are good reasons for his doing
so. Nozick’s negative answer is implicit in the fact that his “historical
entitlement” conception of just holdings incorporates not only rules
for transfer and rectification but also a rule for appropriation. Rules
for transfer and rectification ensure that no one’s holding contains
objects which he forcibly acquired from others. But what is it that
confers the title to those objects upon those others?
Clearly, the requirement of inviolability provides us with a range
of grounds on which, in particular cases, we might claim that an
object justly belongs to one person and not another. Inviolability
entails ownership of one’s own (physical) person and thus, following
Locke and Nozick, ownership of one’s labor. An object may there-
fore be mine on the ground that I made it. If it is mine, I am justly
entitled to dispose of it as I wish, and therefore the title to it may
justly accrue to anyone to whom I donate or sell it. But it is
necessary truth that no object can be made from nothing, and her ce
that all titles to manufactured or freely transferred objects must
derive from titles to natural and previously unowned objects. No
title to a manufactured or transferred object can exist if that object
was manufactured by, or transferred from, someone not entitled to
acquire it or its unaltered antecedents from nature. An individual's
just rights logically extend not only to noninterference by others
with the disposition of what is his but also to the initial acquisition of
what he may then dispose of. So any theory ofjustice that employs
voluntary disposition as the ground of entitlement to already owned
objects must also incorporate a rule forjust initial acquisitions. Such
a rule, as Nozick says, determines the legitimate “first moves”
which may be made in constituting a set of individual holdings.”
The application of such a rule might, conceivably, be a fairly
straightforward matter in a group whose membership is constant;
that is, it might not entail any redistribution (although it is worth
noting that Nozick’s proposed rule for appropriation does entail
redistribution). In human society, however, this appears not to be
the case. For here we have a situation in which membership is
continually changing, and yet we are required to treat each member
as possessed of inviolable rights. What protection is to be afforded to
the appropriate entitlements of individuals originating at a histori-
cally later moment than individuals who are members of the first
generation?
382 Hillel Steiner

One solution is to require that no individual's appropriation of

unowned natural objects may ever be so great as to preclude any
other present or future individual from making a similar appropria-
tion. This stringently conservationist interpretation of the just initial
acquisition requirements, unwieldy as it would be, seems to be the
only one which can circumvent the necessity of periodic redistribu-
tion. Otherwise, we are confronted with the problem of formulating
a different rule which somehow assigns the same kinds of right to all
individuals as were enjoyed by those licensed to make the “first
moves.” Nor will it do to object that this problem is precluded from
arising by an affirmation that the right of free disposition must
include the right of bequest and, hence, that the right of initial
acquisition is exercised on a once-for-all basis, leaving (eventually)
no unowned natural objects over which later individuals might
exercise this same right. For, in the first place, the existence of the
right of bequest requires special proof, inasmuch as the sense in
which a deceased person may be said to have his inviolability
breached is evidently not the same as that in which a living person
undergoes such a transgression. But, more importantly, the
aforementioned problem is not precluded by such an assertion (i)
unless we are prepared to say that the rules of justice vary over time
and between individuals, or (ii) unless we presume that the re-
positories of rights are not, after all, individual persons, but rather
ancestral-chains-of-individual-persons, to which collectives alone
all the rules constituting just holdings apply. (This latter point
applies equally to nonancestral bequeathing “chains.”) That an
individual’s deserts should be determined by reference to his
ancestors delinquencies is a proposition which doubtless enjoys a
degree of biblical authority, but its grounding in any entitlement
conception ofjustice seems less obvious. It is therefore difficult to
detect the presence of any strong connection between an indi-
vidual’s just deserts and his ancestors’ accomplishments.

NOTES
1. Robert Nozick, Anarchy, State and Utopia (New York: Basic Books.
1974), p. 2; Eric Mack, “Distributionism versus Justice,” Ethics 86, no. 2
(January 1976): 145-53.
2. Nozick, p. 151.
 20
Entrepreneurship,
Entitlement, and Economic
Justice
ISRAEL M. KIRZNER

Nozick’s entitlement theory of justice plays a crucial role in his

carefully and brilliantly crafted case for the minimal state. Accep-
tance of the theory, it appears, sweeps away with one stroke all
those demands for the state interference with the market that rest
on the claims of distributive justice. The persuasive elegance with
which Nozick develops his position can leave few thoughtful,
moralist critics of uncurbed capitalism unimpressed, at least, by the
strong claims which Nozick advances, precisely on the grounds of
economic justice, for the free market. It will be the thesis of this
paper, however, that Nozick’s theory of entitlement, important
thought it undoubtedly is for any defense of the morality of laissez
faire, does not—at least without significant reformulation—solve all
the difficulties that may be alleged to exist in respect of the justice of
the market. Pursuing this theme, we will offer a suggestion for
supplementing (or perhaps reformulating) Nozick’s theory which
may not only equip it to handle the difficulties to which the paper
draws attention, but may, in fact, render its defense of the morality

383
384 Israel M. Kirzner

of the market even more straightforward and subject to fewer

qualifications than Nozick apparently believes to be necessary.
Nozick’s theory depends, in its application to the market, largely
on the view that with few definite exceptions, the market reflects
adherence to the principles of justice both in the original acquisition
of holdings (from the natural state) and in subsequent transfers of
holdings in market transactions. The difficulties to which this paper
draws attention pertain to the claim that the market is fully consis-
tent with the principles of justice in transfer. This claim carries
conviction, it will be shown, only if we are prepared to incorporate
into our entitlement theory certain somewhat novel views concern-
ing the morality of the entrepreneurial role. But, we will then
argue, recognition of this aspect of the entrepreneurial role makes
it no longer useful to distinguish, as sharply as Nozick does,
between justice in original acquisition, on the one hand, and justice
in transfer on the other. So that, while the entitlement theory may,
it will turn out, indeed be deployed to defend the morality of the
market—and with fewer reservations, perhaps, than in Nozick’s
own statement—this will have been achieved only through a fairly
substantial reformulation of that theory.

JUSTICE IN TRANSFER, VOLUNTARINESS, AND

ERROR

For the purposes of this paper, we accept Nozick’s entitlement

theory of justice without reservation. If an object was originally
acquired justly from nature, and if all subsequent transfers of the
object have been justly accomplished, then, we will say, the present
holder of that object holds it justly—and no aesthetic or moral
considerations concerning desirable distribution patterns can, with-
out injustice, permit the state to tamper with the rights of the
present holder. The issue that concerns us is whether this theory,
when applied to the results of the free market, can certify these
results as being in accord with just principles. Let us, for the time
being, accept Nozick’s own conclusion! that the free operation of the
market system is consistent with past acquisition of initial holdings.
Our concern will be with the justice of market transfers.
Now Nozick has not, he explains,? attempted in his book the task
of specifying the details of the principles of justice in transfer. The
 Entrepreneurship, Entitlement, and Economic Justice — 385

general outlines of his position on the justice of market transfers are,

however, clearly implied: What has been acquired through market
transaction has been justly acquired for the simple reason that such
transactions are voluntary. The money which a professional athlete
receives from spectators eager to watch his performance is justly his
because those who paid have done so willingly. They preferred to
watch the athlete (at the given money price) rather than to retain the
money (without witnessing the performance).? Because a property
system excludes the taking of anyone’s goods or money without his
consent (whether by theft or fraud)—because, that is, such a system
permits transfer only by gift or voluntary market exchange—it
follows that such a system has room only for transfers that are just.
The view assigning so critical a role to voluntariness in just
transfers is, surely, a highly appealing one, and we will accept it
fully as the basis for our discussion. Involuntary transfers, we have
said, are ruled out in the market by definition—only voluntary
exchanges qualify as market transactions. The question which we
wish to raise has to do with the extent to which error, in the
decisions of market participants to engage in exchange, erodes the
voluntariness and hence the justice—of the transfers effected by
such exchange. Now it might appear that consistent application of
our definition of market transactions as being voluntary, avoids any
difficulty (for the justice of market transfer) arising out of error. If,
say, a seller sells an item in error, then either the error was so
serious as to impair definitively the voluntariness of the sale, or it
was not. That is, either the error on the part of the seller was so
fundamental that we are compelled to say that he did not really wish
to sell at all (i.e., that his consent to sell was given utterly erroneously
and was thus no consent at all) or else it was not sufficiently serious
to impair to voluntariness of the sale. If the error was sufficiently
serious as to render the sale involuntary, then we should simply
pronounce the exchange to be a total mistake and thus not a
permissible market transaction at all. Unjust though we must con-
sider the purchase to have been, it was, we should say, not a market
transaction but simply an involuntary transfer. If, on the other hand,
the voluntariness of the sale was not undermined by the error, then
the transfer remains a just one. Either way, it would seem, the
possibility of error presents no problems for the justice of market
transfers. (Defenders of the market have, after all, always made it
386 Israel M. Kirzner

clear that fraduluent transactions are a form of theft and are with or
without state action—to be expunged from the market system.)* But
surely a fairly persuasive case can be made for a less comfortable
view of that matter. Such a case may perhaps gain plausibility from a
consideration, first, of the degree to which the market process
depends, in fact, on the profitability of entrepreneurial trading with
market participants who have at least to some extent—erred.

EQUILIBRIUM, DISEQUILIBRIUM, AND ERROR

For the economist’s model of market equilibrium it is not only

possible, but indeed necessary, to imagine a world without error. As
Hayek explained forty years ago, equilibrium is defined as the state
of correct foresight.2 Were, then, a market economy to be an
economy continuously in equilibrium, it would be indeed easy to
defend the voluntariness—and hence the justice—of all market
transfers, since every one of them would reflect decisions made with
complete awareness of market conditions. In no way could it be
claimed that any exchanges were entered into out of ignorance.
But it is now well understood that the function of equilibrium
models is hardly to portray the real world states of affairs. Rather
such a model serves to illuminate the nature of the equilibrating
market forces which are at work during the states of disequilibrium
which, in fact, prevail at all times.® And it is of the essence of states
of disequilibrium that the decisions being made are quite different
from those which would have been made in an errorless world. It is,
further, the case that insight into the equilibrating forces of the
market generated by the conditions of disequilibrium reveals them
to operate through entrepreneurial discovery (and exploitation) of
the very errors which are characteristic of disequilibrium. A simple
example is able to illustrate the matter very effectively.
It is one of the features of models of competitive equilibrium that
no more than one price can, in the market for a given period, prevail
during the same period. Before equilibrium has been attained,
however, many prices for the same good may be simultaneously
paid and accepted by different buyers and sellers. The disequilibrium
situation, in which many prices prevail in this way, generates a
spontaneous equilibrating tendency toward the elimination of such
gratutious price differentials. The existence of a price differential
 Entrepreneurship, Entitlement, and Economic Justice 387

constitutes an attractive opportunity for entrepreneurial profit.

Eager seekers of profit will tend to grasp these opportunities until
their competitive activity has squeezed out all such opportunities,
i.e., until prices have become uniform throughout the market.
Consideration of this simple example immediately shows how
heavily the equilibrating process rests on the profitability of acting
to take advantage of the errors of others. We notice at the very
outset that the multi-price situation characteristic of disequilibrium
reflects widespread ignorance on the part of buyers and sellers.
Those who paid the higher prices were clearly unaware of the sellers
who were prepared to accept (and in fact did accept) lower prices;
and those sellers who accepted the lower prices were obviously
unaware of the buyers who offered and paid the higher prices. The
entrepreneur who discovers and moves to exploit the profit oppor-
tunities presented by the multi-price situation is buying at low
prices from those who are unaware of the possibility of selling at
higher prices. He, on the other hand, is buying at the low pricesin
order to sell at higher prices (to those who are, in turn, unaware of
the possibility of buying at lower prices). The equilibrative aspects
of the market process depend, in an essential way, upon the lure of
the profits made possible by the errors of those with whom the
entrepreneur deals. In fact, the insights gained from the simple
illustrative example we have used apply not only to this simple case
but to the most complicated of markets, involving production in any
number of stages, inputs in any number and in any variety, and
outputs of any degree of multiplicity and heterogeneity. The coor-
dination and allocative properties of competitive markets depend
entirely on the attractiveness of pure entrepreneurial profit oppor-
tunities; such opportunities arise only out of the less than perfect
omniscience of those from whom entrepreneurs buy, and of those to
whom they sell.
The question we wish to raise concerning the justice of market
transfers thus emerges fairly clearly. If the market depends heavily
on the exploitation of profit opportunities made possible only by the
errors of others, and if goods purchased from sellers who sold only
as a result of error, and money received from buyers who bought
only as a result of error, be considered unjustly acquired—then
surely the justice of the market has been unsalvageably com-
promised. It will not do to declare that transactions entered into in
388 Israel M. Kirzner

error—being “involuntary’—are excluded by definition from the

class of market transactions—since a market process without “er-
roneous’ transactions is unthinkable. The only logical possibility for
defending the general morality of market transactions must be to
maintain that the errors which characterize disequilibrium markets
do not affect the voluntariness of the transactions completed. The
profits won by entrepreneurs taking advantage of the errors of
others are, one would have to maintain, not unjust, as measured by
Nozick’s “voluntarism” yardstick of justice in transfer. Can such a
position, we must ask, in fact be maintained?

THE MORALITY OF ENTREPRENEURIAL

PROFIT: A BRIEF DIGRESSION

It will be seen that the question we have raised about the justice of
market transfers generally has led us to question, in particular, the
justice of pure entrepreneurial profit.7 The form in which we have
posed this last question makes it rather different from other chal-
lenges to the morality of profits. It may be useful to digress very
briefly in order to explain this difference. The roots of the matter lie,
in the first place, in the theory of profit which one chooses to
embrace, and, in the second place, in the theory of justice which
one wishes to apply.
Most criticisms of capitalism made on moral grounds have de-
nounced profit as unjust. Usually such criticism has rested on one
or other of the theories of economic justice which Nozick has,
persuasively, rejected in favor of the entitlement theory. Entrep-
reneurial profits are likely to violate many of the patterns of distribu-
tive justice that one might wish to promote. In addition, the critics
saw profits as generated by, say, the exploitation of labor, or by the
unfair exercise of economic power; or what they were criticizing as
profit was not pure entrepreneurial profit at all, but the interest on
capital.®
The question which we, on the other hand, have raised in this
paper about entrepreneurial profit (and indeed about all market
transfers) rests on what has been described as an arbitrage theory of
pure profit,? and has been raised against the specific background of
Nozick’s entitlement theory. The arbitrage theory of profit sees
profit as generated by the existence of different prices in different
 Entrepreneurship, Entitlement, and Economic Justice 389

parts of the market for what are, economically if not physically,

identical goods. Such differentials can arise only as a result of
imperfection in knowledge. So that it is this theory of profit which,
in the context of Nozick’s “voluntariness” criterion of justice in
transfer, is responsible for the question which we have raised.

OPTIMUM IGNORANCE, DELIBERATE MISTAKES,

AND GENUINE ERROR

We return to consider the possible challenge to the justice of market

transfers, arising out of the errors on the basis of which market
transactions are completed. In this section we discuss (and reject)
one possible way of dismissing this challenge altogether. In sub-
sequent sections we proceed to examine the question itself.
One way in which one might reject the challenge of injustice
based on error, is to deny altogether the possibility of genuine error.
One might, that is, maintain that while market transactions are
indeed frequently entered into as a result of incorrect knowledge or
expectations, this never involves genuine error. After all, one who
knows that his vacation may be ruined by bad weather and
nonetheless travels to the resort and occupies his hotel room,
cannot, when bad weather indeed arrives, be said to have really
erred (in the sense of having consented to do something which he
did not “really” wish to do). After all, he deliberately erred; he
deliberately risked his money; he gladly took his chances; every
transaction which he entered was a wholly voluntary one. To the
extent that every decision made in error is made either through
deliberately accepting an uncertainty (such as the incidence of bad
weather), concerning which accurate knowledge is simply unavail-
able, or through deliberately choosing not to spend the resources
necessary to remove the possibility of error!°—every decision has
been deliberately and hence voluntarily made. A deliberate deci-
sion not to acquire costly knowledge is, after all, made voluntarily. If
we rule out deception (where, for example, an entrepreneur mis-
leads a seller into thinking that no one else is prepared to pay a
higher price),1 the fact that a seller sells to an entrepreneur at a low
price (knowing full well that diligent search might yield the possi-
bility of selling at a higher price) can surely not raise doubt
concerning the voluntariness of the sale (even though it remains
390 Israel M. Kirzner

true that, were the seller in fact to have known that higher prices
were being paid, he would not have sold at the lower price).
This line of argument, it will be observed, denies that genuine
error (in the sense of a decision being made in unwitting ignorance
of pertinent information) can be made at all. All mistakes are seen as
the result of deliberately assumed risk. No mistakes can raise
questions concerning the voluntariness of decisions made.
The writer has elsewhere argued at length that genuine error can
and does indeed occur.!? Without repeating that discussion here, it
will simply be pointed out that decisions are often made in ignor-
ance of the very need and/or the possibility, of acquiring (possibly
freely available) information. It is one thing to know that one is
ignorant, and to deliberately maintain one’s ignorance because of
the high cost of gaining knowledge. It is quite another to be ignorant
simply because one has no inkling that one is ignorant, because one
has no idea that information exists, or indeed that there is any such
thing imaginable—in the relevant context—as “information.” Surely
the latter kind of ignorance is abundantly present; genuine error is
alive and well. We cannot rule out the possibility that market
decisions have been made, not out of deliberately accepted ignor-
ance, but out of genuine error. The “voluntariness” of such deci-
sions still calls for examination.!%

MISTAKES: LAW AND MORALITY

The question of erroneously made decisions has, of course, been

treated thoroughly by jurists in regard to the law of contracts. And
the kinds of error occuring during disequilibrium that have given us
concern—where, say, sellers would not have sold at the prices they
accepted had they known the true eagerness of buyers elsewhere in
the market—are, in the legal literature, not seen as affecting the
validity of transactions completed. Providing the entrepreneur-
buyer did not explicitly deceive the seller concerning the facts—
extrinsic to the goods sold—about which he has been misinformed
(and providing no fiduciary-type relationship between them exists
that might render the buyer's silence concerning the truth, a form of
implied deception), the law finds no grounds to invalidate market
transactions into which one of the parties has entered under mista-
ken assumptions concerning present or future market conditions.
 Entrepreneurship, Entitlement, and Economic Justice 391

“Tacit acquiescence in the self-delusion of another, if nothing is said

or done to mislead, or silence which does not make that which is
stated false, draws with it no legal liability,”"4 we are told. And the
kinds of selfdelusion referred to include that of the landowner who
sells his land for the price of grazing land when in fact it contains
valuable minerals, or who is ignorant of the fact (known to the
buyer) that a railroad is intended to pass through it. “If the parties
are at arm’s length, neither of them is under any obligation to call
the attention of the opposite party to facts or circumstances which
lie properly within his knowledge, although he may see that they
are not actually within his knowledge.”!5 The law thus takes a
hard-boiled view of commercial transactions—an attitude often
loosely and imprecisely identified as caveat emptor—which does not
see a mistake (except where it was induced by one’s trading partner)
as legitimate cause for the invalidation of acompleted transaction or
commitment. Apparently the law does occupy a position close to
that (rejected in the preceding section) which sees a mistake merely
as the deliberately-assumed gamble that failed, rather than as
representing lack of true will to participate in the transaction as it
turned out. 16
But the legal validity of entrepreneurial transactions in dis-
equilibrium markets is not at all what is of concern to us in this
paper and was never in question. Of course, the legal system within
which the capitalist economy operates recognizes the validity of the
market transactions which make up the system. We have been
concerned with possible challenges to the morality of that very legal
system which sustains capitalism. What answer, we have been
asking, can one give those who might contend that market transfers
violate Nozick’s canons of justice in that error (which invariably
characterizes market transactions) introduces an ineradicable moral
stain of involuntariness into the very fabric of these transactions? It
is noteworthy that the jurists expounding the hard-boiled attitude of
the law toward mistake go out of their way not to defend the
morality of those who benefit by the law’s tough-mindedness. “No
doubt” we are in fact told, “such dealings would be repugnant to a
man of high honor and delicacy . . .”!7 The view of fraud which is
taken by the law is carefully distinguished from the view of
moralists.1® It will not do simply to denounce moralists who fail to
include, in their assessment of the morality of market transactions,
392 Israel M. Kirzner

the enormous social benefits generated by these transactions. To be

sure, these benefits ought not to be overlooked by the moralist.*?
But Nozick’s demonstration of the justice of the free market cannot,
surely, be pronounced complete if the voluntariness of market
transfers, upon which Nozick’s case depends, can possibly remain
under a cloud. What does one say to the critic who argues that the
law permits the gullible to be cheated? (“Fraud is difficult to prove”
is a repeated refrain; and, anyway, what is not technically fraud
may, to men of honor, be seen as cheating all the same.) And, in the
broadest of senses, can it not be said that the market process
depends on (at least a mild form of) “cheating”
??°

ON CHEATING AND THE JUST PRICE

No examination of the possibility of injustice in market transfers can

avoid some reference to recent discussions of the medieval just price
doctrines. Earlier scholars had understood the medieval writers to
have seen cost of production as the criterion for justice in pricing.
An unjust price for a good was one which diverged from its true
value, as defined by production costs (with the latter “determined
by a fixed standard of living on the part of the producers and. . . not
to include any element of interest.”).2! An unjust price was thus
seen as unjust not primarily because it involved deceit by the one
party (or at least an error on the part of the second), but simply
because justice requires that each party to an exchange receive the
true value of what he has given up. If divergence from the true value
is described as involving “cheating,” this must then mean, either,
merely that without deceit it would presumably be impossible to
secure more than the true value of what one gives up, or that to
cheat is to be defined purely in terms of divergence from true value.
(Compare the phrase used by Nozick in describing the old question
about the possibility of profits: “How can there be profits if every-
thing gets its full value, ifno cheating goes on?”)?2 For decades after
1870 economists found it necessary to explain how inadequate such
a conception of justice in transfer, ignoring all demand con-
siderations, must be considered. And in Nozick’s entitlement theory
little room seems to be assigned for divergence from production
costs as a criterion for injustice in transfer. (In referring to the
 Entrepreneurship, Entitlement, and Economic Justice 393

possibility of “gouging,” Nozick seems quite content to leave to

buyers the responsibility of looking out for themselves.)?8
More recently, however, historians of medieval economic
thought have emphasized references in the scholastic writings to
market price as the criterion for justice.24 And such references are
occasionally couched in language suggesting that to take advantage
of imperfect knowledge of market conditions on the part of one’s
trading partner is to violate the canons of the just price.25 But it
would appear to be incorrect to ascribe to the medieval writers the
concern for the possible injustice of disequilibrium market transfers
we have expressed in this paper. Whether, with Schumpeter and de
Roover, one is prepared to credit Aquinas and the medieval writers
with a sophisticated understanding of the relationship between cost
of production and long run equilibrium price, or whether, with
Hollander, one is not prepared to do so, it seems fairly clear that for
Aquinas actual market price is understood as being always the
equilibrium price. It was because the market price was, therefore,
seen as expressing the true value of a good (reflecting “the entire set
of objective and subjective elements which forms the community
estimate6) that it was considered unjust to take advantage of a
buyer's ignorance of the market price. It is true that pure profits
were, as Hollander has explained, generally frowned upon by
Aquinas, but this was clearly on grounds of other than the taking
advantage of the ignorance of one’s trading partners. In other
contexts it was clearly not considered unjust to take advantage of
another's ignorance. A number of writers have drawn attention to
Aquinas’ view that a seller may charge a high price for grain in a
place where it is dear, even though he knows that others are
following with more supplies—a fact which, if known to the buyers,
would have led them to refuse to pay the present higher price.?7
Clearly a price paid only out of ignorance of the true facts does not,
in their view, by itself mark it as unjust.
Aquinas’ justification for this permissive position clearly implies
that the market price at a given instant is the true equilibrium price
relevant to questions of justice. The current high value of the grain
is its true current value. The anticipated arrival of additional
supplies of grain can be expected to lower the market price in the
future, so that a seller who sells at today’s price does not act unjustly
394 Israel M. Kirzner

in failing to disclose what will happen in the future.2® One can

understand that if the market price is considered the just price
because it reflects the current “community estimate’ of value, then
information concerning the future possessed by a single market
participant may not be seen as altering the present community
estimate; and his exploitation of his superior information need not,
therefore, be seen as unjust according to the criterion adopted. But
if, as in this paper, one questions the justice of market exchange
precisely because it occurs under conditions concerning which one
of the parties is ignorant (so that his consent to the deal might be
said to hinge on a wholly erroneous perception of the relevant
circumstances), then the scholastic insight into the justice of market
price has not helped us answer our question. Our awareness that
market prices are never equilibrium prices does not encourage us to
accept the market prices as just because they are somehow expres-
sive ofall relevant circumstances. These prices we must recognize,
necessarily reflect the very errors which have occasioned our con-
cern.

THE BASE FOR THE JUSTICE OF THE MARKET:

THE ETHICAL AND ECONOMIC BUILDING BLOCKS

It will be argued during the balance of this paper that the difficulties
we have raised concerning the justice of disequilibrium market
transactions can be resolved definitively by (a) accepting a particular
ethical judgment, and consistently applying it in conjunction with
(b) the acceptance of a particular economic insight into the nature of
disequilibrium market transactions. The ethical judgment referred
to has been called the “finders, keepers” ethic.22 The economic
insight is that which permits us to perceive the discovery of a
hitherto unknown market use for an already owned resource or
commodity as the discovery of (and consequently the spontaneous
establishment of ownership in) a hitherto un-owned element asso-
ciated with that resource or commodity. We will argue that accep-
tance of the morality and justice of a market system does imply the
acceptance of a market system and economic ways of seeing things.
Many who consider a market system just may not perhaps have
explictly articulated their own position to themselves in precisely
 Entrepreneurship, Entitlement, and Economic Justice 395

these terms, but upon reflection they will probably recognize our
exposition as faithful to their own view. It should be noted that it is
not the purpose ofour discussion of these ethical and economic ways
of seeing things to insist on or persuade the acceptance of these
views. Our purpose is only to show that there exist plausible (and, at
least implicitly, apparently widely accepted) moral and economic
insights upon which a consistent defense of the justice of the market
can be constructed. These insights, we will discover, remove the
difficulties which we have encountered so far. Moreover, they can
be easily grafted onto a suitably reformulated entitlement-theoretic
interpretation of market justice. Let us consider separately and
more carefully each of these ethical and economic insights—
building blocks for the construction of the case for the justice of the
free market.

FINDERS, CREATORS, AND KEEPERS

The “finders-keepers ethic” has been discussed only slightly in the

literature on the ethics of private property.*° In fact, it seems fair to
conclude that most writers on the justice of private acquisition from
nature of hitherto unheld resources do not accept the finders,
keepers ethic. The mere fact that an individual has stumbled on a
rich deposit ofa valuable natural resource does not (without at least
some effort on his part, say, some mixing of his labor with the
resource) entitle the discoverer, on this view, to claim title to it
merely on the grounds that he found the resource deposit. Mere
discovery has not placed the discoverer, on this view, in any kind of
privileged position with respect to the hitherto unheld resources. If
the rest of mankind were seen, up to the present, as enjoying rights
of access to and common use of these hitherto unheld resources,
then these rights are seen as in no way disloged by the mere event of
the discovery.
In order to introduce plausibility into the notion of finders-
keepers,*! it appears necessary to adopt the view that, until a
resource has been discovered, it has not, in the sense relevant to the
rights of access and common use, existed at all. On this view it
seems plausible to consider the discoverer (of the hitherto “non-
existent” resource) as, in the relevant sense, the creator of what he
396 Israel M. Kirzner

has found.? It becomes, then, fairly easy to understand how the

finder can be held justly entitled to keep that which he has
“created.”
It should be noted that ownerships-by-creation is quite different
from ownership-by-just-acquisition-from-nature (as the latter is
spelled out in, say, Nozick’s entitlement theory). Ownership by
acquisition occurs against the prior background of given unheld
resources (even if no one is aware of their very existence). Acquisi-
tion is, in fact, a kind of “transfer” (from nature to the first holder).
Ownership by creation, on the other hand, involves no notion of
transfer at all. The finder-creator has spontaneously generated
hitherto non-existent resources, and is seen, therefore, as their
natural owner.
The adoption of a finders-keepers ethic does not, of course, rule
out scope for acquisition from nature in the usual (Nozick’s) sense.**
The first man to land on Mars can hardly claim title to it as its
“creator. In order to establish just ownership in an unheld resource
the existence of which everyone is fully aware, it is certainly
necessary to follow the criteria considered appropriate to just ac-
quisition from nature.
Moreover (and this will be of some importance later in this
paper), it does not seem necessary to choose between either adopt-
ing the finders, keepers ethic absolutely or accepting it not at all. It
seems possible to view some kinds of “creation-by-discovery” as
conferring just natural ownership, while in other criteria (perhaps
those in which discovery was wholly accidental, or those in which
discovery by one came on the heels of years of exhausting search by
another), one may not be so prepared to recognize the actual
discoverer as the sole just keeper of his find. Certainly the case in
which the ethics of finders’ remaining keepers might be invoked
calls for systematic analysis and classification. Our purpose has
merely been to emphasize the possible role which a finder-creator
view of discovery can play in a theory ofjustice.

ENTREPRENEURIAL DISCOVERY AND CREATIVITY

We turn to consider the second of the building blocks referred to

earlier: the economic insight that the discovery of a hitherto un-
known market use for an already-owned resource or commodity
 Entrepreneurship, Entitlement, and Economic Justice 397

constitutes the discovery of a hitherto un-owned element associated

with that resource or commodity. In the conventional view (appar-
ently shared by Nozick), once a unit of resource has been acquired,
ownership has been established in it with respect to all its proper-
ties and powers, whether these have been known or imagined, or
not. In the view being now considered, on the other hand, those
aspects of a thing which are unknown, remain, so-to-speak, non-
existent. Their discovery constitutes the discovery of a hitherto
unknown, “non-existent,” and hence un-owned dimension of the
thing. An owner owns only those aspects of “his” property of which
he is aware. Acceptance of this way of viewing the matter has
far-reaching implications for the perception of the entrepreneurial
role in the market.
The entrepreneur perceives and exploits opportunities in the
market which others have not noticed. He discovers, for example,
that a quantity of oranges is being sold (for eating purposes)
throughout the market at $5, while consumers would gladly pay a
total of $12 for these same oranges converted (at a total manufac-
turing cost, above that of the oranges, of $4) in the form of orange
juice and marmalade. Entrepreneurial discovery of the $3 profit
opportunity—of buying oranges for $5 and selling them for $8 (i.e.,
the $12 obtainable from the sale of juice and marmalade less the
other costs of $4)—represents, in the view under present discus-
sion, the discovery of $3 value in the oranges which did not
previously exist. Up to the moment when the entrepreneur's vision
“saw the juice and marmalade which the oranges represent,
oranges had value only for eating—a value which the market set at
$5. The entrepreneur has discovered $3 additional value in the
oranges. He may, then, be held to have “created” this additional
value in these oranges.*4 It is as if the enterpreneur found orange
juice and marmalade in nature, where no one had perceived their
existence; he has “created” the orange-resource that can provide
juice and marmalade.
Pursuing the matter further, it may be held that any price
differential discovered and exploited by the entrepreneur consti-
tutes the discovery of hitherto unknown and non-existent value
(even where no new, physically different, use is entailed). If orange
juice can be bought at $3 (in one market) and sold at $4 (with no
additional costs involved) in a second market, this means that those
398 Israel M. Kirzner

who were buying and selling at $3 did not know of the presence of
those sufficiently eager for juice to be willing to pay $4 for it.
Entrepreneurial discovery of this may, then, be seen as the dis-
covery in the first market of a hitherto unsuspected intensity of
potential value in orange juice. The entrepreneur may be held to
have “created” this additional value by introducing these oranges to
the second market.
It should be observed that this view of entrepreneurial discovery
and creativity arises out of an understanding of the entrepreneurial
role in a strictly “arbitrage” sense.*° In this view the entrepreneur
adds nothing to the production process other than his alertness to
the production possibilities already existing. He provides no “ser-
vices,” managerial or other; he simply notices that inputs can be
obtained at a total outlay less than the sales revenue obtainable from
output. We see the entrepreneur as ‘creator’ not in the sense of the
physical producer, but strictly in the sense of his being the dis-
coverer of an available opportunity.

ENTREPRENEURSHIP AND THE EXPLOITATION OF

ERROR

Acceptance of the finders-keepers ethic and of the economic insight

into entrepreneurial discovery discussed in the preceding sections,
clears away the difficulties surrounding the justice of disequilibrium
market transactions to which attention has been drawn in this paper.
The central feature which distinguishes the market in disequlibrium
from the model of market equilibrium can, after all, be stated in
terms of the scope open for entrepreneurship. In the equilibrium
model all profitable opportunities have been already discovered and
exploited, nothing remains for entrepreneurs to discover and to
create. In the disequilibrium market, it is precisely the changes
introduced by entrepreneurial discovery of existing errors (and the
consequent opportunities for profit), which constitute the market
process (and which are, in fact, the meaning of the label describing
the market as a disequilibrium one).3§
It follows that entrepreneurial profits captured during the dis-
equilibrium market process can be defended as “new” value which
entrepreneurs have discovered (and thus “created”). The equilib-
rating market process is thus perceived not simply as economic
 Entrepreneurship, Entitlement, and Economic Justice 399

theory has traditionally shown, as a process tending to correct the

“misallocation of resources,” but as a process of the continued net
creation of values—as it were ex nihilo—as goods tend to move from
lower valued to higher valued ones.
But what of the difficulty, spelled out in earlier portions of this
paper, that such entrepreneurial activity in disequilibrium has
involved transactions with trading partners who would never have
bought or sold (at the prices they accepted) had they known the true
state of the market? What of the possibility that these transactions,
having been made in error, lack the critical element of true volun-
tariness, that the consent given to those sales and purchases was in
reality no consent at all? Reflection shows that, given the basis of the
views discussed on the preceding pages, these difficulties no longer
obtrude.
If a man sells oranges (knowing full well their usefulness for
marmalade and juice) because of some serious misunderstanding on
his part, we may wish to say that the sale “really” lacked consent and
is thus invalid. The oranges (including their potential in producing
juice and marmalade) were his; without genuine consent they
cannot justly become owned by another. But consider the man who
sells oranges for $5, because he is unaware that, as potential raw
material in producing juice and marmalade they are worth $8 to the
entrepreneur to whom he sells. The shadow clouding his consent
arises from the existence of the $3 of additional value concerning
which he is ignorant. But we have seen that this additional $3 value
may well be held never to have been possessed by the seller at all.
This $3 value was discovered (indeed created) by the entrepreneurs
purchase and subsequent sale. So that the error on the part of the
seller (on the basis of which we sought to invalidate the sale) can, on
the present view of things, hardly be held to affect the conclusive-
ness of the consent of the seller to the sale of that which was his to
sell in the first place.
It thus turns out (not at all accidentally) that the reasoning which
justifies pure entrepreneurial profit (on the grounds that the entrep-
reneur has “created” previously non-existent value), is at the same
time able to protect the purchaser, or the seller, of any good at a
disequilibrium price from the charge that it was sold to him or
purchased by him only on the basis of error (and hence of flawed
consent). Market transfers are just, on this view, because no one
400 Israel M. Kirzner

consents voluntarily to a transaction except insofar as it gives him a

satisfactory exchange for that which he sees himself as giving up.
And until someone discovers that what is given up is more than the
owner sees, no more than that exists, in the sense relevant to this
view of economic justice.
It should be observed that this justification of entrepreneurial
alertness to the errors made by others does not extend to the
justification of fraud, properly defined. Nor does it necessarily rule
out the possible view that at least some cases of non-fraudulent
exploitation of error be considered morally questionable. Fraud is
not covered (by the reasoning which justifies market transfers)
because fraud involves the deceitful inducement of error (either
positively, or where a fiduciary-type relationship exists—tacitly)
on the basis of which consent is fraudulently obtained. And even
some cases of non-fraudulent exploitation of error may be condemned
despite an otherwise general acceptance of the justification of
disequilibrium market transfer we have discussed. It seems entirely
possible (as noted earlier) to restrict one’s adoption of the “finders,
creators, keepers’ ethic to some kinds of entrepreneurial finds and
creations only. Perhaps one may feel that to take advantage of one’s
prior knowledge of information that will be commonly known in 5
years time is justified, but that exploiting one’s knowledge of that
which everyone will surely know in 5 minutes is going too far.37 The
point of our discussion has not been to show that all possible forms of
entrepreneurial exploitation of error sanctioned by the law are
rendered immediately morally acceptable by application of the
finders-keepers ethic. The purpose has been to show that this ethic
can plausibly be deployed to rebut possible blanket-condemnation
of market processes on grounds of error and consequent lack of
genuine consent.%8

MODIFICATIONS IN THE ENTITLEMENT THEORY

We are now in a position to spell out the way in which Nozick’s

entitlement theory calls for modification, if it is to serve effectively
to demonstrate the possible justice of the market system. The
entitlement theory maintains the following definitions: (a) a distribu-
tion of holdings is just if everyone is entitled to the holdings he
possesses under the distribution; (b) one is entitled to a holding only
 Entrepreneurship, Entitlement, and Economic Justice 401

(i) if he has acquired it from the unheld state in accordance with the
principle of justice in acquisition, or (ii) if he has acquired it in
accordance with the principle of justice in transfer, from someone
else entitled to the holding.? Our discussion calls for modification
of Nozick’s view that these latter definitions under (b) “exhaustively
cover the subject of justice in holdings.”4°
For Nozick, the justice of holdings depends “historically” on the
justice of the original acquisition from the unheld state, and on the
justice of each of the subsequent transfers of the holding. Our
discussion of the finders-keepers ethic, and its application in the
justice of entrepreneurial creation, indicates first, that Nozick’s
definitions have not definitively covered all cases of holdings that
may be held just, and, again, that the lines drawn by Nozick
between original acquisition and acquisition by transfer, are not as
sharp as Nozick’s discussion suggests.
The framework of Nozick’s definitions sees things as being held
either as the result of original acquisition from an unheld state, or
else as the result of acquisition by transfer from a previous holder.
Our discussion has pointed out a third possibility: that of a thing
being held as the result of the holder's having, in the relevant sense,
“created” it ex nihilo—i.e., by finding it.44 To be sure, the possibil-
ity that a thing has been, at one level of discourse, “created” from
the state of “non-existence,” does not preclude its having been
acquired (either “originally” or by transfer) from what, at a different
level of discussion, is treated as an earlier state of existence. Oil
discovered in an unsuspected location may, at one level, be treated
as not having existed before; but at another level, it already did exist
before. Nozick’s schema is certainly an exhaustive one at this latter
level; but our discussion has shown that discourse may be fruitful
when conducted at a level at which the third possibility we have
mentioned enters as an important additional class of holdings.4?
And recognition of this possibility, the holding of a thing as a
result (not of its acquisition from the unheld state, or from a
previous holder, but) of its having been “created,” introduces a
certain fuzziness in the sharpness of the line drawn by Nozick
between holdings resulting from original acquisition and those
resulting from acquisition by transfer. In Nozick’s schema transfer
involves only the acquisition of a previously held thing. For us
transfer may well involve (besides the acquisition of that which was
402 Israel M. Kirzner

already previously held) the “creation” ofan entirely new dimension

of the holding—something not only, not previously held, but also
something that did not, in the relevant sense, exist previously
altogether.43 It is recognition of this complexity in transfers, espe-
cially in market transfer, which has enabled us to perceive the
possible justice of the entrepreneurial discoveries that may be
expressed in disequilibrium market purchases and sales.
Another implication of our discussion, for the entitlement theory,
relates to the justice of original acquisition. As we will see in the
following section, it appears that some of Nozick’s views which flow
from his treatment of the justice of original acquisition, arise from
his decided lack of enthusiasm for the possibility that many cases of
original acquisition may qualify, at least in part, for justification
under the finder-creator, finder-keep ethic. To the implications of
this possibility we now turn.

NOZICK AND THE LOCKEAN PROVISO

As mentioned earlier in the paper, Nozick has not, in his book,

attempted to spell out in detail the proper principles of justice in
original acquisition or in acquisition by transfer. Nonetheless,
Nozick has devoted a good deal of attention to what he has termed
the “Lockean Proviso.” And, while Locke himself enumerated the
“proviso” in relation to original acquisition from nature, Nozick has
pursued its implications insofar as it introduces complications into
the justice of acquisition by transfer.
Locke's theory of justice in original acquisition as requiring only
that the would-be expropriator of an un-owned object mix his labor
with it, was qualified by the proviso that there be “enough and as
good left in common for others.”44 Nozick explains that by this
qualification Locke meant “to ensure that the situation of others is
not worsened.’4° While Nozick sharply limits the scope of the
Lockean proviso as it enters into his own entitlement theory of
justice,4° he does, without hesitation, strongly accept the principle
that justice in original acquisition requires that such acquisition shall
leave no one else in a worse situation than he would have been
without it. In fact, Nozick seems almost relieved to be able to invoke
this principle in order to deal with cases (involving appropriation of
the entire stock ofa limited, life-giving resource), to which critics of
 Entrepreneurship, Entitlement, and Economic Justice 403

the private property system have traditionally pointed as

exemplifying the injustice of the system. The case of “someone who
comes upon the only water in the desert several miles ahead of
others who also will come to it and appropriates it all’47 is a violation
of the “proviso” surrounding original acquisition. And even, Nozick
argues, where one appropriates only one of many water holes in the
desert, if it subsequently happens that all the other holes dry up,
the Lockean proviso stringently limits what he can do with “his”
hole;
The emphasis which we have placed in this paper on the role of
discovery and “creation” in justifying title to a holding enables us to
question the rather sweeping scope which Nozick, at least, in
principle, assigns to the Lockean proviso. (In practice, Nozick be-
lieves “that the free operation of a market system will not actually
run afoul of the Lockean proviso.”)49 Clearly, once we admit that
the discovery of an unknown thing justifies the holding of it by its
finder on the grounds that, in the relevant sense, he “created” it,-as
it were ex nihilo, the entire basis of the Lockean proviso becomes
vulnerable to challenge. Is it really true that, where a discoverer
appropriates all of a limited deposit of resource, he is worsening the
situation for others—for whom this deposit was completely un-
known and “nonexistent?” This question seems so obviously to call
for a negative answer that indeed Nozick finds himself forced by it to
accept, in effect, a limited finders-keepers ethic. Nozick cir-
cumscribes the Lockean proviso by observing that where a re-
searcher synthesizes a new substance (out of easily available raw
materials) he may justly refuse to sell except on his terms since by so
doing he does not harm others (who are free to do what he had
done). Moreover, Nozick adds, one who appropriates the total
supply of anew substance by finding it “in an out-of-the-way place”
has not worsened the situation ofothers: “ifhe did not stumble upon
the substance no one else would have.” But Nozick immediately
qualifies this by pointing out that in the latter case, “as time passes,
the likelihood increases that others would have come across the
substance,” justifying, Nozick suggests, possible limitation on be-
quest by the first discoverer.°°
Nozick’s limited recognition of the exemption of discovery from
the Lockean proviso does not appear to go nearly far enough. For
Nozick even appropriation of an object following its discovery may,
404 Israel M. Kirzner

we have seen, be considered to worsen the situation of those

(possibly in later generations) who “would have” found the object for
themselves. But our insight into the “creative” aspect of discovery
suggests a different view of the matter, on two separate grounds.
First, we must maintain that one who might at a given date have
“created” an object ex nihilo has hardly been hurt by the fact that a
second individual in fact “created” the object first, at an earlier date.
It was the latter individual who was the “creator” not the former. A
finders-‘creators,” finder-keepers ethic cannot, it must be main-
tained, confer any claim on those who might have—but did not in
fact—‘create.” Nozick’s concern for the “harm” done to those who
would have themselves discovered the new substance, is based on
the view that, whether discovered or not, the new substance has
always “existed” (both for present and future generations). So that
the basis of Locke’s proviso—that others have some claim on
un-owned objects, requiring that they therefore not be harmed by
appropriation—applies also to undiscovered substances. But if we
recognize that an undiscovered substance does not in the relevant
sense, exist for those who are not aware of it, then Nozick’s concern
loses its justification.
But, perhaps, even more important is a second reason why we
cannot share Nozick’s view that the discoverer of anew substance is
justified to the holding of it only to the extent that it would not have
been discovered by others. For Nozick, Lockean acquisition of an
unowned object from nature seems to be held justified only in the
negative sense that such appropriation (where, of course, it does not
violate the “proviso”) has not harmed others. That is, one who has
mixed his labor with the un-owned resource (of which there is ample
left for others) has not acquired just title on the basis of any powerful
positive moral claim. He has acquired title because mixing one’s
labor with the resource is the act of appropriation—and compliance
with the Lockean proviso ensures that no injustice is involved in this
acquisition. But acceptance of a finders-“creators,” finders-keepers
ethic confers just title on the discoverer-“creator” not in the nega-
tive sense (that such title involves no injustice to others) but in the
positive sense that justice requires that the “creator” be recognized
as the owner of what he has “created”; to deny the “creator” title
would be to inflict injustice on him. From this view of the ethics of
“creation” it is by no means clear that Locke’s proviso has necessary
 Entrepreneurship, Entitlement, and Economic Justice 405

relevance to discovery at all. Ifjustice requires that the “creator” of

an object be recognized as its owner, then this may remain true
even if it might be shown that others (who might, say, have
otherwise discovered the object for themselves) can be considered
as having been rendered worse off by the “creation.”
The considerations advanced in the preceding paragraphs tend to
exempt from the Lockean proviso a substantial proportion of the
cases—all those involving discovery—which Nozick includes under
the heading of“original acquisition from the unheld state.” (On this
basis, the troublesome water-hole-in-the-desert cases must, where
they have involved discovery, be viewed as indeed involving no
violation of strict justice—despite whatever other moral structures
one may invoke to criticize selfish behavior on the part of a just
owner, especially in situations involving threat to life.) Moreover, it
should be pointed out that consistent application of reasoning
developed in an earlier section of this paper suggests that the
Lockean proviso cannot claim necessary relevance even for cases in
which discovery does not seem obviously to be involved. It was
argued (in earlier sections of this paper) that genuine entrepreneu-
rial discovery and “creation” may occur even with respect to objects
whose existence is known to all.
Consider then the case (referred to only by implication, in
Nozick’s discussion) of the unheld sole water hole in the desert
(which everyone in a group of travellers knows about), which one of
the travellers, by racing ahead of the others, succeeds in appropri-
ating. For Nozick this case, involving as it does no discovery at all,
clearly and unjustly violates the Lockean proviso: the other travel-
lers who in the absence of appropriation by their fellow, would have
all enjoyed some water without cost, are now forced to pay a price
(even a “monopoly price”) for that same water. For us, however,
this view is by no means the only one possible. We notice that the
energetic traveller who appropriated all the water was not doing
anything which (always ignoring of course, prohibitions resting on
the Lockean proviso itself) the other travellers were not equally free
to do. The other travellers, too, could have raced ahead. Assuming
(for simplicity) that all the travellers were of equal strength and
speed there would have ensued a “gold-rush” in which each would
have, let us say, captured some water. As it happened, the other
travellers did not bother to race for the water. May it not be that
406 Israel M. Kirzner

they were less alert, entrepreneurially, to the possibility that some-

one else might indeed appropriate all of the water than the energe-
tic traveller? Should we not, then, say that the latter was the first to
“discover” the true market value of the unheld water? For the
others the water was indeed known, but the worthwhileness of its
appropriation was not known. (Perhaps they mistakenly thought
there was more water available than could possibly be drunk;
perhaps they mistakenly thought that no one would or could race
across the desert at a faster speed than that at which they were
travelling, or perhaps they gave the water no thought at all.) It does
not seem obvious that these other travellers can claim that they
were hurt by an action which they could themselves have easily
taken, had they been as alert as the successful appropriator. What,
one must ask, even under conditions involving the appropriation of
known substances—is so obviously acceptable about the Lockean
proviso, as interpreted by Nozick?

THE JUSTICE OF THE MARKET

It turns out, then, that the insight into entrepreneurial discovery,

which we have discussed earlier, coupled with the possibility of a
finders-keepers ethic, have not only solved the difficulties raised in
the first half of this paper, but have enabled us to perceive possible
justification for the free operation of a market system with fewer
qualifications than those Nozick, on the basis of the Lockean pro-
viso, was impelled to introduce.
A finders-keepers ethic, we have observed, may not be found
compelling. And even if the basic idea is accepted, there remains
ample room for moral reservations concerning particular cases of
application of the ethic. Nonetheless there does seem to be a certain
plausibility in the notion of ownership through creativity. It is this
plausibility which may help explain how so many observers of the
market appear to find it consistent with economic justice in the face
of the denunciations of the moralist critics of capitalism. This paper
has explored the sources of this apparent plausibility, and has
scrutinized its ability to serve as possible support for the morality of
the market. For this purpose Nozick’s entitlement theory has served
as a crucially important framework. That our discussion has
 Entrepreneurship, Entitlement, and Economic Justice 407

suggested certain modifications in the framework itself, has, it is

hoped, been one of the paper's positive contributions.

NOTES

1. R. Nozick, Anarchy, State, and Utopia (New York: Basic Books,

1974), pp. 178-182.
DO DCA pe loo
3. See Nozick, op. cit. pp. 160-164. See also Nozick’s extensive discus-
sion of the voluntariness of market transactions, pp. 262-265.
4. See the discussion of this point in J. S$. Mill, Principles of Political
Economy (Ashley Edition, London: Longmans, Green and Co., 1909), pp.
796f, M. N. Rothbard, Power and Market (Menlo Park: Institute for
Humane Studies, 1970), p. 34; H. B. Acton, The Morals of Markets
(London: Longmans, 1971), p. 4; Nozick, op. cit. p. 152. See also the
discussion in G. Tullock, The Social Dilemma (Blacksburg: University
Publications, 1974), p. 11.
5. F. A. Hayek, “Economics and Knowledge,” Economica IV (1937), in
Individualism and Economic Order (London: Routledge and Kegan Paul,
1949), p. 42.
6. L. von Mises, Human Action (New Haven: Yale University Press,
1949), p. 249. See also F. H. Hahn, On the Notion of Equilibrium in
Economics (Cambridge: Cambridge University Press, 1973), pp. 7ff.
7. It ought perhaps to be emphasized that it is not only a relatively small
class of business entrepreneurs the justice of whose acquisitions is being
questioned. As von Mises pointed out (Human Action, p. 253) every acting
individual displays entrepreneurial characteristics. Since almost all market
transactions involve less than perfect omniscience, the errors which they
reflect might seem to cast a shadow over the justice of all market transfers.
Pure entrepreneurial profit may be present, to some extent, in any sale and
in any purchase.
8. For some discussion of the morality of profits, and of the arguments of
its critics, see H. B. Acton, The Morals of Markets (London: Longmans,
1971), Chapter 2.
9. See von Mises, “Profit and Loss,” in Planning for Freedom (South
Holland: Libertarian Press, 1952), pp. 108ff; see also I. M. Kirzner,
Competition and Entrepreneurship (Chicago: University of Chicago Press,
1973), pp. 85ff.
10. To do this is again to deliberately accept an uncertainty viz. the
possibility that, without expenditure of the resources needed to remove the
possibility of error, error may occur concerning which accurate knowledge
408 Israel M. Kirzner

is simply unavailable. There is no way, without expenditure of those

resources, to know whether error will or will not occur.
1l. We prefer, for present purposes, not to press this position in even
more uncompromising form—which might maintain that one who has been
deceived has, again, merely refrained deliberately from spending the
resources necessary to ensure against deception.
12. I. M. Kirzner, “Economics and Error” (unpublished paper presented
at Austrian Economic Symposium, Windsor Castle, September 1976). The
view denying scope for error within economic analysis has been stated by
G. J. Stigler, “The Xistence of X-Efficiency,” American Economic Review,
(March, 1976), pp. 213-216.
13. Despite our insistence on the prevalence of genuine error, we must
of course readily grant that many cases in which an entrepreneur appears to
be exploiting the ignorance of others, do result merely from the deliberate
judgment of others (in the face possibly of the very same information
available to the entrepreneur) concerning a future which, they are well
aware, is highly uncertain.
14. G. S. Bower and A. K. Turner, The Law of Actionable Misrepresen-
tation (London: Butterworths, 1974), p. 104.
15. S. E. Williams, Kerr on Fraud and Mistake (6th Edition, London:
Sweet and Maxwell, 1929), p. 76.
16. For a fascinating pioneer discussion of both the morality and the law
surrounding exchanges made on the basis of incomplete or faulty informa-
tion, see Gulian C. Verplanck, An Essay on the Doctrine of Contracts:
Being An Inquiry How Contracts are Affected in Law and Morals by
Concealment, Error, or Inadequate Price (New York, 1825). Verplanck
shows remarkable awareness of the element of pure (entrepreneurial) profit
in all exchanges, and of how this implies the impracticality ofalegal system
which would insist on “full disclosure” in all exchanges.
17. Bower and Turner, op. cit. p. 106.
18. Williams, loc. cit., An oft-cited opinion of Cardozo does, however,
appear to argue that the law simply reflects the “morals of the market
place”; so that where (as in the case of partners) morality calls for loyalty to
one another, or where (as in the case of trustees) morality calls for not
“honesty alone, but the punctilio of an honor the most sensitive. . .” this
should and will find expression in the law. (See the quote in H. Manne,
Insider Trading and the Stock Market (New York: Free Press, 1966), pp.
20-21.
19. See von Mises, Human Action, p. 147. For a recent plea not to reach
conclusions concerning the morality of particular economic practices
(specifically, insider trading) before thoroughly exploring the welfare con-
sequences of these practices, see Manne, op. cit. p. 15.
 Entrepreneurship, Entitlement, and Economic Justice 409

20. See e.g. Manne, op. cit., p. 18.

21. W. J. Ashley on the “Just Price” in R. H. I. Palgrave (Editor),
Dictionary of Political Economy, (London: MacMillan, 1896), Vol. II, p.
500.
22. Nozick, op. cit. p. 262 (italics in original).
23. Nozick, p. 161.
24. B. W. Dempsey, “Just Price in a Functional Economy,” American
Economic Review, (September, 1935), reprinted in J. A. Gherity (Editor)
Economic Thought, A Historical Anthology (New York: Random House,
1965); R. de Roover, “The Concept of the Just Price: Theory and Economic
Policy,” Journal of Economic History (December, 1958), in Gherity, op.
cit.; J. A. Schumpeter, History of Economic Analysis (New York: Oxford
University Press), Part 2, Chapter 2; J. T. Noonan, The Scholastic Analysis
of Usury (Cambridge: Harvard University Press, 1957); S$. Hollander, “On
the Interpretation of the Just Price,” Kyklos 18:4, (1965); M. N. Rothbard,
“New Light on the Prehistory of the Austrian School,” in E. Dolan (Editor),
The Foundations of Modern Austrian Economics (Kansas City: Sheed and
Ward, 1976).
25. See e.g. the view of Cajetan, cited in de Roover, op. cit. p. 29.
26. S. Hollander, op. cit. p. 625.
27. See de Roover, op. cit. p. 28; Hollander, op. cit., p. 624.
28. See Hollander, ibid.
29. H. M. Oliver, A Critique of Socioeconomic Goals (Bloomington:
Indiana University Press, 1954), p. 42.
30. See the writer's “Producer, Entrepreneur, and the Right to Prop-
erty, Reason Papers (Fall, 1974), and its discussion of the views of Oliver
(op. cit.).
31. It seems particularly important to distinguish the “finders, keepers”
ethic from the ethic of “first come, first served” which Professor Vickrey, for
one, has described as of dubious equity. (W. Vickrey, “An Exchange of
Questions between Economics and Philosophy,” 1953, reprinted in E. S.
Phelps (Editor), Economic Justice, Penguin Books, 1973, p. 58.)
32. The notion of the discoverer as creator should, of course, be linked
with the view of F. H. Knight that it is not the factors of production which
produce the output, but the entrepreneur who decides to enlist the services
of these factors. (F. H. Knight, Risk, Uncertainty, and Profit, 1921, p. 271).
See also F. B. Hawley, Enterprise and the Productive Process, (New York:
Putnam, 1907), pp. 85, 102, 112, 127.
33. Below we discuss the (very limited) recognition of the ethical
significance of discovery in Nozick’s system.
34. It was Schumpeter (Theory of Economic Development, Harvard
University Press, 1934, translated from the German work published in
410 Israel M. Kirzner

1911) who emphasized the creative role of the entrepreneur. It should

perhaps be noted that the writer has elsewhere demurred from Schumpe-
ter’s view of the entrepreneur as disrupting earlier states of response (I. M.
Kirzner, Competition and Entrepreneurship, pp. 72f). The position taken
here in the text is not, the writer believes, inconsistent with his earlier
insistence on the equilibrating role of the entrepreneur, seen as responding
to the existence of as yet unexpected opportunities. For individuals in a
society, of whom none has as yet perceived the existence of a profitable
opportunity, this opportunity may properly be said not yet to exist: its first
discoverer may be properly seen as its creator. Nonetheless the theorist
analyzing the social role of entrepreneurial discovery is surely entitled to
point out that, prior to the discovery of a productive opportunity, its
“existence” did mean that society had failed in some sense to achieve its
greatest possible level of output. It is not improper, at this level of
discourse, to insist on the entrepreneur as responding to the opportunities
“out there,” of which he becomes aware.
35. See note 9 above.
36. See Kirzner, Competition and Entrepreneurship, Chs. 1, 2.
37. Samuelson has derided the social utility of the vast speculators’
profits won by entrepreneurs several seconds more nimble than their
fellows in assessing new information (Weltwirtschaftliches Archiv, 79,
December, 1957, p. 209). One may question the validity of Samuelson’s
argument at the utilitarian level (see e.g. Kirzner, Competition and Entrep-
reneurship, p. 224; also “Producer, Entrepreneur, and the Right to Prop-
erty,” Reason Papers, Fall, 1974, p. 13), while recognizing the possibility
that one may not choose to extend one’s finders, keepers ethic (if, indeed,
one subscribes to it at all) to defend the morality of the case discussed by
Samuelson.
38. To put the point somewhat differently, the discussion in the text has
sought to show that all market transactions can be seen as in no way
involving imperfect consent. This was done on the basis of the insight that
all entrepreneurial gain reflects, not exploitation of error, but creation (i.e.
“finding”) of new value. At the same time this may not, of itself, be seen as
justifying all cases of such gain, since one may not wish to recognize the
justice in which every finder, under all conceivable circumstances, becomes
a keeper.
39. Nozick, pp. 150-151.
40. Nozick, p. 151.
41. See the following section for a discussion of the (limited) degree to
which Nozick appears ready to recognize an ethical role for pure discovery.
42. For the biblical version of an entitlement theory based on Divine
Creation, see Psalms 24:1, 2: 115:16.
 Entrepreneurship, Entitlement, and Economic Justice 411

43. In objection, either to the argument of the text itself, or to its

application of the finders, keepers ethic, it can be argued that if I discover
new, hitherto unnoticed value in my neighbor's property, that additional
value ought, on the finders, keepers ethic, be mine without any transfer of
the property at all. To the extent that the “new value” can be consumed
without violating the rights of the neighbor, this seems not unreasonable.
(One thinks of the legal questions raised when it was discovered, through
the invention of flying, that air rights over land were more valuable than
hitherto realized.) In general, where consumption of the newly discovered
value cannot occur without violating existing rights, it will be in the interest
of the discoverer to buy up those rights, in order to enjoy the new values
which he has discovered.
44John Locke, An Essay Concerning the True Origin, Extent, and End of
Civil Government, Chapter V, section 27.
45. Nozick, p. 175.
46. Nozick, p. 178.
47. See Nozick, pp. 179-180, and the footnote on p. 179.
48. Nozick’s position on this point has potentially far-reaching implica-
tions for the justice of monopoly positions in production obtained through
sole acquisition of needed resources. For a discussion not involving the
issue of the justice of such monopoly cases, but from the perspective oftheir
alleged harm to society, see Kirzner, Competition and Entrepreneurship,
pp. 236-242, (where various considerations are weighed against one
another.) The discussion in the subsequent paragraphs in the text (ques-
tioning Nozick’s views) points, on the other hand, not to any vindication of
monopoly producers from possible charges that their activity in some sense
harms society, but to a defense of the moral legitimacy of monopoly
resource ownership (even were it to be shown that the rest of society would,
in some sense, be better off were such monopoly not to be present).
49. Nozick, p. 182.
50. Nozick, p. 181.
 Selected Bibliography

Arrow, K. J. 1978. “Extended Sympathy and the Possibility of Social

Choice,” Philosophia (Israel) 7, pp. 223-37.
. 1978. “Nozick’s Entitlement Theory ofJustice,” Philosophia
(Israel) 7, pp. 265-79.
Axinn, S. 1978. “The Law of Land Warfare as Minimal Government,
Personalist 59, pp. 374-85.
Barber, B. R. 1977. “Deconstituting Politics—R. Nozick and
Philosophical Reductionism,” Journal of Politics 39, 1, pp. 2-23.
Barnett, R. E. 1977. “Whither Anarchy: Has Robert Nozick Justified
the State?” Journal of Libertarian Studies 1, pp. 15-21.
Bernick, M. 1978. “A Note on Promoting Self-Esteem,” Political
Theory 6, pp. 109-18.
Biesenthal, L. 1978. “Natural Rights and Natural Assets,” Philosophy
of the Social Sciences 8, pp. 153-71.
Blackstone, W. T. 1978. “The Minimal State: An Assessment of Some
of the Philosophical Grounds,” Personalist 59, pp. 333-43.
Bowie, N. E. 1979. “Welfare and Freedom,” Ethics 89, pp. 254-68.
Buchanan, A. 1975. “Distributive Justice and Legitimate Expecta-
tions,” Philosophical Studies 28, pp. 419-25.
Childs, R. A., Jr. 1977. “The Invisible Hand Strikes Back,” Journal of
Libertarian Studies 1, pp. 23-33.
Christie, G. C. 1978. “The Moral Legitimacy of the Minimal State,”
Arizona Law Review 19, 1, pp. 31-43.
Coburn, R. 1976. “Relativism and the Basis of Morality,” Philosophi-
cal Review 85, pp. 87-93.
Cohen, G. A. 1977. “Robert Nozick and Wilt Chamberlain: How
Patterns Preserve Liberty,” Erkenntnis 11, pp. 5-23.
Coleman, J. S. 1978.“The Balance Between Rights Individually Held
and Rights Collectively Held,” Arizona Law Review 8, 1, pp.
180-92.
Danielson, P. 1978. “Taking Anarchism Seriously,” Philosophy of the
Social Sciences 8, pp. 137-52.

412
 Selected Bibliography 413

Danley, J. R. 1978. “An Examination of the Fundamental Assump-

tion of Hypothetical Process Arguments,” Philosophical Studies 34,
pp. 187-95.
. 1979. “Robert Nozick and the Libertarian Paradox,” Mind
88, 351, pp. 419-23.
Darwall, S. L. 1977. “Two Kinds of Respect,” Ethics 88, 1, pp.
3649.
Davis, L. 1976. “Comments on Nozick’s Entitlement Theory,” Jour-
nal of Philosophy 73, pp. 836-44.
Davis, M. 1977. “Necessity and Nozick’s Theory of Entitlement,”
Political Theory 5, pp. 219-32.
Degregori, T. R. 1979. “Market Morality—Robert Nozick and the
Question of Economic-Justice,” American Journal of Economics
and Sociology 38, 1, pp. 17-30.
Den Uyl, D., and Rasmussen, D. 1978. “Nozick on the Randian
Argument,” Personalist 59, pp. 184-205.
Diggs, B. J. 1977. “Liberty Without Fraternity,” Ethics 87, pp.
97-112.
Exdell, J. 1977. “Distributive Justice: Nozick on Property Rights,”
Ethics 87, pp. 142-49.
Flemming, A. 1978. “Using a Man as a Means,” Ethics 88, pp.
283-98.
Gill, E. R. 1978. “Responsibility and Choice in Robert Nozick: Sins of
Commission and of Omission,” Personalist 59, pp. 344-57.
Goldman, A. H. 1976. “The Entitlement Theory of Distributive
Justice,” Journal of Philosophy 73, pp. 823-35.
Gorr, M. 1977. “Nozick and the Opposed Preferences Theory of
Exchange,” Theory and Decision 8, pp. 289-92.
. 1977. “Nozick’s Argument Against Blackmail,” Personalist
58, pp. 187-91.
Harris, C. E. 1979. “Kant, Nozick, and the Minimal State,” South-
western Journal of Philosophy 10, 1, pp. 179-87.
Held, V. 1976. “John Locke on Robert Nozick,” Social Research 43,
pp. 169-95.
. 1978. “What is Minimal Government?” Personalist 59, pp.
405-7.
Henley, K. 1978. “Children and the Individualism of Mill and
Nozick,” Personalist 59, pp. 415-19.
Hodson, J. D. 1978. “Nozick, Libertarianism, and Rights,” Arizona
Law Review 8, 1, pp. 212-27.
Holmes, R. L. 1977. “Nozick on Anarchism,” Political Theory 5, pp.
247-56.
414 Selected Bibliography

Johnson, K. 1976. “Government by Insurance Company: The An-

tipolitical Philosophy of Robert Nozick,” Western Political Quar-
terly 29, 2, pp. 177-88.
Kearl, J. R. 1977. “Do Entitlements Imply that Taxation is Theft?”
Philosophy and Public Affairs 7, pp. 74-81.
King, C. 1976. “Seeing Through the Political Realm,” Western
Political Quarterly 29, 2, pp. 179-81.
Kirzner, I. 1978. “Entrepreneurship, Entitlement, and Economic
Justice,” Eastern Economics Journal 4, 1, pp. 9-25.
Ladenson, R. F. 1976. “Does the Deterrence Theory of Punishment
Exist? A Response to Nozick,” Philosophy Research Archives 2,
1090.
. 1978. “Nozick on Law and the State: A Critique,” Philo-
sophical Studies 34, pp. 437-44.
Lieberman, J. K. 1977. “The Relativity of Injury,” Philosophy and
Public Affairs 7, pp. 60-73.
Litan, R. E. 1977. “On Rectification in Nozicks Minimal State,”
Political Theory 5, pp. 233-46.
Lyons, D. 1975. “Welcome Threats and Coercive Offers,” Philosophy
50, pp. 425-36.
Lyons, D. B. 1977. “The New Indian Claims and Original Rights to
Land,” Social Theory and Practice 4, pp. 249-72.
. 1976. “Rights Against Humanity,” Philosophical Review 85,
pp. 208-15.
Machan, T. R. 1978. “Against Nonlibertarian Natural Rights,” Jour-
nal of Libertarian Studies 2, 3, pp. 233-38.
. 1979. “Considerations of the Libertarian Alternative,” Har-
vard Journal of Law and Public Policy 2, pp. 103-24.
. 1977. “Nozick and Rand on Property Rights,” Personalist 58,
pp. 192-95.
Mack, E. 1976. “Distributionism Versus Justice,” Ethics 86, 2, pp.
145-54.
. 1978. “Nozick’s Anarchism,” in J. Roland Pennock and John
W. Chapman (eds.). Nomos XIX: Anarchism, pp. 43-62. New
York: New York University Press.
Maneli, M. 1978. “Expanding the Functions of the State and the
Freedom of the Individual,” Personalist 59, pp. 424-27.
Moffat, R. C. L. 1978. “ ‘Minimal Government’: An Introductory
Appraisal,” Personalist 59, pp. 321-32.
Murphy, J. G. 1978. “Rights and Borderline Cases,” Arizona Law
Review 8, 1, pp. 228-41.
Norton, D. L. 1977. “Individualism and Productive Justice,” Ethics
87, pp. 113-25.
 Selected Bibliography 415

O'Connor, F. W. 1978. “Minimalist Presumptivism: A Response,”

Personalist 59, pp. 420-23.
O'Neil, P. 1979. “Inadequacy of Contract Theory in Robert Nozick,
Anarchy, State, and Utopia,” Personalist 60, 4, pp. 429-32.
O'Neill, O. 1976. “Nozick’s Entitlements,” Inquiry 19, pp. 468-81.
Paul, Ellen Frankel. 1979. “The Time-Frame Theory of Governmen-
tal Legitimacy,” Personalist 60, pp. 151-61.
Paul, J. 1977. “Nozick, Anarchism and Procedural Rights,” Journal of
Libertarian Studies 1, pp. 337-40.
. 1980. “The Withering of Nozick’s Minimal State,” Philosophy
Research Archives 6.
Pazner, E. A. 1978. “Entitlement Principles and the Original Position: A
Rawlsian Interpretation of Nozick’s Approach to Distributive Justice,”
Arizona Law Review 8, 1, pp. 169-79.
Perelli-Minetti, C. R. 1977. “Nozick on Sen: A Misunderstanding,” Theory
and Decision 8, pp. 387-93.
Quest, E. 1977. “ ‘Whatever Arises From a Just Distribution by Just Stee:
is Itself Just, ” Analysis 37, pp. 204-8.
Rabinowitz, J. T. 1978. “Emergent Problems and Optimal Solutions,”
Arizona Law Review 19, 1, pp. 61-157.
Rodman, J. 1976. “Analysis and History—Or, How Invisible Hand Works
Through Robert Nozick,” Western Political Quarterly 29, 2, pp. 197-201.
Rothbard, M. N. 1977. “Robert Nozick and the Immaculate Conception of
the State,” Journal of Libertarian Studies 1, pp. 45-57.
Ryan, C. C. 1977. “Yours, Mine, and Ours: Property Rights and Individual
Liberty,’ Ethics 87, pp. 126-41.
Sampson, G. 1978. “Liberalism and Nozick’s “Minimal State,’ ~ Mind 87,
pp. 93-97.
Sanders, J. T. 1977. “The Free Market Model Versus Government: A Reply
to Nozick,” Journal of Libertarian Studies 1, pp. 35-44.
Scaff, L. A. 1978. “How Not to Do Political Theory: Nozick’s Apology for
the Minimal State,” Arizona Law Review 8, 1, pp. 193-211.
Scanlon, T. 1976. “Nozick on Rights, Liberty, and Property,” Philosophy
and Public Affairs 6, pp. 3-25.
Schrag, F. 1976. “Justice and the Family,” Inquiry 19, pp. 193-208.
Senter, N. W. 1978. “Nozick on Property Rights. To Each According to
Marginal Productivity,” Arizona Law Review 8, 1, pp. 158-68.
Singer, P. 1975. “The Right to be Rich or Poor,” The New York Review of
Books 22, 3, pp. 19-24.
. 1976. “Why Nozick is not so Easy to Refute,” Western Political
Quarterly 29, 2, pp. 191-92.
Steiner, H. 1977. “Anarchy, State, and Utopia: Book Review,” Mind 86,
341, pp. 120-29.
416 Selected Bibliography

. 1974-75. “Individual Liberty,” Proceedings of The Aristotelian

Society 75, pp. 33-50.
. 1977. “Justice and Entitlement,” Ethics 87, 2, pp. 150-52.
. 1977. “The Natural Right to the Means of Production,” Philosophi-
cal Quarterly 27, pp. 41-49.
. 1978. “Nozick on Appropriation,” Mind 87, 345, pp. 109-10.
. Forthcoming. “Slavery, Socialism, and Private Property,” in J.
Roland Pennock and John W. Chapman (eds.) Nomos: Property.
New York: New York University Press.
Sterba, J. P. 1978. “In Defense of Rawls Against Arrow and Nozick,”
Philosophia (Israel) 7, pp. 293-303.
. 1978. “Neo-Libertarianism,” American Philosophical Quarterly 15,
pp. 155-21.
Thomson, J. J. 1978. “Some Ruminations on Rights,” Arizona Law Review
19, 1, pp. 45-60.
Tucker, D. 1979. “Nozick’s Individualism,” Politics 14, 1, pp. 109-21.
Williams, B. 1975. “The Minimal State,” Times Literary Supplement, pp.
4647.
Wolff, R. P. 1978. “Robert Nozick’s Derivation of the Minimal State,”
Arizona Law Review 19, 1, pp. 7-30.
Wood, D. 1978. “Nozick’s Justification of the Minimal State,” Ethics 88, 3,
pp. 260-62.
Yanal, R. J. 1979. “Notes on the Foundations of Nozick’s Theory of Rights,”
Personalist 60, 4, pp. 349-59.
 Contributors

Lawrence Davis is Associate Professor of Philosophy, University of

Missouri (St. Louis).
Douglas Den Uyl is Assistant Professor of Philosophy, Bellarmine College.
Robert L. Holmes is Professor of Philosophy, University of Rochester.
Israel M. Kirzner is Professor of Economics, New York University.
David Lyons is Professor of Philosophy, Cornell University.
Eric Mack is Associate Professor of Philosophy, Tulane University.
Thomas Nagel is Professor of Philosophy, New York University.
Robert Nozick is Professor of Philosophy, Harvard University.
Onora O'Neill is Professor of Philosophy at the University of Essex, Eng-
land.
Ellen Frankel Paul is Assistant Professor of Political Science, the
University of Colorado (Boulder).
Jeffrey Paul is Associate Professor of Philosophy, Bowling Green
State University.
Douglas Rasmussen is Assistant Professor of Philosophy, St. John’s
University.
Cheyney C. Ryan is Assistant Professor of Philosophy, University of
Oregon.
Thomas Scanlon is Professor of Philosophy, Princeton University.
Samuel Scheffler is Associate Professor of Philosophy, the University
of California (Berkeley).
Peter Singer is Professor of Philosophy, Monash University, Australia.
Hillel Steiner is on the Government Faculty of Manchester Univer-
sity, England.
| 417
418 Contributors

Judith Jarvis Thomson is Professor of Philosophy, M.I.T.

Bernard Williams is Provost of King’s College, Cambridge.
Robert Paul Wolff is Professor of Philosophy at the University of Massa-
chusetts (Amherst), and is currently Visiting Professor at Brandeis
University.
 1970s, Robert Nozick’s Wes Stale and tenia:
Sean anne been a rr cers’ OF 1e n
state and laissez-faire economy and include contr
Bernard Williams, Robert Paul Wolff, Judith nav: if
Thomas Nagel, Thomas Scanlon, David Lyons, as well
Robert Nozick himself.
____ Besides explicating the central themes of Nozick’s work
book provides the student with an excellent eee
contemporary political philosophy, showing thes ,
liness and its relevance to matters of eractival ra
concern. He

jeffrey Paul is associate professor of philosophy at

_ Green State University.
aan

Cover design by Helene Berinsky

omen

ISBN 0-8476-6280-2

ROWMAN ANID LIT

TUERIELD
Totowa, New Jiersey