Title: On a question of the tenability of capital punishment within the framework of social contract theory
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00102719/00001
 Material Information
Title: On a question of the tenability of capital punishment within the framework of social contract theory
Physical Description: Book
Language: English
Creator: Sarver, Vernon Thomas, 1943-
Copyright Date: 1994
 Record Information
Bibliographic ID: UF00102719
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 33027224
ltuf - AKL5552

Full Text







Copyright 1994


Vernon Thomas Sarver, Jr.


Several people have provided me with assistance and

encouragement at various stages of my work for the


Very early on, Professor Thomas Simon read and

critiqued papers of mine on Hobbes, Locke, and Rousseau, and

many of his comments and suggestions have been helpful to me

in the development of my chapters on these theorists.

In the early stages I also benefitted greatly from many

wide-ranging discussions I had with Professor Robert Ziller

of the Department of Psychology. I have brought many of the

insights I have gained from these conversations to my work

for the dissertation.

I would like to recognize the contribution of Professor

Charles Collier of the College of Law. At the outset of my

writing, Professor Collier met with me and offered many

helpful suggestions for bibliography. Over the course of my

work he remained available to me for consultation should I

have had occasion to stray into the area of legal theory.

Throughout the writing of this dissertation, I have had

the very good fortune to have the advice and counsel of

Professor Michael Radelet, a nationally recognized expert on

the sociology of the death penalty. I am particularly

grateful for his assistance in locating proper sources of

documentation for most of the empirical claims I make in the


Professor Thomas Auxter's expertise in the area of

Kantian studies has proved to be an invaluable source of

help and inspiration in my examination and critique of

Kant's arguments for the death penalty. It was also

Professor Auxter who first suggested to me that I might

explore the possibility of a general solution to the central

question of the dissertation.

Very few people who write dissertations can claim to

have for the chair of their committee someone of Professor

R. M. Hare's professional reputation and scholarly

accomplishment. At every stage of my writing, he has taken

a very active interest in my work and, because of this, I

have been saved from many of my own mistakes, both large and

small. Of course, for those errors which remain, I alone am

responsible. Yet, even so, I have no doubt at all that the

final draft of my dissertation is a far better piece of

writing than would have resulted without Professor Hare's


Finally, I would like to express my appreciation to

Mrs. Barbara Smerage, who typed the manuscript.



ACKNOWLEDGMENTS ......................................... iii

ABSTRACT ................................................ viii


1 INTRODUCTION ..................................... 1

1.1 Central Question ............................ 1
1.2 Relevant Literature ......................... 2
1.3 Purpose and Scope of Inquiry ................ 5
1.4 Organization of Topics ...................... 6
1.5 Conceptual Bases for the Study .............. 7
1.5.1 Types of Consent ..................... 7
1.5.2 The "State of Nature" ................ 13
1.5.3 "Natural Laws" as a Source of
Constraint.............................. 14
1.5.4 Punishment and Capital Punishment..... 15
1.6 Importance of the Study ..................... 17

2 HOBBES AND THE JUS NATURALIS ..................... 20

2.1 A Query for Hobbes' System .................. 20
2.2 The Hobbesian State of Nature ............... 22
2.3 Empowerment of the Sovereign and Capital
Punishment .................................. 23
2.4 Capital Punishment and the Cotenability
Issue ....................................... 25
2.5 Attempts to Augment or Amend Hobbes'
Analysis .................................... 28
2.5.1 Appeal to Hohfeld's Distinction ...... 28
2.5.2 The Lifeboat Lottery Analogy ......... 37
2.5.3 Gauthier's Amendment ................. 51
2.6 A Rejection of Hobbes' Analysis............... 67
2.6.1 Inference from Contractors'
Psychological Incapacity ............ 69
2.6.2 A Reductio Ad Absurdum on Hobbes'
Analysis ............................. 75


3.1 The Lockean Right of Self-preservation ...... 81
3.2 The Skeptical Challenge to "Natural Laws".... 85
3.3 Difficulties with Locke's Reliance on
Biblical Texts ............................. 88
3.4 A Problem with Punishment of the Innocent.... 90
3.5 Summary and Perspective...................... 97

REQUIREMENT ........................................ 99

4.1 Stages in the "State of Nature".............. 100
4.2 Rousseau's Submission Requirement and the
Death Penalty ............................... 102
4.3 A Two-step Argument for Capital
Punishment ................................. 104
4.3.1 A Rejection of the First Step......... 105
4.3.2 A Counter to the Second Step........... 112
4.4 A Hint of Diffidence in Rousseau's Account... 116


5.1 The "Juridical State of Society" and
Punishment ................................. 117
5.2 Kant's Arguments for the Death Penalty........ 120
5.2.1 The "Principle of Equality"........... 121
5.2.2 The "Inner Viciousness" Argument....... 127
5.2.3 The Suppression of Quibbles
Consideration........................... 131
5.2.4 The Counterargument to Beccaria....... 132
5.3 Systemic Difficulties with the Death
Penalty .................................... 135
5.3.1 Kant's Use of Contract Language....... 136
5.3.2 The Vacuity of the Expanded Criterion. 150
5.3.3 Kant's "Natural Law" Foundation and
Punishment of the Innocent............. 153


6.1 Relevant Features of Rawls' Theory........... 160
6.2 Cooper and King-Farlow's "Dying Innocents"
Rationale................................... 162
6.3 Donnelly's Imperilment of Liberty Argument... 166


7.1 The "Initial Bargaining Position"............ 181
7.2 Gauthier's Rationale for Punishment........... 188
7.3 A Rejection of Capital Punishment within
Gauthier's Framework ......................... 192

8 CONCLUSION........................................

8.1 Fundamental Presuppositions and the Generic
Framework.. ............................. ... 204
8.1.1 Status as Living Beings .............. 204
8.1.2 A Change "For the Better"............. 208
8.1.3 Self-Advocacy ........................ 211
8.1.4 Unanimous Consent .................... 212
8.1.5 Termination of the Agreement.......... 212
8.1.6 The Generic Framework................. 213
8.2 A General Conclusion on the Question of
Tenability .................................. 216
8.2.1 Punishment and the Ends of Constraint. 218
8.2.2 The Uniqueness of the Death Penalty... 219
8.2.3 The Problem of the Shotcaller.......... 220
8.2.4 The Idealization of the Hypothetical
Bargaining Situation .................. 223
8.2.5 Conclusions Regarding the Central
Question ............................. 229
8.3 A Practical Observation and Challenge........ 230

REFERENCES .............................................. 232

BIOGRAPHICAL SKETCH ..................................... 237



Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Vernon Thomas Sarver, Jr.

April 1994

Chair: Richard M. Hare
Major Department: Philosophy

This study explores a question of the tenability of

capital punishment within the framework of social contract

theory. In this connection, the theories of Hobbes, Locke,

Rousseau, Kant, Rawls, and Gauthier are examined and all of

them are found to be fundamentally incompatible with an

affirmative stance on capital punishment. During the course

of the study, specific arguments for the death penalty

receive attention, including those advanced by Hobbes,

Locke, Rousseau, and Kant, as well as those advanced on

behalf of Rawls by others, and on behalf of Hobbes by

Gauthier. All of these are rejected. Finally, an attempt

is made to provide a general solution to the central

question of the study, and here, the conclusion is reached

that any social contract theory based on hypothetical


consent would be fundamentally incompatible with use of the

death penalty.


1.1 Central Question

Is the execution of criminals a legitimate practice

under a government whose raison d'etre springs from social

contract assumptions, in particular, from the premise that

"Governments are instituted among Men, deriving their just

powers from the consent of the governed"?l In other words,

are there any compelling reasons to think that individuals

who have given consent (in some form) to their being bound

together in a society, and derivatively, under a government

emanating from that society,2 have, by so doing, committed

IThomas Jefferson, "The Declaration of Independence,"
in Stuart Gerry Brown, ed., We Hold These Truths, Harper &
Brothers, 1941, p. 37. It is not my intent here to narrow
the topic of the study to an inquiry about American politi-
cal institutions. Indeed, I cite Jefferson only because he
expresses in concise and familiar language what I take to be
the unifying theme of the various theories I will examine
(see section 1.5.1, infra). The influence of contractarian
assumptions on American political institutions, of course,
has been quite well documented; on this, see, e.g., Andrew
C. McLaughlin, The Foundations of American Constitutional-
ism, New York University Press, 1932, Ch. III, "The Social
Compact and Its Significance in America"; J. W. Gough, The
Social Contract, Greenwood Press, 1978, ch. XIV, "The Con-
tract in American Thought"; and Thad W. Tate, "The Social
Contract in America, 1774-1787," William and Mary Ouarterlv
22 (1965), pp. 375-391.

2At least implicit (and at times explicit) in the
contractarian literature is the presence of a distinction
between the contract for government and the contract for


themselves to the possibility (at least) that some day any

one of them may be (justly) put to death by the very society

for which they have given their consent?

1.2 Relevant Literature

In reply to this sort of query, the most prominent and

influential philosophers among the contractarian theorists

of the seventeenth and eighteenth centuries, Hobbes, Locke,

Rousseau, and Kant, have advanced, each of them without

exception, arguments in support of the death penalty, and

moreover, have done so from positions articulated within the

framework of their respective social contract theories. I

refer, of course, to specific texts which appear in Hobbes'

Leviathan,3 Locke's Second Treatise on Civil Government,4

Rousseau's The Social Contract,5 and Kant's The Metaphysical

Elements of Justice.6,7

society. As Vicente Medina explains, the latter concept
plays a more fundamental role in the theories themselves
(Social Contract Theories, Rowman & Littlefield, 1990, pp.

3Collier Books, 1962, pp. 105, 110.

4In Social Contract, Sir Ernest Barker, ed., Oxford
University Press, 1962, pp. 8-9.

5In The Essential Rousseau, Lowell Bair, trans., New
American Library, 1974, ch. V, "The Right of Life and

6Bobbs-Merrill Company, 1965, pp. 99-107.

7Any comprehensive survey of social contract theory
would have to begin with an account of its historical roots
in ancient Greece, and would include some discussion of
Socrates' argument at Crito 51 for an implied agreement with
the people of Athens as the ground for his not evading his

After Kant, with the rise of utilitarianism and German

idealism in the nineteenth century, little incentive could

be found to argue the merits of capital punishment under a

theory which had entered a period of protracted, though not

interminable, decline.8 It would not be until the latter

half of the twentieth century that a substantial rekindling

of interest in social contract theory would take place, an

interest keyed (by and large) on the publication of two

books, John Rawls' A Theory of Justice in 19719 and David

own death sentence; Plato's presentation of what Ron
Replogle calls the "proto-contractarian speculations" of
Glaucon at Republic 357A-361B (Recoverina the Social Con-
tract, Rowman & Littlefield, 1989, p. 2); and Aristotle's
allusion to, and comment on, the view of the sophist,
Lycophron, at Politics 1280B (A useful discussion of these
classical references can be found in David G. Ritchie, "Con-
tributions to the History of the Social Contract Theory," in
Darwin and Heael, Swan Sonnenschein & Company, 1893, pp.
196-199; see also Gough, op. cit., ch. II). None of these
classical sources, however, yield enough of a theoretical
perspective to warrant inclusion in the present study. Much
the same could be said with regard to later classical and
medieval sources (on these, see Ritche, op. cit., pp. 200-
205; and Gough, op. cit., chs. II, III, & IV). It is, of
course, a much different matter with regard to the seven-
teenth and eighteenth centuries, which, as Gough observes,
achieved distinction as "the heyday of the social contract"
(ibid., p. 1). These centuries were to produce Hobbes,
Locke, Rousseau, and Kant, whom Patrick Riley describes as
"the four most eminent members of the social contract
tradition" during this era (Will and Political Legitimacy,
Harvard University Press, 1982, p. vii). In a very similar
vein, Ron Replogle reports, "contractarian theory came into
its own only in the seventeenth and eighteenth centuries,
when, in the hands of Hobbes, Locke, Rousseau, and Kant," he
urges, "it served as the vehicle for the era's most impor-
tant political philosophy" (op. cit., p. 2).

8See Gough, on. cit., pp. 183-185 and ch. XII; and
Michael Lessnoff, Social Contract, Humanities Press Inter-
national, 1986, pp. 97ff.

9Harvard University Press.

Gauthier's Morals By Agreement fifteen years later.10 While

both of these books are silent on the subject of the death

penalty, other recent developments have forged links between

the theories they express and contemporary defenses of

capital punishment.

One of these developments consists of an attempt by an

American law professor, Samuel J. M. Donnelly, to develop

criteria for use of the death penalty by appeal (largely) to

Rawls' contractarian views.11 More recently, two philoso-

phers, W. E. Cooper and John King-Farlow, have published an

article entitled "A Case for Capital Punishment," which

likewise makes seminal use of Rawls' theory.12

In still other developments, Gauthier himself has

forged links to his own theory in Morals By Agreement with

the publication of two articles on Hobbes, "Taming Levia-

than" in Philosophy and Public Affairsl3 and "Hobbes' Social

Contract" in Nous.14 These papers reflect the general

emphasis of an earlier book, The Logic of Leviathan,15 in

1oClarendon Press, 1986.

11"A Theory of Justice, Judicial Methodology, and the
Constitutionality of Capital Punishment: Rawls, Dworkin, and
a Theory of Criminal Responsibility," Syracuse Law Review 29
(1978), pp. 1109-1174.

12journal of Social Philosophy 20 (Winter 1989), pp. 64-

1316 (1987), pp. 280-298.

1422 (1988), pp. 71-82.

5IClarendon Press, 1969.

which Gauthier offers an improvement on Hobbes' argument for

the death penalty. His affirmative stance in this context

lends credence to the idea that use of the death penalty

would at least not be incompatible with the theory of Morals

By Agreement, a theory which Gauthier himself depicts as a

"modern-day successor" to that of Hobbes.16

1.3 Purpose and Scope of Inquiry

In this study I will undertake a critical assessment of

the theories of the major contributors to the contractarian

tradition, beginning with Hobbes and continuing through

Gauthier, for the very limited purpose of ascertaining

whether or not any of these theories are compatible with an

affirmative stance on the death penalty. Centrally at issue

in my assessment will be the question of whether or not con-

tractors under any of these theories may be (plausibly)

assumed to have given consent (in some form) to the possi-

bility of use of the death penalty by a government emanating

from a social contract.17 It will not be my purpose in this

study, however, to pursue any question of the morality of

capital punishment. Instead, the scope of my inquiry will

extend only to exploring a question of the compatibility of

the death penalty with a particular way of doing political

philosophy. Still another limitation of my approach will be

that no assessment will be done regarding the compatibility

16Op. cit., p. 17.

17See infra, 1.5.1, for a discussion of the various
types of consent.

of the death penalty with other ways of doing political

philosophy. A relevant consequence of this limitation is

that, whatever the practical import may be of any conclusion

I may reach, it would apply only to governments of

contractarian extraction.

1.4 Organization of Topics

With regard to the theories of the seventeenth and

eighteenth centuries, my inquiry will be focused on the

specific arguments for the death penalty offered by Hobbes,

Locke, Rousseau, and Kant (chs. 2, 3, 4, & 5, respectively).

As for the contemporary theories of Rawls and Gauthier,

which are silent on capital punishment, I will, in the case

of Rawls, evaluate the attempts of Cooper and King-Farlow

and of Donnelly to construct arguments for the death penalty

based on A Theory of Justice (ch. 6); and, in the case of

Gauthier, assess the adequacy of the theory of Morals by

Agreement to support an affirmative stance on capital punish-

ment, in particular, along the Hobbesian lines Gauthier

defends elsewhere (ch. 7).

My study will conclude with an attempt to generalize

upon the results of my earlier chapters (ch. 8). In this

connection I will make an attempt to characterize the

(generic) conceptual framework for any social contract

theory by attending to the interplay of presuppositions

common to them all. After this has been done, I will then

seek to evaluate the fitness of this framework to support an

affirmative stance on the death penalty. In this manner I

would hope to arrive at a general conclusion regarding the

compatibility of capital punishment and social contract

theory. Another concern of mine in this closing chapter

will be to mention ways in which the conclusions I will have

reached in the study may be applied to the practical reality

of a world where governments of the contractarian origin do

at times impose the death penalty on some of their citizens.

Finally, in the remaining sections of the present

chapter, I will elaborate on those concepts which are both

of focal interest and in need of clarification as a pre-

liminary to the proper conduct of the study (1.5); and I

will present those reasons I have to believe that the

central question of the study engages a topic of some

considerable theoretical and practical importance, and one

that requires a treatment which is thorough and rigorous


1.5 Conceptual Bases for the Study

1.5.1 Types of Consent

Consent (in some form or other) is, of course, the

leitmotif of social contract theory. Indeed, the very idea

of deriving the "just powers" of government from "the

consent of the governed" is what imparts to theories of this

kind whatever prima facie appeal they have. Yet, having

called attention to this common feature, it is also

important to emphasize that these same theories vary among

themselves regarding the particular type (or types) of

consent on which they rely. On this score all of the

theories I will examine may be found to exhibit a reliance

on one or more of three different types, viz., actual

consent and two nonactual varieties, hypothetical and ideal


The most sustained discussion of actual consent among

these various theories appears in Locke's Second Treatise.18

Here, Locke distinguishes between two forms of consent, both

of which, I believe, he regards as actual, and to which he

applies the labels "express" and "tacit," respectively. By

"express" consent, I understand him to mean a voluntary form

of consent for which the mechanism of conveyance is some

observable sign, as, for example, when a person says, "I

consent thus to such and such," or less formally, signals

the same by a raising of the hand or a nodding of the head.

Some reliance on this form may also be found in Hobbes

Leviathan19 and perhaps even in Rousseau's The Social

Contract.20 None of these theorists, however, appears to

rely exclusively (or even predominantly) on this form.

Moreover, this may be just as well in light of the conclu-

sion reached by A. John Simmons (and shared by many others),

that "'express' consent is not a suitably general ground for

180. cit., pp. 69ff (ch. VII, sect. 117ff).

190O. cit., p. 106 (ch. 14).

200p. cit., see, e.g., bk. III, ch. XII.

political obligation."21 But I will have more to say about

this in later chapters.22

"Tacit" consent is the other form of consent recognized

by Locke and it, too, appears to be actual. Here, I under-

stand him to mean a form which is no less voluntary, but

unlike the "express" variety, occurs without the accompani-

ment of any observable act of commitment by the consenting

person. To illustrate this form, he cites the case of a

person's "possession or enjoyment of any part of the domin-

ions of any government," which, for Locke, includes every-

thing from the "possession . of land" to one's "lodging

only for a week."23 A similar point is made by Rousseau when

he affirms that "residence indicates consent."24 Note,

however, that it is not the act of "possession or enjoyment"

or "residence" which eludes observation by others; rather,

it is the act of commitment these observable acts are sup-

posed to convey which lacks observable "expression." Yet,

notwithstanding this, in any particular instance it may well

be that the actor in question has previously agreed (i.e.,

prior to any act of "possession" or whatever) that were an

act of this sort to occur, it could be regarded as a token

21Moral Principles and Political Obligations, Princeton
University Press, 1979, p. 79.

22Infra, 2.5.2 (pp. 45ff); 4.3.2 (pp. 114f).

230p. cit., p. 71 (ch. VIII, sect. 119).

243p. cit., p. 88.

of his or her consent to be bound under the terms of some

social contract. In such an instance, of course, it would

be proper to regard the act of "possession" (or whatever) as

a token of the actor's actual consent. But what I find so

problematical about the notion of "tacit" consent is that it

may not always be easy to tell in a particular instance

whether or not some observable act is in fact a mechanism

for conveyance of that person's prior actual consent to a

social contract. The difficulties I see here will be

brought to light by the following anecdote. Suppose I am

walking along a path with my lunch in hand and I happen upon

a place where you are standing. You stop me and draw a line

in the sand, and then you say to me, "By crossing this line

you will have given 'tacit' consent for me to take possess-

ion of your lunch." I say nothing to you and then proceed

to cross your line. At this point you demand that I sur-

render my lunch. Yet, what claim could you possibly have on

my lunch under these circumstances? It is not so easy to

tell without additional information. Of course, it could be

the case that we have a prior actual agreement that, under

these circumstances, I will surrender my lunch. To be sure,

in a case of this sort my "tacit" consent would be actual.

But now, suppose in the story I have related that there is

no prior actual agreement between the two of us. In the

absence of any such agreement, how am I to know that I am

obligated to surrender my lunch? Evidently here, the only

answer still based on my consent would require supposing

that I am bound under the terms of some nonactual (i.e.,

hypothetical or ideal) agreement with you. Evidently, too,

when Locke and Rousseau speak of "possession and enjoyment"

and "residence," respectively, as a basis for consent to the

social contract, they refer to some nonactual form of agree-

ment. That is to say, it can hardly be assumed that these

theorists intend to represent by the notion of "tacit" con-

sent that every contractor has participated in some prior

actual (i.e., "express") agreement regarding what will count

as the mechanism for conveyance in situations where no (on

site) "expressions" of consent will be given. Instead, it

must be supposed that they ultimately rely on some nonactual

form of agreement in their respective discussion of "tacit"

consent. Yet, even this "clarification" leaves unsettled a

question of which variety should be imputed to them, i.e.,

whether hypothetical or ideal. In light of these observa-

tions, therefore, it should be easy to see that I would be

ill advised in my inquiry to make any use of a term so

ambiguous as this one. Much the same could be said regard-

ing the notion of "implied" consent, which, too, has been

employed with reliably ambiguous consequences in discussions

of social contract theory. Anything of importance I will

have to say, therefore, in my study about consent, will have

to be said relying on less problematic terms than these.

Nonactual forms of consent, as I have reported, may be

classified as one of two varieties, i.e., hypothetical or

ideal. By the former variety, I refer to the sort of con-

sent people would have given under specified conditions.

Exactly what these conditions are purported to be varies

from theory to theory. By the latter, I refer to the sort

of consent people ought to have given, i.e., where 'ought'

is understood in a moral sense. At some time or other, all

of the theories I will examine have been linked to the

notion of hypothetical consent. Ron Replogle, for example,

is moved to offer a general assessment: "The contractarian

regards institutions just insofar as they could be the

object of a hypothetical agreement."25 In a more specific

comment, Vicente Medina identifies Hobbes, Rousseau, and

Kant as theorists who utilize "the concept of contract in a

purely hypothetical way."26 Similarly, A. John Simmons

remarks even of Locke that he relies on a contract which is

"not necessarily actual (i.e., possibly hypothetical)."27

Finally, of course, the contemporary theories of Rawls and

Gauthier would appear to be expressly hypothetical in their

conceptions of the social contract.28 Ideal consent, of

250p. cit., p. 2.

260p. cit., p. 137.

270p. cit., p. 85.

28A Theory of Justice., op. cit., p. 11; Morals By
Agreement, op. cit., p. 9.

course, is most often linked to the theory of Kant,29 though,

on occasion, it does receive mention in connection with the

theories of Rousseau30 and even Rawls.31 It is perhaps worth

noting, too, that ideal consent can be expressed in a hypo-

thetical mode, as indeed it would be if it were character-

ized as the sort of consent people would give if they were

to do what is morally required of them. Later, I will have

more to say about this in connection with Kant's theory.32

1.5.2 The "State of Nature"

The early social contract theorists rely on the notion

of a "state of nature" as a mechanism for describing the

social conditions which obtained prior to the advent of any

contract. For Hobbes, these conditions are described in

almost horrific terms;33 with Locke, there is some

improvement over the conditions which Hobbes describes, but

not enough to warrant a continuation of the status quo;34 and

for Rousseau, while the conditions are idyllic at first,

they deteriorate substantially over time and eventually

29See, e.g., Michael Lessnoff, Social Contract, Humani-
ties Press International, Inc., 1986, pp. 90ff.

30Ibid., p. 74.

31Ibid., p. 134.

321nfra., 5.3.2.

330D. cit., ch. 13.

340D. cit., ch. II.

become intolerable.35 These quasi-historical accounts, of

course, are presented by these three theorists as an

argument for the rational basis of abandoning the "state of

nature" in favor of a civil society. The later theorists,

Kant, Rawls, and Gauthier, all exhibit a shift in emphasis

away from accounts of this sort and toward a purely

conceptual discussion of the rational constraints on what

may be negotiated in a hypothetical or ideal pre-political

context. Only Kant, among these later theorists, continues

to speak of a "state of nature"; for Rawls and Gauthier, the

terms of choice are "the original position" and "the initial

bargaining position," respectively.

1.5.3 "Natural Laws" as a Source of Constraint

All six of the theories I will examine rely (at least

partially) on assumptions which are not the stuff of any

agreement among contractors. Among the early theorists,

these assumptions are often called "natural laws," and they

tend to operate as a source of constraint on what is

permitted under the terms of a social contract. Hobbes, for

example, derives a number of "natural laws" from his more

fundamental notion of a "right of nature" (the jus

naturalis). These "laws," in turn, provide the foundation

upon which his social contract is erected.36 Locke's system,

too, is constrained by the effects of certain "natural

35Infra, 4.1.

3600. cit., chs. 14 & 15.

laws." Unlike Hobbes, though, who relies principally on a

rational analysis of the implications of the Jus naturalis,

Locke turns primarily to scripture for inspiration. Yet

another approach is taken by Rousseau, who, after

recognizing a number of "natural laws," seeks to obviate

their effects in civil society by appeal to his conception

of a "general will." Among the later theorists, discussions

of "natural law" either take the form of (as with Kant), or

have given way to (as with Rawls and Gauthier), an analysis

of what rationality requires of a person in a pre-political

context and, of course, later in civil society. Indeed, for

the contemporary theories of Rawls and Gauthier, rational

choice considerations emerge as a dominant theme. With

regard to all of these theories, I would say, the potential

(at least) for a certain antagonism arises between what may

be required of social contractors as a consequence of what

they have agreed to and what may be required of them because

of the operation of certain "natural laws" or other

intuition-based assumptions. Several of the relevant

consequences of this sort of antagonism will receive

attention in my chapter on Kant (5.3.3).

1.5.4 Punishment and Capital Punishment

Finally, I should say a few words about the standard

way of justifying use of punishment (generally) within the

framework of social contract theory; and beyond this, about

what may be additionally required of the theorist to justify

use of the death penalty as a particular form of punishment.

On this way of doing political philosophy, of course, it is

assumed that the contracting parties would recognize the

necessity for some form, or forms, of constraint on their

own behaviors in "civil society." Indeed, under this

assumption, they are all persuaded that it is their

willingness to constrain some of their own behaviors which

in fact promises a "better life" for each of them than they

presently have in the "state of nature." They are moved to

draw this conclusion because they have been able to connect

the lack of constraint on their behaviors in the "state of

nature" with their present difficulties in pursuing the

various sorts of life each of them would prefer to have.

Now, punishment must be included among the forms of

constraint upon which nearly every political theorist of any

importance has relied; and, in this regard, no exceptions

arise for theorists of contractarian disposition. Indeed,

for many theorists, including those in the social contract

tradition (and, among these, the six I will examine), the

use of punishment may be described as the "constraint of

choice."37 Of course, there is a burden on every theorist

(so disposed) to explain why punishment is the "constraint

37This suggestion is perhaps open to debate with regard
to Rawls and Gauthier, though I think my analyses in chs. 6
and 7 will provide some support for its applicability to
them, as well.

of choice," that is to say, why punishment is to be

preferred over other forms of constraint.38

Within the context I have described, then, any use of

capital punishment would emerge as the most extreme form of

constraint which could be imposed. This is because death

results in the cessation of all of a person's behaviors, not

merely in those which may have been in need of constraint.

Hence, it would appear that an even greater burden falls on

the social contract theorist to account for why this punish-

ment would be approved by contractors seeking a "better

life" for themselves.

1.6 Importance of the Study

One value of the study I have proposed will be the

potential it may have as an alternative approach to the

contemporary discussion of capital punishment. Presently,

of course, the literature is saturated with articles aiming

to assess the moral status of the practice either inde-

pendently altogether of any political theory or as a prelude

to the introduction of some political theory. To be sure,

this might be a fruitful way of proceeding were it not for

the wide divergence of opinion on what constitutes a proper

assessment of moral status, and beyond that, on what con-

stitutes a proper understanding of the function of moral

language. My approach is less ambitious, though, I think,

perhaps more promising. What I propose to do is to take

38E.g., providing rewards as incentives for self-

each of several versions of social contract theory and then

to inquire of each version whether or not its presupposi-

tions would be compatible with an affirmative stance on

capital punishment. This approach, of course, entails

neither my having to defend a set of normative assumptions

nor my having to take a meta-ethical position on the

function of moral language. Another edge I have over moral

status assessments is that by narrowing the focus of my

inquiry, not just to political theory, but indeed, to a

specific variety of it, I have placed the theoretical arena

in much closer proximity to opportunities for practical

application. But now, this is only another way of saying

that there are governments of contractarian extraction in

the world today which do in fact employ the death penalty.

Finally, I urge, without a major study along the lines I

have proposed, there may be the temptation, not merely in

academic circles, but also in the public sector (where the

consequences may be more keenly felt), to take for granted a

particular outcome on the question of the legitimacy of

capital punishment under social contract theory. In support

of this observation, I would just mention the remarks of

Frank Kelly, a New York City attorney, whose article

entitled, "The Social Contract: Murder and the Death

Penalty," appeared in a recent issue of the New York State

Bar Journal:

The death penalty is a conceptually necessary
component of the social contract. . When

imposing the death penalty upon the murderer,
society, in the form of government, is doing only
what it must do to preserve the peace and fulfill
the contract.39

The difficulty I have with Kelly's article is the super-

ficiality of the several arguments he offers in defense of

his claims,40 arguments which reflect only the most fleeting

acquaintance with the dynamics of social contract theory.

In fairness to Kelly, of course, he may be right in what he

claims; however, the conclusion he reaches can be safely

drawn, if at all, only after a very thorough and rigorous

examination of the contractarian tradition, indeed, an

examination, I would think, very much along the lines of

what I will try to provide in this study.

39(July 1989), pp. 45f.

40For example, he argues: Because a person may repel
with deadly force an attack on his or her life, society,
too, may employ deadly force in the form of capital
punishment against the attacker whenever the attack is


2.1 A Query for Hobbes' System

None of the claims in Hobbes' Leviathan are more funda-

mental to the operation of his system than (a) his appeal to

an initial act of authorization by which contractors under

his theory vest in their ruler(s) the power to command

obedience of them and even require of them the sacrifice of

any of their lives;1 and (b) his steadfast insistence upon

every contractor's inalienable right of self-preservation

(the ius naturalis), a right which authorizes any action at

all taken by any one of them in an attempt to preserve his

or her own life.2 My purpose in what follows will be to

explore a question of whether or not the claims at (a) and

(b) are cotenable in light of Hobbes' (intrasystemic)

defense of capital punishment.3 To lay a foundation for

iCollier Books, 1962 (originally published: 1651), ch.
17, p. 132; ch. 21, p. 161.

2Ibid., ch. 14, p. 103.

3Hobbes' political theory in Leviathan includes a
defense of capital punishment as an integral feature of
his system; hence, I have employed the term 'intrasystemic.'
By contrast, I will take no account of his defense in De
Cive (Sterling P. Lamprecht, ed., Appleton-Century-Crofts,
1949 (originally published: 1642), ch. 2, sect. 18), which,
for my purposes, is extrasystemic, notwithstanding that

this query of Hobbes' system, preliminary attention will be

given to certain other features of his theory which provide

an essential framework both for his claims at (a) and (b)

and for his theory of punishment upon which he relies for

his defense of the death penalty. My prefatory remarks, in

this connection, will take the form of an elaboration of

Hobbes' view of the "state of nature" (2.2) and his empower-

ment thesis regarding the institution of political authority

(2.3). In conjunction with this latter topic I will lay

stress upon the relationship between Hobbes' empowerment

thesis and his theory of punishment, and I will explain why

this relationship is thought (by Hobbes) to provide a theo-

retical basis for his defense of the death penalty. With

this foundation in place I will then take up the question of

whether or not Hobbes' defense of the death penalty can be

reconciled with (a) and (b) (2.4), and here my concern will

be, not only with Hobbes' own account, but also with the

efforts of others who have undertaken to augment or amend

his original analysis (2.5). Finally, I will assess the

status of two counterarguments to Hobbes' analysis, both of

which purport to raise serious doubts about the reconcil-

ability of (a) and (b) with any defense of the death penalty

based on his theory (2.6).

his argument there closely parallels that given in

2.2 The Hobbesian State of Nature

On Hobbes' view every man is at war with every other

man so long as he remains in a state of nature. While he

continues in this state, "every man has a right to every

thing; even to one another's body" (Hobbes' First Law of

Nature).4 However, this right "to every thing" should not

be taken in the sense of what today would be called a "claim

right," or a right which imposes on others a correlative

duty not to interfere with a person's exercise of it.

Instead, what Hobbes appears to mean here would much more

closely approximate what today would be called a "permission

right," or a right whose exercise does not impose on others

any duty of noninterference.5 But now, as Hobbes explains,

since everyone in a state of nature is endowed with a right

"to every thing," it is all too apparent that the social

costs of its exercise are too high, for people invariably

compete and fight with one another for many of the same

things. In this intolerably precarious and otherwise deplor-

able condition, he continues, people discover (under the

impress of reason) that they ought to seek the security of a

covenant with one another, and this for their mutual

4Hobbes, Leviathan, loc cit.

5After Wesley Hohfeld's Fundamental Legal Conceptions,
Greenwood Press, 1978, pp. 35ff. Hohfeld's original term,
'privilege,' has given way to either 'liberty' or
'permission,' or 'permission right' in contemporary
discussions of his distinction. The last of these has been
employed in my own analysis.

protection and benefit (the Second Law of Nature).6 This

discovery by the people, for Hobbes, marks the beginning of

the end of a state of nature.

2.3 Empowerment of the Sovereign and Capital Punishment

It is to secure the peace, therefore, and to put an end

to a state of nature where life is "solitary, poor, nasty,

brutish, and short," that all but one (or just a few) among

the people agree to relinquish their individual right "to

every thing; even to one another's body."7 The one (or

ones) whom the others allow to retain this right, on Hobbes'

account, are elevated (de facto) to a position of supreme

power and authority over all the rest. This empowerment of

a "sovereign" is intended for the purpose of establishing a

mechanism for the legislation and enforcement of laws which

can then provide for the peaceful settlement of disputes

among people, and for the restraint and punishment of acts

deemed contrary to the goal of peace. Obviously, there is a

considerable sacrifice of freedom on the part of those who

covenant together in this manner, and who, as a consequence,

pledge loyalty and obedience to their sovereign; neverthe-

less, it is Hobbes' contention that this is the only

reasonable alternative to the condition of war which exists

6Hobbes, Leviathan, OD. cit., ch. 14, p. 104.

7Ibid., ch. 14, p. 103.

perpetually, and to which one is otherwise committed, in a

state of nature.8

It is no very great problem to discover, within this

framework, the basis for Hobbes' defense of capital punish-

ment. A general authorization to employ punishment arises

simply as an extension of the sovereign's (retained) right

"to every thing," even to the body of another; and any

specific punishments, including the death penalty, all

receive their theoretical imprimatur under the umbrella of

this general authorization.9 Moreover, since the sover-

eign's (retained) right "to every thing" is authorized at

(a), it follows that (a) provides both the justificatory

basis of the right to punish and an enforcement mechanism in

the form of a de facto monopoly on the use of power.10 The

latter serves to deprive anyone slated for punishment of the

opportunity for effective resistance; and the former, of any

cause for complaint, as someone punished under color of (a)

8"And though [sovereign authority is] of so unlimited a
power, men may fancy many evil consequences, yet the
consequences of the want of it, which is perpetual war of
every man against his neighbour, are much worse" ibidd., ch.
20, p. 157).

9"[T]here is a question to be answered, of much
importance; which is, by what door the right or authority of
punishing in any case, came in"; "before the institution of
a commonwealth, every man had a right to every thing, and to
do whatsoever he thought necessary to his own preservation;
. this is the foundation of that right of punishing,
which is exercised in every commonwealth" ibidd., ch. 28, p.

10"[C]ovenants, without the sword, are but words, and of
no strength to secure a man at all" ibidd., ch. 17, p. 129).

would be an "author of his own punishment."11 The two

together are intended to provide the sovereign with all that

is necessary to secure the commonwealth against the threat

of crime and domestic violence.

2.4 Capital Punishment and the Cotenability Issue

The general authorization to employ punishment, there-

fore, which is both justified under, and facilitated by,

(a), activates a latent tension in Hobbes' system whenever

the life of a contractor is actually threatened by the

sovereign. I say this because it is also the case, as I

have described at (b), that each of Hobbes' contractors is

endowed with an inalienable right of self-preservation (the

ius naturalis), a right which authorizes its bearer to take

any action whatever in defense of his or her own life.12

Hobbes himself appears well aware of this latent tension in

the context of his specific remarks on the death penalty.

Here, he argues that when a man who stands condemned to

death by the sovereign is led to the gallows, that man never-

theless retains "the right of resisting them, that assault

him by force, to take away his life."13 It is for this

reason, he explains, that criminals are led to execution by

"armed men, notwithstanding that such criminals have

1Ibid., ch. 18, pp. 134ff; "[H]e that complaineth of
injury from his sovereign, complaineth of that whereof he
himself is author" (p. 136).

12Ibid., ch. 14, p. 103.

13Ibid., p. 105.

consented to the law, by which they are condemned."14 Simi-

larly, in another passage of Leviathan, Hobbes claims that,

while a man can covenant, "Unless I do so, or so, kill me,"

in conformity with the right of the sovereign "to every

thing," even to the body of another, he cannot covenant,

"Unless I do so, or so, I will not resist you when you come

to kill me," by virtue of the inalienability of his right of

self-preservation.15 The tension evident between (a) and (b)

in these passages has led some commentators to conclude that

Hobbes' system reaches impasse in cases involving capital

punishment.16 While these commentators are right, of course,

about the tension in Hobbes' system, and may even react with

understandable perplexity to the peculiarity of his having

contractors who are both authors of, and rightful resisters

to, any punishment the sovereign may threaten, it follows

neither from the tension nor the oddity of these cir-

cumstances that a charge of theoretical impasse can be

14Ibid., p. 110.


16See Gordon J. Schochet, "Hobbes and the Voluntary
Basis of Society," in Thomas Hobbes and Political Theory,
Mary G. Dietz, ed., University Press of Kansas, 1990, p. 62;
Leo Strauss, "On the Spirit of Hobbes's Political
Philosophy," in Hobbes Studies, K. C. Brown, ed., Harvard
University Press, 1965, pp. 25f; Gregory S. Kavka, Hobbesian
Moral and Political Theory, Princeton University Press,
1986, ch. 8; and David Gauthier, whose views I discuss later
in the chapter (2.4.3).

sustained against his system.17 To demonstrate impasse, I

submit, Hobbes' critics must show that his claims at (a) and

(b) are not cotenable in light of his defense of capital

punishment; and this demonstration, by and large, they have

neither provided nor even attempted to provide.18 In fair-

ness to the concerns of these critics, however, it must be

admitted that Hobbes himself, though alert to the tension

within his system and sensitive to the peculiarity of his

contractors' circumstances,19 fails nevertheless to supply a

rationale for these features adequate to preclude the

specter of impasse. Indeed, that task has been left for

later adherents of his system who have suggested various

ways of augmenting or amending his original analysis. In

the remainder of this chapter, therefore, I will consider,

first of all, the suggestions of these later adherents to

see whether or not they can improve on Hobbes' own analysis;

and secondly, with the results of this inquiry in hand, I

will entertain two challenges to his system which do in fact

appear to raise serious concerns about the cotenability of

(a) and (b) in light of his defense of the death penalty.

17For other reports of theoretical tension in Hobbes'
system, cf. Michael Lessnoff, Social Contract, Humanities
Press International, 1986, pp. 101f; and John Watkins,
Hobbes's System of Ideas, Hutchinson, 1973, pp. 97f.

18An exception to this would be David Heyd's recent
article, "Hobbes on Capital Punishment," History of
Philosophy Quarterly 8 (April 1991), pp. 119-134; see my
discussion of Heyd's views infra (2.6.1).

19Hobbes, Leviathan, op. cit., ch. 28, p. 229.

2.5 Attempts to Augment or Amend Hobbes' Analysis

2.5.1 Appeal to Hohfeld's Distinction

One way of augmenting Hobbes' analysis makes use of the

distinction between a "claim" right and a "permission" right

to which I have already alluded.20 It is conceded, on this

approach, that if it were necessary to regard the rights at

(a) and (b) as "claim" rights, then Hobbes' system would be

at impasse. A moment's reflection will suffice to confirm

the basis for this concession. For example, if the

sovereign's right "to every thing" at (a) were a "claim"

right, then the right to punish, which derives from that

right, would impose on others a duty of noninterference.

Since this duty would extend even to those singled out for

punishment, (a) would be at loggerheads with (b). This is

because (b) confers a right of resistance upon anyone whom

the sovereign threatens with punishment.21 Similarly, if a

contractor's right of resistance at (b) were itself a

"claim" right, then everyone else would have a duty of

noninterference and so even the sovereign would have to

stand aside, which is contradictory in light of the

20Supra, n. 5; see, e.g., the discussion of V. Bobbio's
views in Mario A. Cattaneo, "Hobbes' Theory of Punishment,"
in K. C. Brown, ed., Hobbes Studies, Harvard University
Press, 1965, pp. 292f.

21"[N]o man is supposed bound by covenant, not to resist
violence; and consequently it cannot be intended, that he
gave any right to another to lay violent hands upon his
person. In the making of a commonwealth, every man giveth
away the right of defending another, but not of defending
himself" (Hobbes, Leviathan, op. cit., ch. 28, p. 229; see
also, ch. 21, p. 164).

authorization given at (a). What emerges from this

concession, then, on this way of augmenting Hobbes'

analysis, is that the rights derived from (a) and (b) are

much more plausibly construed as "permission" rights, and so

no duty of noninterference is conferred upon anyone in

virtue of their exercise. Moreover, since no duty of

noninterference arises in conjunction with the exercise of

these rights, neither does any "theoretical impasse" occur

when the sovereign clashes with a contractor who resists

punishment; and, of course, this would apply to any

punishment at all the sovereign threatens, including the

death penalty. So, at least on this interpretation, the

specter of impasse may be retired by the simple expedient of

embracing the distinction between a "claim" right and a

"permission" right, and then recognizing that the rights at

(a) and (b) are instances of the latter rather than the


Does this way of augmenting Hobbes' analysis serve to

strengthen his position on the death penalty? Much depends,

I urge, on how the notion of a "permission" right is under-

stood in connection with the rights affirmed at (a) and (b),

and on whether this manner of augmentation correctly

identifies the source of tension in Hobbes' system. What I

propose here is to provide an account of the standard

interpretation of this notion in the philosophical litera-

ture, and then to show that the rights at (a) and (b) have

certain inherent features which resist that interpretation.

In this manner I will seek to expose the problematic charac-

ter of any reliance upon the distinction between a "claim"

right and a "permission" right as a means of augmenting

Hobbes' original analysis.

An example which invokes the standard interpretation of

a "permission" right is the case of a person who seeks

election to public office.22 Here, a person's right to get

oneself elected is construed as a "permission" right since

others have no obligation to stand aside and permit this to

happen. Indeed, others may even oppose such a person by

supporting other candidates or seeking office themselves.

Now, on this account, what is germane to my inquiry is that

a person's "permission" right to get oneself elected is

ordinarily accompanied by certain other rights which

function in a facilitating role with regard to the "permis-

sion" right and which themselves may be accorded status as

"claim" rights. Among these are included a right to collect

signatures or meet other criteria necessary to have one's

name placed on the ballot, a right of equal access to the

media, and, when the election is over, a right to a fair and

impartial tabulation of votes. Of course, what makes these

latter rights "claim" rights is that others do have a duty

of noninterference with respect to their exercise by a

candidate for public office. This observation is germane

22From R. M. Hare's Moral Thinkina, Clarendon Press,
1981, pp. 149f.

to my inquiry because it calls attention to a feature of

this example not present in Hobbes' right of self-

preservation at (b). This is just to say, at least in

capital cases, a condemned person's right of self-

preservation is unaccompanied by any (claim) rights which

facilitate its exercise. In other words, a person under

sentence of death, on Hobbes' system, has no claim against

the sovereign or anyone else that requires concessions which

facilitate exercise of the right at (b).23 What this sug-

gests is that not all "permission" rights are rights whose

exercise is necessarily protected in some respects by the

(contemporaneous) presence of certain "claim" rights, (as, I

think, would be the case with the right to get oneself

elected). Evidently, then, if the right at (b) is to count

as a "permission" right, it cannot be so in virtue of its

being a non-claim right whose exercise is supported by other

rights which are themselves "claim" rights, as this cri-

terion would exclude (b). Therefore, an appeal must be made

to some other feature (or features), necessarily present in

all "permission" rights, upon which to rely in certifying

the right at (b) as such a right.

The only candidate for this role I have found in the

philosophical literature is a feature which functions to

afford every bearer of a "permission" right (in effect) the

23Hobbes even requires that contractors under the
authorization agreement at (a) are obliged "to assist him
who hath the sovereignty, in the punishing of another"
(Leviathan, OD. cit., ch. 28, p. 229).

accompaniment of a "claim" right not to be punished merely

for exercising the "permission" right.24 Note that the

accompanying "claim" right in this instance is not one

which facilitates exercise of a "permission" right but is

instead one which gives rise to a claim against others only

after the "permission" right has been exercised. Now again,

either this feature is present in the right at (b) or it is

not. If it is present, then this would mean that a con-

demned person has a claim against the sovereign not to be

punished merely for exercising the right of self-

preservation.25 Hobbes, however, clearly rejects this

possibility when he asserts the illegitimacy of claims of

any kind brought by a contractor against the sovereign.

Mincing no words in this regard, he states: "Whatsoever

. [the sovereign] doth, it can be no injury to any of

his subjects; nor ought he to be by any of them accused of

injustice."26 This idea of the sovereign's immunity to

claims extends also to those who act on his (or their)

behalf and under his (or their) direction; of them, Hobbes

writes: "Nothing the sovereign representative can do to a

24See, e.g., Samuel Stoljar's discussion of defensive
rights in An Analysis of Rights, St. Martin's Press, 1984,
ch. 2.

25Such a claim, I suppose, would be intended to block,
e.g., the sovereign's substitution of a more painful method
of execution as a means of punishing both the capital
offense and acts of resistance by the condemned person.

26Hobbes, Leviathan, oP. cit., ch. 18, p. 136.

subject, on what pretence soever, can properly be called

injustice, or injury."27 Evidently, therefore, this feature,

too, is one that cannot be found in the right of self-

preservation at (b); and so, a fortiori, neither can its

presence be cited as a basis for certifying the right at (b)

as a "permission" right.

It would appear, then, at least in the case of a person

under sentence of death, that no "claim" rights of any kind

accompany Hobbes' right of self-preservation at (b). All

that remains, therefore, as a basis for certifying the right

at (b) as a "permission" right, is the presence of the

defining feature of such a right, viz., that its bearers

have no grounds for any claims against others that they not

interfere with its exercise. The difficulty I have, how-

ever, with a reliance on the manifestation of only this

feature is that I do not see how its presence can in fact be

distinguished in a capital case. To understand why I say

this, imagine the circumstances of two persons who live in

parallel universes, both of whom are threatened with execu-

tion by their respective sovereigns. Assume that the

relevant circumstances of these two persons are exactly

similar in every respect except that one of them bears

Hobbes' right at (b) and the other does not. Assume, too,

that the relevant similarities include that they both

undertake acts of resistance to being put to death; that

27Ibid., ch. 21, p. 161.

neither bears any "claim" rights against others that certain

conditions be allowed in order to facilitate their acts of

resistance; and that neither bears any "claim" right against

his or her sovereign that he or she not be punished for

these acts. Now, recall that what is supposed to distin-

guish the person who bears Hobbes' right from the one who

does not is that the consequence for the former is that no

duty of noninterference is imposed on others in virtue of

its exercise. However, what should be no less apparent, I

urge, is that no duty of noninterference is imposed on

others also in virtue of the latter person's not having any

such right. That is to say, no duty whatever is imposed on

others in either case, whether it be in virtue of having or

not having the right at (b). Since the presence of this

feature may be detected for both a bearer and a nonbearer of

the right at (b), it follows that its presence alone cannot

be utilized as a basis for according status as a "permis-

sion" right, as this would require the absurdity of

according such status to one who bears no right at all in

this regard.28

I would conclude, therefore, that efforts to strengthen

Hobbes' argument for the death penalty by means of appeal to

the notion of a "permission" right may well land those who

are committed to this strategy on the horns of a dilemma.

28To be sure, it is sufficient reason for why a person
has no claim against others not to interfere with that
person's exercise of the right at (b) that that person has
no such right to begin with!

That is, either they will accept the accompaniment of one or

more "claim" rights as the feature (or features) that

certify the right at (b) as a "permission" right, in which

case they must also accept the contradictory implications

that follow for other features of Hobbes' system, as I have

explained; or they must pin their hopes for qualification

entirely on the defining feature of such a right, again,

that of not having any basis for a noninterference claim

against others, in which case they will be vulnerable to the

criticism that nonbearers of the right at (b) qualify for

"permission" right status just as readily as bearers, which

is absurd, as I have also explained.

Finally, I would just note that appeal to the notion

of a "permission" right is no more compelling for the right

at (a) than it is for the one at (b). I say this because

the sovereign's right "to every thing" at (a), which is the

foundation for his (or their) right to punish, finds its

ultimate ground in the sovereign's own right of self-

preservation (the ius naturalis).29 What this means is that

both the sovereign and a person under sentence of death

exercise the very same right (the ius naturalis) when the

former attempts to kill the latter and the latter counters

29"For the subjects did not give the sovereign that
right ['to every thing' with they possessed]; but only in
laying down theirs, strengthened him to use his own, as he
should think fit, for the preservation of them all."
(Leviathan, op. cit., ch. 28, p. 229).

with resistance.30 It would appear to follow, therefore,

that the analysis I have given to show that the right at

(b) fails to qualify as a "permission" right would also

apply to the right at (a). Perhaps the only challenge to

this conclusion would be to insist upon "permission" right

status for the right at (a) in connection with the authori-

zation given by Hobbes' contractors. That is to say, a

feature of (a), in addition to and conjoined with the

sovereign's right of self-preservation (the jus naturalis),

is (in effect) the contractors' establishment of a "claim"

right, on behalf of the sovereign and against all others

(except for a person under threat of punishment), not to

interfere with his (or their) right of self-preservation

and, by extension of course, with his (or their) right to

punish. The chief difficulty I have with use of this

criterion (i.e., the accompaniment of a "claim" right) to

qualify the right at (a) as a "permission" right is not that

any conflict arises with the right at (b) which leads to

impasse,31 but rather that the right at (b) fails of

qualification altogether (having no "claim" rights to

accompany it) and, of course, this provokes the more telling

30Cattaneo writes: "At the moment when the death penalty
is inflicted, the rights of the sovereign and the subject
are placed on the same plane, and there is a return to the
state of nature, whereby at that moment the conflict between
the sovereign and the subject takes on the character of a
state of war." (op cit., p. 282).

31No impasse results here because the sovereign's
"claim" right of noninterference applies only to those who
are not under threat of punishment.

question of why one should qualify and the other fail of

qualification when in fact they are both instances of the

very same right.

Finally, notwithstanding all that I have said,

advocates of Hobbes' system may nevertheless take some

encouragement from my analysis up to this point. I say this

because I have (in effect) argued that the instances of the

jus naturalis at (a) and (b) are weaker (in a normative

sense)32 than even a "permission" right (on the standard

interpretation) and so would not seem likely candidates to

generate any sort of conflict which could lead to theo-

retical impasse. But then, I will have more to say about

this later. In the meanwhile what I have shown in this

section is that the notion of a "permission" right (on the

standard interpretation) is not a particularly useful

concept in the service of those who seek to augment Hobbes'

argument for the death penalty.

2.5.2 The Lifeboat Lottery Analoav

Another attempt to augment Hobbes' analysis consists of

an argument from analogy where Hobbesian contractors are

compared to participants in a lifeboat lottery.33 Here, no

320ne author even goes so far to conclude that Hobbes
"deprives the notion of rights of virtually all its
normative significance." (J. Finnis, Natural Law and Natural
Rights, Oxford University Press, 1980, p. 208).

33Kavka urges that "it may be wrong . for a partici-
pant in a lifeboat lottery to refuse to jump overboard once
he has lost," (op. cit., p. 316); and later, analogizing on
this scenario and others, concludes, "rational parties [in a
state of nature] aware of all the relevant facts might well

effort is made to provide a conceptual account of how the

rights at (a) and (b) are inherently consistent (as before

with appeal to the notion of a "permission" right); instead,

the intent is to show that, as a practical matter, the best

way of securing the conditions necessary for a contractor's

(successful) exercise of the right at (b) is for that

contractor to participate in the authorization agreement at

(a).34 The argument offered for this conclusion rests on an

appeal to the desperate circumstances of lifeboat occupants

who find themselves adrift at sea following a shipwreck. It

is stipulated, on this account, that these occupants all

agree that the sacrifice of one of them is essential to the

survival of the rest. Moreover, they also agree unanimously

upon a procedure for conducting a lottery to select the

hapless person. Next, with regard to these circumstances,

it is further understood that these occupants have arrived

at this agreement in light of their observation that each

person's overall prospects for survival would be worse under

any other (available) course of action. Finally, the

crucial step of the argument consists in urging that it

would be comparably rational of Hobbesian contractors to

incur the risk of death by execution in conjunction with the

. [prefer] the small risks of execution to the larger
risks of a state of nature" (p. 328).

34Cf. Rousseau's claim, "Everyone has the right to risk
his life in order to preserve it." (The Social Contract in
The Essential Rousseau, Lowell Bair, trans., New American
Library, 1974, p. 31.

authorization agreement at (a), i.e., on the assumption that

every contractor would have assessed his or her (overall)

prospects for survival to be worse under any alternative to

that agreement.35

Now, a tempting response to this argument would be to

ignore its analogical features altogether and instead chal-

lenge the assumption upon which Hobbesian contractors rely,

viz., that their prospects for survival would be worse under

any alternative to the authorization agreement at (a).

Here, the point to be made is not that they would fail to

embrace this assumption, (indeed, they could hardly be

expected to do otherwise on Hobbes' system),36 but rather

that, in doing so, they would be simply mistaken, i.e., with

regard to (at least) some sets of empirical circumstances.

This response keys on the observation that not every pre-

political, social set of circumstances in which people

either have found (or could find) themselves (what Hobbes

calls a "state of nature") would be such that every person

so situated would have worse prospects of survival than they

would have if they embraced an autocratic government

pursuant to the authorization agreement at (a).37 Moreover,

35Supra, n. 8.

36Hobbes, Leviathan, op. cit., chs. 13 & 14.

37Historical examples of societies which have elected to
remain in a "state of nature" include the Eskimos of the
North American Arctic region, the Pygmy hunters who dwell in
the rain forests of the interior of Zaire; the Ifugao who
live on the island of Luzon in the Philippine chain; and the

even in those circumstances where some form of government

would have greater survival value than a prolongation of a

"state of nature," it hardly follows that the sort of

totalitarian regime Hobbes commends in Leviathan would be in

all (or even most) of these instances that particular form

of government which affords every person so situated his or

her best prospects for survival.

While these observations may well provide an imposing

challenge for two of Hobbes' bedrock assumptions,38 still,

they neither directly address nor bear on a question of the

plausibility of the lifeboat lottery analogy. I say this

because nothing in the lifeboat scenario excludes the

possibility that occupants of this craft are themselves

mistaken in their assessment that every alternative to a

lottery would afford each of them worse prospects for

survival. Moreover, I submit, no harm would be done to the

analogy even if all the occupants of the craft were, say,

oblivious to the availability of a far better alternative in

comparison to a lottery. What matters most within the

framework of a contractarian approach (i.e., insofar as the

creation of an obligation is concerned) is not that the

empirical assumptions upon which these occupants rely may be

Land Dayaks of Sarawak in Borneo. For a detailed discussion
of these societies, as well as others which are similarly
disposed, see Alex Comfort, People Without Governments, Kahn
& Averill with Cienfuegos Press, 1982, chs. III-VI.

38For a recent critique of these assumptions, see
Vicente Medina, Social Contract Theories, Rowman &
Littlefield, 1990, ch. 2.

mistaken but rather that they have unanimously agreed upon a

particular course of action which they have all embraced

come what may. Later, when one of them is selected by

lottery for sacrifice under this course of action, no objec-

tion could then be raised either by or on behalf of the

person so designated which would have any merit whatever

within a contractarian framework, i.e., so long as all the

procedures followed adhered to the terms of their agreement.

Similarly, insofar as only the analogy itself is considered,

it makes no difference that Hobbesian contractors may be

mistaken in what they have assumed about the "state of

nature" and the necessity for a totalitarian regime. Here

again, what matters within a contractarian framework (rela-

tive to the analogy) is that they have all embraced the

authorization agreement at (a) come what may. Later, should

one of them be condemned to death by the sovereign, no

objection could then be raised on (strictly) contractarian

grounds, i.e., again with the proviso that the agreement

itself had been faithfully implemented. No discredit,

therefore, redounds to the argument from analogy because of

its failure to exclude the possibility of (empirical) error

by those involved (whether they be lifeboat occupants,

Hobbesian contractors, or both).

Other considerations, then, will provide the basis for

my criticism of this argument. Of these, none will figure

more prominently in my analysis than a matter which has so

far been taken for granted in my discussion, viz., that the

lifeboat scenario relies on the circumstances of an actual

agreement among the craft's occupants. Implicit in the

scenario itself, the presence of this feature is otherwise

suggested by the observation that no one would seriously

propose that these occupants could be bound by the terms of

some hypothetical agreement which, say, harkened back to the

occasion of their having boarded an ocean liner at a time

well before any mishap. Indeed, at that time, I submit, the

most rational action which could be imputed to them would be

for them to defer any decision about participation in a

lottery until after a lifeboat scenario arises. Only then,

I urge, would it make sense for any of these occupants to

reach a decision about whether or not it would be in his or

her overall best interest to convene a lottery.39 When this

latter feature is considered in conjunction with the other,

more explicit features which I introduced earlier, it

becomes evident that the lifeboat scenario is one which

relies for its plausibility on the circumstances of an

actual agreement, unanimously embraced and developed in

response to the exigencies of a particular empirical


39Moreover, even after they find themselves aboard a
lifeboat, I submit, even people of equivalent rationality
may arrive at different conclusions about a lottery based on
attention to relevant, personal dissimilarities. For
example, a young and healthy person who finds himself among
others who are elderly and frail may be strongly inclined to
reject the risk of a lottery while the others may be
strongly disposed to accept it.

Now, for the lifeboat scenario to have analogical value

relative to the circumstances of Hobbesian contractors, I

suggest, it must also be the case that these contractors,

too, have all given their actual consent to the authori-

zation agreement at (a), and moreover, have done so in

response to the exigencies of a particular "state of

nature." It will not do to argue from the rationality of

the lifeboat occupants' embrace of an actual agreement to

conduct a lottery (if that can be supposed) to the ration-

ality of merely assuming that Hobbesian contractors would be

bound by the terms of an agreement to which they have never

actually committed themselves. While the plight of these

occupants and contractors is similar (ex hypothesi, at

least) in that they both face worse conditions in the

absence of their respective agreements, still, this

similarity alone cannot explain why a hypothetical agreement

which is merely supposed to apply could bind a group of

people who have in actuality made no such commitment.

So, for the time being at least, I am prepared to

concede a measure of plausibility to the argument from

analogy, i.e., as long as the parallel between lifeboat

occupants and Hobbesian contractors may be regarded as one

where all the members of both groups embrace an actual

agreement keyed on responses to their respective sets of

(empirical) circumstances. But then, this is a concession I

am prepared to make in any case, as, even without the

analogy, I see no basis for any (strictly) contractarian

objection to a group of people getting together and all

deciding to embrace the authorization agreement at (a), that

is to say, under conditions whereby everyone knows about and

accepts the included risk of death by execution.40

The argument from analogy, therefore, shows promise

only if Hobbesian contractors may be assumed to have

embraced an actual agreement at (a). In view of this, it

becomes critically important to examine the relevant texts

in Leviathan and to ascertain from these whether or not any

support may be found for this assumption. Undoubtedly,

Hobbes' most illuminating remarks, in this connection,

appear in his chapter entitled, "Of the First and Second

Natural Laws, and of Contracts," where he distinguishes two

types of contract: one, "express," and the other, "by

inference."41 Clearly, for Hobbes, the first of these is

actual, as the contracting parties rely on "spoken words"

for an exchange of "promises." Though he does not discuss

written contracts here, I believe they, too, would merit

classification as "express." It is the second type, of

course, which is properly called "hypothetical." What makes

use of this label appropriate is evident from Hobbes'

description of the contract "by inference" as one which

40Theoretical consistency requires this concession; for
an explanation of the dearth of historical examples, see my
mention of Filmer's criticism, infra, p. 45f.

41Hobbes, Leviathan, OD. cit., ch. 14, p. 106.

"sufficiently argues the will of the contractor," i.e., in

the absence of any "express" sign.42 On the basis of this

text, then, it would seem to follow that the authorization

agreement at (a) may be construed either as an actual

exchange of "spoken words" (or written assurances) or

(hypothetically) as that which merely "argues the will" of

those who have given no "express" (or actual) sign of their


If the agreement at (a) is construed as an actual

contract, then the argument from analogy is better served

but Hobbes' system as a whole becomes the target for a

barrage of criticisms which have been around for two or

three centuries. In this regard David Hume's eighteenth

century essay, "Of the Original Contract," is worthy of

special mention because of its influence and cogency.43

Essentially, what Hume contends is that no existing govern-

ment can be traced to an actual contract, and moreover, even

if one could be, there would still remain the difficulty of

accounting for the basis on which subsequent generations and

immigrants would be bound.44 In another vein, a seventeenth

century critic, Sir Robert Filmer, argues that it is simply

unrealistic to suppose that the occupants of some territory


43In Social Contract, Sir Ernest Barker, ed., Oxford
University Press, 1962, pp. 147-166.

44Ibid., pp. 149ff.

would in fact all get together and agree to a social con-

tract.45 Because of criticisms such as these, those who

would defend Hobbes' system today are much more likely to

construe the authorization agreement at (a) as a contract of

the hypothetical, rather than actual, variety.46 Further-

more, entirely apart from these criticisms, the text of

Leviathan appears to support what is essentially a hypotheti-

cal construal, anyway. An important passage, in this

regard, appears in chapter seventeen, where Hobbes identi-

fies two, equally legitimate ways in which a commonwealth

can emerge. One of these, he calls a "commonwealth by

institution"; the other, a "commonwealth by acquisition."47

Under the former way, people "choose their own sovereign,"

albeit "for fear of one another" in a "state of nature."48

Here, I suppose, Hobbes would grant that the contract may be

either actual or hypothetical. By contrast, under the

latter way, "sovereign power is acquired by force," whereby

authorization occurs, it would seem, not in virtue of any

actual consent of the people qua contractors, but instead,

de facto, as a consequence of their acquiescence to

sovereign rule from "fear of death, or bonds" qua citizens

45See Lessnoff, op. cit., p. 85.

46See, e.g., Medina, op, cit., p. 137; and Kavka, Op.
cit., p. 182.

47Hobbes, Leviathan, oP cit., ch. 17, p. 133.

48Ibid. ch. 20, p. 151.

who have been subdued or conquered.49 Notwithstanding this,

I suppose, their collective acts of acquiescence could be

construed as a token of an actual (but implicit) agreement,

on their part, to "embrace" their circumstances of subjuga-

tion. Yet, this seems an improbable interpretation: what

is, after all, the point of agreeing to "embrace" circum-

stances which have already been imposed? It would appear,

therefore, that the authorization agreement at (a), i.e.,

with regard to a commonwealth by acquisition, could be much

more plausibly construed as having a hypothetical character

or, in Hobbes' own words, as that which "sufficiently argues

the will" of those who have been subdued or conquered.50 Now

since, as one author wryly observes, "a sovereign by institu-

tion, whatever his rights may be by the act of institution,

will always be able to convert himself into a sovereign by

acquisition, with unlimited right,"5s and further, since

Hobbes himself acknowledges in chapter twenty, that "the

rights and consequences of sovereignty, are the same in both


5sIbid., ch. 14, p. 106. That is, hypothetically, if
Hobbes' citizens were confronted with a choice of whether to
continue in a "state of nature" or to accept sovereign rule
under conditions where they had been subdued or conquered,
they would all agree to the latter; however, in actuality
they have no choice in the matter since sovereign rule over
them will be established in any case.

51David Gauthier, The Logic of Leviathan, Clarendon
Press, 1969, p. 137.

[types of commonwealth],"52 it would appear to follow that

Hobbes' theory ultimately rests on a characterization of the

authorization at (a) as the product of a hypothetical,

rather than actual, agreement. Of course, if I am right

about this, then the argument from analogy misses the mark,

because, as I have explained already, it will not do at all

to argue from the rationality of an actual agreement (aboard

a lifeboat) to a rationality which merely assumes of

Hobbesian contractors a willingness to embrace an agreement

at (a), and further, does so in the absence of any actual

commitment from them. But even if I am wrong about this,

and the most fundamental interpretation of Hobbes' theory

requires a characterization of the agreement at (a) as an

actual contract, then, as I have already explained, while

the analogy itself fares better, Hobbes' system as a whole

would be vulnerable to the criticism of Hume, Filmer, and

many others.53

Yet another area of difficulty for the argument from

analogy may be suggested. Here, I appeal to the observation

that Hobbesian contractors do not know nearly so much as

lifeboat occupants do about the risks assumed subsequent to

their respective agreements. Indeed, under a hypothetical

construal of the agreement at (a), all that can be imputed

to Hobbesian contractors would be a willingness to assume a

52Hobbes, Leviathan, op. cit., p. 151.

53For the others, see J. W. Gough, The Social Contract,
Greenwood Press, 1978, ch. XII, passim.

particular tvDe of risk, viz., death by execution. From

this imputation, however, no inference can be drawn regard-

ing the knowledge these contractors would have about the

odds on one of them actually being put to death in a

particular (empirical) context. Instead, the rational basis

for their agreement would presumably key on their commitment

to an entire range of empirical possibilities. Moreover,

the range would be so wide, I suggest, that it would include

everything from the circumstances of a benign dictatorship

where, say, the death penalty is either seldom or never

employed, to a Caligula-like reign of terror where the

slightest whim of the sovereign may mean certain death even

for innocent subjects.

Now, to be sure, what knowledge may be attributed to

Hobbesian contractors is more substantial with a shift from

a hypothetical to an actual construal of the agreement at

(a). That is to say, if the agreement may be taken as

actual, then it would be plausible to think of these

contractors as knowing something (in advance) about the

person (or persons) whom they intend to elevate to a

position of sovereignty, as well as something about other

relevant features in the empirical context. Even so, I

hasten to add, any very precise calculation of the odds on

being put to death would still be rather difficult for them

to achieve. I say this because any calculation of this sort

would be tantamount to an attempt by them to predict the

behavior of the sovereign once a virtual monopoly on the use

of power had been granted. Here, I need hardly draw upon a

wealth of examples from the history of the world to inspire

an appreciation for how unreliable calculations of this sort

can be! In marked contrast, however, with either of these

construals of the agreement at (a), occupants aboard the

lifeboat may be thought to know rather precisely the odds on

any one of them being sacrificed pursuant to a lottery. For

example, if there are seven passengers on the craft, and

they all agree that each of them will have an equal chance

of being selected, and further, to this end they all embrace

procedures which are fair and reliable, then each passenger

would be in a position of knowing rather exactly his or her

own odds on being sacrificed (viz., one in seven).

Now, for lifeboat occupants and Hobbesian contractors

alike, I submit, part of what commends risk-taking as a

rational enterprise may be expressed in terms of what these

people actually know about the odds on being put to death

under their respective agreements. If I am right about

this, there should be a wide gulf in the degree of

rationality which may be properly assigned the decision of

Hobbesian contractors to embrace the agreement at (a) in

comparison to that assigned the decision of lifeboat

occupants to conduct a lottery. Moreover, it would follow

from this that the highly equivocal sort of commitment to

risk in which Hobbesian contractors engage could not be

cogently defended by analogy to the very concrete, almost

mathematically precise sort of risk-taking reflected in the

decision of the lifeboat occupants.

Finally, there is another difficulty which should be

mentioned. Even if the argument from analogy were not

subject to any of the objections I have raised, still,

merely showing that the risk of death (by execution) associ-

ated with the agreement at (a) is a risk worth taking,

(i.e., because Hobbesian contractors are thereby afforded

the best chance they will have for the successful exercise

of their rights at (b)), will do nothing whatever to reverse

a lack of cotenability regarding (a) and (b), i.e., if such

a lack can be otherwise shown to obtain as a result of

Hobbes' defense of the death penalty.54 That is to say, the

argument from analogy proves helpful to Hobbes's theory only

if my earlier objections fail and the cotenability of (a)

and (b) cannot be successfully challenged. In this latter

connection, of course, the argument itself cannot be used as

a means of defending against any charge of a lack of


2.5.3 Gauthier's Amendment

I will now consider an amendment to Hobbes' system

which David Gauthier proposes in his book, The Logic of

Leviathan.55 This amendment is introduced as a replacement

540n this, see my discussion of two counterarguments to
Hobbes' analysis, infra, 2.6.

550p. cit., pp. 148f.

for, rather than an augmentation of, Hobbes' analysis of

punishment (generally) and the death penalty (specifically).

In effect, what Gauthier argues is that the system of

Leviathan arrives at impasse if Hobbes' analysis is allowed

to stand. As a remedy for this, he introduces and defends a

modified version of the authorization agreement at (a). In

what follows I will provide an account of why Gauthier

believes Hobbes' analysis leads to impasse, and then I will

describe and comment upon the particular amendment he

proposes as a solution for this difficulty.56

For Gauthier, the problem with Hobbes' analysis stems

from the latter's claim that every contractor may be

regarded as an "author of his own punishment."57 For his

criticism, Gauthier relies on a distinction which Hobbes

himself draws between a "natural" person and an "artificial"

person. An individual functions in the capacity of a

"natural" person, on Hobbes' account, whenever that indi-

vidual's "words or actions are considered . .as his own."

By contrast, an individual whose "words or actions are

considered . as representing the words or actions of

another man" assumes the role of an "artificial" person.

"Of persons artificial, some have their words and actions

56Gauthier's book reflects an attempt to retain the
formal structure of Leviathan, while selectively rejecting,
and then replacing, certain of Hobbes' assumptions. The
amendment with which I am presently concerned represents
only one of several moves by Gauthier to revamp parts of
Hobbes' system.

57See supra, n. 11.

owned by those they represent," Hobbes continues, "and then

the person is the actor: and he that owneth his words and

actions, is the AUTHOR: in which case the actor acteth by

authority."58 Now, Gauthier takes Hobbes to task, not for

drawing this distinction, but for failing to recognize its

debilitating consequences for his system. Nowhere are these

consequences more easily discerned than in the context of

Hobbes' remarks on punishment in general and the death

penalty in particular. Yet, before I discuss the relevant

texts in this regard, preliminary attention to Hobbes'

account of the authorization agreement at (a) may prove

helpful. He writes: "The consent of a subject to sovereign

power . is contained in these words, I authorize, or

take upon me, all his actions: in which there is no

restriction at all, of his own former liberty."59 I find

this passage useful in introducing Gauthier's criticism

because the sovereign is here cast in both of the roles

which Hobbes distinguishes. At this text, of course, it is

easy to see that Hobbes features the sovereign in the role

of an "artificial" person, or as one with respect to whom

every contractor "authorizes," or "takes upon" himself or

herself, "all his actions." Yet, this is only to recognize

that, while the sovereign is the actor in all he does, every

contractor is the author, or owner, of all those actions.

58Leviathan, oD. cit., ch. 16, p. 125 (emphasis

59Ibid., ch. 21, p. 164 (emphasis Hobbes').

But now, this emphasis, although primary, is not exclusive;

for, at this same text, Hobbes also appears to cast the

sovereign in (at least) the subsidiary role of a "natural"

person. This is suggested by his depiction of the sovereign

as one whose (unrestricted) actions are explained by appeal

to "his own former natural liberty." So depicted, of

course, the sovereign would be regarded as both actor and

author (or owner) of what he does. At other texts, where

Hobbes discusses either punishment or the death penalty, the

sovereign is presented in one or the other (but not both) of

these roles.

In chapter 28 of Leviathan, Hobbes asserts:

It is manifest . that the right which the
commonwealth, that is, he, or they that represent
it, hath to punish, is not grounded on any
concession, or gift [or perhaps authorization?] of
the subject.s6

Instead, Hobbes explains, the ground for the "right of

punishing" is the sovereign's own natural right "to every

thing" (the ius naturalis).

[B]efore the institution of commonwealth, every
man had a right to every thing, and to do
whatsoever he thought necessary to his own
preservation; subduing, hurting, or killing any
man in order thereunto. And this is the
foundation of that right of punishing, which is
exercised in every commonwealth. For the subjects
did not give the sovereign that right; but only in
laying down theirs, strengthened him to use his
own, as he should think fit, for the preservation
of them all.61

60Ibid., p. 229.


After this, Hobbes discusses the wide variety of crimes to

which the sovereign's "right of punishing" (so construed)

applies, including capital crimes.62 Clearly, then, in these

passages I have cited or noted from chapter 28, it is appar-

ent that, whenever the sovereign exercises his "right of

punishing," he does so in his capacity as a "natural" person

and, as such, he is both actor and author regarding his


In an earlier chapter of Leviathan, however, Hobbes

leaves quite a different impression of the basis for the

"right of punishing." Here, his initial comments are made

with reference to a particular sort of crime, viz., the

attempt of a subject "to depose his sovereign."

[I]f he that attempteth to depose his sovereign,
be killed, or punished by him for such attempt, he
is the author of his own punishment, as being by
the institution [of a commonwealth], author of all
his sovereign shall do.63

Within this context, the ground for the "right of punishing"

is evidently linked to every contractor's participation in

the authorization agreement at (a), and so the sovereign

functions, not as a "natural" person, but as an "artificial"

person who is the actor, though not the author, of any

punishments he may impose on his subjects. A few pages

later, in a more general vein, Hobbes makes essentially the

same point when he remarks, "[B]y this institution of a

62Ibid., ch. 28, p. 232.

63bid., ch. 18, pp. 134f.

commonwealth, every particular man is author of all the

sovereign doth." From this, Hobbes draws the amusing,

though impeccable conclusion, "[H]e that complaineth of

injury from his sovereign complaineth of that whereof he

himself is author."64

Now, Gauthier's criticism of Hobbes is not that the

latter provides two, very different accounts of the theo-

retical foundation for the "right of punishing" nor even

that the sovereign is cast in two, very different roles as a

result of these distinctive accounts. These observations

alone would hardly inspire criticism, for, disparate as

Hobbes' two accounts may be, nothing in what he has to say

about punishment (per se) precludes the possibility of two,

very different, but entirely compatible theoretical deriva-

tions for the "right of punishing." Nor is there any reason

to suppose that the sovereign cannot alternately assume the

roles of a "natural" and an "artificial" person, or even

both of these roles, simultaneously.65 Instead of this, the

criticism of Hobbes Gauthier advances is that the sover-

eign's role as an "artificial" person, i.e., apropos of the

"right of punishing," requires a construal of the authori-

zation agreement at (a) which is incompatible with the

affirmation of every contractor's inalienable right of self-

preservation at (b). In other words, the claims at (a) and

64ibid., p. 136.

65In fact, the assumption of both roles will be an
essential feature of Gauthier's amendment (infra, p. 61).

(b) are not cotenable, on Gauthier's view, insofar as (a)

requires the sovereign, whenever he punishes, to act in the

capacity of an "artificial" person. The basis for this

assessment may be easily grasped, as Hobbes is adamant in

what he has to say about the right at (b). In chapter 14 of

Leviathan, he writes, "[A] man is forbidden to do that,

which is destructive to his life, or taketh away the means

of preserving the same, "66 and later in the same chapter,

[A] man cannot lay down the right of resisting
them, that assault him by force, to take away his
life; because he cannot be understood to aim
thereby, at any good to himself. The same may be
said of wounds, and chains, and imprisonment;67

and much later, in another chapter, he writes:

[N]o man is supposed bound by covenant, not to
resist violence; and consequently it cannot be
intended, that he gave any right to another to lay
violent hands upon his person.68

It follows from these texts and others, on Gauthier's

account, that insofar as the sovereign is cast in the role

of an "artificial" person, Hobbes' commitment'to (b) is

inimical to his commitment at (a), pursuant to which he

claims that every contractor is "an author of his own

punishment." To be sure, of course, an account of the

"right of punishing" in a commonwealth may also be provided

by appeal to what the sovereign may do in his role as a

66Hobbes, or. cit., p. 103

67Ibid., p. 105.

68Ibid., ch. 28, p. 229.

"natural" person; and, as I have already stated, nothing

precludes his functioning simultaneously in the capacities

of both a "natural" and an "artificial" person. Indeed,

later, when I present Gauthier's amendment to Hobbes'

theory, I will report that his innovation in fact takes for

granted that the sovereign would be cast in both roles at

one time. Yet, however this may be, what matters presently

for Gauthier's criticism of Hobbes is the idea--inescapable

in Leviathan--that regardless of any additional roles the

sovereign may assume, clearly, pursuant to the authorization

agreement at (a), he always functions as a "artificial"

person vis-a-vis his subjects, because, in Hobbes' own

words, "every particular man is author of all the sovereign

doth."69 The upshot of this is that Hobbes' commitments at

(a) and (b) taken together, on Gauthier's view, both require

and prohibit a contractor's authorization of his or her own

punishment.70 Moreover, since Hobbes is, evidently, as

69Ibid., ch. 18, p. 136 (emphasis added).

70"Require" is easier to show than "prohibit." In his
rejection of Hobbes' analysis, Gauthier assumes that a
relationship exists between every contractor's inalienable
right of self-preservation (the ius naturalis) and certain
passages of Hobbes which suggest constraints apply to what
contractors either can or may do. These passages would
include those which I cited earlier, i.e., specifically,
those where Hobbes affirms, "[A] man is forbidden to do
that, which is destructive of his life, or taketh away the
means of preserving the same"; and where he insists, "[I]t
cannot be intended [by the covenant] that . [anyone]
gave any right to another to lay violent hands upon his
person" (supra, pp. 57). What Gauthier assumes, and, I
believe, rightly so, is that the constraints imposed by
these passages receive their impetus, in some fashion, from
the operation of the right at (b). Yet, in the context of

intractable in his commitment to the sovereign's role as an

"artificial" person under (a) as he is to every contractor's

right of self-preservation at (b), and further, since these

commitments would appear difficult to reconcile in their

implications for his defense of punishment (generally) and

the death penalty (specifically), Gauthier concludes that

Hobbes' system arrives at impasse. In light of this conclu-

sion, and since Gauthier, in any case, wishes to retain the

"right of punishing" for the sovereign (including the right

to employ the death penalty), he is confronted with an obvi-

ous choice of strategies for his attempt to amend Hobbes'

theory. That is to say, he must either choose to weaken the

claim at (a), or the one at (b), or both. As it happens,

Gauthier elects to tinker only with the claim at (a). In

what follows I will present the amendment which arises from

his implementation of this strategy; and I will urge that,

if his criticism of Hobbes may be allowed to stand, then his

proposal will prove vulnerable: and this because, as I will

show, the general drift of his attack on Hobbes may be

turned against his own amendment.

his attack on Hobbes' analysis of punishment (generally) and
the death penalty (specifically), what is not so clear in
Gauthier's account is the exact mechanism by which these
constraints operate in virtue of the jus naturalis. In the
concluding section of this chapter (2.6.2), I will have
something more to say about this.

Because Gauthier regards the authorization agreement at

(a) as an essential feature of Hobbes' system,71 it is not at

all surprising to find that, in his attempt to accommodate

Hobbes' defense of the "right of punishing," he adopts the

goal of doing as little damage as possible to the claim at

(a). Moreover, since there is no way of escaping the idea

that the sovereign functions in (at least) the role of an

"artificial" person under the agreement at (a), i.e.,

because, in Hobbes' own words, "[N]o man is obliged by a

covenant, whereof he is not author,"72 it follows that

Gauthier must find a clever means of embracing this idea

generally, while nevertheless rejecting its specific

applicability for contractors whom the sovereign threatens

with punishment. It is therefore, in pursuit of this

ostensibly elusive objective, that Gauthier proposes the

following account as an amendment to Hobbes' theory:

The right to punish may . easily be explained.
Each man authorizes, not his own punishment, but
the punishment of every other man. The sovereign,
in punishing one particular individual, does not
act on the basis of his authorization from that
individual, but on the basis of his authorization
from all other individuals.73

71The stated objective for chapter four of Gauthier book
is "to set out the structure of Hobbes's mature political
theory in a manner which will show the essential role played
by authorization" (op. cit., p. 121).

720p. cit., ch. 16, p. 126.

730p. cit., p. 148.

Gauthier acknowledges that his proposal would "mean that

each subject is not author of everything the sovereign

does," and so, on this interpretation, "the right to punish

is subtly different from all other rights of the sovereign,

in that the sovereign in each act of punishing is not

exercising a right given him by all his subjects."74 He

believes, however, that an alternative defense of the "right

of punishing" may be invoked (when necessary) by appeal to

the sovereign's exercise of his right "to every thing" in

his capacity as a "natural" person. What this means is

that, whenever the sovereign threatens a contractor with

punishment, he does so in the primary role of an "artifi-

cial" person insofar as all other contractors are concerned,

but then, simultaneously, he functions only in the role of

a "natural" person regarding the one whom he threatens.

Finally, Gauthier admits that his amendment "does require a

complication in Hobbes' theory of authorization," but then

he adds, "[I]t is a complication which preserves the essen-

tial character of that theory, and so we may allow it."75

Gauthier's amendment gives rise to some unanswered

questions. Not the least of these is a general question

which may also be asked of Hobbes: What, exactly, confers

(political) legitimacy upon any action by the sovereign?

Or, more specifically put, Does every action by the

7 4;hiid


sovereign require every contractor's "authorization" in

order to be accorded status as a legitimate action on

Hobbes' system? Or does it suffice to defend the legitimacy

of some, or perhaps all, of the sovereign's actions solely

by appeal to his own right "to every thing" (the Jus

naturalist ? Or lastly, are there any circumstances under

which these two distinctive avenues of appeal may be indi-

vidually necessary, but only jointly sufficient, for the

purpose of conferring legitimacy upon a particular action by

the sovereign? While these questions are not specifically

addressed in Leviathan, it may be possible to provide some

indication of how Hobbes would respond. My clearest impres-

sion is that he would not be likely to retreat from the

position, reflected at certain texts,76 that all the sover-

eign's actions may be defended solely on the basis of appeal

to his own natural right "to every thing."77 But now, if

this is so, it invites the question of whether or not the

agreement at (a) functions at all as a ground of legitimacy

for Hobbes' system? If the answer is "yes," it would appear

that his system has been overdetermined insofar as such a

ground is concerned and, indeed, may even reflect an

unnecessary redundancy in this regard. Of course, if the

answer is "no," then an even more difficult question arises

76In Hobbes' system, the sovereign "cannot be accused by
any of his subjects of injury" (op. cit., ch. 20, p. 151);
similarly, "whatsoever he doth, it can be no injury to any
of his subjects" ibidd., ch. 18, p. 136).

77Supra, p. 35.

of why both Hobbes and Gauthier assign so prominent a posi-

tion, in their respective analyses, to the authorization

agreement at (a). My own view of the agreement at (a),

which runs counter to Gauthier's interpretation, is that it

serves only the practical purpose in Hobbes' system of

generating a monopoly on the use of power.78 Yet, perhaps

this is no small contribution to his system, for it is

precisely the collective impact of every contractor's

authorization of the sovereign which ensures that his

exercise of his natural right "to every thing" will be

effective, in addition to being, on its own merits,

legitimate. Still, nothing of what follows in my assessment

of Gauthier's amendment will depend on my being correct in

my conjecture about how Hobbes would respond to the ques-

tions I have raised. What may prove helpful, however, is

that I have utilized this occasion to introduce the various

ways in which Hobbes' system may be thought to ground its

claim to legitimacy.

Perhaps the most plausible interpretation of Gauthier's

amendment is that he requires that the sovereign's natural

right "to every thing" operates as a ground for the legi-

timacy of his action in threatening a contractor with

punishment, i.e., insofar as that particular contractor is

concerned; but that, insofar as all other contractors are

78For a view similar to my own, cf. Jean Hampton, Hobbes
and the Social Contract Tradition, Cambridge University
Press, 1986, ch. 5.

concerned, the authorization agreement at (a) is the appro-

priate ground.79 This interpretation, however, would seem to

pose more questions than it answers. In particular, it

poses but does not answer a question of why, if the agree-

ment at (a) is a ground for everyone with regard to some of

the sovereign's actions, then, does it not provide a ground

for everyone with regard to all of his actions. Indeed,

what seems so attractive about Hobbes' original analysis, I

suggest, is that every contractor gives his or her consent

for all of the sovereign's actions. Now, what Gauthier's

amendment does is to introduce a modification of Hobbes'

theory under which the latter's commitment to a theoretical

foundation of consent and authorization is withdrawn in

those cases (and only those cases) where a particular con-

tractor is threatened with punishment by the sovereign (and

then only with regard to that particular contractor). Yet,

I wonder, under Gauthier's amendment, what motivation any of

Hobbes' contractors would have, given that they ordinarily

value their acts of consent and authorization so highly on

79For the former ground, Gauthier writes: "[I]n punish-
ing, the sovereign is no longer acting as the representative
of the person punished, and so . he is placing himself
in the position of an enemy with respect to that person";
and, of course, just prior to this, Gauthier emphasizes:
"[I]n the state of nature . every man has the right to
kill, injure, or subdue others" (op. cit., p. 184; cf.
Cattaneo's observation, supra, n. 30). For the latter
ground, once again I quote Gauthier's innovative move: "The
sovereign, in punishing one particular individual, does not
act on the basis of his authorization from that individual,
but on the basis of his authorization from all other
individuals" (loc. cit.).

Gauthier's construal, for excepting those instances which

threaten them with the greatest possible encroachment upon

their personal liberties (as with imprisonment) and even

perhaps upon their very lives (as with capital punishment).

It seems to me, then, if any value at all is to be placed on

having a foundation of consent and authorization for Hobbes'

theory, these commitments would most certainly include those

areas where the personal liberties and even the lives of

contractors may be so profoundly compromised. Of course,

Gauthier's reaction to my analysis would be to insist that

his amendment is necessary if Hobbes' theory is to avoid

impasse.80 In reply to this, I can only say that, if

Gauthier is to be permitted to cast aside selectively what

he himself regards as an essential feature of Hobbes'

theory,81 and moreover, to do so only at that place where its

applicability would appear to have its greatest plausibility

and justification, and further still, to do so only for the

purpose of repairing Hobbes' theory, then I, for one, cannot

see that the theory of Leviathan fares any better under

Gauthier's amendment that it does on Hobbes' original


Yet another criticism of Gauthier's amendment should be

briefly mentioned, and this one takes account of the full

range of possibility with regard to how the ground of

soIndeed, this is the only argument he gives on behalf
of this amendment!

81Supra, n. 71.

legitimacy may be construed on Hobbes' system. In particu-

lar, this criticism should be effective against any interpre-

tation of Hobbes' theory which attempts to show that no

authorizing function may be imputed to Hobbesian contractors

who are threatened with punishment by the sovereign; and so,

no conflict between (a) and (b) arises on this basis.82 My

essential point, then, is this: Even if a contractor does

not "authorize" his or her own punishment in the narrow

sense of granting legitimacy to the sovereign's threats in

this respect, still, in a broader sense, every contractor

(without exception) "authorizes," pursuant to the agreement

at (a), the very conditions which make it possible for the

sovereign to threaten whomever he pleases with punishment,

and moreover, for him to do so with an awesome efficiency

which practically ensures the outcome. My point, then, is

here largely an epistemological one. Hobbesian contractors

are not fools and so they can plainly see that the sovereign

is just about as powerful with Gauthier's amendment as he is

under Hobbes' original analysis. Moreover, it follows from

this that "authorization," broadly conceived, runs as easily

afoul of Hobbes' commitment at (b) as does the narrower

conception on Hobbes' own analysis; hence, if the system of

Leviathan arrives at impasse on Hobbes' original analysis,

as Gauthier purports, then it would appear to do so just as

readily on the amendment Gauthier proposes.

82For an example of this general line of interpretation,
see Hampton, loc. cit.

2.6 A Rejection of Hobbes' Analysis

To this point I have examined three attempts to augment

or amend Hobbes' analysis and I have explained why they have

been unsuccessful, i.e., insofar as they purport to show how

use of the death penalty can be reconciled with Hobbes'

claims at (a) and (b). While the appeal to Hohfeld's

distinction (2.4.1) merits attention for the (limited)

purpose of showing that the right of self-preservation

(Hobbes' just naturalis) cannot be an instance of a "claim"

right, still, all this reveals is that the interests of

those who bear this right may be radically opposed

(illustrated, e.g., by the conflict between a person

condemned to death and his or her executioner) and yet not

be the basis of any theoretical impasse. This would be a

commendable use of Hohfeld's distinction except for the

impression it leaves that the specter of impasse may be

retired altogether by this revelation. In point of fact,

however, a much more adequate basis for a challenge to the

integrity of Hobbes' system emerges in connection with an

entirely different sort of conflict, as Gauthier rightly

observes (2.5.3), and as I myself have characterized in my

earlier discussion (2.4). I refer, of course, to the

tension between every contractor's inalienable right of

self-preservation, affirmed at (b), and the act of

authorization, affirmed at (a), by which every contractor

incurs the risk (at least) of being condemned to death and

executed. Any threat to the integrity of Hobbes' system, I

suggest, arises as a consequence of this conflict, rather

than the one which Hohfeld's distinction is supposed to


Problems of a different sort arise in connection with

the lifeboat lottery analogy, many of which I have discussed

at length already (2.4.2). One point I have not touched

upon, however, is that this analogy may be very plausibly

viewed as imposing (almost surreptitiously) a relaxation of

the inalienability feature of the right affirmed at (b).

That is, (in effect) it may be construed as urging that

every contractor's right of self-preservation be partially

alienated so that use of the death penalty may be allowed

and everyone's (overall) prospects for survival thereby

improved. If this analysis is correct, then, the lifeboat

lottery analogy "resolves" the tension in Hobbes' system by

amendment, albeit, by amendment in the guise of an

augmentation; and so, Hobbes' own analysis stands corrected.

Of course, I have also taken pains to show that the analogy

itself fails on its own merits (i.e., apropos of a hypo-

thetical contract), and my assessment, in this regard, may

be evaluated independently of whether or not Hobbes' system

has been amended in the way I have suggested.

Finally, I have examined Gauthier's critique of Hobbes'

analysis, which, I believe, correctly identifies the source

of tension.in Hobbes' system (2.4.3). Also, I have

discussed the manner in which Gauthier expressly amends

Hobbes' analysis by tinkering with the authorization

agreement at (a). Here my conclusion has been that the

tension in Hobbes' system is not alleviated by his tinkering

since the critique Gauthier offers may be effectively turned

against his own amendment. In what remains of this chapter,

I will examine two other critiques of Hobbes' analysis, both

of which purport to raise serious doubts about the reconcil-

ability of (a) and (b) with any defense of the death penalty

based on his theory.

2.6.1 Inference from Contractors' Psychological Incapacity

In a recent issue of the History of Philosophy

Ouarterly, David Heyd presents the first of two critiques I

will examine.83 In this journal Heyd describes his purpose

as one of exposing "an inner inconsistency in Hobbes' justi-

ficatory argument" for the death penalty.84 To achieve this

objective, he introduces a counterargument which may be

given the following characterization:

(i) In a civil society each of Hobbes' contractors
would be psvcholoaicallv incapable of failing to
resist any attempt by the state to put him or her
to death;85

830p. cit., pp. 119-134.

84Ibid., p. 120.

85"The general vein of Hobbes's argument is typically
naturalistic. It relies on the psychological impossibility
that one should willingly subject oneself to acts of
violence, even where these are "normatively" justified, as
typically in the case of capital punishment" ibidd.)

(ii) At the time of their initial agreement (i.e.,
the social contract), each of Hobbes' contractors
would know that (i) is true;86

(iii) In light of (i) and (ii), it would be
irrational for any contractor to strike an
agreement exposing him or her to the risk of being
put to death by the state;87

(iv) At the time of their initial agreement,
Hobbes' contractors would not do that which they
know to be irrational for them to do;

(v) Hence, no contractor would strike an agreement
exposing him or her to the risk of being put to
death by the state.

There is, of course, a textual basis in Leviathan for

what Heyd attributes to Hobbes at (i). That is to say,

Hobbes does appear to embrace the view that "man by nature"

will always offer resistance to being put to death.88 Heyd's

interpretation, in this regard, finds support in Hobbes'

remark that a man cannot covenant not to resist death

because "man by nature chooseth the lesser evil, which is

danger of death in resisting; rather than the greater, which

86"[T]he real question is whether it would be rational
on their part to give their consent to . laws
[establishing capital punishment] in the first place,
knowing in advance that they will have a reason (or rather
an irresistible cause) to violate the law" ibidd., p. 123,
emphasis Heyd's).

87Infra, p. 72.

88"[A] man is forbidden [in a psychological sense, on
Heyd's account] to do that, which is destructive of his
life, or taketh away the means of preserving the same; and
to omit that, by which he thinketh it may be best preserved"
(Leviathan, op. cit., ch. 14, p. 103); arriving at Heyd's
conclusion, Kavka imputes to Hobbes the view that "it is
psychologically impossible not to resist when attacked" (Op.
cit., p. 330).

is certain and present death in not resisting."89 For this

reason, Hobbes continues, criminals are led to execution by

"armed men, notwithstanding that such criminals have con-

sented to the law, by which they are condemned."90 In a

later chapter Hobbes writes:

If a man, by the terror of present death, be
compelled to do a fact against the law [e.g.,
resist his executioner], he is totally excused;
because no law can oblige a man to abandon his own
preservation. And supposing such a law were
obligatory; yet a man would reason thus, If I do
not. I die presently; if I do it. I die
afterwards; therefore by doing it, there is time
of life gained; nature therefore compels him to
the fact.91

Here, I think, Heyd quite plausibly construes a criminal's

compulsion (by nature) to resist being put to death as the

equivalent of a psychological incaDacity not to resist; and

so, accordingly, I have adopted at (i) this manner of charac-

terizing the claim he imputes to Hobbes. What I am not

prepared to accept, however, is the claim itself, which may

be easily refuted by counterexample. Indeed, in another

context, Heyd himself successfully refutes the claim at (i)

upon citing the case of "a Socratic principled refusal to

avoid execution. "92 What I find a little odd here is that he

would take pains to refute a claim which he needs as a

89Leviathan, op. cit., ch. 14, p. 110.


91Ibid., ch. 27, p. 223 (emphasis Hobbes').

92Heyd, oP. cit., p. 122.

premise for his counterargument against Hobbes. Still,

perhaps this can be explained by supposing that the point of

his attack on Hobbes is merely one of showing that the

latter's defense of the death penalty fails even if (con-

trary to fact) (i) were true.

As the claims at (ii) and (iv) are both credible as

interpretations of the text of Leviathan and plausible on

their own merits, I will concede them to Heyd without

further comment.93

The chief difficulty I see with Heyd's counterargument

relates to the inference from psychological incapacity at

(iii). Here, Heyd's own words should be reported: he

states, "[T]he fact that a penalty is psychologically

impossible to endure entails that the individual [who knows

this] cannot rationally agree to introduce it into the

contract."94 It is with this (alleged) entailment that I

have a problem. It would appear to me that Heyd is simply

mistaken to infer any sort of entailmentt" at (iii), whether

it be of the logical or psychological variety. Any thought

of a logical entailment, of course, may be dismissed upon

93Both rationality and a knowledge of one's own
limitations would seem reasonable imputations regarding
Hobbes' contractors.

94Ibid., (emphasis added). It may well be that Heyd
intends only to say here that Hobbes (and not Heyd himself)
is committed to the entailment at (iii) and, therefore, the
whole point of the inference in context would be to show
that Hobbes' own logic should lead him to reject the death
penalty; yet, if this is so, why not simply a frontal attack
on the inference itself, which, as I myself show in due
course, can be undertaken successfully?

even a cursory examination of the claims at (i) and (ii).

In all probability, therefore, what Heyd intends to suggest

at (iii) is an entailmentt" of the psychological variety.

But now, consider this: suppose I am standing on a railroad

trestle with a fast train approaching and no way of avoiding

certain death except by jumping into the lake below.

Suppose, too, that I have enough time to make a rational

decision about what to do. Under these circumstances, I

suggest, I might very well be rationally disposed to make

the jump even though the lake is populated with hungry

alligators and I cannot bear the thought of being torn to

shreds by the menacing creatures who await my plunge. There

is, after all, a chance--however remote--that I may survive

the ordeal unscathed. In much the same vein, then, even

though Hobbes' contractors may find it "psychologically

impossible" for them to endure a death by execution (as

Heyd contends), still, this incapacity, in and of itself,

need hardly present any psychological impediment to their

accepting the risk of being put to death in this fashion if

they believe themselves to be otherwise doomed or at least

otherwise seriously compromised insofar as their (overall)

prospects for survival are concerned.95 While I am prepared

to concede that there may be some people who may be inclined

to reject any risk of being executed even though they

95But see my rebuttal to Samuel J. M. Donnelly's
imperilment of liberty argument, infra, 6.3.

otherwise face certain death from other causes,96 I know of

no empirical studies which show this to be true of all human

beings or even most of them. Accordingly, I am equally

unaware of any psychological law or principle to which Heyd

could appeal in defense of his inference at (iii); moreover,

since his claims at (i) and (ii) cannot by themselves

provide adequate support for any sort of psychological

entailmentt" at (iii), his conclusion at (v) must be


Heyd's critique fails, but there is a silver lining.

His attack may be applauded, I submit, insofar as he

explores a question of whether it would be rational of

Hobbes' contractors to expose themselves to the risk of

being put to death by the state. Ultimately, what Heyd

tries to show is that, because of certain psychological

impediments, Hobbes' contractors cannot be rationally

disposed to authorize a permission for a (derivative)

society to employ the death penalty pursuant to their

agreement at (a). That he fails in this endeavor has been

demonstrated by the ease with some of the premises he

requires may be challenged. Yet, there is a more promising

way of exploring the question of the rationality of the

death penalty in this context. An alternative approach

96In later chapters I move beyond mere concession and in
fact seek to establish a rational basis for this point of
view, that is, of course, with regard to the circumstances
of a hypothetical contract; on this, see infra, 5.3.3, 6.3,
and 7.3.

would be to rely on a much weaker initial premise than does

Heyd. What I would suggest, in this regard, would be to

assert merely that it cannot be excluded, on Hobbes' theory,

that at least some of his contractors would have a rational

basis for rejecting the death penalty. If this weaker

premise could be sustained, of course, then it would follow

that no unanimous decision to authorize a permission for a

(derivative) society to employ the death penalty could be

safely imputed to Hobbes' contractors as a feature of their

agreement at (a). In later chapters, I do in fact attempt

to establish this weaker premise, both for the purpose of

criticizing another theory (7.3) and with the aim of

achieving a general conclusion (8.2). For now, I would

simply note that if I am successful with those efforts, my

results there could also be usefully applied as a challenge

to the cogency of Hobbes' defense of the death penalty.

2.6.2 A Reductio Ad Absurdum on Hobbes' Analysis

In this final section I will present what I consider to

be a telling argument against Hobbes' defense of capital

punishment. In the course of this presentation I will

introduce analysis of my own to show that Hobbes allows that

the authorization agreement at (a) would be effectively

dissolved for any contractor upon whom the death sentence is

pronounced. An important consequence of this dissolution, I

will argue, is that any use of the "death penalty" would in

fact be extraneous to the set of permissible actions

authorized by the agreement at (a). In other words, on this

account, no contractor of Hobbes can be supposed to have

agreed, in agreeing to (a), that (a) includes a permission

for a (derivative) society to employ the death penalty. For

the support of these contentions I will appeal to the text

of Leviathan. Ultimately, my purpose will be to show that

Hobbes' position on the death penalty is contradictory and,

therefore, theoretically untenable.

There can be no doubt, of course, that Hobbes under-

stands the authorization agreement at (a) to include a

permission for a (derivative) society to employ the death

penalty. In his own words, "it may, and doth often happen

in commonwealths, that a subject may be put to death by the

command of the sovereign power."97 In another context, he

reports that "[condemned] criminals have consented to the

law by which they are condemned;"98 and, in still another, he

affirms that "consent of a subject to a sovereign power"

implies a permission (from the subject's own perspective)

for "him [i.e., the sovereign] to kill me."99

But now, Hobbes also affirms in Leviathan that the

agreement at (a) remains in effect for his contractors only

as long as the sovereign provides them with protection. He


97Leviathan, o. cit., ch. 21, p. 161.

98Ibid., ch. 14, p. 110.

99Ibid., ch. 21, p. 164 (emphasis Hobbes').

The obligation of subjects to the sovereign, is
understood to last as long, and no longer, than
the power lasteth, by which he is able to protect
them. For the right men have by nature to protect
themselves, when none else can protect them, can
by no covenant be relinquished.100

The death sentence, for Hobbes, is a case in point, and this

is for the obvious reason that the sovereign's protection of

a contractor ceases whenever the former moves against the

latter with a design to take his or her life. Moreover,

with this cessation of protection, I submit, Hobbes evi-

dently intends that a dissolution of the agreement at (a)

occurs insofar as the condemned person is concerned. So,

for example, Hobbes can say, "If the sovereign command a

man, though justly condemned, . not to resist those who

assault him, . yet hath that man the liberty to dis-

obey."101 Several commentators have observed, correctly in

my view, that the circumstances of a person under sentence

of death in Hobbes' system are those of a person who has

been returned to the "state of nature."102 The importance of

this observation, for my purposes, is that, ipso facto, any

"death penalty" confrontations necessarily occur beyond the

100oobid., p. 167.

1011bid., p. 164.

102"The relation between the individual and the sover-
eign in the case of capital punishment is that characteriz-
ing the state of nature. The sovereign has a 'right' only
in the sense of power (and the same applies to the indi-
vidual trying to escape the death sentence)" (Heyd, op.
cit., p. 123); see also Cattaneo, supra, n. 30, and
Gauthier, The Loaic of Leviathan, op. cit., p. 148.

purview of what is authorized by the condemned person at

(a). In effect, then, Hobbes' sovereign first of all

cancels the agreement between the person to be condemned and

the rest of society, and then and only then moves to kill

that person. From the perspective of the person to be

condemned, therefore, the authorization agreement at (a)

does not include any permission for a (derivative) society

to employ the death penalty.103 It merely includes an

implied permission for the sovereign to return a contractor

to the "state of nature." So viewed, the sovereign's

subsequent action in taking the life of the one who has been

disfranchised in this way is hardly a "punishment" in any

ordinary sense of the word, but is instead more like an act

of war committed against an external enemy of the society.104

Indeed, Hobbes himself appears to make just this point when

he writes, "Hurt to revolted subjects is done by right of

103This is the mechanism, implied by the authorization
agreement at (a), by means of which Hobbes tries (unsuccess-
fully, as I will show) to make room for his commitment at
(b); on this point, cf. my analysis of Gauthier's amendment,
supra, pp. 58ff, and see especially n. 70.

104"[I]f a subject shall by fact, or word, wittingly,
and deliberately deny the authority of the representative of
the commonwealth, .. he may lawfully be made to suffer
whatsoever the representative will. For in denying subjec-
tion, he denies such punishment as by the law hath been
ordained; and therefore suffers as an enemy of the common-
wealth; that is according to the will of the representative.
For the punishments set down in the law, are to subjects,
not to enemies; such as are they, that having been by their
own acts subjects, deliberately revolting, deny the sov-
ereign power" (Leviathan, OD. cit., ch. 28, p. 231). Also,
see Cattaneo's comment, supra, n. 30.

war, not by way of punishment."105 But now, this is all from

the perspective of the one who has been disfranchised. From

the perspective of everyone else in society, the sovereign's

action in taking the life of this person is an action pur-

suant to the authorization agreement at (a). After all,

killing an external enemy of society would seem to be one

way in which the sovereign provides protection for these

people. Yet, even from this perspective, all that is author-

ized pursuant to (a) is the killing of an external enemy,

not the Dunishing of a subject by taking his or her life.

Again, in Hobbes' words, "[H]arm inflicted upon one that is

a declared enemy falls not under the name of punishment."l06

Regardless of perspective, therefore, one very important and

highly pertinent conclusion emerges: no permission is given

by anyone pursuant to the agreement at (a) for a (deriva-

tive) society to employ the death penalty against one of its

citizens. On my analysis, then, Hobbes' position on capital

punishment in Leviathan is contradictory because he both

expressly affirms and denies by implication (as I have

shown) that the agreement at (a) yields such a permission.

Moreover, since a contradiction follows from the agreement

at (a), by reduction (a) itself is untenable and therefore

cannot be cotenable with (b). It strikes a telling blow at

the very foundation of his theory, I would conclude, for

105Ibid. (emphasis Hobbes').


Hobbes to advance an affirmative argument for the death

penalty in the manner he does.107 Evidently, then, no

contractarian justification for the death penalty can be

sustained, either by Hobbes' own analysis (as I have

explained in this section), or by the efforts of others to

augment or amend his theory (as I have explained in the

previous section). What is left to be considered, there-

fore, in the chapters which remain, is whether or not the

same could be said for the other (major) theorists in the

social contract tradition.

107As a practical matter, of course, Hobbes' contractors
do knowingly (I would say) expose themselves to the risk of
being put to death by the sovereign, since he is evidently
authorized by them to return any one of them to a "state of
nature" and there to do with them "whatsoever he will." A
question which merits asking, then, is why any fully
rational contractor would ever participate in an agreement
where the terms are subject to dissolution at the whim of
another person, and moreover, are so under circumstances
where so great an imbalance in power obtains that near
certain death may well await any person who has been
disfranchised in this way. However this may be, nothing in
what I have just noted implies any authorization whatever by
Hobbes' contractors for use of the death penalty under the
terms of a social contract.


3.1 The Lockean Right of Self-preservation

In his Second Treatise on Civil Government John Locke

defines "political power" as "a right of making laws with

penalties of death."1 Locke's account of the death penalty

relies on the notion of a person's forfeiture of the right

of self-preservation and, as such, represents a substantial

departure from Hobbes' handling of the jus naturalis. While

he agrees with Hobbes that a person cannot relinquish, trans-

fer, or otherwise abandon the right of self-preservation, he

disagrees with him in allowing that an individual can

"forfeit" this right by committing certain acts of (suf-

ficient) "heinousness."2 With the exception of murder,

Locke does not specify what these acts are; although he does

state that when a murder is committed, the perpetrator "may

lIn Social Contract, Sir Ernest Barker, ed., Oxford
University Press, 1962 (Locke's Second Treatise originally
published: 1690), p. 4 (1.3) (emphasis added); note that
Locke's own chapter and section numbers appear in
parentheses after my citation of Barker's text (this is a
practice I will observe throughout this chapter).

2Ibid., p. 50 (VII.87); see also pp. 15f (IV.23).

be destroyed as a lion or tiger, one of those wild savage

beasts with whom man can have no society nor security."3

Locke's departure from Hobbes is all the more striking

because the notion of "forfeiture" removes any possibility

of an intrasystemic conflict between the ius naturalis and

the agreement by which a social contract is generated.

Another feature of the Second Treatise, which contrasts with

Hobbes' account, is Locke's unequivocal appeal to a unitary,

extrasystemic ground of legitimacy for the death penalty,

viz., "that great law of nature, Whoso sheddeth man's blood,

by man shall his blood be shed."4 Prior to the advent of a

social contract, Locke explains, everyone who lived under

this law in a "state of nature" possessed a "right of punish-

ment" which permitted any one of them to put a murderer to

death; however, with the emergence of a civil society, he

continues, everyone agreed to transfer their individual

rights to impose "penalties of death" to the one (or ones)

whom they elevated to a position of authority over all the

rest. This they did with the expectation, under this

arrangement, that fewer abuses and errors of judgment

would be encountered, and so the lives of all of them would

3Ibid., pp. 8f (II.11).

4Ibid., p. 9 (II.11); Locke quotes Genesis 9:6 (King
James Version of the Bible); the emphasis is his.

be rendered more secure.5 This account may be sharply

distinguished from that of Hobbes, in which there is no

shortage of ambiguity about the ground of legitimacy (as I

have shown),6 and, moreover, in which whatever ground or

grounds there are require an intrasystemic affirmation

that takes the form of an "authorization" by Hobbesian


Finally, the right of self-preservation itself must be

assigned a different status on the two accounts. The ius

naturalis of Leviathan, as I have argued,7 should be con-

strued as something weaker than even a "permission" right,

while the Lockean right of self-preservation, in marked

contrast to this, merits recognition as a "claim" right. In

this vein, Locke may speak for himself.

Every one, as he is bound to preserve himself, and
not to quit his station willfully, so by the like
reason, when his own preservation comes not in
competition, ought he as much as he can to
preserve the rest of mankind.8

I take this to mean that Lockean contractors, as well as

people in a "state of nature," all have not merely a right

5"I easily grant that civil government is the proper
remedy for the inconveniences of the state of nature, which
must certainly be great where men may be judges in their own
case, since 'tis easy to be imagined, that he who was so
unjust as to do his brother an injury, will scarce be so
just as to condemn himself for it" ibidd., pp. 9f) (11.13).

6Supra, 2.4.3, especially pp. 61ff.

7Supra, 2.4.1.

SLocke, op. cit., p. 6 (11.6) (emphasis added).

but a duty of self-preservation; moreover, they each have a

claim against everyone else not to interfere with that right

and duty so long as neither a state of war obtains nor

forfeiture occurs.9

These differences, then, on which I have now elabo-

rated, give rise, in turn, to a different set of problems

from those I discussed in Chapter Two. In what follows it

will be my purpose to address three major areas of concern I

have with Locke's defense of the death penalty. All three

of these areas will be seen to emphasize one or more of the

several differences with Hobbes upon which I have commented.

The first of these areas has to do with a question of

whether there are any "natural laws," and, if so, whether

any of these operate to compel use of the death penalty

under a social contract (3.2). After attention to this

area, I will then take up the issue of Locke's reliance on

biblical texts (especially Genesis 9:6) as a conduit of

"natural law" regarding capital punishment. Here (3.3), I

will explore whether or not the exegetical principle upon

which he relies can be plausibly defended. My final area of

concern will feature certain difficulties I perceive regard-

ing how punishment of the innocent would be construed on

Locke's theory (3.4). My analysis in all three of these

areas will then be summarized and presented in conjunction

9As Michael Freeden astutely observes of Locke, "What
is on one dimension a duty to God becomes on another a right
against other men" (Rights, Open University Press, 1991, p.

with a concluding perspective on Locke's defense of the

death penalty (3.5).

3.2 The Skeptical Challenge to "Natural Laws"

It is certainly a question worth asking whether or not

there are any "natural laws" which operate to bind the

actions of human beings. The quintessential challenge, in

this regard, confronting all "natural law" theorists (includ-

ing Locke), has been concisely expressed in the colorful

words of the Danish jurist, Alf Ross:

Like a harlot, natural law is at the disposal of
everyone. The ideology does not exist that cannot
be defended by an appeal to the law of nature.
And, indeed, how can it be otherwise, since the
ultimate basis for every natural right lies in a
private direct insight, an evident contemplation,
an intuition?10

On the heels of this assessment, Ross invites the conclusion

he presses by means of a further question, "Cannot my intui-

tion be just as good as yours?"11 To be sure, of course, if

Ross's question is answered in the affirmative, then there

is no reason to prefer Locke's appeal to the "natural law"

over the account of another theorist, or even over the

incantations of a soothsayer or perhaps an ancient oracle.

On the other hand, however, if an answer is given in the

negative, then, I urge, after the manner of Ross, that a

burden falls on those who make these appeals to show every-

one else (who needs convincing) just why their intuitions

100n Law and Justice, University of California Press,
1959, p. 261.


(and not mine or someone else's) reveal laws which are just

the ones competent to bind the actions of human beings.

Still, it is not my intention here to defend Ross's skepti-

cal view regarding the existence of "natural laws." Such an

enterprise would be unnecessary given the more limited aims

of my present inquiry. That is to say, it will be enough

for my purposes if I am able to show that Locke fails to

meet the burden of making a convincing case in only one

instance of an alleged "natural law," viz., that particular

appeal on the basis of which he intends to argue for a

societal permission to impose the death penalty. If I am

successful with this, as I expect to be in the section which

follows (3.3), then it would hardly matter (for my aims)

that a rigorous defense of Ross's skeptical view may also be

possiblel2 and, if implemented effectively, may well have

untoward consequences for Locke's entire system.

Yet another aspect of Locke's use of "natural law"

should be mentioned. Suppose, for the sake of argument,

that Locke does succeed in convincing everyone (who needs

convincing) that there is some "natural law," say X, which

confers upon individuals in a "state of nature" a right to

impose "penalties of death" for a certain crime (or crimes).

Assume, too, that this right is transferable to authorities

in a civil society. Now, what I want to emphasize here will

12Ross acquits himself rather well in this regard; on
this, see his ch. 11, "Analysis and Criticism of the
Philosophy of Natural Law" (ibid.).

seem obvious enough, though some elaboration will be helpful

to show its relevance within the present context. That is

to say, the right at X (on Locke's account) would be just

that: a right and not a duty. It confers upon individuals,

and by transfer, upon the authorities in a civil society, a

permission to impose "penalties of death," but not an

obligation to do so. In effect, what this means is that it

is a prerogative of individuals in a "state of nature," and

of the civil authorities under a social contract, whether or

not to impose the death penalty in particular cases. Locke

so much as confirms this when he writes, "[A murderer] may

be destroyed as a lion or a tiger,"13 where the operative

term is 'may' rather than 'must'; and later, when he dis-

cusses specific circumstances where the right to employ the

death penalty would not be exercised.14 Also, he devotes an

entire chapter of the Second Treatise to a defense of the

sovereign's prerogative in matters which include actions

that would "pardon some offender."15 So, while the right at

X may be very clear about when a person has forfeited the

right of self-preservation, it does not follow, on Locke's

13Locke, op. cit., pp. 8f (II.11).

14"Indeed, having by his fault forfeited his own life by
some act that deserves death, he to whom he has forfeited
it, may, when he has him in his power, delay to take it, and
make use of him to his own service" ibidd., pp. 15f (IV.23),
emphasis added).

15Ibid., ch. XIV, "Of Prerogative"; see especially p. 95

account, that a person so disposed must in fact be put to

death. The difficulty I have, then, with a "natural law"

such as X, which confers on individuals and then civil

authorities a right whose exercise is a matter of human

prerogative, is that, even if X is based on intuitions which

everyone has (or can accept), still, in its implementation,

X relies on human discretion which, in turn, may reflect the

whims, the biases, and, under the most unfortunate of circum-

stances, even the pernicious motives of particular human

beings. So even on this account, I suggest, which concedes

much to Locke, there may yet be room for a skeptical view

after the manner of Ross.

3.3 Difficulties with Locke's Reliance
on Biblical Texts

In this section I will explore a question of whether or

not Locke's reliance on biblical texts (especially, Genesis

9:6) can perform the pivotal function it must perform in

order to sustain his argument for the death penalty.16

Earlier, I made the point that forfeiture of the right of

self-preservation arises, under Locke's theory, whenever a

person commits an act of (sufficient) "heinousness." Since

he makes appeal to Genesis 9:6 in the case of murder, it

would appear safe to infer that the exegetical principle by

which he is guided in his selection of this text is some-

thing like this: "If p is cited in a biblical text as an

16Locke also cites Genesis 4:14, "Every one that findeth
me shall slave me" (KJV) ibidd., p. 9 (II.11), emphasis

act worthy of death, then p is sufficiently heinous to

warrant the death penalty." Since Locke himself provides no

guidance at all about how heinous an act would have to be in

order to be sufficiently heinous to warrant the death pen-

alty, the principle of selection I have imputed to him does

at least have the virtue of clarifying that point. But now,

it will be quickly seen that, if this principle is applied

to certain other biblical texts (from the Old Testament),

rather peculiar, and in fact morally counterintuitive, conse-

quences follow. For example, at Exodus 21:15 there is the

pronouncement, "Whoso strikes his father or mother shall be

put to death," and at 21:17, "Whoso curses his father or

mother shall be put to death." Also, at Exodus 22:18 there

is the statement: "You shall not permit a sorceress to

live," and at Leviticus 21:10, "If a man commits adultery

with the wife of his neighbor, both the adulterer and the

adulteress shall be put to death."17 Even though these texts

identify death as the appropriate punishment in each case,

still there is not any reason to think that Locke seriously

intends that unruly children, sorceressess," adulterers, and

adulteresses should follow murderers to the gallows.18 In

17Citations by the author are from The Oxford Annotated
Bible (Revised Standard Version), Oxford University Press,

18Nor, evidently, was this the intention under Talmudic
law, which, according to Steven S. Schwartzschild,
"abolished the death penalty de facto two-thousand years
ago" ("Kantianism on the Death Penalty," ARSP LXXI (1985),
p. 345; for additional bibliography, see Schwartzschild's
n. 10 ibidd.); see also J. K. Mikliszanski, "The Law of

fact, he appears to block this possibility when he states in

the Second Treatise that "lesser breaches of the law [than

murder]" should be punished with such severity as to cause

the offender "to repent."19 In the absence of any reason why

Genesis 9:6 should be singled out as a conduit of "natural

law" on capital punishment, while the other texts I have

cited may be safely ignored in this respect, even though

they also specify acts sufficiently heinous to warrant

"penalties of death" (or at least do so under the exegetical

principle I have imputed to Locke), I am left with one of

two conclusions, neither of which bodes well for Locke's

system. That is to say, either (1) he embraces the highly

peculiar and counterintuitive side effects of the exegetical

principle I have imputed to him (which would require unruly

children and others to follow murderers to the gallows), and

this, in opposition to what he himself alleges regarding

"lesser breaches of the law"; or (2) he abandons altogether

his appeal to biblical texts upon which his defense of

capital punishment ultimately rests.

3.4 A Problem with Punishment of the Innocent

The question of Locke's reliance on "natural law" and

biblical authority aside, there are yet other difficulties

which arise on his account of the death penalty. Of these,

the difficulties which pertain to his handling of matters

Retaliation and the Pentateuch," Journal of Biblical
Literature LXVI (1947), pp. 295-303).

19Locke, op. cit., p. 9 (11.12).

relating to punishment of the innocent would seem to merit

closest scrutiny. I suggest this because he appears to link

the notion of forfeiture to the act (itself) of murder,20 and

not to a finding of guilt by others, and yet, it is the

latter and not the former which in fact determines who will

be put to death, and which allows for the possibility (at

least) of punishment of the innocent.21 It may then be asked

of Locke: Does acceptance of the risk which this possibil-

ity entails have its basis in the social contract, or is it,

too, compelled of Lockean contractors (like forfeiture) by

some (as yet unspecified) manifestation of "natural law"?

Here, I will concede, for the sake of argument, that Locke

is able to provide a plausible ground for his notion of

forfeiture, contrary to what I have already shown. That is

to say, here my only concern will be to explore the question

of whether or not his notion of forfeiture can be reconciled

20Infra, pp. 95f.

21Worry over the possibility of error is not merely a
theoretical concern: Michael L. Radelet, Hugo Adam Bedau,
and Constance E. Putnam have provided extensive
documentation for more than 400 instances of wrongful
conviction for capital crimes in the United States alone; of
this number, twenty-three were in fact subsequently executed
(In Spite of Innocence, Northeastern University Press,
1992). Interestingly, too, infrequency of use is no
guarantee against the possibility of error. The state of
Israel, whose courts have rendered only two death sentences
in its 45-year history, evidently erred in one of these
cases (Adolf Eichmann was hanged in 1962, following his
celebrated trial; John Demjanjuk's 1988 conviction (also for
war crimes during the second world war), however, was
reversed by the Israeli Supreme Court in 1993 (New York
Times, July 30, 1993, p. A4)).

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs