Freedom under an Order of Public Law: From Hobbes through Hayek
to Republicanism David Dyzenhaus1
INTRODUCTION
In Two Concepts of Liberty, Isaiah Berlin distinguished between
negative and positive libertyrespectively freedom from and freedom
to--and argued for the priority of negative liberty. 2 Berlin
argued that positive liberty is not liberty at all but rather a
misleading description of a plurality of values, for example,
social equality, that compete with negative liberty; hence, we
should be wary of claims that any sacrifice in liberty to promote
one of these goods is no sacrifice because a gain in the good is by
definition a gain in liberty. Berlin was clear that such sacrifices
are often morally justifiable, even required, but adamant that the
moral cost in making the sacrifice should be recognized and given
its proper weight. Berlin considered his defence of negative
liberty as within the liberal tradition of the classical English
political philosophers, and specifically endorsed Thomas Hobbess
definition of liberty: a free-man is he that, in those things which
by his strength and wit he is able to do, is not hindered to do
what he hath the will to do.3 From this definition, Berlin thought
it followed that Law is always a fetter, even if it protects you
from being bound in chains that are heavier than those of the law,
say, some more repressive law or custom, or arbitrary despotism or
chaos.4 Put differently, any law, no matter how laudable its goal,
must be seen as a restriction of liberty. At the end of Two
Concepts of Liberty, Berlin spoke of an ideal of choosing ends
without claiming eternal validity for them and the pluralism of
values connected with this.
Electronic copy available at:
http://ssrn.com/abstract=2211608
1
He suggested that this connection might be the late fruit of our
declining late capitalist civilization. But, he said, no sceptical
conclusions followed. Principles are not less sacred because their
duration cannot be guaranteed. Indeed, the very desire for
guarantees that our values are eternal and secure in some objective
heaven is perhaps only a craving for the certainties of childhood
or the absolute values of our primitive past. To realize the
relative validity of ones convictions, said an admirable writer of
our time, and yet stand for them unflinchingly, is what
distinguishes a civilized man from a barbarian.5 To demand more
than this is perhaps a deep and incurable metaphysical need; but to
allow it to determine ones practice is a symptom of an equally
deep, and more dangerous, moral and political immaturity. 6 This
warning, written during perhaps the most fraught period of the Cold
War, might well seem of a piece with other jeremiads of the time,
notably Friedrich Hayeks famous warning in 1944 that the forces of
collectivism, whether they march under the banner of equality or
nationalism, would put Europe on the road to serfdom.7 This book,
along with others, earned Hayek the reputation of the foremost
exponent of the position that negative liberty should be given not
a complex, but an almost absolute priority. However, Hayeks
position was more nuanced than often both his followers and his
detractors acknowledge. He did not so much argue that negative
liberty should be given an absolute priority as assert a tight
connection between liberty and the rule of law, as in his famous
statement of the rule-of-law ideal in The Road to Serfdom.8 My
focus in this chapter is on the connection between law and liberty,
specifically on the connection between law and the Republican idea
of liberty as non-domination which takes the idea of negative
liberty as its foil. In my view, Republicans have done political
2
Electronic copy available at:
http://ssrn.com/abstract=2211608
philosophy an immense service both because they have moved
attention back to classical themes about liberty and because they
have so persuasively argued that the focus of this attention should
be the theme of non-domination. Moreover, Philip Pettit, the
leading philosophical spokesman for contemporary Republicanism has
made the connection between law and liberty central to his account
of legitimate political authority. However, as I will argue here,
the main result of this contribution is to uncover resources in the
liberal tradition, sometimes in surprising places, not to provide
an alternative to it. Republicans regard Hobbes as the founder of
negative-liberty liberalism and as having put forward his
understanding of liberty in order to contest and suppress that of
the Republicans of his day. However, as I shall show below, for
Hobbes the chief virtue of the rule of law is that it secures
non-domination. Moreover, his argument about why it does so might
well prove superior to that of Republicans in that it shows in what
way the rule of law is constitutive of liberty and not merely
instrumental to it, an insight that resonates in the twentieth
century in the work of such arch-liberals as Hayek.
FREEDOM AS NON-DOMINATION
In Republicanism,9 his philosophical reconstruction of the
Republican tradition, Philip Pettit argues for a conception of
freedom as non-domination, which he regards as different from both
the negative and positive concepts of liberty discussed by Berlin.
Pettit regarded Berlins articulation of the connection between
negative liberty and the conception of law as a fetter as proof
positive that for Berlin those not attached to positive liberty
allied themselves invariably with the Hobbesian tradition. That
Berlin had in this way to ally himself with Hobbes demonstrated for
Pettit that Republican liberty had not only been lost 3
to political thinkers and activists; it had even become
invisible to the historians of political thought.10 Freedom as
non-domination can be explained by using a standard example in
Republican literature. A slave with a benevolent master might be
given lots of latitude to make choices. But the slave is un-free in
the sense of freedom as non-domination because the master is
entitled to interfere arbitrarily at any moment in the slaves
decisions. In other words, domination consists in the subjection of
an individual to the whims or arbitrary will of another, whether of
the state or of some private person or group, and freedom, on its
best understanding, is freedom from dependence on the will of
another.11 Freedom as nondomination differs from negative liberty
is that it does not require that any actual or potential desire of
an individual be thwarted to register a loss of libertyall that is
required is the capacity of someone else to interfere arbitrarily
in the decisions that individual might choose to make. Pettit also
argued that whilst on a view of freedom as non-frustration freedom
is always removed by interference, freedom as non-domination not
only discerns a loss of liberty in some situations in which there
is no actual interference, but also does not suppose that
interference always removes freedom. If interference is
non-arbitrary in that it tracks the will of the individuals
interfered with, it is also non-dominating. As Pettit has
conveniently summarized this argument: In the classical republican
tradition, this thought is most commonly expressed in the refrain
that an empire of law, unlike an empire of men, is not a dominating
regime. The laws may impose taxation on all, coerce all with the
threat of punishment for disobedience, and impose penalties on
those who in fact disobey, but still, so the idea goes, such
interferences will not be arbitrary if they are framed in
accordance with 4
the will of the subjects and framed so as to protect those
subjects from domination by others.12 And he quotes with approval
Blackstone, who was soon to be challenged by Bentham: Laws, when
prudently framed, are by no means subversive but rather
introductive of liberty.13 So far we can see that the Republican
critique of liberalism has two strands to it. First, they dispute
the claim that interference is necessary for there to be a loss in
freedom. Second, they argue that actual interferences are not
sufficient to reduce freedom, because if the interferences are
non-dominating there is no loss to freedom. A third criticism
arises out of Berlins observation that negative liberty is not
incompatible with some kinds of autocracy, or at any rate with the
absence of selfgovernment. Liberty in this sense is principally
concerned with the area of control, not with its source.14 This
observation might seem to make Berlin at best suspicious of
democracy, at worst hostile (as Hobbes clearly was) to it. We would
then have evidence for the claim of critics of liberalism that
liberalism of a certain sort is an enemy of democracy or rule by
the people, since its defenders wish to establish ramparts around a
protected area of negative liberty to repel raids by a tyrannous
majority. Moreover, the spatial description of liberty adopted by
liberals like Berlinits area-- might seem to imply both that what
is important is the size of the area of choice and that the bigger
the area the better. Proponents of negative liberty are often
thought to embrace both of these implications. The task of the
state--the nightwatchman state--is confined to maintaining order
only to the extent that successful human interaction requires
background conditions of order and stability, and a democracy acts
illegitimately if it enacts laws that go beyond securing these
conditions.
5
In contrast, for Republicans it matters a great deal that
particular laws have a democratic provenance in order that citizens
not be subject to alien control.15 And Pettit argues in
Republicanism that the institutional requirements for freedom as
non-domination include at least the following: that decisions about
the collective good be made by a democratically elected parliament;
that the executives decisions about how to implement the
legislative programs enacted by that parliament be subject to
rule-of-law requirements, supervised by an independent judiciary;
and that the political rights of citizens should be guaranteed by
an entrenched bill of rights, again supervised by an independent
judiciary. Indeed, whilst Pettit wishes to avoid institutional
mechanisms that go as far as giving individuals an individual power
of veto of legislation, he also wishes to give them as many
opportunities as possible to edit government policy, in order to
ensure that it tracks their interests, on their conception of what
it is in their interests. The more democracy is contestatory, the
better.16
BERLINS BREAK WITH HOBBES?
In a recent article, Pettit has offered a more nuanced
understanding of Berlins position, and thus of liberalism in
general. He argues that Berlins own conception of freedom is not
one of negative liberty in Hobbess sense, which Pettit calls
freedom as non-frustration, but located on the path between
negative liberty and liberty as non-domination. Berlin, according
to Pettit, puts forward an idea of freedom as non-interference
which occupies an unstable place between these two other
conceptions, and which can, Pettit argues, be stabilized only if it
moves in the direction of freedom as non-domination.17
6
Berlin, according to Pettit, explicitly broke with Hobbes in his
own commentary on Two Concepts of Liberty by moving from freedom as
non-frustration to a conception of freedom as non-interference, and
moreover suggested by implication in the actual essay just this
break. For in the commentary Berlin said: The extent of a mans
negative freedom is, as it were, a function of what doors, and how
many are open to him; upon what prospects they are open and how
open they are.18 Pettit comments that the important point in this
metaphor is that the freedom of a choice turns, not just on whether
the door you push is open, but on whether all the doors are open,
including those you might have pushed on but didnt.19 In sum, it is
implausible to suppose that you can make yourself more free by
adapting your preferences to the point that you prefer only the
options in fact available to you. However, Pettit argues that
Berlins freedom as non-interference is vulnerable to the same kind
of objection Berlin ends up making against freedom as
non-frustration. For freedom as non-frustration seems to permit
that one can make oneself more free by ingratiating oneself with a
powerful person who could interfere with ones options in order to
get that person to allow one to have ones way.20 The problem with
both adaptation and ingratiation is that the freedom the individual
has remains dependent on the will of another.21 Hence, Pettit
supposes, one should be led by Berlins argument for freedom as
non-interference to adopt freedom as non-domination. Not only must
the doors be open, but there must also be no doorkeeper who can
close a door more or less without cost.22 Thus Pettit finds it
unsurprising that Berlin also accepted other commitments dear to
the Republican heartthat freedom is the area of free action we have
limited only by the law, an area that is artificially carved out by
the state, and that the freedom we have should be available equally
to each, and should be as large as possible consistently with the
7
existence of organized society.23 But Pettit suggests that one
reason why Berlin did not proceed on the logical path from freedom
as noninterference to freedom as non-domination is that he feared
that such progress entailed embracing a doctrine of positive
liberty.24 Berlins view in this regard is that doctrines of
positive liberty have an inherent logic that leads to the
conclusion that incursions into the protected space of negative
liberty do not amount to any loss of liberty because they are
justified by what the individual would rationally prefer even if as
a matter of fact the actual or empirical individual would not. It
is Berlins concern about this logic that led him to give priority
to negative liberty. That is, whilst Berlin recognized both that
some valuable goods are pursued in the name of positive liberty,
notably equality, and that their pursuit might necessarily and
justifiably limit negative liberty, he wanted to emphasize that in
calculating whether the gain in the good is worth the cost to
negative liberty, the cost must always be openly acknowledged and
given its due weight. The problem for Berlin, then, with
conceptions of positive liberty is not per se the fact that
negative liberty is limited. Rather, the problem is that the fact
is not appropriately acknowledged, and hence not properly weighed,
because conceptions of positive liberty suppose that there is no
cost to liberty at all. But it follows, as Pettit rightly sees,
that Berlin would have been uncomfortable with the thought that the
ideal of freedom as nondomination is the logical terminus of his
argument against freedom as non-frustration. Moreover, I think it
is clear that Berlin would have worried that freedom as
nondomination perpetrates the same magical transformation or
sleight of hand in denying the loss to liberty as do conceptions of
positive liberty.25 Here we should note that Berlin also argued
that such sleights could be perpetrated with conceptions of
negative liberty, despite
8
the fact that they do not easily lend themselves, as do positive
conceptions, to splitting the personality into the empirical and
the real or rational or ideal self.26 The obvious response open to
Pettit is to say that it is just a liberal hangover to keep
thinking that the law infringes your freedom simply because it
interferes with you. However, Pettit is as averse as Berlin to the
paternalism in positive liberty claims to the effect that an
individual is not the best judge of his or her own interests. On
his view, there is no external standard of excellence to which an
individuals choices must conform in order for that individual to be
truly free or rational. All that matters is that the individual is
able to exercise free, that is, non-dominated choices, whatever he
or she decides to choose. Indeed, in his most recent work Pettit
says that the most harmful form of domination is actual
interference with actual desires, thus showing that something like
the idea of liberty as non-frustration has a central place in his
political thought.27 These ideas undermine his argument that
liberalism goes wrong in supposing that there is something
inherently wrong with actual interferences even when that wrongness
is compensated for by the gain in some other good. Hence, I tend to
agree with arguments made by some critics of Republicanism who see
it as in no major respect distinctive from central strands within
the liberal tradition.28 I will add to this criticism the reason
that at the low altitude29 of institutional design, by which I mean
the design of institutions to ensure that government is according
to law or in accordance with the rule of law, Republicans, with
some exceptions, find it difficult to distinguish themselves from
the liberal tradition. However, the exceptions are significant
because they embrace a conception of the rule of law that fits well
the kind of negative liberty version of liberalism to which
ironically they are most opposed. In addition, this embrace creates
a problem that is not confined to this subgroup. It displays a
tension in contemporary Republicanism itself that explains both
some 9
Republicans very partial reading of Hobbes and, more generally,
a profound ambivalence in Republican theory about the nature of
political and legal authority. I turn for this argument to the
figure Republicans believe to be the cause of all the negative
liberty trouble, Thomas Hobbes. As I will show, Hobbes is as
concerned as Blackstone with the prudent framing of laws so as to
make them by no means subversive but rather introductive of
liberty, or as Pettit himself puts it, with the Republican thought
that to be a free person just was to be a citizen incorporated in
the matrix of protection for certain basic choices that is afforded
to eachin theory, afforded equally to eachby the rule of law.30
HOBBES ON LAW AND LIBERTY
Pettit has argued more elaborately in other work along with
Quentin Skinner that freedom as non-domination is a chronologically
prior conception of freedom to the other two. For, as Republicans
understand things, freedom as non-frustration was cooked up by
Hobbes in his bid to subvert the Republicans of his own day and
their conception that one can only be free in a free society, often
equated with a political order of parliamentary democracy.31 Hobbes
serves as the main foil for Republicans because they think that his
conception of liberty in Leviathan is designed to debunk the
Republican ideal of a freeman. They emphasize what they take to be
Hobbess view that those who are subject to the arbitrary will of an
all-powerful sovereign are by definition free. As long as legal
subjects live under his rule and are not slaves or in chains, they
must take themselves to have consented to his rule and to regard
all of his laws as just. It is rational for them to do so, because
however obnoxious32 subjection to the will of such a person seems,
it is better than being 10
subject to the wills of all other individuals in a state of
nature. But even if a subject fails to appreciate the rational
force of the argument for his obligation to obey the law, he will
be able to appreciate the force of the sanction with which he is
threatened should he disobey. Finally, from the perspective of
liberty, it does not matter what kind of political regime one lives
under, whether it is democratic or monarchical, because the kind of
political regime does not affect the quality of ones liberty. There
are, according to Pettit, two surprising claims built into Hobbess
definition of liberty. The first is that you are only hindered in
your choice if you actually prefer the obstructed option. The
second is that to be a freeman is to escape all such external
hindrance in the options you prefer to take.33 The two claims are
surprising because the first makes it too easy to be free, whereas
the second makes it too hard. Pettit suggests that the second
surprise, that one cannot be a freeman since we cannot escape all
hindrances, contains Hobbess fundamental challenge to the
Republican argument that to be a free person just was to be a
citizen incorporated in the matrix of protection for certain basic
choices that is afforded to eachin theory, afforded equally to
eachby the rule of law.34 The challenge is fundamental because it
strikes at the very heart of Republicanism by redefining the
Republican idea of the freeman. According to Pettit, as we have
seen, the substance of the challenge resides in defining freedom in
such a way that it becomes impossible. However, it is important to
notice that the challenge works better on the basis of the first
kind of claim, one that makes it too easy to be free, with the
upshot that one is always free, as long as one is living under the
protection of a sovereign, indeed, that one enjoys the same freedom
whether ones sovereign is a dictator, or the ruler elected by the
citizens of a self-governing republic.
11
This way of putting the challenge is closer in fact to what
Hobbes says in the chapter in which he gives the definition of
freedom, if not to the definition itself, and it is the way that
Skinner supposes it mainly works.35 According to Skinner, Hobbes
has a two-pronged argument. In one sense of liberty, the sense of
the definition, legal subjects are always free because civil laws
do not restrict liberty as they do not amount to physical
impediments to action. In another sense, legal subjects are always
free because their civic liberty is simply the freedom to act as
they will in so far as the civil law does not prescribe a course of
action, and so freedom under the law is always the same
everywhere.36 That two such eminent exponents of contemporary
Republicanism, who happen also to be leading Hobbes scholars, agree
that Hobbess discussion of liberty is aimed at the destruction of
freedom as non-domination, yet seem to disagree about Hobbess
strategy, raises questions about how the strategy could have had
the success they attribute to it. I will argue below that the
answer is that their disagreement about Hobbess strategy indicates
that they have both likely mistaken it. But the issue I wish to
focus on goes beyond a claim that Pettit and Skinner have attacked
something of a straw man in choosing Hobbes as their target.
Rather, it is that their conception of freedom as non-domination is
more a reconstruction of Hobbess conception of liberty as liberty
under an order of public laws than it is a retrieval of a
conception that Hobbes sought to subvert. In addition, to the
extent that their attack on Hobbes hits home in that their
conception of freedom of nondomination differs from his, Hobbess
conception, or so I will argue, might have some advantages. This
argument has to contend with the fact Hobbes does put in place all
of the major ingredients of the negative liberty view of freedom
and law that Republicans contest. For example, in The Elements of
Law he says that the sovereign authority, when legislating 12
prohibitions on what was lawful to individuals in the state of
nature, must ensure that well meaning men may not fall into the
danger of laws, as into snares, before they be aware.37 That is, in
this passage, Hobbes does suggest both the image of law as a fetter
and that the image fits with his understanding of freedom as the
natural liberty of the state of nature. Moreover, since he also
says that the sovereign should not put in place a restraint that
goes beyond what is necessary for the good of the commonwealth,38
he might be taken to be hinting at the kind of libertarian view
that says that the less there is by way of restraint the better.
The sovereign should seek to put in place only what is necessary to
maintain order and stability. (It could, of course, be no more than
a hint since for Hobbes the extent of liberty is entirely up to the
sovereign to decide.) But, as we have seen, Pettit and Skinner seem
undecided about how Hobbess strategy works. They divide over
whether Hobbes makes it too easy be to free or too difficult, with
Skinner supporting the first option, and Pettit the second. Whilst
Skinner would say that for Hobbes legal subjects are equally free,
no matter the nature of their political regime, Pettit must say
that they are equally unfree--that subjects are in substance the
slaves of their sovereigns.39 The political effect of the
strategies is the same, as they both issue in the conclusion that
the nature of ones political regime is irrelevant to the question
of how free one is. However, as I suggested, Hobbess challenge
works better on the basis of the claim that makes it too easy to be
free. For the upshot of that claim is that one is always free, as
long as one is living under the protection of a sovereign, indeed,
that one enjoys the same freedom whether ones sovereign is a
dictator, or the ruler elected by the citizens of a self-governing
republic. In addition, this claim fits better with the main
elements of Hobbess picture of politics, as just painted, in which
law is conceived as a fetter on natural liberty. Subjects 13
consent to the way in which the sovereign fetters their natural
liberty with his laws because that gives them the security to enjoy
what remains of our natural liberty; that is, subjects may do as
they like insofar as the law is silent as to their obligations. And
this seems to be exactly Hobbess view in The Elements of Law.
However, even though Skinners view is to be preferred, it does
encounter a major problem, as does the claim that Hobbes sees law
as a fetter. For Hobbes does not think that a civil law is capable
of fettering natural liberty since he holds that fear of coercion
does not count as an obstacle to the will. As Pettit and Skinner
have discussed elsewhere, Hobbes did not consider the sovereigns
laws to be constraints on liberty since one is physically free to
disobey the laws. Fear, in this case fear of punishment, is for
Hobbes not a fetter or constraint on liberty.40 Moreover, when we
fill in the picture of Hobbess political and legal theory with
finer brush strokes, with Leviathan as our palette, the image of
law becomes more nuanced. And with nuance our understanding of how
Hobbes conceived of the arbitrary will of the sovereign has to
become more complicated, with the result that, paradoxically, the
sovereigns will has to be exercised in ways that ensure that, in
Pettits words, to be a free person [is] to be a citizen
incorporated in the matrix of protection for certain basic choices
that is afforded to eachin theory, afforded equally to eachby the
rule of law.41 The first point to note is that Hobbes is crystal
clear that the radical insecurity of the state of nature resides
more in the uncontrolled capacity of others to interfere than in
actual acts of interference. In his discussion of the state of
nature in Leviathan, he says: [D]uring the time men live without a
common Power to keep them all in awe, they are in the same
condition which is called Warre: and such a warre, as is of every
man, against every man. For WARRE, consisteth not in Battell onely,
or the act of 14
fighting; but in a tract of time, wherein the Will to contend by
Battell is sufficiently known: and therefore the notion of Time, is
to be considered in the nature of Warre, as it is in the nature of
Weather. For as the nature of Foule weather, lyeth not in a showre
or two of rain; but in an inclination of many dayes together: So
the nature of War, consisteth not in actual fighting, but in the
known disposition thereto, during all the time there is no
assurance to the contrary. All other time is PEACE.42 Commentators
on this famous passage usually focus on the point about warre, as
is of every man, against every man. They thus fail to notice that
for Hobbes it is not the interferences that actual fighting amounts
to that typifies the state of nature, but the lack of assurance
that someone who has the capacity to interfere will not interfere.
Thus the point of establishing civil society is to provide that
assurance, and the social contract should be entered into in order
to achieve that end. The second point to note is that the sovereign
who is created by the social contract is an artificial person who
is under a duty to serve the interests of those subject to him, and
he does so by ruling through law. In chapter 30 of Leviathan,
Hobbes discusses the Office of the Soveraign, which he says
consisteth in the end, for which he was entrusted with the
Soveraign Power, namely, the procuration of the safety of the
people. The sovereign is obliged by the law of nature to procure
this end and Hobbes adds that safety is more than a bare
Preservation, since it includes all other Contentments of life,
which every man by lawful Industry shall acquire to himselfe.43 He
continues: To the care of the Soveraign, belongeth the making of
Good Lawes. But what is a good Law? By a Good Law, I mean not a
Just Law: for no Law can be Unjust. A good Law is that, which is
Needfull, for the Good of the People, and withall Perspicuous.44
Hobbes then offers the following account of the function of law in
civil society: 15
For the use of Lawes, (which are but Rules Authorised) is not to
bind the People from all Voluntary actions; but to direct and keep
them in such a motion, as not to hurt themselves by their own
impetuous desires, rashnesse, or indiscretion; as Hedges are set,
not to stop Travellers, but to keep them in the way. And therefore
a Law that is not Needfull, having not the true End of a Law, is
not Good. Unnecessary Lawes are not good Lawes; but trapps for Mony
45 The last line expresses a concern that bad laws are vulnerable
to capture by powerful private elites. The first line gives us an
image of law as a hedge that is more nuanced than Berlins image of
law as a fetter, or Hobbess in The Elements of the Law as a
restraint, that is, more nuanced than the image of law-as-fetter
that both Berlin and Pettit suppose follows from Hobbess definition
of the free man. For the image of law-as-hedge suggests that hedges
are public goods that neither impede our motion nor do they impose
an ultimate on our motion. Rather, hedges make mobility possible.
The image thus presupposes that there can be such a thing as a
non-instrumental law: a law that facilitates individual action. It
does not do so primarily because it clearly sets out in advance any
prohibition so that one can take the prohibition into account when
planning. Indeed, it need not prohibit anything. Rather, what it
does, whether it prohibits or not, is put in place the conditions
that make it possible for individuals to follow through on their
own plans.46 The best modern articulation of the idea of
non-instrumentality in law is in Michael Oakeshotts somewhat
neglected essay, The Rule of Law, where Oakeshott says that the
expression the rule of law, taken precisely, stands for a mode of
moral association exclusively in terms of the recognition of the
authority of known, noninstrumental rules (that is, laws) which
impose obligations to subscribe to adverbial conditions in the
16
performance of the self-chosen actions of all who fall within
their jurisdiction.47 By this, Oakeshott seems to mean the
conditions that attach to performing actions that are made possible
by, though not dictated by, the law. Thus: If you wish to drive
from Budapest to Vienna, you must do so on the right hand side of
the road does not tell you to make that journey, but makes that
journey possible in a way that is otherwise be unavailable to you;
If you wish to buy that house you must follow these formalities in
order for the contract to be valid, does not tell you to buy the
house, though it makes it possible for you to do so; If you wish to
interact with other individuals, do not murder, commit other acts
of violence, steal, does not tell you whether and otherwise how to
interact with them, but it does make possible civilized interaction
between you and others. In all of these cases, the subject is at
liberty to decide whether to drive, etc. Notice that none of these
liberties, whether to bring some state of affairs into being, or
simply to get something done under conditions of security, is even
possible in a state of nature. Hence, the image of law-as-hedge
captures the way in which law of a particular form can be
introductive of civic libertyliberty under an order of public
law--that otherwise could not be had. Moreover, the choices that
this kind of law makes possible create the framework for particular
exercises of choice, where to travel to, what legal relationships
to enter into, how to interact with others. In sum, this kind of
law provides a resource that makes choice in general possible and
so also particular exercises of choice. It can thus be said to be
constitutive of the liberty to choose for oneself, rather than the
instrument of particular goals. It is important to see how civic
liberty differs from other conceptions of liberty. First, it is not
negative in nature since it is not liberty from the law but liberty
introduced by law. It 17
is true that law in making possible the choice could also be
said to circumscribe it. But it would be odd to think of this kind
of law as a fetter on the choice, just as it would be odd to think
of the rules that, for example, constitute the game of chess as
fettering the choices that one makes when one plays chess. Second,
civic liberty is not positive in nature. The liberty the law makes
possible for legal subjects provides them with a resource that they
may use or not, at their discretion. Even in the case of the
criminal law which enacts proscriptions on certain behaviours that
are harmful to others, thus dictating important terms of social
interaction, the terms are best understood as making it possible
for individuals to choose on their own terms how to interact with
others.48 Finally, civic liberty is not a subset of the natural
liberty that according to Hobbes individuals enjoy in the state of
nature to do whatever it takes to ensure their own survival. That
is, Republicans and others take the liberty Hobbes ascribes to the
subject in civil society to be the residue of the liberty enjoyed
in the state of nature that is left to the individual in so far as
the civil laws do not create obligations for subjects. But that
cannot be right. Natural liberty is equally and fully enjoyed by
subjects in civil society and so they are always equally free in
its sense to break the civil law. Their disobedience might result
in their liberty being physically restrained, but at that point it
is not the law that constrains them but the prison walls or their
physical bonds. One has to appreciate these differences to make
exegetical sense of Hobbess extensive discussion of liberty in
chapter 21 of Leviathan.49 Whilst Hobbes starts the chapter with
the definition of the freeman quoted at the outset of this chapter,
he goes on to describe what he clearly regards as a distinct kind
of liberty, the Liberty of Subjects, which is a liberty in relation
to [the] Bonds that are made up of the Artificiall Chains or civil
laws, 18
which are issued by the artificial person of the sovereign.50
With perhaps the Republicans of his day in his sights, Hobbes
comments caustically that it would be absurd to clamour for liberty
from such laws, if liberty is taken as corporeal liberty, since
that liberty is one that people manifestly enjoy. That is, to
repeat, civil law does not physically obstruct. And it is equally
absurd for subjects to demand the liberty to be their own masters,
to demand an exemption from the civil laws,51 because that would be
to demand a return to the condition of masterless men, that is, to
the perpetuall war52 of the state of nature. It is true that in
these passages in chapter 21 there are clear echoes of the
law-asfetter image that dominates The Elements of the Law, most
notably, when Hobbes says the Liberty of a Subject, lyeth therefore
only in those things, which, in regulating their actions, the
Soveraign hath praetermitted: such as in the Liberty to buy and,
and sell, and otherwise contract with one another 53 He also says
later in the chapter that other Lyberties depend on the Silence of
the Law since in cases where the Soveraign has prescribed no rule,
there the Subject hath the Liberty to do, or forebeare, according
to his own discretion. And therefore such Liberty is in some places
more, and in some lesse; and in some times more, in other times
lesse, according as they that have the Sovereaignty shall think
most convenient.54 But even here the space is what we might think
of as discretionary libertythe facility of judgment or choice under
conditions of security and stability that is made possible by the
law. It is not, to repeat, natural liberty since that consists in a
space limited only by physically disabling factors. In sum, for
Hobbes the point of individuals consenting to live under the
authority of an all-powerful sovereign is to procure a condition of
civic liberty. Civic liberty is a different kind of liberty from
the natural liberty of individuals in a state of nature to do
anything they consider appropriate to ensure their survival. It
does not consist in a freedom from physical 19
obstacles, first, because the bonds of the law are not such
obstacles.55 Second, it is not even accurate to understand civic
liberty as a liberty from the law, even though Hobbes himself is
tempted at times to say just this, because it is a liberty which we
would not and could not have were it not for the fact that the law
introduced it. Rather, law is necessary to create civic liberty;
and as long as that is what the laws of a legal order do, they will
create the same quality of liberty even though the space they make
for the exercise of discretion by subjects will vary greatly across
both time and space. Notice that what I called above discretionary
liberty, the judgment by subjects about whether and how to use
their civic liberty, is a direct descendant of the radical or
unfettered discretion of individuals in the state of nature to
decide on what actions will best ensure their survival. It is a
descendant because Hobbes assumes the equality of all individuals
in the state of nature to decide on what is best for themselves
and, as I will sketch below, carries forward this assumption into
the structure of civil society. But there are salient differences
because subjects in civil society unlike individuals in the state
of nature do not have to exercise their judgment about how to
survive, since survival is no longer at stake. Instead, their
judgments will be about how to make their lives go as well as they
can on their own terms, within the framework of civic liberties
established by the law. As David Gauthier has argued, the
transition from the state of nature to civil society enables
subjects to replace what we can think of as the natural reasoning
appropriate to the state of nature with a form of public reasoning
appropriate to life in a civilized society.56 The quality of civic
liberty is thus more important for Hobbes than the quantity; and
that a universal quality is secured, even though quantity is
dictated by sovereign will, is important to seeing why for Hobbes,
as I am about to explain, sovereign rule is not arbitrary in the
way his critics allege. This explanation serves to show why the
transition to civil 20
society makes possible a kind of public reason. Moreover, in
seeing this, we can also appreciate why Hobbess legal theory,
contrary to his reputation as someone who holds an austerely
positivist view of law as the commands backed by sanctions of a
legally unlimited sovereign, is not part of the legal positivist
tradition.
THE QUALITY OF CIVIC LIBERTY
The first respect in which sovereign rule is non-arbitrary (or
non-dominating) from the perspective of subjects is that for the
sovereign as an artificial person to will he has to will through
law. The conditions of civic liberty have to be created by, and can
only be changed by, public, prospective legislation, which in words
taken from Friedrich Hayeks well-known formulation of the ideal of
the rule of law, makes it possible to foresee with fair certainty
how the authority will use its coercive powers in given
circumstances.57 By itself, this requirement goes a long way to
ensuring non-arbitrary rule in one of two important senses that of
rule by law. Neither the sovereign nor his agents are entitled to
use coercion against legal subjects unless they can show a prior
legal warrant. However, whilst this criterion for a valid exercise
of sovereign authority is necessary for the rule of law to be
non-arbitrary in that it requires prior legal authorization for
valid exercise of such power, it cannot be sufficient. For example,
the officials might be authorized to act arbitrarily, either in
that they are given what lawyers call an unfettered discretion, a
discretion to do as they please, or they are explicitly authorized
to perform substantively arbitrary acts, that is, acts that
perpetuate or create relationships of domination. Thus one eminent
philosopher of law and legal positivist, Joseph Raz, has suggested
that the rule of law is essentially a negative value: the law
inevitably creates a danger of 21
arbitrary power, so that the rule of law is designed to minimize
the danger created by the law itself.58 Raz, as I will argue below,
is wrong to suppose that the rule of law is essentially a negative
value, one which does not cause good except through avoiding evil
and the evil which is avoided is evil which could only have been
caused by the law itself.59 However, his suggestion about negative
value does go some way to capturing one aspect of the rule of law.
As I will now go on to show, the other, more positive or
constitutive aspect is well captured by Hobbes. And once that
aspect is revealed, we have not only a better understanding of the
negative aspect, but also, and despite the fact that Hobbes is the
main foil for Republicans, a grip on the way in which law connects
to a conception of freedom as non-domination in a constitutive
fashion. It does so by meeting a second criterion, that of
legality. Razs suggestion trades on the idea that the highly
centralized power of modern governments, which includes the
monopoly on the legitimate use of coercion, is a kind of power that
is legally constituted. Hence, law creates the kind of governmental
power that preoccupies political philosophers who take as their
principal concern the limits on state action, and who find in the
rule of law one way of enforcing such limits. But whilst it is true
that law does organize power in a way that makes it an effective
instrument of those who wield power, it does not merely organize
power, but also endows it with authority. Those who wield legal
power claim more than the de facto ability to force compliance with
their directives; they claim, de iure, the right to require
compliance from those subject to their authority. It is for this
reason that Hobbes in his definition of law does not refer to the
coercive power of the sovereign, but to his right to obedience: law
is the command not of any man to any man: but only of him, whose
Command is addressed to one formerly obliged to obey him.60 Of
course, the claim to authority, and moreover to legitimate
authority, might serve to make legal power seem even more
dangerous, which is 22
why legal positivists such as Raz are anxious to debunk any
claim that law has some intrinsic legitimacy. But that those who
wield legal power claim authority for their directives entails that
they lose authority, and with that loss fail to exercise power,
when they do not act according to law. However, as suggested above,
the rule-by-law sense of non-arbitrary rule permits rules that
introduce what we can think of as substantive arbitrariness into a
legal order, for example, rules that require that one group of
legal subjects be treated as less than equal. Hence, the
requirement that the conditions of civic liberty be set by publicly
enacted rules of sufficiently determinate content is necessary but
not sufficient for there to be freedom under a public order of law.
The proviso I added to the claim that for Hobbes the bonds of the
law create civil liberty--as long as that is what the laws do--is
important because the history of legal orders is littered with
examples of laws that far from constituting civic liberty have
either undermined or destroyed it. Most Hobbes scholars do not
think that Hobbes made room for such a proviso in his legal theory.
They take as evidence, amongst other things, his remarks to the
effect that the sovereign is accountable only to God for his
infractions of the laws of nature.61 But these scholars, including
the Republicans amongst them, neglect Hobbess extensive discussion
in chapter 26 of Leviathan of the interaction within civil society
of the laws of nature with enacted law, an interaction over which
subordinate judges preside. Moreover, this interaction shows why
for Hobbes to use law to undermine civic liberty is an abuse of
legal form that will not only create tensions internal to legal
order, but bring into question the relationship of authority
between sovereign and subject that has to be maintained for a legal
order to persist.
23
As I have argued elsewhere,62 the sovereign as ultimate judge is
constrained by the laws of nature, not because he owes duties to
his subjects, and despite the fact that Hobbes rejects arguments
for the separation of powers. Rather, the constraints come about
because of the duty his subordinate judges owe to him to interpret
his enacted law in the light of their understanding of the laws of
nature. The sovereign as first and ultimate judge can either
preclude or override such interpretations. However, that does not
make the constraints cease to be such; it simply makes them
overridable by very explicit words. And if the sovereign should
choose to override the constraints, he risks bringing about the
limit case in which the subject is entitled to treat him as no
longer maintaining the essential relationship between protection
and obedience, that is, as no longer a sovereign. Oakeshott,
perhaps alone among the major Hobbes scholars of the last century,
was sensitive to this aspect of Hobbess legal theory. Not only is
Hobbes the primary inspiration for his own essay on the rule of
law, but he also saw that in Hobbess legal theory the laws of
nature give us the content of legal right, or ius, to which enacted
law, or lex, must aspire to conform.63 The proviso is entailed in
that kind of legal theory. It amounts to what I call the legality
proviso--that it is not sufficient for an enactment to comply with
public criteria of validity to count as law. The content of the
enacted law must also be interpretable by judges in light of the
laws of nature. And when the legality proviso is met, one has in
place more than rule by law, government by lex, since one has the
rule of law, government in accordance with principles of legality
or ius. Only then, only that is, when there is a public order of
law, meaning of particular laws that are interpretable in way that
meets the legality proviso, will the laws that make up the order be
introductive of civic liberty. Such interpretations condition the
content of the laws as they apply to particular subjects, and thus
affect the relationship between the subject 24
and the sovereign on the basis of the laws of nature, the set of
which Hobbes calls the true and onely Morall Philosophy.64 Among
the most important of Hobbess laws for understanding the content of
ius are the laws that require impartial and equitable judgment and
proportionality in punishment. In spirit, they overlap considerably
with the list of the desiderata of the rule of law that Lon L.
Fuller argued made up an inner morality of law, with the major
difference being that for Hobbes a principle of equality is central
to his account of the laws of nature.65 The legality proviso thus
indicates that there is more to the rule of law than Hayek conveyed
in his famous formulation in The Road to Serfdom: Nothing
distinguishes more clearly conditions in a free country from those
in a country under arbitrary government than the observance in the
former of the great principle known as the Rule of Law. Stripped of
all of its technicalities, this means that government in all of its
actions is bound by rules fixed and announced beforehandrules which
make it possible to foresee with fair certainty how the authority
will use its coercive powers in given circumstances and to plan
ones individual affairs on the basis of this knowledge.66 For the
formulation rules which make it possible to foresee implies that
all that matters is that we have as much certainty as possible
about what obstacles the laws might place in our way, so that we
can avoid being surprised by them. Hence, the rule of law gets
announced with great fanfare in the first line, but is reduced to
the rule of rules with determinate content in the second.67
However, in other work Hayek referred to a nomocratic or
law-governed conception of a free society, which he contrasted with
a telocratic or purpose-governed conception.68 He credited Oakshott
with this distinction and said that the great importance 25
of the abstract order of a nomocracy is that it extends the
possibility of peaceful coexistence of men for their mutual benefit
beyond the small group whose members have concrete common purposes,
or were subject to a common superior, and that it thus made the
appearance of the Great or Open Society possible.69 And it was
clearly the nomocratic conception that Hayek had in mind when in
The Road to Serfdom he drew a distinction between the rule of law
as the rule of general and formal rules and rule by substantive
rules in terms of providing signposts and commanding people which
road to take.70 For the image of law-as-signpost seeks to convey
what Hobbes had in mind in giving us the image of law-as-hedge. The
image of law-as-fetter works best in the position of those who
think that the only important sense of liberty is liberty as
non-frustration. For them law is best conceived as no more than a
determinate constraint on negative liberty which will be legitimate
in so far as it constrains our liberty as little possible, just
enough to secure peaceful social interaction, and in so far as the
constraint is as determinate as possible, so that we can plan
around it. The loss in liberty will be justifiable or not depending
on whether the value of the goal of which the law is the instrument
compensates sufficiently for the disvalue in the loss of liberty.
Pettit and Berlin himself71 considered that Bentham was a central
figure in this tradition of political thought. They are right to do
so. But it is important to see that Bentham, and those who follow
after him in his legal positivist understanding of law, either
reject the legality proviso or have considerable difficulty
integrating it into their theories. Bentham rejected the legality
proviso for at least two reasons. First, legal order has in his
view no intrinsic legitimacy. Law is no more than a mechanism for
transmitting the commands of the powerful to legal subjects and
whether it is legitimate or not will depend both on its content and
its provenance. Because Bentham thought that good content, i.e. a
26
correct calculation of overall utility, was most likely to be
achieved by having a democratically elected assembly decide on the
content of the law, provenance and content were in his view linked.
Second, and relatedly, because the legality proviso requires judges
to seek to interpret enacted law in light of the principles that
make up the legality proviso, Bentham thought that it gives to an
unelected and conservative elite a resource to second guess those
with legitimate authority to decide on overall utility. It is
interesting that this kind of legal positivism gets support from
some contemporary Republicans.72 For them, it is a sufficient
condition for law to be legitimate, that is, non-dominating, that
it has been enacted by a properly constituted democratic
legislature. They thus reject arguments that require that the
content of the law be determined in accordance with any
morally-infused interpretative test such as the legality proviso,
as interpreted by a staff of judges independent of the executive,
because they claim that such arguments give a license to the judges
to second guess democracy. The point of law, on their view, is to
transmit as effectively as possible a content that has been
determined outside of the legal order, which means that its
legitimacy is wholly derivative, that is, dependent on whether or
not it has the right content. Put differently, and somewhat
paradoxically, law has to be content-independentto have a content
that can be determined independently of any moral considerations
and arguments--in order that it can do the job of transmitting the
content that moral tests without the legal order have determined
should be so transmitted.73 All the moral action, then, takes place
before the enactment of the statute and legitimacy is
content-dependent, dependent on the right content having been
determined in the first step. For Republicans of the sort just
described, right content is whatever is determined to be right by
majority legislative vote.. For other Republicans, notably Pettit,
not only does the 27
law have to have the right democratic provenance, but it must
also be subject to judicial review for compliance with the rule of
law and with an entrenched bill of rights. And, as I already
indicated, Pettit does not stop there. Because for him ultimately a
law is nondominating only if it tracks the subjects interests on
his or her own conception of interest, as many opportunities must
be given to the subject as possible to edit or contest the law.74
As one political theorist has pointed out, the profusion of checks
and balances in this picture of the good polity amounts to a
fulsome embrace of institutional sclerosis that will likely
paralyze any attempt to enact and implement the kinds of
legislative programs necessary to do away with social domination.75
Hobbes, we should note, said that his main aim in Leviathan was to
find a way to pass unwounded between those that contend on one side
for too great Liberty, and on the other side for too much
Authority.76 In his view, the Republicans of his day embraced the
first horn of this dilemma, whilst the seditious clergy--the
proponents of divine right-embraced the second. We can already see
the temptation in some Republicans to follow their predecessors in
Pettits idea of endless contestation--that one may deny a law
authority, no matter the basis for its claim to authority, simply
because one disagrees with its content. In others, however, there
is a tendency to grant a right to democratic legislatures
comparable to the divine right that was once accorded to kings.77
Republicans thus divide into two camps, an anarchist one and a
parliamentary absolutist one, each embracing one horn of the
dilemma Hobbes believed a viable political theory in an age of
pluralism has to resolve. Hobbess resolution of this dilemma
requires that there be a moment of decision or institutional
closure on any contested political issue. The moment comes when, at
least for the time being, subjects accept not only that the content
of the law is as those charged with ultimate authority to interpret
it say, but also that they are under an obligation to obey it
28
despite their strong disagreement with its content. As I have
argued, that the law must be interpretable in light of the laws of
nature is a significant constraint on the content of the law, at
the same time as it is a basis for subjects to accept that they are
under an obligation to obey the law. Moreover, that basis is such
because the constraint serves an ideal of nondomination. Hobbess
ideal is, however, for good reason less exacting an ideal than that
adopted by the anarchistic strain within Republicanism. As Lars
Vinx has pointed out, it is plausible to infer from Hobbess
arguments in Leviathan a fear that the pursuit of a thick and
demanding ideal of non-domination that requires endless
contestability would turn out to be incompatible with the existence
of well-functioning political institutions capable of securing
public order and social co-ordination.78 It is also, however, more
exacting, at least at the low altitude of institutional
requirements, than that presupposed by the parliamentary absolutist
camp in Republicanism, because the legality proviso requires that
any enactment be interpretable in light of the laws of nature. If
the enactment is so interpretable, then the subject has reason to
comply with it, even if he or she disagrees with its content. That
reason is based on the fact that the law is legal, that is, it
meets the requirements of the rule of law or legality. That the law
has a democratic provenance is another, perhaps stronger reason for
compliance. There is at the least an elective affinity between the
two kinds of reasons, democracy and the rule of law, since they
both take the point of legal order to be to construct a society in
which individuals may interact with each other in conditions of
security and stability that enable each to live in accordance with
his or her sense of the good life. It might even be correct to
argue that freedom under a public order of law cannot be secured in
the absence of democratic controls on government.79 For example, an
order that 29
was put in place by a benevolent despot would remain in place
only if the despot so willed, thus making the freedom enjoyed by
legal subjects under that order dependent on the despots continued
benevolence, which is anathema to a conception of freedom as
nondomination. In addition, one can argue that questions about the
quantity of civic liberty are best worked out in the forum of
democratic politics, since that will increase the chances that the
answers will track the interests of subjects as they themselves
understand their interests. Indeed, Hobbess own hostility to
democracy has less to do with the spirit of his account of
sovereignty than with his historically contingent worry that the
political factionalism of civil war can be ignited by the open
political contests of democratic politics. If one abstracts from
that worry, democracy will seem, as it does to many Republicans, as
the natural companion to the requirement that the sovereign rule by
law and in accordance with the rule of law. After all, the major
political advance of Leviathan resides in the idea that the
authority of the sovereign is the authority not of a natural
person, but of the artificial person of the state who speaks to
subjects through law, that is, through general, public laws that
are interpretable in light of the laws of nature. The great bargain
that individuals make in the state of nature is to swop their
subjection to the potential for uncontrolled interference by any
other individual for the controls inherent in legal government. At
times Hobbes talks as though the swop is beneficial only because it
is better to be subject to the whims of a single natural individual
than to the whims of everyone, notably, when he says that one might
object that the Condition of Subjects is very miserable; as being
obnoxious to the lusts, and other irregular passions of him, or
them that have so unlimited a Power in their hands. 80 However, his
answer to the objection is an implicit rejection of the terms in
which it is stated. It is that no matter the character of a
government, democratic or monarchical, subjection to it has to be
better than the awfulness of the state 30
of naturethe dissolute condition of masterlesse mennot because
there is now one natural master, but because the master is a
sovereign who governs by law with the coercive power to enforce the
law centralized in his hands.81 Hobbess inquiry into the character
of rule by the single artificial person of the state goes a long
way to showing why such rule should be considered legitimate by
those subject to it because it will be non-dominating. He is not
disabled in doing so by the fact that on his view of freedom only
actual interferences are normatively relevant. That freedom as
nonfrustration is so precarious in the state of nature provides
everyone with a reason to exit that state, but the cure for that
precariousness is not premised on freedom as non-frustration.
Rather, it is premised on securing equal civic liberty under an
order of public law. Moreover, whilst the particular laws that make
up this order will be fetters on negative liberty in Berlins sense,
they are incapable of fettering natural liberty in Hobbess sense,
and can at most make it more probable through their sanctioning
mechanisms that subjects who do not understand why it is rational
to obey the law because of its intrinsic legitimacy will choose to
obey out of fear. However, subjects who understand why it is
rational to obey the law because of its intrinsic legitimacy have
it because they grasp that a public order of law is introductive of
civic liberty, and so cannot be counted as an interference with
that kind of liberty. As Hobbes says, to clamour for freedom from
such laws is to make at least three mistakes: first, a category
mistake in that one is demanding natural liberty when in issue is
civic liberty; second, a logical mistake since one in fact has
natural liberty in a civil society; third, a moral and rational
mistake, since the state one is clamouring for is one where the
only liberty one has is natural liberty, but that is a not worth
having because one is in the dissolute condition of masterlesse
men.82 31
The contours of such liberty, what it takes to ensure the
quality of civic liberty, can, then, be worked out apart from the
design of the democratic political order.83 And it is best that the
two tasks are not muddled, or worse collapsed as happens in the
case of the absolutist strain within Republicanism or the most
libertarian strains within liberalism. In the former case, the rule
of law is reduced to the rule by law of the political elite or
alliance of elites that happens to command a majority in the
legislature. In the latter, democracy is reduced to the rule of
law, since democracy is advocated only to the extent that it seems
the best, more accurately, the least bad, means of ensuring that
the rule of law is secured, where by the rule of law is meant not
only that a certain quality of liberty is secured, but the greatest
possible quantity consistent with maintaining order and
stability.
CONCLUSION
We saw at the outset that Berlin set great store on the ability
to realize the relative validity of ones convictions at the same
time as being ready to stand for them unflinchingly because that
ability is what distinguishes a civilized man from a barbarian.84
This somewhat paradoxical claim might seem some fifty years on
uncomfortably paired with a distinction between civilized men and
barbarians.85 But the tragedy of liberty, the topic of this
collection of essays, has plausibly to do with our sense these days
of the fragility and contingency of a particular form of society
that makes possible what Hobbes thought of as civic liberty. As I
have argued, the conception of liberty at stake here is not a
residue of the liberty of the state of nature, but is freedom under
a public order of law. It is in large part produced by the
particular enactments of the sovereign, but these have to be part
of the public order, a 32
requirement that imposes the discipline of legality on the
enactments. In such an order, the individual has to understand that
his or her own convictions are relative in the sense that the law
supplies what Hobbes called the publique Conscience, by which [the
subject] hath already undertaken to be guided.86 That is,
individuals must subordinate or relativize their sense of right and
wrong to the public sense, at least in so far as they conform in
their actions to that sense, even whilst they maintain their own
convictions. But that subordination is subject to the limit case,
in which the natural individuals who people the office of
sovereignty step beyond the limits of their authority and those
subject to the law are entitled to conclude that the law has lost
its authority because it is no longer doing the job of constituting
their equal freedom. The Republican revival is of immense value in
an understanding of these issues in that it focuses our attention
on the major mark of a civilized society as one in which individual
interaction is untainted by domination. But, or so I have argued,
they should see that perhaps their most helpful ally in trying to
understand the design of such a society is the figure they usually
take as their enemy number one, Thomas Hobbes and the tradition of
modern thought about the discipline of legality which he founded.87
Hobbes is not the only figure in the history of political thought
who can prove helpful in this regard, in part because a suspicion
of or even hostility to democratic controls creates an incentive to
inquire more deeply into the controls of legality. Consider that
Hayek is one of the figures in the liberal tradition most closely
associated with the political ideology of libertarianism and thus
with hostility to the welfare state, most famously expressed in his
The Road to Serfdom.88 However, in Republicanism, Pettit is willing
to recognize, albeit in footnotes, that even Hayeks conception of
freedom under a system of public laws might not look at the low
33
altitude of institutional requirements that different from
freedom as non-domination.89 Here I have argued that this
coincidence comes about not because liberals are tempted to adopt
institutional forms that are premised on a conception of freedom as
non-domination despite the fact that they espouse some version of
negative liberty that cannot support such forms. Rather, they adopt
those forms precisely because on closer inspection it appears that
they share the basic idea behind the Republican ideal of
non-domination--that above all it is important that individuals
have secured for them the ability to exercise what I called above
their discretionary liberty, their freedom to decide for themselves
how to live independently of otherss arbitrary interferences, both
actual and potential. Thus, Hayek in his later work placed his own
conception of the rule of law firmly within the tradition of
thought embraced by Republicans, when he claimed Cicero as the main
authority for modern liberalism because he had provided many of the
most effective formulations of freedom under law: To him is due the
conception of general rules or leges legum, which govern
legislation, the conception that we obey the law in order to be
free, and the conception that the judge ought to be merely the
mouth through which the law speaks. No other author shows more
clearly that during the classical period of Roman law it was fully
understood that there is no conflict between law and freedom and
that freedom is dependent upon certain attributes of the law, its
generality and certainty, and the restrictions it places on the
discretion of authority.90 Moreover, Hayek relied on exactly the
contrast between slavery and freedom so central to Republican
thought. His The Constitution of Liberty is presented as an
exercise in the recovery of the original meaning of freedom. Man,
or at least European man, enters history divided into free and
unfree; and this 34
distinction had a very definite meaning. The freedom of the free
may have differed widely, but only in the degree of independence
which the slave did not possess at all. It meant always the
possibility of a persons acting according to his own decisions and
plans, in contrast to the position of one who was irrevocably
subject to the will of another, who by arbitrary decision could
coerce him to act or not to act in specific ways. The time-honored
phrase by which this freedom has often been described is therefore
independence of the arbitrary will of another. 91 Here Hayek
emphasized, more important than the fact that this was the original
meaning of freedom, is that it describes a state which is desirable
for reasons different from those that which make us desire other
things called freedom, for example, freedoms from and freedoms
to.92 Finally, Hayek is clear that whilst there is an important
question about how many courses of action are open to a person,
this is a different question from the one that should be the
primary focus of political and legal philosophy. This is the
question of how far in acting an individual from that of how far in
acting he can follow his own plans and intentions, to what extent
the pattern of his conduct is of his own design, directed towards
ends for which he has been persistently striving rather than
towards necessities created by others in order to make him do what
they want. Whether he is free or not does not depend on the range
of choice but on whether he can expect to shape his course of
action in accordance with his present intentions, or whether
somebody else has power so to manipulate the conditions as to make
him act according to that persons will rather than his own. Freedom
thus presupposes that the individual has some assured private
sphere, that there is some set of circumstances in his environment
with which others cannot interfere. 35
For Hayek, then, as we saw for Hobbes, the kind of liberty
produced by a public order of law has the following
characteristics: (a) it is different in kind from negative and
positive liberty; (b) it is the kind that should be the primary
focus of political and legal philosophy because of the desirability
of the social and political condition in which it is secured; (c)
its value depends not on its extent or quantity but on its quality;
(d) that quality is affected negatively by the capacity for
arbitrary interference as well as by actual arbitrary
interferences. Once that quality of civic or legal liberty is
attained, there are further tasks for political philosophy: for
example, whether and to what extent there should be a welfare
state. It is significant that Pettit finds himself in Republicanism
having to distinguish himself from Hayek, considered one of the
leading proponents of negative-liberty liberalism of the twentieth
century, by pointing out that his own version of freedom as
non-domination is as much concerned with the potential for social
arbitrarinessfor example, domination in the workplace and the
family--as it is with the potential for political arbitrariness.93
However, it is not that clear that Hayek was unconcerned with
social domination. Rather, he was worried about the potential for
domination in giving an extensive role to the state to use law to
combat the kinds of social domination to which Pettit points. That
is, attempts at extensive state regulation necessarily multiply the
amount of state interference in our lives, and thus the occasions
when we might find ourselves subject to the coercive and arbitrary
judgment of some state official. That concern becomes heightened
when we notice that the main influence in complex legislative
regimes designed to combat social domination is quite likely
36
to be that of powerful elites, which might be what Hobbes had in
mind when he said. As we saw above, that bad laws are but traps for
money. Put differently, the use of law as an instrument to combat
social domination carries inherent risks of creating new forms of
social domination. And in this regard it is worth noting that Hayek
shifted from a position totally opposed to the welfare state at the
time of The Road to Serfdom to one that recognized that the problem
was not about state interference as such, but about whether state
interference can be made subject to the discipline of legality.
That question remains as pressing today as we consider contemporary
manifestations of threats to liberty that highlight the fragility
and contingency of a particular form of society that makes possible
what Hobbes thought of as civic liberty. And the lesson of my
chapter is that there is much to be learned from those thinkers in
the Western tradition who, perhaps because of their suspicion of,
even hostility to democracy, focused their attention on the way in
which legality might help to secure freedom as non-domination.
1
Professor of Law and Philosophy, Toronto. For discussion of this
paper, I thank the
participants in the Tragedy of Liberty conference at the Central
European as well as the audience at a Department of Philosophy
Workshop at the University of Vienna. My greatest thanks are due to
Ariel Zylberman, whose comments on a revised draft of that paper
led to rather radical changes. 37
2
Isaiah Berlin, Two Concepts of Liberty, in Berlin, Four Essays
on Liberty (Oxford: Oxford
University Press, 1969) 118.3
Ibid, 123, note 2. Berlin, Two Concepts of Liberty, 122, note 2.
I believe the quote to come from Joseph Schumpeter. Berlin, Two
Concepts of Liberty, 172. FA Hayek, The Road to Serfdom (Chicago:
University of Chicago Press, 1994). Ibid, 80. Philip Pettit,
Republicanism: A Theory of Freedom and Government (Oxford: Oxford
University
4
5
6
7
8
9
Press, 1999).10
Ibid, 50. Ibid, 31-5. See Frank Lovett and Philip Pettit,
Neorepublicanism: A Normative and Institutional
11
12
Research Program (2009) 12 American Review of Political Science
11, 16.13
Ibid. Berlin, Two Concepts of Liberty, 129. For the idea of
alien control, see Philip Pettit, Republican Freedoms: Three
Axioms,
14
15
Four Theorems in Ccile Laborde and John Maynor, eds,
Republicanism and Political Theory (Oxford: Blackwell, 2008)
102.16
Pettit, Republicanism, 292-97. Philip Pettit, The Instability of
Freedom as Noninterference: The Case of Isaiah Berlin,
17
(2011) 121 Ethics 693. In the article, terms such as
non-frustration are not hyphenated, but I will use hyphens for the
sake of uniformity. 38
18
Berlin, Four Essays on Liberty, xlviii and for the implicit
break, see Two Concepts of
Liberty, 127, as set out by Pettit, Instability, 698.19
Ibid. Pettit, Instability, 704-5. Ibid, 706. Ibid, 709. Ibid,
712, quoting from Berlin, Two Concepts of Liberty. Pettit,
Instability, 712. Berlin, Two Concepts of Liberty, 134. Ibid.
Philip Pettit, On the Peoples Terms: A Republican Theory and Model
of Democracy (Cambridge
20
21
22
23
24
25
26
27
University Press, forthcoming).28
For a distinguished example of this kind of criticism, see
Charles Larmore, The Meanings
of Political Freedom in Larmore, The Autonomy of Morality
(Cambridge: Cambridge University Press, 2008) 168.29
I owe this term to Ian Shapiro, On Non-Domination, (2012) 62
University of Toronto Law
Journal 293.30
Note XX above. Pettit, Republicanism, 41-5, and Philip Pettit,
Liberty and Leviathan, (2005) 4 Politics,
31
Philosophy, & Economics 131; Quentin Skinner, Hobbes and
Republican Liberty (Cambridge: Cambridge University Press,
2008).32
Hobbes, Leviathan, 128. For the full quote, see the text below.
Pettit, Instability, 696. 39
33
34
Ibid, 711-12. Skinner, Hobbes and Republican Liberty, 160-61.
Ibid, 151-77. Skinner says of the first sense that it amounts to
the most outrageous
35
36
moment of effrontery in the whole of Leviathan; ibid, 151.37
Thomas Hobbes, The Elements of Law Natural and Politic (Oxford:
Oxford University Press,
1994), in De Corpore Politico, 173-4.38
Ibid. See Pettit, Liberty and Leviathan, Thomas Hobbes,
Leviathan (Cambridge: Cambridge University Press, 1997, Richard
Tuck,
39
40
ed.), 146. See Skinner, Hobbes and Republican Liberty, chapter
5.41
Note XX above. Hobbes, Leviathan, 88-9. Ibid, 231. Ibid, 239.
Ibid, 239-40. See Scott Shapiro, Legality (Cambridge, Mass.:
Harvard University Press, 2011). Michael Oakeshott, The Rule of Law
in Oakeshott, On History and Other Essays
42
43
44
45
46
47
(Indianapolis: Liberty Fund, 1999) 129, 148.48
In a draft companion paper, Liberty and Legal Form, I argue that
this kind of law is the
basic legal form, underpinning both kinds of law identified by
HLA Hart--duty-imposing and power-conferring rules. That is, both
these kinds of law, albeit in different ways, instantiate the legal
form that introduces a particular kind of liberty.
40
49
For a compelling critique of Pettits analysis of Hobbes on
liberty as well as an argument in
support of the claim about exegesis, see Lars Vinx, Hobbes on
Civic Liberty and the Rule of Law, in David Dyzenhaus and Thomas
Poole, eds., Hobbes and the Law (Cambridge: Cambridge University
Press, 2012) 145.50
Hobbes, Leviathan, 147. Ibid. Ibid, 149. Ibid, 148. Ibid, 152.
Hobbes says the bonds of the law are in their own nature but weak
though he adds that
51
52
53
54
55
they may neverthelesse be made to hold, by the danger, though
not by the difficulty of breaking them. Ibid, 147.56
David Gauthier, Thomas Hobbes and the Contractarian Theory of
Law (1990) 16
Canadian Journal of Philosophy 5.57
Hayek, The Road to Serfdom, 80. Joseph Raz, The Rule of Law and
its Virtue in Raz, The Authority of Law: Essays on Law and
58
Morality (Oxford: Clarendon Press, 1983) 210, 224.59
Ibid. Hobbes, Leviathan, 183. Raz has extensively analysed this
feature of law, but his analysis is
60
constrained by his positivistic commitmentssee especially,
Joseph Raz, Authority, Law, and Morality in Raz, Ethics in the
Public Domain: Essays in the Morality of Law and Politics (Oxford:
Oxford University Press, 1994) 194; and for incisive critique, see
Stephen Darwall, Authority and Reasons: Exclusionary and Second
Personal (2010) 120 Ethics 257 and 41
Authority and Second Personal Reasons for Acting, in David Sobel
and Steven Wall, eds., Reasons for Action (Cambridge: Cambridge
University Press, 2009) 135.61
For example, immediately after Hobbess claim that the sovereign
is bound by the law of
nature to make good law; Leviathan, 231.62
Most recently, in Hobbes on the Authority of Law, in Dyzenhaus
and Poole, Hobbes and
the Law, 186.63
As he put it, the laws of nature should not be seen as
independent principles which, if
followed by legislators, would endow their laws with a quality
of justice; they are no more than an analytic break down of the
intrinsic character of law, the jus inherent in genuine law which
distinguishes it from a command addressed to an assignable agent or
a managerial instruction concerned with the promotion of interests;
Oakeshott, The Rule of Law, 173.64
Hobbes, Leviathan, 110. Fuller discussed eight such desiderata:
generality, promulgation, non-retroactivity, clarity,
65
non-contradiction, possibility of compliance, constancy through
time, and, the one which he took to be the most complex, congruence
between official action and declared rule. See Lon L. Fuller, The
Morality of Law (New Haven, Conn.: Yale University Press, 1969,
revised edition), chapter 2. He argued that a legal system that
fails completely to meet one of these, or fails substantially to
meet several, would not be a legal system. It would not qualify as
government under lawas government subject to the rule of law.
66
Ibid, 80. Hayek thus made himself an easier target for Razs
critique in The Rule of Law and Its
67
Virtue, an essay in which Hayek and Fuller are Razs main foils.
42
68
FA Hayek, The Principles of a Liberal Social Order, in Hayek,
Studies in Philosophy, Politics
and Economics (New York: Simon and Schuster, 1969), 160, at
163.69
Ibid. His emphasis. Note that the book is dedicated to Karl
Popper. In a review article of a
collection of Poppers essays, Katrina Forrester traces Poppers
intellectual journey from a position of opposition to communism but
not to social democracy to an unrelenting libertarianismToqueville
anticipated me, (2012) 34 London Review of Books, 26 April 2012,
42.70
Hayek, The Road to Serfdom, 82. See note XX above. See Richard
Bellamy, Political Constitutionalism and the Human Rights Act
(2011) 9
71
72
International Journal of Constitutional Law 86. See further,
Samantha Besson and Jos Luis Mart, Law and Republicanism: Mapping
the Issues in Besson and Mart, eds., Legal Republicanism (Oxford:
Oxford University Press, 2009), 3, at 32-33: legal republicanism
ought to encompass a positivist theory of law because it cannot
rely on the existence of a natural, prepolitical validity.73
See Raz, Authority, Law, and Morality, 202-4, for an explanation
of why a positivistic
account of authority requires the notion of content-independent
reasons.74
Pettit is more ambiguous on this point than I allow in the text,
especially in On the Peoples
Terms: A Republican Theory and Model of Democracy. He seems
undecided whether interests can be determined pre-institutionally
in accordance with some impartial, third personal test, or whether
they have to be determined institutionally, in the course of
individual interaction, that is, secondpersonally. Perhaps it is
his view is that the interest in not being dominated is determined
on the first method, and everything else falls out of the second.
As Ariel 43
Zylberman has pointed out to me, there is a problem for
consequentialist Republicans like Pettit if what ultimately matters
is a social condition of non-domination. For it is not then
immediately clear that the rule of law is a necessary means to
attain the desired Republican outcome. In my view, which I will not
defend here, Hobbess version of non-domination whilst perhaps also
consequentialist in that non-domination is the goal of civil order,
is not vulnerable to the same form of critique.75
Shapiro, On Non-Domination, 331. Hobbes, Leviathan, Dedication,
3. I discuss these ideas further in How Hobbes Met The Hobbes
Challenge, (2009) 72
76
77
Modern Law Review 488.78
Vinx, Hobbes on Civic Liberty and the Rule of Law, 163-4. This
is the argument of much of Pettits new book, On the Peoples Terms:
A Republican
79
Theory and Model of Democracy.80
Hobbes, Leviathan, 128. Ibid, 128. Ibid, 147-8. In fact, this is
precisely how Pettit proceeds in On the Peoples Terms: A Republican
Theory and
81
82
83
Model of Democracy, as the requirements of social justice are
set out in chapter two, and the analysis of political legitimacy
begins in chapter 3.84
See note XX above. Here we should note Steven Pinkers intriguing
argument in The Better Angels of Our Nature:
85
Why Violence has Declined (New York: Viking, 2011), in which he
takes Hobbes along with Kant to be the main philosophical influence
in the civilizing process that leads to the decline 44
of violence that is the theme of his book. On Pinkers view,
Hobbes explains more the civilizing process within states, whilst
Kant explains more the process between states.
86
Hobbes, Leviathan, 223. Perhaps the best exploration of these
issues in the twentieth century is Hans Kelsen, Vom
87
Wesen und Wert der Demokratie (Aalen: Scientia Verlag,
1981).88
FA Hayek, The Road to Serfdom (Chicago: University of Chicago
Press, 1994). Pettit, Republicanism, note 9 at 50, note 2 at 89. F
A Hayek, The Constitution of Liberty (London: Routledge & Kegan
Paul, 1960), 166-7. Ibid, 12. Ibid. Pettit, Republicanism,
89-90.
89
90
91
92
93
45