Carl Schmitt's Defense of Sovereignty
Lars Vinx, Department of Philosophy
Bilkent University
H.L.A. Hart once remarked that a sovereign, according to the classical doctrine of
sovereignty, is "as essential a part of a society which possesses law, as a backbone is of a
man."1 Not least as a result of Hart's own attack on Austin's theory of sovereignty,
analytical legal theorists today agree that a sovereign is not just unnecessary for but even
incompatible with the existence of legal order. To explain the variety and persistence of
legal norms, the continuity of legal order, as well as the evident possibility of legal
constraints on legislative power, so the contemporary legal-theoretical consensus, legal
order must be regarded as rule-based. As a result, sovereignty can at best be an office
defined by positive law. There can be no meta-legal, purely political power that is legally
illimitable and yet functions as the source of all positive law.2
Nevertheless, the doctrine of sovereignty is not without its contemporary
defenders. The authors in question, though, rarely address the legal-theoretical worries
about sovereignty head on. Rather, they argue that sovereignty is essential to the
legitimacy of a modern democratic constitution. To be democratically legitimate, it is
claimed, a constitution must be the product of an exercise of a constituent power that is
1
H.L.A. Hart, The Concept of Law (Oxford: Oxford University Press, 1961) 49.
2
See Hart (n. 1 above) 18-76; Joseph Raz, The Concept of a Legal System. An Introduction to the Theory
of Legal System (Oxford: Oxford University Press, 1970) 27-43; Pavlos Eleftheriadis, ‘Law and
Sovereignty’, in Law and Philosophy 29 (2010) 535-569.
1
prior to all positive law (including positive constitutional law) and that functions as the
legitimating source of all positive legality.3 A popular sovereign that stands above all
positive law, hence, is the cornerstone, according to Martin Loughlin, of any modern
public law.4
However, if the doctrine of sovereignty conflicts with key features of legal order,
a constitutional theory based on the concept of sovereignty must be flawed. And if the
idea of a sovereign authority as the source of legal order can't possibly make any sense,
we will have to let go of the claim that a constitution must be legitimized by reference to
constituent power. Those who want to hold on to the notion of sovereignty, because they
think it essential to a democratic constitutional theory, must first establish that the
concept of sovereignty is jurisprudentially meaningful.
Carl Schmitt's theory of sovereignty is the obvious place to look for a defense of
the continuing jurisprudential relevance of the concept of sovereignty. Schmitt's famous
definition of sovereignty - sovereign is he who decides on the state of exception5 - offers
a surprisingly sophisticated response to the legal-theoretical challenge to sovereignty.
Schmitt managed to show, I will argue, that the presence of a legally illimitable sovereign
whose decisions condition the applicability of law is not incompatible with the existence
3
See for example: Dieter Grimm, Souveränität. Herkunft und Zukunft eines Schlüsselbegriffs (Berlin:
Berlin University Press, 2009) 99-123; Olivier Beaud, La puissance de l'état (Paris: Presses Universitaires
de France, 1994) 199-491; Paul W. Kahn, Political Theology. Four New Chapters on the Concept of
Sovereignty (New York: Columbia University Press, 2011).
4
See Martin Loughlin, Foundations of Public Law (Oxford: Oxford University Press, 2010).
5
Carl Schmitt, Politische Theologie. Vier Kapitel zur Lehre von der Souveränität, 2nd edition (Berlin:
Duncker&Humblot, 1996) 13.
2
of rule-based legal order. Still, Schmitt's theory of sovereignty falls short of a full
rehabilitation of the classical doctrine of sovereignty. His argument does not establish that
a sovereign is as necessary to law as a backbone is to a man. Like his contemporary
followers in constitutional theory, Schmitt claims instead that a sovereign is necessary for
the existence of a legitimate legal order.6
To assess Schmitt's theory of sovereignty, as well as contemporary efforts to
defend the relevance of sovereignty for democratic constitutional theory, we therefore
have to ask whether there is good reason to hold that a legal order can only be legitimate
if it derives from a sovereign power above the law. I will argue that there is not. Schmitt's
defense of sovereignty, despite its partial success against the legal-theoretical criticism of
sovereignty, is a dead end for the doctrine of sovereignty.
The aims of Schmitt's theory of sovereignty will remain opaque unless they are
situated in the context of the history of the doctrine of sovereignty. I will therefore begin
by offering a brief discussion of the classical doctrine of sovereignty in the work of
Thomas Hobbes, and then go on to outline the criticism of the doctrine of sovereignty in
the legal theories of Hart and Kelsen.
I. Hobbes and the Sovereignty of the State
In explaining how a social contract is made, and what its effects are, Hobbes talks
about the state before he talks about the sovereign. The state, we hear in chapter 17 of
Leviathan, is "One Person, of whose Acts a great Multitude, by mutuall Covenants one
6
See Hasso Hofmann, Legalität gegen Legitimität. Der Weg der politischen Philosophie Carl Schmitts, 4th
edition (Berlin: Duncker&Humblot, 2002); David Dyzenhaus, Legality and Legitimacy. Carl Schmitt, Hans
Kelsen and Hermann Heller in Weimar (Oxford: Oxford University Press, 1997) 38-101.
3
with another, have made themselves every one the Author,"7 in the expectation that the
artificial person thus created will use the united power of all to give protection from the
dangers of the state of nature to each. The sovereign, in turn, is introduced in the next
sentence as "he that carryeth this Person."8 The sovereign, in other words, is a
representative, the representative of the person of the state. Consequently, the powers that
Hobbes attributes to the sovereign are not the private property of the natural person or
group of natural persons who are sovereign. They are essentially public powers.
Hobbes's discussion of essential powers of sovereignty in the next chapter
reinforces this point.9 In order to arrive at a list of the powers that are "inseparably
annexed"10 to the office of sovereignty, Hobbes asks what powers a sovereign must be
able to wield in order to secure the survival of the artificial person of the state. Hence, it
is the interest of the state, not the interest of the person or group of persons who hold
sovereign power, that determines the material content of the power of sovereignty.
Of course, Hobbes was an absolutist and an authoritarian political thinker. The list
of essential sovereign powers that Hobbes offers in chapter 18 of Leviathan includes an
impressive array of competences. Hobbes is insistent, moreover, that the state will be able
to provide peace and security to its subjects only if all these powers are united in the
hands of one person or one group of persons. Any form of separation or division of
powers, in Hobbes's view, would raise the danger of an irresolvable conflict between
7
Thomas Hobbes, Leviathan, ed. Richard Tuck (Cambridge: Cambridge University Press, 1996) 121.
8
Ibid. 121.
9
Ibid. 121-9.
10
Ibid. 127.
4
different organs of state.11 Finally, Hobbes puts strong emphasis on the claim that the
sovereign is not legally accountable to his subjects in any way. Not being a party to the
social contract, the sovereign cannot rightfully be deposed, accused of injustice in a court
of law, or be punished.12
It would nevertheless be mistaken to infer that the powers of Hobbes's sovereign
are legally unlimited. Throughout Leviathan, Hobbes repeatedly emphasizes the claim
that the sovereign lacks the legal power to alienate any of the essential rights of
sovereignty. In particular, Hobbes is very anxious to hammer home the point that a
sovereign cannot possibly be bound to decisions of previous sovereigns that appear to
have granted away sovereign powers, such as the power to tax without seeking consent.
Any such grant, in Hobbes view, is to be regarded as void, unless it went along with an
explicit renunciation of sovereignty.13
Needless to say, an individual person holding the office of sovereignty may well
have a private interest in alienating a power of sovereignty. The sovereign's legal
incapacity to alienate any essential power of sovereignty results from the fact that such an
alienation would endanger the survival of the collective person that we call the state. A
sovereign power cannot be alienated, in other words, because a sovereign person, in
taking the decision to alienate it, could never be understood to act in the name of the
state. It follows that Hobbesian sovereignty must be an attribute of the state, and not the
private property of those who hold the office of sovereignty.
11
Ibid. 127.
12
Ibid. 122-4.
13
See ibid. 127, 153, 222.
5
The inalienability of sovereign power in Hobbes illustrates an important feature of
Hobbes's theory of the state, a feature that is also evident in the claim that one can draw
up a list of essential powers of sovereignty, or in the claim that no system of government
can possibly accommodate a separation of powers and still fulfill its essential function. In
Hobbes's view, the social institution that we call the state has an essence or nature that is
determined by its function. The function of the state is to provide peace and security to its
members, by ending the state of nature between them, and by protecting them against
external enemies. Any institution that could possibly serve this purpose with a degree of
success sufficient to deserve to have its claim to authority recognized must, according to
Hobbes, instantiate the constitutional framework laid out in Leviathan. Both the
sovereign and the parties to a social contract are bound, according to Hobbes, to what I
will call a ‘constitution in natural law.'
Hobbes's constitution in natural law is concerned, in large part, with the
organization of the institutional structures through which political power is to be
exercised. It primarily aims to enhance the efficiency of the exercise of power, and to
brush aside the relics of a feudal constitutionalism that Hobbes believes must frustrate the
sovereign's ability to secure peace and order. Thus, Hobbes famously claims that all
positive laws or laws ‘properly so called' are sovereign commands, which implies that a
sovereign's legislative power cannot be restricted by positive law.14 Still, the dependence
of positive legality on sovereign authority is only one aspect of the relationship between
law and sovereignty embedded in the constitution in natural law. The second key aspect
14
See ibid. 184-5.
6
of that relationship is given by the tight connection that Hobbes forges between the
function of the state and the ideal of the rule of law.15
Hobbes clearly believes that it is one of the sovereign's primary tasks to use his
legislative power to concretize the laws of nature discussed in chapters 14 and 15 of
Leviathan, and to make the moral code expressed in the laws of nature applicable,
through adjudication and enforcement, to the solution of concrete social conflict. Hobbes
goes so far as to claim that the laws of nature have legal force in every possible state,
regardless of whether the sovereign took care to incorporate them explicitly into the
positive law.16 He suggests, moreover, that a sovereign lacks the authority to enact an
explicit legislative ouster of judicial recourse to the laws of nature, though his
understanding of the laws of nature will, of course, prevail against anyone else's.17 The
sovereign's attempts to govern his subjects, finally, are bound to the rule of law and to
principles of natural justice, in the sense that his authority over subjects who are not
treated in accordance with the rule of law must necessarily lapse.18
The laws of nature, though they are not proper laws without the sword of the
sovereign, provide the concrete content to Hobbes's conception of peace and security, and
15
For some recent work affirming the importance of the rule of law for Hobbes see chapters 6-10 in
Hobbes and the Law, ed. by David Dyzenhaus and Thomas Poole (Cambridge: Cambridge University
Press, 2012); Perez Zagorin, Hobbes and the Law of Nature (Princeton: Princeton University Press, 2009)
84-98.
16
See Hobbes, Leviathan (n. 7 above) 185.
17
See ibid. 190-5; David Dyzenhaus, ‘Hobbes and the Legitimacy of Law', in Law and Philosophy 20
(2001) 461-98.
18
See the discussion of punishment in Hobbes, Leviathan (n. 7 above) 214-21.
7
thus to his understanding of the essential functions of the state. The Hobbesian state,
therefore, is necessarily committed to the goal of creating a stable and dependable legal
order, and a sovereign cannot openly disavow the pursuit of that goal, or constantly act in
ways evidently incompatible with it. Hobbes would have resisted the idea that
exceptional exercises of extra-legal power reveal the true essence of sovereign power.
That essence is defined by the function of the state, which is to provide for peace and
security through the creation and maintenance of legal order.
Hobbes bears a degree of responsibility, though, for the fact that the limitative
consequences of his constitution in natural law have often been overlooked by
commentators. His obsession with the prevention of internal political discord has made it
difficult to recognize the sovereignty of the state in the powers attributed by Hobbes to
the sovereign person or group of persons. Hobbes argued that a state can be sovereign (or,
what amounts to the same thing, that it can exist) only if it is represented by a sovereign
organ, by a person or group of persons who hold all the essential powers of the state in
their own hand. For all practical intents and purposes, Hobbes identified the sovereignty
of the state with the authority of the sovereign person, so as to rule out any internal
dispute about who is to speak in the name of the state.
This identification is based on a number of empirical assumptions that have
turned out to be false. Historical experience has shown that Hobbes was wrong to believe
that a state, in order successfully to pursue the goals of providing a rule of law, public
security, and protection against external enemies, must be represented by a sovereign
authority that unifies all powers of the state in the hands of one person or group of
persons. A latter day Hobbesian would have to argue, presumably, that modern states
8
whose constitutional structures do not fit Hobbes's constitution in natural law are not
really states at all, that they merely conceal a continuing state of war among different
social groups. Such a claim is too preposterous to merit refutation.
It does not follow, however, that it is wrong to think that the state has essential
purposes - including the purposes of providing peace, security, and the rule of law - or to
hold that it must claim an authority that can reasonably be called ‘sovereign' to be able to
achieve those purposes. Arguably, a state, to create peace and security, must still take
itself to be the supreme arbiter of all social conflict in a certain territory, it must still
successfully monopolize the use of legitimate coercive force in that territory, and its
decision-taking must still be organized in such a way as to ensure that there is a clear and
unambiguous procedural path towards the final settlement of any potential dispute.19 The
dismissal of the Hobbesian sovereign, I conclude, does not establish that the state cannot
or need not be sovereign. Rather, it should help us to put the focus of the doctrine of
sovereignty where it belongs: on the state.20
19
See for more detailed discussion Lars Vinx, ‘Constitutional Indifferentism and Republican Freedom', in
Political Theory 38 (2010) 809-37.
20
Hobbes, at any rate, is not the only author who developed an idea of a constitution in natural law. As
Martin Loughlin has shown in his magisterial study on the foundations of public law, many early modern
authors tried to develop theories of a constitution in natural law or, to use Loughlin’s terminology, a
‘science of public right’. See Loughlin, Foundations of Public Law (n. 4 above) 91-180. In Loughlin’s
view, however, the failure of early modern natural law theorists to agree on a particular conception of the
constitution in natural law suggests that the project of a science of public right is doomed to fail. See ibid.,
157-64. Public law, Loughlin concludes, must be based on a prudential and not on a scientific discourse of
politics. I am not convinced that we should dismiss the idea of a constitution in natural law so quickly. The
mere fact that authors have put forward different accounts of a constitution in natural law (or, what amounts
9
II. Austin, Hart, and the Problem of Legislative Sovereignty
John Austin's theory of sovereignty is too well-known to require a lengthy
rehearsal. Austin's most fundamental assumption is that all laws are general commands.
Austin understands a command as a mandatory directive, issued by a superior to a less
powerful inferior, and backed by the threat of a sanction in case of non-compliance. He
distinguishes commands that are proper laws from other general commands by reference
to their enactment by a sovereign, who is defined as a person or group of persons
habitually obeyed by the majority of the members of a society and not paying habitual
obedience to anyone else.21
Though Austin presents his legal theory as a development of Hobbes's ideas, his
approach is much narrower in its focus and in its concerns than Hobbes's theory of the
state. Take the view that there must be a sovereign wherever there is law, or the view that
a sovereign's authority is not limitable by positive law. For Hobbes, these claims are
grounded in substantive claims about the essential function of the state, and the necessary
institutional means of their attainment. In Austin, by contrast, they are reduced to
implications of stipulative definitions of ‘law' and ‘sovereignty.' Austin, for instance,
declares limited sovereignty to be a "flat contradiction in terms" 22 for the simple reason
to the same thing, different accounts of the nature of the state) hardly suffices to show that the project of a
science of public right is bound to fail. It may as well be the case that its concepts have yet to be elevated to
the requisite level of generality.
21
See John Austin, The province of Jurisprudence Determined, ed. Wilfrid Rumble (Cambridge:
Cambridge University Press, 1995) 18-37, 164-241.
22
Ibid. 212.
10
that a sovereign, by definition, does not pay habitual obedience to anyone else and thus
cannot be subject to laws that are conceived as commands of a superior power.
Of course, the point here is not that Hobbes would disagree with the view that
sovereignty limited by positive law is a contradiction in terms. The point is that Hobbes's
theory of the state attempts to explain why all positive laws or laws ‘properly so-called'
must be sovereign commands, and why subjection to positive law would destroy
sovereignty (and thus all orderly government); by arguing that the state must take a
certain institutional form - a form structured by the principle of sovereignty - in order to
be able to provide peace and security. Such a substantive explanation of the need for
sovereign authority in a theory of the state is nowhere to be found in Austin.
What is more, Austin's reductive account of the sovereign as a natural person or
group of natural persons receiving habitual, de facto obedience from the bulk of a
population makes it impossible to conceive of the sovereign as a representative of the
state. Austin's legal theory, as Kelsen rightly observed, simply lacks the conceptual
resources to draw the distinction between the sovereignty of the state and the de facto
power of a sovereign person.23 As a result, Austin fails to recognize the legal limitations
of sovereign authority that might flow from a constitution in natural law.
As Hart has shown with admirable clarity, Austin's theory of sovereignty cannot
account for the continuity of legal system, and the persistence of laws, in cases where a
sovereign is succeeded by a new sovereign. It also mistakenly assimilates all laws to
sovereign commands and thus fails to acknowledge the diversity of legal rules, and in
23
See Hans Kelsen, ‘The Pure Theory of Law and Analytical Jurisprudence', in Hans Kelsen, What is
Justice? Justice, Law, and Politics in the Mirror of Science (Berkeley: University of California Press, 1957)
266-87.
11
particular the existence of secondary rules that determine how mandatory legal rules are
to be introduced, changed, or authoritatively applied.24 The fact that all legal systems
contain secondary rules, and most importantly a rule of recognition setting out the criteria
for the validation of law, serves to explain, in Hart's theory, how even the legal powers of
a legislator who is not subject to sanction-backed legal duties can come to be limited: a
rule of recognition may withhold from a legislator the authority to legislate in certain
ways and thus create limitations of legal authority in the form of legal incapacities.25
Austin's sovereign, a legislator who can legislate in whatever way it pleases, now appears
as only one of many possible forms of legislative authority under a rule of recognition.
I do not wish to take issue with the cogency of Hart's criticisms of Austin. What I
would like to suggest, though, is that Hart's critique of Austin remains beholden to the
narrowing of jurisprudential perspective that is implicit in Austin's reduction of the theory
of the state to a mere theory of legal system. As I pointed out, Austin's theory lacks the
conceptual resources to recognize that sovereignty is an attribute of the state, and this
limitation carries over into Hart's legal theory. Hart established that if sovereignty is
understood exclusively as legislative authority that inheres in a particular natural person
or group of natural persons, in virtue of overwhelming de facto power, the appeal to
sovereignty must fail to explain certain key features of legal system. But Hart, like
Austin, does not address the substantive claims of Hobbes's theory of public law, which,
as we have seen, conceives of sovereignty as an attribute of the state that is grounded in
the essential functions of the state.
24
See Hart, The Concept of Law (n. 1 above) 18-76.
25
See ibid. 64-9.
12
As a result, Hart's criticisms of the doctrine of sovereignty are not as general in
their jurisprudential bearing as may appear at first glance. They do not apply to versions
of the doctrine of sovereignty that conceive of sovereignty as an attribute of the state, and
that allow for the possibility of a sovereign state not governed by a sovereign person or
group of persons. Likewise, they do not apply to versions of the doctrine of sovereignty
that do not understand sovereignty as a legislative power. Consequently, Hart's attack on
sovereignty fails to show that the doctrine of sovereignty is no longer jurisprudentially
meaningful in any of its possible forms.26
III. Kelsen on the Sovereignty of the State and the Sovereignty of Law
Kelsen's reflections on sovereignty engage the problem of the sovereignty of the
state much more directly than Austin's or Hart's. In his major work on sovereignty - Das
Problem der Souveränität und die Theorie des Völkerrechts - Kelsen did not reject the
doctrine of sovereignty. Rather, he tried to work out the constitutional consequences of
the view that sovereignty is an attribute of the state.27
Kelsen's reconstruction of the doctrine of sovereignty starts out from the
observation that all variations of the doctrine of sovereignty - despite their lack of
agreement on a clear conception of sovereignty - concur that the subject to which
sovereignty is attributed is to be regarded as a highest or a supreme authority to which all
other persons or agents are subject, while it is not itself subject to any other authority.
26
The same holds for Eleftheriadis, ‘Law and Sovereignty’ (n. 2 above).
27
Hans Kelsen, Das Problem der Souveränität und die Theorie des Völkerrechts. Beitrag zu einer reinen
Rechtslehre (Tübingen: J.C.B. Mohr-Paul Siebeck, 1920) 1-101.
13
Consequently, if the state is regarded as sovereign, then the state must be "something that
occupies a highest position."28
Kelsen points out that the concept of sovereignty, so understood, is ambiguous.29
It could be taken to refer to an empirical fact of domination, i.e. to the psychological or
sociological fact that the will of a sovereign state (or rather of those who exercise its de
facto power) is sufficient to motivate the will of its subjects, and thus to cause them to act
in a certain way. Kelsen associates this view with Georg Jellinek, who claimed that the
state can be regarded under two aspects: as a social fact of power and as a normative
order under law. Jellinek gave priority to the first of these two aspects, and argued that
the state as a fact of power creates a legal order, and then proceeds to bind itself to the
law, though not irrevocably, through its own free decision.30
A conception of the sovereign state as a fact of power, in Kelsen's view, must end
in failure. To say that the state is sovereign as a fact of power, Kelsen argues, is to say
that the will of the state determines the motivations of its subjects without being itself
determined by antecedent causes. But such a claim is meaningless, since any
psychological or sociological description of the causal relationships between, for
instance, the will of the state and the will of its subjects is committed to the assumption
that the will of the state (or of those who exercise its powers) must itself be caused by
28
Ibid., 5.
29
See ibid. 4-9.
30
See Georg Jellinek, Allgemeine Staatslehre, 3rd edition, ed. Walter Jellinek (Berlin: O. Häring, 1914)
367-75. For a very interesting and unduly neglected criticism of Jellinek’s theory see Leonard Nelson, Die
Rechtswissenschaft ohne Recht. Kritische Betrachtungen über die Grundlagen des Staats- und Völkerrechts
(Leipzig: Veit&Comp., 1917) 6-76.
14
some antecedent causal factor or set of factors. It is a condition of the intelligibility of the
causal world that every event has a cause.31
The claim that the sovereign state is something that occupies a position of
supremacy, however, can be understood in a fundamentally different way that does not
lead into incoherence. The supremacy of the will of the sovereign state over that of its
subjects can be interpreted normatively.32 To say that the will of the state is supreme must
then mean that its decisions are binding or obligatory for its subjects, whereas the will of
the state itself is not bound to conform to the decisions of any yet higher will. To assume
that the decisions of the will of the state are binding on its subjects is to assume that the
will of the state has normative authority. Such authority could not be a result, Kelsen
argues, of a mere de facto power of coercion. Rather, it must result from the fact that the
state is empowered by an antecedent norm to issue binding directives to its subjects.
Since we cannot derive ought from is, a norm can only be validated by another
norm, and not by recourse to a social or psychological fact. To avoid an infinite regress of
validation, there must, therefore, be a fundamental norm that is the source of the validity
of all other norms of any complete legal order. This fundamental norm, Kelsen will later
call it the ‘basic norm', is not itself derived from any other norm. Its validity, rather, is
presupposed by legal science, as a necessary condition of the normativity of law.
Sovereignty can now be understood as a property of a complete legal order. A legal order
31
See Kelsen, Das Problem der Souveränität (n. 27 above) 6-7.
32
See ibid. 8-9.
15
is supreme if and only if its normative authority is not derived from an external source,
but rather from its very own basic norm.33
What we call the sovereignty of a state, from a normative point of view, simply is
the normative independence of the state's legal system. The state, from a legal point of
view, is to be regarded as identical with its legal system.34 The state, Kelsen argues, is not
a real person. It can act only where it is possible to attribute acts performed by natural
persons to the state. Such attribution, however, requires that the natural persons in
question be legally authorized to act in the name of the state. Since Kelsen rejects the
idea of a constitution in natural law, at least in its narrow, Hobbesian form, which claims
that all the authority of the state must always be concentrated in the hands of one person
or group of persons, such legal authorization must now be based on positive law. Kelsen
concludes that there can be no such thing as an illegal act of state. An act not authorized
by positive law simply could not be recognized as an act of state. The sovereignty of a
state, therefore, can be no more than the personification of the unity of its positive legal
order in which we picture the normative independence of its legal system.35
This view of the relation of law and state implies that the state cannot be a
transcendent creator of law. In order for anyone to exercise legislative powers in the
name of the state, there must already be a foundational norm that confers legislative
authority, but that norm cannot itself be the result of an exercise of a power to make laws.
Jellinek's view that the sovereign state is a fact of power prior to law that creates a legal
33
See ibid. 9-16.
34
See ibid. 12-3. For further discussion of the identity thesis see Lars Vinx, Hans Kelsen's Pure Theory of
Law. Legality and Legitimacy (Oxford: Oxford University Press, 2007) 78-98.
35
See Kelsen, Das Problem der Souveränität (n. 27 above) 16-21.
16
system out of legal nothingness is therefore false.36 Though Kelsen continues to regard
the state as sovereign, his claim that the sovereignty of the state is the normative
independence of its legal order implies that organs of state will be in a position invoke the
sovereignty of the state to justify their acts only as long as what they do turns out to be
authorized by the positive law.37 The sovereignty of the state, hence, is internally related
to a principle of legality. The view that sovereignty is an attribute of the state, taken to its
logical conclusion, leaves no room for a sovereign above the positive law.
Kelsen’s argument is often subjected to criticisms that, in my view, echo Schmitt’s
polemics against Kelsen38 somewhat too uncritically. In preparation for our discussion of
Schmitt, it might be helpful to briefly discuss a recent example. Martin Loughlin claims
that the question of how the “sovereign entity of the state” can “be subject to law”
“simply evaporates” if “the state is the legal order tout court”.39 Kelsen would be well
within his rights to object to this assessment. In his view, the problem does not at all
disappear. Rather, it transforms into the question of how to make sure that those who
(claim to) act on behalf of the state do not violate or exceed the legal boundaries imposed
by the positive norms that authorize their activity, a question that is plainly analogous in
36
See ibid. 40-7.
37
See ibid. 16-7, 24-7.
38
See Schmitt, Politische Theologie (n. 5 above) 26-9; Carl Schmitt, Verfassungslehre, 8th edition (Berlin:
Duncker&Humblot, 1993) 8-9; Carl Schmitt, Der Hüter der Verfassung, 4th edition (Berlin:
Duncker&Humblot, 1996) 38-40, n. 2.
39
See Loughlin, The Foundations of Public Law (n. 4 above) 217.
17
form to the question of how a Hobbesian sovereign person can be made to stick to the
constitution in natural law. Kelsen had quite a lot to say about that issue.40
The mistaken claim that Kelsen defined the problem of the legality of acts of state
out of existence seems tied to a questionable interpretation of modern legal positivism.
According to Loughlin, “the public realm now presents itself as autonomous, it cannot be
anchored in either divine law or natural law. The public realm must function according to
laws that we have given ourselves”. It would be tempting but wrong, Loughlin goes on to
argue, to “to follow Hobbes” and to infer from this that “law means simply the command
of the established law-making authority”. To do that, Loughlin holds, would imply that
“the conditions under which [...] law-making authority is exercised are matters of politics
that lie beyond juristic knowledge”. 41
As I have tried to show, this claim goes wrong even as an interpretation of
Hobbes, for the reason that Hobbes acknowledges that the valid exercise of legislative
authority is conditioned by a constitution in natural law that, in Hobbes’s view, surely is
not beyond juristic knowledge. But it is also wrong as a charge leveled against the legal
positivism of Kelsen (or, for that matter, of Hart), notwithstanding the fact that the
assessment can appeal to the authority of Schmitt. According to Kelsen laws are not
mandatory commands issued by an un-commanded commander. They are rules that
authorize organs of state to employ legitimate force. As such, they lay down the
conditions, among other things, of the rightful exercise of legislative power. It is therefore
40
See for instance Hans Kelsen, ‘Wesen und Entwicklung der Staatsgerichtsbarkeit’, in Veröffentlichungen
der Vereinigung der Deutschen Staatsrechtslehrer, Heft 5, Berlin and Leipzig 1929, 30-88.
41
Loughlin, Foundations of Public Law (n. 4 above) 158.
18
plainly false to claim, with Schmitt, that the conditions of the rightful exercise of political
power lie beyond juristic knowledge for Kelsen.42
What seems to be going on here is that Loughlin wrongly attributes an Austinian
view of law to Kelsen (and to positivists and ‘normativists’ in general), in order to
chastise Kelsen for not recognizing that there is a law that enables and conditions the
exercise of political power. This maneuver is made all the less convincing by the fact that
Loughlin himself seems implicitly attached to an Austinian picture of law. While he
attacks Austin’s claim that all law is sovereign command, Loughlin at times seems to
express his agreement with the view that, in a modern legal order, all positive law is
sovereign command.43 He also holds that ‘droit politique’, the meta-positive law that is
said to constitute the state, is not judicially enforceable and that its content, since it
cannot be drawn from a science of public right, is wholly subject to the unrestricted
choices of the (popular) sovereign.44 In substance, this is an Austinian picture of the
42
For an overview of Kelsen’s treatment of these issues see Vinx, Hans Kelsen’s Pure Theory of Law (n. 34
above) 78-100. Kelsen’s most elaborate reply to the Schmittian criticisms of ‘normativism’ is to be found in
Hans Kelsen, ‘Wer soll der Hüter der Verfassung sein?’ in Die Wiener rechtstheoretische Schule. Schriften
von Hans Kelsen, Adolf Merkl, Alfred Verdross, vol. 2, ed. by Hans R. Klecatsky, Rene Marcic, and
Herbert Schambeck (Wien: Verlag Österreich, 2010) 1533-73.
43
See Loughlin, Foundations of Public Law (n. 4 above) 186, 196, 209-16.
44
For instance ibid. 229: “Political power is absolute, in the sense that the authority of the people to fashion
the political world is unbounded: the authority exercised through the public sphere cannot be limited by
claims of history, custom, or inherited religious beliefs. The only constraints are immanent, those that the
people or their representative governors determine to be in the public interest.” Loughlin’s argument, here
and elsewhere in Foundations of Public Law, seems to me to misunderstand the idea of an immanent
constraint. Let us assume for the purpose of argument that Hobbes’s conception of a constitution in natural
19
relation between the sovereign and the law, veiled by the rhetorical choice to refer to
what Austin would have called constitutional morality as ‘droit politique’.
This rhetorical choice, however, begs the question why ‘droit politique’ deserves
to be recognized as a species of law. The most convincing answer I can think of (pace
Schmitt) is that it consists of rules that – while they are not mandatory, sanction-backed
directives to do or not do something or other for whose disregard one can be punished in
criminal court – condition the valid exercise of public power. But if that is what makes
‘droit politique’ into law, it is perfectly possible, as Kelsen has shown, to conceive of it as
part and parcel of the system of positive law - since the system of positive law as a whole
is best understood as a system of authorizations for the use of force - and to make it
judicially enforceable, by giving judges the power to void purported acts of public power
that lack adequate legal authorization.
The real issue here, I submit, is not that positivists like Kelsen have nothing to say
about the foundations of public law or the problem of the legality of acts of state. What
really bothers Loughlin (and others) about Kelsen’s thesis of the identity of law and state
law is sound. It will follow that it is an immanent constraint of the activity of state-building that the
institution that is to be created must contain a sovereign person. This restriction is not based on ‘history,
custom, or inherited religious beliefs’. But it is not a restriction that the people, i.e. those who embark
together on the project of building a state, can simply choose to waive if they want to be successful.
According to Hobbes, one cannot create a state without sovereign just as one cannot create a house without
walls or a roof. Similar conclusions will result for any conception of a constitution in natural law or of the
nature of the state. Hence, it doesn’t follow from the rejection of tradition, custom, or religious belief that
the people, in creating a state, are subject only to ‘constraints’ that they ‘determine to be in their interest’.
There may be (and very likely in fact are) constraints that flow from the nature of the activity of statebuilding, whether a people determines that to be in its interest or not.
20
is that it appears to leave no room for a certain kind of popular sovereign: If there can be
no sovereign person who is a transcendent creator of the law, and yet completely
unbound from it, then there can be no people that acts as a transcendent creator of its own
law, and thus, according to Hans Lindahl, no collective freedom, understood as a group’s
absolutely unrestricted power to determine the conditions of its communal life as it sees
fit.45
Turning to Schmitt, I will argue that a conception of popular sovereignty as a
power transcendent to law is not necessarily incoherent, inconceivable, or fictional, as its
legal positivist critics like Hart and Kelsen have claimed. The normative arguments that
Schmitt as well as his contemporary followers put forward on its behalf, however, will
turn out to be unconvincing.
IV. Sovereignty as the Power to Decide on the Exception
We are now in a position to outline the aims of Schmitt's theory of sovereignty, as
well as to understand the challenges it had to face. The discussion so far suggests that any
attempt to conceive of sovereign authority as a source of positive law will drive us
towards the view that sovereign authority, if there such a thing, must itself be legally
constituted.
Though Hobbes conceives of the sovereign person as a transcendent source of all
positive law, he takes sovereignty to be an attribute of the state, and he holds that the
authority of the sovereign is defined and limited by a constitution in natural law. For
Hobbes, there was no tension between the claim that sovereignty is an attribute of the
45
See Hans Lindahl’s contribution to this volume.
21
state and the claim that there must be a sovereign person who is above all positive law.
Hobbes held, after all, that a state can exist only where it is represented by a sovereign
who unites all the powers of the state in his hand, and he thought that his constitution in
natural law could supply criteria of the validity of a sovereign's claim to represent the
state, even in the absence of enforceable constraints on sovereign power in positive law.
Hobbes theory of sovereignty, however, is an unstable synthesis. We know,
though Hobbes didn't, that a state can exist and be successful without being represented
by a sovereign authority that stands above all positive law and that concentrates all the
powers of the state in its hands. If, like Hart, we take the doctrine of sovereignty to claim
that, wherever there is law, there must be a sovereign person or group of persons whose
legislative decisions are the sole source of law, the failure of Hobbes's constitution in
natural law will lead us to reject the doctrine altogether. The alternative is to follow
Kelsen and to adopt a depersonalized reading of sovereignty, as the normative
independence of a state and its law. But in the absence of a Hobbesian constitution in
natural law, the state, as we have seen, cannot be prior to the positive law. Under either
option, hence, there is no room for a meta-legal sovereign authority that creates all
positive law. Hart and Kelsen both concede, or course, that a positive legal system could
be structured in such a way as to confer wide-ranging, materially unlimited powers of
lawmaking on one person or group of persons. However, such a power could not, in
either approach, be regarded as transcendent to positive law. It would depend, rather, on
the contingent content of a particular rule of recognition or basic norm.46
46
See Hart, The Concept of Law (n. 1 above) 66-8; Kelsen, Das Problem der Souveränität (n. 27 above)
27.
22
So here is the challenge that Schmitt took up in developing his theory of
sovereignty: Is there a way, without taking resort to a Hobbesian constitution in natural
law, to conceive of positive law as grounded in the decisions of a sovereign person or
group of persons, while avoiding the legal constitution of sovereign authority? 47
At first glance, the chances of meeting that challenge appear very slim indeed. It
seems that to meet the challenge, one will have to deny - in order to avoid the legalist
conclusion of Kelsen's analysis - that sovereignty is an attribute of the state, and to
conceive of sovereign authority as an attribute of a sovereign person or group of persons.
The only way to do that, presumably, is to understand sovereignty as an overwhelming de
facto power of that person or group of persons to compel obedience. But then one will
47
Adrian Vermeule and Eric Posner suggest that Schmitt’s theory of sovereignty can be interpreted as a less
ambitious, more pragmatic project. See Adrian Vermeule and Eric Posner, ‘Demystifying Schmitt’,
forthcoming in the Cambridge Companion to Carl Schmitt. In their reading, Schmitt’s point is simply that
any legal order stands in need of a residual decision taker, due to the limited foresight of rule-makers, who
can iron out conflicts of rules, deal with problems of over- and under-inclusiveness, or overcome decisional
paralysis. I agree that Schmitt might have found an authority of that sort useful, or perhaps even
pragmatically necessary. But Schmitt would not have described it as sovereign. Schmitt refers to a residual
decision-taker within an already existing constitution as a commissarial dictator or a guardian of the
constitution, offices that Schmitt aims to distinguish from full-blown sovereignty. See Carl Schmitt, Die
Diktatur. Von den Anfängen des modernen Souveränitätsgedankens bis zum proletarischen Klassenkampf,
6th edition (Berlin: Duncker&Humblot, 1994); Carl Schmitt, Der Hüter der Verfassung (n. 38 above)
chapter III. As the discussion in Schmitt’s Verfassungslehre makes clear, the sovereign is clearly supposed
to do more than just to stabilize an already existing constitution. He is to be its legitimating source, the
bearer of constituent power. In this capacity, Schmitt’s sovereign is to decide on the fundamental structure
of the constitution, and to create the situation of social normality or homogeneity that makes constitutional
law applicable in the first place. These tasks go far beyond the powers of, say, a Lockean prerogative.
23
have to face the criticisms that Hart leveled against Austin. If it is the mere possession of
de facto power to issue general commands and to compel obedience that endows a person
or group of persons with sovereign authority, it will become impossible, if Hart's
argument against Austin is sound, to explain the continuity of legal system, as well as the
permanence and diversity of legal norms.
Schmitt avoids the first horn of this dilemma by conceiving of sovereignty in
personal terms. His famous definition of sovereignty states that the sovereign is he who
decides on the state of exception.48 This defines sovereign authority as the power to take
a decision on the exception. In addition, the definition carries the implication that there
must be a personal sovereign, a real or concrete will capable of taking an actual decision
on the exception. A rule-bound, impersonal institution cannot be the primary bearer of
sovereignty. The power of sovereignty, according to Schmitt, ultimately belongs to him or
to them who can in fact take the decision on the exception, not to a legally constituted
artificial person of the state.49
48
See Schmitt, Politische Theologie (n. 5 above) 13.
49
To be sure, Schmitt suggests, in Politische Theologie, that sovereignty is an attribute of the state. For
instance, he says that in the state of absolute exception, "the state remains, while the law recedes" and that
it is "the state that suspends the law" (ibid. 18). But the term ‘state', in this context, cannot refer to an
institution structured by any kind of law, if, as Schmitt makes clear, the decision on the exception is at least
potentially a decision that suspends absolutely all law, perhaps with the view of establishing an altogether
new constitution. In his Verfassungslehre, Schmitt defines the state as “the political unity of a people”,
which manifests itself in legally unregulated acts of constituent power. See Schmitt, Verfassungslehre (n. 38
above) 21. Schmitt’s talk of the state, then, is perhaps best understood as referring to a Schmittian political
community, i.e. to a group of people united only by a shared disposition to distinguish between friend and
24
One might object here that it does not follow from the claim that the power of
sovereignty is vested in the person who can take the decision on the exception that
sovereignty can't be an attribute of the state. A Hobbesian would argue, after all, that the
sovereign person is he who holds all essential powers of sovereignty in his hands, without
thereby intending to imply that sovereignty is not attributable to the state.
This objection fails, due to unavailability of a Hobbesian constitution in natural
law. In Schmitt, the content of sovereign authority is reduced to the power to take a
decision on the absolute exception. Schmitt's definition of the power of sovereignty no
longer contains any reference to a positive list of essential rights of sovereignty derived
from an account of the function of the state. As a result, Schmitt's sovereign can no
longer play a role defined by a constitution in natural law and be publicly recognized to
do so. In the absence of such a role, there is simply no basis for attributing the decisions
of the sovereign person to the artificial person of the state. Schmitt was perfectly aware of
this fact. During the 1920's, he came to argue explicitly that the state-as-institution is
nothing but a derivative expression of a group's willingness to draw a distinction between
friend and enemy.50
enemy in the same way. See Carl Schmitt, Der Begriff des Politischen. Text von 1932 mit einem Vorwort
und drei Corollarien (Berlin: Duncker&Humblot, 1963) 20-6. However, to call such a group a state is
misleading at best, if it is essential to the state to be an artificial person or an organized community.
50
See n. 49 above. I do not think that it makes a difference to claim that the state, for Schmitt, is a concrete
order. The term ‘concrete order’, which Schmitt didn’t use before 1934, is just another name for what
Schmitt, in his earlier works, refers to as the situation of normality that, allegedly, is a condition of the
legitimate applicability of legal norms. Schmitt’s decisionist approach during the Weimar period reflects the
perception that there is no uncontested concrete order, and that a conscious sovereign decision is necessary
25
Schmitt holds that the power to take a decision on the absolute exception is not to
be understood as a power conferred by law, but rather as a de facto power. Constitutional
law, to be sure, may contain acknowledgments of the existence of such a power (or it
might, for that matter, try to deny the existence of such a power). But the power exists
wherever there is someone who can in fact take a decision on the total exception.51 It is a
misinterpretation of Schmitt’s definition of sovereignty to read it as though it claimed that
one ought to look to who is made competent by a positive constitution to decide on the
state of emergency to find out who is sovereign. The existence of such a constitutional
provision, for Schmitt, is neither a necessary nor a sufficient condition for sovereignty: It
is not necessary since there may obviously be an agency that can bring about a decision
on the total exception even without being formally authorized to do so. It is not sufficient,
because constitutional emergency-provisions, for rather obvious reasons, will typically
fail to confer a power to declare an absolute exception. Their point, as Schmitt points out,
to determine, in the first place, which concrete order is to form the substance of German political life.
Schmitt’s shift to ‘concrete order thought’ in 1934, as Hofmann pointed out, is therefore not a rejection of
the decisionist theory of sovereignty. It is premissed on the assumption that the sovereign decision as to
what is to count as normal is no longer open. See Hasso Hofmann, Legitimität gegen Legalität (n. 6 above)
68-78. The identification of the state with concrete order, in any case, runs into the difficulty that it is hard
to see how a concrete order could have a capability to act. Hans Lindahl’s suggestion (see the contribution
to this volume) that ‘concrete order’ refers to a form of collective agency in Margaret Gilbert’s sense would
solve that problem. But one would have to add that, for Schmitt, a political community is a collective agent
of a rather specific kind, namely one that can draw a friend-enemy distinction, which takes us right to the
interpretation of Schmitt’s understanding of ‘state’ proposed here.
51
See Schmitt, Verfassungslehre (n. 38 above) 75-6.
26
is to tame and to domesticate the power to decide on the exception.52 What is more, the
claim that Schmittian sovereignty is a competence conferred by a positive constitution
obviously conflicts with Schmitt’s theory of popular sovereignty. If the people are to be
sovereign, and if this sovereignty is to include the power to determine the content of their
own constitution as they see fit, and to continue to do so in the future, popular
sovereignty (or sovereign dictatorship exercised on behalf of the people) can’t be a power
or an office or a competence conferred by a positive constitution. 53
52
See Schmitt, Politische Theologie (n. 5 above) 18.
53
Loughlin claims that when Schmitt’s sovereign takes a decision on the exception, “positive law recedes,
but droit politique remains” (Loughlin, Foundations of Public Law [n. 4 above] 401). He goes on to argue
(ibid., p. 402) that droit politique manifests itself, for instance, in the fact that “in the Weimar Republic that
power [to take a decision on the exception] was vested in the president under Article 48 of the Constitution,
a common arrangement under modern constitutions.” Schmitt, however, explicitly denies that the
dictatorship of the president under article 48 is a power of sovereignty, precisely because the former, but
not the latter, is a competence allocated (and limited) by positive constitutional law. See Carl Schmitt, ‘Die
Diktatur des Reichspräsidenten nach Artikel 48 der Weimarer Verfassung’, in Schmitt, Die Diktatur (n. 47
above) 212-57. Hence, the positive constitutional rules that confer circumscribed powers of dictatorship to
a president (or some other constitutional office) cannot be the order that Schmitt, in Politische Theologie,
claims will remain even in a state of total exception. What remains in the state of total exception, according
to Schmitt, is not a legal order, but the distinction between friend and enemy.
In George Schwab’s English translation of Politische Theologie (Chicago: University of Chicago Press,
2005), Schmitt is made to say, on p. 12, that in a state of exception “order in the juristic sense still prevails,
even if it is not of the ordinary kind.” However, what Schmitt really wrote (in my translation) is that “since
the state of exception is still something other than a mere anarchy or chaos, an order in the juristic sense
still exists, though not a legal order” (Schmitt, Politische Theologie [n. 5 above] 18: “Weil der
Ausnahmezustand immer noch etwas anderes ist als eine Anarchie oder ein Chaos, besteht im juristischen
27
Note as well that the power to take a decision on the total exception is not a
legislative or adjudicative authority, in any accustomed sense of these terms. It does not
create legal norms or modify legal relationships between legal persons, and neither does
it apply norms. All it does, so to speak, is to switch the law as a whole on and off.54
Despite its nature as a de facto power, however, the power to take a decision on the total
exception is a power that affects the law, as it conditions the law's applicability. The
decision on the exception, then, is jurisprudentially relevant, even though it does not
legislate or adjudicate: Every ordinary application of the law, according to Schmitt,
presupposes a prior decision to the effect that the situation is normal and not exceptional
and that it is therefore possible and appropriate to rely on legality for the solution of
social conflict.55
These observations explain how Schmitt avoids the second horn of the dilemma
outlined above. Hart's attack on Austin's notion of sovereignty, as we have seen, is an
attack on the claim that a mere de facto power can become the source of all positive legal
norms by assuming the role of an un-commanded commander. Schmitt, however, is
Sinn immer noch eine Ordnung, wenn auch keine Rechtsordnung.”) What Schmitt wants to say here is clear
enough. What still exists in the state of exception is the state, understood as a political community (see the
discussion in n. 49 above), which, according to Schmitt, is a subject of juristic thought, but not the law,
political or otherwise. Schwab’s translation simply drops Schmitt’s explicit statement that there is no longer
a legal order in the state of exception, and replaces it with the invented claim that there is still a legal order,
but that it is not ‘of the ordinary kind’. Loughlin’s view that, according to Schmitt, ‘droit politique remains’
in the state of exception appears to be based on Schwab’s faulty translation. See Loughlin, Foundations of
Public Law (n. 4 above) 401.
54
See Schmitt, Politische Theologie (n. 5 above) 18.
55
See ibid. 18-9.
28
clearly not guilty of the mistake of conceiving of laws as sovereign commands. A
sovereign who does not legislate obviously does not issue commands that claim legal
authority. Hart convincingly argues, as we have seen as well, that Austin's theory fails to
explain the continuity of legal system and the permanence of legal norms. But since
Schmitt's sovereign is not a legislator and does not enact positive legal norms Schmitt's
theory simply does not give rise to the puzzles concerning continuity and permanence
that Hart so successfully deploys against Austin.
Let me explain these claims in a little more detail. Schmitt's sovereign is perfectly
able to coexist, in times judged non-exceptional, with a working positive constitutional
order, perhaps even a liberal-democratic one.56 Such an order would determine a
procedure of legislation for the production of individual legal norms, norms which in turn
guide the activity of the courts and of administrative agencies. Hence, no positive legal
norm, under circumstances of normality, need be validated by recourse to sovereign
authority. A norm's validity, rather, will rest on the fact that it has been enacted in
accordance with the rule of recognition determined by the constitution. Since there is no
need for Schmitt to portray positive laws so validated as commands, his theory is well
able to accommodate Hart's insights into the importance of rules and rule-following and
into the diversity of legal norms.
Of course, Schmitt argues that all this presupposes that the sovereign judges the
general situation to be non-exceptional, so that law can apply. He also takes it that legal
norms, though validly enacted, will not apply as long as the sovereign judges the situation
to be exceptional. So perhaps a Hartian might argue that there is still room for a
56
See Carl Schmitt, Verfassungslehre (n. 38 above) 98-9.
29
continuity-puzzle of the sort that Hart deploys against Austin to arise. Imagine that the
sovereign suspends legality altogether, in a global state of emergency, and that legality is
later re-established, after the sovereign, through the use of de facto force, has managed to
produce a situation he judges to be normal. Should we say that the legal system after the
state of emergency is the same as the one that was in force before the emergency? Or
should we say that a new legal system has been created? Questions of this sort will not
embarrass Schmitt, for the simple reason that Schmitt will answer such questions in line
with the criteria that a Hartian would apply to judge of questions of continuity.
A sovereign's actions in a state of exception, presumably, could lead to two
different results. A sovereign might re-establish the social condition that underpinned the
old constitution, so as to make it possible, once again, to apply the law of the old
constitution. Else, a sovereign's actions could lead to the establishment of a new
constitution. This, presumably, will tend to happen if the old constitution has been made
obsolescent by social change, so that it is no longer possible to preserve the condition of
normality that underpinned it.57 In the first of these two scenarios Schmitt would claim
that there is legal continuity, while he would deny continuity in the second.58 Hart will
arrive at exactly the same result, since, in the first case, judges and officials are going to
continue to apply the old rule of recognition, while in the second they will follow a new.
Let me draw a preliminary conclusion of our discussion of Schmitt's theory of
sovereignty so far. Schmitt's theory of sovereignty seems to offer a coherent response to
the legal-theoretical claim that a sovereign authority above the positive law, an authority
57
See Schmitt, Verfassungslehre (n. 38 above) 21.
58
Of course, Schmitt thinks that the continuous existence of political community is not tied to legal
continuity. See ibid. 91-9.
30
that conditions the law's applicability, is incompatible with legal order. However,
Schmitt's argument clearly does not establish that a meta-legal sovereign authority is as
necessary to the existence of legal order as a backbone is to a man. There would appear to
be no good reason to deny that a society characterized by the absence of a Schmittian
sovereign could have a functioning positive legal system. Schmitt, then, has fought the
legal-theoretical criticism of sovereignty to a standstill. But where does that leave us?
It is crucial to Schmitt's argument, as we have seen, that the power of deciding on
the exception - though it is not legislative or adjudicative - be seen as a power that
conditions the applicability of positive law. If sovereign power is not interpreted as
conditioning the applicability of positive law, Schmitt's conception of sovereignty risks
becoming purely political and jurisprudentially irrelevant.
A Kelsenian would press Schmitt on precisely this point. The Kelsenian will
admit that a society might contain a person or group of persons who have a de facto
power altogether to suspend legality, to completely interrupt the normal operation of the
law. But why should this fact be jurisprudentially significant? As long as the law does not
recognize the power to take a decision on the exception, the existence of that power, or its
successful exercise, will be no more than a mere fact of political sociology. In making
this claim, one does not have to deny that politically powerful groups attempt, from time
to time, to interrupt the application of law, and that they sometimes succeed in breaking
legal continuity. Neither does one have to contest that a legal system can exist and
operate only where it is sufficiently effective, i.e. where its operation is not successfully
challenged by de facto powers aiming to suspend or block its application. If the view that
31
the power to take a decision on the exception conditions the applicability of law boils
down to the banal insight that law must be sufficiently effective to exist, how can that
power be portrayed as any kind of legal power, as a power that is of jurisprudential
concern?
To meet this challenge, Schmitt adds an important qualification to the view that
the sovereign decision conditions the applicability of law. He claims that it is the
legitimate applicability of the norms that belong to some positive legal system that
presupposes a sovereign who can decide on the exception.59 This response, if defensible,
opens a way out of the impasse we just pressed on Schmitt. The power to take a decision
on the exception must obviously be regarded as jurisprudentially relevant if the existence
of that power is indeed a necessary condition of the legitimate applicability of positive
law. But of course, Schmitt will now have to explain why a sovereign who can take a
decision on the exception is necessary for the existence of legitimate positive law.
Schmitt's answer to this question, in a nutshell, is that the sovereign's decision on
the exception expresses the political existence of a people. The sovereign decision, in
other words, is the only form in which popular sovereignty can be actualized, and popular
sovereignty, Schmitt argues, is the only modern basis for the legitimacy of law.60 It is to
the assessment of this normative claim that we must now turn to resolve the standoff.
59
See ibid. 87; Carl Schmitt, Legalität und Legitimität, 6th edition (Berlin: Duncker&Humblot, 1998).
60
See Schmitt, Verfassungslehre (n. 38 above) 77-82.
32
V. Schmitt's Sovereign and the Legitimacy of Legal Order
Schmitt's claim that a sovereign authority is necessary to secure the legitimate
applicability of law makes a first appearance, at least implicitly, in Schmitt's early work
Gesetz und Urteil, which was published in 1912.61 In this book, Schmitt is concerned to
outline the conditions of legal determinacy in judicial decision-taking. The starting point
of the argument is a rejection of a formalist picture of adjudication. The application of
statutes to particular cases by the courts, Schmitt argues, cannot be portrayed as a process
of logical deduction, in which general statutory rules clearly determine their applicative
instances. Rather, application will, in many instances, require a judgment to the effect
that a case can be brought under a concept, a judgment which is not itself guided by the
legal rule that is to be applied.62
Schmitt does not think, however, that the rejection of formalism should lead a
judge to embrace the self-conscious use of judicial authority as an instrument of social
reform. Schmitt holds that judicial decision-taking is legitimate only as long as it is not
dependent on potentially controversial moral or political judgments on the part of the
judge.63 The fact that legal determinacy cannot be ensured by a judicial commitment of
fidelity to statute does not imply, in Schmitt's view, that legal determinacy cannot be
achieved at all. Judicial practice, Schmitt claims, has developed an alternative means of
assuring legal determinacy: a shared sense of appropriateness, grounded in the common
61
Carl Schmitt, Gesetz und Urteil. Eine Untersuchung zum Problem der Rechtspraxis, 2nd edition
(München: C.H. Beck, 1969).
62
See Schmitt, Verfassungslehre (n. 38 above) 21-43.
63
See ibid. 42, 99.
33
educational background of legal officials, in the common experiences of those who hold
judicial office, and in the convergent ethical assumptions of members of the judiciary.64
If it is the social homogeneity of the judiciary that ensures legal determinacy, and
if legal determinacy is desirable, we will have to conclude that a justifiable judicial
decision is one that conforms to the expectations of other legal officials. This is why
Schmitt claims that a legal decision is correct if and only if we can assume that another
judge would have taken the same decision.65
Schmitt's argument in Gesetz und Urteil does not yet explain why legal
determinacy is desirable; so desirable, in fact, as to be declared the sole basis of a
standard of the correctness of judicial decision, and more desirable, apparently, than the
moral betterment of the law that might result from a less constrained use of judicial
authority in cases not determined by statute. It is possible, however, to gather an answer
to this question from Schmitt's later works. A legal decision not determined in advance by
a shared practice would amount to an instance of domination or, as Schmitt prefers to put
the point, to a form of ‘indirect rule’.66 The authority and institutional independence of
the judiciary, Schmitt frequently insists, depends on the presupposition that it does no
more than to apply the law. A judge who takes decisions conditioned by his personal
moral or political judgment, under the guise of applying the law, exercises a form of
arbitrary and unaccountable political rule. Not surprisingly, Schmitt harbors a strong
64
See ibid. 68-114.
65
See ibid. 68-9.
66
See for the notion of ‘indirect rule' Carl Schmitt, Der Leviathan in der Staatslehre des Thomas Hobbes.
Sinn und Fehlschlag eines politischen Symbols (Stuttgart: Klett-Cotta, 1982) 99-118.
34
suspicion against judicial review.67 But at the same time, he argues that an unrestrained
parliamentary legislator is in danger of becoming a mere instrument of the illegitimate
rule of partial interests that have found ways to corrupt the legislature.68 From Schmitt's
point of view, both of the fundamental positions in contemporary normative
constitutional theory - a rights-oriented constitutionalism arguing for judicial review as
well as a political constitutionalism concerned to protect the democratic legislature from
judicial interference - fall equally short of preventing domination.
The reason, according to Schmitt, is that both these positions overlook the
importance of legal determinacy as a condition of the legitimate applicability of law,
including constitutional law. Liberal constitutionalism will have to admit that judicial
decisions that enforce constitutional rights have the character of undemocratic judicial
impositions, as they will often turn out to be controversial, unexpected, and not grounded
in an established judicial practice that is already accepted as appropriate by the
community at large.69 Political constitutionalism, or what Schmitt calls the legislative
state,70 must own up to the fact that parliamentary decision-taking, especially in a
67
See Carl Schmitt, Der Hüter der Verfassung (n. 38 above) 22-48; Carl Schmitt, ‘Das Reichsgericht als
Hüter der Verfassung', in Carl Schmitt, Verfassungsrechtliche Aufsätze aus den Jahren 1924-1954 (Berlin:
Duncker&Humblot, 1958) 63-109.
68
See Carl Schmitt, Die geistesgeschichtliche Lage des heutigen Parlamentarismus, 2nd edition (Berlin:
Duncker&Humblot, 1926) 41-63; Schmitt, Der Hüter der Verfassung (n. 38 above) 73-91; Schmitt,
Legalität und Legitimität (n. 59 above) 19-37.
69
See Carl Schmitt, ‘Grundrechte und Grundpflichten', in Schmitt, Verfassungsrechtliche Aufsätze (n. 67
above) 181-231, at 217-24; Carl Schmitt, ‘Die Auflösung des Enteignungsbegriffs', in Schmitt,
Verfassungsrechtliche Aufsätze, 110-23.
70
See Schmitt, Legalität und Legitimität (n. 59 above) 19-28.
35
pluralist society, need not reflect the will of the people and that it may well come to
benefit partial interests over the common good. In either case, the appeal to the
constitutionality of the decision - be it judicial or legislative - is merely going to paper
over the fact that genuine democratic self-determination remains unrealized.
Schmitt does not hold that such failure of self-determination is an inevitable result
of judicial review or of parliamentary legislation. But if there is no failure, he argues, the
reason must be that the judicial or legislative decision takes place in the context of a
shared social understanding of what is appropriate that makes decisions both expectable
and acceptable. Legal determinacy, in other words, indicates the absence of conflict
between social groups, and hence the absence of domination of one social group by
another. Determinacy, however, is based on the presupposition of social homogeneity, a
presupposition that cannot be taken for granted under modern social conditions and that
cannot be guaranteed or protected by the law itself.71 Once a decision is controversial, we
must conclude that determinacy, as the condition of the legitimate applicability of the law,
no longer obtains. And whenever the condition of homogeneity is unfulfilled, decisiontaking under legal or constitutional procedures must turn out to be dominating to some.
The rule of law will become an instrument of the indirect rule of a part over the whole
and veil it at the same time. Where decisions that take place within the constitutional
system have become deeply controversial, social homogeneity must first be restored in
order to make the law legitimately applicable. This restoration, Schmitt argues, requires a
71
See Schmitt, Politische Theologie (n. 5 above) 13-21.
36
sovereign decision on the exception as well as dictatorial action in the state of
exception.72
We already know what the sovereign decision on the exception does. It declares
the law as a whole to be non-applicable. As Schmitt makes clear, the decision on the
exception thereby opens the space for a sovereign dictatorship that operates without any
legal restraints of any kind and that uses its unbounded discretion to create a situation of
normality or homogeneity; if necessary by the use of force that eliminates dissent.73 This
activity need not be conservative. It may well turn out to be revolutionary. He who
decides on the absolute exception also decides what is to be regarded as normal or
exceptional, and thus defines what kind of homogeneity is to be brought about.74
It would be wrong, however, to think of the sovereign's decision as a one-way
imposition of authority. For the sovereign to be successful in the attempt to define
normality, the decision must express some widely shared substantive identity that is prior
to the law and to the state as a legal expression of community. This identity will become
political, as Schmitt argues in The Concept of the Political, only if a sufficient number of
members of a society are willing to fight and die for the defense of that identity against
those whom they perceive as its internal and external enemies.75
Schmitt's famous criterion of the political is clearly motivated by the aim to
portray the constitution of political community as a process that does not involve legality.
72
See Schmitt, Die Diktatur (n. 47 above) 95-167.
73
See ibid.
74
See Schmitt, Politische Theologie (n. 5 above) 19.
75
See Schmitt, Der Begriff des Politischen (n. 49 above) 26-8.
37
Schmitt insists, in The Concept of the Political, that we cannot define ‘the political' with
reference to the state, for example as the fight for control over the state.76 The state, as we
have seen, is inextricably bound up with law. If sovereignty and political community are
to be prior to law, then sovereignty and political community must be explicable in terms
that do not make reference to the state. It must be possible to explain what a political
community is, what distinguishes it from other kinds of community, without taking resort
to the implicitly legal notions that are the building blocks of our idea of the state:
representation, authorization, or the idea of a social contract. This is why Schmitt claims
that political community is defined by a pre-legal distinction between friend and enemy.77
In successfully dividing society into those who support or reject a decision on the
exception, and the definition of normality implied by it, the decision on the exception
proves that some shared identity has political quality. It does that by forcing people to
take sides; to reveal themselves as friends or enemies, in a space outside the law, relative
to the identity highlighted by the sovereign as a marker of political community.78
A group exists as a political community, Schmitt concludes, as long as (and only
as long as) it remains capable of taking a decision on the exception and thus to determine
its own political identity.79 The decision must be the group's own, not one imposed on the
group, whether by force or fraud, by its internal or external enemies. Since every decision
on the exception draws a boundary between insiders and outsiders, we can also conclude
76
See ibid. 20-6.
77
It is also why Schmitt emphasizes that an exercise of constituent power isn't bound to any particular
legal form. See Schmitt, Verfassungslehre (n. 38 above) 82-7.
78
See Schmitt, Der Begriff des Politischen (n. 49 above) 46-7 on the internal enemy.
79
See ibid. 51-4.
38
that a group will enjoy existence as a political community only as long as it has the power
to determine its own membership. It follows that those who would attempt to subject a
political community to some normative standard of inclusion must be seen as attacking
its very existence. And a political community, Schmitt argues, must take itself to have a
right to reject and to repel their proposals.80
In making this claim Schmitt does not put forward a view about the instrumental
value of political community to its members. He does not argue, say, that membership in
a political community that enjoys unrestricted self-determination, including the right to
redraw its boundaries as it sees fit, is necessary to realize the good life of individuals, and
that we should therefore recognize each political community's right to determine its own
identity. Any such argument would raise the obvious question why it should be morally
permissible for the decision on the exception to preemptively exclude those who are
defined by the sovereign as internal enemies from the community.
Schmitt addresses himself only to those who already see each other as the true
members of some pre-legal political community, or who can be brought so to see
themselves, and who already accept the claim that nothing should be allowed to thwart
that group's political existence.81 Schmitt's aim is simply to raise awareness, amongst
readers or listeners who fit that description, of the danger that an unconditional
commitment to legality may threaten the existence of the political community to which
they take themselves to belong, and which they hold to be supremely valuable. Or to put
the point slightly differently: Schmitt is trying to stop his audience from confusing their
80
See ibid. 45-51.
81
See ibid. 27.
39
political community with their state. The latter is a legally constructed entity whose rules
are likely to give some sort of standing, perhaps even citizenship, to some who do not
truly belong to the political community, and who undermine the homogeneity that is the
necessary condition of legal determinacy.82 Hence, one should recognize that the concrete
will of political community, as manifested in the decision on the exception, must have the
power to prevail against the rule-based authority of the state.
In raising awareness of the conditions of the preservation of political community,
Schmitt takes himself to be defending democracy. Pretending to take his cues from
Rousseau, Schmitt defines democracy as the identity of ruler and ruled, and he argues
that the successful sovereign decision on the exception is the most perfect realization of
that identity.83 In taking a decision on the absolute exception, a pretender to sovereign
authority proposes to a group of people to define their political identity in a certain way.
But he will be successful, as already pointed out, only if a sufficient number of the
addressees concur with the sovereign's distinction between friends and enemies, and are
motivated to draw the requisite practical conclusions from it. Hence, the sovereign
decision on the exception, if successful, is really a communal decision. Through that
decision a group manifests its existence as a political community, in the willingness of its
members to treat some characteristic that they share as a political identity, an identity for
which one must fight and die.
82
See Schmitt, Der Leviathan (n. 66 above) 61-78.
83
For Schmitt's theory of democracy see Schmitt, Die geistesgeschichtliche Lage (n. 68 above) 13-20, 30-
41; Schmitt, Verfassungslehre (n. 38 above) 223-38.
40
A successful sovereign decision on the exception, Schmitt suggests, is the
paradigm case of collective self-government.84 The sovereign, in taking the decision on
the exception, is not a representative whose will is imputed by fiction to all those who are
regarded as members of an artificial social body, though their individual wills may differ
from that of the sovereign. The successful sovereign decision, rather, is the concrete
manifestation, according to Schmitt, of the real, the unanimous will of a people; the only
one there can be.85 And it is only as long as the identity that finds political expression in it
is preserved, in the day-to-day business of constituted, legally regulated politics, that the
law remains legitimately applicable.86
We are now in a position to offer an assessment of Schmitt's normative claim.
Does Schmitt present a convincing argument for the claim that the legitimate
applicability of law requires the existence of a sovereign authority?
It is not difficult to see why Schmitt's theory of popular sovereignty is
problematic. Schmitt claims that the sovereign's decision on the exception perfectly
realizes the democratic identity of ruler and ruled. But this is so only for the trivial reason
that any successful decision on the exception draws the boundaries of political
84
See Schmitt, Verfassungslehre (n. 38 above) on the contrast between exercises of constituent power and
constituted politics.
85
See ibid. 20-23, 76, 84, 87; Schmitt, Der Begriff des Politischen (n. 49 above) 27.
86
Schmitt's critique of the Weimar constitution claims that the revolution of 1918 had failed to bring about
homogeneity. See Schmitt, Verfassungslehre (n. 38 above) 28-36, 106-9. Once the Weimar Republic
entered its political crisis, Schmitt argued that a renewed exercise of sovereign authority might be called for
to solve the problem. See Schmitt, Legalität und Legitimität (n. 59 above) 82-91.
41
community in such a way as to remove from the polity all those who do not follow and
support the sovereign's interpretation of its identity. The identity of ruler and ruled that
manifests itself in the successful decision on the exception is an artifact of antecedent
exclusion, in a space outside of law, of all interesting political difference and all deep
dissent.
The view that democracy is the identity of ruler and ruled is normally understood
to raise a more interesting claim. Rousseau, for instance, argues that democratic
institutions, attitudes, and practices will allow citizens who differ in their private interests
and their social identities, to benefit from the realization of common interests, and to do
so in a way that ensures a proper respect for the individual freedom of each. For
Rousseau, democratic identity results from the reasonable acceptability to all of laws
produced by a legal and constitutional system that successfully implements the ideal of
civic equality.87 In Schmitt's view, on the other hand, the identity of ruler and ruled is no
longer a goal to be pursued through the instrumentality of settled democratic political
practice. Rather, a democratic legal and constitutional system is said to presuppose a
political identity that it cannot itself establish or protect, and that must therefore be
created through the prior dictatorial homogenization of society. The identity of ruler and
ruled is no longer realized in a general will but in a particular will that is opposed to the
‘encrustations' of rule-based legality.88 For Schmitt, democratic politics in the state of
normality does little more than to raise a danger of the corruption of the people's
antecedent identity. The most that we might ever be able to claim in its favor, in
87
See Joshua Cohen, Rousseau: A Free Community of Equals (New York: Oxford University Press, 2010).
88
See Schmitt, Politische Theologie (n. 5 above) 20-1.
42
circumstances where the sovereign sees no need to take a decision on the exception, is
that it does no harm.
What seems to start with the laudable aim to achieve non-dominating law thus
turns into a paean for a collectivist version of the right of the stronger. It may well be true
that judicial or legislative decisions, in a society that has been made perfectly
homogeneous by sovereign dictatorship, would no longer imply a danger of a domination
of the ruled, for the simple reason that judicial and legislative decisions could no longer
be very controversial among the rulers and the ruled. But this result, to repeat, is achieved
only through the prior violent repression and exclusion of all ethical and cultural diversity
in the space of exception. The determinacy that results from perfect homogeneity does
not make law legitimate as much as it turns the legitimacy of law into something that can
now be dispensed with, for the reason that all those who are addressed by law, as well as
those who make it, share a political identity and will never disagree about the wisdom of
any important decision. All others, by definition, are enemies who are outside the polity.
The question of how to treat them is not one of legitimacy but of power-politics.
Let me emphasize that this critique of Schmitt's theory of popular sovereignty is
not primarily a moral critique. The point is not that Schmitt's conception of popular
sovereignty is to be regarded as morally incorrect because it licenses disregard of
individual rights or minority rights, however true that may be. The point is that Schmitt's
argument, though it claims to ground the legitimacy of law in an appeal to popular
sovereignty, is self-defeating as an account of the legitimacy of law.
Imagine you question the legitimacy of the laws of your society, laws that the
sovereign, together with the majority of the members of your community holds to be
43
legitimately applicable. It would be useless, presumably, for someone to tell you that the
laws are legitimate because they express the identity of the people. The fact that you raise
a complaint already establishes that you do not belong to the people. If you did belong to
the people, on the other hand, you would not raise the complaint, and no explanation of
the legitimacy of the law would have to be given to you. Appeals to the legitimacy of the
law or the constitution, even if based on the notion of constituent power, become
meaningless in Schmitt's framework. So whatever Schmitt thinks he has shown, he
cannot have shown that the existence of a sovereign authority is a condition of the
legitimate applicability of law.89
It will likely be objected that this criticism misunderstands Schmitt's claim that
the legitimate applicability of law requires a sovereign authority. Schmitt, it might be
argued, is not concerned with a situation in which the positive law does express the
people's identity, but with a case where it does not. If law does not express the identity of
the people it will run counter to and frustrate the people's collective self-determination,
and thus be illegitimate. This is what happens, for instance, where a parliamentary
legislature is captured by partial interests that have the power to bring about legislative
decisions which frustrate the true will of the people, as expressed in a past decision on the
exception. The claim that the legitimate applicability of law requires the existence of a
89
For a fuller statement of this argument see Lars Vinx, ‘The Incoherence of Strong Popular Sovereignty',
in International Journal of Constitutional Law 11 (2013) 101-24.
44
sovereign authority is to be understood as the claim that such a situation can only be
prevented where there is a sovereign who can switch the law off.90
This reinterpretation of the claim that sovereignty is a precondition of the
legitimate applicability of law concedes that Schmitt is not concerned to offer reasons to
a dissenting minority why it should defer to the law though it rejects the law's content. He
is now portrayed as concerned, rather, to warn the majority against accepting the
deliverances of the positive legal system's procedures as final. In doing so, the majority
would alienate its collective autonomy to a positive legal system that provides legal and
perhaps political standing and influence to minorities or interest groups that do not share
the majority's political identity. The reason why a sovereign authority is needed, then, is
not that a sovereign is necessary for the constitution of authoritative law, of law that binds
even those who criticize its content. Rather, a sovereign authority is necessary to make
sure that the results of legal procedures cannot prevail against the will of the majority, of
those who claim that they truly belong to the people.
One is inclined to reply that, strictly speaking, it makes no sense for Schmitt to
argue that the positive law might frustrate the political self-determination of a people.
According to Schmitt, a political community exists if and only if it has the capacity to
take a decision on the exception. But in that case the law can not be an impediment to
self-determination, because the sovereign's decision on the exception can switch it off. If,
on the other hand, a group is unable to bring about a decision on the exception it simply
does not exist as a political community or a people and it consequently can not make any
sense to complain that the group's self-determination is impeded by law.
90
For this perspective see Paul W. Kahn, Political Theology (n. 3 above).
45
Admittedly, this reply is a little too quick. Schmitt's writings on sovereignty
during the Weimar-era address a situation in which it is unclear, not least to Schmitt
himself, whether there still is a sovereign authority. Put differently, they address a
situation in which it is unclear whether the German people still exists as a political
community in Schmitt's sense. The answer to these questions must depend on whether the
German people, suitably led, are still capable of bringing about a decision on the
exception that will, if necessary, prevail against the corrupted positive legality of the
Weimar constitution. And whether the German people are still capable of bringing about
a decision on the exception depends, in turn, on whether they are willing to support such
a decision, in light of the conviction that to do so is a precondition of the preservation of
their collective autonomy. Schmitt's theory of sovereignty, then, is perhaps best seen as an
exhortation to Germans to be willing to take the decision on the exception.91
This alternative reading of Schmitt's claim that the existence of sovereign
authority is a condition of the legitimate applicability of law, in contrast to the one we
already rejected, is not self-defeating. But it is also devoid of jurisprudential interest.
We can admit that if political existence is understood in Schmitt's way, it will
follow that a political community must always be ready to put aside its commitment to
legality if it wants to preserve itself. We can also admit that if political existence, as
Schmitt defines it, is desirable, we should fight against an ideology of legalism that
wishes to make the law out to be the final arbiter of all social disputes. But what is all that
to the law? Why should the law, faced with Schmitt's conception of political existence,
91
See Schmitt, Die geistesgeschichtliche Lage (n. 68 above) 77-90 on the need for a nationalist political
myth.
46
abandon its own claim to normative finality? Why should it recognize the Schmittian
sovereign's decision on the exception as a legally relevant decision, as a decision about
the legitimate applicability of the law? The law, as far as I can see, would have to
recognize the sovereign's decision as legally relevant only if the existence of an authority
that can take a decision on the absolute exception was necessary for the law to be able to
achieve its own essential purposes. That, however, does not seem to be the case.
Arguably, a system of law must be backed up by a state to be sufficiently effective
to achieve whatever essential purposes it might have. That was one of the key claims of
the classical doctrine of sovereignty. But as should be clear, the law may very well have
the backing of a state even where there is no sovereign capable of taking a decision on
the absolute exception. The presence of a sovereign, moreover, is not just unnecessary to
secure the law's effectiveness. It is also likely to frustrate the laws essential purposes, on
some of the most plausible accounts of what these might be. In recognizing the
sovereign's decision as legally relevant, the law would, for example, have to betray the
aim that is peculiarly its own according to the classical discourse of sovereignty: namely
the aim to subject social conflict to peaceful arbitration and to suppress the employment
of violence not licensed by the law.92 We arrive at the same result, obviously, if we follow
Fuller and take the law's essential purpose to be the establishment of an inviolable rule of
law that will make exercises of power predictable to those affected.93 To maintain the
claim that the existence of a Schmittian sovereign is necessary to allow the law to achieve
its essential purposes one would, it seems, have to adopt the rather silly view that the
92
See Hobbes, Leviathan (n. 7 above) 117-21.
93
See Lon L. Fuller, The Morality of Law, revised edition (New Haven: Yale University Press, 1964).
47
essential purpose of law consists in not getting in the way of the decision on the
exception.
Legal positivists who question the view that the law has essential purposes are
unlikely to arrive at a different conclusion. In recognizing the legal relevance of the
Schmittian sovereign, the law would, for instance, betray its claim to Razian authority,
since the law, to claim authority, must take its own decisions to be final.94 And if we deny
that the law necessarily claims authority, or perhaps that it has any nature at all, there is
simply nothing left to build on in trying to establish the jurisprudential relevance of the
decision on the absolute exception. Schmitt is not in a position to adopt an instrumental
conception of law and to argue that the law's essential purpose is to serve whatever goal
the sovereign decides to pursue. The sole purpose of sovereign action, according to
Schmitt, is to create homogeneity, and Schmitt is firmly committed to the view that this
goal can only be achieved through dictatorial action freed of all legal restraints. Schmitt's
conception of sovereignty, I conclude, turns out to be too purely political to be of any
jurisprudential relevance.
Schmitt's basic claim is that legality can never make any positive contribution to
the legitimate settlement of profound social conflict. He holds, as we have seen, that
whenever the answer to a political question is not determined by a background of social
agreement, its arbitration in a legally regulated form must involve a hidden exercise of
domination under the guise of the rule of law. The liberal idea that suitably constructed
94
See Joseph Raz, ‘Authority, Law, and Morality’, in Joseph Raz, Ethics in the Public Domain. Essays in
the Morality of Law and Politics, revised edition (Oxford: Oxford University Press, 1995) 210-37.
48
legal procedures could ever lead to the fair settlement of political conflict must
consequently be a form of false consciousness; a consciousness that must be fought
because it might stop the majority, those who truly belong to the people, from asserting
their identity against a minority that relies on the rule of law to thwart the strong and to
subject them to the ‘indirect rule’ of the weak. Schmitt's point, of course, is not that the
rule of the strong is more justifiable, from a moral point of view, than the rule of the
weak. His point is simply that, as a member of the majority, one should reject the
constraints of legality on the majority's power once one has seen them for what Schmitt
thinks they are: the impositions of alien groups whose members ought to be regarded as
enemies, and be done away with, in the interest of securing substantive homogeneity.
I hope it is clear that it would be a grave mistake to regard Schmitt's purported
rehabilitation of the doctrine of sovereignty as a continuation of the classical discourse of
sovereignty. In that discourse, sovereignty and law are seen as essentially arbitrative.
Hobbes's sovereign creates unity, and thus peace and security, through representation, not
through antecedent exclusion. His will displaces the wills of those who enter into a social
contract, it must be owned, as Hobbes says, even by those who disagree with the wisdom
of the sovereign's decisions.95 But this displacement, as Schmitt himself loudly
complained in his book about Hobbes, does not destroy the individual will, to fuse it into
a true collective identity based on a friend-enemy distinction.96 It only excludes or
preempts it for the time being, for as long as an individual has reason to prefer sovereign
protection to the danger of the state of nature. Sovereign representation, thus, turns
95
See Hobbes, Leviathan (n. 7 above) 120, 124.
96
See Schmitt, Der Leviathan (n. 66 above) 79-97.
49
plurality into unity without eliminating difference. A modern liberal-democratic state,
despite the fact that it does not contain a personal sovereign, does much the same,
provided it has the capacity to finally settle all social conflict.
In Schmitt's theory of sovereignty, by contrast, sovereign representation, as a
principle of political unity, is replaced with pre-legal exclusion, and the very possibility
of political difference within a legally constituted and pacified political unity is denied.
Or to be more precise: Schmitt's legal theory deliberately attempts to create an attitude, in
those to whom it addresses itself, that will make political difference within legal unity
impossible. This political project, if I am correct, has little to teach us about the nature of
law and the conditions of its legitimacy.
The critique of Schmitt presented in this paper does not imply that we ought to
abandon the doctrine of sovereignty. My conclusion is more limited: After the breakdown
of Hobbes's unstable synthesis of personal and institutional sovereignty, the doctrine of
sovereignty must take its start from the insight that sovereignty is an attribute of the state
or perhaps of the state's law, as Kelsen claimed. Whether the doctrine of sovereignty is
still jurisprudentially relevant as a doctrine about the state or about the law is a question
that remains to be answered. As I suggested above, I do not think that Hart's criticism of
Austinian sovereignty even addresses the issue. Perhaps a general Staatsrechtslehre is
still possible, if we make the effort to free our understanding of sovereignty of the
theological and anthropomorphic baggage that, as Kelsen argued, has tended to force it
into an inadequate theoretical straightjacket. What I have tried to show is that Schmitt's
attempt to conceive of sovereignty as a personal or ‘concrete’ authority that is prior to
both state and law is a dead end. Contemporary defenders of popular sovereignty, and
50
those searching for foundations of modern public law, are well advised to look for a
different source of inspiration.
51