Schmitt’s denial of the justiciability of domestic conflict

Carl Schmitt frequently claimed that both constitutional and international law are forms of political law and consequently cannot be applied or guaranteed, at least not fully, by courts.Footnote ⁵ According to Schmitt, both constitutional and international law rest on the presupposition of a situation of normality that can only be brought about and guaranteed by purely political action.Footnote ⁶ This view implies that the law and legal institutions are powerless actively to promote order, peace, and justice or to make a positive contribution to the solution of social or intercommunal conflict. The rule of law can gain a foothold only where all non-trivial conflicts have already been overcome in some other way.

Schmitt, to be sure, did not put forward the implausible claim that it is literally impossible for a constitutional legislator or for members of the international community to create judicial institutions formally tasked with applying international or constitutional law. Schmitt’s point, rather, is the more subtle one that such juridification could not possibly be legitimate; at least not if there are profound disagreements, among social groups or members of international society, as to how the constitutional or international law that is to be applied is to be understood. Far from realizing the ideal of the rule of law, any full juridification of constitutional or international law will inevitably undermine the separation of law and politics and lead to politically unaccountable ‘indirect rule’ under a veil of legality.Footnote ⁷

The following quotation from The Guardian of the Constitution, a work in which Schmitt, in response to Hans Kelsen’s call for the creation of a constitutional court in Weimar Germany,Footnote ⁸ attacked the legitimacy of constitutional review, contains a characteristic expression of this line of thought:

The demand for an ‘adjudicative settlement of all political questions’, Schmitt argues, results from a ‘misunderstood and abstract conception of the rule of law’ and should therefore be rejected.Footnote ¹⁰

Schmitt repeats this assessment in a number of other key works, throughout the course of his whole career. It would therefore be wrong to suppose that the passage quoted above is unrepresentative of Schmitt’s general constitutional theory. The deep hostility towards constitutional review is clearly a leitmotif of Schmitt’s work as a constitutional theorist.Footnote ¹¹ What is more, it found practical expression when Schmitt became involved in the most important case of constitutional adjudication that took place in the Weimar Republic; the decision of the Weimar Staatsgerichtshof on the legality of the so-called ‘Preussenschlag’.

On 20 July 1932, the conservative federal government led by Franz von Papen, authorized by an emergency decree signed by president Hindenburg, deposed the social-democratic government of the state of Prussia and appointed federal commissioners to bring the powers of the Prussian government under the control of the central government. The emergency decree authorizing the measure appealed to Article 48 of the Weimar constitution, which gave the president the power to force states to fulfil their constitutional duties towards the Reich with the use of armed force, and to take all measures necessary to restore public safety and order. The Prussian government denied that it had violated any constitutional duties or that it was unable to secure public order in Prussia, and asked the Staatsgerichtshof in Leipzig, a tribunal empowered by Article 19 of the Weimar Constitution to hear cases concerning conflicts between the federal government and the individual states, to issue an injunction against the Preussenschlag. While the court refused to issue such an injunction, it did initiate a full trial, to determine whether the Preussenschlag had been constitutional or not.Footnote ¹²

Schmitt was appointed as one of the Reich’s legal representatives in Leipzig and was centrally involved in preparing the government’s defence.Footnote ¹³ Schmitt’s line of argument focused on the question whether the Staatsgerichtshof was in a position to second-guess the president’s judgment that public order and security in Prussia were threatened so that a federal intervention had become necessary. In his closing statement in court, Schmitt admitted that the Staatsgerichtshof could be called a ‘guardian of the constitution’. But he emphasized that the court’s guardianship ought to be limited to questions properly subject to judicial decision. To decide whether a threat to public order and security requires the use of dictatorial powers, in Schmitt’s view, is a political question, and thus not for courts to decide. Hence, the president’s use of Article 48 ought to be regarded as the exercise of a specifically political guardianship of the constitution not subject to the scrutiny of the courts. Since the dictatorial powers of the president under Article 48 in Schmitt’s view undoubtedly included the power to suspend the government of a state if necessary, he suggested that the court should not have let the case come to trial, for the reason that the commissioners appointed by the federal government ought to have been recognized as the authentic representatives of the state of Prussia.Footnote ¹⁴

This argument did not fully win out. The court upheld the legality of a temporary assumption of the executive powers of the Prussian government by the Reich, on the ground that public order and security had indeed been threatened in Prussia. But the court rejected the claim that the Prussian government had violated its constitutional duties towards the Reich and held that the Prussian government could not be stripped, under Article 48, of its power to represent the state of Prussia in the Reichsrat, i.e. in the federal process of legislation. This compromise failed to satisfy Schmitt as well as other prominent legal observers, and it did little to stabilize the Weimar Republic in the face of its national-socialist enemies. While the decision put the actions of the president under a cloud of legal suspicion, it did nothing to practically reverse the conservative coup d’état in Germany’s largest and most important state.

However, the diagnoses as to why the court had botched the case differed widely. Hans Kelsen argued that the court should not be held responsible for its hesitant and unsatisfactory decision. The root of the problem, as Kelsen saw it, was the defective character of the constitution itself, which had failed to create a robust framework of constitutional review, and which had given the president far too unrestricted powers of emergency. These constitutional deficiencies, Kelsen argued, had been caused by the tendency of German jurisprudence to draw a sharp distinction between legal and political questions, in order to exempt the latter from judicial control.Footnote ¹⁵ In Kelsen’s view, this distinction is meaningless, since most legal conflicts in constitutional or international law clearly have a political dimension, while so-called ‘political conflicts’ usually turn out to be amenable to legal arbitration if the parties to a dispute prefer such arbitration to the resort to violence or coercion.Footnote ¹⁶ The fathers of the constitution, then, ought to have created a court with stronger powers of review, instead of clinging to the authoritarian fiction that the nature of political conflict inevitably renders a constitution partly unenforceable.

This was decidedly not Schmitt’s perspective, in whose view there had already been too much constitutional review. In a retrospective assessment of the causes of the downfall of the Weimar Republic published in 1935, Schmitt portrayed the trial in Leipzig as the culmination of a process of constitutional decay induced by liberal constitutional thought.Footnote ¹⁷ Liberal constitutionalism, in Schmitt’s view, abstracts from the substantive and pre-legal political identity of a people that can alone provide constitutional legitimacy and thus turns the constitution into a mere mechanism of positive legal rules that can no longer distinguish between friend and enemy. The attempt to subject the Preussenschlag to judicial review, Schmitt argued, was the logical endpoint of the process of political neutralization inspired by liberal ideology and at the same time its reductio ad absurdum. From the point of view of liberal constitutionalism, which demands that all exercises of political power be subject to legal control, the Staatsgerichtshof was not in a position to refuse to pass judgment on the president’s actions under Article 48 of the Weimar constitution. But since a court’s legitimacy, in Schmitt’s opinion, inevitably rests on the claim that it merely applies the law, and does not take political decisions, the judges were not in a position to exercise political leadership and thus had to refuse a clear decision. The completion of the liberal demand for full judicial control of the state thus came to coincide with the end of any possibility of genuine political leadership that might have saved the crumbling constitution. Constitutional review, Schmitt mockingly concludes, is the crowning achievement of a ‘bourgeois constitutionalism that destroys any possibility of genuine government’.Footnote ¹⁸

Given this deep hostility to judicial review, it should not occasion surprise that Schmitt’s argument in The Guardian of the Constitution is formulated in a thoroughly essentialist manner. Schmitt claims that some questions in constitutional or international law are political in their very nature and can never be a proper subject matter for adjudicative settlement. Moreover, Schmitt tells us that a political question cannot be turned into one that is properly subject to adjudicative settlement simply by endowing a court with the formal authority to decide it. To do so would leave the political nature of the question unaffected, and thus do nothing more than to undermine the separation of law and politics.Footnote ¹⁹

What, then, are the characteristics of political questions? And how do they explain the essential non-justiciability of such questions? The first place to look for an answer to this question, of course, is Schmitt’s analysis of the political in The Concept of the Political. As is well known, Schmitt claims that political community is grounded in the distinction between friend and enemy. A political community is a social group of a very special kind. It is defined not merely by a shared identity, but, additionally, by the willingness, on the part of the members of the group, to defend the group’s existence as a self-governing entity at the risk of their own life. In other words, a political community is a group – and, for every one of its individual members, the only group – that has the capacity to order its members to kill or to sacrifice their own lives in the group’s fight against non-members. In successfully drawing a distinction between friend and enemy, a political community manifests its own existence, while at the same time defining its membership.Footnote ²⁰

Schmitt’s understanding of the political expresses the conviction that unrestricted self-determination is an essential feature of political community. A social group exists as a political community only as long as the group decides for itself who its enemies are. Hence, a group that submits to a friend–enemy distinction drawn by an external power, in exchange for a promise of protection or in reaction to a threat of force, forfeits its political existence. If its members are unwilling to continue to endorse a friend–enemy distinction that is the group’s own, and to risk their life in the fight against those the group has declared to be enemies, the group will lose the status of political community.Footnote ²¹

A political question, accordingly, must be a question that concerns the very existence of a social group as a political community and thus raises the spectre of the friend–enemy distinction. This is a purely formal conception of what it means for a question to be political. As Schmitt himself is at pains to point out, the intensity of the political and the friend–enemy distinction can combine with very different markers of collective identity. Groups initially defined in religious, economic, moral, or ethnic terms can all attain the status of political groups. It is therefore impossible, Schmitt argues, to offer a substantive criterion of the political. Any question can become political if a group judges it to concern its own political existence.Footnote ²²

Schmitt’s understanding of the political in The Concept of the Political explains why political questions are essentially non-justiciable. If political communities exist only in the mode of full self-determination, and if self-determination is to be understood as a group’s capacity to draw a friend–enemy distinction that is its own, and to decide for itself whether it is necessary, in a given case, to use force to defend its own existence, it will follow that political communities, to maintain their existence, must insist that political conflicts cannot be subject to the compulsory arbitration of third parties. The judgment as to whether a conflict is political, and thus non-justiciable, can only be taken by the group itself as subjection to the judgment of third parties in this matter would forfeit political existence.Footnote ²³ In a truly political conflict between two hostile groups, the only available options for third parties are to either take sides or to avoid getting involved in the conflict at all. This rules out the possibility of legitimate judicial decision. A court called upon to decide a truly political conflict will either fail to be impartial or it will have to avoid a clear decision.

Note that this argument carries implications not just for international politics but also for the domestic constitutional sphere. A community, on pain of losing its own political existence, cannot alienate its power to draw a friend–enemy distinction to a written constitution, construed as a system of positive legal norms.Footnote ²⁴ Schmitt clearly argues that a political community, to maintain its existence, and to protect its identity, must have the capacity to declare members of society, and not just members of other political communities, to be enemies that should be denied the protection of the law.Footnote ²⁵ A modern written constitution, in a liberal and democratic state, however, is obviously unlikely to make explicit allowance for that possibility. Rather, it will typically presume that all those who are legally recognized as citizens must be afforded the protection of the rule of law, regardless of whether they fully identify with the identity endorsed by the majority of the members of society. This demand, in turn, normally involves the further presumption that social conflicts within society can and must be resolved through the proper employment of the constitution’s rules, whether in the legislative or the judicial process.

If a written constitution, on the basis of these presumptions, is taken to block the state from drawing a friend–enemy distinction to deal with a situation of internal crisis it will, from Schmitt’s point of view, become a danger to the existence of political community. The attempt to enforce the legalistic presumptions of liberal constitutionalism, by subjecting all internal disputes to the judgment of courts, amounts to an implicit denial of the political community’s power to protect the political expression of what it takes to be its identity as it sees fit. The existence of a sovereign power capable of taking a decision on the total exception, in the name of the political community, is therefore a necessary condition of a stable and legitimate legal order, Schmitt argued as early as 1922, even where such a power is not provided for by positive constitutional law.Footnote ²⁶

It might be objected that this interpretation of Schmitt’s view of internal conflict is at odds with the constitutional theory put forward in Constitutional Theory. Schmitt appears to adopt a constructive stance towards liberal constitutionalism in this key work, with a view to developing a conception of liberal constitutionalism mindful of the underlying conditions of the stability of liberal democracy. This impression, in my view, is misleading.

Schmitt’s argument in the Constitutional Theory rests on a fundamental distinction between two different understandings of the constitution. According to what Schmitt calls the ‘relative concept of constitution’, the constitution is nothing but the set of legal norms that we can identify as constitutional laws, on the basis of their being protected by a special rule of amendment in the written constitution. Schmitt argues that this concept of constitution is deeply implausible for two main reasons. To begin with, it is perverse to claim that the special status of constitutional norms is a result of the fact that they enjoy special procedural protection against legislative change. A plausible conception of constitution should argue the other way around, and invoke the substantive importance of constitutional laws to explain and justify their entrenchment. Moreover, to come to Schmitt’s second point, the relative concept of constitution abstracts from the constitutional relevance of constituent power as a source of constitutional legitimacy. The reason why the constitution can legitimately be treated as a higher law, in Schmitt’s view, is that the constitution is an expression of the constituent power of the people. The relative concept of constitution does not acknowledge this fact, and thus lacks the resources to justify constitutional entrenchment.Footnote ²⁷

To avoid these problems, Schmitt introduces a ‘positive concept of constitution’.Footnote ²⁸ This concept conceives of a constitution as the concrete form of collective existence that an already existing political community chooses for itself. The act of constitution-giving, then, does not primarily consist in the making of particular legal norms with a special procedural status. Rather, that act is to be understood as a popular sovereign’s identity-defining choice for a basic blueprint of social and political order. The nature of this choice will usually be expressible in a number of basic constitutional principles. In the case of the Weimar Republic, for instance, the German people, according to Schmitt, decided for a democratic, republican, federal, and parliamentary political system committed to the rule of law.Footnote ²⁹ These basic principles, however, are not mere constitutional laws, i.e. particular legal norms contained in the written constitution and protected by an amendment clause. They stand above any mere constitutional law and define a constitutional core or essence that Schmitt claims is shielded from constitutional amendment. Though the Weimar Constitution did not explicitly exempt any of its provisions from amendment, Schmitt argued that it would be unconstitutional to use the constitutional procedure of amendment to, for example, turn Germany into an absolute monarchy or a communist society. For constituted powers to do so, even if acting with a supermajority, would be to usurp the political community’s prerogative to determine (and to re-determine) its concrete form of existence.Footnote ³⁰

The positive concept of constitution opens the door for Schmitt’s peculiar understanding of constitutional guardianship. If the positive constitution is the main object of constitutional guardianship, constitutional guardianship cannot consist in the attempt to ensure constitutional legality, i.e. conformity of exercises of public power with particular constitutional laws contained in the constitution. Rather, constitutional guardianship must guarantee that decisions taken by constituted powers do not violate the fundamental principles that define the positive constitution. We might be inclined to think that this conception of guardianship should lead to a justification for judicial review, as protection of fundamental constitutional principles against parliamentary majorities. But Schmitt argued that the positive constitution cannot be guarded by a court, on the grounds that legitimate adjudication must always apply particular laws with a determinate content. Schmitt therefore held that the task of guarding the positive constitution is a political and not a judicial task, and that it must, therefore, fall to the president, making use of his powers of dictatorship under Article 48.Footnote ³¹

In order to show that courts cannot legitimately apply constitutional principles, Schmitt does not directly rely on the friend–enemy conception of the political. In The Guardian of the Constitution (as well as in other works directly concerned with the legitimacy of constitutional review), he offers an alternative account of what makes a question political, on the basis of an argument from disagreement that resembles, at first glance, a currently popular argument against judicial review associated with the work of Jeremy Waldron.Footnote ³²

Schmitt’s concrete discussion of the problem of constitutional review might be described as an attempt to limit the implications of Justice Marshall’s famous argument in Marbury v. Madison.Footnote ³³ Schmitt does not appear to deny the truth of the key premise of that argument: He admits that constitutional norms constitute a higher law that governs the legality of ordinary legislative action, such that ordinary laws that conflict with constitutional norms must lack legal validity. What is more, Schmitt seems to agree with Marshall’s view that a judge’s duty to decide cases arriving in his court in accordance with the law would require a judge not to apply an ordinary law that violates a constitutional norm and thus lacks legal validity.

What Schmitt is concerned to claim in The Guardian of the Constitution is that this argument only suffices to justify an extremely restricted form of judicial review. Schmitt plausibly maintains that the question whether an ordinary law violates a constitutional provision will sometimes be difficult to answer. To be sure, there might be instances in which the conflict between a constitutional law and an ordinary law will be glaringly obvious. But we should expect that there will be cases in which even knowledgeable and reasonable observers who judge in good faith will disagree with respect to the question whether some ordinary law conflicts with a constitutional provision. Marshall’s argument, Schmitt claims, can only apply to trivial cases of obvious conflict between constitutional norms and ordinary laws, and it therefore fails to establish the legitimacy of judicial review in the harder cases that are characterized by reasonable disagreement over the meaning of the constitution.Footnote ³⁴

Schmitt’s reasons for so restricting the implications of Marshall’s argument are based on a peculiar, extremely narrow understanding of what it means for a judge to decide in accordance with or on the basis of the law. Though he elsewhere emphasizes the idea that all judicial decisions contain a moment of pure discretion,Footnote ³⁵ Schmitt comes close, in The Guardian of the Constitution, to denying that judges can ever have any legitimate decisional authority:

According to the view outlined here, a judge, strictly speaking, cannot legitimately decide. All a judge can legitimately do is to execute or apply decisions already taken by someone else. Of course, such avoidance of judicial decision-taking is possible only if the execution or application in question does not require any interesting form of practical judgment that might give rise to disagreement.

Another way to put the same point is to say that judges, in Schmitt’s view, necessarily lack legitimate authority. According to the way the term ‘authority’ is usually understood in contemporary legal and political philosophy, a person or institution A has legitimate authority over another person or institution B if and only if A’s decisions in some matter bind B, within certain limits, regardless of whether B thinks the decision was substantively correct.Footnote ³⁷ Needless to say, courts are usually taken to possess authority, so understood, with respect to the interpretation of the law. However, Schmitt seems to deny that courts can have authority, at least where it seems most relevant, namely within the sphere of disagreement about how to interpret the law. Wherever the results of applying the law are perfectly ‘measurable’ or ‘calculable’, decisional authority won’t be needed to justify the results of application. But wherever authority would have to be invoked to justify the results of application, because these results are no longer fully ‘measurable’ or ‘calculable’, courts, in Schmitt’s view, have no business to interfere, lest the rule of law be undermined.

Schmitt’s overall conclusion concerning the legitimacy of constitutional adjudication, then, comes to this: Unless the conflict between a constitutional norm and an ordinary law is so trivial as to obviate the need for potentially controversial judgment, judges must not be allowed to strike down or not to apply ordinary laws suspected of unconstitutionality. If the case is not altogether trivial, a judgment that denies the constitutionality of some ordinary law will require political decision-taking, and political decision-taking, Schmitt claims, cannot legitimately pertain to the courts.Footnote ³⁸

Schmitt’s argument from disagreement does not apply merely to the courts. Modern critics of judicial review, like Schmitt, hold that judges lack authority to decide on the interpretation of the constitution in the sphere of disagreement, but they typically argue that the constituted parliamentary process has the authority to do so.Footnote ³⁹ For Schmitt, this approach to the problem of disagreement about constitutional issues is foreclosed. Legitimate legislative activity, for Schmitt, is as dependent on the presupposition of normality as legitimate judicial activity, since a legislator will be unable to formulate general laws that are applicable by judges in a measurable and calculable way if a society is affected by profound practical disagreement. Every general law will have to contain vague general terms that will lead to determinate results in particular cases only if there is a cohesive practice of interpretation that does not give rise to practical disagreement. Hence, Schmitt claims that both the system of legislation as well as the courts can work properly only relative to an already existing background of practical agreement or, to use Schmitt’s term, of homogeneity.Footnote ⁴⁰ In a deeply pluralistic society, Schmitt argues, majority voting among party-affiliated representatives is likely to do nothing more than to empower a part of society to decide on a question that concerns the whole while following its sectional interest.Footnote ⁴¹ A legislative decision on the meaning of the constitution must then be as illegitimate as judicial review.

Now clearly, even if both Schmitt’s arguments against judicial review and his arguments against the ‘parliamentary system’ of ordinary legislation are sound, they won’t make the problem of disagreement go away. So don’t we need an exercise of authority at some point? It seems that Schmitt’s attacks on judicial review and on the ‘parliamentary system’ only defer that exercise, by denying authority to the courts or parliament. But where does the buck stop, with respect to the authoritative settlement of social disagreement?

In one sense, the answer to this question is of course obvious. In general, Schmitt argues that the buck stops with the sovereign who can decide on the exception, or, put differently, mobilize the community to draw a friend–enemy distinction, and use extralegal means to bring about what he defines as a situation of normality. In the particular context of Schmitt’s interpretation of the Weimar Constitution, constitutional guardianship is taken to fall to the president: The president, even while he does not have the power to change the positive constitution, and is thus not strictly speaking a sovereign, does have the power, under Article 48, to suspend all constitutional laws and to use dictatorial means to restore a situation of normality adequate to the core of the positive constitution. The president’s plebiscitary legitimacy, and his alleged neutrality in party-political conflict make him the only constituted power, in Schmitt’s view, that can speak on behalf of the popular sovereign and that can interpret and safeguard the positive constitution. The dictatorial powers he wields under Article 48 are best understood as a residue of the sovereign dictatorship exercised by the constitutional assembly that drafted and enacted the Weimar Constitution.Footnote ⁴²

This positive conception of executive constitutional guardianship, however, does not solve the problem of the source of legitimate decisional authority. Recall that an authoritative decision does more than just to end a conflict. It purports to bind those to whom it is addressed regardless of whether they agree that the outcome is substantively correct. In other words, to exercise authority is not a way to create a situation of normality, at least not if normality is understood to involve the absence of disagreement. Rather, successful exercises of authority will make normality or homogeneity of opinion dispensable. Where authority is successfully claimed and exercised, people will recognize that they are bound to a collective decision even if they disagree with its wisdom.

Schmitt’s president is not exercising authority, so understood. To create or maintain normality, Schmitt’s president will suspend the rule of law and rely on dictatorial means. As should be clear, the decision on the exception establishes a distinction between friend and (internal or external) enemy. So in effect, the sovereign’s decision on the exception will declare those who disagree to be enemies.Footnote ⁴³ But a president who draws a friend–enemy distinction does not even claim to have authority over the enemy. He does not claim, in other words, that those he declares to be enemies ought to recognize his commands as legitimate exercises of authority which bind them to obedience. Rather, the president – to use Hobbes’s phrase – returns the dispute to the sword to eliminate a threat. In drawing the friend–enemy distinction, he is excluding the enemy from the political community. Such exclusion signals the end of a political relationship of belonging that might give rise to duties of deference to public authority on the part of the dissenter, while it also deprives the excluded of any form of legal protection.Footnote ⁴⁴ Neither does the president, in drawing the friend–enemy distinction, claim authority over those he considers to be friends. Friends will follow the president’s lead because they agree with the way in which his decision conceives of the community’s substantive identity, and not because they attribute to the president a power to take decisions about the community’s identity that would bind even if they were wrong. But this is really just another way of claiming that the president does not have authority, in the sense defined above, and that his leadership can never do more than to express and politically mobilize a group’s prior agreement on the nature of its substantive identity.

At the end of the day, Schmitt believes that the kind of constitutional disagreement that cannot legitimately be tackled by the courts – and this includes, as we have seen, very much all interesting constitutional disagreement – can only be dealt with by treating the dissenter (which, practically speaking, means the weaker party) as an enemy. When it comes to the authoritative settlement of social disagreement, the buck stops nowhere, since Schmitt simply does not believe that deep social conflicts can ever be settled through exercises of legitimate authority. They can only be removed through extralegal violence. The successful exercise of such violence, one that creates sufficient homogeneity, is a necessary condition of legitimate or sufficiently ‘calculable’ legality, but it cannot itself be subjected to the rule of law.

Let me now come to the promised assessment of Schmitt’s argument against the legitimacy of constitutional review. Recall that Schmitt’s criticism starts out from the seemingly plausible assumption that judges lack the authority to take controversial political decisions and that one cannot turn a question that is political into a question that is properly subject to adjudication simply by giving a court the competence to decide it. But it now seems that judges lack the legitimate authority to settle political conflicts created by practical disagreement only because everyone does, and I am inclined to think that this shows that something has gone wrong.

Let me try to pinpoint the problem more precisely. Schmitt’s understanding of what it means for a question to be political, and thus to be non-justiciable, is beset by ambiguity. On the one hand, Schmitt explains what it means for a question to be political with reference to the argument in The Concept of the Political. According to this conception of the political, a question is political if and only if it is taken to concern the very existence of a political community. However, this understanding of what makes a question political differs from the one that Schmitt is working with in The Guardian of the Constitution. In the latter work, as we have seen, a question is taken to be political whenever its solution requires potentially controversial practical judgment.

These two conceptions of a political question do not appear to be coextensive. Members of a political community, clearly, can disagree about how their constitution is to be understood, and disagree deeply, while keeping the peace and recognizing the authority of a court or of a constituted legislature bindingly to interpret the constitution. This would seem to leave considerable space for judicial (or, for that matter, legislative) exercises of authority and thus to undercut Schmitt’s claim that non-trivial social conflicts can never be settled through legally or constitutionally defined procedures. To make his case against constitutional adjudication and for executive constitutional guardianship, Schmitt must therefore identify the two senses of what makes a question political. He must pretend, implausibly, that every constitutional dispute that gives rise to profound social disagreement inevitably marks a friend–enemy distinction and thus calls for the exercise of dictatorial power.

To be sure, the settlement of disagreement through constituted procedures will not always be free from elements of discretion or even arbitrariness. Schmitt is correct to argue that the judicial settlement of constitutional conflict will not typically take the form of an automatic application of determinate law with calculable outcomes. However, few authors today believe that arbitration of non-trivial practical disagreement in judicial form can never be legitimate. Legal theorists, whether of a Hartian, Dworkinian, or legal-realist stripe, have long abandoned the view that the application of the law in modern liberal democracies can be portrayed in formalist terms. But even those who take this to imply that judges are frequently engaged in interstitial legislation do not usually draw the conclusion that judicial decisions in hard cases must, therefore, automatically be illegitimate.Footnote ⁴⁵ And it is surely even less plausible to deny all capacity legitimately to deal with political disagreement to the constituted legislative institutions. Even those who – like Jeremy Waldron and Richard Bellamy – deny the legitimacy of constitutional review, and prefer social conflicts to be settled by a parliament, would therefore almost certainly reject Schmitt’s radical scepticism about constitutional procedure.

Of course, these observations leave us with real questions as to where to draw the line between legitimate and illegitimate judicial or legislative arbitration of political conflict. Schmitt, however, is not very interested in addressing those questions, since it is not really his intention to offer a viable theory of the limitations of legitimate judicial or legislative power. Rather, by conflating the two senses of what makes a question political, Schmitt wants to shift as much power as possible to the executive, in order to give the executive elbow-room to engage in homogenizing dictatorial action. As I have shown, this tendency should not be understood as merely a context-bound reaction to the terminal crisis of the Weimar Constitution. It is rooted in Schmitt’s theory of politics and is characteristic of all his works in legal and constitutional theory.

A defender of Schmitt’s view might reply that there can be no agreed-upon criteria for deciding whether a political conflict constitutes an existential threat to political community and therefore requires extralegal action. If it is admitted that some conflicts are political in that sense, and, consequently, not amenable to judicial arbitration, a political community, or political leaders acting in its name, must be entitled to draw a friend–enemy distinction and to take a decision on the exception. The view that a question must be regarded as non-justiciable whenever the executive decides that it is will then be unavoidable.

This reply is based on a non sequitur. That it is possible for societies to fall into civil war does not entail that a good constitution ought to endow the executive with a standing power to declare civil war, preemptively, on those who are perceived as dissenting from a community’s substantive identity. It is of course imaginable that a community might come to understand its own identity in such a way that the majority of its members will experience all dissent from the community’s momentarily prevailing ethos as a mortal threat to their collective existence that can only be violently eliminated if the community is to survive. If a community so defines itself, the two senses of the political that I distinguished above will indeed coincide, as Schmitt claims, and its members are likely to be willing to accept executive dictatorship as a standing constitutional possibility.

However, most of us would surely find this ideal of community highly unattractive. If constitutionalism and the rule of law are to be more than empty words, they must at the very least imply that those who are citizens, under declared constitutional laws, and who behave law-abidingly, cannot simply be disenfranchised or deprived of legal protection through executive acts, on the ground that their opinions, ideals, or way of life are perceived as a threat to a community’s prevailing substantive identity. A liberal and democratic community betrays its own ideals if it does not trust itself to be able, at least within certain limits, to renegotiate its own identity in peaceful and lawful form.

One might argue that this option was no longer available in the late Weimar Republic. Responsibility for this condition, however, must rest in large part with those who, like Schmitt, were all too keen to narrow down the sphere in which legal conflict-resolution might have been able to operate, and to resist the establishment of a fully developed system of constitutional review. The failure of the Staatsgerichtshof, as Kelsen rightly pointed out, did not result from a failure on the part of court to remain mindful of the essential limits of legitimate judicial activity. It was the result, rather, of the success of the regressive constitutional ideology advocated by Schmitt.

Schmitt’s denial of the justiciability of international conflict

Schmitt’s writings throughout the Weimar period manifest a strong concern with the problem of the rule of law in international conflict. The context of Schmitt’s initial interest in this issue is the political reorganization of Europe after the First World War. The treaty of Versailles was widely believed to be illegitimate in Germany, and the governments of the Weimar Republic, regardless of ideological orientation, were all committed to the goal of revision of the treaty. There was disagreement, however, over the question whether such revision should be pursued peacefully, through Germany’s integration into the international legal framework established after the war, or through a later use (or threat of the use) of military force.Footnote ⁴⁶

Schmitt’s major early writings on international legality are concerned with two specific questions that arose in this context. First, Schmitt wrote extensively, and from a highly sceptical perspective, on the question whether Germany should become a member of the League of Nations and be willing to support the mechanisms for the judicial arbitration of international conflict envisaged by the covenant of the League.Footnote ⁴⁷ Second, Schmitt developed trenchant criticisms of the international initiatives for the legal prohibition of aggressive war, and urged that Germany resist those initiatives.Footnote ⁴⁸ In both cases, Schmitt’s point of view was strongly influenced by the assumption that international law is closely analogous to constitutional law and subject to the same essential limitations.

Schmitt’s assessment of the League of Nations provides a good illustration of how he perceived the analogy between constitutional and international law. The Core Question of the League of Nations, first published in 1926, the year in which Germany became a member of the League, concludes with a fundamental criticism of international juridification that matches the argument against constitutional review almost word for word. Expressing doubts concerning the wisdom of Germany’s entry into the League of Nations Schmitt writes:

Once again, we are told that there are political questions that are inherently non-justiciable. To subject such questions to a judicial procedure would only mix up law with politics and thus necessarily lead to illegitimate results. Schmitt goes so far as to say that ‘political justice’ does not just lack legitimacy, it is also unjust. And the injustices that must result from an attempt to subject all international disputes to legal regulation, Schmitt adds, would be of enormous magnitude. This assessment concludes a general critique of the League of Nations, a critique which is based on the view that the conditions of the legitimacy of an international regime that aspires to the legal regulation of the use of force among states must be very similar to the conditions of the legitimacy of a domestic constitution. The core question concerning the League of Nations, in Schmitt’s view, is the question whether it is a real union (in German, whether the ‘Völkerbund’ is a genuine ‘Bund’) or only a looser international organization that facilitates certain forms of coordination among sovereign states.Footnote ⁵⁰

Schmitt’s discussion of this question should be read in the context of the theory of federalism offered in Constitutional Theory.Footnote ⁵¹ This theory takes up a question that had greatly exercised the constitutional theorists of the Wilhelmine Empire, namely whether the unified German state that had been created in 1871 ought to be regarded as a Staatenbund, a federation of independent states retaining their individual sovereignty, or as a Bundesstaat, a sovereign federal state divided into a number of subordinate constituent states that are no longer sovereign. Schmitt argues that neither of these two options captures the essence of a true Bund or union.

The first key characteristic of a union is that its constituent states have at least partially given up their ius ad bellum. On the one hand, they no longer retain the power to make war against each other, but are committed to settling all their disputes with one another peacefully and on the basis of the constitutional law laid down in a treaty of union. On the other hand, the government of a union acquires the right to draw friend–enemy distinctions that are binding on all constituent states, while the latter are bound to support any war declared by the federation against external enemies. Nevertheless, to come to the second key characteristic of a union, the reason for which constituent states enter into a treaty of union is to preserve their political existence as independent states. For this reason, every true union, Schmitt argues, must guarantee the preservation, and in particular the territorial integrity, of all constituent states.

These two key features of a true union, Schmitt claims, stand in tension with one another. The possession of a ius ad bellum, as we have seen, is the definitive mark of political existence.Footnote ⁵² This would appear to imply that, in a true union, it is the union and not its constituent states that has political existence. However, Schmitt insists that it would be wrong to assimilate the union to a centralized state with a single sovereign. Such a state, after all, would not have to regard its own division into constituent states, or the separation of competences between the central and the local level, as anything more than an administrative convenience, open to change on the basis of the unilateral decision of a unitary sovereign. It is therefore essential to a true union, in Schmitt’s view, that the question of ultimate sovereignty be permanently deferred or left open.

Such deferral, Schmitt argues, is possible only on the condition that the constituent states of a union exhibit sufficient political homogeneity to rule out the possibility of violent conflict within the union. In particular, it is necessary that all members of a union mutually recognize their respective forms of government and their territorial possessions as legitimate and normal. On the condition that there is sufficient homogeneity, it will become possible to settle conflicts among constituent states through judicial procedures, on the basis of the treaty of union. The background of agreement will make the process of interpreting and applying the treaty’s provisions sufficiently calculable to count as proper adjudication, and the process can be expected not to adversely affect the vital interests of any constituent state.Footnote ⁵³ Schmitt’s condition for the legitimacy of a supranational rule of law, in other words, is the same as the condition of the legitimacy of a domestic constitution, namely the antecedent removal, by extra-legal means, of the potential for serious internal conflict and political disagreement among the states associated by a treaty of union.

The Covenant of the League of Nations, in Schmitt’s view, deliberately fails to settle the question whether the League is or is intended to become a real union. If the covenant was to be understood as constituting a real union, Schmitt argues, German accession to the League of Nations would have to be interpreted as implying a recognition of the legitimacy of the territorial status quo in Europe. For Germany to offer such recognition would be acceptable, Schmitt holds, only if the League of Nations exhibited the homogeneity of a real union, and if it offered Germany an opportunity to pursue what Schmitt takes to be its right to territorial revision within a shared constitutional framework capable of producing genuinely collective decisions. This condition, however, is unsatisfied, as Germany’s traditional enemies would never allow it to employ the mechanisms of the covenant to pursue revision of the status quo. Schmitt concludes that the League is nothing but a charade by which the Western powers seek to deprive Germany of what he takes to be its legal and moral right to forcibly revise a status quo that is patently unjust and too abnormal to carry a legitimate legal order capable of judicial enforcement.Footnote ⁵⁴

The parallels of Schmitt’s argument against international legalization to his attack on constitutional review should be obvious enough. Schmitt does not argue that the problem of disagreement is typically even deeper in international politics than in domestic politics, and he does not claim that the institutions of international law are presently defective in some remediable way. Schmitt claims, rather, that the international rule of law is in principle incapable of making any positive contribution to the solution of non-trivial international disputes. Just as a judicially enforced domestic constitution, if it is to be legitimate, can do no more than to express an agreement or a homogeneity that has already been brought about through non-legal means, legitimate international adjudication can only ratify the results of a shared perception of normality arrived at through political means.Footnote ⁵⁵ Schmitt holds that any other use of international legality for the binding settlement of international conflict can be nothing more than a veiled instrument of oppression.

It is now time to ask whether this sweeping thesis should be considered more acceptable in the international than in the constitutional case. I will argue that it is not. Schmitt’s attack on international legality rests on the same unconvincing assimilation of all political disagreement to existential conflict that undercuts his argument against the domestic rule of law.

What are the catastrophic injustices that Schmitt thinks must inevitably result from the binding judicial settlement of non-trivial international conflict, as envisaged by the Covenant of the League of Nations? And how can a non-legalized international politics avoid those injustices? As far as I can see, there are three different lines of argument in Schmitt’s writings that are meant to answer these questions: the claim that a scheme for the judicial settlement of international conflict may come to protect a substantively unjust status quo, the claim that such a scheme threatens political self-determination, and the claim that it will only fuel violent international conflict instead of limiting it.

The first of these three worries certainly deserves to be taken seriously. It fails, however, to establish that binding international adjudication is an inherently illegitimate method of conflict-resolution. Admittedly, an international legal order that outlaws the use of force to bring about revision of the legal status quo may come to protect a substantively unjust status quo that merely enshrines the morally contingent outcome of an earlier resort to force. This problem will likely be particularly pressing if that order also tends to foreclose peaceful revision of the legal status quo, by endowing those who benefit from the status quo with a power to veto legal change. Under such conditions, the attempt to make resort to international adjudicative institutions compulsory in all cases of international conflict would, by itself, fail to provide an adequate remedy against legal injustice.Footnote ⁵⁶

Schmitt was certainly not alone in perceiving the legal order created by the Treaty of Versailles and the League of Nations in this light, though more cautious critics argued for the adoption of a forward-looking approach that would accept the legitimacy of the status quo so as to give Europe the chance of a peaceful future.Footnote ⁵⁷ But let us grant to Schmitt that legally unauthorized resort to force might be morally legitimate in extreme cases, where there is no other option for a community to defend its vital interests or for bystanders to help protect basic human rights. This concession, as recent debate on the legality of international uses of force has made clear, does not necessarily imply that a legal regime that may, in exceptional cases, have to be violated on moral grounds ought to be abandoned or fundamentally changed. Even where there is sufficient moral reason to break a legal rule in a particular case, it may well be better to continue to uphold that legal rule.Footnote ⁵⁸ Schmitt, in any case, does not want to restrict the legitimate use of force to exceptional instances, in which it might be the only way to overturn a severely unjust legal status quo. Rather, he argues that international law ought to continue to recognize that a state’s decision to go to war should under no circumstances be held to be illegal.Footnote ⁵⁹ There seems to be little reason, to put it charitably, to expect that this alternative would produce results more substantively just than any conceivable framework for the judicial settlement of international conflict. If the legal status quo is likely going to reflect unjust or morally contingent distributions of power, then so, a fortiori, are the results of the exercise of an unrestricted ius ad bellum.

Instead of criticizing the legal status quo as unjust, Schmitt’s second argument claims that international legalization is incompatible with political self-determination. The Schmitt of The Concept of the Political argues, as we have seen, that in order to enjoy true political existence, a community must consider itself to be free from any international legal obligation that purports to bind it against its own continuing will. This requirement has both a material and a procedural aspect. On a material level, a community’s willingness to consider itself bound to norms of ius cogens – norms from which a derogation is not permitted, so that treaties violating them have no legal forceFootnote ⁶⁰ – would be incompatible with political self-determination, since such subjection would curtail a community’s freedom to express its own political identity in whatever form it sees fit. It would establish the indirect rule of those who determine the content of ius cogens. On a procedural level, a community, if it is to enjoy genuine self-determination, must refuse to accept subjection to any regime of international legal norms of which it is not itself the ultimate interpreter or from which it is not permitted to withdraw unilaterally by invoking its vital interests. This much would seem to follow from Schmitt’s committment to the view that a people’s sovereignty cannot be alienated through the consent of its constituted governmental institutions. Footnote ⁶¹

It should give us pause, however, that many contemporary states, including some that we would surely not hesitate to describe as politically independent, seem to fail Schmitt’s standard, for the reason that they recognize norms of ius cogens, including the prohibition of legally unauthorized resort to war, and support the cause of building international institutions endowed with conclusive authority. So Schmitt must explain, to make his second criticism work, why it would be mistaken for jurisprudence to operate with a less radical notion of political self-determination that does not take independence to be incompatible with subjection to a system of international law enforced by international courts. However, Schmitt’s argument from self-determination, by itself, has very little to say about the issue. Schmitt insistently claims that a community that no longer takes its own decision on friend and enemy, unhampered by constraints of legality, no longer exists as a political community. A community’s attempts to defend its political existence in whatever way it sees fit, Schmitt adds, do not stand in need of normative justification. Such matters, we are informed, are ‘existential’ and not ‘normative’.Footnote ⁶²

The problem with this line of argument is that it begs the question against the proponent of international legalization. The argument assumes that there is no meaningful way to give a less demanding definition of political existence or independence than Schmitt is willing to countenance, for example as a legal status of sovereign equality conferred and protected by international law.Footnote ⁶³ But this assumption is precisely what is at issue between Schmitt and the proponent of international legalization, and Schmitt’s argument from political existence simply does not speak to the question. It presupposes the claim that legal institutions can never legitimately settle any non-trivial practical disagreement, that all subjection to law must really be subjection to some other community’s dominating will, a view that has already turned out to be implausible in the constitutional context. Or, to put the point in a slightly different way, Schmitt’s argument from self-determination assumes that we already agree with his conception of the political, a conception that defines a community’s political existence in terms of the possession of an unrestricted ius ad bellum and in terms of non-subjection to binding international (or constitutional) law. Hence, the argument will hold no attraction to those who think that Schmitt failed to give a convincing account of the nature of the political.

Schmitt, to be sure, does put forward a further consideration that would, if it were sound, support the conclusion he wants to draw from his conception of self-determination. Schmitt cautions against the introduction of a ‘discriminating concept of war’Footnote ⁶⁴ or, in other words, against the attempt to outlaw aggressive war as a legitimate instrument of policy.

Attempts to outlaw aggressive war, culminating in the Kellog–Briand Pact, played a prominent role in the diplomacy of the inter-war period. This political movement towards outlawing aggressive war was supported, moreover, by important theorists of international law, for example by Hans Kelsen and Hersh Lauterpacht.Footnote ⁶⁵ Both Kelsen and Lauterpacht emphatically rejected the claim that any part of the conduct of states should be seen as exempt from the law on the ground of a distinction between legal and purely political questions. If international law is to be regarded as a genuine legal system, it must claim to regulate all international use of force. Any use of force in the international sphere, in other words, ought to be interpreted either as a sanction against a violation of law or as a punishable delict. Whether a use of force is justified as a sanction or amounts to a delict is a legal question that can, in principle, always be decided by a court, since a resort to force must amount to a delict, once the assumption of completeness is accepted, where the opposing state has been found not to have violated its international obligations. Kelsen and Lauterpacht consequently advocated the subjection of international conflict to the binding adjudication of a system of international courts, and argued that the limits of international legalization are not set by inherent limitations of legality but by the political will of the sovereigns.

The introduction of a discriminating concept of war, Schmitt held against Kelsen and Lauterpacht, is not going to pacify international politics. Rather, it is only going to make war more total and destructive. Schmitt’s argument sets out from a historical reflection, and claims that the public international law of the Westphalian epoch was highly successful in limiting and restraining the destructive consequences of warfare. Though the period of classical early-modern statehood was by no means pacific, the belligerent states tended to observe the restraints of the ius in bello rather carefully. This limitation of the consequences of warfare, Schmitt argues, depended on a non-discriminating concept of war. Classical public international law did not draw a distinction between those who had a just cause to go to war and those who lacked it, at least none that had any legal consequences. Rather, all parties to a conflict, as long as they could lay claim to the status of sovereign statehood, were presumed to be legitimate belligerents.Footnote ⁶⁶

Schmitt goes on to observe that the attempt to subject the international use of force to the rule of law, through the demand that any use of force must either be regarded as a delict or as a sanction, will inevitably undermine the idea of mutually legitimate belligerency. The sanction–delict schema forces us to regard at least one of the parties to any violent international conflict as a violator of the law. The conditions for the legality or illegality of uses of force, Schmitt fears, will likely be drawn from the tradition of just war theory, so that the question of legality will become fused with the question of justice. As a result, those who claim to have the law on their side will, in case of armed conflict, feel entitled to loosen the restraints of the ius in bello in trying to overcome their allegedly unjust and criminal enemy. The de-legitimized opponent, knowing that he will be punished as a criminal in case he loses the war, will come to lack an incentive to observe the constraints of the ius in bello on his part. Hence, the attempt to subject the use of force to the rule of law will unleash a destructive dynamic towards total war.Footnote ⁶⁷

Schmitt’s claim that a discriminating concept of war is only going to make international conflict worse is clearly meant to suggest that the Western powers that advocated the introduction of such a concept of war were the real aggressors in the global conflicts of the twentieth century. When the vast majority of the members of the League of Nations, invoking the covenant, organized sanctions to prevent fascist Italy from successfully concluding its aggressive war against Ethiopia in 1935, some legal observers argued that non-members of the League of Nations not participating in the sanctions – a category that included Germany, which had left the League of Nations in 1933 – should not be allowed to invoke the traditional rights of neutrality in the dispute between Italy and the League of Nations.Footnote ⁶⁸

In his response to these arguments, Schmitt correctly observes that this position assumes that the coalition arrayed against Italy is acting on behalf of the international community of states, and that it consequently possesses the authority to take a decision concerning the legitimacy of Italy’s resort to force which binds all other states. The assumption of such authority, for Schmitt, is nothing more than a disguised claim to global hegemony on the part of Britain and France that carries an implicit denial of the independent political existence of the German people.Footnote ⁶⁹ To maintain its independence, Germany must, therefore, be willing to defend the classical, non-discriminating concept of war against the universalist pretensions of the Western Allies.Footnote ⁷⁰ The aggressions of Nazi Germany, Schmitt would argue in 1941, should therefore be seen as a legitimate attempt to create a new condition of normality that would be able to support a global version of the classical, non-discriminating concept of war that had governed the relationships of European sovereigns. The war, Schmitt argued, would establish a new territorial division of the earth into ideologically homogeneous spheres of hegemony, each controlled by a great power strong enough to continue to exist politically under the forbidding circumstances created by modern military technology.Footnote ⁷¹

Arguably, Schmitt’s polemic against the discriminating concept of war should not be rejected solely because Schmitt used it to justify Nazist aggression. However, Schmitt’s argument fails even if we focus on its substantial claims, as it blurs the distinction between two very different forms of discrimination in the assessment of the justifiability of the international use of force. On the one hand, a discriminatory concept of war might distinguish between legal and illegal uses of force. On the other hand, it might distinguish between substantively just and substantively unjust uses of force, on the basis of a moral theory of just war. Though Schmitt typically assumes that these two forms of discrimination will go hand in hand, it should be evident that they have a rather different ideological basis and may well lead to very different evaluations of one and the same international conflict.

According to the current legal regime, international uses of force not authorized by the UN Security Council are considered impermissible unless undertaken in self-defence. The point of this regime, one might plausibly argue, is not to establish a moral disqualification of one or another party in an international conflict that would allow for a loosening of ius in bello constraints. The point – much as in the case of the state’s claim to a domestic monopoly of legitimate power – is to stop violence from escalating and to provide room for a diplomatic or legal, and hence a peaceful, resolution of conflict.Footnote ⁷² The scheme for the settlement of international conflict envisaged by the Covenant of the League of Nations was based on the same idea, in that it tried to commit states to undergo a compulsory process of arbitration before resorting to war.Footnote ⁷³

Moral theorists of just war, by contrast, frequently claim that aggressive uses of force that are impermissible under current public international law may well be justified or even required on moral grounds and that the law, in such cases, should give way to the demands of justice. Such views have gained in popularity in recent years, in the context of attempts to justify Western military interventions in developing countries that cannot be grounded on the right of self-defence and that are not authorized by the Security Council. It is argued, for instance, that coalitions of morally superior democratic states should be permitted to engage in aggressive war at their own discretion and in disregard of current international law in order to protect human rights or to unseat morally unsavoury regimes.Footnote ⁷⁴ Some authors involved in the renaissance of moral theories of just war have explicitly defended the view, as predicted by Schmitt, that the moral discrimination between just and unjust causes for war on the level of ius ad bellum is incompatible with the idea of mutually legitimate belligerency that characterizes the traditional ius in bello. Those who fight without a just cause cannot have a moral right to kill their opponents, and it must therefore be morally wrong, at least from the perspective of ideal theory, to extend equal in bello-protection to both sides of a conflict and to shield those who fight without a just cause from punishment as long as they do not commit war crimes or crimes against humanity.Footnote ⁷⁵

Clearly, Schmitt’s attack on the discriminatory concept of war makes good sense if it is understood as a cautionary note on the renaissance of moral theories of just war, and on the idea that such theories should directly guide the international use of force. However, this does not show that it also makes good sense if understood as a critique of the attempt to subject international uses of force to the rule of law. The latter project, after all, is based on the conviction that no state should be allowed to use aggressive force at its own discretion, whether on the ground of reasons of power or interest or on the ground of purported moral concerns. Global legalism, insofar as it aims to prevent all legally unauthorized resort to force, may well be the best antidote to the dangers of a self-serving moralism of the powerful.Footnote ⁷⁶

None of Schmitt’s three attacks on the legitimacy of global legalism, I conclude, are fully convincing. This result should not occasion surprise. Schmitt’s real aim in attacking international legality is not to criticize the legal status quo after the First World War for being an unjust expression of mere power or to find a principled solution to the problem that international law has a tendency to bend to the hegemonic interests of the powerful. Schmitt evidently believed that Germany – at least if it was freed of what Schmitt described as its ‘spiritual subjection’ to liberal ideologyFootnote ⁷⁷– would be strong enough to maintain a true political existence, by drawing its own friend–enemy distinctions, in a world of power politics not subject to legal control. His denial of the very possibility of legitimate international legality under conditions of profound disagreement or conflict, then, was born of the same intention that guided his attempt to deny the possibility of constitutional conflict-resolution: To rally Germans to the fight against those whom Schmitt considered to be Germany’s internal and external enemies, by portraying all political conflict as existential conflict.

To admit that international conflict might be amenable to legal resolution, and to advise Germany’s constructive participation in the international legal order that could have come into being in the inter-war period, would have defeated Schmitt’s attempt to preserve political enmity by undermining the prospect of pacification through law. Schmitt’s position, as in the realm of domestic constitutional law, is not based on some profound insight into the inherent limits of legality. It is based, rather, on the questionable normative assumption that it is undesirable for politics to be subjected to the rule of law.

Conclusion: The analogy between constitutional and international law and modern scepticism about international law

Though I have been critical of Schmitt’s arguments against legalization, I do not contest the observation that there is an analogy between constitutional and international law. As Jack Goldsmith and Darryl Levinson have recently argued, both constitutional and international law are relatively indeterminate even today, in the sense that decisions taken under constitutional or international law continue to be less predictable and require more open moral and political judgment than decisions taken on the basis of ordinary municipal law. What is more, compliance with both contemporary constitutional and international law cannot be explained as a result of the fear of a powerful sovereign capable of enforcing the law against all those who are expected to conform to it. Finally, both constitutional and international law, Goldsmith and Levinson argue, still stand in tension with the idea of the sovereignty, insofar as they purport to constitute and to bind sovereign power.Footnote ⁷⁸ Given this contemporary support, it would be wrong to dismiss the analogy of constitutional and international law as a symptom of the underdevelopment of constitutional and international law in Schmitt’s time.

What is problematic about Schmitt’s view, I submit, is the normative prescription that Schmitt wants to draw from the analogy, namely to reject legal methods of conflict-resolution as inherently illegitimate whenever the condition of a perfect homogeneity of value and interest remains unfulfilled. Constitutional law has undoubtedly shown itself to be capable, in spite of the fact that it shares in the key features of international law, of dealing with at least some of the normative disagreement and conflict of interest that Schmitt argues must prevent legitimate legal process, and of doing so in a way that is obviously preferable to Schmitt’s alternative of executive dictatorship. Moreover, if non-trivial domestic conflict is in principle amenable to the rule of law, the analogy of constitutional and international law should lead us to expect that non-trivial international conflict is also, in principle, open to settlement under binding legal rules.

The analogy of constitutional and international law, then, supports the project of international legalization. If legalization is desirable where it is feasible, and if the analogy between constitutional and international law implies that the problems of feasibility that afflict global legalism do not differ in kind from those that afflict domestic constitutionalism, we ought to give support to the project of fully subjecting international politics to the rule of international law. Not to do so would be incompatible with our commitment to domestic constitutionalism, to the goal of subjecting all domestic conflict to the rule of constitutional law.Footnote ⁷⁹ To resist this conclusion, the modern sceptic about international legalization would have to embrace Schmitt’s view that constitutional and international law are equally incapable legitimately to resolve any non-trivial political conflict. He would have to deny the legitimacy of domestic constitutionalism along with the legitimacy of the rule of international law.

Of course, this argument invites the counter that Schmitt, for all his other faults, was quite right to read the analogy in the way he did. Though Goldsmith and Levinson, for instance, are careful not to draw very explicit normative conclusions from their claim that constitutional and international law are analogous, they do explicitly argue for a return to an Austinian perspective that regards both constitutional and international law as defective or non-paradigmatic forms of law, since neither is enforced by a meta-legal sovereign.Footnote ⁸⁰ Goldsmith and Levinson agree that it is wrong to claim that international law compares unfavourably with domestic constitutional law, but this is so, they argue, only because both are forms of ‘public law’, i.e. of a political law that purports to constitute and to bind sovereignty. Such law, they suggest, will always lack the determinacy, enforceability, and democratic legitimacy of the legal commands that an already existing popular sovereign issues to his individual subordinates. The conventional view that international law is a defective form of law, then, is wrong only insofar as it fails to recognize that constitutional law is equally defective. Such a reading of the analogy, needless to say, lends itself to the attempt to justify executive claims to strong extralegal powers, and some influential authors have in fact come to combine scepticism about international law with scepticism about constitutional law.Footnote ⁸¹

It would be uncharitable to suggest that contemporary attempts to read the analogy in an anti-constitutionalist key merely repeat the arguments of Schmitt’s that have been criticized above. Nevertheless, such attempts, much like Schmitt’s arguments, tend to move rather too quickly from the descriptive observation of the relative indeterminacy or the non-enforceability of compliance to the normative conclusion that constitutional and international law should not be regarded as comprehensive or conclusive.

As I have argued above, it is not at all clear why legal determinacy should be regarded as a necessary condition of the legitimacy of legal conflict-resolution. To hold that legal conflict-resolution can only be legitimate if its outcome is completely determined by legislative or constitutional decisions already taken is just another way of denying that the law can ever legitimately arbitrate genuine conflict. However, if the law can never legitimately settle genuine conflicts, social conflict will have to be settled in some other, purely political way. It should be evident that the observation that judges might have to deal with legal indeterminacy in hard cases does not answer the question whether that alternative is preferable to the rule of law.

The claim that compliance with ‘public law’ cannot be enforced by a sovereign able to keep all affected parties in awe of his irresistible power, though descriptively true, does not imply that constitutional or international law are inherently limited in their legitimate scope. After all, it is the very project of constitutionalism to create a legal order that does away with the need for a transcendent sovereign power. If modern public law was dependent on enforcement by a transcendent and potentially dominating sovereign power, then something would have gone wrong. Admittedly, a system of modern public law can only work if the affected groups, though lacking in perfect homogeneity, are willing to cooperate with one another in supporting the constitutional system, even where it creates outcomes they might not like. But experience shows that such cooperation is perfectly possible, and it would be perverse, given the goals of constitutionalism, to interpret the dependence of a system of public law on that attitude as a defect.

Finally, the argument of this paper will have made clear that constitutionalism, in its global and domestic form, is indeed incompatible with a certain conception of popular sovereignty. If a theory of popular sovereignty merely transfers the attributes of an absolutist sovereign to an imaginary constituent power, so that those who purport to speak for the latter can inherit the powers of an absolute sovereign not bound to the rule of law, constitutionalism will, of course, conflict with popular sovereignty. But it should be clear that popular sovereignty need not be understood in this way. Schmittian theories of popular sovereignty are normatively unattractive as well as conceptually confused,Footnote ⁸² and they are not our only theoretical option. The mainstream of democratic constitutional thought, stretching from Rousseau and Kant to Habermas, has always emphasized that popular sovereignty is internally related to the rule of law.Footnote ⁸³

Kelsen and Lauterpacht had it right: It is perfectly possible to interpret the analogy between constitutional and international law in a constitutionalist key. The question is whether we want to do so, or whether we prefer to organize our political life around the distinction between friend and enemy.