Jump to ContentJump to Main Navigation
The Legacy of PluralismThe Continental Jurisprudence of Santi Romano, Carl Schmitt, and Costantino Mortati$

Mariano Croce and Marco Goldoni

Print publication date: 2020

Print ISBN-13: 9781503612112

Published to Stanford Scholarship Online: January 2021

DOI: 10.11126/stanford/9781503612112.001.0001

Show Summary Details
Page of

PRINTED FROM STANFORD SCHOLARSHIP ONLINE (www.stanford.universitypressscholarship.com). (c) Copyright Stanford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in SSO for personal use. Subscriber: null; date: 15 April 2021

Carl Schmitt and the Concrete Order

Carl Schmitt and the Concrete Order

Chapter:
(p.99) Chapter Three Carl Schmitt and the Concrete Order
Source:
The Legacy of Pluralism
Author(s):

Mariano Croce

Marco Goldoni

Publisher:
Stanford University Press
DOI:10.11126/stanford/9781503612112.003.0004

Abstract and Keywords

Chapter abstract: This chapter centers on Carl Schmitt’s influential theorizing by exploring his thoroughgoing revision of his previous decisionist paradigm. It investigates the major theoretical move whereby, based on Maurice Hauriou’s and Santi Romano’s teachings, Schmitt dispensed with his famed theory of the exception and put forward his “concrete order and formation thinking.” While his persisting obsession was with the homogeneity of the political community, he importantly changed his mind as to how it can be attained and how it should be preserved. These pages also shine a light on the difference with Santi Romano’s idea of order, especially as to how their disagreeing conceptions of it led to disagreeing conceptions of pluralism. Schmitt’s revision of his own theory, juxtaposed to Romano’s firm conceptualization of the juristic point of view, teases out what is at stake in the late-modern relation between the juridical and the political.

Keywords:   concrete order, decisionism, exceptionalism, institutionalism, legal pluralism, Romano, Schmitt, social pluralism

THE NOTION OF ORDER LENDS itself as a counterpoint to the notion of system: the whole legal debate of the first three decades of the twentieth century can be read through the prism of the dialectic between these alternative modes of the legal. An idea of law as an organized normative complex that brings order to society through the production and implementation of statutory rules (system) rivals an idea of law as something that arises out of people’s practices and gets incorporated into state-based institutional categories, figures, and models (order). “Order” gestures to something that can be neither created ex nihilo nor exhaustively verbalized, as it escapes human beings’ limited law-making faculty and can only be interpreted—albeit partially—as it gets discovered. While “system” gives a sense of something that is artificially laid down to provide guidance for action—as people perform practical activities and try to foresee what others will do—“order” conveys a sense of material intertwinement of social life and the law. Accordingly, a system belongs to a higher order of normativity with respect to social normativity, in the sense that the former claims to guide, mold, constrain, regulate, govern the latter from outside and from above. Quite the reverse, the order is not severable from the practice of daily life, in the sense that it springs from it and never claims to hover over it. The order is a normative force that governs by gathering motivational resources from the very practices it regulates. Therefore, the forms of observance that system and order imply are antithetical to each other. While (p.100) the system always makes room for frictions between the reasons for action it provides (mainly in terms of action planning, threat of punishment, and reward) and rule-abiders’ private reasons, the order’s main reason for action is that it is internal to the practice it governs from within.

This prism is the guideline we chose to follow in exploring Carl Schmitt’s influential theorizing. It is our claim that Schmitt’s overall theory was profoundly affected by his thoroughgoing revision of the role of the order in the creation and maintenance of a political community. Needless to say, at no point was he a supporter of the idea of a system, as he fiercely chastised the systematic idea of law advocated by his fiercest intellectual adversary, Hans Kelsen. However, Schmitt’s critical take on the concept of a system changed significantly at the end of the 1920s, as he gave an institutional twist to his theory of law and politics. This chapter will investigate this major theoretical change whereby Schmitt dispensed with his famed theory of the exception and put forward a theory of the concrete order. The scrutiny of the different types of criticisms he leveled at the normativity of the system will allow us to show that his main concern was with pluralism as an ongoing threat of dissolution. While Schmitt’s persisting obsession was with the homogeneity of the political community, he importantly changed his mind as to how it can be attained and how it should be preserved. This analysis will also shine a light on the difference with Santi Romano’s idea of order, especially as to how their disagreeing conceptions of it led to disagreeing conceptions of pluralism. As will also become clearer in Chapter 5, Schmitt’s revision of his own theory, juxtaposed to Romano’s firm conceptualization of the juristic point of view, teases out what is at stake in the opposition between juristic and political conceptions of law.

The Traces of Institutionalism

The name of Carl Schmitt has long been associated with the notions of exception and decision and the corresponding paradigms of exceptionalism and decisionism.1 While this is obviously not mistaken, we believe this interpretation omits remarkable variations in his intellectual biography and the productive tensions between the multiple facets of his work as a whole. (p.101) Certainly, in this context it is impossible to retrace his voluminous work to sketch the complete trajectory. What we set out to do instead is foreground Schmitt’s moving away from decisionism and the conceptual reasons for his eventual adhesion to his own version of legal institutionalism, which he dubbed “konkretes Ordnungs- und Gestaltungsdenken” (concrete order and formation thinking). This short section aims to pinpoint a few, sparse traces of an institutional inflection before the 1930s, while the subsequent one will discuss a possible interpretation of Schmitt’s decisionism that, though partly mistaken, ferrets out significant conceptual tools that will prove useful to the understanding of his thinking. This will pave the way for a more nuanced analysis of Schmitt’s progressive inclusion of the institutional elements that severed the ties with his previous exceptionalism.

As early as 1910 Schmitt licensed his dissertation on criminal law titled Über Schuld und Schuldarten (On guilt and types of guilt). The dissertation was supervised by Fritz von Calker, who was professor of criminal law at Kaiser-Wilhelm-Universität in Strasbourg and was a member of the National Liberal Party. In this text, Schmitt sought to clarify conceptually the notion of guilt in order to put forward an entirely juristic definition of it. In an antisubjectivistic fashion, he claimed that the concept of guilt cannot rest on subjectivistic motives, such as dispositions, impulses, or inclinations, but has to be founded on legal categories that thoroughly integrate it into the legal order. Guilt is not something pertaining to subjective reasons that are external to the legal order, nor does it merely have to do with the infringement of legal norms. Rather, it is a disposition to act in such a way that the order itself objectively qualifies the performed action as inadmissible. Importantly, in this context Schmitt explained that not only legal rules but the whole legal order can be assigned specific and concrete ends; which is to say, a sort of normative guidance illustrating what it is that guilt is directed against. Guilt is never an individual act, but a hindrance to the general ends of the legal order. A crime is a conduct that is objectively antithetical to the norms reflecting the order’s ends and the guidance it offers. While this original though minor text could be interpreted as an early manifestation of Schmitt’s subsequent decisionism,2 Andrea Salvatore rightly avers that Schmitt offered us two main cues to an alternative reading. First, (p.102) as we already remarked, the existence of specific ends that the legal order pursues as such. Second, the fact that these ends are encapsulated by institutional instances that the order encompasses and makes legal. In this light, the conflict between the legal order and the criminal is a conflict between two types of ends and therefore between the instances of two rival orders.3

Schmitt’s subsequent book, Gesetz und Urteil (Statute and judgment) (1912), went down the same path. It marked a clear departure from Laband’s conclusion that judges are the “viva vox legis” as they apply statutory law—nothing other than logical operations and subsumptions. While what is at stake in this writing is the judicial decision, the term decision in this context bears no resemblance to the notion that Schmitt adopted a few years later as the cornerstone of his 1920s decisionism. For he criticized Laband’s idealization of a gapless legal order that leaves no elements of individual discretion. Still, it is worth noting that the judge’s decision is not a sovereign, political decision, but a judicial one, which does not institute a legal order—as is the case with the sovereign decision Schmitt’s later decisionism centers on. This decision plays off of the indeterminacy that every legal norm entails. No doubt, also in this case one would be justified in detecting a decisionist tinge in this book. The idea that all that counts is the judge’s making a decision appears to “foreshadow Schmitt’s major themes of the 1920s: orientation towards the exception, the emphasis on the value of the act of deciding, and the distinction between legality and legitimacy, which he plays off against one another.”4 However, as Salvatore notes, Schmitt is particularly concerned with the far less decisionistic notion of judicial normality as something that provides the ground for a stable decisional practice among judges. The judge is called upon to adjudicate in a way that conforms to the current practice with a view to making it uniform and consistent. According to Schmitt, a judge’s decision is correct when another judge could in principle arrive at the same decision. It is the normal judge, the normal legal expert. Not an ideal judge, but the average judge.

However, at no point in the decision may the judge follow an absolutely free discretion, his particular subjectivity, his personal belief as such; the “other judge” is the normal, legally cultivated judge, where (p.103) the word “normal” is used in a quantitative-average sense, not describing an ideal type, not qualitative and teleological.5

Interestingly, the normal judge is entrusted with safeguarding the order as a whole by reiterating a practice that incarnates a standard. In this sense, it is evidently at odds with any variation of exceptionalism. The fact that this is Schmitt’s main concern before the 1920s supports Martin Loughlin’s (2016) argument that Schmitt’s overall oeuvre should be read as an exercise in political jurisprudence, that is, a type of jurisprudential inquiry that aims to capture the character of the basic laws of the political. These basic laws can hardly be discovered in the unstable circumstances where normality is suspended and the law awaits a sovereign decision to be brought into life. Rather, Loughlin continues, one fails to latch on Schmitt’s key concerns unless one pays due heed to the texts in which he came to elaborate on his concrete-order thinking. We believe this interpretation does justice both to Schmitt’s self-understanding as well as the efforts he put into solving the puzzles his previous decisionism had generated.

While Schmitt’s espousal of institutionalism as well as the theoretical path that led him to it have been abundantly debated in the last few years,6 what we set out to do in the successive sections is to identify the frictions between Schmitt’s exceptionalism and his later institutionalism. As we will clarify, we will not go so far as to argue, as some have done, that exceptionalism was merely apparent and that he was always concerned with the concreteness of the order. We believe it is more likely that early in the 1920s he genuinely thought the exception could convincingly explain what law is and that he progressively became conscious of the limits of this conception. He was certainly always sensitive to how the conditions of normal, everyday life is produced, as attested by the early writings we have briefly examined. However, in his decisionist phase he was genuinely inclined to believe that the intervention of a sovereign decider was foundational to the political community and that a sovereign decision is needed to set the possibility conditions for normality. It was not until the end of the 1920s that, he came to realize he had underestimated the vital role of concrete, institutional, daily practices in the production of the legal order. Based on (p.104) this, he broke with exceptionalism and overhauled decisionism by introducing elements of an institutional theory of law. The subsequent sections home in on this transition.

The Exceptionalist Reading

As one thinks of Schmitt, two books spring to mind: Political Theology, published in 1922, and The Concept of the Political, first published in 1928. In the course of a few years, Schmitt deployed an influential theory of the legal order as the creation ex nihilo of legal normality through a decision that cuts out that which a sovereign decider identifies as a lethal threat to the political community. In light of this conception of the legal order and how it emerges, it comes as no surprise that most interpreters postulated a link between these two books. This section intends to make the claim that this exceptionalist reading does justice to Political Theology but severely misreads The Concept of the Political and misses out on Schmitt’s main preoccupations in this latter text.

In Political Theology Schmitt developed two conceptual devices that are commonly known as “decisionism” and “exceptionalism.” It should not be neglected that he elaborated on them as a response to the normativism of positivist thinkers. For from the very beginning the exception plays out as a “borderline concept,” that is, a concept that permits testing the tenability of a general theory of sovereignty and the state. He reasoned that no general theory can be premised on the normal case, because the latter is always parasitic on the “borderline case.” Normality is a by-product of a decision. To make this point, Schmitt presented the legal order as the interplay between two elements: norm and decision. Importantly, in Political Theology he only took into consideration these two elements, and as we will see later on, this is key to untangling Schmitt’s subsequent revision of his own theory. For now, however, let us concentrate on the reasons for the exceptionalist reading. The initial pages of Political Theology set the stage for a harangue against the normality of everyday life where the law carries out its tasks of regulation and coordination. Schmitt insisted that the contribution of jurisprudence cannot be limited to the study of the “recognizable” (12), namely, (p.105) that which ensures the regularity and predictability of people’s behavior. Such an undue separation between jurisprudence and sociology prevents positivism from getting to the core of the legal phenomenon.

Schmitt was by and large right in insisting that legal positivists believed the legal order is intelligible and can be accounted for by scrutinizing its formal structure. For example, Kelsen’s influential version of legal positivism, known as “pure theory of law,”7 claimed that legal scientists can make sense of the unity of the legal system only insofar as they postulate the existence of a “basic norm,” whereby all norms appear as linked to each other and finally anchored to the constitution in force. By preserving the boundaries of such a “pure” methodology, Kelsen intended to expunge all extralegal (that is, political and/or moral) considerations from the conceptual toolkit of legal science. All that is required for legal scientists to produce a workable and consistent description of the legal system is contained in the theoretical toolkit of legal science. The study of law does not and should not accommodate any extralegal notions—such as justice, sovereignty, will, power, and others—that are not reducible to the concepts of legal norm and legal sanction.

Schmitt’s main preoccupation in his analysis of legal positivism is that the (alleged) self-sufficiency of the positivist theoretical toolkit omits to consider the source of the legal order and exclusively places emphasis on the normal case. He averred that legal positivists took the state for granted and only contemplated the legal order when it is already up and running. Positivism deliberately ignored how the state came about, as if this formative process were not part of the law and hence were not of the legal scientist’s concern (and importantly, Schmitt viewed this as a conceptual flaw). However, he went on to say, the state, its existence, and its jurisgenerative force can hardly be taken for granted. He wrote: “The existence of the state is undoubted proof of its superiority over the validity of the legal norm.”8 If this reading makes sense, chapter 1 of Political Theology should be interpreted as a conceptual critique of legal positivism by which Schmitt reintroduced the issue of the law’s origin as a key jurisprudential concern and depreciated normality as ancillary. The overemphasis on normality made legal positivists blind to the foundational nature of the exception as a foremost jurisprudential (p.106) concept. The exception is a circumstance where the normality of everyday life is suspended because a novel, emerging configuration of social life has grown so pervasively that the law cannot be complied with and the normal status quo is subverted.

It deserves attention that, on this account, the state of exception in Schmitt does not signify the primacy of politics over law, but the conceptual relevance of the original moment when the law is brought into life along with the normality of social life in which the legal order can work properly. Moreover, Schmitt’s insistence on the origin of law was instrumental in disproving the identification of law and the state. Pace Kelsen and prior authors, in Schmitt’s view paying heed to the exception demonstrated that the law does not coincide with the state. Importantly, as he pointed out, what he was concerned with was not a “causal or psychological” origin of law.9 Rather, he was concerned with a conceptual definition of law in the light of how it comes to life. He wanted to pinpoint not the legal order’s factual conditions of existence but the conceptual ones—that is to say, the element without which no defensible conceptualization of law can be advanced. This is why he attached the utmost importance to the sovereign decider who makes the decision that creates the order without any previous power-conferring norm.

Schmitt’s sovereign is not a predefined institutional figure (although in some circumstances it can be, when the constitution provides for special prerogatives of the executive in circumstances of extreme peril). Rather, the sovereign is the one who decides on the state of exception; which is to say, the person or group who effectively and successfully imposes an order after suspending the order. On this account, the exceptionalist reading of The Concept of the Political might appear to be correct to such an extent that the connection between this book and Political Theology turns out to be decisive. It is this connection that makes it clear that the sovereign’s decision coincides with an act of identification of the enemy and that this decisional activity creates the friend. In effect, The Concept of the Political portrays the friend as contingent on the enemy because of a polemical and polemogenic process that is required to bring about the order.10 This is the gist of Schmitt’s incipit in Political Theology where he says that the “exception (p.107) in jurisprudence is analogous to the miracle in theology.”11 The creation of an order requires the intervention of a supreme entity that generates it, while this entity carries out this creational task by identifying an enemy that stirs the friend’s willingness to stick together and to fend off the enemy. On one hand, the sovereign fulfils her duty if and only if she brings the friend to believe that the enemy is posing a lethal threat to them, one that menaces to wipe out their form of life. On the other, the friend turns into a self-conscious political entity if and only if people are disposed to give their life and take the life of their enemy, if such an extreme step needs to be taken.

Based on the connection between Political Theology and The Concept of the Political, two enlightening notions come to the fore: the formal metamorphosis of nonpolitical conflicts and the performative conception of the enemy. First, in The Concept of the Political Schmitt clarifies that the political is not so much a field as it is the effect of an ultimate distinction “to which all action with a specifically political meaning can be traced.”12 Famously, as we mentioned above, this ultimate distinction is the friendenemy criterion. It signifies “the utmost degree of intensity of a union or separation”13 of an association or dissociation of humans “whose motives can be religious, national (in the ethnic or cultural sense), economic, or of another kind and can effect at different times different coalitions and separations.”14 This implies that any opposition, whatever the realm in which it occurs, can potentially escalate into a political conflict. As soon as this metamorphosis comes about, the conflict becomes a political one and the friend-enemy distinction takes over. The opposition between the opponents ceases to be moral, religious, or economic and turns into a genuinely political conflict. This means that it can no longer be accounted for as moral, religious, or economic, but as quintessentially political. This implies that the distinctive trait of political antagonism is the fact that the stake of the conflict undergoes an essential transformation: it no longer concerns the material substance around which it orbits (e.g., moral disagreement or economic competition), but the existential menace it poses to both parties. People no longer dispute over the stake, but associate their existence and their subsistence (as a group) with this stake to the extent that they are ready to kill and to be killed when it comes to disputing over it. We (p.108) defined this metamorphosis “formal” because the content can well stay the same, whereas the group’s disposition to stand up for it intensifies so much so that they are ready to kill and die.

This leads to the second notion we mentioned above. If the political is the utmost degree of intensity, and not a socially observable field or a specific type of conflict, it is arguable that it serves as the performative condition for a group to come into life. The formal metamorphosis of a conflict urges the members of a group to conceive of themselves as parties to the same entity. In a way, this could be likened to an activity of “performative citation” whereby the talk over something brings this something into existence by dint of being mentioned as something.15 On this reading, the degree of intensity that marks the political prompts a group to verbalize their condition in such a way that a new entity becomes the pivot of their verbalizations. This citational performance parallels the creational activity Schmitt mulls over in Political Theology, whereby one’s pointing the finger at the menace posed by the enemy performatively constitutes the group of friends as the latter look at the situation as a group and conceive of themselves as such.16 If this is the case, despite Schmitt’s emphasis on the concrete and existential nature of the conflict, the friend-enemy antithesis is first and foremost a conceptual device; or better, it is essentially hallmarked by the virtual possibility of an existential conflict. The insistence on the virtual and the possible means that it is not the actual conflict as such that counts, but its bare likelihood, that is, the group’s tendency to conceive of a conflict—regardless of its stake—as something that involves the disposition to kill and to die.

This means that more than the polemical aspect, it is the polemogenic one that sets the citational activity in motion. In fact, if the possibility of a concrete pòlemos is central, it is the friends’ acquired awareness that the conflict requires their readiness to kill and to be killed that turns it into a political conflict. The decision is therefore the polemogenic (in the sense of generating a fundamental disposition to fight) recognition of a specific intensity of the conflict, which urges the friends to join together and to acquire a new awareness of who they are and who they have to fend off. This interpretation of Schmitt’s conception of the political squares with Vinx’s (p.109) nuanced interpretation of Schmitt’s conception of how a people comes about. In his “Introduction” to the Schmitt-Kelsen debate on the guardian of the constitution, Lars Vinx avers that for Schmitt,

a people’s “political existence” is manifested in its willingness to assert itself…. [T]he unified will of the people … in fact exists for as long as a people is willing to take (or rather to support) genuinely political decisions; decisions on the exception that constitute political community, in an extra-legal space, by drawing the line between friend and enemy.17

Based on this, it is our claim that an exceptional reading of Schmitt’s work of the 1920s is conducive to an epistemic understanding of the law and the political. While in Political Theology Schmitt aimed to debunk positivism as an approach to the proper conceptualization of law, in The Concept of the Political he intended to illustrate how a group, a population, a people, a community come into life. As far as the legal order is concerned, understanding its generative dynamics entails going beyond the constitution in force and scrutinizing the creational moment when the order is established. As far as the political community is concerned, the analysis of the institutive, creational moment demonstrates that those generative dynamics rest upon the jurisgenerative force of a group of people conceiving of itself as a group of people vis-à-vis an incipient threat. This epistemic understanding of Schmitt’s theorizing shines an interesting light on a far less substantive notion of politics than one might attribute to him.

Although this epistemic understanding makes sense of many of Schmitt’s concerns in his decisionist phase, we believe the connection between Political Theology and The Concept of the Political is looser than the exceptionalist reading would have us believe. In particular, while the epistemic interpretation accurately foregrounds the virtual and performative nature of the friend-enemy opposition, it fails to grasp Schmitt’s main concerns at the ends of the 1920s. The following section will suggest approaching The Concept of the Political from an alternative angle. Political Theology marked a transitory phase in Schmitt’s thinking, one that was heavily indebted to (p.110) the notion of exception. A few years later, he extensively revised decisionism in a way that led him to drop exceptionalism and to concentrate on the more salient aspects of normality and the normal case.

Toward a Concretist Reading

As we mentioned above, recently some scholars have argued that traces of an institutional thinking were already manifest in Schmitt’s writings of the 1920s, such as Political Theology and Roman Catholicism and Political Form.18 For example, Leila Brännström maintains that an institutional attitude tinges Schmitt’s exceptionalism because it is “only when the authorized subject of sovereignty makes an effective decision on the state of exception that ‘actual power’ and ‘the legally highest power’ come together. This, however, presupposes that the authorized subject can be identified prior to the decision.”19 However intriguing this interpretation may be, we have already explained why it does not withstand scrutiny. Indeed, we believe it is imperative to cast light on the different background against which Schmitt gradually came to espouse an institutional theory of law. This analysis brings to light a remarkable change in how he conceived normality and its role in yielding the contents of the legal order.

There is no denying that Schmitt was concerned with the concreteness of the political community before the end of the 1920s. For example, it is true that in Roman Catholicism and Political Form he extolled the exemplar way in which the Catholic Church had long enacted the principle of representation in “antithesis to the economic-technical thinking” that (he claimed) dominated his time. He clearly dismissed a notion of representation as delegation (Vertretung) and insisted on genuine representation being “the concrete foundation for a substantive form.”20 This is evidence that the problem of the concrete, material constitution of the social is a recurring trope in his theorizing, which is hardly reducible to bare existentialism. However, Schmitt returned more intensely to the issue of the concrete foundation of the order in between the end of the 1920s and the early 1930s. More in particular, as we will explain in detail later on, between (p.111) 1933 and 1934 he put forward his “concrete order and formation thinking” and clarified that decisionism cannot stand on its own feet, because, like positivism, it neglects important societal elements that are foundational to the law. Before examining the outcomes of this crucial transition, we would like to point out early symptoms of his dissatisfaction with exceptionalism towards the end of the 1920s. To do this, we will present The Concept of the Political as a first decisive break with the exceptional paradigm that saturates Political Theology.

In the 1963 “Preface” to The Concept of the Political Schmitt rebuked those who claimed he had disproportionately paid heed to enmity and overlooked friendship. Those critics, he wrote, “neglect that all movements of a legal concept, by dialectic necessity, always proceed from negation. In the life of law as well as in legal theory, entailing negation implies no ‘primacy’ of that which is negated.”21 Then he went on to say that in criminal law the concept of punishment presupposes wrongdoings, but this does not imply criminal law’s eulogizing crime. While also in this case enmity is presented as a conceptual possibility condition for the constitution of the friend (just as the crime is a conceptual presupposition of criminal law), what matters here is that enmity is portrayed as a threat to an existing order. If we take this remark as the pointer to an alternative reading of The Concept of the Political Schmitt offered in 1963, it would be helpful to understand what the threat was that he had in mind more than thirty years earlier. We believe a further telling lead can be found in the first of the three corollaries that appear in the 1963 German edition of Der Begriff des Politischen. This corollary, originally written in October 1931, discussed alternative meanings of the notion of state neutrality. One of them, the fourth, relates to the state granting equal consideration and equal opportunities to the various substate groups. In this passage Schmitt concerned himself with social, cultural, and religious pluralism and tackled the question of whether the state can really be neutral when it comes to this key issue. His argument was that this type of neutrality “is feasible only with regard to a relatively small number of legitimate groups and only by a relatively uncontested allocation of power and influence among equally legitimized partners.”22 In other words, the (p.112) state has to recognize some substate groups of different sorts (cultural, religious, or driven by other associative interests), but it behooves it to identify a criterion for selecting some of them and selecting out others.

A few important conclusions can be reached based on these passages, in particular one concerning Schmitt’s social ontology and the other concerning his view on how the state should grapple with the question raised by this social ontology. First, Schmitt was sociologically more realist than one might think. As he had already clarified in one of the most revealing, and yet neglected, writings, “State Ethics and the Pluralist State,” which appeared in Kant-Studien in 1930, Schmitt rejected the Hobbesian ontology of a civil society comprising discrete monads that stand in a direct relationship to the state. Schmitt acknowledged that pluralism is a permanent condition of all complex societies. Drawing from a few key Western thinkers, from Aristotle to Hegel, he acknowledged that social reality is composed of a variety of groups and associations and that no political government, however centralized and powerful it may be, could or should completely homogenize them. This is a key indication of one of Schmitt’s main concerns, namely, pluralism and its thorny relation to the political center. If this is the case, then the second important conclusion is that, for Schmitt, there is no easy recipe to handle the innate plurality of social life. In fact, if the state gets symbolic and material resources out of the rich plurality of substate groups, it would be both myopic and counterproductive to exert authority in such a way as to damage their microsocial life and to disrupt their intimate normative texture. Needless to say, however, this does not mean that Schmitt had converted to a pluralist theory of the state. Rather, this means that he was trying to solve this theoretical predicament by striking a balance between a tenable conception of social pluralism and the need for political homogeneity. Keeping this, let us make our way into The Concept of the Political.

Schmitt begins with a basic warning: “The concept of the state presupposes the concept of the political.”23 The message conveyed by this famous opening is that statehood is conditional upon the concept of the political and not the other way around. As we illustrated in Chapter 1, most German jurisprudence and public law theory between the last decades of the (p.113) nineteenth century and the first three decades of the twentieth century had revolved around the state as the source of both sociality and legality. It was presented as the fundamental ground for the existence of a political community. Moreover, the common view was that the state is not only the cradle of politics but also its destiny. The political is contingent on there being a state as the supreme law-giving entity, outside of which there can be no politics whatsoever. Schmitt wanted to debunk this view because it postulated a transhistorical link between the state as a political form and the political as a conceptual category and a field of human existence. It neglected the transient nature of the state as the entity that gained the monopoly on the political and consequently on the jus belli (that is to say, the right to use force when it comes to ban an internal enemy or to fight off an external one). Schmitt rather believed that the state is nothing other than a transitory configuration of the political, with a beginning and an end, one that could cease at any moment to be the political entity par excellence. It is worth stressing that, for Schmitt, the state monopoly on the jus belli was inseparably tied to its capacity to inhibit the formal metamorphosis of nonpolitical conflicts, and thus to prevent intersectional social conflicts of different nature (moral, economic, religious, and so on) from escalating into political conflicts. When the state loses this key prerogative, it can no longer claim to be the political entity par excellence. By the same token, when the members of a group “decide” on their own that the conflict they are engaged in requires their disposition to kill and be killed, at this very moment this group morphs into a political group and (whether intentionally or not) lays a claim to the jus belli. This performative act produces a collision between the symbolic and material aspects of the political: a group’s performative claim entailed by their decision that the conflict is a political one turns at once into a challenge to the state as the supreme political entity. The mere claim—or better, the mere possibility of a particular group’s envisaging a concrete pòlemos—sets off the collapse of the state as the supreme political entity. This means that control over pluralism entails control over the conceptual grid whereby substate groups conceive of themselves.

If this interpretation holds true, Schmitt’s chief concern in The Concept of the Political was not with explaining how a political community comes (p.114) about at the moment it counters an enemy, but with that which the state has to forestall if it wants to preserve its monopoly on the political. The exceptionalist reading that glues together Political Theology and The Concept of the Political presents the friend-enemy antithesis as a condition that the state has to nurture in order for the group of friends to be assembled. Quite the reverse, the concretist reading we champion here presents this antithesis as something that merely plays a theoretical and a pragmatic function. From a theoretical vantage point, the friend-enemy one is a fundamental conceptual criterion that allows determining what the entity is that enjoys the monopoly on the political. From a pragmatic, political vantage point, it identifies a virtual possibility that the state has to ward off. Schmitt elaborated on this criterion to draw attention to the fact that, if at a given moment in time the supreme political entity is the state, this does not mean that it will continue to be the state forever. Indeed, one day it might turn out that, by applying the friend/enemy criterion, we will have to conclude that there are more (or different) supreme political entities, that is, all the entities which, at that moment, will be laying claim to the right to engage in political conflicts. If this is the case, pragmatically the friend-enemy antithesis is something that should never materialize for the state to continue to hold the monopoly on the political. It is a state of things that radically changes the configuration of the social because of the rise of new political actors.

In sum, whereas Political Theology portrays the state of the exception as the original moment in which normality is created, The Concept of the Political portrays the opposition between friend and enemy as a condition that the political community has to obviate strenuously. This idea that Schmitt’s core concern was with the friend and a community’s internal stability rather than war and the enemy is also advocated by Ernst-Wolfgang Böckenförde in his perceptive article on Schmitt’s constitutional theory.24 He explains that in The Concept of the Political Schmitt’s objective was to show that the chief end of the state—what makes the state the fundamental political entity—is “to relativize domestic antagonisms, tensions, and conflicts in such a way as to facilitate peaceful debates as well as solutions and ultimately decisions in accordance with procedural standards of argumentation and public discourse.”25 The state has to preclude the possibility that one or more components of the civil society get to nurture the desire (p.115) for or feel the need to intensify their opposition to the state or to other groups. If the state failed to do so, an effect of political self-constitution would be triggered, and the state political unity would ipso facto be relativized. This is why, Böckenförde rightly points out, “the political in the Schmittian sense is what underlies, facilitates, and shapes the political unity as unity: a degree of intensity of that association which supersedes conflicts and antagonisms in such a way as to provide both form and organization and furnish and maintain a working political order.”26

Why did Schmitt introduce such a significant amendment to his previous conceptualization of the exception? Our hypothesis is that since the late 1920s Schmitt’s interest in pluralism had grown significantly. For one thing, he devoted close attention to pluralist theories of the state. Schmitt was two-minded about them. On one side, he found them enlightening especially as to the conception of the social they propounded. On the other, he sensed that their advocates did not understand the lethal risks that pluralism involves. Schmitt viewed Gierke’s idea that the state is an association of associations as a “decisive step” to the pluralist conception of the state. It opened the door to the further demotion of the state to a neutral apparatus governing the relations between associations. While taking issue with Duguit’s theory, Schmitt recognized that he rightly debunked “uncritical metaphysics of the state and personifications of the state.” Yet Duguit eventually missed “the actual political meaning of the concept of sovereignty.”27 The same applied to Anglo-Saxon theories of the pluralist state. According to Schmitt, in one way or another they all failed to tackle the key political question of “which social entity … decides the extreme case and determines the decisive friend-and-enemy grouping.” However, in Schmitt’s eyes, this was not the neglect of the sovereign decider, as was the case with Political Theology when he argued that positivist legal theories failed to theorize the original moment at which a law is established. Rather, his argument was that if the state relinquishes its monopoly on the extreme case, then it is destined to perish as a political form:

That the state is an entity and in fact the decisive entity rests upon its political character. A pluralist theory is either the theory of state which arrives at the unity of state by a federalism of social associations (p.116) or a theory of the dissolution or rebuttal of the state. If, in fact, it challenges the entity and places the political association on an equal level with the others, for example, religious or economic associations, it must, above all, answer the question as to the specific content of the political…. The state simply transforms itself into an association which competes with other associations; it becomes a society among some other societies which exist within or outside the state. That is the pluralism of this theory of state. Its entire ingenuity is directed against earlier exaggerations of the state, against its majesty and its personality, against its claim to possess the monopoly of the highest unity, while it remains unclear what, according to this pluralist theory of state, the political entity should be.28

In other words, the concretist reading we are advancing in this section submits that The Concept of the Political, and more generally the works Schmitt produced between the last couple of years of the 1920s to the first years of 1930s, hinges on a “constructive” idea of the political. He intended to explain what the state should do to maintain the preeminence it had acquired as the key political form of the modern age. In this light, the enemy as well as war are but conceptual devices that proceed from that which is negated in order to cast light on that which is affirmed. If this is the basic assumption of the concretist reading, our interpretive scheme has to take a further step to show what the friend really consists in. To achieve that, it is imperative to pinpoint what Schmitt regarded as the most serious menace to the state. In fact, if he thought that substate intersectional conflicts might destabilize the existent state order, why did he think the mechanism of the exception was no longer enough to neutralize the internal enemies and ban them? Why could the state government not take matters in hand and declare the state of exception to purge society of its internal vulnerability? The main reason that Schmitt came to believe that this solution would fall through is that he had embraced a more nuanced notion of the friend. Or better, he had come to pay more heed to the societal element that lies beneath any social and legal order. His conception of what a society is had evolved to accommodate elements of an institutional theory of law, one that (p.117) laid stress on the order being composed of social practices and the normative webs developed within them. In other words, the abstractive nature of an unjustified and unjustifiable decision came to appear in Schmitt’s eyes as the main shortcoming of his theory of law and politics.

Institutionalism versus Decisionism

The revealing 1930 essay, “State Ethics and the Pluralist State,” which engages with pluralist theories of the state, offers a telling clue to his renewed preoccupation with state and public law theory from a point of view that bespeaks a noticeable change of mind. As we will discuss in some detail, Schmitt abandoned the view of the social that underlay his work of the early 1920s and embraced a more realist conception particularly alert to the normativity of the social. However, quite unsurprisingly, it is exactly the normativity of the social that became his major cause for concern. It is worth expanding on this theoretical evolution to grasp the decisive paradigm change that led Schmitt to espouse the model that can be called “institutionalist decisionism,” as the combination of a radically antipluralist institutionalism and an amended decisionism.29

Although Schmitt’s initial discussion focuses on the virtues and flaws of the pluralist paradigms that we examined in Chapter 1, what should strike the reader is his brisk rejection of the theoretical monism of state-based conceptions of politics and law, which he straightforwardly presented as abstract and misguided:

Political unity can never be understood as absolutely monistic and destructive of all other social groups…. When constitutional lawyers speak of the “omnipotence” of the sovereign—the king or the parliament—their baroquely exaggerated formulas should be understood as owing to the fact that in the state of the sixteenth to eighteenth centuries the issue was overcoming the pluralist chaos of the churches and estates. One makes one’s task too easy if one adheres to such idioms…. State unity was always a unity from social pluralities. At various times and in various countries it was very different but (p.118) always complex and, in a certain sense, intrinsically pluralist. A reference to this self-evident complexity can perhaps refute an extravagant monism but does not solve the problem of political unity.30

Schmitt’s notion that any society includes a variety of groups builds on two implicit premises. First, social subjects acquire their identities within group-based normative contexts, while the social world is but an aggregation of these normative contexts. Second, these practical sites cannot either be reduced to their individual components or dispersed into greater normative entities. In this regard, he bought the social ontology of the champions of pluralist theories of the state. This means that he did not disagree with them on what the social world is composed of. He rather disagreed on the conclusion they reached based on this social ontology: “They aim not only to negate the state as the highest comprehensive unity, but above all to negate its ethical claim to be a different and higher sort of social relation than any of the many other associations in which people live.”31 From the bare fact that individuals are members of a series of distinct groups that most of the time overlap, pluralist theories of the state inferred that the state is also one of those groups. It is at this point that Schmitt relied on his conception of the political to reject this conclusion:

Among pluralist theorists of the state as nearly everywhere, an error prevails that generally persists in uncritical unconsciousness—that the political signifies a specific substance, next to the substance of other “social associations”; that it represents a specific content besides religion, economy, language, culture, and law; and that, therefore, the political group can be understood as standing coordinately next to the other groups—to church, combine, union, nation, cultural and legal communities of all sorts. Political unity thus becomes a special, new substantial unity, joining other unities. Any debates and discussions on the nature of the state and the political will become confused as long as the widespread idea prevails that a political sphere with its own content exists side by side with other spheres.32

(p.119) Evidently, while Schmitt’s objection to monist theories of the state had to do with their conceptualization of the social world, his objection to pluralists concerned the political. Moreover, this also attests to the theoretical frame in which he was elaborating his view of the political, that is, the increasing pluralism of contemporary societies. For it is at this point that Schmitt defined the political decision not as that which creates the friend by identifying the enemy, but as that which prevents “all other opposing groups from dissociating to the point of extreme hostility (i.e., to the point of civil war).”33 As the concretist reading emphasizes, the political is primarily oriented towards internal politics, in the sense that it has to forestall the formal metamorphosis of conflicts. The stable coexistence of groups, Schmitt went on to say, enables the stabilization of the normal condition. From this viewpoint, Schmitt at the end of the 1920s can be theoretically defined a social pluralist who thought social pluralism should never turn into political pluralism. In his own words: “It is not a question of denying this existing pluralism and violating it with universalism and monism, but rather of correctly placing pluralism.”34 Put otherwise, contrary to Romano, Schmitt was convinced that the state should be an entity with no normative rivals, one that establishes what is worth fighting for and thus what conflicts possess a political character.

Before we focus on the details of Schmitt’s theory of institution, it is worth briefly insisting on the gradual paradigm change in his conception of law and the political. While it is mistaken to overstress his institutional inclination before the end of the 1920s, it is just as mistaken to speak of a precipitous conversion to institutionalism. In the works he wrote late in the 1920s, and above all in Schmitt’s most systematic work, Constitutional Theory, published in 1928, there are quite a few traces of a deep revision in progress. In this latter book, Schmitt introduced the concept of a substantial constitution as the containing structure of a political entity molded in a determinate conformation. More tellingly, if in this context one concentrates on Schmitt’s presenting the constitutional guarantees as a set of concrete institutional guarantees (such as the various, intertwined regulations of the justice administration, religious associations, the educational system, (p.120) etc.), one can easily come to the conclusion that this famous book is a first, manifest sign of an institutional thinking.

In the two subsequent years, Schmitt expanded on the doctrine of constitution he had advanced in 1928 and emphasized the foundational role of institutions. It is at this juncture that Schmitt upturned his previous exceptionalist perspective. He dispensed with the idea that the institutional guarantees are party to a more comprehensive set of constitutional guarantees and embraced a view of the latter as elements of a broader set of institutional guarantees. These exceed the written constitution, as they are embedded in the normative practices of daily life. This decisive change came to surface in an (usually overlooked) essay Schmitt wrote in 1931, Freiheitsrechte und institutionelle Garantien der Reichsverfassung (The liberty rights and the institutional guarantees of the Reich Constitution). In the formal revision he made some decades later (1958), Schmitt himself explicitly averred that this writing marked a turning point in the doctrine of institutional guarantees, originally elaborated in Constitutional Theory and subsequently summarized in an essay published in 1932, Grundrechte und Grundpflichten (Basic rights and basic duties). From 1932 on, Schmitt came to think of the constitution not as something that rests on its institutional guarantees, but rather as something that is composed of its institutional guarantees.35 As a consequence, he clearly reoriented his analysis towards the normal situation. What was at stake was the relation between norm and normality.

However, as we will explain in more depth at the end of this chapter, rejecting exceptionalism did not entail rejecting decisionism. For Schmitt held onto a basic element of the latter, that is, the genealogical relation between norm and normality: all legal norms hinge on there being unbroken continuity between law and normality. Importantly, he no longer saw normality as the product of a sovereign who decides on the exceptional case. By making a conspicuous theoretical amendment to his conception, Schmitt got to understand normality as a set of general models of conduct that are produced and accepted in a society’s ordinary life. This is a crucial institutionalist feature that found no room in Schmitt’s previous position and led him to introduce a third type of legal thinking that he had neglected until the late 1920s. He pointed this out in two texts dated 1993 and 1934—the (p.121) period when he made the most despicable political choices. In the preface to the second edition of Political Theology, written in November 1933, he wrote: “I now distinguish not two but three types of legal thinking; in addition to the normativist and the decisionist types there is the institutional one,” which “unfolds in institutions and organizations that transcend the personal sphere.”36 The subsequent year, in On the Three Types of Juristic Thought he advanced his concrete-order thinking. On his account, this type of legal thought emphasizes the relation between the legal order and the social context in which it unfolds.

Before we go into the details of Schmitt’s conception of the relation between normality and the law, it is worth noting that his adhesion to institutionalism was mainly inspired by Hauriou and Romano, whom he regarded as his predecessors: “Hauriou, like Santi Romano, are my masters…. Perhaps, rather than masters, it is more appropriate to say predecessors.”37 However, it is easy to realize that Schmitt’s peculiar understanding of Romano reduced the latter’s nuanced theory to a minor variation of Hauriou’s institutionalism. Whether consciously or not, Schmitt neglected the decisive differences between these scholars and especially failed to grasp Romano’s point on the innate plurality of the legal world. As we will see, this will prove crucial to capturing the limits of Schmitt’s conception of law vis-à-vis pluralism.

The brief sketch of Hauriou’s theorizing that we offered in Chapter 2 helps us understand what it is that Schmitt found intriguing in the groundbreaking paradigm advanced by this French scholar. Hauriou’s state-centered institutionalism took the state to be the “institution of institutions” and “the legal order of legal orders.” Hauriou held a view of the institution as the demiurgic principle of law—“a project-idea underlying an undertaking”—that is accomplished over time through the law in a given social context. In particular, Schmitt seized on Hauriou’s fundamental ideas of a “directing idea” and the formation of a structuring power, along with the notion of superlegality. In doing so, he radicalized the politicization of institutionalism that can be considered as the upshot of Hauriou’s theory of institution, especially between the first (1910) and the second edition (1916) of Principes de droit public (Principles of public law). Both Hauriou’s (p.122) and Schmitt’s institutional frameworks espoused the view that the state is the highest institution, coupled with the antivoluntarism that necessarily follows from such a monopoly on the legitimate enforcement of the legal order. By capitalizing on the inclusion of these two elements and the corresponding exclusion of the liberal connotations of Hauriou’s legal theory, Schmitt coalesced the subjective element of the social (individual beliefs shared by social subjects) and the objective and suprapersonal element of an order created by the state.

On Hauriou’s account, the “directing idea,” in conjunction with the “organized power,” conveys the task that a body has to accomplish and how it can accomplish it through self-organization. The organization is thus instrumental in the realization of the directing idea, while the latter is hegemonic toward the former.38 This is an insight that led Schmitt to make sense of how an assemblage of people turns into a collective body. This transformative process is the source of legality. This means that in both Hauriou’s and Schmitt’s institutional conceptions institutions have not a legal but a protolegal character. Contrary to Romano, Hauriou thought that the institution is the fountainhead of law, in that it is a social project that is meant to achieve a particular objective in light of the organized structure that the directing idea imposes. In this sense, while institution is not law, law never produces institution: “Institutions make juridical rules; juridical rules do not make institutions.”39 The institution is a demiurgic principle of a group project that the law permits accomplishing. Therefore, the law is nothing other than the visible conformation of a social project that seeks to achieve its structuring goal. In the same vein, Schmitt never claimed institutions are legal entities. Rather, the new key role they took in his amended theoretical framework is that they are the source of social normality. At the same time, this theoretical move assigned normality an unprecedent task. While in the previous exceptionalist framework, normality is the product of an unjustifiable, groundless decision on the part of a sovereign, as we will illustrate shortly, in the institutional thinking, normality becomes the lynchpin of a society’s concrete order.

In this new conceptual setting, norms are nothing but standard cases that are given a stable form through legislation. Legal norms are a set of (p.123) stabilized typical cases, declared as binding, which originate from reiterated patterns of conduct. They characterize the contexts in which they are practiced and acquire normative force. They are exempla, exemplar instances offering guidance for conduct to all those who are involved in the kinds of activities that the norms claim to govern. Institutions, as a recognizable ensemble of typical models within practical frameworks, are key components of a concrete form of life, which are reflected in the political form and must be preserved by legal means. In a genuinely Hauriouian fashion, institutional practices can be said to be at one and the same time originally related to, but ultimately independent of, both subjective wills and individual beliefs. They are constituted of a mix of subjective wills, individual beliefs, and reiterate activities, but are not constituted by them. But before we analyze the main elements of Schmitt’s institutionalism, we would like to illustrate in what sense this novel position was intended to overcome the essential flaws that affected the decisionist solution to the problem of legal and political order.

As we noted above, decisionism implies a decision being made regardless of any justifiable criterion. Based on this, the criterion of recognition required to determine the state of exception relies on a self-founding logic that characterizes the sovereign action. The exception brings itself about and provides the unit of measurement for its own application and enforcement. This eventuates in an absolute lack of criteria for determining if a decision is adequate. The conclusion is a sort of irrationalist nihilism: an a priori rejection of any ratio decidendi. This proves a serious flaw, as decisionism fails to indicate what fundamental bonds are shared by the members of a given community, how they emerge, and how they put together the members of a given political reality. In the end, Schmitt’s decisionism never tells us what a friendship relation is. It sheds light only on the polemical and polemogenic function of the political. However, it never accounts for the factors that induce unrelated individuals to team up and give life to a political association. Decisionism only clarifies that a population’s enemy is whoever negates that population’s way of life. But who are the friends? Under what conditions can an ensemble of human beings be considered to share one way of life? How can something be determined by comparison with something undetermined?

(p.124) In sum, the decisionist approach does not indicate the positive content (i.e., the friendship relationship) able to substantiate the concept in question (i.e., who the friend is) independently of any reference to its opposite (e.g., the friend as the nonenemy). It is also to unravel this impasse that Schmitt overhauled his decisionism and set forth his institutionalist theory as a solution to it. In short, institutions, depicted as consistent and widely accepted patterns of conduct, are social devices able to turn scattered individuals into interacting agents who form the “political friend.” Institutions achieve this “quasi-ontological” aim by singling out, among the whole set of descriptive features characterizing a given social context, a particular set of basic and essential elements that are believed to constitute the distinctive traits of such and such course of action. This shift permitted Schmitt to leave the metaphysical level of a groundless and essentially imagined community and to reach the political level of concrete social agency. In this frame, it is no longer a predetermined identity that permits to identify the distinctive way of life adopted within a given social reality. Quite the contrary, it is the joint nature of interactional practices that shapes a social entity and confers on it a distinctive political relevance and identity. What ultimately characterizes individuals is that which they concretely do in their daily, ordinary life, that is, the adoption of reiterated, recognizable patterns of conduct. All individuals who share the same patterns of conduct and the same interactional practices can be said to be “friends,” since they are bound together by common interests and relevant aspects of their everyday life that greatly contribute to making them a social group and (at least potentially) a political actor essentially different from all the others.

Law as a Sieve

The gist of the analysis above is that Schmitt’s endorsement of institutionalism was not an opportunistic move, despite what a few interpreters submit.40 As David Bates pointedly notes in his account of Schmitt’s adhesion to institutionalism, his “institutional theorization in 1933 and 1934 was not therefore merely opportunistic, nor was it intrinsically fascist.”41 This does not mean that his relationship with this theory of law was unproblematic. (p.125) Despite his attempt to show that a few key figures of the German tradition (such as Luther and Hegel) were conscious of the importance of institutions to the good health of the German culture, he knew that legal institutionalism was foreign to the German jurisprudential tradition. His professed skepticism of the word institution bespoke Schmitt’s reluctance to any facile adoption of a non-German jurisprudential thinking as well as his fear that the Nazi establishment might dislike legal institutionalists’ extolling the virtues of a self-organizing society.

Schmitt’s misgivings about the lexicon of institutionalism are manifest in his change of mind as to the term that was meant to express his novel view of the legal phenomenon. As we pointed out above, in the preface to the second edition of Political Theology, he had declared that he wanted to revise decisionism by integrating institutionalism into his revised jurisprudential paradigm. However, in 1934 Schmitt was clear that none of them fully captured his idea of institution. He noticed that the word “institution” in German has a problematic semantic status, as three discrete words might be used to translate it: Einrichtung, Institution, Anstalt. Einrichtung was too general and only referred to the factual existence of a social entity. Anstalt by that time, he claimed, had acquired a polemical meaning that made it unusable. Institution and its Latin root evoked a fixity and rigidity that was at odds with the flexibility of institutional practices.42 He eventually concluded that “concrete-order and formation thinking” was unequivocal. It is highly likely that Schmitt reckoned that an institutional conception of law entails a bottom-up view of politics and thus was at variance with the Nazis’ concentration and centralization of power. It is no coincidence that in that very preface to Political Theology he admitted that “an isolated institutional thinking [i.e., without the support of normativism and decisionism] leads to the pluralism characteristic of a feudal-corporate growth that is devoid of sovereignty.”43 In reality, as again Bates remarks, Schmitt’s endorsement of institutionalism coupled with his moving away from decisionism evoked a theory of law that sets a limit on the decider’s sovereign power: “Schmitt suggested that an ‘institution’ marks the legal form of a ‘concrete order’ that defines both the limit of sovereign decision and the essential foundation of any normative constraint.”44

(p.126) In short, on one side, Schmitt got acquainted with the limits of decisionism in the light of institutionalism; on the other, his endorsement of the latter was by no means unreserved, because institutionalism is coupled with the pluralist view of social life that he had excoriated in “State Ethics and the Pluralist State.” Thus, he had to tease out what he needed and to throw away what he deemed unacceptable. The text in which Schmitt took issue with this theoretical conundrum is On the Three Types of Juristic Thought, published in 1934. Although erratic and slim, this text offers a core conception of law that intends to integrate three notions that, for him, reflect three material features of law that cannot be overlooked: norms, decisions, and institutions. The problem, he reasoned, is that legal theories tend to attach privilege to one aspect to the disadvantage of the others, while a complete picture of the legal phenomenon cannot do away with any of them. Norms, decisions, and institutions are equally key to there being a legal order, because they all contribute to making sure that the concrete order gets reflected and incorporated into the law. Therefore, a legal theory that aspires to be complete should be able to reflect the way these elements integrate each other. While sovereign decisions are required only in specific, particularly tragic moments of a community’s life, norms and institutions are devices that draw out and solidify the normativity of social life on a daily basis. In this new theoretical scenario, law is no longer the top-down imposition of a sovereign decision, but a protection for those social institutions that bring a political community into life and make it flourish. At the same time, what emerges as the mainstay of this more complex image of law is the notion of normality. To acquire a sound conception of Schmitt’s institutional theory and his take on pluralism it is important to pinpoint three elements: his novel critique of positivism, the notion of institutional standards, and the legal instrument that is known as “general clauses.”

Schmitt’s Novel Critique to Positivism

In On the Three Types of Juristic Thought Schmitt advanced a theory based on a mix of transcendentalism and contextualism. For he makes two claims that point to partly divergent directions. First, normativism, decisionism, and concrete-order thinking are three fundamental aspects of law that make the (p.127) law what it is everywhere—regardless of if and how different legal traditions privilege this or that aspect. Second, every legal tradition takes a developmental trajectory of its own that is conducive to a particular configuration of the law and the relative legal science. This means that, on one hand, there is no law that does not comprise norms, decisions, and institutions. On the other, law is always contextual. In short, the contextual nuance is determined by the specific mix of the three basic elements of law. This was particularly relevant for Schmitt’s purposes as he was at pains to renovate legal science for it to be able to account for the serious modifications that German law was undergoing after the Nazi takeover. Based on this mix of transcendentalism and contextualism, Schmitt unfolded a two-pronged study based on conceptual and historical analysis.

What matters for our purposes is the type of critique he addressed to positivism and its spurious normativism. While in his decisionist phase he charged positivists with ignoring the relevance of a seemingly nonjuristic notion to the emergence of the legal order, in On the Three Types of Juristic Thought he mainly concerned himself with the relation between legal norms and social reality. His argument came to be that positivists misread their relation and construct a mistaken hierarchy between them. In a way that distinctly recalls Romano’s critique of rule-based theories in The Legal Order, Schmitt argued that the legal order is neither norm nor a combination of norms and metanorms.45 Normativism is a theoretical attitude that screens off the decision and the concrete order and seeks to found the whole legal order on the alleged self-sufficiency of norms and norms over norms. However, more than decision, Schmitt insisted on the concrete order and how it shapes norms. It is important to stress that “concrete” here gestures to the actual organization of the juridical machinery: “For a law cannot apply, administer, or enforce itself. It can neither interpret, nor define, nor sanction itself; it cannot—without ceasing to be a norm—even designate or appoint the concrete men who are supposed to interpret or administer it.”

Therefore, it is no surprise that Schmitt made a rather liberal use of Romano’s theory, which he defined as “very significant.” At the end of the first chapter, he quoted his Italian colleague as the latter wrote that “the legal order, taken as a whole, is an entity that partly moves according to (p.128) the norms, but most of all moves the norms like pawns on a chessboard—norms that therefore represent the object as well as the means of its activity, more than an element of its structure.”46 It should not go unnoticed that Schmitt translated Romano’s expression “comprensivamente inteso” (taken as a whole) into “ist ein einheitliches Wesen” (is a unitary entity). As we argued in Chapter 2, this betrays a questionable interpretation of Romano’s notion of order and at the same time lays bare Schmitt’s idiosyncratic misuse of his institutionalism. For he wanted to integrate Romano’s theory of institution into his concrete-order thinking and, at the same time, to expunge what in his eyes was its most detestable consequence, that is, the intrinsic link between institutionalism and pluralism.47 This differentiates Romano’s and Schmitt’s approach to normality, as we will discuss later on with regard to this latter notion. Despite this, however, there is no doubt that the two scholars agreed on blaming normativism’s unjustified elision of the machine-like character of law to the exclusive advantage of one aspect.

Normativists—and above all the school of legal positivism, which pushed normativism to its extremes—defended two basic tenets. First, there is no ontogenetic relation between norms and social reality, as they dwell in discrete realms. Legal norms lie in the realm of the ought and are impermeable to all that happens in the realm of the is. Norms are nothing but a technique of description that is valid regardless of whether individuals follow or break them. Second, legal norms can be used to describe social reality in an entirely different way than the way the ordinary language of everyday life would do. For they give life to a metalanguage by which experts can speak about reality in a special manner, that is, according to the set of fact-types that are yielded by the legislator. As we discussed in Chapter 2, that the language of law is a special language is something that Romano fundamentally concurred on with positivists, while Schmitt did not. He identified a major sticking point between institutionalism and positivism on how these paradigms regard law’s specialty. For Schmitt, the error of normativism is that it portrayed law as something norm-abiders use for interpreting and predicting the conduct of the other norm-abiders. Law is reduced to a coordination mechanism, just like a “the framework of a scheduled railroad traffic.”48 Interestingly, Schmitt rebuked this view because it obliterates the relation between norms and the behaviors that norms are intended to (p.129) repress or promote. If the law is a special technique, this is not because it is a regulatory framework for the rational agents of the bourgeoisie, but because it is a potent means for consolidating the conducts that make a community that community. Accordingly, norms are not indications for rational agents to plan their actions, but pieces of a concrete order for the promotion of conducts that are beneficial to the political community and the eradication of those that are detrimental to it.

Therefore, it makes sense to say that it is the connection between law and conventional morality that Schmitt was concerned with. If the law is reduced to a coordination mechanism, a norm prohibiting a crime such as murder is turned into an undesired consequence intended to discourage one from killing another. In other words, the norm prohibiting murder threatens a disadvantage, i.e., a punishment that occurs if somebody kills somebody else. Murder then is not an act that defies and jeopardizes the concrete order, but an instrument for one to plan one’s action whereby one is put in the position to consider the advantages and disadvantages of doing something. Instead, Schmitt thought legal norms encapsulate behavioral standards that are produced by social institutions within ordinary life. Like Romano, Schmitt denied a norm being the sanction that accompanies a behavioral standard, as the sanction is a piece of the machinery as a whole. As Romano put it: “The sanction can be neither contained nor threatened by any specific norm: it can be immanent and latent in the mechanisms, in the organic apparatus of the legal order taken as a whole, it can be a force operating indirectly, a practical guarantee that does not give rise to any subjective right, and thus to any norm from which this law emerges, a constraint that is inborn in, and necessary to, social power.”49 While, as we illustrated in Chapter 2, Romano did not think that norms are related in any way to morality and looked at the machinic nature of sanction within law, Schmitt gave this view an ethical twist: the concrete order is an all-encompassing ethical system.

On this account, conventional morality is the condensation of moral standards produced by those institutions that are compatible with the concrete order. We will shortly see that at the basis of this interplay between social institutions and the concrete order there is still a decider playing a pivotal role—although the decision’s role is strikingly different than the one (p.130) Schmitt had in mind in the 1920s. What matters here is the relation between legal norms and moral standards—which is indeed more correct to define institutional standards. For the adjective “institutional” is more appropriate than “moral,” or, as Schmitt himself wrote, normality, as the latter term conjures the idea of a spontaneity of social habits, while on the contrary institutional standards are the outcome of a selection process. While this will become clearer as we go along, it now leads us to the notion of normality.

Institutional Standards

In Schmitt’s view, institutional standards are prototypical types of behavior, stabilized and established as binding within the normative contexts in which they arise. These are frequently recurring patterns of conduct that characterize the contexts in which they are generated and applied. As they get incorporated into legal norms, the law renders them into exemplar cases offering guidelines for conduct to all those who are involved in the kinds of activities that legal norms claim to govern. They are concrete models “growing out of the order of the concrete ‘conditions.’”50 It is widespread normality that generates typical models of conduct embodied by norms, not the other way around:

A legal regulation presupposes concepts of what is normal, which develop so little from the legal regulation that the norming itself becomes so incomprehensible without them that one can no longer speak of a “norm.” A general rule should certainly be independent from the concrete individual case and elevate itself about the individual case, because it must regulate many cases and not only one individual case; but it elevates itself over the complete situation only to a very limited extent, only in a completely defined sphere, and only to a certain modest level. If it exceeds this limit, it no longer affects or concerns the case which it is supposed to regulate. It becomes senseless and unconnected. The rule follows the changing situation for which it is determined.51

While in the 1920s Schmitt thought that normality explained nothing and the exception everything, he completely revoked his distrust in normality, (p.131) so much so that it became the keystone of legal regulation. In this new institutional frame, the legal order is a filter that is set to select in institutional standards that are instrumental in the homogeneity of the political community and to select out those that threaten to jeopardize it. The “good head of family,” the “brave soldier,” the “duty-conscious bureaucrat”52 are examples of models that the law, on Schmitt’s account, has to incorporate and foster. Certainly, these standards cannot be displayed in any easy manner. No propositional statements comprising the text of a norm is ever able to specify all of their characteristics: “The cohabitation of spouses in a marriage, family members in a family, kin in a clan, peers in a Stand, officials in a state, clergy in a church, comrades in a work camp, and soldiers in an army can be reduced neither to the functionalism of predetermined laws nor to contractual regulations.”53 Although they are not designed to detail institutional standards, legal norms are instruments in the hands of officials to support and stabilize these models. Officials are required to make sure that the norm may always follow the changing situation and that the latter is adequately reflected in legal regulation.

To put it otherwise, legal norms are tasked with determining those that in specific institutional settings establish themselves as normal cases. The law makes them binding and confers on them the force of legitimate authority. Social reality and the law are entangled in a special relation of sieving and stabilization whereby a normative content is drawn out from social institutions and transformed into a stable evaluation standard for all members of the political community. Norms are supposed to follow reality, while reality is molded in accordance with selected legal standards. If this is the case, normality cannot be identified with normativity. Though the former is the cradle of the latter, these two spheres remain distinguished and distinguishable. Interplay does not entail coalescence. Based on this, we can now discuss how this interplay occurs.

General Clauses

In On the Three Types of Juristic Thought Schmitt wrote that “the extent to which the age of juristic positivism has ended is most clearly recognizable today in Germany” because the positivistic idea of certainty had been superseded by the “so-called general clauses” (Generalklausen).54 Those are (p.132) “indeterminate concepts of all kinds, reference to extralegal criteria, and notions such as common decency, good faith, reasonable and unreasonable demands, important reason, and so on.” The notion itself of general clauses entails giving up certainty and calculability. They allow officials to perform the job Schmitt thought pertains to the law as a filter and a stabilizer. The main advantage of general clauses is that they do not predefine any given contents, but leave it to the judges to determine them case by case.

In German law, general clauses had become subject of controversy since they were first introduced by the 1900 German civil code, the Bürgerliches Gesetzbuch. General clauses were generic principles or maxims addressed to officials for them to be able to deal with circumstances not specifically regulated by the code. Thomas Krebs comments that in fact these clauses were a double-edged sword. On one side, the legislator meant to use them to make sure that legal officials might cope with unregulated cases. On the other, general clauses inevitably imply various degrees of discretion, and thus empower judges significantly. As Schmitt himself underlined, these concepts were characterized by a high degree of fuzziness and thus granted much flexibility to judicial activism.55 Yet it was not fuzziness itself that Schmitt appreciated, but the way-out of positivism general clauses suggested. As he explained in State, Movement, People, German law was bound to navigate through the Scylla of total uncertainty and the Charybdis of an impracticable return to positivism.56

It is here that Schmitt gave away the decisionist flavor of his institutionalism. As the relation between social reality and the law is based on a controlled filtering, general clauses are the main instrument in the hands of officials to see to this process. However, officials cannot be granted complete freedom, as they are the instrument in the hands of an ultimate decider who is able to identify those institutional standards that fit the political community. This means that judges are to be formed in such a way that they are naturally sensitive to the inputs of a leader as the ultimate decider. Schmitt put it as follows:

There is only one road. The National-Socialist State has been treading it with great firmness, and the Secretary of State Freisler has given it (p.133) the clearest formulation in the call: “no reform of justice but reform of jurists.” If an independent administration of justice must continue to exist, even though a mechanical and automatic commitment of the judge to predetermined regularization is not possible, then it all depends precisely on the breed and type of our judges and civil servants. Never has the question “quis iudicabit” had any such crucial importance as today.57

What Schmitt had in mind was a pyramidal edifice where the various nodes of a well-structured network are connected to the various subleaders up to the ultimate leader. As a spontaneous convergence on the meaning of the law is impossible, “we demand their commitment without which all the guarantees and freedom, all the independence of the judges, and above all, that ‘creativity’ would be but anarchy and an especially noxious source of political dangers.”58 This kind of leadership feeds off an intrinsic relationship between the leader and his following, one that has to do with ethnic identity. The followers’ loyalty does not hinge on any sense of administrative responsibility but on a substantive identity that ensures social homogeneity. Obviously, this is at odds with an institutional view that is sensitive to the spontaneous emergence of social patterns of conduct and models of life. Spontaneity endangers the relation between the leader and the people.

As a final aside, it is worth stressing that this bespeaks a major change in Schmitt’s key notion of homogeneity. The law is the product of a selection of institutional standards from practice-based normative sources that devoted officials handle with recourse to general clauses and in compliance with the leader’s view of the community. The leader is the interpreter of the common ethnic identity, which, as a concept, owns a “systematic force … that pervades all the judicial deliberations.”59 This makes Schmitt’s institutionalism an “institutionalist decisionism,” as the conjunction of an antipluralist state monism with an amended decisionism. In such a revised decisionist framework, deprived of its previous exceptionalist allure, the homogeneity of the social cannot be obtained through recourse to forceful, abrupt decisions on the part of the leader. Rather, it is a result to be achieved over time, with patience and accuracy in the selection of institutional standards (p.134) through norms and general clauses—and importantly, in keeping with the view of a leader. As the social naturally tends to produce differences, the legal order is called upon to protect the uniformity of concrete order, based on the identity between the leader and the people.

Concluding Remarks

In this chapter we concerned ourselves with Schmitt’s telling revision of his own legal theory, and the comparison with Romano’s proved enlightening in a few junctures. While either was always reluctant about the notion of law as system, Schmitt was the one who pushed the notion of law as order to the extreme. Romano was obviously on the side of the order versus the system, but his view made no room for the idea that law incorporates a form of life and promotes social homogeneity. While for Romano the task of law is that of making orders compatible with each other through technical forms of negotiation, Schmitt tasked law with preventing the rise of an order vis-à-vis another within state borders. The law, Schmitt believed, is an instrument for the leader and his loyal officials to tease out the institutional standards that feed into the ethnic homogeneity of the people and shore it up. The order, then, graphs onto a form of life that has an ethical and an ethnic nature. Therefore, it is hardly surprising that two theories of law as order, despite many key aspects in common, came to irreconcilable visions of pluralism.

Still, the juxtaposition of Romano and Schmitt is useful for another reason. It shines a light on their blind spots. Romano’s trust in the autonomy of legal science obliterated the issue of political power. An order always requires someone selecting among a variety of normative sources that coexist even in a context of plurality of pluralities. Romano thought the law can rely on its special nature to handle the potentially uncountable conflicts between institutions. The order is implicitly associated to a spontaneity of a functional mechanism of self-production and self-regulation. Schmitt overturned this view. His conception casts light on the relationship between the production of legal norms and the reverberating effects of selection and social normalization that these norms bring about on the (p.135) social world. Even within liberal democracies, the core of legislation scarcely lies in the production of shared standards that are neutral vis-à-vis private lifestyles and alternative conceptions of the good. On the contrary, legal norms are instrumental in fostering some lifestyles and conceptions of the good to the detriment of others. In doing so, Schmitt’s theory serves as a warning to those who neglect the mechanisms that turn certain institutional standards, produced in the here and the now of particular contexts, into universally valid norms.

Based on Schmitt’s analysis, the legal field appears as a platform where social identities are given social legibility and political “speakability” insofar as they abide by the “rule of the game,” that is to say, the acceptance of law’s normative preeminence over any other type of normative repertoire. The utmost duty of political institutions is to “tame” the innate force of social practices and to impose on them a common understanding of what is normal. The “normal” is given by the individual conducts and interactional models that are formed within the most widespread and age-old institutions of a given country (such as the church, the army, or the bureau). Schmitt was particularly adroit at showing that the “normal” stems from an arbitrary process of selection that defies the passage from “is” to “ought” by transmuting social patterns of behavior into binding legal standards. While he obviously thought this is indispensable for a healthy community to survive and thrive, this is a lesson that those who appreciate pluralism should take to heart to lessen detestable effects of exclusion.

Notes:

(1.) For recent, influential works that in a way or another take issue with Schmitt’s seminal contribution to the study of the state of exception and emergency powers, see G. Agamben, State of Exception (Chicago: University of Chicago, 2005); D. Dyzenhaus, The Constitution of Law: Legality in a Time of Emergency (Cambridge: Cambridge University Press, 2006); C. Fatovic, Outside the Law: Emergency and Executive Power (Baltimore, MD: John Hopkins University Press, 2009); N. C. Lazar, States of Emergency in Liberal Democracies (Cambridge: Cambridge University Press, 2009). See also A. W. Neal, Exceptionalism and the Politics of Counter-Terrorism: Liberty, Security and the War on Terror (Abingdon, UK: Routledge, 2010), chapter 3.

(p.217) (2.) See e.g., G. Balakrishnan, The Enemy: An Intellectual Portrait of Carl Schmitt (London: Verso, 2000), 15–16.

(3.) A. Salvatore, “Normalità e prassi giudiziale. Per una rilettura delle opere giovanili di Carl Schmitt (1910–1914),” Politica & Società 1 (2018): 131–52.

(4.) S. Korioth, “The Shattering of Methods in Late Wilhelmine Germany. Introduction,” in Weimar: A Jurisprudence of Crisis, ed. A. J. Jacobson and B. Schlink (Berkeley: University of California Press, 2000), 48.

(5.) C. Schmitt, Statute and Judgment, in Jacobson and Schlink, Weimar: A Jurisprudence of Crisis, 64.

(6.) See e.g., D. Bates, “Political Theology and the Nazi State: Carl Schmitt’s Concept of the Institution,” Modern Intellectual History 3, no. 3 (2006): 415–42; L. Brännström, “Carl Schmitt’s Definition of Sovereignty as Authorized Leadership,” in The Contemporary Relevance of Carl Schmitt: Law, Politics, Theology, ed. M. Arvidsson, L. Brännström, and P. Minkkinen (Abingdon, UK: Routledge, 2016), 19–33; M. Croce, “Does Legal Institutionalism Rule Out Legal Pluralism? Schmitt’s Institutional Theory and the Problem of the Concrete Order,” Utrecht Law Review 7 (2001): 42–59; M. Croce, “The Enemy as the Unthinkable: A Concretist Reading of Carl Schmitt’s Conception of the Political,” History of European Ideas 43, no. 8 (2017): 1016–28; M. Croce and A. Salvatore, The Legal Theory of Carl Schmitt (Abingdon, UK: Routledge, 2013); M. Croce and A. Salvatore, “After Exception: Carl Schmitt’s Legal Institutionalism and the Repudiation of Exceptionalism,” Ratio Juris 29, no. 3 (2016): 410–26; J. Meierhenrich, “Fearing the Disorder of Things,” in The Oxford Handbook of Carl Schmitt, ed. J. Meierhenrich and O. Simons (New York: Oxford University Press, 2016), 171–216; M. Loughlin, “Politonomy,” in Meierhenrich and Simons, The Oxford Handbook of Carl Schmitt.

(7.) Kelsen’s celebrated book on pure theory (H. Kelsen, Introduction to the Problems of Legal Theory. A Translation of the First Edition of the Reine Rechtslehre or Pure Theory of Law [Oxford: Oxford University Press]) was written in 1934, but his conception of law was making deep inroads into the contemporary debate well before the 1930s—two highly influential works of his prior to 1934 are Hauptprobleme der Staatsrechtslehre (Tübingen: Mohr, 1911) and Allgemeine Staatslehre (Berlin: Julius Springer, 1925). His ongoing polemic with Schmitt concerned key issues such as sovereignty and the constitution. See e.g., the recent translation of the Kelsen-Schmitt debate on the guardian of the constitution, with a most helpful introduction by the translator: L. Vinx, The Guardian of the Constitution: Hans Kelsen and Carl Schmitt on the Limits of Constitutional Law (Cambridge: Cambridge University Press, 2015).

(8.) C. Schmitt, Political Theology: Four Chapters on the Concept of Sovereignty (Chicago: University of Chicago Press, 2005), 12.

(10.) As we will explain in more detail below, it is important to underline that this process is not only “polemical,” that is to say, thoroughly contingent upon (p.218) there being a conflict between two parties, but also “polemogenic,” in that envisaging the existential conflict between friend and enemy, or better, considering it as a practical possibility, is enough to ignite the political mechanism that brings the friend into life.

(12.) C. Schmitt, The Concept of the Political (Chicago: University of Chicago Press, 2007), 26.

(15.) This important theory was developed by David Bloor with reference to the late Ludwig Wittgenstein and Barry Barnes’s bootstrapping paradigm (see B. Barnes, “Social Life as Bootstrapped Induction,” Sociology 17, no. 4 (1983): 524–45). Bloor argues that standards to understand and criticize practices emerge out of a self-referring activity, that is, an activity of citing whereby a given performance becoming a standard is determined by one’s “commenting on the performances of others, and of one’s self” (D. Bloor, Wittgenstein, Rules and Institutions [London: Routledge, 1997], 33). Such a creational activity is thoroughly resolved into the practice itself when people draw their attention to a given performance and provide it with a stable, objectified, and transmissible description of it.

(16.) As Loughlin suggests, the concept of the political is constituted by the criteria that enable us to identify a group as a group, as an organized political entity (see Loughlin, “Politonomy”).

(18.) See e.g., Brännström, “Carl Schmitt’s Definition of Sovereignty as Authorized Leadership”; Meierhenrich, “Fearing the Disorder of Things”; S. Weber, Targets of Opportunity: On the Militarization of Thinking (New York: Fordham University Press, 2005).

(20.) C. Schmitt, Roman Catholicism and Political Form (Westport, CT: Greenwood, 1996), 8. On this issue, see J. P. McCormick, Carl Schmitt’s Critique of Liberalism. Against Politics as Technology (Cambridge: Cambridge University Press, 1997), chapter 4.

(21.) C. Schmitt, Der Begriff des Politischen. Text von 1932 mit einem Vorwort und drei Corollarien (Berlin: Duncker & Humblot, 1963), 15. Authors’ translation.

(24.) E.-W. Böckenförde, “The Concept of the Political: A Key to Understanding Carl Schmitt’s Constitutional Theory,” Canadian Journal of Law & Jurisprudence 10, no. 5 (1997): 5–19.

(29.) The analysis here draws from the argument put forward in Croce and Salvatore, The Legal Theory of Carl Schmitt.

(30.) C. Schmitt, “State Ethics and the Pluralist State,” in Jacobson and Schlink, Weimar: A Jurisprudence of Crisis, 306.

(35.) Contrary to this reading of Schmitt’s essay, see B. A. Schupmann, Carl Schmitt’s State and Constitutional Theory. A Critical Analysis (Oxford: Oxford University Press, 2017).

(37.) C. Schmitt, “Un giurista davanti a se stesso. Intervista a Carl Schmitt,” in Schmitt, Un giurista davanti a se stesso. Saggi e interviste (Vicenza: Neri Pozza, 2005), 166–67.

(39.) A. Broderick (ed.), The French Institutionalists: Maurice Hauriou, Georges Renard, Joseph T. Delos (Cambridge, MA: Harvard University Press, 1970).

(40.) See e.g., I. Maus, “The 1933 ‘Break’ in Carl Schmitt’s Theory,” Canadian Journal of Law and Jurisprudence 10 (1997): 125–40.

(42.) C. Schmitt, On the Three Types of Juristic Thought (Westport, CT: Praeger, 2004), 89.

(46.) S. Romano, The Legal Order (Abingdon, UK: Routledge, 2017), 7.

(47.) On the relation between Schmitt and Romano, see Croce and Salvatore, The Legal Theory of Carl Schmitt, 109–24; M. de Wilde, “The Dark Side of Institutionalism: Carl Schmitt Reading Santi Romano,” Ethics & Global Politics 11 (2018): 12–24; A. Salvatore, “A Counter-mine that Explodes Silently: Romano and Schmitt on the Unity of the Legal Order,” Ethics & Global Politics 11 (2018): 50–59.

(55.) T. Krebs, Restitution at the Crossroads: A Comparative Study (London: Cavendish, 2001), 20.

(56.) C. Schmitt, State, Movement, People: The Triadic Structure of the Political Unity (Corvallis, OR: Plutarch, 2001), 50.