The tension between constituent power and a liberal theory of constitutionalism is important in constitutional theory as they paint very different perceptions of the nature of authority in a constitution. Constituent power is “the power through which the prince’s power to rule was authorised,”1 and its primary function is to “specify in constitutional language the ultimate source of authority of the state.”2 This article focuses on the work of Carl Schmitt, as it is suggested as a “powerful corrective to the distortions of normativism.”3 It is clearly anti-liberal and decisionist, providing a direct ground for critical evaluation of the alleged tension. In a similar vein, calls for executive action and/or dominance in times of emergency or crisis renders it necessary to understand whether the assumption of (near) absolute power by the executive in such times entails the death of liberal constitutionalism.4 This article argues that the concept of constituent power is only anathema to liberal constitutionalism if their relationship is conceived as dichotomous without any mutually inclusive aspects. Three arguments will be made. Firstly, by analysing Schmitt’s conception of constituent power and its perceived contradistinctions against liberal constitutionalism, it will be shown that the relationship between constituent power and liberal constitutionalism is dichotomous only if constituent power is conceived as external to the legal order. It will then narrow down the focus to the issue of emergency, contending that it does not require the law to recede and be totally suspended, and the concept of constituent power can be reconciled with liberal constitutionalism. Last but not the least, using the recent rumors of the use of the Emergency Regulations Ordinance (Cap. 241) (“ERO”) by the Hong Kong Government to deal with various protests and clashes since early June 2019, it will be shown that limitations on the Executive’s power is crucial in order for Schmitt’s concept of constituent power and a liberal theory of constitutionalism to be reconciled. This article will come to the conclusion that Schmitt’s refusal to recognise the broader political and legal contexts underpinning the concept of constituent power is the main reason why it may be anathema to liberal constitutionalism.
Schmitt’s unnecessarily dichotomous conception of the relationship between constituent power and a liberal theory of constitutionalism
For Schmitt, the sovereign bears constituent power: “sovereign is he who decides on the exception,”5 which cannot be derived solely from a general norm. It is a power which has the ultimate say in a political regime, unconstrained by any norms. His insistence on constituent power as the subject of a constitution denies the possibility of a closed, purely normative constitutional system.6 From this stance, it is unmistakable that Schmitt is antagonistic against a liberal theory of constitutionalism. The sovereign decision is the absolute beginning. His conception of constituent power is clearly anathema to liberal constitutionalism. Taken to its logical extreme, Schmitt’s conception of constituent power in Political Theology forms the basis for an “authoritarian exceptionless exception.”7 Based on the omnipotence of the sovereign, it has the absolute discretion to determine what “constitutes” an exception, without any constraints from any other parties, rules, or norms whatsoever. The untrammeled absolute power the sovereign enjoys here in determining what constitutes an exception and what to do with such “exception” illustrates the possibly endless power the sovereign may be said to possess. This is an accurate observation as the exception which the sovereign decides on may be easily normalised as a daily occurrence. The power to authorise so is unchecked. The utilisation of extraordinary powers in cases of the now-normalised “exceptional” circumstances is antagonistic against a liberal constitutionalist’s propositions for checks and balances and rooting the foundation of extraordinary powers in a legal source.
Before proceeding to scrutinise Schmitt’s propositions, a liberal theory of constitutionalism should first be introduced. In contradistinction to Schmitt’s absolutist picture, Friedrich Hayek offers a remarkably liberal and norm-based theory of constitutionalism. For Hayek, the authority of a legislator should rest on something distinguished from “an act of will on a particular matter in hand.”8 Such authority cannot rest solely upon the discretion of the legislator himself. Instead, such power “rests on a common opinion about certain attributes which the laws he produces ought to possess.”9 The power must be shaped and/or contributed by such attributes, instead of stemming from any unpredictable and variable absolute discretion of a legislator. Moreover, there is no logical necessity that the ultimate power must be omnipotent. Such ultimate power is “a negative power, but as a power of withholding allegiance it limits all positive power.”10 The ultimate authority authorising all actions of the government constraints the government’s actions. The only true side of law is nomos, and it should neither dictate nor direct the actions of the people. For Schmitt, the true side of law is thesis — the law of will: imposed organisation. Schmitt’s main attack on liberalism is based on the inevitable existence of the exception. There is no normative basis which it may be seen to rest or be constrained on.
What Schmitt fails to realise, however, is that his dichotomous conception of the relationship between constituent power and liberal constitutionalism has blinded him from the position the former has within the latter. Loughlin recognised11 that for Schmitt, the significance of constituent power lies in the fact that the state continues to evolve, expressing “the principle of dynamic emergence of political unity, of the process of constantly renewed formation and emergence of this unity from a fundamental or ultimately effective power …”.12 Constituent power also expresses the formative process by which that sovereign will exhibits itself through time.13 Here is where Schmitt misses the point: the formative process of exhibition and decision-making can and necessarily have recourse to legal norms; there can be fundamental norms which the sovereign is subjected to. The authorisation from constituent power in terms of the decisions that can be adopted can nevertheless be subjected to liberal norms, such as the rule of law. For example, the collective autonomy of the people to uphold the rule of law can control the exercise of ultimate authority for the purpose of upholding the desired norms. Schmitt, however, only focused on the formation of the unity itself, but not the role of this unity in conferring power to the government. The formation of the unity and its authorisation of the ruler’s power are separate questions, but Schmitt’s conflation of them has blinded him from the possibility of liberal norms underpinning the authorisation of power. The collective autonomy of the people as vested in the ruler to make decisions does not have to be dichotomised with the norms that the community of people decides to uphold. From this lens, the exercise of constituent power is part of the liberal constitutional order. This gives crucial insight into the second inquiry of this article: the conception of constituent power in the context of emergency.
The peaceful co-existence of constituent power and liberal constitutionalism in times of emergency
The recent expanded uses of emergency power that suspend normal constitutional protections have rejuvenated interest on Schmitt’s distinction between the norm and exception as the analytical framework for relevant constitutional measures.14 For Schmitt, the exception can be “best characterised as a case of extreme peril…It cannot be circumscribed factually and made to conform to a preformed law.”15 If actions in an emergency are not subject to controls, which a liberal theory of constitutionalism would require, it should become clear who the sovereign is.16 In order for the executive to respond to emergencies effectively, constituent power must be unlimited in authorising the range of responses it can offer, regardless of extremity. The exception is characterised by principally unlimited authority, entailing the suspension of the entire existing legal order: “the state remains, whereas the law recedes.”17 Constituent power gives unlimited authority for the ruler to do whatever he deems necessary: it is not subjected to controls of any sort. This contrasts with a liberal constitutionalist’s, for example Hayek, claim that such exercise of power should still be constrained. The key is to limit the power that authorises the government’s powers and actions.
The need to decide on the exception emphasises the central role of political decision-makers who decide on how to deal with it on a case-by-case basis. Schmitt’s conception of constituent power “denies that law’s authority can be found on the intrinsic qualities of legal order.”18 His account, however, is flawed; by correcting this logical fallacy, the tension between constituent power and liberal constitutionalism can be resolved. To begin with, Dyzenhaus noted that Schmitt was unable to locate authority in something external because it is “drawn to claim that the basis is quasi-legal.”19 This is reflected in Schmitt’s insistence for a normal situation to exist in order to determine whether the nation is still in a normal situation or it has entered into a state of emergency. In deciding so, the executive has to start from the norm, which delineates the normal situations of a constitutional order. This is because the power authorising the government’s power to make emergency measures has, at the very minimum, part of its basis in the norm itself. Without recourse to or starting from the general norm, the exception may become the norm itself, as it has nothing to be “exceptional” from. Given the necessary involvement of a “norm,” the concept of constituent power can therefore be reconciled with liberal constitutionalism. The power which authorises the executive to adopt emergency measures does not spring out from nowhere, but from the norm underpinning the normal situation as recognised by the liberal constitution. For example, for a liberal state to declare martial law, it is necessary for the power making the decision to enter into a state of emergency to find in the general norm situations or conditions which allow such state of emergency to be declared. The move from a state of normal law to martial law has to start with an examination of the normal law itself, and whether its inability to tackle the alleged emergency warrants exceptions to be made to the normal law.
However, to further rescue the concept of constituent power from Schmitt’s extreme stance, it is necessary to briefly examine his views on the influence of politics on law. Schmitt’s systematic thinking, in reducing the concept of the “political” to “friend” and “enemy,” has led him to conceive emergency, based on constituent power, as a “battering ram”20 against liberal democracy. His conception of law, that “all law is ‘situational law,’“21 is “intrinsically bound up with a political struggle.”22 Once the “struggle” of the exception triumphs over the normal situation, the latter is seen as “defeated;” the exception’s “victory” awards the executive an unlimited authorisation to enact emergency measures. It is questionable, however, whether the concept of “political” has to be reduced to “friend” and “enemy” in order to justify the application of emergency measures. As Scheuerman observes, Schmitt’s analysis of emergency government “neglects its complex political origins.”23 It focuses excessively on the formation and constant renewal of the power itself, neglecting any common interests which competing concerns in politics may have or compromise on. Here is where liberal constitutionalism can provide an answer: to safeguard the sustainability of the norms the liberal constitutional order is found on. Once the focus of constituent power is broadened to encapsulate the objectives of the authorising the use of emergency powers and “malleable recent social trends,”24 instead of only pestering on the formation and history of constituent power itself, it may be re-conceptualised as the starting point for ending the exceptional situation and return to the normal order. Historical givens, despite providing a solid ground for a framework of exception, is inadequate. It tends to neglect the reasons for emergency and the objectives of emergency measures. The stakeholders within the “political” do not have to be dichotomised into an opposing relationship of friend and enemy.
Once the competing interests underlying norms and exceptions are no longer contradistinguished from one another in the Schmittian sense, constituent power can be reconciled with a liberal theory of constitutionalism in acknowledging the existence of mutually competing interests. No particular stance and interests would enjoy an absolute status over another. In order to reconcile with the concept of constituent power, liberal constitutionalists should view the state of emergency as “a crucial tool enabling public reassurance in the short run without creating long-run damage to foundational commitments to freedom and the rule of law.”25 Constituent power, as the ultimate authority for the executive to act for the purpose of safeguarding commitments to liberal ideals, works alongside liberal constitutionalism to correct the state the exceptional situation and return it to normal.
The Emergency Regulations Ordinance (Cap. 241)
The ERO empowers the Chief Executive of the Hong Kong Special Administrative Region to enact a wide but exhaustive range of regulations26 “whatsoever which he may consider desirable in the public interest.”27 Any regulations made under section 2(2) would, moreover, be “continue in force until repealed by order of the Chief Executive in Council.”28 Moreover, any regulations enacted under section 2(2) would “have effect notwithstanding anything inconsistent therewith contained in any enactment.”29 When read together, sections 2(2) and 2(3) or 2(4) would suggest that unlimited power is bestowed upon the Chief Executive in the selected area(s) (s)he chooses to legislate on. Instead of being limited by any external restraints (say, for example, a judicial review action which, given the generality of the language under the ERO, is unlikely to succeed), the Chief Executive has unlimited power to legislate in any terms (s)he desires to be in the public interest. Moreover, the language of section 2(1) epitomises Schmitt’s conception of the exception. Instead of requiring any approval by any other institutional bodies, for example the Executive Council or the Legislative Council, the power to make regulations under section 2(1) only refers to the Chief Executive himself and himself only. This opens the possibility of a significant threat being posed to the liberties which citizens would ordinarily enjoy – in times of alleged emergency or public danger, civil liberties may be disregarded significantly (if not completely) and there are no executive or legislative checks against it.
Section 2(3) provides the only statutory method for terminating a regulation enacted under section 2(1) and 2(2) of the ERO. This means that the termination of any emergency regulation would, on a literal interpretation of the Ordinance, be at the absolute discretion of the Chief Executive in Council. Whilst any predictions to the Chief Executive’s use of the Ordinance is beyond the scope of this article, it is sufficient to note that the civil liberties of over seven million Hong Kong citizens would lie in her sole discretion. The lack of any principles or rules guiding the return to normalcy under the Ordinance further exposes the potential dangers to liberal constitutionalism posed by the Ordinance. Instead of being required to repeal any emergency regulations over a particular period of time, or inserting mandatorily such regulations with sunset clauses, the Ordinance provides no hints as to normalcy can and should be achieved (or returned to) once the emergency regulations are enacted. Any approaches to the Ordinance and the enactment of regulations must, for at least the purpose of restoring the stability and orderliness of the Hong Kong society, bear in mind the protection of a liberal constitutional order and the integrity of the legal system.
Constituent power is a powerful concept in illustrating the ultimate authority of the state and justifying the ruler’s authority to act. Schmitt’s wickedly ingenious illustration of the unlimitedness of the exception and its authority to respond to emergencies, whilst compelling, risks unnecessarily dichotomising the relationship between constituent power and liberal constitutionalism. Although a thorough examination of Schmitt’s conception and critique against it cannot be examined in the limited space of this article, it is clear that the concept of constituent power is not anathema to a theory of liberal constitutionalism: they can work together to uphold the liberal constitutional order, and safeguard it from falling into a permanent state of emergency.
 Martin Loughlin, “The Concept of Constituent Power” (2014) 13 European Journal of Political Theory 218, 219.
 ibid 220.
 Martin Loughlin, “Why Read Carl Schmitt?” in Christoph Bezemek, Michael Potacs and Alexander Somek (eds), Vienna Lectures on Legal Philosophy, volume 1 (Oxford; Hart, 2018) 63.
 A recent example is rumors of use of the Emergency Regulations Ordinance (Cap. 241) by the Hong Kong Government to deal with the ongoing protests in the city against, inter alia, (i) the attempted (now failed) legislation Fugitive Offenders and Mutual Legal Assistance in Criminal Matters Legislation (Amendment) Bill 2019 and (ii) alleged incidents of police violence since early June 2019. “Hong Kong leader Carrie Lam says protests will be addressed with local laws, amid rumors of emergency legislation” (Hong Kong Free Press, 27th August 2019) (https://www.hongkongfp.com/2019/08/27/hong-kong-leader-carrie-lam-says-protests-will-addressed-local-laws-amid-rumours-emergency-legislation/) (Accessed 27th August 2019). For an article discussing the controversial 2019 extradition bill and its perceived defects (as a result of, inter alia, excessive executive discretion in the extradition process and ineffective judicial control), see Thomas Yeon and Claudia Tam, “Extraditing suspects and safeguarding human rights – a tangled tale” LSE Law Review Blog (25th June 2019).
 Carl Schmitt, Political Theology (G Schwab tr, Chicago: Chicago University Press, 2005) 5.
 Hans Lindahl (2007), “Constituent Power and Reflexive Identity: Towards an Ontology of Collective Selfhood” in Martin Loughlin and Neil Walker (eds), The Paradox of Constitutionalism: Constituent Power and Constitutional Law (Oxford University Press 2007) 12.
 Owen Gross, “The Normless and Exceptionless Exception: Carl Schmitt’s Theory of Emergency Powers and the Norm Exception Dichotomy” (2000) 21 Cardozo Law Review 1825, 1829.
 Frederick Hayek (1973), Law, Legislation and Liberty, Volume 1 (London: Routledge and Kegan Paul, 1973) 92.
 ibid 93.
 (n 1) 225.
 Carl Schmitt, Constitutional Theory (1928; Duke University Press 2008) 61.
 ibid 62.
 (n 3) 51.
 (n 5) 6.
 ibid 6.
 ibid 11.
 David Dyzenhaus, “Constitutionalism in an Old Key: Legality and Constituent Power” (2012) 1 Global Constitutionalism 229, 233.
 Bruce Ackerman, “The Emergency Constitution” (2004) 113 Yale Law Journal 1029, 1044.
 (n 4) 13.
 (n 3) 64.
 William Scheuerman, “Emergency Powers and the Rule of Law after 9/11” (2006) 14 Journal of Political Philosophy 61, 64.
 (n 20).
 Section 2(2) ERO.
 ibid section 2(1).
 (n 26) section 2(3).
 (n 26) section 2(4).