The Dynamics of Proportionality: Constitutional Courts and the Review of COVID-19 Regulations.

Abstract

The COVID-19 pandemic has made it clear that even when using trusted legal tools, courts may run into challenging problems. Governments reacted to an unprecedented (at least in the context of post-WW2 era of fundamental rights) global crisis by adopting measures that drastically limited fundamental rights in order to protect the lives and health of many. Courts, of course, were entrusted with protecting fundamental rights against governmental overreach. The question was, how strict should the courts be when reviewing governmental acts. On the one hand, they could have relied on substantive proportionality assessment. This option, however was virtually ignored and most courts have opted for a deferential approach. This article analyzes both of these approaches, their strengths and weaknesses, but ultimately it argues that a third option - semiprocedural review - is the best way out of this judicial conundrum. Relying on comparative as well as theoretical arguments, it argues that semiprocedural review is the best way to deal with challenging empirical question - even under conditions of epistemological uncertainty.

The principle of proportionality (and its subsequent analysis) are staples of contemporary approaches to fundamental rights in modern democracies. They are recognized and applied by constitutional courts virtually all over the democratic world, save, perhaps, for the United States, and can be considered one of the basic building blocks of global constitutionalism.¹ As a result, proportionality is one of the most dissected, analyzed, and researched topics of constitutional law (or human rights law). It would seem redundant to write yet another text dealing with the question of “how do constitutional courts test proportionality,” even though more empirical analysis will always be welcome.

However, the global COVID-19 pandemic made clear that even when using trusted legal tools, courts may run into challenging problems. Governments reacted to an unprecedented (at least in the context of the post-WW2 era of fundamental rights) global crisis by adopting measures that drastically limited fundamental rights in order to protect the lives and health of many.² These measures included, inter alia, various forms of lockdowns (closing businesses, prohibiting public worship or other public gatherings), quarantines, social distancing, the wearing of face masks, the introduction of mandatory vaccination, and the deployment of contact-tracing technologies (by private or public entities).

Courts – including constitutional or apex administrative courts – had to find a way to respond to these measures, which were often challenged by those affected individuals or political opposition. However, in light of the high degree of uncertainty, conflicting interests, and the dynamically evolving scientific knowledge regarding the characteristics of the pandemic, courts struggled to define fitting standards for reviews of the governmental measures that restricted fundamental human rights. While it was clear (in most jurisdictions) that proportionality was the default approach, recourse to the standard application of the three-prong test was challenging, given the shift from rigorous evidence-based analysis to best available guesses, pressure to react swiftly to the clear and present danger, and the high cost of error. More specifically, the courts had to decide how much practical “bite” proportionality review should carry. Should courts maximize their role as guardians of fundamental individual rights and enforce stringent substantive standards, thereby potentially slowing down the deployment of measures designed to protect the public interest? Or should courts recognize the complexity of the situation, derived from the gravity of the risk to lives and public health and informed by their relative advantage over other branches in terms of their political legitimacy and access to expertise, and opt for a deferential stance? Traditional research suggests that the initial reaction of courts to challenges of this sort – for example, in the context of national security – leads them to adopt evasive techniques, such as resorting to threshold doctrines or delaying tactics (such as extending the time to issue a given decision), or by applying lax standards of review. This Article explores this judicial conundrum: how can, and perhaps how should (constitutional) courts respond to legal challenges of rights-infringing state measures enacted in the context of empirically dynamic and complex situations (often emergencies) while still fulfilling their judicial duty as guardians of rights?

We first briefly sketch the problems surrounding proportionality review in empirically complex and dynamic situations to answer this question. Afterward, we present the two intuitive responses to the conundrum: the strict and deferential approaches. We explore arguments both for and against these approaches. As we will argue, both approaches have some weaknesses that make them suboptimal. While choosing the lesser of two evils is often the dilemma that the courts face, we argue that, in this case, it is a false dilemma; there is a third option that supplements the aforementioned two. Courts – when reviewing governmental measures that limit fundamental rights during emergencies – may review the “procedural rationality” of these measures. This approach places emphasis on reviewing the policy formation process (including risks of capture or other biases) as well as the empirical evidence behind the measures. It is especially fitting for judicial decision-making in dynamic, complex, and polycentric situations where the public faces what appears to be a grave harm if the state remains aloof.

A. The Dynamics of Pandemics: A Challenge for Courts

The core of judicial challenges to pandemic measures has been intimately intertwined with the very logic of the standard three-step proportionality test used by constitutional courts.³ Therefore, even if the structure of the proportionality test is well-known, we will briefly sketch the steps and their role within the context of judicial review of pandemic regulations.

The starting point of proportionality analysis is that the state established a pressing (and legitimate) goal – such as protecting public interest and rights – the achievement of which would trigger rights-infringing measures. In the case of COVID-19, this was not much of a challenge, as relatively early on – albeit not necessarily early enough – evidence of the harm brought about by the pandemic was present. That being said, under this prong, the judicial process is tasked with siphoning out measures that, while related to COVID-19, are either motivated or likely to achieve other goals that may not be permissible. Upon ensuring that this is not the case, the well-known three-step proportionality test was applied by various courts. In the first stage, the court had to assess the limitations of the particular fundamental right through the lens of appropriateness (rationality). The question is whether the governmental measure that limits a certain fundamental right is “rational” in a practical sense – in other words, whether the measure can fulfill the legitimate aim put forward to justify the measure. In COVID-19 scenarios, the question would usually be whether the reviewed governmental measure can reasonably contribute to protecting the lives and the health of others and general public health.

The second step of proportionality analysis reviews the necessity of the measure in question. Specifically, the courts assessed whether there were alternative measures that would achieve the intended objective while – at the same time – being less detrimental to fundamental rights. If the aim pursued by the regulation could be achieved by alternative and less intrusive measures, it would generally be the constitutional duty of the legislator or the executive to use those alternative means. In the context of the COVID-19 pandemic, courts could have used this step to determine whether some alternative – for example, a less intrusive substitute to a general lockdown – would have been comparably effective in halting the pandemic.⁴

Finally, the third step of the proportionality test is the assessment of proportionality in the narrower sense of the word (balancing); that is, a kind of cost-benefit analysis that compares the relative weight of the conflicting gains and losses to rights or the public good. This would often turn into an argument about whether saving lives and protecting the health of an estimated number of people outweigh the limitations of the general public's rights and the associated negative externalities of these limits. It is relatively easy to surmise that such a balance requires some evidentiary basis to be meaningful.

Even though it is fairly easy to describe these steps in the abstract, their application has presented significant challenges for courts – some of which have proven to be especially relevant in the context of the COVID-19 pandemic. In the following section, we analyze these problems and try to provide recommendations on how to overcome them. However, the key issue to which all problems are tied is the question of the judicial branch’s attitude towards the government’s assessments and the role of such attitudes in proportionality analysis. Even outside the recent pandemic, constitutional courts have had to consider what kind of room for maneuvering they should leave to legislators or executive bodies and what conditions should be relevant when deciding whether or not to defer to legislative and executive bodies. Therefore, before we turn to answer the main questions of this Article within the specific COVID-19 context, we consider it important to at least roughly outline the debates surrounding judicial deference, its structure, and its main arguments. Thus, we first introduce the topic of judicial deference (or, in other words, the topic of the degree of scrutiny) within the context of the separation of powers. Afterward, we highlight the key problems of judicial deference within the context of the COVID-19 pandemic and explain how these problems might have influenced the behavior of courts that reviewed regulations designed to halt the pandemic.

I. Deference and separation of powers in general

The separation of powers (or its particular reflections) is perhaps the most important factor that can affect the degree of judicial scrutiny of legislation affecting fundamental rights. As Kriele, among others, has noted, the strictness of a fundamental rights review is closely linked to the issue of separation of powers. As judicial review becomes stricter, it reduces the latitude given to the legislature or the executive in assessing the optimal measure and judicial discretion regarding the appropriate manner of exercising their functions according to the country’s constitution. ⁵

Each branch of government is constitutionally entrusted with a particular function and is, or should be, endowed with appropriate powers and resources to carry out that function. In performing these functions, no branch of government can be wholly supplanted by another. Applying the proportionality test undoubtedly limits the discretion of the legislature or the executive quite significantly. This is evident in applying the necessity test and the related search for the means that interfere least with a given fundamental right. While the strict application of this test (and especially the strict insistence that the state bears the burden of proof to show that no other alternative is available to achieve the purpose) does not negate legislative or executive discretion, ⁶ it nonetheless significantly limits it. Therefore, in the context of the doctrine of proportionality (and in its practical application by constitutional courts), the discretion of the other branches, particularly the legislature, must be considered and respected. Otherwise, the principle of separation of powers runs the risk of being irreversibly undermined. ⁷ Consequently, some jurisdictions have followed the US approach and established a different degree of scrutiny in certain areas, to the extent it is understood that, within these areas, the courts do not enjoy a defensible institutional advantage over the legislature and the executive. In other areas, the risk of excessive or superfluous infringement of core rights is greater. Therefore, the judicial process should demand exacting justification from the elected representatives or the professional civil servants. In such jurisdictions, the emerging matrix reflects the importance of the right to the democratic process and the national set of values, the contextual characteristics of the social area regulated by the measures in question, and the particular institutional design of agencies governing that area (and its perceived ability to check against such excessive or superfluous infringement of rights). Where a case presents before a court a challenge, the resolution of which places the court at a relative disadvantage, the refusal to apply a strict proportionality test would not contravene the principle of practical concordance. On the contrary, if scrutiny of the legislative action in the performance of a particular essential function (for example, in the field of taxation) is too strict, a core aspect of the separation of powers could be destroyed, which could violate the principle of practical concordance as a consequence.

In practice, the separation of powers factor is particularly relevant in situations where the application of strict proportionality (especially the strict necessity test) would place overly stringent demands on other public authorities, in the sense that some measures in some contexts would be beyond the reach of the authorities to an extent that would not be easily reconcilable with the underlying design of the constitution. Excessive restrictions on the discretion of other branches in areas that are fundamentally within their powers, and where the judicial process may not be ideally suited to generating a well-calibrated balance between individual rights and the public interest, is likely to create unnecessary friction and potentially undermine public confidence. At the same time, adopting an overly cautious approach across the board would similarly undermine the constitutional structure and accord the legislature or executive the leeway to infringe upon core rights even in cases where such an infringement could have been avoided without incurring overly complex burdens.

In principle, the separation of powers as an argument against applying the proportionality test in a particular area can have two dimensions. We can call them 1) the dimension of democratic accountability and 2) the dimension of epistemological (in)adequacy. The first argument emphasizes the greater democratic legitimacy of the legislative and executive institutions as elected institutions. The second argument is based on the epistemological limits of judicial decision-making (compared to the other two branches). Essentially, this argument spotlights the practical ability of courts to evaluate complicated policy issues in cases where understanding the evidence requires expertise that rests with the bureaucracy or the legislature, and its findings, methods, and implications cannot be made easily accessible in a timely manner, through the judicial process, to the courts. During (public health) emergencies specifically, we consider the latter premise concerning the applicability of the proportionality test to be more relevant. The argument that the legislature and the executive are endowed with democratic legitimacy that constitutional courts lack is a general one; that is, it applies to the relationship between the constitutional court and the legislature in the field of human rights as a whole and does not have any special force with regard to the problem at hand, ⁸ save, perhaps, for the question of the price on an error (to which we will return in a moment). Therefore, in this text, we focus mainly on the problems related to the epistemological argument; that is, how the epistemological limits of courts (especially those that have become apparent during pandemic-related emergencies) shape the nature and standards of review.

II. The epistemological limits of courts in the context of pandemics

Proportionality analysis is, of course, a legal tool. The balancing stage concerns a hierarchy of values. One might thus wonder why we are talking about epistemological difficulty in the context of proportionality review. However, it is important to understand that at least the first two steps of proportionality analysis (suitability and necessity) are based on empirical questions and their answers. ⁹ In the context of the COVID-19 pandemic, courts that assessed the proportionality of governmental measures were required to ask themselves essentially empirical questions and find appropriate answers. Has the government established that a certain variation of lockdown works? Is there credible evidence that face masks (and which types) prevent or reduce the spread of coronavirus inside buildings? In the open air? Did the government have a “better” solution than it had adopted? Even the third step – the cost-benefit analysis – is difficult to conduct in practice if real-world evidence, consequences, and potential impact do not inform it. ¹⁰ One may hardly properly balance conflicting values when it is uncertain to what extent (empirically) these values are affected by the pandemic, governmental measures, etc.

Constitutional courts have faced similar problems in the past and across a wide array of issues. For a long time, for example, scholars have questioned the institutional capacity of courts to adjudicate fundamental social-rights claims related to highly complex social security systems or health care systems. ¹¹

But the problem within the COVID-19 pandemic context lies not simply in the fact that courts had to address empirical questions in order to perform their duties or that a certain area of the law was more complex in the sense that it required proficiency in methodologies related to the interaction of social processes. What is unique about the context of COVID-19 is that 1) both the challenges brought about by the pandemic and the scientific knowledge reflecting these challenges evolved very dynamically; 2) there was at times a lack or at least a perceived lack of consensus amongst the scientific community; ¹² 3) credible voices within the relevant professional communities stressed that an urgent regulatory response was needed and that the price of error in terms of public health would be considerable (i.e., a tangible risk of significant magnitude) and 4) the relevant regulatory response transcended a single specific area of law but instead formed a complex, wide-ranging, polycentric scheme. These factors created an almost unprecedented situation and a real judicial conundrum.

Many authors thus share the conviction that because of the complexity of, in particular, the empirical questions relating to the costs of lockdowns, it is almost impossible to reach a confident conclusion regarding the proportionality of such measures. In this regard, Kai Möller mentioned the relevance of the “Swedish approach” and reflections of it within necessity testing. He posited the question of:

whether lockdowns were necessary in that there was no less restrictive but equally effective alternative. Critics of lockdowns often point to the ‘Swedish way’ of dealing with the pandemic, which focused less on legal prohibitions and more on recommendations, or the approach advocated by the Great Barrington Declaration, which proposed ‘shielded protection’ of those most at risk from the virus but otherwise no restrictions, in order to build up herd immunity which would then also protect the vulnerable. But whether these approaches really are equally effective is empirically unclear. ¹³

Another example of a complex and unclear (at least for a certain amount of time, before enough studies could be conducted) empirical question may be to what extent vaccination could prevent the spread of the virus, related to reviews of mandatory vaccination measures. The complexity of this issue is, of course, connected to the fact that anti-COVID measures have not usually been binary but rather occur on a spectrum. This further complicates any assessment of their relative effectiveness. Such examples highlight the difficulty presented by uncertainty, which society usually addresses via (scientific) experimentation. But experimentation requires time, and the COVID-19 attack was immediate. Under these circumstances, is it the judicial role to insist on such experimentation, given the price they may exact?

B. Strict Review or Substantive Deference: A False Dilemma?

As noted, during the COVID-19 pandemic, some commentators have argued that the general problem of the separation of powers is key to understanding the institutional dynamics of responses to the pandemic. ¹⁴ Many authors focused mainly on the problem of executive overreach, ¹⁵ but the search for the appropriate role of courts has also been considered a crucial issue. ¹⁶

In most jurisdictions, the challenge of finding “an appropriate role for courts” boils down to the question of “how strict should an appropriate standard be?” While excluding courts from the equation might be a distinct theoretical notion, ¹⁷ it is hardly feasible in liberal democracies that permit judicial review of government actions. ¹⁸ In the context of the COVID-19 pandemic, the two options of strict and deferential review modes bring to mind the mythical Scylla and Charybdis.

At first glance, the strict proportionality approach looks appealing – at least from the perspective of human rights protection. Since regulatory responses would limit rights significantly, and since the risk of overreach is palpable (given the sense of impending grave danger), it would seem preferable to calibrate the judicial process to ensure that any response is narrowly tailored to address the risk. In the same vein, judicial interventions may enhance the quality of regulatory measures (as well as the policymaking process itself) and the level of public trust. ¹⁹

On the other hand, there are compelling arguments against strict substantive review of various governments’ pandemic measures. These arguments are, of course, inherently tied to three of the four general problems that we have outlined above, namely that: 1) the pandemic situation evolved very dynamically, 2) there was at times a lack or at least a perceived lack of consensus amongst the scientific community, and 3) the regulation enacted in response to the pandemic was wide-ranging, complex, and polycentric. We may add a fourth element that is more specifically relevant to the issue of legitimacy: the cost of judicial overreach is considerable. ²⁰

In the context of the COVID pandemic, a judicial decision that would employ an overly strict understanding of proportionality (and mainly of necessity) regarding a complex empirical question would be highly problematic for several reasons. In addition to the general problems of legitimacy and epistemology outlined above, time itself is a crucial element. In the pandemic emergency context, courts were often forced to review executive orders under severe time pressure. Complementing the executive bodies that were expected to adopt policy solutions quickly and effectively, the courts were asked to conduct swift reviews ²¹ of these far-reaching and often unprecedented executive acts. This was all the more problematic when, besides the time pressure, substantive decisions would often need to be reached on uncertain empirical grounds.

Under such conditions, both the executive and the judicial branches could make mistakes (which are usually discerned with the benefit of hindsight). However, unlike executive overreach, an overly invasive approach by hurried, hasty, and empirically uncertain courts could potentially lead to a public health care disaster, the deaths of many people, and irreversibly damage the health of others.

This last argument follows – to a certain degree – the logic of the precautionary principle. In its proper sense, the precautionary principle is applied to “positive” governmental policy changes, such as changes in the status quo. In its traditional form, the precautionary principle states that under conditions of uncertainty, the policymaker should be allowed and may even be obligated to take steps that prevent (potential) irreversible future harm. ²² In this sense, it has been applied, inter alia, to environmental and health issues. ²³ The precautionary principle per se does not deprive the state of the duty to respect fundamental rights nor deflect the duty to conform to the principle of proportionality. More specifically, the judicial function must ensure that the precautionary principle does not invade its chambers, lest it undercuts the judiciary’s checking role. At the same time, however, we cannot ignore the epistemological challenges that, in the context of COVID-19 pandemic emergencies, might make a substantive assessment of proportionality virtually impossible. Courts should also reflect this fact when they assess whether the executive branch (or, in some cases, the legislature) has fulfilled its constitutional duties.

To wit, it is important to stress that we do not argue that courts reviewing emergency pandemic measures are bound by the precautionary principle in its proper sense. Our argument is rather that while the precautionary principle often guides legislators and executive bodies, the courts have a complimentary duty to appreciate and reflect the empirical uncertainty that underlies the use of the precautionary principle and tailor their reviews accordingly, thereby shifting their attention to focus on best available proxies with which to bridge the epistemological gap that hinders classic application of proportionality analysis. ²⁴

When we apply this logic to the question of the judicial review of pandemic measures, it follows that the courts should not substantively intervene if the author of the measure under review acted on uncertain evidence unless there is strong evidence that they are wrong or unless the measure is on its face clearly disproportionate, and other, less disproportionate alternatives, have not been convincingly shown by the government to fall significantly short of achieving a realistic goal. Under conditions of uncertainty, it is, of course, possible that measures will later retrospectively be viewed as overzealous (or unsuitable, unnecessary, or disproportionate within the context of proportionality). Still, the combined effect of epistemological limits, time pressure, and the price of “judicial overprotection error” dictates that we should generally navigate towards an attitude that is aware of the risk of governmental overreach but focuses on the procedures taken by the government to ensure against such an overreach, and is attentive to the balance of evidence as it stands when the decision is taken. Simply put, if courts are too strict in their proportionality testing of governmental measures within the context of the pandemic, they might prevent the government from saving the lives and the health of many people. Arguably, because of these reasons, no constitutional court has – to our knowledge – adopted a substantively strict version of proportionality testing that would replace the empirical assessment of the policymaker with that of a court.

Does it follow that courts are, therefore, obligated to adopt a position of substantive deference (and not much else)? We argue that this option is also suboptimal. Taking the “navigate towards the safer option” logic would essentially make judicial review meaningless, as fundamental rights that found themselves in the way of an executive pandemic measure would almost surely always lose. Even though strict judicial review of executive measures might impair the executive’s advantages in managing emergencies (such as decisiveness, swiftness, or capacity for action), it is not prudent to leave fundamental rights at the mercy of the executive, simply because the precautionary principle would direct their efforts to avoid risks. On the contrary, especially in times of emergency, the traditional justifications of judicial review – bounded rationality – seem most relevant. As has been noted before:

judicial review could be seen as a precommitment for more sober policy formation and enforcement processes. Policy formation and enforcement processes are subject to short-term (but nonetheless considerable) pressures by heated emotional reactions to threats and provocations. Moreover, policy formation and enforcement processes tend to prefer short-term gains over long[-]term considerations in part because politicians stand for election (and bureaucrats stand for promotion) in relatively short time cycles and therefore are usually evaluated by gains (or losses) they were able to achieve (or prevent) during their term of office. This tilt calls for a mechanism designed to compensate. Such a compensating mechanism is especially needed in times of emergency. ²⁵

If courts adopt a deferential stance, society may question whether they fulfill their primary role as guardians of rights. For example, as Thulasi Raj observed in her report on India’s response to the pandemic, the highly deferential approach of the court may have damaged public trust in the courts almost as badly as it had during the 1970s emergency period. ²⁶

Despite this, many (constitutional) courts have chosen the deferential option. Besides India, other authors have pointed out many courts that have practically refused to challenge COVID-19 measures on substantive grounds. For example, the Belgian Council of State has been described as giving the government “‘the widest possible discretion’ to protect the public health and safety in this ‘unseen and most serious’ health crisis” while reviewing its lockdown measure. ²⁷ In the case of Canada, as Paul Daly put it:

Canadian courts have repeatedly emphasized the importance of deference, often in the context of applications for injunctive relief for allegedly over-inclusive public health measures but also when challengers have complained that public health measures are under-inclusive. Where courts have intervened to interfere with pandemic measures, they have done so at the margins and not called into question governmental policy decisions on how to combat COVID-19. ²⁸

Even though some measures have been struck down, even by “deferential” courts, the reasons for the judicial intervention were not substantive but rather a lack of legal standing or an inappropriately chosen legal form of the measure. ²⁹

C. “Semiprocedural” review as a distinct third-way

Because of the challenges presented above, it is somewhat understandable that – perhaps seeing no other feasible option – courts have generally drifted towards deferential review. But as we will argue, the tertium non datur choice between a substantively deferential and a substantively strict review is a false dilemma since there is a distinct third option based on semiprocedural rationality review. This third option is faithful to the notion of judicial review as a “Socratic contestation.” In the words of Mattias Kumm, “The point of judicial review … is to institutionalize a practice of Socratic contestation legally. Socratic contestation refers to the practice of critically engaging authorities in order to assess whether the claims they make are based on good reasons.” ³⁰

As Ittai Bar-Siman-Tov has noted, in recent decades, courts have developed:

judicial doctrines that integrate an examination of the legislature’s decision-making process into the judicial tests for determining the permissibility of constitutional infringements. Moreover, courts themselves create certain heightened procedural requirements when particular rights or values are infringed. Judicial review of the legislative process in these cases does not completely supplant the traditional balancing tests that courts use to determine the permissibility of infringements. Rather, the procedural review typically supplements the traditional balancing tests and is integrated into them. ³¹

These heightened procedural requirements might include the quality and record of parliamentary debates, ³² the existence and quality of empirical evidence, ³³ or even, in an intra-judicial context, judicial treatment of the case law of the ECtHR. ³⁴ It may also include a review of the existence of processes to refine the evidentiary basis in order to ensure that the best available guesses are not entrenched but are continuously challenged, knowing that the evolution of the evidentiary basis will likely lead to revisions of the regulatory measures (and exposure of assumptions proven wrong).

Semiprocedural review, despite being conceptualized by Ittai Bar-Siman-Tov and others, is not a clear-cut doctrine but rather a trend or umbrella term that refers to a relatively broad set of judicial practices in various jurisdictions. ³⁵ The “hard look” review – a doctrine developed by the US federal courts in the 1960s and 1970s ³⁶ – is one of the related practices. Around this time, Mathews notes, judges on the D.C. Circuit Court of Appeals began to intensify the scrutiny associated with “arbitrary and capricious” review, demanding that agencies took a “hard look” at the salient issues (they were not satisfied with a mere lack of arbitrariness) and engaged in reasoned decision-making. ³⁷ The doctrine was also adopted by the US Supreme Court, among others, in the State Farm decision. ³⁸ The “hard look” review very much resembles semiprocedural review – or rather, it is an example of the same phenomenon, as it requires that the decision-maker’s fact-finding, reasoning process, etc. are plausible and sound. However, it is only applied to the decision-making of administrative agencies and not to legislative action. ³⁹ This factor is not very relevant in the pandemic context because most of the relevant regulations were adopted by administrative bodies, often governments. But from a broader perspective, it is a relevant distinction. The notion of semiprocedural review as an emerging trend at European constitutional courts and beyond is not by definition limited to the review of administrative acts. Legislators can – as a rule – also be subject to heightened semiprocedural scrutiny. ⁴⁰

Still, decades of experience with the “hard look” review offer us a chance to reflect on its problems and possible pitfalls that might be prudent to avoid when developing a proper place for any form of semiprocedural review in a particular jurisdiction. Mathews noted that with regard to the State farm case:

For example, the challenge in State Farm was not that the agency regulated too much—the classic complaint in proportionality—but that it regulated too little. And because American courts do not credit post hoc rationalizations of agency action, agencies’ justifications must be contemporaneous with their actions. Some scholars have long argued that hard look review has contributed to the “ossification” of rulemaking in the United States by imposing unreasonable justificatory burdens on agencies that complicated the regulatory process. ⁴¹

It is important to stress that semiprocedural review (at least the kind of semiprocedural review we analyze here and suggest as a helpful judicial tool during emergencies) does not reject or fully replace proportionality analysis. ⁴² The shift to semiprocedural review, even though it may be forced by external factors such as time pressure and epistemological limits, does not, in principle, relieve policymakers of the duty to conform with the proportionality requirement. Nor, consequently, does it relieve courts of their duty to review it. The core logic of semiprocedural review, as we understand and advocate it, is not based on abandoning proportionality (or similar standards that the courts might have developed) but on asking the key questions differently. Indeed, as Alberto Alemanno observed, “what is central to this new form of scrutiny [semiprocedural review] is the instrumental use of the evidence gathered during the decision-making process in order to verify the adequacy and quality of that process.” ⁴³ Thus, unlike the “undirected hard look review,” the nature and “direction” of the questions concerning rationality of the decision-making process, asked by the courts, would be guided by the underlying standard (or test) applicable in a particular jurisdiction. Courts that have adopted proportionality analysis as a go-to tool for assessing constitutional conformity of legal acts can still use their standard three-step test. Instead of looking for a substantive answer, they would look for a semiprocedural one. For example, the courts will not try to determine whether the policymaker had a “better” (that is, more comparably effective and more human-rights-friendly) alternative at their disposal. Instead, it would ask whether the policymaker 1) brought forward enough reliable empirical evidence to conclude that no such alternatives existed, 2) adopted the measure in question in a rational process where the question of necessity had been rigorously addressed and the risk of overreach or potential hidden agendas, motives or impacts were seriously discussed, and 3) the policy-maker has in place a mechanism to search for less restrictive measures as the evidence is actively developed. The court would, of course, need to be context-sensitive and adjust its demands according to the level of empirical uncertainty at the time. As noted, a cy-prē`s version of the test’s third prong would allow the court to press the government when the measure is prima facie disproportionate, and there is little evidence that less drastic measures would fail to achieve a significant portion of a realistic goal.

By shifting from a “choosing the optimal solution” mindset to pressing for procedural rationality – that is, the quality of deliberations and the use of reliable empirical evidence – constitutional courts can avoid the pitfalls mentioned above, particularly their epistemological limits in times of emergency. Under this course of action, a (constitutional) court, as the proverbial Socratic guardian of rationality, does not focus on assessing the outcome of the political process (provided, as noted, that the set of measures chosen is not egregiously disproportionate to the alleged risk). The burden of selecting appropriate measures is left to the political institutions – typically the government (as the policymaker) and the parliament (which provides the necessary legal framework for the chosen policies). These institutions are accountable, via elections, for the political solutions they devise (at least those solutions that can be defended as procedurally rational). They also have resources and processes at their disposal to better address the epistemological challenges (and, therefore, can be expected by courts to demonstrate that they harnessed such resources and processes). Moreover, semiprocedural review, if approached correctly and rigorously (more on this below), could provide an almost ideal structure for addressing the epistemological problems outlined above.

When courts review the rationality of the decision-making process or the quality of the empirical evidence, they do precisely what they are institutionally and personnel-wise created to do:

A court’s activity is not focused on the active construction of elaborate theories but on a considerably more pedestrian form assessing the reasons presented by others to determine their plausibility.… [T]his engagement takes place as a public procedure leading to a public judgment, while institutional rules relating to judicial independence ensure that it is immunized from the pressures of the ordinary political process. ⁴⁴

By turning to procedural rationality review, constitutional courts embrace the “culture of justification,” according to which it is the role of the courts to ensure that every act of the state that affects a person is substantively justifiable. ⁴⁵

Of course, the devil lies in the details. Even if we accept the logic and the underlying justification of semiprocedural review, the question remains precisely what content courts would fill it with. In this regard, it is important to stress that semiprocedural review does not simply mean a deferential review of the rationality of the decision-making process. While we cannot rule out that some courts might adopt a very deferential understanding of semiprocedural review, such an approach is by no means the only one. In the semiprocedural review of emergency measures, we argue that such an approach would be inappropriate. As we have already indicated above, the shift to semiprocedural review (or the “good practice of semiprocedural review”) should not be viewed primarily as a shift in strictness but rather as 1) a shift in the structure and the focus of the review, and, consequently, 2) a slight shift in the wording of the concrete judicial test. The strictness of semiprocedural review may vary based on how ambitiously a particular court works with the concept of rationality. The practical approaches (while they still could be categorized as semiprocedural review) could include both a deferential standard for what would constitute a reasonable decision-making process, reasonable evidence, and a rigorous maximalist standard that would be very challenging to meet in a practical situation. If a semiprocedural review is seen as a functional and meaningful alternative to substantive deference, we argue that even in emergencies, the courts should push policymakers to justify and substantiate their measures rigorously. The resulting standard might come close to “justify[ing] [these actions] as rigorously as practically possible.” This is not an absurdly high standard. During the COVID-19 emergency, executive bodies severely limited the rights and freedoms of numerous individuals. Hence, the quality of the decision-making process and the reliability of empirical evidence should correspond to the gravity of the situation. Moreover, to the extent there is an indication that the executive or the legislature may seek to deploy severe measures in a context that raises concerns of capture or mission creep, namely that the measures may be deployed in service of goals other than the sole mission of combating the acute emergency, the court should undoubtedly apply an exacting standard when reviewing the procedure.

If understood like this, a semiprocedural review can dispel the fears that rights will always be lost in a conflict of values and principles during emergencies. A demanding standard of semiprocedural review can identify more obvious executive overreach, such as mission creep (which is by no means limited to emergency times), and it can also meaningfully address the finer points of the specific measure under review. Before we even turn our attention to specific cases of semiprocedural review in the COVID-19 pandemic, this point can be demonstrated in a well-known example of semiprocedural review: the Hartz IV judgment ⁴⁶ of the German Federal Constitutional Court. While this case concerns the area of positive duties of the state with regard to social rights, it shows the potential of semiprocedural review. In German constitutional doctrine, policymakers have a recognized obligation to justify their acts, especially those that limit fundamental rights. While the obligation to capture the reasons for a legal provision in advance and in a formalized form is not generally recognized, the obligation to have such reasons available and readied to present them to the Constitutional Court in the event of a constitutional review does indeed exist. In the Hartz IV judgment, the German Federal Constitutional Court reviewed the newly set subsistence minimum from the point of view of its consistency with human dignity.

While the German Federal Constitutional Court did not claim the legitimacy to set a constitutionally acceptable subsistence minimum, it subjected the legislature to a relatively stringent procedural rationality test:

(1) whether the legislature has considered and described the objective of ensuring an existence in human dignity doing justice to Article 1(1) GG in conjunction with Article 20(1) GG; (2) whether it has, within the boundaries of its latitude, chosen a fundamentally suitable method of calculation for assessing the subsistence minimum; (3) whether in essence, it has completely and correctly ascertained the necessary facts; and (4) whether it has kept within the boundaries of what is justifiable within the chosen method and its structural principles in all stages of calculation with plausible figures (consistency requirement). ⁴⁷

The legislature must disclose its thought process and the methods and stages of calculation employed in the legislative procedure. If it fails to do so, the lack of rationality and transparency leads to a declaration of unconstitutionality. ⁴⁸

What is important with regard to the potential strictness of semiprocedural review is that the German Constitutional Court was not content with the mere fact that the legislature devised and put forward a fairly sophisticated statistical model to calculate the subsistence minimum ⁴⁹ but engaged with the statistical model in a very detailed manner (such as the inclusion of specific expenses, the calculation of benefits for children as a certain percentage of adults, etc.) and, due to a lack of empirical evidence, was unconvinced by certain very specific solutions and justifications put forward by the legislature. ⁵⁰ Although dealing with a substantially different topic (and thus not directly analogous to the problem at hand), the German Constitutional Court has nonetheless shown that semiprocedural review is not just another name for deference but a standard of review with potential bite.

Semiprocedural review, which we paint as a third distinct alternative to judicial deference and strict substantive proportionality analysis, did not remain a purely theoretical notion in the case of the COVID-19 pandemic. Despite the general tendency to apply a deferential approach, decisions by the Czech and Austrian constitutional courts and a shift in Israeli Supreme Court case law provide very interesting examples.

The Austrian Constitutional Court declared two provisions on the obligation to wear face coverings (these provisions were issued during May 2020) illegal by two judgements issued in October (the provision concerning the general to wear face masks) and December 2020 (the specific obligation concerning schools) respectively, as the competent Federal Minister had not explained their necessity sufficiently in the preparatory files of the respective ordinance. ⁵¹ In the latter judgment, ⁵² the Austrian Constitutional Court reviewed the obligation to wear face masks in schools. While it did not try to assess whether such an obligation was necessary for the protection of public health, it stressed that the process of formulating any pandemic measure had to be transparent. It stipulated what kind of information basis a given directive had to incorporate, and it emphasized that policymakers must demonstrate that they had weighed all relevant interests.

The Czech case is perhaps even more interesting because the Czech Constitutional Court, despite its initial reluctance to challenge COVID-19 measures, eventually shifted towards procedural rationality review and even, to an extent, reflected on the reasons for this shift. In Pl. ÚS 106/20, ⁵³ the Czech Constitutional Court was faced (not for the first time) with the question of the constitutionality of a type of lockdown (closing down selected shops and numerous service providers). Technically, the lockdown measure was formulated as a general prohibition of the in-person sales and provision of services, complemented by a list of exceptions. The Czech Constitutional Court reviewed the challenged measure from a point of view focusing on the prohibition of discrimination and limitations on the right to engage in enterprise. Even though the court considered the question of strictness of review, it essentially based its approach on basic Rechtsstaat (rule-of-law) principles. ⁵⁴ It argued that the state and its institutions have a duty to adequately justify acts that restrict fundamental rights and freedoms, even during a state of emergency:

[T]he requirement of a rational and discernible justification for a measure affecting fundamental rights, in a manner which produces different effects among comparable subjects, is an immanent part of the test of discrimination, i.e., the assessment of whether the difference in treatment is sufficiently justified and proportionate. Only in such a case can the difference in treatment be regarded as justified. Under the rule of law, it is inconceivable that any act of a public authority that interferes with fundamental rights should not be rationally and convincingly justified, or at least that that justification should not be discernible in the context of a subsequent judicial review. ⁵⁵

For the purposes of this Article, it is notable that the Czech Constitutional Court tried to address many of the epistemological challenges we formulated in the previous section, such as the dynamic and evolving nature of the pandemic, the surrounding empirical knowledge, and the scientific debate. It also acknowledged that the “time factor” might influence the possibility (and quality) of the measure’s justification:

The specific requirements for the rationality of the solution or the requirements for the completeness and reliability of the justification… depend on the concrete factual situation and must reflect the reality. Thus, when reviewing a legal regulation regulating the rights and obligations of persons, the Constitutional Court must also reflect what information the public authority in question could and should have had at its disposal and what factual situation it was in when formulating the specific challenged measure. It is thus obvious that the Constitutional Court can place different (higher) demands on the rationality and soundness of a law which was adopted in “calm times” after extensive parliamentary debates, which was accompanied by a reasoned report, and which was not adopted in under time pressure caused by objective external circumstances. On the other hand, when reviewing a normative measure that, for objective reasons, had to be adopted “overnight”, so to speak, and which responds to a complicated factual situation whose development is difficult to predict, a certain degree of restraint is appropriate in view of its subsequent judicial review. ⁵⁶

Similarly, the Czech Constitutional Court also reflected on the problem of scientific consensus (or its perception):

It is not the task of the Constitutional Court to require the government, in the context of constitutional review, to find and perfectly justify the (rather hypothetical) “optimal solution” and the optimal distribution of burdens associated with the restriction of the fundamental rights of certain groups of the population, if there is no practical consensus even among experts on the assessment of the current situation and the forecasts of its possible development (the requirement for justification must not be excessive). However, from a constitutional point of view, the other extreme cannot be admitted either. In laconic terms, even practical uncertainty and a lack of perfect information do not mean that the government can do anything and rely only on instinct or political compromise. Indeed, the government’s decision must be based on expert advice, reflect the maximum available knowledge of the disease and its spread. It is entirely at the government’s discretion from what sources and in what way it will draw that information, and in that respect the Constitutional Court must exercise great restraint. However, in view of its duty to protect fundamental rights, it must insist that these reasons, which demonstrate the need for (just such intensive) interference with fundamental rights by means of a government decision (emergency measure), must be discernible, which specifically means that they should be publicly available. It should be borne in mind that any crisis measure is a political decision, which must of course be based on expert evidence, but the responsibility for it lies with the government, not with its expert advisers. At the same time, the government must consider not only the specific expert evidence at its disposal but also the overall context and the impact of its measures on other areas of social life, both in the short and long term. ⁵⁷

By reflecting on the dynamics of pandemics and the evolution of scientific consensus, the Czech Constitutional Court (perhaps unwittingly) essentially subscribed to the dynamic model of judicial review during emergencies. ⁵⁸ Initially, an emergency is usually characterized by a lack of information and interference with rights lasting only a short time, so the courts should be deferential in their review. However, a continuing emergency (coupled with lasting limitations of rights and evolving scientific knowledge) justifies a shift to a more active role for judges. ⁵⁹

It is significant that the Czech Constitutional Court (at least in this case) did not stay at the level of general proclamations but actually applied the procedural rationality review with a certain bite. This is in contrast to the approach of the Belgian State Council, which “took this position [procedural rationality review] in theory, but showed more reluctance in practice.” ⁶⁰ The Czech Constitutional Court, on the other hand, took into account that the government (by February 2021, a year into the pandemic) had had sufficient time to give more thought to the measures and justify them properly, unlike in March 2020, when the pandemics had just begun. The Czech Constitutional Court even stated that there had to be a certain correlation between the intensity and justification of prohibitions and the passage of time. ⁶¹ The rationale for this stricter requirement was both the fact that the government had had much more information, practical experience, and time to think through and systematically justify the challenged regulation but also the fact that long-term, repeated interference with a fundamental right (in this case, the right to engage in enterprise) is much more invasive and “painful” than a short-term, temporary restriction.

As noted above, the government formulated the lockdown as a general ban on all retail sales and services while simultaneously providing numerous exemptions, which resembled a “telephone directory” (36 in total). The Czech Constitutional Court emphasized that, in light of the procedural rationality review (even though the court itself did not use this term), this approach’s fundamental shortcoming was that no relevant source could demonstrate what basis the government employed to arrive at this particular solution. In this respect, the government did not refer to any relevant scientific sources, although the Constitutional Court repeatedly invited it to do so. The Czech Constitutional Court stressed that it was not enough to present the mere assertion (based on foreign scientific sources) that reducing people’s movement and limiting interpersonal contact in shops and public establishments was essential to stopping the spread of COVID-19. Although the court accepted this assertion as prima facie rational, it noted that the government provided no evidence (or even a good faith attempt to obtain it) as to whether a comparable objective could not have been achieved by using less restrictive measures. Similarly, the government’s references to research results from the USA merely illustrated that restaurants, gyms, and cafés were among the most problematic locations (as regards transmission of COVID-19), which, however, clearly missed the point of the crisis measure under consideration. The reference made by the government to the importance of restricting the association and assembly of persons was also considered logical by the Czech Constitutional Court, but it did not feel that the government had put forward enough evidence as to why it had resorted to this particular form of ban on sales and provision of services.

Another question was what specific procedural obligations the general “obligation to justify” entailed. The Czech Constitutional Court was convinced that it was unnecessary for the justification to be captured ex ante at the time of adopting the measure in question in a formal document. However, it stressed that any public authority (including the government or the parliament) must have the relevant reasons and supporting documents available and, at least in the event of a review of the constitutionality of such measures, must be prepared to submit them to the court that has jurisdiction. If it fails to do so, the court will simply hold that the measure in question constitutes an arbitrary limitation of a fundamental right.

A similar shift in the Israeli Supreme Court case law occurred at a similar time. It is important to note the context that Israel faced around that time (February 2021); it was facing the COVID-19 crisis and a political crisis with constitutional implications. ⁶² The courts had reason to question whether the government may harness the COVID-19 crisis as a justification to curtail demonstrations more than was necessary. Of course, this made any deferential treatment of governmental measures prima facie suspicious. While, in general, the courts (and the legal advisors of the government) would not have any reason to distrust the governmental COVID-19 policies, when it came to matters relating to the political process, this attitude was less sustainable; the government was incentivized to over-value the risk of COVID-19. A case in point addressed the limits placed on assembly (above and beyond the requirement that demonstrations be held while respecting a 2-meter gap between demonstrators). Interestingly, this suspicion of the government’s motivation found its expression in a kind of semiprocedural review. The Israeli Supreme Court, while reviewing the limitation of the right to assembly, ⁶³ explicitly stated that:

Against the gravity of this harm [to rights] stands a benefit whose exact degree is unknown and unproven… As the respondents themselves have stated, they do not have any data on the extent of infections in demonstrations. Thus, the attempt to hinge on to the decrease in general morbidity after the imposition of closures, as a fact justifying the imposition of restrictions relating to demonstrations, suffers from the fact that it does not indicate a proven causal link between the two. ⁶⁴

Due to the lack of evidence provided by the government, the Israeli Supreme Court held that the measure did not satisfy the “near-certainty of harm to the public wellbeing,” and the limitation of fundamental rights was thus found to be unconstitutional. ⁶⁵

A similar formulation can be found in a subsequent decision of the Israeli Supreme Court concerning the prohibition of entering Israel. ⁶⁶ The Court openly criticized the procedural rationality of the government’s decision-making process:

During the hearings held in the petitions, it became clear to us that the process for adopting the regulations and restrictions set forth therein also suffered from a lack of a relevant factual infrastructure. As is well known, any decision of an administrative authority, including a decision to enact secondary legislation, must be based on a sufficient factual basis. From the arguments heard before us, it became clear during the discussions that the government does not have any data on the number of citizens abroad seeking to return to Israel. This basic data, which could have illuminated the extent of the expected infringements, was not available to the government during the entire period in which the decisions were made and not even after the filing of the petitions and the holding of hearings on the petitions. ⁶⁷

Finally, in late 2022, a similar approach was adopted by the German Bundesverwaltungsgericht in two cases that reviewed COVID-19 measures in Saxony and Bavaria, respectively. ⁶⁸ In the Bavarian case, a rather strict lockdown was imposed, and the Bundesverwaltungsgericht declared that it contradicted the proportionality principle. The court emphasized that, in assessing the necessity of measures to protect against COVID-19, the responsible agency must enjoy a certain maneuvering space. However, its assessment must be based on sufficiently sound grounds, and the outcome must be plausibly justified. The court has explicitly acknowledged the relevance of empirical uncertainty in proportionality assessment and attempted to formulate a suitable approach to overcome it:

If the encroachment serves to protect weighty constitutional goods and if, in view of the actual uncertainties, it is only possible to a limited extent for the legislator to form a sufficiently certain picture, the constitutional review is limited to the justifiability of the legislative forecast of suitability … A measure is deemed to be appropriate if, in the overall weighing of the severity of the encroachment and the weight and urgency of the reasons justifying it, the limit of reasonableness is still observed. An appropriate balance must be struck between the weight of the encroachment of the measure and the objective pursued as well as the expected achievement of the objective. ⁶⁹

While the Bundesverwaltungsgericht accepted that Bavaria had a worse epidemiological situation than the rest of Germany at the time the lockdown had been ordered, and this fact was substantiated by evidence, including data from the Robert Koch Institute, it was not convinced that a sufficiently sound justification extended to the actual measure (lockdown) in question:

The ban on leaving one’s home to spend time outdoors could therefore be justified only if it could itself make a significant contribution to achieving the goal. In estimating this contribution, the legislator had - as in the examination of necessity - a factual margin of appreciation; however, the defendant would have had to plausibly demonstrate such a substantial contribution in the factual instance. .. . [The legislator] has not demonstrated before the administrative court that the ban could make a relevant contribution to reducing cross-household contacts. ⁷⁰

The above-summarized decisions, while not representing the majority of cases where the court was significantly more deferential by resorting to a host of threshold doctrines, nonetheless offer a taste of the potential of semiprocedural review but also highlight some problems. Perhaps the most important inherent advantage of the semiprocedural judicial review evident in the pandemic context is that it provides a viable model of judicial review under conditions of uncertainty. Strict substantive review requires the courts to set a specific and empirically grounded threshold for the contested measure, which is simply not feasible in such conditions. The reactive nature of semiprocedural review allows courts to assess the rationality or plausibility of the governmental policy, including the quality of empirical evidence, the quality of the process leading to its adoption, etc. Specifically, it also allows courts to assess how prudent the policymaker was when applying the precautionary principle. Questions such as, “Are we really in a situation of scientific uncertainty?”, “Is it a case of reasonable uncertainty?”, “What are the limits of the uncertainty?”, or “Did the policymaker provide enough evidence to substantiate the application of the precautionary principle?” can all be answered within the boundaries of semiprocedural review.

Here, the aforementioned problem of strictness with respect to semiprocedural review comes into play. As we have stressed above, the notion of semiprocedural review itself does not determine any particular strictness of rationality review. Its practical application by the courts may range from a deferential understanding of reasonability (almost anything goes) to a demanding pressure on rigorous reasoning and evidence-based policymaking (of the Hartz IV kind). A few scattered examples from the case law of the three courts we have covered will provide no conclusive answers. However, it seems very much possible to devise a fairly challenging standard, even in the context of the pandemic. The Czech Constitutional Court, for example, was not convinced by the government’s references to foreign scientific studies and by general arguments about the necessity to limit the movement of people (which, in all fairness, might pass a more deferential version of semiprocedural review) and demanded a more nuanced and rigorous justification and evidence. We must concede, however, that the meaningful use of semiprocedural review of this kind creates certain challenges for courts and judges. At a personnel level, a proper evaluation of empirical evidence, scientific studies, and the rationality of the decision-making process would demand that the judges adjust their mindset ⁷¹ or, ideally, even undergo specific training on the methodology of empirical sciences. At the court level, it would be more advisable to use oral hearings, ⁷² correspondence with amici curiae, and other similar tools to realize the potential of semiprocedural review fully.

Another general argument for semiprocedural review is that it indirectly creates pressure on the legislature (or the executive) to follow good practices, employ evidence-based decision-making, and engage in a transparent and rational debate. This is also relevant in the COVID-19 context: this was emphasized by the Czech Constitutional Court in the judgment summarized above. This argument is less relevant in jurisdictions with highly competent public service personnel and a sophisticated political culture; however, this prospective “educational” aspect of semiprocedural review cannot be underestimated in many jurisdictions.

One of the issues not tied to the semiprocedural review in general but, rather, specifically to the extremely dynamic nature of the COVID-19 pandemic in recent years is the impact of the development of our scientific (empirical) understanding on the content and outcome of semiprocedural review. Under normal circumstances, when reviewing a piece of legislation, a court would set a certain substantive threshold, and the legislative measure would either fail or succeed in stepping over that threshold. The threshold would – as a rule – not be time-sensitive. ⁷³ Within the COVID-19 context, however, our empirical understanding is expected to develop as further information about the variants, their susceptibility to treatments, and their modes of spread becomes available. Hence, a measure that might succeed in semiprocedural review at a certain point might fail a month later simply because the previously accepted conditions of empirical uncertainty were reduced (or had altogether vanished). This was reflected by the Israeli case dealing with recourse to contract tracing by the General Security Services. At first, the Court found that this measure is authorized under the GSS statute, empowering the GSS to act to protect national security since a pandemic affects national security. However, as the picture regarding the scope of the risk to the population became clearer, and as other methods to protect the public were better assessed, and the relative advantage of the GSS tools became more concrete, it was modest at best – the Court read the same words in the statute as insufficient to support an ongoing deployment of the GSS tools and sent the government to seek explicit authorization. ⁷⁴ Similarly, the Czech Constitutional Court stated that:

After several months of experience with the development of the pandemic in the Czech Republic and elsewhere in the world, the level of practical uncertainty was not as high as in the first months of the pandemic, and it is therefore to be expected that the government could and should have had a number of relevant bases for rational and justified decision-making. ⁷⁵

This problem might also have a procedural dimension in some jurisdictions. Is it possible to review the same measure multiple times, or would res iudicata be established after the first review, barring any review in the future? The logic of semiprocedural review would posit that if the external review conditions changed, no formal res iudicata would be established. ⁷⁶

D. Conclusion

In this text, we have presented how constitutional courts have responded to the challenges of the COVID-19 pandemic and argued how they should have reacted. Specifically, we have focused on the courts’ approach to reviewing anti-pandemic measures and specifically on proportionality analysis with respect to these measures.

First, we have presented two “traditional” options: the courts can either apply a standard strict proportionality test or employ a deferential approach. Not surprisingly – given the specific circumstances of the pandemic and the convincing arguments against strict substantive review – most courts have opted for the deferential approach. By contrast, evidence of strict substantive review is virtually non-existent.

But, as we further argue, the perceived dilemma between strict and deferential substantive review is false and can be escaped by opting for a semiprocedural review. In this regard, courts escape the necessity of setting substantive standards by reviewing the rationality of the legislative or executive process that ultimately led to the adoption of the measures in question, including the rationality of evidence and the rationality and transparency of the legislative debates. By doing this, the relationship between legislators and courts can be one of fruitful cooperation, as everyone can do what they are designed to do. The government or parliament can use their personal and financial resources to gather and analyze empirical evidence and conduct a rational and transparent debate. On the other hand, courts can assess the rationality of policymakers’ thought processes and function as the “Socratic” guardians of public reason.