1.1.1 Constitutional Mechanisms of Personal Control

The threat of removing rulers has long been the most important check on governmental power and the primary means to combat bad governance. Today, the obvious mechanism for removing bad rulers is an election. As Karl Popper argued – and Schumpeter subsequently affirmedFootnote ⁷ – democracy ensures that voters can remove the government without bloodshed, that is throw the rascals out.Footnote ⁸ Inefficient and incapable politicians will pay the cost at the ballot box and will simply be voted out of office, or so goes the theory.

Yet, we know that reality does not always neatly correspond to this idealized template. Elections provide some measure of governmental accountability, but they provide no comprehensive protection against politicians and administrators who fail to do their jobs conscientiously. Elections and election results are dependent on a host of factors, and existing failures are likely to be only one argument among many in the competition for votes. Different political regimes also come with different degrees of accountability. The possibility of holding politicians accountable depends on many things, including on how clearly responsibility can be attributed in the first place.Footnote ⁹ Coordination problems and capacity issues may lead to inefficiency and sometimes failure, which in turn may be hard to attribute to a particular government and harder still to remedy. And in some regimes, democratic processes themselves are largely dysfunctional, and thus elections provide only very limited means to change governmental policy and improve administrative capacity.Footnote ¹⁰

Of course, elections are not the only means of controlling governments and removing incompetent or unwilling officials. In parliamentary systems, parliaments hold the power to force governments, and sometimes individual ministers, to resign. This is not true for presidential systems, but here too, legislatures typically have some means to remove heads of government or state and sometimes other officials by impeaching them. Not all of these mechanisms are easy to operationalize in practice, however. It is rare for parliaments to force the resignation of governments as this requires the majority party to act against its own party members. Where parliaments take that step, it will typically be in situations where a political coalition has come apart – but we would expect a governmental failure of some magnitude before this would be the response to it.

Something similar is true for impeachment provisions. Whether they are suited to dealing with cases of dysfunction or failure will depend both on their concrete framing and interpretation by the relevant actors as well as political will. As Tom Ginsburg, Aziz Huq and David Landau argue, impeachment today often serves as a means not just to remove officials in cases of criminal offences but to address broader political crises where incumbents have lost support, which may involve dysfunction.Footnote ¹¹ But while some constitutions explicitly include broader clauses, allowing for impeachment in light of persistent governmental dysfunctions, not all do. As a result, we sometimes find arguments for an expansive reading of impeachment provisions.Footnote ¹² But impeachment often comes not just with high legal but also practical hurdles, given the influence and interests of political parties. The US example is a case in pointFootnote ¹³ and by no means exceptional, if we consider, for example, the South African ANC’s refusal to impeach then-President Zuma for corruption. Political parties often have little incentive to impeach their own leaders. Given that impeachment can be abused, the high hurdles that come with it are also not unreasonable, especially considering that (in presidential systems) it undermines the democratic choices of voters by removing directly elected officials and so should be a means of last resort. As a matter of addressing failure, impeachment may also not be the ideal tool because failure will often be less a matter of bad intent than a mix of incapacity, neglect and coordination problems. Add to this that both no-confidence votes and impeachment of central figures are nuclear weapons in politics, triggering major changes.Footnote ¹⁴ As such, they will likely be deployed only to deal with problems of a certain magnitude rather than as a tool to address discrete cases of failure and dysfunction.

Within the administration, too, there are means for removing officeholders. Administrative oversight typically allows superiors and sometimes ministers to replace inefficient or corrupt officials with other more suitable candidates, though not all officials can be removed easily or without specific grounds for doing so. Typically, political appointees who ‘serve at the pleasure of the President’ may be dismissed more easily, whereas career civil servants enjoy stronger protections, both against being moved to different offices and even more so against dismissal. This would suggest that systems with a higher number of political appointees – such as the United States – overall do better than those operating largely with a career civil service, like Britain. Often, however, we think the reverse is true, and the more recent political science literature offers support for this intuition. Thus, for example, David Lewis shows how politicization in the United States enables and drives patronage appointments as well as leads to higher turnover in staff, both of which often hurt performance in the longer run.Footnote ¹⁵ And there are a range of factors that influence whether and how politicians might get involved in such oversight in the first place, including the broader political context, the nature of the problem, time pressures, etc.Footnote ¹⁶ Not all decisions on personnel are made by politicians, of course, but overall it is often difficult to dismiss career civil servants without a specific cause, which may be hard to establish. And moving them to other offices, if possible, may create new problems. More importantly still, an exchange of personnel will not always suffice if the problem at hand is structural and thus not tied to any particular officeholder. Broader legislative or administrative remedies will then be required.

1.1.2 Issue-Specific Mechanisms of Control

Another way of grappling with institutional failure is not by exchanging personnel but by intervening directly in the issue in question. There are a range of constitutional and administrative mechanisms providing for such interventions. Sometimes, they merely involve the rights to publicly question officials about their and their departments’ actions, as parliaments usually enjoy. Sometimes they provide the means to actually change existing policies, directly or indirectly.

The standard mechanisms for intervening directly on specific issues are judicial review and administrative oversight. Administrative oversight comes in different forms and may empower both politicians and administrative superiors to intervene at lower levels. Yet, there are many reasons why such mechanisms might not work: Bureaucrats may be overworked or may not care; bureaucracies may be too big to control effectively or be captured by certain lobby groups; close personal ties and personal loyalties may make intervention difficult; and tight budgets and coordination problems with other actors may get in the way of efficient oversight, even in reasonably well-functioning systems. And politicians, too, may not get involved for all these reasons and more. As Rosalind Dixon points out, even in well-functioning democratic systems there are routine malfunctions that may inhibit governmental responsiveness.Footnote ¹⁷ Current legislators have to consider a great deal of issues, often involving difficult and technical questions. Blind spots result from these pressures for time and attention within the ordinary legislative process, which may lead to some problems simply falling through the cracks. And previously declared priorities or the necessity for coalition building may prompt legislative inertia, with the result that some issues and rights violations may be ignored or sidelined, even in cases where they would arguably enjoy majority support.Footnote ¹⁸

Judicial review, too, grants courts powers to intervene, subject to certain conditions. However, courts need support from the other institutions routinely tasked to implement their decisions – and such support may or may not be forthcoming. (As we will see in the second part of this chapter, it is precisely when this assumption of support no longer holds that we see courts intervene more expansively, on the basis of governmental failure.) Judicial review is also limited by the fact that it typically happens ex post and in response to individual cases – and this in turn can make it a less than ideal tool to address broader structural and systemic problems, as we will see in Chapters 4 and 5.

Finally, there is a range of more indirect and subtle means of intervention, both within the legislative and administrative branches. Such mechanisms are often specific to individual legal systems, working to establish accountability within the bureaucracy and thus dealing with failure when it arises.Footnote ¹⁹ Within the legislative branch, too, there are indirect powers of accountability, for example through budgetary control over government spending, which parliaments may use to increase leverage over the executive branch. Here, too, though, the scope and impact of such powers will depend on the specific legal and constitutional context.Footnote ²⁰ And as we have seen, some constitutions explicitly provide federal governments with powers to intervene in individual states under certain circumstances or contain emergency clauses that shift powers to the executive branch in some situations, which I address briefly in Section 1.2.3.

1.1.3 Summary and Consequences

This brief overview shows that there is a range of formal and informal mechanisms to respond to failure by exchanging personnel or by intervening directly through mechanisms of oversight and control. Yet, it is also easy to see how some cases of failure or dysfunction might slip through the cracks. Not all available mechanisms will be adequate to all kinds of failure, and some may not themselves be working, for pragmatic or structural reasons. This is especially true for the kind of failures that are not likely to be remedied by exchanging officeholders but are structural or systemic in nature. Addressing these often requires significant efforts, including coordination between multiple branches. Not surprisingly therefore, administrators and politicians have sometimes sought to take over issues and deal with them outside of those legally prescribed channels. Doing so, of course, then implies a certain degree of legal flexibility.

1.2.1 Arguments from Failure

A prototypical example for a rare legalized argument from failure is rules on federal coercion, such as Art. 356 of the Indian Constitution. India is not an outlier in this regard. Rules for federal coercion exist in a host of states, sometimes framed in more explicit, sometimes in less explicit terms, allowing federal states to intervene in cases where sub-national units such as states fail to fulfil certain duties.Footnote ²¹ My own home country, Germany, provides for this possibility in Art. 37 of the German Basic Law:

This rule has a considerable tradition in German constitutional history, going back to the German Empire.Footnote ²² The takeover of the Prussian government by the federal German government under Chancellor Franz von Papen in 1932, the so-called coup or putsch in Prussia (Preussenschlag), represents an infamous example for the exercise of that power. It led to a prolonged legal battle in court (Staatsgerichtshof), which ultimately upheld the measure, albeit with some modifications.Footnote ²³ The measure was based on the second paragraph of Art. 48 of the Weimar Constitution, which concerned cases of the disturbance of public order and security, and thus somewhat differs from Art. 37 of the Basic Law. But Art. 48 allowed intervention also in cases where states did not fulfil their duties towards the Reich, and, perhaps surprisingly, this provision survived National Socialism. Both provisions essentially constitutionalized arguments from state failure to the benefit of the respective federal governments. However, unlike in Spain, where the federal government made use of the similarly phrased – and German-inspired – Art. 155 of the Spanish constitution to take over the Catalan government in the wake of its secessionist actions in 2017,Footnote ²⁴ in Germany, this provision has never been used in the post-war era, perhaps due to its abuse in the Weimar Republic.Footnote ²⁵ India’s Art. 356, by contrast, has been relied on in many controversial cases, with the Indian Supreme Court declaring some instances unconstitutional, in spite of claiming only (limited) powers of review.Footnote ²⁶

Beyond cases of federal coercion, where we typically have a legal basis for intervention, there are many further examples for arguments from failure in a number of different contexts. They range from arguments for why courts need to undertake structural reform and intervene in expansive ways in failing institutions to examples for triggering intervention by different international bodies and organizations, among many other contexts and areas.

Both the reasons for failure and its magnitude, however, may be very different. Sometimes, the original design may have been faulty from the start. Sometimes, institutional inertia blocks necessary adaptation, or institutions are captured by influential groups. Sometimes there may be coordination problems between different offices that need to work together. Failure can be limited to concrete instances of regulatory oversight, or it can be of longer duration. In the latter case, it often makes more sense to speak of dysfunction, though I use both terms interchangeably here. Sometimes, it arises because legislators or administrators actively want to harm certain groups, or because they do not care, since those groups lack political power; sometimes because they are unsure what to do to address a problem or feel they do not have the means to address it. Arguments from failure or dysfunction appear in all of these different circumstances, as a rhetorical device to justify expansions of authority – ‘power grabs’ – in certain exceptional situations. And while they are often blurred with arguments about implied powers or emergencies, they are not identical with either, nor should they be treated as if they were.

1.2.2 Teleological Interpretation and Implied Powers

Arguments about implied powers are, strictly speaking, one important example of a purposive or teleological approach in interpreting institutional competences. As such, they share the broader problems associated with purposive or teleological interpretation. Those problems are exacerbated when dealing with the interpretation, not of rights, but of competences. Teleological interpretation entails an interpretation of law in light of the key purpose or telos of the specific rule, which is typically deduced not from the historical drafter’s original intent, though such intent may play a role, but from a broader reading of the respective norm in its legal and political context. Thus, while teleological interpretation seeks to realize the primary aim or goal associated with a specific norm, there is always a risk of such an interpretation going considerably beyond the legal text. For example, the European Court of Justice has used the doctrine of the so-called effet utile to justify a ‘functionalist’ readingFootnote ²⁷ of many Treaty provisions in ways that have dramatically strengthened individuals’ economic rights under the Treaty, as well as the scope of Union law more broadly. This has sometimes occurred in a context of perceived political failure where the Court has felt it necessary to step in.

While such teleological arguments are primarily or ostensibly about the realization of certain rights (under a constitution or treaty), they often have significant and sometimes unintended consequences for our understanding of institutional roles and competences, which can be far-reaching. They may also blur into more specific arguments about the roles of particular institutions, for example in the context of structural reform litigation, where courts are called upon to ensure the effective protection of rights.

In other cases, the argument is more explicitly about the scope of the competences or jurisdiction of specific institutions. This complicates matters insofar as such competences are meant from the outset to be limited by the constitutional framework.Footnote ²⁸ Nevertheless, public lawyers have often adopted broad approaches to the interpretation of competences as well, where this has seemed necessary to exercise governmental functions, for example, by developing the doctrine of implied powers. Today perhaps most commonly used in the realm of international law, the doctrine was originally developed by the US Supreme Court in McCulloch v. Maryland, an early decision discussing whether the federal government had the power to establish a (second) federal bank for the United States, in the context of a state’s efforts to undermine this effort by taxing the bank.Footnote ²⁹ The US Supreme Court famously argued that while the power to establish a bank was not a power specifically enumerated in the US Constitution, it was implied by others that were. The general clause entitling the federal government to make ‘all laws which shall be necessary and proper for carrying into execution the foregoing powers’ therefore provided it with such power.Footnote ³⁰ Similarly, albeit in slightly different terms, the German Constitutional Court has developed the concept of federal competence in those cases where the issue to be regulated is per se a federal concern (Kompetenz qua Natur der Sache) – and many other courts have come up with similar ideas.Footnote ³¹ In international and European law, too, ideas of implied or inherent powers play an important role.Footnote ³²

Arguments about implied powers are sometimes blurred with emergency arguments (which I will discuss below). This is regrettable because the two are ultimately based on different rationales. Not infrequently, such a blurring of argumentative frames indicates that we are really dealing with a case where an argument from failure would be most appropriate. This becomes clear once we understand the purpose of implied powers arguments. That purpose is to read competences into an institution’s governing framework, ‘not in order to modify it or add to the members’ burdens, but in order to give effect to what they agreed by becoming parties to the constitutional treaty’.Footnote ³³ Giving effect to what the parties agreed on may imply filling in some gaps, but only where this is possible in light of the overarching framework in place rather than providing a new mechanism to deal with unforeseen events, such as emergencies. In other words, in cases of emergencies and indeed when we encounter arguments from failure, the special powers sought are typically not already provided for by the existing legal materials. This is different from arguments about implied powers, which are geared towards giving effect to the original purpose of the existing text, albeit by interpreting it in a broad manner.

In contrast, the less we can draw on text and standard interpretative methods, the more we move into the realm of failure and emergency arguments, where we will typically encounter efforts to distinguish the situation from other, more ‘normal’ ones and appeals to necessity. Moving in that direction means moving into a grey area of legality. Whereas implied powers and teleological interpretation are today widely relied upon and accepted as ‘legal’ methods/doctrines, the same is not true for arguments from failure and indeed emergency arguments, unless they are explicitly provided for in the legal text. As we have seen in the introduction, failure in particular seems too vague a notion, too politically and ideologically polarizing to count as a legal concept as opposed to a convenient rhetorical trope.

1.2.3 Emergency Arguments

Perhaps we should thus simply understand arguments from failure as a specific case of emergency arguments and defend them in those terms.

Contemporary constitutions today routinely include emergency clauses. These typically transfer powers to the executive and often allow for the abridgement of certain constitutional rights in cases of emergency. The database of the Constitute Project lists emergency provisions in 180 constitutions.Footnote ³⁴ Indeed, the US Constitution is today an outlier in not providing a specific regime of emergency powers, but here too a set of constitutional doctrines and statutory provisions lay down specific rules to apply in different kinds of emergencies. Many modern constitutions include emergency provisions in the text itself. Consider, for example, the Indian Constitution (Art. 352 ff.), the South African Constitution (Art. 37) and the Colombian Constitution (Art. 212 ff.), all of which transfer certain competences to institutions that do not normally hold them. The Colombian Constitution stands out insofar as it endows the president with emergency powers not only in the standard cases of war or internal violence but also in situations that ‘disrupt or threaten to disrupt in serious or imminent manner the economic, social, or ecological order of the country or which constitute a grave public calamity’ (Art. 215).

But even where such explicit rules are lacking or indeed unclear, we may encounter emergency arguments seeking to defend acts that would otherwise be illegal in terms of necessity.

Indeed, several authors defend a specific category of failure arguments, namely arguments for certain kinds of structural judicial intervention, in terms of a response to an emergency. Thus, in his work on the separation of powers, after criticizing courts that initiate judicial proceedings by themselves, as happens in some cases of public interest litigation, Christoph Möllers argues:

Möllers’ framing suggests that we may be moving outside of the realm of constitutional law in such cases and instead be operating within the realm of morality, albeit that it is unclear in what kind of cases this verdict might apply and whether it is confined to cases where judges initiate proceedings on their own.

Möllers is not alone in invoking an emergency argument in this context, though his qualification of the practice as extra-legal is not widely shared. Thus, in his work on the separation of powers and the role of courts, Dimitrios Kyritsis approaches failure as a case of emergency, but one that can be accommodated as legal under certain conditions:

These arguments are not surprising. Arguments from failure share an argumentative structure with emergencies and can accordingly be hard to separate from the latter. As Christian Kreuder-Sonnen argues in his book on international emergency powers, ‘the state of emergency can be understood as a legal institution regulating the suspension of (certain provisions of) the normally applying constitutional order – the state of normalcy’.Footnote ³⁷ Kreuder-Sonnen argues that exceptional cases in this sense presuppose two things: First, there needs to be a ‘constitutionally deviant expansion of political power for executive actors increasing the reach and/or the intrusiveness of authority (i.e. the adoption of emergency powers)’; and second, this expansion of powers will be justified in terms of ‘necessity’.Footnote ³⁸ Moreover, emergency clauses typically respond to anticipated failure or at least inefficiency of the institution ordinarily responsible by delegating power to another institution.

This closely parallels the kind of arguments we encounter in cases of institutional failure, and it is therefore no surprise that the two are often blurred and confused. However, there are several reasons why we should treat arguments from failure as distinct from emergency arguments.

First, unlike emergency arguments, arguments from failure are in practice not always or necessarily geared towards increasing executive power. Indeed, quite a few of the cases discussed in this book involve shifting powers away from the executive branch. To be fair, this difference would not seem decisive if we are trying to justify failure as an emergency insofar as shifts to other branches may be normatively less problematic. In other words, we might think that if we accept unwritten emergency powers for the executive branch, we should have even less of a problem with granting emergency powers to other, less dangerous branches and institutions.

Emergencies also often involve the suspension, or at least significant limitation, of individual rights, whereas arguments from failure are often directed towards better protecting individual rights, as we will see in later chapters. Once again, however, this would presumably be an argument for rather than against defending arguments from failure. More importantly though, while emergencies are notoriously difficult to define in advance, they are typically characterized by two qualities: scale and urgency.Footnote ³⁹ Urgency, however – that is, a rapidly escalating situation where things will get significantly worse unless speedy action is taken – will not always and not necessarily be present in cases where we encounter arguments from failure. As we will see in the course of this book, arguments from failure are often arguments about governmental omissions over a certain time period. Accordingly, the situation is often much less urgent, in the sense of not arising suddenly and requiring an immediate response to ward off further damage, leaving no time to reflect.

Moreover, the existence of unwritten emergency powers as such is already a highly problematic proposition and their legality is heavily contested in public law.Footnote ⁴⁰ While there are some who assert the need for such powers and defend them in terms of unwritten rules or indeed on the basis of a non-legal concept of emergency powers, following Carl Schmitt,Footnote ⁴¹ many scholars are rightly sceptical, given the high risks of abuse.Footnote ⁴² A range of theorists have emphasized how the toggle-switch paradigm of the emergency as an exception as opposed to a state of normalcy neither fits the very different realities of the cases that arise (ranging from natural disasters to terrorist attacks to cases of war and martial law) nor indeed provides a normatively convincing template to deal with them in terms of public law.Footnote ⁴³ What we need, therefore, are differentiated responses to different problems, and this entails recognizing arguments from failure as a category of their own, as opposed to treating them in terms of a dichotomous emergency framework.

Moreover, the general risks associated with emergency powers are heightened if we rely on them to address institutional failure. Crisis and failure are, as we know, hard to measure objectively. Blurring failure and emergency arguments plays into the hands of populists, emphasizing the ‘out of the ordinary “character”’ of events in order to justify radical breaks with existing rules. Famous examples include the fire of the German parliament, allowing a further clamp-down on political opponents by the German National Socialists in the 1930s, as well as the more recent military coup and its suppression by the Turkish government and its role in the installation of a new, more authoritarian constitution.Footnote ⁴⁴ Crises and their performances are important instruments in the emergence and strengthening of populist movements, as Moffitt shows drawing on Laclau’s work.Footnote ⁴⁵ Consider in this context also Bhuwania’s troubling account of Indian Public Interest Litigation, which he argues has enabled a form of judicial populism (see Chapter 4).Footnote ⁴⁶

Such risks of abuse loom especially large because the central justificatory framework when it comes to emergency arguments is necessity: Something truly dreadful will happen if the executive does not seize extraordinary powers to act. This also implies that the greater the danger, the more can be justified – opening up a dangerous slippery slope that may justify the adoption of authoritarian measures whenever a transgression of legal competences seems warranted as a matter of output. If we are seeking to ‘abuse-proof’ legal doctrines, adopting an emergency framework is therefore clearly the wrong way to go.

Finally, the emphasis placed on necessity in emergency arguments only partly captures what is going on in arguments from failure, which are, inter alia, concerned with inter-institutional relationships or at the very least should be understood in these terms (see more below). Thus, while considerations of necessity are hardly irrelevant to arguments from failure, they should not be understood as their animating spirit. The term failure – which is irrelevant to a straightforward emergency argument – emphasizes that.