Donald Trump’s decision to challenge the result of 2020 presidential election has raised grave concerns about the future of democracy in the United States. In the years leading up to the election, many comparativists and democratic theorists began to sound the alarm (Ackerman, 2010; Bermeo, 2016; Daly, 2019; Levitsky & Ziblatt, 2018; Scheppele, 2018). These theorists pointed out the increasing prevalence of charismatic leaders in countries like Hungary, India, Russia, Turkey, and Venezuela. They accused these leaders of turning democracy against itself, weaponizing democratic legal systems to accumulate power for themselves and for their friends. Increasingly, these theories are being applied to the United States, to Donald Trump, and to Donald Trump’s potential successors in the Republican Party. Levitsky and Ziblatt (2021) now argue that the ‘bulk’ of the Republican Party is ‘behaving in an antidemocratic manner’. For these theorists, American democracy faces a legitimacy crisis, one that calls its very survival into question. But is this really the situation in the United States today?

Political theorists understand crises, and particularly crises of legitimacy, in different ways. Some theorists—like Reinhart Koselleck (1988)—think of crises as acute phenomena, in which there is a direct threat to the survival of the political system. For Koselleck, a latent crisis is simply a crisis that has not yet escalated to the acute level. During the latent crisis, a moral dualism forms, as the state’s values become estranged from the people’s. The state has the opportunity during a latent crisis to listen to the people and rectify the dualism. But if it ignores the dualism, it risks allowing the latent crisis to escalate into an acute crisis in which there is an ‘actual moment of revolt’ (Koselleck, 1988, p. 155). For Koselleck, the latent crisis is a time when the state has an opportunity to avert the acute crisis. The ‘threat’ comes from the possibility that the latent crisis could become acute, not from the latent crisis itself.

Habermas’ (1975) theory of legitimacy crisis similarly describes a legitimation crisis as a ‘natural fate’ that the state tries to ‘put off’ or ‘avoid’. But recently, other theorists—such as Andrew Gamble (2014), Wolfgang Streeck (2014), and David Runciman (2018)—have begun to consider the possibility of a crisis that may never become acute, in which the political system is tolerated with some reluctance, because people see no viable alternative to it and/or fear the consequences of acute rejection.

Gamble makes a distinction between ‘situational’ and ‘structural’ crises. For Gamble, ‘situational’ crises are about a particular moment that requires an immediate response, with ‘existential’ crises involving a threat specifically to survival (Gamble, 2014, pp. 28–31). In contrast, ‘structural’ crises are about a condition rather than an event, in which there are ‘long-term and persistent deadlocks’ that lead to repeated situational crises (Gamble, 2014, p. 31).

At first blush, Gamble’s existential crises resemble Koselleck’s acute crises, and Gamble’s structural crises resemble Koselleck’s latent crises. For Koselleck, the latent crisis produces the acute, and for Gamble, structural crises can lead to a series of situational crises. But there is a further distinction here. For Koselleck, an acute crisis is necessarily existential, because it involves a moment of revolt. Gamble, however, specifically draws a distinction between situational crises that are existential and those that are not, implying that a structural crisis can generate a series of non-existential situational crises.

What if American democracy is in a structural crisis, but one that generates non-existential situational crises? I call this kind of structural crisis a ‘chronic legitimacy crisis’ or ‘chronic crisis’ for short. Runciman (2015) argues that veteran democracies exhibit high levels of ‘confidence’. Having managed difficult situations before, the citizens of an experienced democracy are confident they can ‘muddle through’ crises. Even when these democracies perform poorly for long periods of time, citizens find it hard to imagine abandoning democracy in favor of some other political system. Democracies that have this kind of resistance to acute, existential crises are ‘embedded’. They are deeply rooted, and difficult for charismatic leaders to dislodge.

It is not obvious which real-world democracies are ‘embedded’ in this sense. Runciman, Gamble, and Streeck often discuss the United States, the United Kingdom, and some of the countries of the European Union. In this article, when I use examples, I draw mainly on the United States and the United Kingdom, because those countries use very old democratic procedures (the U.S. Constitution and the Westminster system) that have been in force in some form for over two centuries. If any democracies are embedded, they are. But I encourage researchers to consider whether there are other embedded democracies, especially in other parts of the world. When I say that the US and UK are embedded, I am not saying that they are ‘more’ democratic or ‘better’ than other democracies. By many different standards, the US and UK are less progressive than many younger democracies, in part because of their age. Embedded democracies began long ago. They are very old. They have been through a lot. And yet, they are still here. Major parts of their populations either approve of their democratic procedures or can imagine no viable alternatives to them.

The life of an embedded democracy is like a television series with multiple seasons during its run. Each season consists of multiple episodes, but the end of one season does not necessarily mean the end of the series—instead, it gives way to an interval without episodes, until the next season is ready to air. A chronic crisis is like a season of television, and the non-existential situational crises it generates are like episodes of the show. What we call stable politics occurs in the intervals between seasons, between chronic crises. Most TV series don’t survive past the first season, but if they make it to season 2, there’s a good chance they will last for quite a few seasons. In the same way, an embedded democracy has survived a chronic crisis before. Its procedures have been tested, and it is more likely to survive further chronic crises. An embedded democracy is like a popular hit series—it will have a number of seasons before it is cancelled, each flush with episodes.

A heavily embedded democracy may become like The Simpsons. Season 33 of The Simpsons is, by all accounts, much worse than Season 1. If season 33 were the first season of the show, it probably wouldn’t have been renewed for a second season. But the distinguished history of The Simpsons allows it to perform increasingly poorly without getting cancelled. Most theories of crisis are interested in the kinds of crisis we see in young, unestablished democracies. This article is interested in the crises we see in subsequent seasons, when the show is getting older. These will only rarely contain the series finale, the episode of crisis in which the show finally meets its demise. Most of the time, there will not only be further episodes, there will be further seasons. What if American democracy is past its prime, but still very far from the end? What if American democracy is in the equivalent of The Simpsons’ season 13 or season 23? It is certainly far too old to be in season 1.

The above allegory conveys a general idea, but it is too aphoristic to do the work of a precise argument. This article will endeavour to offer a more detailed, precise account of this ‘chronic legitimacy crisis’. It will do this in three main ways. First, the chronic crisis must be conceptually distinguished from stable politics. It needs to be shown in what sense this chronic legitimacy crisis is a crisis. Second, because a chronic crisis does not involve existential or acute crisis, and therefore does not involve a moment of revolt against the state, it must be demonstrated in what sense a chronic legitimacy crisis threatens a democracy, in what sense it involves a rejection of legitimacy. Finally, the article will examine more closely how a chronic crisis works, temporally. The article will handle each of these things in sequence. So let us now turn to what precisely distinguishes a period of stability from a period of chronic crisis.

Stable Politics Versus Crisis Politics

One way to visibly distinguish stable politics from crisis politics would be to distinguish how people act during a chronic crisis from how people act during a period of stability. The roots for building such a distinction can be found in Daron Acemoglu and James Robinson’s (2013, p. 84) discussion of the way losers respond to losing. For them, inclusive procedures are often blocked because people will oppose inclusive procedures if those procedures cause them to lose economically and politically. Acemoglu and Robinson (2013, p. 85) use the example of aristocratic opposition to the procedural reforms that granted merchants and businessmen access to political power during the era of industrialization. In this, Acemoglu and Robinson are making an important observation—when people feel they are losers in the state’s order, they may start politicising state procedures, either to preserve, maintain, or restore prior procedures that they associate with winning or to create or develop new procedures that they hope may enable them to win in future.

This can be the basis for a distinction between stable politics and crisis politics. In stable politics, losers may disagree in principle with state decisions, but they continue to accept the procedures through which those decisions were taken. In crisis politics, the procedures are politicised. In an acute crisis, the entire system of the procedures is politicized all at once. But it is possible to politicise some procedures without politicizing all of them.

This provides a clear delineation between stable behaviour and crisis behaviour. During an interval of stability, political procedures are depoliticised, and political debates become more issue and policy oriented. During a crisis period, some or all of these procedures are repoliticised. Subjects move from emphasising what they decide to how they decide. This is a change in what political conflict is about. It is not merely polarisation or an accentuation of extremes. When Bernie Sanders proposes a medicare-for-all single payer healthcare system, he is proposing a policy that might be radical in the American context. But in proposing a policy reform, he is implicitly suggesting that America’s political procedures are sufficiently adequate in their construction to enact the policy reform. By contrast, when Sanders calls for changes to campaign finance law or for new voter access laws, he is suggesting that the way America picks leaders itself must change if it is to have leaders who can reliably make good decisions. He is expressing a frustration with losing, and he is proposing procedural rule changes that will allow his side to win. Here, Sanders makes the democratic procedures themselves the subject of reforms—his reforms become ‘procedural’ rather than ‘policy’.1

Nevertheless, in proposing to modify the procedures through reforms that the procedures themselves must enact, Sanders is expressing a level of baseline confidence in the procedures by which American democracy modifies its procedures. If the political system were completely broken, American democracy’s procedures would block the procedural reforms intended to address the problems with those procedures. In that kind of scenario, fixing the procedures would require more than procedural reforms. It would require a new political system. Despite its nomenclature, Bernie Sanders’ (2017) ‘political revolution’ still orients itself around procedural reforms rather than extra-legal challenges to the political system.

In stable politics, policy reforms will dominate the agenda, while in crisis politics, procedural reforms driven by frustration with losing will dominate. But domination should not be taken to imply exclusivity. People will still propose policy reforms during crisis periods and procedural reforms during stable periods. It is the ratio of the one to the other and the priority given to each that shifts. A crisis has more procedural reform proposals on the agenda, and the procedural reform proposals that are on the agenda will be higher on the agenda. For instance, Brexit is in part a procedural reform, and there have been people who have wanted the UK to leave the European Union for a while. But it is only in recent years that Brexit became a priority issue, ahead of many policy reforms that previously would have taken precedence for parties and governments. Insofar as British politics has become more oriented around the fate of this procedural reform in recent years, it has moved away from stability and toward crisis. This remains true even if there is little meaningful risk of an existential crisis, of an actual moment of revolt that puts an end to the political system.

Because the distinction between crisis and stability is grounded in domination rather than exclusivity, a period of stability will never truly eradicate procedural disputes, and a period of crisis will never truly eradicate policy disputes. Full stability is itself consequently something of a utopian notion. Real politics is never entirely freed from the procedural disputes that take centre stage during periods of crisis. Procedures attempt to achieve the level of legitimacy necessary to reach full stability and may move the system closer to it, but it is never fully grasped. The difference between the crisis and something approaching full stability is mainly one of intensity and priority. This makes it very difficult, in practice, to draw a sharp line between the two. It’s a scalar distinction, and it will often be unclear where precisely on the scale a particular state sits. Political actors can easily misread the situation, placing emphasis on procedural reforms at a time when the public is more interested in policy reforms or vice versa. The difference is clear at the extremes, but harder to perceive at the margins.

This might appear to concede the argument. If full stability is unobtainable, states are always experiencing crisis to some degree. States are always managing change, and they never get into a situation that is truly static, where there are no losers and no procedural reform proposals motivated by a frustration with losing. But insofar as the reader is willing to take scalar concepts seriously, we can speak about ‘full stability’ and ‘acute crisis’ as two ends of a scale, with embedded democracies never occupying either end, but moving noticeably up and down the scale.

The reader might ask why I do not define ‘procedures’ more narrowly, perhaps in terms of Rawls’ (2001) constitutional essentials. There are three issues with that move. First, Rawls’ constitutional essentials often include matters that are substantive, that are not purely procedural, like freedom of religion. Second, in different contexts different parts of the procedural schema are deemed ‘constitutional’, with some constitutions attempting to cover much more procedural ground than others. Third, a constitutional crisis is much more likely to result in an acute crisis, insofar as a constitutional crisis is a direct challenge to democracy as a political system as it has been understood in the context in question. Procedural reformers are not necessarily proposing to throw out constitutional essentials. Sometimes they make suggestions that would require constitutional amendments, such as abolishing the Supreme Court or the Senate, but they advocate achieving those aims through extant constitutional processes. In most cases the reforms they are suggesting are not prohibited by the constitution, focusing instead on things like the independence of central banks or the number of justices on the court.

The reader might reply that this crisis doesn’t appear severe enough to be worthy of the name ‘crisis’. What is so threatening about an emphasis on procedural reforms that are often consistent with constitutional essentials as they stand or with the amendment mechanisms that those essentials have spelled out? It is hard to tell which procedural reforms are compatible with maintaining a democratic political system. Political parties have strong incentives to politicize the distinction. In a chronic crisis, there will be widespread support for the democratic political system, but widespread desire for procedural reforms. If you want to stop someone else from enacting procedural reforms, it will often be effective to accuse them of trying to destroy democracy. Different political parties will advocate different procedural reforms. They portray their own reforms as strategies for saving democracy, and they portray the reforms of their opponents as efforts to destroy it. More and more political energy is caught up in the struggle to save democracy from all the other factions that are also trying to save democracy.

How does this happen? For theorists like Bermeo (2016) and Scheppele (2018), existential threats to democracy often begin in ‘executive aggrandisement’ and ‘autocratic legalism’. Democratically elected leaders manipulate the constitution to increase their power. They reform procedures to improve their own positions while maintaining a democratic veneer. When procedural reforms enhance the power of the executive, they can easily look like an existential attack on democracy. In countries like Hungary, India, Russia, Turkey, and Venezuela, that is often precisely what procedural reforms have been used to do. Bermeo (2016, p. 12) describes Turkish president Recep Tayyip Erdoğan’s use of procedural reforms to gut the independent judiciary:

In 2010, Erdoğan passed two-dozen constitutional changes via national referendum. The president received power to name fourteen of the seventeen Constitutional Court judges, while decisions about which parties are legal and allowed to field candidates for office were shifted from the courts to the legislature. In 2014, the government passed legislation giving the justice minister power to directly appoint members to the High Council of Judges and to control the inspection board that disciplines judges. Within six months, more than three-thousand sitting judges had been removed. The courts suffered another blow from a law that gave the National Intelligence Organization (headed by a presidential appointee) power to collect ‘all information, documents or data from any entity in Turkey’ without having to seek judicial permission or submit to judicial review. All these changes were made by democratically elected officials with a strong popular mandate to rule. Because many of the new measures challenged military and civilian elites with less than perfect democratic credentials of their own, they cut through the old order with what even critics describe as ‘a democratizing edge'.

But procedural reforms can also be used to strengthen the democratic character of the political system. In the UK in 1910, the House of Lords used its suspensory veto to block a land tax, frustrating government policy. The Prime Minister—David Lloyd George—pushed through the Parliament Act 1911, reducing the power of the lords to veto legislation. Most British democratic theorists don’t consider the Parliament Act 1911 to be anti-democratic. But Bermeo (2016, pp. 10–11) defines executive aggrandizement this way:

This more common form of backsliding occurs when elected executives weaken checks on executive power one by one, undertaking a series of institutional changes that hamper the power of opposition forces to challenge executive preferences. The disassembling of institutions that might challenge the executive is done through legal channels, often using newly elected constitutional assemblies or referenda. Existing courts or legislatures may also be used, in cases where supporters of the executive gain majority control of such bodies. Indeed, the defining feature of executive aggrandizement is that institutional change is either put to some sort of vote or legally decreed by a freely elected official—meaning that the change can be framed as having resulted from a democratic mandate.

By reducing the power of the lords, the Liberal government modified British constitutional essentials to increase its own power. It reduced the institutional checks on the power of the Prime Minister, hampering the power of opposition forces to challenge executive preferences. But today, that procedural reform looks like an extension of British democracy rather than a distortion. The lords were a small minority, and they used their veto to protect their own estates from tax.

The recent debate over voter access in the United States shows how easy it is to muddy the waters. In 2020, the Democrats sought to make it easier for voters to vote by mail and by absentee ballot. They worried that the COVID-19 pandemic would depress voter turnout. Many Democrats have long believed that our democratic procedures would be fairer if it were easier for people to vote. It has also sometimes been argued that Democrats do better when turnout is higher, though the evidence on this point is contested (Fowler, 2015; Hansford & Gomez, 2010; Thompson et al., 2020).

When President Trump was defeated in 2020, he and some of his supporters alleged that the procedural reforms that were enacted to make it easier to vote from home were a form of election manipulation. Many Republicans have sought to reverse these electoral changes. Some have sought to use the frustration surrounding the 2020 election to make it even harder to vote than it was before.

The Democrats, in turn, regard the Republican procedural reforms as a voter suppression strategy. They hope to pass a federal law that would stop the Republicans from passing voter suppression laws. The Republicans, in turn, regard that bill as an attempt by the federal government to usurp the rights of the states guaranteed under the 10th amendment.

Each party accuses the other of trying to rig elections and destroy democracy. But both parties say they want fair elections and a strong democracy. The more the Democrats try to make elections fair, the more the Republicans accuse them of trying to enable voter fraud, and the more the Republicans try to make elections fair, the more the Democrats accuse them of suppressing the vote. The more one party tries to save democracy, the easier it is for the other party to portray it as a destroyer.

Chronic crises are tricky because they are periods in which the definition of democracy is renegotiated. The very procedural reforms that can easily be framed as democratic backsliding are framed by others as tools for defending, extending, and purifying democracy. This creates a lot of legitimate, honest disagreement about which procedural reforms are good for democracy and which are bad. But that honest disagreement doesn’t look honest to those participating in it.

The chronic crisis looks as if it might be an existential crisis, and fear of an existential crisis prompts defenders of democracy to take drastic measures to save the political system from phantom foes. The Democrats increasingly fear the effects of fake news. For Levitsky and Ziblatt (2018, p. 56), Donald Trump won the 2016 primaries in large part because cable news and social media allowed him to build a following and to raise money without having to go through traditional party channels. Levitsky and Ziblatt (2021) believe that Trump and the Republicans are hostile to democracy. If they are right, American democracy faces an existential threat, and it is necessary to enact procedural reforms that strengthen gatekeepers in the political parties and in the media. But if they are wrong, the reforms they propose will exacerbate definitional disputes. Everything they do to save democracy from the Republicans looks to the Republicans like an attack on a democracy. Levitsky and Ziblatt’s reforms inspire the Republicans to pursue their own reforms, and those reforms in turn look anti-democratic to Levitsky and Ziblatt. That drives them to try even harder to pass their own reforms. A vicious cycle ensues.

This mutual suspicion makes it harder and harder for Americans to talk about substantive issues. The more intense the fight about democratic procedures becomes, the less room there is to use the procedures to solve problems. The other party looks authoritarian. Its supporters look like useful idiots, or worse. There is little possibility of reaching across the aisle. The deeper the chronic crisis becomes, the more gridlock there is. Procedural disputes crowd out substantive issues. State capacity declines. As the unsolved problems build up, those unsolved problems generate innumerable forms of malaise.

This section has established what distinguishes the chronic crisis from stability. It has also shown that this threat to democracy can be said to exist without our having to describe this threat as existential, and thereby we have preserved the distinction between the chronic crisis and the more traditional acute/latent crisis. It’s also drawn attention to a visibility problem—the fact that the same procedural reforms can simultaneously appear to some subjects to be necessary to save democracy while to other subjects they appear to be a distorting or even an existentially threatening force. But if there is no existential threat to democracy, in what sense is the chronic crisis a legitimacy crisis?

What Makes the Chronic Legitimacy Crisis a Crisis of Legitimacy?

To show that the chronic crisis is a crisis of legitimacy, I need to connect it with existing accounts of legitimacy. It is not possible to connect the chronic crisis with every account of legitimacy. In this article, I work off Bernard Williams’ theory. I use Williams’ theory both because it is relatively prominent, and because Williams frames legitimacy around securing a bare minimum level of stability. Compared with other, fuller theories of legitimacy, it is pretty minimalist. If the chronic crisis is a crisis of legitimacy in Williams’ sense, it will be a crisis of legitimacy in many other senses.

Williams argues that the state is legitimate when it meets the ‘basic legitimation demand’ (BLD). For Williams (2005, p. 4), meeting the BLD requires the state answer what he calls ‘the first political question’ in an ‘acceptable’ way. The first political question is a question about how to secure stability. Specifically, it is about how to secure ‘order, protection, safety, trust, and the conditions of cooperation’ (Williams, 2005, p. 3). It must be answered before any further political questions can be asked. Acceptability, for Williams, is about whether the answer to the first political question is justified to each of the state’s subjects. The state’s subjects do not have to be fully satisfied with the justification the state offers, but they have to tolerate the state, despite feelings of reasonable resentment. This means that for Williams there are at least two ways in which legitimacy can be threatened:

  1. (1)

    The state can fail to answer the first political question, i.e., it can fail to secure order, protection, safety, trust, and the conditions of cooperation.

  2. (2)

    The state can answer the first political question but do so in an unacceptable way, i.e., in a way that provokes too much resentment for citizens to tolerate it.

Williams, like Rawls, is interested in the stability of political systems, but he is much less interested in whether states are stable for the right reasons (Rawls, 2001, pp. 184–189). Acceptability is therefore about whether even a stable modus vivendi is possible, let alone the kind of stability that citizens might positively endorse. I follow Williams rather than Rawls because Gamble (2014), Streeck (2014), and Runciman (2018) focus mainly on whether, and in what sense, democracy can maintain stability, rather than on whether this stability can or should be endorsed. They focus on whether and in what sense democracy can continue to work even as a modus vivendi. In that way, they are focusing on democracy’s ability to maintain the bare essentials of stable order. On Rawls’ (2001, pp. 192–195) account, stability for the right reasons is more than a mere modus vivendi, and a modus vivendi is a prerequisite for more normatively satisfying kinds of stability. Since these crisis theorists discuss whether and in what sense democracy can work even as a modus vivendi, their discussion is a prerequisite for the normative discussion associated with Rawls. For that reason, this article brackets the question of whether democracy is or can be stable ‘for the right reasons’, focusing instead on whether and in what sense it is or can be stable full stop.

To argue that the chronic crisis is a crisis of legitimacy, I draw on the distinction Williams makes between reasonable resentment and rejection. Williams (2005, p. 82) describes resentment as follows:

…the restriction of our activities by the intentional activities of others…can give rise to a quite specific reaction, resentment; and if resentment is not to express itself in more conflict, non-cooperation, and dissolution of social relations, an authoritative determination is needed of whose activities should have priority…

Williams (2005, p. 87) says resentment is ‘specific’, but he also emphasises that it ‘so readily merges into other negative feelings, such as anger and dislike’. He considers associating it with the feeling that we are being coerced without rightful cause, but instead describes it as a kind of alienation, in which we cease to identify with the state’s decision, regardless of whether or not we deem it to be rightful. To be clear, Williams (2005, pp. 88–89) does not himself use the term ‘alienation’ here, but it effectively gets at the kind of experiential non-identification he describes when he says:

Someone who disapproves of a measure in principle but not on procedural grounds is less identified with it than someone who approves of it in both these respects. Someone who finds it both procedurally and in principle objectionable is even less identified with it, and one who thinks that all the procedures are a sham is less identified still. At the end of this line, when the action that constrains someone is experienced as nothing but coercion, sheer force in the interests of others, the lack of identification is total, and this certainly is resentment. But right from the beginning of this progression there is room for the idea that the action, whatever there is to be said for it, is a limitation of someone’s liberty, to the extent that he identifies with the desires and projects which this action will frustrate.

In sum, the subject’s identity is alienated from the state’s action and is instead associated with whatever desires and projects the state’s action is perceived to obstruct. But there are many different levels of severity of this resentment qua alienation. We start with a kind of resentment that is regularly experienced by lots of people all the time—disapproving of a measure in principle but not on procedural grounds—and end with a total lack of identification in which the state’s actions are regarded as sheer force.

For Williams resentment cannot be purely about liberty, because in his view the concept of liberty is too normatively loaded (2005, p. 89). Instead, Williams associates resentment with alienation. This is a subjective feeling that any given loss of liberty may or may not trigger. When subjects are not identifying with the state’s decisions to some degree, they aren’t seeing these decisions as fully acceptable. Insofar as the state’s order is comprised of its decisions, the less acceptable the decisions are, the less acceptable the order becomes. If subjects don’t believe the state secures order in an acceptable way, they don’t think the state is legitimate (Williams, 2005, pp. 3–4). So, for Williams, threats to the state’s legitimacy must begin in the feeling of resentment, a feeling that for Williams can be experienced even when subjects are still regarding decisions as procedurally acceptable but substantively unacceptable.

Williams is not alone in using the word ‘resentment’ and its variants to describe the experience subjects have when they feel alienated from decision-making. Friedrich Nietzsche (2007) uses ‘ressentiment’ to similarly describe the way the slaves experience the power of the masters. For him, this ressentiment drives the slaves to reject the values of the masters and fashion an alternative morality:

The revolt of the slaves in morals begins in the very principle of ressentiment becoming creative and giving birth to values — a ressentiment experienced by creatures who, deprived as they are of the proper outlet of action, are forced to find their compensation in an imaginary revenge.

Note here that the ‘revolt’ of the slaves is not an attempt to change the political system, but a form of moral protest. The slaves reject the legitimacy of the masters, but their rebellion is ideational rather than physical. Alienation also plays a role—the slaves do not identify with the values of the masters and the decisions that result from them, and that drives the experience of resentment. Along similar lines, Bernard Meltzer and Gil Richard Musof (2002, pp. 240–255) emphasise that most conceptions of resentment—including Nietzsche’s, but also many of the theorists who followed him—involve a hostility that is protracted in large part because it is accompanied by powerlessness. Those who experience ressentiment are not capable of retaliating, at least in the near-term, because they are deprived of decision-making power. The association of ressentiment with impotence makes it an especially relevant concept for discussing a chronic crisis, in which the acute response is blocked.

But Meltzer and Musof do not exclusively associate ressentiment with impotence—they also argue that it can itself generate forms of response. Where Nietzsche argues that the resentful engage in an ‘imaginary revenge’, Meltzer and Musof (2002, p. 251) argue that ‘ressentiment may issue in action when the conditions from which it derives become defined as mutable and defeasible’. Thus, ressentiment is associated with the hostility and frustration of the weak, but not exclusively with impotence—insofar as there are responses available, ressentiment can drive the weak to respond.

Streeck (2014, pp. 23–24) makes a similar argument regarding expectation gaps. On his account, the delegitimation created by unmet expectations stems from a sense of impotence on the part of the owners of capital and the earners of wages. They are unable to get the state to issue the right decisions or to deliver on the right conditions, and they feel alienated from these decisions and from these conditions. The feeling of impotence, however, results in genuine political action, albeit action that falls short of acute rejection. So Streeck can also be positioned as part of this tradition of thinking of legitimacy as involving identification with decisions or at least with the results they bring. The experience of unmet expectations goes hand-in-hand with feelings of alienation and powerlessness.

A number of theorists have written about resentment as a driver of contemporary American politics (Cohen, 2019; Cramer, 2016; Engels, 2015). For Cohen (2019), resentment leads to ‘electoral authoritarianism’ and to a kind of acute crisis. But Williams’ idea that resentment can be reasonable suggests that a more sympathetic interpretation is possible. Michelle Schwarze (2020) argues that in liberal democracies it is important to sympathize with the resentment of our fellow citizens. As Schwarze (2020, p. 5) puts it:

When we elide resentment with violence, we obscure its practical and historical role in effecting positive change in both politics and our everyday affairs.

These political responses don’t have to destroy democracy. There are a lot of procedural reforms resentful subjects can pursue without wavering in their commitment to democracy as a political system.

Most of these theorists are using ‘resentment’, ‘ressentiment’, or ‘expectations’ to describe the experience of disempowerment, of alienation from decisions and/or from their results, and of the attendant political action that follows from this feeling. So, subsequently in this article, when the term ‘resentment’ is used, or when gaps in ‘expectations’ are discussed, it is this kind of experience that is being referenced. If we follow these theorists, insofar as states are able to maintain legitimacy, they are also able to produce feelings of identification with decisions and outcomes, and to meet the expectations that subjects may have. Insofar as states struggle to maintain legitimacy, they produce feelings of alienation from decisions and outcomes, and they leave subjects’ expectations unmet.

Insofar as we are willing to accept this kind of account of legitimacy, we can establish that when we are dealing with resentment, we are potentially dealing with the beginning of a legitimacy problem. But this article has not yet done enough to show that the chronic legitimacy crisis features sufficient resentment to constitute a crisis of legitimacy per se. When subjects resent a decision because they disagree with it substantively, but accept it procedurally, they experience resentment, but that resentment will likely still come alongside stable, non-crisis political behaviour. Subjects who accept procedures cannot respond to a substantive defeat by moving to modify the procedures, because they by definition continue to accept those procedures. But once the resentment goes a bit deeper than this and extends to the procedures, crisis behaviour becomes possible. Resenting procedures can give rise to proposing and prioritising procedural reforms, while resenting substantive policies alone does not suffice for this. It is in this way that a particular form of resentment—resentment of procedures—gives rise to the politicisation of procedures and increased emphasis on procedural reforms that this article associates with the chronic crisis.

This clarifies the argument in two reciprocal ways. On the one hand, we can see where legitimacy fits into the chronic crisis. The kind of resentment Williams associates with rejecting the state also produces the behaviour associated with the chronic crisis. At the same time, we are able to use our understanding of crisis to revise Williams’ distinction between ‘reasonable resentment’ and the kind of resentment that results in a legitimacy problem. Williams (2005, p. 126) describes a reasonably resentful man this way:

He is reasonably resentful of what is happening to him, because he is being coerced against his will; and these are the complaints of someone who accepts, not only some political order, but this one.

The reasonably resentful man who accepts this political order is precluded from revolting against the state, but Williams’ explanation permits the reasonably resentful but order-accepting man to exist both under conditions I have described as ‘stable’ and under the conditions I have described as a ‘crisis’. Williams does not distinguish between the reasonably resentful man who accepts, not just the democratic order, but its particular procedures, and the reasonably resentful man who accepts the democratic order, but rejects some of its particular procedures, and targets them with procedural reforms. When he discusses whether the state’s order has been accepted or rejected his treatment is binary—either the order is accepted or it is not. I have complicated this binary by illustrating that there are cases where reasonable resentment coexists with both stability and chronic crisis. It is the fact that the chronic crisis is about resentment—that the losers who respond to losing by politicising procedures and proposing procedural reforms are resentful—that makes it a crisis of legitimacy.

We see just how distinctive this kind of crisis is when we contrast it with other circumstances in which procedures are modified, but in the absence of resentment. Take for instance Runciman’s (2018, pp. 157–158) discussion of the impact of Facebook and Mark Zuckerberg. Zuckerberg is not described as resentful, but as an unintentional reformer:

Facebook—starting with Zuckerberg—has expressed genuine surprise at discovering how its technology can be used to spread fake news. The architects of its system are stumbling across its pitfalls with the rest of us. There is every reason to believe Zuckerberg when he says he wants to make the manipulation stop. He didn’t intend for it to happen. That’s the problem: no one did. It is just a side effect of being in the advertising business.

Insofar as Facebook has changed democratic procedures, it has done this by accident, rather than as a deliberate political response to feelings of resentment. The reforms to democratic procedures that Facebook brings about are incidental—they are not the result of a legitimacy crisis. These accidental reforms might cause or contribute to a legitimacy crisis in reaction to the conditions the reforms create. We might find what Facebook has done to be delegitimating, and we might set about attempting to reform our procedures to correct for Facebook’s influence. But here Facebook’s accidental reforms cause the legitimacy crisis and are not caused by it. This means that the mere fact of procedural change is not sufficient to conclusively establish the existence of a legitimacy crisis. These reforms must be a political priority due to resentment. They must be deliberate reforms rather than accidental reforms.

This article has conceptually separated the chronic legitimacy crisis from stable politics and from acute/latent crisis in several significant ways. It has shown that in a chronic legitimacy crisis in an embedded democracy, procedures are more heavily politicised, and procedural reforms are given greater priority. These procedural reforms are motivated by resentment of extant policies and procedures, and this resentment stems from feelings of powerlessness, alienation from the political, and perceived gaps in expectations. However, the reforms are not motivated by resentment of the entire political system—of democracy writ large. Instead, the reforms are framed by their proponents as devices for unblocking or purifying the democratic procedures. These reforms may still threaten to distort or undermine democracy on particular normative conceptions of democracy, but do not pose a straightforward existential threat to its existence. It is now clear what is crisis-making about the chronic crisis (the proliferation of high-priority procedural reforms that are potentially distorting) as well as the respect in which it is a crisis specifically of legitimacy (these reforms are motivated by resentment of procedures by those who experience alienation and perceive themselves to be political losers).

Having conceptually separated the chronic crisis from stable politics and from acute/latent crises, the article now takes a closer look at the temporal composition of chronic crises.

Temporal Windows: Intervals Between Episodes and Intervals Between Crises

Earlier the article highlighted the possibility of ‘episodes’ of crisis, of non-existential situational moments within periods of chronic crisis. But to this point the article has discussed chronic crises as periods of procedural contestation, without highlighting these specific punctuating moments. There has been a lack of detail on the temporal component of crisis theory, and this has left it unclear how long chronic crises might ordinarily be, and what kind of temporal rhythm they might have. To this end, the article now distinguishes between two different sorts of intervals. One occurs between episodes of crisis, and the other occurs between crises. During the first kind of interval, a crisis continues in the background even in the absence of episodes, while during the second kind of interval the crisis ends and there is a period of stability.

When Gamble (2014, p. 43) discusses his ‘structural crises’, he typically frames structural crises as quite long, protracted things. For Gamble, the last century contains three structural crises—those of the 1930s, the 1970s, and the 2010s. The first of these begins with the stock market crash of 1929, the second begins with the floating of the dollar in 1971, and the third begins with the 2008 financial crash (Gamble, 2014, p. 43). These go on for quite some time. Gamble (2014, pp. 44–45) doesn’t call the crisis of the 1930s finished until 1945. He doesn’t issue a formal end date for the crisis of the 1970s, framing its dissolution in ‘the 1980s’ as a kind of controlled burn rather than a firm end (Gamble, 2014, pp. 46–47). As his book’s title suggests, Gamble is unsure whether the crisis of the 2010s will end, or what its end might look like. For Gamble we can be in a crisis for a very long time, at least a decade or more. As implied by the use of the word ‘chronic’ in the name ‘chronic crisis’, this article agrees with Gamble about this. Chronic legitimacy crises go on for a long time. The period of procedural contestation, wherein procedural reform proposals will proliferate and increase in political priority, can last for years.

Within this period of crisis, there are moments when the government will have to decide whether to accept or reject particular procedural reform proposals. These moments are not existential, but they are decision points that shape how democratic procedures will evolve. So, for instance, in Britain since 2008, there have been a number of procedural proposals that have risen to the top of the political agenda and been either accepted or rejected. These include alternative vote, Scottish independence, reforms to the Labour Party’s MP and leader selection mechanisms, and most obviously Brexit. Each time, British procedures were asked to make determinations about how, going forward, British procedures would make determinations. These decisions brought a level of resolution to pressing procedural questions, but they did not necessarily end the period of procedural contestation. Some of these procedural questions felt more pressing than others, because some procedural reforms were more controversial than others. Procedural reforms are perceived to be controversial when significant numbers of people believe they would distort democracy.

Not every episode of crisis is highly controversial, because not every procedural reform inspires the same level of anxiety. To briefly return to the TV metaphor from the beginning of the article, some episodes of a show generate a lot of buzz and some episodes do not, but an episode is an episode regardless of whether those watching the show get especially excited over it. Brexit might generate a lot more hype than debates over the Labour Party’s internal rules, but both are procedural reforms and both might be said to count as episodes of a chronic crisis, particularly when the decisions regarding these procedural reforms occur in near proximity.

The British state is not issuing decisions about accepting or rejecting procedural reforms every moment of every day, but it is existing in a period in which these questions come up a lot. The gaps between decisions are still periods in which the public is frequently thinking and deliberating about procedural reforms, and so these intervals still take place during a political climate that is meaningfully a crisis climate. It is a climate where procedural reforms are high on the agenda, even when some months or even a year or two passes between episodes in which state procedures feel compelled to make public decisions about particular reforms. If a TV show puts out a new episode every week during its season, these are the intervals between the episodes, when we might reasonably anticipate that more episodes are coming.

A period of stability is very different, insofar as the procedures we have are only rarely questioned and the state procedures only rarely feel compelled to issue decisions about whether to accept or reject particular procedural reforms. During such a period, the procedures begin to feel like they may never really change, and commentators are induced to speak of things like a post-war consensus, a liberal consensus, a consensus on the ‘constitutional essentials’ or ‘basic structure’, a consensus about what ‘now and around here’ legitimacy requires, or even an ‘end of history’ (Fukuyama, 1992; Rawls, 2001, p. 183; Williams, 2005, p. 8). A period of stability is like a hiatus between seasons of a TV show, insofar as we may start to feel uncertain if the show will ever return, if there will ever be another crisis, especially if the stable period goes on for a very long time. But because democracies tend to produce resentment and tend to produce losers, we ought not to be surprised when periods of stability come to an end.

If a break between episodes goes on long enough, it might give the appearance of a return to stability. Gamble (2014, pp. 67–69) speculated that we might be recovering from the crisis of the 2010s in a lasting way beginning as early as 2013. Colin Crouch (2011) wrote of neoliberalism’s ‘strange non-death’ as early as 2011, arguing that while there had been a crisis post-2008, that crisis had not materialised much in the way of procedural reforms, and that the pre-crisis procedures remained largely intact and were likely to remain so. When Crouch (2011, p. 179) asked ‘what remains of neoliberalism after the financial crisis’, the answer he gave was ‘virtually everything’. But subsequent procedural showdowns, like Brexit, may indicate that if the post-2008 crisis is an instance of a chronic legitimacy crisis, it may have some legs left in it. It is easy to think the crisis is over, because there is little that visibly distinguishes a ‘long break’ from ‘stability’, aside from the fact that during a long break there will still be lots of people like Crouch talking about the need for procedural reforms and pushing them up the political agenda. Indeed, in a chronic crisis, worrying that the crisis may have ended without the necessary procedural reforms is itself a means of making the kind of procedural reform demands that are so common in crises. To be sure, it’s possible that Crouch could agitate for procedural reforms during a period of stability. As a left-wing academic, Crouch is the sort of person who would probably be dissatisfied with neoliberal procedures in both periods of crisis and stability. But the fact that Crouch is dissatisfied—and that his dissatisfaction can sell books—may also be a sign that there remains enough resentment to continue fuelling the chronic crisis.

The article has now made a firm distinction between the interval between episodes—which is a mere break, an eye in the storm—and the interval between chronic crises, which is a hiatus, a prolonged period of stability. In so doing it has clarified the temporal boundaries between the chronic crisis and stability, and between the climate or atmosphere of the crisis and the discrete moments when decisions are taken.

Conclusion

By interrogating the relationship between crisis and legitimacy, and by laying out a precise formulation of the chronic legitimacy crisis, this article carves out conceptual space for more nuanced and complex forms of legitimacy crisis. It also helps to clarify the kind of crisis that features in the long-term, slow burn theories of Gamble, Streeck, and Runciman.

The article also helps comparativists think about some of the possible differences between embedded democracies and their younger counterparts. By considering which contemporary democracies are embedded, and which ones are experiencing chronic legitimacy crises, comparativists can avoid falling into a conceptual binary. Too often, crises in democracies are ignored or treated as five-alarm emergencies. But in embedded democracies, there is a middle ground. The sources of resentment must be found and addressed. If the sources of resentment are ignored, the procedural disputes will intensify. If the crisis is mistakenly treated as existential, drastic measures to ‘save’ democracy may backfire. If they increase feelings of resentment, they will only heighten the definitional tensions at the heart of this kind of crisis.

In the United States, it has become increasingly difficult to pass important legislation. Francis Fukuyama (2014) worries that the American political system has become too ‘vetocratic’. Fukuyama (2014, p. 505) advocates procedural reforms to trim veto points:

Many of these problems could be solved if the United States moved to a more unified parliamentary system of government, but so radical a change in the country’s institutional structure is inconceivable. Americans regard their Constitution as a quasi-religious document, so getting them to rethink its most basic tenets would be an uphill struggle. I think that any realistic reform program would try to trim veto points or insert parliamentary-style mechanisms to promote stronger hierarchical authority within the existing system of separated powers.

But even in the UK, where ‘a more unified parliamentary system’ exists, the endless struggle over procedural reforms has made governing the country increasingly difficult. The procedural dispute about whether Britain should continue to be in the EU began taking over almost immediately after the Referendum Act was passed at the end of 2015. After the referendum the ruling Conservative Party changed leaders twice, in 2016 and 2019. The UK endured two elections, in 2017 and 2019. There was a further crisis in 2019 about whether Prime Minister Boris Johnson could ‘prorogue’ (or suspend) Parliament during the Brexit negotiations. Ultimately, the Supreme Court of the United Kingdom ruled that he could not, and Parliament resumed session. The Brexit withdrawal agreement was not signed until 2020, and the EU–UK Trade and Cooperation Agreement did not come into force until May 2021. Only the COVID-19 pandemic succeeded in pushing Brexit off the top of the political agenda.

There are procedural disputes still to come. Partially in response to Brexit, the Scottish National Party has pledged to run a second referendum on Scottish independence. The pro-independence parties very nearly had the numbers to demand a new referendum following the 2021 Scottish Parliament elections, with the SNP coming just one seat shy of an outright majority.

The United States and the United Kingdom are two of the most experienced democracies. If, as I suspect, they are embedded, the procedural disputes that increasingly bog down American and British politics are not an indication of a widespread willingness to abandon the democratic system. Instead, they speak to intense feelings of resentment. To get to the bottom of the legitimacy crises in these countries, the sources of resentment must be identified and addressed.

Cramer (2016) argues that much of the resentment driving the success of the right in the United States comes from ‘rural consciousness’, a sense that rural areas are neglected both culturally and economically. If her account is right, efforts to use procedural reforms to further weaken rural areas—by, for instance, abolishing the electoral college or the Senate—would only deepen resentment and intensify the crisis. Streeck (2014) links resentment to long-term structural changes in the political economy of wealthy countries. If his account is right, procedural reforms that allow these underlying economic problems to continue festering will not resolve resentment.

In a chronic legitimacy crisis, there is lots of time to sort out what is driving resentment and address it in subtle and sophisticated ways. If they can show sympathy for reasonably resentful subjects and take concrete, substantive steps to mitigate resentment, embedded democracies can muddle through chronic crises. They can bring these periods of intense procedural contestation to a close. But if that proves impossible and procedural deadlocks intensify, this chronic crisis may truly become a crisis without end.

Note

  1. 1.

    The ‘democratic procedures’ include those procedures through which leaders are selected and through which decisions are taken, such as the electoral system, the party system, the branches of government, the civil service, the level of state centralization, or the membership of regional and international institutions that circumscribe the state’s policy options, like the European Union.