I. Democracy as contested and pluralist

As noted in the Introduction, a puzzling challenge arising from populism is to explain, as it might appear, how there can be too much democracy. A basic etymological description of democracy is the people ruling themselves, or rule by the people.Footnote ⁸ It is key to pay attention to “rule” and “ruling” as well as “people” in this description, a task to which I will return in the next section after noting the contested nature of the concept of democracy.

The constitutional review debate assumes a polity committed to being democratic. This assumption allows participants to bypass deeper political philosophy questions about the justification of the state and of democracy in order to focus on the institution of constitutional review. Differing emphasis and understanding of the value and shape of democracy typically emerges and does important work in driving argumentation about the worth of constitutional review. It is clear that democracy cannot be kept as an uncontroversial constant in the debate. It is common to recognize democracy as an essentially contested concept; that is, a concept about which there is endless philosophical reasoned dispute about the core or essential features of the concept, not just the margins.Footnote ⁹ That democracy is essentially contested should not warrant quietism or relativism about its use as a label and about its content. Democracy as an evaluative concept can be recognized as essentially contested while insisting on standards to be met for its usage. We are able to spot misuse of the democracy description by pseudo-democratic regimes. There must be minimum conditions for a political system to get into the zone of what can be called democratic. A benign dictatorship, while it might exhibit features celebrated in some conceptions of democracy, is still not a democracy because it is missing elections and voting.Footnote ¹⁰ Democracy is emptied as a concept if it just tracks whatever system produces the best laws, on some view, or the most equality-respecting laws. For democracy, there must be some way in which government action and the polity’s laws are responsive and accountable to popular will on an ongoing basis, as captured in a formal majoritarian process such as voting to elect representatives to a legislative assembly. This is the claim that in a democracy laws are produced democratically.

The idea that the concept of democracy is essentially contested and yet not plastic—in the sense that it could take any shape imposed on it—is consistent with the recognition of democracy as pluralist. Within liberal political philosophy in the broadest sense there are multiple justificatory accounts of democracy (this points to democracy’s justification as pluralist) and there are multiple accounts of its shape. That is, there are many and varying forms democracy may take. Democracy is pluralist in its constitution and also in its outlook in that it recognizes and fosters value-pluralism as the liberal idea that there are multiple and substantially incompatible ways of leading a good life,Footnote ¹¹ where no one’s culture or value-system has a monopoly on the good.Footnote ¹² Even if a democratic majority has laws in place that are partial between value or cultural divisions, such laws ought—by democracy’s principles—remain open to contestation and change via future majoritarian processes. Today’s losers could be tomorrow’s winners. In this way, democracy is offended by populism’s antipluralism: Against the preceding idea, populists tend to close down political competition for the future via a claim to definitive monistic value embedded in a unitary people.Footnote ¹³ Populism’s antipluralism in this sense is a leading reason for its incompatibility with democracy. In this Article, I identify a more basic explanation of populism’s incompatibility with democracy based on its inability or refusal to recognize that democracy is made a working reality only through law, and law necessarily restrains popular will. I will illustrate this point in the next section via reflection on Cas Mudde’s and Cristóbal Rovira Kaltwasser’s account of populism’s relationship to democracy.Footnote ¹⁴

II. Democracy needs law

Mudde and Kaltwasser claim that populism is necessarily at odds with liberal democracy but not with democracy per se, or “democracy (sans adjectives),” as the authors also put it. Democracy per se respects “popular sovereignty and majority rule,” as does liberal democracy, but liberal democracy also “establishes independent institutions specialized in the protection of fundamental rights.”Footnote ¹⁵ Populism stands against the constraint of “the will of the (pure) people” represented by fundamental rights protections;Footnote ¹⁶ it does not want democracy per se diluted by liberalism or other values. The conceptualization offered by Mudde and Kaltwasser is unobjectionable, and they develop insightful analysis of how, in authoritarian regimes, populism can be a positive driver of democratization, as well as a driver of de-democratization in established or new democracies.Footnote ¹⁷ What is underappreciated—and what disrupts the claim that populism is not at odds with democracy per se—is that while democracy per se is a coherent theoretical concept, it is a fiction; it is not a realizable concept. There is no such thing as unconstrained popular will. Law dulls democracy, but democracy is realized only through law. Every law is undemocratic in that it is a past political decision governing the present, ousting contemporary popular will and self-governance by the members of a polity. Yet without law—at least for large groups—democracy could not be a reality because it needs rules to operate and it needs rules by which the people are ruling themselves. Democracy must accept law, and, what is more, laws can be prospectively changed to reflect changes in popular will, and still function as laws while in force. This is achieved through majoritarian processes such as representative parliamentary legislatures. It is democracy in action to change laws or make new laws via legally prescribed processes that credibly claim to be democratic in the sense that they give expression to popular will through majoritarian processes. The values that underpin democracy under-determine democracy’s operation;Footnote ¹⁸ it needs to be realized and concretized in rules.

III. Constitutional arrangements

A pluralist understanding of democracy rejects dogmatism about democracy’s contours and, as such, leaves space for judicial intervention in law-making processes overall, while also countenancing limitation and modesty in that effort. There is no magic formula for democracy; its features are open to debate. Within western democracies it is contested whether a proportional representation electoral system is better than a plurality system; whether gender quotas for parliamentary parties are acceptable; whether there should be an upper house of parliament and its composition. As Richard Bellamy notes, debates about the form of democratic processes are underwritten not just by arguments of what is most procedurally fair, but engage a wide range of deep questions of substance in political morality.Footnote ¹⁹ Appreciation of this position has implications for debates about the judicial role in reviewing and shaping democratic processes. It indicates that there is no end goal arrangement that the courts can push things towards. This militates against judicial activism in the role. Electoral and other arrangements brought to the courts’ attention must be given a chance to be recognized as reasonable democratic features notwithstanding apparent sub-optimality on some democratic view. Equally, the malleable, evolving nature of democracy points to a shared role in the development of democratic arrangements. The occasional prod from the courts to cause a change may be a democratically acceptable contribution and, as such, the idea that law should be produced democratically must be understood as having its own flexibility.

The phrase “judicial intervention” is used in the preceding paragraph as a generic term to refer to courts’ empowerment—as well as choice—to move towards reviewing political process arrangements rather than away from them. There are many and varying rules of jurisdiction, justiciability, standing, and avoidance that greatly affect the scope for judicial intervention. Rarely if ever is it simply a judicial discretion to intervene or not. Nevertheless, rules around the courts reviewing laws are more flexible than may seem. Doctrines such as judicial review of legislation and unconstitutional constitutional amendment have been judicially fashionedFootnote ²⁰ as well as explicitly set up in a written constitution. In addition, the judicial intervention envisaged may have ordinary laws governing political processes as its subject matter, but it also embraces scope for more or less creative interaction of courts with the constitutional laws that set parameters for those ordinary laws governing political processes.

A key marker of populism, as Müller emphasizes, is its antipluralism, specifically demonstrated in populists’ claims to uniquely represent and know what is best for a population.Footnote ²¹ This logically leads where populists have sought to take it, that is, to the alteration of rules of governance to close down future political contestation and competition.Footnote ²² This can range from seemingly subtle changes to electoral arrangements to sweeping constitutional amendment or replacement.Footnote ²³ In the hands of populists, democracy is especially vulnerable to self-degradation. Generally speaking, democracy can be degraded in different ways. Thugs might intimidate voters around polling booths, for example. This would be an exogenous, unlawful prevention of democracy, and this is not the way populism typically operates. Electoral laws might be changed to favor one party over others, and the change may well be ostensibly consistent with populist rhetoric in a way that masks its artfulness better than, say, 19^th Century gerrymandering, which demonstrates on its face electoral districts constructed to favor certain voting factions.Footnote ²⁴ This is the special threat of populism: The prima facie lawful change of democratic processes that in fact degrades democracy despite having a stateable democratic rationale. Populists exalt referendum results (that go their way) and say it would disrespect the people’s vote and thus be undemocratic to revisit or re-run the referendum.Footnote ²⁵ For many, this may be a convincing argument from democracy, but it is in truth highly undemocratic because it purports to cut down future political debate and change.

By democratic logic, the rules by which democracy operates—the democratic processes—should themselves be made democratically. Democracy makes itself. Of course, this recursivity is practically escapable. The birth of a new democratic constitutional order may or may not include a plebiscite or some other attempt to provide popular legitimation, and it is easy to imagine an autocratic or imposed ruling that, from this day on, the state is to be democratic and here are its set-in-stone ground rules for democratic governance.Footnote ²⁶ This may be a successful exercise of authority, but the thrust of democratic thinking will always question why the people are not able to decide or adjust for themselves the structure of their democracy. On this view, entrenched constitutional rules that can never be changed, or that can be changed only by special processes different from what are taken to be the appropriate democratic processes for legal change generally, are a compromise on democracy.Footnote ²⁷ Constitutionalism—here understood as the practice of entrenching laws against ordinary legal change—limits democracy, but it is also needed to some extent for democracy to be realized in the first place and to be robust. Constitutionalism is often—but not necessarily—oriented around a written constitution. An unwritten—but understood—constitutional order underpins all legal systems and stands to preserve their character, whether that is democratic or not. The democratic-pluralist endeavor in the modern era is to keep constitutionalism between untrammeled popular will (a mythical position seemingly espoused by populists) and comprehensive constitutionalism (where the ground rules for political competition, and many other areas, are off limits to majoritarian processes of change). The space between these poles is wide and referred to with terms such as “constitutional democracy” and “democratic constitutionalism.”Footnote ²⁸

IV. Democracy’s vulnerability

The above account says that the shape of democracy should be recognized as mutable and its maintenance and change should be democratically engineered. There is an ever-present danger of degradation through democratic voting. Democracy’s openness to change thus leaves it vulnerable to populism. Nonetheless, democracy’s mutability guards against democratic decay. As illustrated by the common experience of disproportionate electoral districts, changes in population distribution in static voting districts may leave people with unequal voting power.Footnote ²⁹ While this situation can simply materialize through unpredictable population changes—quite apart from any attempt at gerrymandering—it nevertheless can be taken advantage of to the detriment of democratic competition. Accordingly, voting districts need be revisable in order to preserve reasonable parity of voting power, known as the one person, one vote principle. This helps to show how democracy’s operational rules must not be set in stone. The immutability of the grounding political rules brings its own particular risk: That the shrewd manage to take advantage of them over time.Footnote ³⁰ The democratic adjustability of democratic processes is not only to be valued in the people getting to say how they would wish their political will to be counted, but also disincentivizes and counters partisan lock-ups, helping preserve political competition. Sometimes change is warranted to democratic processes not because a superior arrangement is presented, but because a reboot is needed to return the functioning of the existing arrangement. Democracy needs to be able to protect itself, and to present itself as ready and able to do so.

Regarding the example above, it may be pointed out that for voting districts the details of their precise shape are part of the ordinary law that makes democracy function, while the one person, one vote principle is the constitutional law. The former requires ongoing adjustment, while the latter is unchanging, so it seems. But in reality, the constitutional law also requires maintenance and adjustment over time. The one person, one vote history reveals later case law struggling to decide whether, for examples, variances from exact parity of voting power, and strangely shaped geographical districts, can be justified in the name of affirmative action.Footnote ³¹ Experience shows that first order questions of democracy inevitably arise.Footnote ³²

Democracy faces a standing threat from within through degradation or decay.Footnote ³³ Even if a polity’s ground rules of democracy are off-limits to ordinary democratic change, it still faces the possibility of externally caused degradation or destruction of the democracy that it has.Footnote ³⁴ Even clear and comprehensive ground rules for a democratic system are going to come into question in their interpretation and application at some point. Courts will be called to adjudicate and, accepting only weak legal indeterminacy theses, judges will have some degree of legally unconstrained choice as to which way some aspect of democracy should be structured. As noted above, these “first order” questions of political governance arise commonly.Footnote ³⁵ The above reflections are preliminary in opening up space within democratic theory for an express judicial role to review democratic processes. Democracy needs protection. Since democracy is realized through law, judges are going to have to get involved in any event. Accordingly, judges should be prepared, and it should be known that judges are so prepared, to protect democracy.

I. Waldron’s challenge

Jeremy Waldron’s The Core of the Case Against Judicial Review Footnote ³⁶ is a stalking horse of the contemporary constitutional review debate, claiming that constitutional review of legislation, where courts can invalidate laws, is not a political institution worth having in a reasonably well working democracy. This well working democracy’s legislature and courts are in good shape, it is assumed for the argument, and there is genuine respect for—and disagreement about—rights. Waldron builds a detailed case resting on two main arguments: (1) That rights-respecting legislatures untrammeled by judicial review produce substantively better laws in terms of rights protection, other things equal, than rights-respecting legislatures subject to judicial review, and (2) that the untrammeled legislative process better instantiates equally dispersed, maximized political liberty than the judicially trammeled legislative process.

Two criticisms of Waldron’s case are brought into focus by reflection on the distinctive worry that populism, as discussed in Section B above, risks accelerated self-degradation of democracy under the guise of popular will. The first criticism is how Waldron handles John Hart Ely’s process theory of judicial review. The second is the limited recognition of legislatures and courts as collaborators. For the first criticism, Waldron acknowledges that his argument does not apply where the legislature fails to be representative. As such, his assumption of a legislature being in good shape does not hold, and he admits that John Hart Ely’s representation-reinforcing defense of judicial review may successfully apply.Footnote ³⁷ This concession to Ely on Waldron’s part is more significant than Waldron acknowledges. Having courts empowered and available to address problems in the operation of democracy entails having them also available when democracy is operating satisfactorily. As argued earlier in this Article, democracy always carries a danger of degradation. A high level of stability at any one time will reduce the risk of degradation in the short term but not eliminate it and, in the long term, the risk might increase and become the new normality. We can grant Waldron’s assumptions, including a legislature in good shape at a particular time, but we should not grant him a legislature guaranteed to stay in good shape—this assumption is question-begging. Waldron’s case cannot avoid dealing with Ely on the merits by purporting to assess circumstances where Ely’s theory does not apply, because, as I will later argue, Ely’s theory can be decontextualized and applied generally.

The second critical evaluation arises from Dimitrios Kyritsis’s emphasis on courts as collaborators, not antagonists with the legislature, to which I now turn.

II. Kyritsis’s separation of powers defense of constitutional review

Kyritsis breaks out of the courts-versus-legislatures orientation of the constitutional review debate to appreciate the advantages of the court and legislature operating as partners.Footnote ³⁸ He relies on separation of powers precepts. The separation of powers is motivated by a division of labor, with the institutions of government performing the tasks of governing to which they are suited. The separation of powers is also motivated by a checks and balances principle where one institution can help reduce the risks of bad governance that other institutions carry. On this view, “legislatures are meant to combine the demands for popular support and moral innovation, whereas courts, by virtue of their independence, are good candidates for assuming the task of legislative supervision.”Footnote ³⁹

Kyritsis characterizes the position of an elected member of a legislature as that of both accountability to, and independence from, his or her electorate. The deputy has a duty to channel and act in accordance with constituents’ wishes, but there is also scope for independent judgment within a fixed term of office; matters will come up which require independent thinking, and even if the electorate ultimately declines to re-elect the deputy, the decisions that were made during their term still count. When elected representatives innovate, “electoral accountability gives them a strong institutional incentive to do their best to explain the desirability of change to citizens.”Footnote ⁴⁰ These factors, for Kyritsis, mean that the legislature can justifiably have the initiative in governing citizens because it combines the demands for moral innovation and popular support. But this does not mean it ought to always have the last word, come what may. Courts’ independence makes them suitable for performing a supportive and supervisory role in respect of the legislature’s output.Footnote ⁴¹ This role respects that the legislature has the political initiative; that is, the legislature is active in governing and the court by comparison is receptive. Compatible with the legislature having this initiative, the courts can work out levels of deference and intervention.Footnote ⁴² Also compatible with the legislature’s initiative is the principle of subsidiarity that courts acknowledge.Footnote ⁴³ Thus, Kyritsis explains constitutional review’s compatibility with the legislature maintaining the political initiative. For quality supervision, the supervising body must have distance from the decision body it is reviewing. Reasons why the courts have this distance include the judges’ secure tenure and pay not being at the discretion of the bodies they supervise.Footnote ⁴⁴ Kyritsis points to the reason-giving duty of courts as a control mechanism.Footnote ⁴⁵ He acknowledges that this is self-control rather than another external control. Because no one is infallible, there is conceivably always a place for another check, but it is not practical to always have an external check in every category of performance: The checks must stop somewhere when a decision is needed and things must move on.Footnote ⁴⁶ Later in the Article, I return to this problem of how to call an end to the checks.

Kyritsis does not claim that constitutional review ought to exist in every constitutional structure. Rather, depending on the circumstances—and importantly that the courts truly have independence—the overall task of governing may be better performed as a joint enterprise between the legislature and the courts, where the courts have the subsidiary role in checking the legislature. Kyritsis’s description of how the courts’ role in reviewing legislative output is subsidiary and receptive was not fully acknowledged or anticipated in Waldron’s case against judicial review. While Waldron says his case does not rule out a weak form of judicial review—for example, where the courts can make declarations of rights violations but not invalidate legislation—he does not sufficiently address the claim that a strong form of judicial review, when circumscribed, may be worth having. As Aileen Kavanagh noted, the judicial power applies over a limited range of decisions despite appearing a greater power (in that a court can strike down a legislative measure).Footnote ⁴⁷ Constitutional review is not all-or-nothing; “the exact kind and degree of participation that is desirable is subject to debate.”Footnote ⁴⁸ Different kinds and degrees of judicial review might be recommended in different states, but what is not typically recommended is an all-encompassing review power for courts where any legislative or executive decision is liable to be second guessed, and where the standard for intervention is simply that the impugned decision is suboptimal by reference to an unrestricted range of evaluative considerations. For Kyritsis, it all depends on the circumstances. Kyritsis’s work does not seek to especially promote judicial political power, and yet it does not supply the principles to rule out high intensity judicial review across the board. Hence, I will suggest modification after briefly recapping why Kyritsis’s theory is promising and persuasive. Furthermore, I will qualify how Kyritsis’s idea of judicial subsidiarity ought to be understood.

III. An elysian gloss on kyritsis’s separation of powers rationalization of constitutional review

The two points of critique of Waldron’s case outlined above are: (1) that Waldron cannot reasonably assume for his argument that a legislature will stay in good shape; and (2) that there needs to be specific attention to the scope for a circumscribed, subsidiary and collaborative judicial role in checking the legislature. Combining these points, I suggest a broadly Elysian modification to Kyritsis’s rationalization of constitutional review as the answer to Waldron’s challenge. This modification purports to reinforce Kyritsis’s claim that constitutional review is something we can live with in democratic terms with differentiation as to the subject matter of review: That is, constitutional review is more clearly acceptable in democratic terms when it is involved in or connected with protecting political processes than with rights or interests generally. I also argue that the intensity of review and degree of intervention should be scaled back from what Kyritsis implies is tolerable. The next Section elaborates.

I. Constitutional review of political processes

I have endorsed Kyritsis’s evaluation of judicial checking of the legislature as largely persuasive, but, inspired by John Hart Ely, I depart from Kyritsis in that I draw a sharp distinction between courts reviewing democratic processes and courts reviewing laws generally. I also seek to be more prescriptive about the restrained role judges should take. Kyritsis’s evaluation is relevant and persuasive because it is a neat distillation of separation of powers theory, which is designed to address the problem of concentrated political power. This is precisely the problem faced permanently by democracy that I have referred to as the danger of degradation. As Müller quotes Berthold Brecht, “[a]ll power comes from the people, but where does it go?” Democracy has a power problem and separation of powers theory offers a power solution.

John Hart ElyFootnote ⁴⁹ defended the constitutional review practiced by the Supreme Court of the United States as compatible with, and indeed supportive of, United States democracy. Ely’s elegant theory derived from preceding academic workFootnote ⁵⁰ and celebrated 20^th Century caselaw.Footnote ⁵¹ Constitutional review in those cases was about “clearing the channels of political change” and “facilitating the representation of minorities.”Footnote ⁵² For Ely, this approach to judicial review does not have courts impose substantive values; it only insists on the fairness of the political processes, and this answered, according to Ely, the counter-majoritarian difficulty that saw the Supreme Court as the deviant institution of American government.Footnote ⁵³ Constitutional review, according to process theory, despite involving unelected judges displacing legislative will, was overall helpful to American democracy. When constitutional review was not geared towards political processes it remained highly undemocratic.

Kyrtsis’s argument can benefit from the Elysian distinction between substance and procedure. Notice how important institutional competence is for Kyritsis: The standard picture of the legislature making the law and the courts implementing, and occasionally checking it, rests on the idea that the legislature—because of its representativeness with scope for independent thinking and moral innovation—is best placed to combine popular support and moral innovation. But if competence as to certain subject matters varies, then the persuasiveness of the division of labor varies. Judicial supervision of certain areas may be more competently undertaken than in other areas. This point may be underappreciated when independence is the key judicial competence because, where judicial independence genuinely exists, it is invariant across the board and subject-neutral.Footnote ⁵⁴ Independence as an attribute does not engage on its own. An independent judgment on a topic is of little use if the judgment-maker lacks understanding and appreciation of that topic. Consider the adjudicator in an architecture prize competition: He or she must be independent—unbiased between competitors—and must also understand principles of architectural quality. Furthermore, independence does not imply political neutrality all the way down. Suppose legislation attempts a balance between social security and property rights and has run into questionable constitutional territory such that either way a court review will involve innovative constitutional interpretation. It is questionable whether judges are really neutral on this kind of property rights balance despite their best efforts. The judiciary tends not to be a diverse body;Footnote ⁵⁵ if not diverse, there will be a concentration of ideological presuppositions that pull in certain ways, not likely to cancel each other out as they might in an assembly of representatives, assumed to be reasonably representative and accountable, or at least more so than the judiciary. Again, this is why the legislature has the initiative with law-making. Even where the independence condition has been met, the court as supervisor may be more or less able to complement and collaborate in governing with the legislature depending on the subject area.

Ely claimed that lawyers and judges are especially skillful in assessing the fairness of processes and procedures. This is in contrast to lawyers’ lack of special expertise regarding substantive political questions. There is a core truth to the claim that judges have expertise as to process; yet the process-substance distinction has been the most fiercely criticized aspect of Ely’s theory. Laurence Tribe,Footnote ⁵⁶ among others, said that process theory claimed to avoid the court imposing substantive values when performing judicial review, but to perform process based judicial review, an understanding of substantive values and rights is needed. The criticism points out that process based judicial review is not neutral as to substantive values. Even seemingly purely procedural positions instantiate and reflect positions on which substantive rights and values are to be entrenched against legislative majoritarian encroachment.Footnote ⁵⁷ Procedural protections for criminal defendants—for instance protections such as bail and jury trial—may negatively affect the accuracy and efficiency of a criminal trial, yet they are required and this must be based on concerns for the freedom and dignity of persons accused of crimes.Footnote ⁵⁸ So far as process review sees judicial review as being about protecting process, it seems to value the process in itself because it disclaims that the process is merely a means to some substantive end. This creates an irony, says Tribe, because if the process is intrinsically valuable, then the process becomes substantive, an end in itself.Footnote ⁵⁹ Tribe at this point is arguing beyond the particular features of US constitutional law. He is making a general point about the impossibility of pure proceduralism. Michael Walzer similarly claimed that “[n]o procedural arrangement can be defended except by some substantive argument.”Footnote ⁶⁰

The short answer in defense of process theory is to concede the accuracy of the criticism, but to maintain the theory. Process theory can still claim virtue in being minimally substantive, if not substance-neutral. To be sure, it inevitably reflects substantive values, but such values are reasonably limited to widely shared core values. Tribe, in later writing, acknowledged that his critique of Ely was overstated insofar as it criticized Ely for purporting to have a value-neutral theory while in fact not having a value-neutral theory.Footnote ⁶¹ Tribe acknowledged that Ely was likely not actually claiming his theory was value-neutral in this way, but just that it minimized judicial incorporation of extraconstitutional values.Footnote ⁶²

The substance-process distinction criticism hits more pressingly at the point of its implementation. Even if the substance-process distinction is defensible as a proxy for differentiating a minimal or core set of ultimately substantive values that are aptly judicially policed, where is the line to be drawn? Questions of political participation are not tightly circumscribed. Free speech will strike many as an intrinsically valuable liberty right, but it is also well known as a feature of healthy democracy for uninhibited political debate, including robust criticism of government. Facilitating the representation of minorities—as one strand of process theory has it—conceivably extends to socio-economic conditions: How can educationally deprived people effectively participate in politics? The long-standing criticism of process theory is that it morphs back to the comprehensive substantive constitutional review power it was intended to avoid. This challenge for process theory is not going to go away, but it is important to not expect too much from a theory. Process theory is a guide that gives indication of the appropriate orientation and not a detailed instruction manual that tells a judge precisely where to stop. Its application will depend on the circumstances, such as the historical and current status of certain groups in society. An affirmative action measure in one polity may be a political participation necessity, while in another polity the same measure is a liberal luxury. Luke McLoughlin recently defended Ely’s theory:

This follows the observation that “representation reinforcing” exists on a continuum, and this was observable in Democracy and Distrust, if not explicitly acknowledged. McLoughlin identifies some Elysian principles that help judges to not slide to unbounded substantive review. These Elysian principles operate in modern US election law, principally in the “political markets” or structural approach to election law judicial review,Footnote ⁶⁴ with emphasis on the importance of “empirics” in electoral case law.Footnote ⁶⁵ This means that the Court, to perform constitutional review, must be supplied with empirical data that clearly proves a structural harm in the democratic process.Footnote ⁶⁶ Requiring strong empirical data is a counter to the malleability in the formation and application of judge-made standards and thus the imposition of the judge’s will.Footnote ⁶⁷ This unsurprising observation is a reminder of the importance of facts in judicial review. The court must have a clear picture of how things are really playing out in practice. The seemingly elusive search for manageable standards in judicial review of political processes will be not need to be undertaken if there is a clear view of the impugned wrong or harm in the case at hand. Process theory has still not provided the instructions for where to draw the theory from to combine with the empirical data in order to give a judgment, but that is part of the point McLoughlin is making. Ely did not give such instructions because it is for the courts to work out as a contribution to democracy, which is always a work-in-progress.

II. How to characterize the judicial role

Ely provided accessible images of justified constitutional review, seeing the courts’ job as to “police the rules of the democratic game.”Footnote ⁶⁸ A problem with the policing image is that there are different levels of policing. Some will think of policing as a hemmed in, reactive role in apprehending criminal offenders. Others might see the police officer as integrated in the community, building relationships, and attempting to help people’s lives in ways quite removed from the prevention or prosecution of crime. The position of the police might be a residual role, or it might be proactive, or anywhere in between. The policing metaphor repeats, rather than clarifies, the ambiguity between activist and restrained judicial roles. It is the restrained role I wish to advocate for the courts, and therefore I hesitate to endorse the policing image as a simple description.

Consider the image of the quality control supervisor described by Laurence Sager, quoted approvingly by Kyritsis, as indicating the judge’s role in performing judicial review.

This image captures the way in which the judge’s role should be reactive or “subsidiary,” as Kyritsis explains.Footnote ⁷⁰ The supervisor is not getting involved in the construction process as a matter of course; rather, she is intervening only where a defect arises. An inapt aspect of this image, however, is that any deviation from the set specifications warrants intervention or possible rejection by the quality control supervisor. To provide this image as the model for judicial review of democratic processes misses the debatable nature of the details of such processes. The quality control supervisor is not at liberty to overlook a minor defect in a product, such as a car, given the concrete specifications that a car must conform to. Indeed, like the task of proof-reading a text, picking up a tiny imperfection is to excel in the job. Additionally, the worker-supervisor image carries the implication that the supervisor is the greater expert as to the thing being made. This translates to the constitutional review debate as the implausible idea that judges can perceive more clearly the demands of political justice and then tell legislators where they are going wrong.

Returning again to Ely:

This competition image animates the recent democratic structures theory in the USFootnote ⁷² that is said to be the progeny of Ely’s process theory.Footnote ⁷³ Here we are closer to the recommended description. This passage invites the old objection of hidden substantive theory masquerading as substance-neutral theory. As noted, while expressing a truth, this does not destroy Ely’s process theory. That difficulty aside, this passage reasonably indicates an appropriate readiness to intervene: Not where things are imperfect but where things are at, or are approaching, democratic malfunction.

III. Who guards the guardians?

Constitutional review of democratic processes relies on the value of judicial independence so that the courts are well positioned to make a call on where political leaders are failing to either repair or maintain democracy, or where democratic processes are otherwise degrading or threatened. While tenure and pay for judges are often said to preserve judicial independence, the more directly effective feature of court practice that maintains independence is the constant duty of the judge to give reasons for decisions. This restrains arbitrary decisions and those motivated by improper reasons. Decisions must be made to appear as an application of the law, properly identified, and using legal reasoning where appropriate, to the true facts of a case. Of course, the artful judge can make a decision appear a legally compelled one when in reality it was not, and hidden motivations might apply. But this can only go so far; legal reasoning is not nearly so radically indeterminate as has been claimed. Close analysis by outsiders, as will be suggested, can cut through spurious window dressing in judgments.

It may be that the felt duty to give reasons among judges has a precarious existence. What stops judges handling cases in a cursory fashion and giving incomplete explanations for their decisions? An answer which suggests a review board of judicial performance or political supervision of judges invites responses pointing out the regress of setting up guards to guard the guardians and also the danger of compromising judicial independence. In addition to the sense of self-worth that presumably applies among members of the judiciary, it can be said that lawyers, judges themselves, law professors, and other observers and commentators on the workings of the legal system are all looking at judges’ judgments. They are reading them, and to varying degrees, scrutinizing, criticizing, and commending them.Footnote ⁷⁴

A reasoned judgment conforms to the norms of legal argumentation. Such argument is logically valid and sound in that it has no fallacies and can genuinely claim to be persuasive. A mere cursory listing of reasons designed to connect with applicable legal norms likely will not be persuasive in this sense; there would typically be explanation required, though it may be tersely delivered. This argumentation also operates within the established methods and practices of the court. For instance, argumentation uses methods of interpretation that are recognized within the judicial community as legitimate.Footnote ⁷⁵ Methods of interpretation are tools available to judges in the reasoning process. There are, of course, multiple methods available and they may pull in different directions. Judicial decisions in constitutional law, as elsewhere, will inevitably be creative at times in that the result of the case was not something determined by existing law. There is tension between saying that a judge should be grounding his or her decision entirely in law, on the one hand, and, on the other hand, desiring transparent judgments that can be seen for the innovation they often are, and the political choices they engendered. There is no easy resolution to this tension. Judges expressly admitting where they are being creative may have an unsettling effect on the law; it abandons the “no gaps” postulate—law presents itself as having an answer to every question—that in turn plays a crucial part in law’s settlement function.Footnote ⁷⁶ When required, judges must attempt to reason their way to new legal norms using relevant extant law. Performance can be assessed when a reasoned judgment is supplied. The reasoning must be transparent and capable of being evaluated; the judgment should not be pro forma or cursory. The evaluation of a judgment is not always clear-cut and there will be disagreement among commentators, but courts’ errors in legal reasoning can be highlighted by commentators or corrected by other courts.

IV. Constitutional prophylactic rules

United States constitutional theorists have identified prophylactic constitutional rules.Footnote ⁷⁷ As described by Mitchell Berman, these are judicially fashioned constitutional law imperatives that have the feature of sweeping more broadly in limiting state action, in their constitutional protection, than seemingly needed.Footnote ⁷⁸ These rules are not coterminous with the actual constitutional requirement. They protect in instances where protection is not really needed as the problem is not present, but they are not redundant or gratuitous for they are preemptive. They set up a barrier to stop a problem developing in the first place. Constitutional prophylactic rules are also said to take into account institutional realities and practice and procedure.Footnote ⁷⁹ These rules are not peculiar to constitutional requirements on political processes such as legislative elections.Footnote ⁸⁰ Democratic processes nevertheless have a special need for them, recalling the ever-present threat of degradation.

Constitutional prophylactic rules and boundary setting are an apparently uncomfortable extension of Elysian terms, which speak of courts fixing things when they have gone wrong. Yet, it may be essential to intervene when a democratic process is on the way to breaking rather than leave it to break first, always bearing in mind whether or not it is likely to be repaired via democratic means. For example, a court might fashion a constitutional prohibition of public expenditure on the advertising of a particular referendum result not because the case made before it is especially problematic, but because if there is no limit on such a practice it could be ramped up or used by political incumbents to secure their re-election.Footnote ⁸¹ It may be a fine balance between intervening to fix a growing problem or standing back to see if the political process can remedy itself.Footnote ⁸² Recall the notion of reinforcing representation in Ely’s terms, which both embraces preemptive action while not encouraging a pursuit of perfection through innovation. Constitutional prophylactic rules are about looking ahead to future dangers and trying to preserve political competition indefinitely. For this reason, they are especially pertinent tools for resisting populist driven democratic degradation.

V. Transnational democratic anchoring

Another method for judicially resisting democracy’s degradation in a suitably restrained manner is the suggestion that courts look to core democratic values and arrangements at a trans-jurisdictional level, referred to by Rosalind Dixon and David Landau as “anchoring.”Footnote ⁸³ A minimal shared conception of the core of democracy is observable around the world. Whether or not textually entrenched in the domestic constitution in question, a transnational consensus of sorts about what must be so in a democracy is a powerful reference point for judges to resist efforts to degrade democracy. It avoids the philosopher-king problem of judges deciding what democracy is—which was a strong concern of Ely—by linking what they insist on to observable practices elsewhere; though of course there is a substantial discretion in the tasks of survey, selection, and comparison.

VI. Beyond domestic constitutional review

Ely’s theory and the constitutional review debate revisited in this Article were concerned with a traditional separation of powers structure in an autonomous nation state. Yet the ideas about courts protecting democracy are relevant for potential contemporary supra-national institutional intervention on populist-led degradation of constitutional democracy in the European Union. Populists in power in Poland have in recent years implemented changes, largely through constitutionally valid procedures, that very substantially affect the composition of the Polish judiciary.Footnote ⁸⁴ The common critical assessment of this development is that it serves, and was designed, to diminish judicial independence, which in turn tends to nullify the courts as a bulwark against the closing down of democratic competition by the populist PiS party. Among various ways that the EU institutions may seek to help halt the erosion of constitutional democracy in Poland, there is contemplationFootnote ⁸⁵ of the Court of Justice of the European Union (CJEU) evaluating and finding systematic rule of law deficiencies in Poland by reference to Article 7 of the Treaties on European Union pursuant to an Irish court’s refusal to extradite a suspect to Poland out of fair trial and rule of law concerns.Footnote ⁸⁶ This would involve reformulating the Irish court’s reference to the CJEU; as such, it is acknowledged that it would seem “activist.”Footnote ⁸⁷

As Andrea Pin observes, the Polish development, among others, exposes a historical lack of anticipation of the scope for regression, rather than simply progress, in political developments in terms of the core political-constitutionalist values of the European integration project.Footnote ⁸⁸ We cannot go back in time, but relevant institutional actors can innovate in a limited way to protect the European project. So far as the rule of law has been operationalized as justiciable,Footnote ⁸⁹ it becomes a valuable tool for careful judicial use. The rule of law is under contest in the Polish case, but at issue is the erosion of constitutional democracy. The rule of law, as it happens, has been made justiciable and is perhaps a less contestable concept than democracy,Footnote ⁹⁰ and, as such, the more secure standard, or set of standards, to enforce. The sense in which the rule of law is said to have been operationalized shows a process consistent with Dixon and Landau’s anchoring method, as noted above. The rule of law is accordingly a suitable value under an Elysian framework for protecting democracy—that is, insisting on the rule of law being maintained as a guard against the erosion of democracy—notwithstanding that the rule of law does not necessarily entail democracy. CJEU intervention in the Polish development would clearly be controversial and liable to be called undemocratic. Elysian analysis helps address this understandable democratic concern. Its pertinent basic advice is that democracy can be protected by unelected bodies’ intervention, operating in a restrained manner, consistent with democratic ideals.