I.

It is difficult to disagree with the intent to make sense of the rule of lawFootnote ¹ and to extend its value onto the international realm, especially in relation to the European Court of Human Rights (ECtHR, or the Court). Moreover, given that the rule of law is a contested concept, the choice to take a step back is to be welcomed. For Follesdal, this is the methodological starting point to reassess the concept and propose a ‘philosophical’ understanding in the hope of finding some sort of practicable consensus.

As I understand his argument, Follesdal’s core concern about the rule of law is the avoidance of unchecked discretion and arbitrariness, which would transform legitimate authority into domination. Follesdal asks which interests of individuals are the ‘true grounds’ for the rule of law. This reminds us of Martin Krygier, who suggests that we should concentrate on the ‘ends’ of the rule of law instead of endlessly disputing its possible features or requirements (its ‘institutional anatomy’).Footnote ²

Follesdal also recalls the work of Phillip Pettit, for whom those decisions taken according to the decision maker’s ‘own judgements’ without reference to the interests of the affected are arbitrary.Footnote ³ Follesdal hence suggests that the rule of law expresses a comprehensive interest in ‘non-domination’ through the abuse of power and in ensuring the predictability of the behaviour of others.Footnote ⁴ From this premise, many further requisites can justifiably follow, such as impartiality, non-retroactivity, legal limitation of prerogative power and, of course, compliance with rules.

In my view, these arguments are in many respects sound and familiar, but they invite further reflection. In what follows, I consider the tenets of Follesdal’s arguments scrutinizing their sufficiency to represent and realize a rule of law ideal. On one side, we should ask what notion of the rule of law is and should be at stake; on the other side, we must understand what the rule of law would imply in a setting beyond the state, and eventually in the peculiar institutional context of the ECtHR.

Admittedly, Follesdal’s description highlights many elements that perfectly fit Max Weber’s idea of the features and functions of law in the European ‘legislative’ state of the nineteenth and twentieth centuries.Footnote ⁵ Law bestows legitimacy to the modern Rechtsstaat, provided that it works under the principle of legality, which means that the state acts through formal-rational channels and law-regulated processes. This entails that the law shall avoid problems of abuse of power and of uncertainty. It shall enable us to guide our actions and calculate certain consequences, thereby fostering common expectations protected by legal rules. These virtues – concisely referable as the principle of legality – are clearly necessary to exclude arbitrariness.

However, at this point I think one should ask whether all that really suffices to prevent ‘domination’. In short, in my opinion all this alone does enable neither non-domination nor the rule of law. The law might be non-retroactive, public and adjudicated by an impartial judiciary, and may treat equals equally, but all this does not impede law from becoming a dominating tool and a mere instrument in the hands of those in power.

II.

If one restricts the requirements of the rule of law to certain qualities mentioned above, one might argue that my objection inappropriately overstates the potential of the rule of law by assuming that it implies some specific contents of the law. Yet I do not believe that this is true. Arguably, we may be inclined to equate the principle of ‘legality’, non-arbitrariness and the rule of law, and believe that both legality and the mentioned institutional ‘anatomy’ requirements will guarantee non-domination. However, the straightforward equation overlooks that the features mentioned by Follesdal alone do not prevent the law from becoming a tool of domination. This is so because these features do not take into account how the law is structured and produced, do not touch upon the question of the hierarchy of legal sources and do not address the plurality of law-generating sources. Evidently these further concerns, which I believe are important for the realization of the rule of law, do not regard the substance and contents of the law (although they might have consequences for them).

There is no list of substantive elements to be included a priori in a definition of the rule of law (that would then become the rule of the good law).Footnote ⁶ The rule of law, understood in the sense of the original English ideal, availed itself of – and cherished – the systemic and institutional aspects of its legal context. English law established a composite order, which included conventions, common law, Acts of Parliament and judicial precedents. There was a plurality of legal sources and institutions that had cumulative but also countervailing effects. As I have attempted to show in detail elsewhere,Footnote ⁷ the rule of law presupposed more than a single jurisgenerative source, more than a sole sovereign rule-making power and more than a single authoritative protagonist. Its rationale and import are intrinsically linked to a legal and institutional notion of non-domination.

The rule of law is a credible and meaningful ideal only if it overcomes the famous Hobbesian dilemma: a sovereign who can change the law on a whim is not under the rule of law, but exercises his own rule. The Weberian maxim to govern through law, which Follesdal cherishes, ensures predictability but does not prevent or limit the law from being within the purview of a majoritarian legislator. The rule of law affords more than that. It adds a structural organization of legality by limiting the law of the sovereign through another law that the legislator cannot override. In this very sense, the sovereign is not legally allowed to dominate through law.

There are two sides of ‘positive’ law: one side remains instrumental to sovereign power, the other remains beyond its purview. By referring to this structure as a duality of law, I mean that besides the law of the sovereign legislator there exists ‘another law’ that the former cannot directly control or arbitrarily change.Footnote ⁸ Such a duality, resembling the medieval pairing of Bracton’s concept of jurisdictio-gubernaculum,Footnote ⁹ permits that beyond the monopoly of the sovereign legislator wider interests of society are voiced and legally protected without or against the will of the sovereign. It is in this sense that the rule of law protects against unilateral and arbitrary decisions and displaces the rule of men (the sovereign).

If the law were under the exclusive control of a more or less ‘gracious’ sovereign authority, then the fabric of law would be monolithic and there would be no competing jurisgenerative source. Where the law is under the monopoly of one supreme sourceFootnote ¹⁰ (a dominating authority), the rule of law is structurally absent, regardless of whether the content of the law is valuable or worthless.

The rule of law, then, is not only about the separation of powers and corresponding institutional set-ups; it is also about the structural organization of the law itself: where the rule of law prevails, legality entails a duality of law where neither side can supersede the other. At the core of the idea of the ‘rule of law, not men’ lies a deliberate tension between equally legitimate legal sources that are both concurrent and competing. Another law (not mere natural law) limits the law that the most powerful can make and use.

The existence of ‘another law’ – a ‘positive’ law – which the sovereign cannot lawfully overwrite features in different systems and in varied ways. For example, contemporary European continental constitutions provide that some principles and rights are of equal (or higher) legal strength as (democratic) legislative authority.

After clarifying this conception of the rule of law, we can now return to the question of which interests of the individual the ideal of the rule of law protects. As mentioned before, at the centre of Follesdal’s paper lie interests in avoiding abuse of power and arbitrariness. In my opinion, the answer needs to be based on the historical development of the ideal, which points us to the persistent concern for the fundamental value of liberty.Footnote ¹¹ Since Follesdal evokes non-domination as an essential aspect of liberty, we should elaborate this notion in order to understand its whole import. As Pettit aptly explains, non-domination and liberty mean not only non-interference, but the lack of a capacity to interfere on an arbitrary basis.Footnote ¹² Non-domination requires that the lives of individuals are not under the good will of someone else. If, according to Pettit, domination can exist even without actual interference, then we can say that it can exist also without formally infringing the law, without non-compliance with rules, retroactive application of law or an unfair judiciary. The question, then, is how the mere capacity of arbitrary interference can be countered.

This can occur either through the organization of law or through the organization of political power. Non-domination implies either the legal protection against domination by others or, according to the political ideal proposed by Pettit, the equality of power held by all people in a society. The rule of law concerns the organization of law. Pettit focuses on a political remedy to domination.Footnote ¹³ He does not try to limit the law in the hand of an omnipotent sovereign through the strength of some other countervailing law. Instead, he addresses directly the ‘political’ question and wants to shape the ‘sovereign’ as a good and fair authority based on his valuable theory of republican democracy. The cherished ideal at the centre of his theory is not law, but rather a republican ideal of sovereign authority. As I have tried to explain above, I contend that if we want to understand which contribution law and the rule of law can make to non-domination, then we need to acknowledge that law itself must possess a particular organizational quality of its own, which I identified in the ‘duality’ of law.

The fact that the sovereign legislator has legal limits that it cannot legally override is not merely about a certain distribution of power in a political sense: it does not affect how the government is organized, but regards the law generation and its sources. The duality cannot be equated with the separation of powers alone. It does not concern the simple fact that the judiciary is only subject to the law (and not to executive power, for example), but regards the sources of the law that the judiciary has to apply and the legal limits to the law-making by those in power. Such legal structures protect the weaker parts of society, such as minorities, and other fundamental interests by limiting the sovereign’s contingent legislative will.Footnote ¹⁴

III.

The second main issue in Follesdal’s article concerns the ECtHR. The question here is how this supranational court can fulfil its functions without itself becoming a dominating and arbitrary authority. Here the margin of appreciation ‘doctrine’ lies at the centre of the discussion. The ECtHR uses this doctrine to mitigate the institutional confrontation with national legal systems in areas where it cannot identify a widespread consensus between member states.

Follesdal investigates what the rule of law requires in this context in light of the interests in non-domination and non-arbitrariness, and focuses in particular on questions of judicial accountability and impartiality. The central tenet is clear: should the supranational judges be biased and act arbitrarily due to their personal preferences, ideological and cultural beliefs or a rigid fidelity to their home states, then the rule of law would be at risk.

I think there is no question that the requirement of non-arbitrariness applies also to international courts, which require judges to comply with the law, to remain independent and to be impartial. But again, what is necessary is not always sufficient. By focusing upon the use of the margin of appreciation doctrine at the ECtHR, Follesdal introduces a further and slightly different issue, one that in my view points to the wider environment and the context of reference for the international rule of law. This has to do with the institutional role of the Convention and with the soundness of a technique of argumentation. Before I return to the rule of law and to the margin of appreciation in this context (see section IV), let me thus posit first some general observations concerning the interpretive power and judicial discretion.

We know that judging is always open to contestation, even when it is done by an impartial court whose Bench is appointed through the best possible procedure. Judges can always disagree among themselves, and for a number of reasons their decision-making might appear discretionary. Some observers may always suspect that the rule of law is lacking in the work of a court.

Yet, when the independence and impartiality of a court are generally accepted, the exercise of legitimate interpretations does not impair the rule of law. The rule of law cannot (and should not) require that judges do not disagree and that judgments are not disputed.Footnote ¹⁵ Questions of indeterminacy in law are the affair of any court, and judicial discretion is a classic recurrent issue for legal theorists (which they answer in varied ways). Most law students would think, for example, of the exchanges between Herbert Hart and Ronald Dworkin in this context. This comment is not the place to recount these debates. I simply point out that a common rule of law requirement in legal adjudication is that judges show a respectful ‘cognitive’ attitude towards ‘positive law’ and exercise their profession impartially.Footnote ¹⁶ Furthermore, the work of courts is not a series of isolated dictums, it is a continuous web of interactions among past decisions, new circumstances and implicit (or sometimes even explicit) dialogue between the parties involved and the beliefs of the wider public: it is a web of ideas where courts have a temporary but on occasion decisive primacy vis-à-vis other actors and political authorities. The judicial standards under these premises become relatively knowable and predictable, apart from those innovative moments of highest courts at the juncture of epochal turns.Footnote ¹⁷ Nor are divergences between judicial decisions and the opinions of the affected people unusual, given the normative function of courts. The same applies to the ECtHR: opposition from member states that are found to have violated Convention’s rights is part of the institution’s work. Courts have by definition a different raison d’être than just following a democratic majority decision.

It is against this background that we need to think about the question of whether the ECtHR’s margin of appreciation doctrine, included in Protocol 15 to the Convention,Footnote ¹⁸ poses a danger to the rule of law. We cannot expect that the doctrine provides a ‘solution’ to the recurrent problems and tensions inherent to legal interpretation, nor should we assume that the doctrine is the main cause of contestation despite the many reasons for its criticism that Follesdal aptly collects. With this caveat in mind, we can now return to the question concerning what the rule of law ‘means’ in a supranational context.

IV.

As I said earlier, in order to guarantee non-arbitrariness and non-domination, the rule of law requires an organizational fabric of legality that reflects law’s intrinsic ‘duality’, one that has been identified since medieval timesFootnote ¹⁹ and that consists of complementary and at times conflicting relations between the law that powerful authorities can ‘generate’ (and use) and a coexistent law that works as a limitation to that. While contemporary constitutions attempt to fulfil this function in the domestic sphere, similar structures of a separate but permanent legality developed more slowly in the supranational context. In fact, the mentioned ‘duality’ emerges in the international order as well. Beyond the states’ prerogative to negotiate their own interests in the traditional domain of treaty-making, an ‘other’ international law developed in the last 70 years or so, especially starting from the Universal Declaration of Human Rights (1948), and subsequent legal instruments like the International Covenant on Civil and Political Rights (1966), or the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (1984), to mention a few, and the ECHR of course belongs to this line. Some international obligations have acquired erga omnes force and some norms have jus cogens status, so that states cannot override them at will.Footnote ²⁰ The establishment of multilateral treaty regimes in the area of human rights extends this duality of law beyond the domestic. We now have a ‘pluralism of legalities’ – local, regional and international – whose constituent ‘parts’ have diverse raisons d’être and complement and compete with each other at different levels.

In my view, also in this particular transnational setting, we must make sense of a rule of law ideal: non-domination here can result from fair and reasonable relations among legal orders, but such relations do not follow a strict normative or institutional hierarchyFootnote ²¹ that establishes some kind of highest authority. These relations rather result from continuous interactions, and the notion of a rule of law canon in such a context includes a fair balance both between different orders’ normative claims and between the will-dependent and will-independent laws that are contextually relevant. Insofar as it includes such a duality, even international law is increasing its capacity to mirror the rule of law. We cannot overlook the mutual otherness, the irreducible diversity that characterizes different legalities. A supranational human rights convention and, for example, the security regime of the United Nations are different from states’ legal orders. They do not possess the same normativity and social embeddedness. The international rule of law must reflect this particular setting within a pluralistic and often heterarchical network of different legalities, protecting concurring or competing values in (and of) different legal regimes.

Thus the non-domination caveat included in our ideal of the rule of law requires us to recognize the continuous interdependence among different systems and the distinct normative reasons (and ways of reasoning) that apply to the different actors in their interplay. This is the rationale behind such ‘doctrines’ as the margin of appreciation and subsidiarity. Another example is the famous ‘solange’ method and its ‘equivalent protection’ requirement that are already well known and practised by national and supranational courts at the wider European level.

The margin of appreciation can represent a non-domination caveat: it conveys the premise that an earnest appraisal of different orders and their own ‘normativities’ is necessary in the ‘beyond the State’ setting. The ECtHR has multiple tasks: it adjudicates rights enshrined in the Convention while taking into account the justified normative claims raised by member states, and either it ensures a common protection standard (throughout the Council of Europe) or it decides to protect diversity. The margin of appreciation appears as a synthetic expression for these multiple substantive and institutional concerns.

Follesdal aptly lists this oscillation between attitudes of the Court. The ECtHR at times follows a rights-enhancing deontological approach and at other times prioritizes a deference to states’ democracy. Such a deference leaves to member states some discretion about the interpretation of rights and the relation of these rights to other relevant public interests and values.

The quest for an equilibrium in these transnational settings is not necessarily harmonious, but nor is it hostile. The structural extension of the duality of law that I emphasized in the context of the rule of law can also be seen in Article 53 of the ECHR, which stipulates that, ‘Nothing in this Convention shall be construed as limiting or derogating from any of the human rights and fundamental freedoms which may be ensured under the laws of any High Contracting Party or under any other agreement to which it is a Party.’ In other words, the ‘authority’ of the Court cannot undermine a higher protection standard in a member state. It is ‘only’ a check of fundamental requirements and standards.

In principle, the confrontations and possible ‘dialogues’ between legal orders (e.g. domestic and regional) and between institutions (e.g. the ECtHR and national supreme courts) do not undermine the rule of law.Footnote ²² On the contrary, these interactions reinforce its spirit, impose a double-check upon the discretional power of either side, reduce the possibility of arbitrary interference and thus confirm the duality-based organization of the law. In my opinion, this structure is precisely what the rule of law requires.Footnote ²³

Nonetheless, as we will see in the next section, should the decision to grant a margin of appreciation be based on unreasonable criteria, the judicial practice could indeed become arbitrary. The same can happen also by granting no margin at all despite varying practices in member states or by granting an excessively wide margin that favours a particular member state. This would violate the rule of law because it affects the institutional balance (of acknowledged reasons) that needs to be safeguarded in the interplay between different normativities.

V.

Admittedly, the structural virtue of the margin of appreciation does not necessarily solve practical problems. Whether a consensus between member states actually exists, and thus whether the margin may be granted and to what degree, leaves the Court with considerable discretionary power. But is it also possible that this becomes an arbitrary power?

We have seen already that the doctrine recognizes the role of peoples and their legal systems in interpreting rights and values of the Convention, especially where no minimum consensus exists between member states. On the other hand, granting this margin can undermine the counter-majoritarian function of the Court and reduce its strength – for example, in protecting minorities. But we also need to recall that the Court does not grant any margin if the alleged interference against Convention’s rights is disproportionate to the objectives sought, that is, if the state fails the proportionality test.

All in all, as an argumentative doctrine, the margin of appreciation clearly retains some vagueness, one that might weaken the belief in balanced relations between member states and the Court and might jeopardize the Convention’s fundamental mission of effective rights protection.Footnote ²⁴ In fact, the invocation by the Court of such a ‘margin’ as a kind of independent and pre-defined border or as a ‘pre-existing’ principleFootnote ²⁵ at times proves to be confusing: it alludes to something that should be allegedly ‘found’ as fixed in advance and accordingly capable to limit or restrain the Court’s discretion.

The inclusion of the principle of subsidiarity and the ‘constitutionalization’ of the margin of appreciation in Protocol 15 are often deemed to establish a more precise limitation of the discretion of the Court,Footnote ²⁶ but they can also be seen as a final conflation of inter-institutional concerns (regarding some respectful attitude of the Court vis-à-vis the states that have signed the Convention) with rights adjudication that distracts the Court from its primary judicial function. In other words, the margin is also regarded as a legalistic tool to disguise ‘political questions’ as issues of rights interpretation. Indeed, this can become a danger for the rule of law.

As Follesdal writes, the practice as a whole would become arbitrary when it is based on inconsistent reasons, ideological bias or unjustified deference to a member state. Of course, that does not hold where such a behaviour is purely occasional, or if sporadic interpretive disagreements arise between the Court and a member state: these can be managed reasonably through further ‘communication’ between national courts and the ECtHR.

The margin of appreciation is a delicate instrument, since it is based on discretion insofar as it is for the Court to decide on a flexible basis whether and to what degree to grant a margin. But at the same time the margin also restrains the ECtHR and prevents it ‘from exercising further oversight into the laws of the member states for the future’.Footnote ²⁷ It is also worth considering that, on the contrary, if no margin was to be granted whatsoever, this would result in an inflexible jurisprudence that would betray the institutional character of the ECHR and thus also be arbitrary. So far, the practice of the Court has successfully avoided the appearance of systemic arbitrariness.Footnote ²⁸

In light of this, on the one hand I agree with Follesdal that the primary concern for a court should be impartiality: supranational courts should not be biased and should not act arbitrarily due to the personal preferences of their judges. Their ideological and cultural beliefs or a rigid fidelity to the interests of their home states would put the rule of law at risk. Yet, on the other hand, one needs to consider further aspects beyond this concern: the specific context of the ECHR requires even from the perfectly impartial and independent judge a careful consideration of the intertwined deontological and institutional concerns. In concrete terms, this might require, among other things, a better argumentative justification of the margin of appreciation that, for example, avoids the impression that a (fictitious) margin ‘pre-exists’ and is merely ‘identified’ by the Court according to some controversial criteria. As part of its mission, the (use of a) ‘margin of appreciation’ not only has to be but also ought to look fair – just like Caesar’s wife, who ought to remain beyond suspicion.Footnote ²⁹ I would uphold the hope that the Court will increasingly prevent and dispel the fear of arbitrariness and be able to develop and refine its ‘doctrine’.