Constitutional courts and relations between parliaments and executive authorities.
The limits to court regulation of a system of government
Armel Le Divellec
The contemporary doctrinal craze for constitutional justice and the exaltation of ‘constitutional democracy’ considered primarily as a regime for guaranteeing individual rights has somewhat overshadowed a non-negligible problem – the problem of the role of the constitutional court in relations between governing organs, namely the deliberative assemblies and the authorities wielding immediate power, in other words the (poorly named) ‘executive’ authority (government and, as the case may be, the head of state). Now, unless it is to do no more than deal in generalities, any reflection on the possible transformation of the representative regime by the advent of constitutional justice must include and so inquire into systems of government proper, that is, nothing less than the question of by what (relatively) stable arrangement, framed by law and more or less accepted by the majority, a country is supposed to be governed.
It seems that legal doctrine wavers on this matter between two extremes. One strand that generally exalts the incursion of the courts into these relations (‘politics grasped by law’) stands opposed by the strand that is largely silent about this presence. These two positions are not without their drawbacks, the main one being the failure to describe a reality, namely that there are nowadays few (if any) countries where, contrary to what was long the case, the courts do not intervene in some way or other in relations between parliaments and executives. Just how extensive the phenomenon is remains to be seen.
The initial title of this paper, suggested by the organisers – ‘Transformations in the fusion of parliamentary and governmental organs under the authority of the constitutional court: court-regulated law-making capacity’ –, sides with the first strand, the one claiming there has indeed been a transformation under the influence of the constitutional court.
Moreover, the theme was planned for the half-day session entitled ‘From the balance of power to the supremacy of the constitution’. This wording, too, suggests a transformation. It infers that before (in the absence of institutions of constitutional justice) relations between organs of government were self-regulating, implicitly through the balance of power (the inference being in a relatively mechanized manner, arising from a balanced organic and functional arrangement, but dependent in part, too, on the relative political clout of the actors) and that henceforth the constitutional court, as a third-party to the relation between governing organs, should take on the crux of such regulation by giving effect to pre-established legal rules. Accordingly, on this hypothesis, the modern representative regime is marked by both a ‘juridicisation’ (the existence of legal rules governing such relations) and a jurisdictionalisation (there is now a court to enforce those rules) of the arrangement between authorities. Consequently, by this hypothesis, the principle of the balance of power has been superseded as it were by the principle of the supremacy of the constitution and, in that, it could be considered that we are witnessing a modernisation of the representative regime.
However, this way of analysing what is at issue in contemporary representative regimes is wide open to challenge. Admittedly, there is no denying that this movement exists (and has done since the nineteenth century) and that relations between parliaments and executives are exercised in rather different legal circumstances than before; but there is no denying either that the movement is only partial and that it does not (and cannot) see this logic of juridicisation and jurisdictionalisation through to the end. The movement can only be partial because the nature of this part of constitutions (‘the arrangement of powers’) does not lend itself either to complete formalisation by legal rules or to overly stringent court supervision, especially in parliamentary type systems. However, the subject does tell us something about the function of the constitutional court and ultimately, about the concept of constitution.
Most contemporary representative democracies are marked by what could be termed a constructivist rationale: the normative framework that is supposed to organise democratic life is viewed as the outcome of a construction, by human will power, using rules laid down at least in outline by that eminent human will that is represented by the constituent power. There is hardly any longer any question of considering – as was long the case – that this normative framework is primarily the outcome of history and the product of not exactly spontaneous developments but at least developments resting on “the order of things”. In addition, it is a certain idea of law that drives the constructivist conception: the rules set out are (and must be, according to mainstream thinking) above all legal rules, which are supposedly more objective than simple ‘political’ or informal arrangements.
Moreover, ever since its origins, constitutional justice has intervened in the interplay of the powers, more exactly in the relations between the organs of government, whether directly or indirectly. But the situation varies somewhat from country to country. The same mechanisms are not to be found everywhere; some countries go further than others in this respect. Contemporary democracies still remain today therefore quite asymmetric on this point.
It is not a moot point to observe that historically, in some countries at least, the earliest constitutional courts in the nineteenth century had been established precisely to take charge of certain questions pertaining to relations among organs much more than questions about the control of the law (e.g. judgement on the responsibility of ministers for ‘violation of the constitution’ or determining the controversial interpretation of an article of the constitution). Contemporary doctrine has lost sight of this, dazzled by the glare of the question of the hierarchy of norms.
In any event, although the mechanism of court ‘resolution’ of disputes among organs by way of direct application to the court has not been extended to all constitutional democracies by a long stretch, the fact remains that the constitutional court may be led to determine such disputes among organs or to state the ‘proper interpretation’ (that is the interpretation that, further to Kelsen, we customarily call ‘authentic’, whereas logically, the term should be reserved for the interpretation by the institution deemed to have laid down the norm, namely the constituent power) of a stipulation of the formal constitution in the course of other proceedings: judgment on the rules of a parliamentary assembly (automatic jurisdiction under the French Constitution, as is known, but that may be exercised besides either upon application or indirectly), ex-ante or ex-post examination of a referendum law (France, Italy, German Länder, etc.), (abstract or concrete) supervision of norms (in the course of which precedence goes to examination of any complaints of breach of parliamentary procedure), or by direct constitutional appeal (e.g. the Verfassungsbeschwerde in Germany). In any case, the dimension is both organic and functional (but this is how Bagehot understood the fusion of powers).
I. Growing court control over the interpretation of sporadic norms of positive law framing the relation between parliament and executive
There is no doubt that the existence of constitutional courts (broadly defined, that is, in material terms) in Western parliamentary democracies affects the law governing relations between deliberative assemblies and the organs of executive power. Beyond the diversity of circumstances that vary by country (depending on the variety of competencies of the court, the way cases are brought before them, the propensity of constitutional actors to bring cases before the court in actual practice, the readiness of the courts to respond to those applications), there is no denying the hold of the courts on the interpretation of positive law relating to it, which has been growing since the second part of the twentieth century.
One of the reasons is, apart from the actual creation of the courts (mostly after 1945 for the courts specialised in constitutional matters), the growing juridicisation of relations between the executive and the parliamentary assemblies in twentieth-century constitutions. Whereas the parliamentary system of government has flourished, within the unchanged framework of limited or dualist monarchies, by practices and then conventions, that is, without legal rules explicitly enshrining them, its principles (and sometimes some of its working arrangements) have since been codified in many countries. After the constitutional laws of France’s Third Republic in 1875, it was only with the wave of new constitutions in Europe after 1918 that there came legal provisions mentioning the political accountability of the government before parliament and various rules deriving therefrom as a positive legal principle. Commentators have spoken of the ‘rationalisation’ of parliamentary government (Mirkine-Guetzévitch). Although it probably failed to encompass all aspects of the question, at least it produced a major shift in the formal legal framing of relations between the executive and the assemblies.
Under these circumstances, examples of court intervention in relations between executive and deliberative assemblies are countless. Italy’s Constitutional Court accepted in 1996 that the Senate could debate and vote on an individual motion of censure against a minister (the 1947 Constitution is silent on this point). In Japan, the Supreme Court ruled in 1953 on the extent of the right of dissolution (it validated a broad interpretation, favourable to the government).
Generally, it can be observed that the principle of autonomy of parliamentary assemblies is less vigorous than it used to be under the dual combined effect, in many countries, of the growth of constitutional texts framing relations between the parliament and the executive organs and the advent of constitutional justice. Change is perceptible even in states conforming to the Westminster model (and even to some extent Israel).
In France, everyone knows about the ‘paradigm shift’ brought about by the creation of the Constitutional Council in 1958, particularly as it was principally designed as a ‘weapon against misdirection of the parliamentary regime’ (Michel Debré), that is, as an institution to prevent the assemblies from circumventing or violating the numerous constitutional provisions limiting their status and their role and, correlatively, to safeguard the new margins of autonomy afforded to the executive. It is widely known how successful the operation was and even if the Constitutional Council subsequently invented a new role for itself in protecting rights and liberties, it has continued ever since to adhere closely to its initial mission. The tone was set as early as June 1959 with the decisions about the assembly regulations (Parliamentary Standing Orders) which, in particular, censured the very principle of political resolutions (whereas the formal Constitution itself did not say a word about them) and the restrictions on the government’s unlimited right to speak (adhering to the letter of article 31(2)(C)). The Constitutional Council clearly set out to serve the ‘spirit’ of the Constitution called for by General de Gaulle. It has occasionally alternated the methodological registers, sometimes making a literal application of the texts, sometimes, on the contrary, departing far from the letter of the texts and consequently being highly inventive; but it has essentially been constant in its severity with respect to the chambers and generous in maintaining the autonomy of the governmental function. And yet (and this is not in contradiction with the foregoing) the Constitutional Council has also been careful to defend the ‘rights of Parliament’ to some extent, essentially in the legislative (notably with the question of the negative absence of jurisdiction of the legislature or the right of amendment of members of parliament) and budgetary fields. But, as shall be seen later, this case law resulted above all in making allowance for parliamentary debate itself and therefore in giving the floor to the opposition; it sometimes proves as troublesome for the government as for the parliamentary majority, without really affecting the core of their relations. It should be added that the court’s intervention may also concern the president of the republic, although more seldom so. On this point, the Constitutional Council tends to give precedence to a very peculiar interpretation of the Russian-doll principle that it calls ‘the separation of powers’.
In any event, we shall confine ourselves here to a neutral and quantitative observation: the creation and then the progressive developments of the Constitutional Council since 1958 (and those of the State Council) have led it to have a greater hold over the law of relations among the organs of government, weaving a tight mesh that constrains the organs of government not just in their public policy decisions but also in their mutual relations. In this simple respect, the circumstances of French constitutional law have plainly changed substantially in just short of six decades.
Thus France has joined in with a movement begun earlier in countries that already have well-developed constitutional courts. The most eloquent – and probably most radical – example is Germany. The Federal Constitutional Court was set up in September 1951 and has since played an extremely active part in German democracy as everyone knows. It is important both qualitatively and quantitatively (the 135th volume of its collected decisions came out early in 2015, which says plenty about the scope). While upholding fundamental rights accounts for most of its activity in terms of volume, the Karlsruhe Court has a string of attributions entitling it to intervene in almost any domain and in all circumstances of constitutional life, and especially, of course, in relations between the assemblies and the federal executive.
Although the West German democracy founded in 1949 quickly found its marks and developed harmoniously all in all, in a radical contrast with its Weimar forerunner, which hardly ever ceased to be tormented by constitutional controversy and conflict, relations between organs of government under Germany’s Basic Law have been the subject of multiple court rulings and primarily rulings of the Karlsruhe Court.
These include observance of the rights of the minority in commissions of inquiry, government refusal to disclose evidence in the inquiry procedure, the extent of rights that the Bundestag rules can recognize for groupings (Gruppen) that do not have enough members to form groups (Fraktion), the composition and powers of the joint Bundestag and Bundesrat conciliation committee in legislative matters, voting arrangements in the Bundesrat, election of the Federal President, the allocation of speaking time in the Bundestag. Because each Land has its own constitutional court, case law is not confined to Karlsruhe. Thus a judgment of Saxony-Anhalt Constitutional Court of 29 May 1997 had to determine whether the extreme left PDS group that supported without being part of the ‘red-green’ (SPD-Ecologist) minority cabinet should continue to receive the share of public financing reserved for official opposition groups, that is, ‘those that do not support the Land’s government’ (under article 48 of the Land’s Constitution).
Is a general direction discernible in this case law, which is more abundant in functional than organic terms?
In the early days of the Federal Republic of Germany, the Karlsruhe Court, haunted by the failure of the Weimar Republic, sought to avert possible excesses of the Bundestag by limiting its intervention and drawing a sort of sphere specific to the federal government. This concern did not vanish in the following decades, even when the actual balance of the German parliamentary system was consolidated, marked as it was by a degree of stability and government authority never achieved under the Weimar parliamentary regime. The Court stood by an interpretation in terms of a fundamental balance between parliament and government, considering in particular that under the Basic Law there was no ‘general supremacy of Parliament for fundamental decisions (kein allumfassender Vorrang des Parlaments bei grundlegenden Entscheidungen) nor any overall parliamentary reserve (kein allumfassender Parlamentsvorbehalt), nor any monism of power (Gewaltenmonismus), but on the contrary that a core of government responsibility (Kernbereich exekutiver Verantwortung) should be recognized’.
And yet, at the same time the Court began to seek to defend the Bundestag’s prerogatives in legislative matters ‘against’ the Federal Government as it were so as to limit any overextension of the domain of regulatory power. In this sense, it developed a ‘theory of what is essential’ (Wesentlichkeitstheorie), by which ‘all fundamental questions directly concerning citizens must be determined by a formal statute’ – a move that contributed greatly to legislative inflation. Outside of the legislative domain, it has progressively extended this idea to that of ‘parliament’s reserve’ (Parlamentsvorbehalt): all decisions that are fundamental and essential for the community must be legitimised by Parliament. It materialised this strikingly over dispatching armed forces (Bundeswehr) abroad to participate in international peacekeeping operations. There followed a long string of decisions in particular over European integration. But it should be emphasised that it was extremely constructive case law, not to say purely judge-made, with no precise grounding in the text of the Basic Law, as shall be seen below.
In any event, there should be no misunderstanding about the meaning of this type of case law. In defending the formal competencies of the Bundestag, it is primarily the parliamentary majority and the federal government that were targets and the opposition (the minority) that was thereby defended since each parliamentary debate and vote provides it with a platform. Admittedly the Constitutional Court’s insistence on the formal competencies of the chambers is one way to strengthen the position of the parliamentary majority with respect to the government which it in principle supports, and means the government listens more closely to its majority. But the appreciation of this jurisprudence extends beyond the government/assembly divide alone and concerns equally (and often more so) the divide between the majority and the opposition.
If we look calmly at this vast deployment of case law and bring it into proper proportions in terms of the structure of the legal order, it has to be acknowledged that what the constitutional court does here is to specify some point or other of positive law. Sometimes by giving effect to the statements of the formal constitution, converting them into operative positive legal norms, and sometimes by achieving the same outcome in the absence of any textual statement. Now, as observed above, case law is very often ‘constructive’, ‘inventive’ and even ‘judge-made’. In many instances (especially in Germany but also in the USA and Israel) and in radical contrast with French practice, the court does not refrain from going into a genuine argumentative (or justificatory) dissertation, administering on almost each occasion a form of lesson in constitutional law. But whatever the argumentation of court decisions, whether they are highly developed or laconic, the situation is the same. In deciding on this or that question, the constitutional court is always compelled – albeit in different degrees and whether it says so openly or not – to reason beyond the letter of the statements of the constitutional parchment. For, in almost all of these cases, such statements taken alone are unable to provide an obvious solution and are even generally inadequate for framing a question or problem. They presuppose something else, they are dependent on other factors, that are not to be found in the text of the formal constitution. If this is the case, it is because there is in some sense ‘a constitution behind the constitution’ as Matthias Jestaedt so elegantly put it.
II. Imperfect juridicisation of the system of government
The constitutional court is, admittedly, often in a position to set out a number of rules of positive law, the rules that together make up what one might call, as indeed Dicey did ‘the law of the constitution’. But do such rules alone usher in a system of government? The answer is obviously ‘no’ – even if this ‘obviousness’, although implicitly shared by part of doctrine, is virtually never discussed in doctrine. And the matter of principle is compounded in some parliamentary type systems by specific reasons that make the problem even more intractable.
A. Rules of positive law (law of the constitution) and system of government
The system of government designates the relatively coherent and stable way in which the power to direct a political body should, by general or near general admission, be exercised. However, it does not derive automatically or mechanically from the collection of disparate norms of the law of the constitution. These provide only the bare bones of what should make up a (relatively incomplete) structural framework through which to articulate relations among the various institutions (or organs) of government. This general articulation results from combining together the various norms of the law of the constitution but also from combining them with other factors external to the norms, and in particular mental representations, possibly ideologies and literary theories, actual deeds and certain behaviours of the actors involved. The law of the constitution, in the narrow sense, of itself cannot ‘operate’ alone; what ‘operates’ (at best) is the system of government, which arises from these various interactions.
In a way, Benjamin Constant back in 1797 had spotted the difficulties of the problem inherent in written constitutions: ‘When one adds, nothing but the constitution!, one adds an ineptitude. The Constitution, the whole Constitution and whatever is necessary to make the Constitution work, that alone makes sense.’ Making the constitution work, that is making its legal provisions (which sometimes require secondary texts to clarify the first) work, relates not just to the effective character of each legal norm in isolation but most certainly above all to the comprehensive and dynamic operation of norms within a set, which, for want of a better term, we can call a system.
In truth, although usually very imprecise on this question, legal doctrine implicitly feels the need to consider the overall product of the combined norms since, instead of saying ‘the norms of the constitutional text’, it speaks more often than not of ‘the constitution’ – in other words it uses a hold-all term – and it readily resorts to the term ‘political regime’ (which is besides frequently used as a synonym for ‘constitution’) to refer to the system resulting from (or ‘produced by’) those norms.
The system of government is part of a greater whole that may be termed constitutional order, which comprises, alongside the mode of government, a system to guarantee individual rights and other structuring components – for example, in economic and social matters. The idea of constitutional order may be likened to that of legal order and borrow from it the definition proposed by Charles Leben: ‘Legal order is the name given to the set of all elements, arranged into a system, entering into the consideration of a body of law governing the existence and operation of a human community’. He adds, ‘the expression “arranged into a system” means that the legal order is not simply a collection of ill-assorted objects but an organised set of interdependent elements forming a unit, a whole, that is more than the simple addition of its component parts’. It is probably not trifling that certain constitutional texts themselves or even case law feel the need to use the term constitutional order, to convey something that is more than the collection of formalised legal norms and which (but this point requires further explanation) does not exist ‘on paper’. As a part of the constitutional order, the system of government falls within an analogous province to this, and one that cannot readily be analysed by an exclusively normative approach.
B. The necessary search for an implicit notional but relative constitution
Even if one wanted to analyse the way in which a country’s government should be led by concentrating solely on formalised – by written legal rules – relations among the main organs in the shape of the deliberative assemblies and the ‘executive’ organs, and leaving aside the notion of system of government (and constitutional courts are generally reluctant to refer directly to it), we almost ineluctably come up against the obstacle of how to interpret many of the utterances and even the concepts evoked by the constitutional text but that are not (and cannot be entirely) explained by the text itself. This is the case, for example, with core notions such as ‘executive power’, ‘monitoring’, ‘accountability’ and ‘trust’. Most constitutional texts do not even attempt to explain the role of functions attributed to an organ; they confine themselves to stating sporadic attributions. Admittedly, sometimes statements of principle are to be found, such as ‘The executive power shall be vested in a President of the United States of America’ (US Constitution, Article II, Sect. I-1) or ‘The President of the Republic is the head of the State and represents national unity’ (1947 Italian Constitution, art. 87(1)). But that just shifts the question. The same is true when the expressions are slightly more developed: articles 5 and 20 of the 1958 French Constitution, without it being possible to determine exactly what they mean. And the same again for article 24, the statement that Parliament ‘shall monitor the action of the government’.
In any event, the great majority of statements in a constitutional text presuppose numerous representations about the role devolved in principle to each constitutional organ, relations among organs and what they are supposed to be for the drafters of the text (and there can be countless compromises and misunderstandings in this respect). In other words, the statements necessarily presuppose a notional scheme of things forming a sort of implicit notional constitution. Without this, the statements would be largely unintelligible; they only become meaningful when resituated intellectually in the more general context of a system of government that shows how to impart some consistency to the rag-bag of normative statements. But this system, this ‘implicit notional constitution’, regardless of the degree of consensus among the actors in charge of the various institutions (as well as the diffuse and elusive ‘public opinion’ of commentators and citizens), is never completely unequivocal: even constitutional law specialists may not share identical views about its exact contours. It is therefore relative while essential to the coherence of the whole.
Furthermore, this intellectual schema, this ‘implicit constitution’ never comes about of itself; it requires an energy input from outside. This is supplied by the deeds and behaviour of the actors occupying the various constitutional institutions (parliamentarians, ministers, heads of state, senior civil servants, etc.) and unfolds over time, with inevitable variations, which is another factor of relativity.
Accordingly, even scrupulous compliance with the letter of the constitutional text, which, once interpreted, forms the law of the constitution, never leads to a single overall ‘outcome’, a perfectly defined system of government, that is seamless and uncontested. Thus, for example, compliance with all the principles laid down by Germany’s Basic Law led to this mixture of ‘chancellor democracy’ and ‘coalition democracy’ consistently practised since 1949. But it could equally well have resulted in a system of weak and unstable governments, or a cabinet of technical staff governing with the support of majorities of variable compositions in the Bundestag. Or again a federal cabinet inspired by the Federal President (Adenauer himself contemplated becoming head of state in 1959 and guiding the chancellor’s policy from that office). The successive changes of the Weimar Republic from 1919 to 1933, for example, bear eloquent witness to the variety of systems of government that are materially possible based on the unchanged legal framework of the 1919 Constitution. The case of the Fifth French Republic illustrates more than many others this problematic dimension of relations between law of the constitution and system of government, that cannot be dealt with in the reductive mode of narrow and ‘pure’ juridicisation.
C. Factors specific to parliamentary systems of government
The general difficulties inherent in any analysis of the complex relations that an actual system of government maintains with the rules of law are compounded in the particular instance of parliamentary systems (that is, those openly or implicitly recognised as attachable to it).
In the others, whether the US system or the directorial Swiss system (to take just the principal stable alternative types), the fundamental articulation of relations among governing institutions is comparatively easy to decipher (for the actors involved, judges, and observers). Not that these systems of government do not have their share of subtlety. But for the essential reason that they have maintained the general principle of the structural autonomy of each organ, without any structural – legal and political – solidarity among them. The classical and initial idea of the organic distinction between the assemblies and the executive has not yielded despite changes in context (rise of universal suffrage, development of political parties). The rationale of a balance of power in the sense of a legal system of checks and balances has basically been maintained and continues to operate freely. The various informal practices that inevitably go along with it have invariably been operative and still are operative mostly in line with this rationale. Accordingly, in such systems of government, it is still relevant (for a constitutional court as for an observer) to analyse the relations between assemblies and executive in terms of opposition between organs, an organic stand-off that is essentially framed by formal legal rules.
The situation is quite a bit different in states practising the parliamentary system of government. This system, through the diversity of its variants (both in terms of formal organisation and political dynamics), rests on a certain institutional logic, that of unity, arising from the principle of political accountability and the intrinsically political connection of trust that derives from it, that is permitted, affirmed and institutionalised by the organic and functional interpenetration between the government and assemblies, their interweaving, their ‘interlacing’ (Necker), in short their legal and political solidarity, what Bagehot (following Guizot in that) named the ‘fusion’ of powers (a term chosen on purpose to refute the common and false descriptions constructed on the idea, taken literally, of ‘separation’). This solidarity is legal, first of all, that is, structural (which is all too often overlooked), in that the ministers who are agents of the ‘executive power’ must in some instance be members of the assembly (British model) and in other instances may be members (dominant European continental model), or at least have a right to enter and speak in the chambers; they hold de facto or de iure the right to introduce legislation, participate in parliamentary debate on law making and even in any political discussion, and so bring their weight to bear on proceedings at all times. This solidarity is extended by a still strong institutional linkage of a dynamic kind: the designation (by convention or by virtue of a formal rule, depending on the country) of the prime minister by the assembly (the ‘elective function’), which is the expected face of political accountability, its logical end-point (Bagehot was one of the first to dare to raise this function into a fundamental principle of what he called ‘cabinet government’), even if it is sometimes subsequently curbed by political difficulties (for example, the hypothesis of ‘governments of technicians’) or must be reconciled with another source of legitimacy as in France’s Fifth Republic.
Fusion but not confusion. Bagehot made it clear: ‘the close union, the nearly complete fusion’. Despite this organic and by the same token partly functional interlacing, the cabinet and assembly remain formally separate in essence, admittedly, with the result that obviously there is room for relational mechanisms among these organs, mechanisms formalised by the law of the constitution. But the core of this intimate relationship goes beyond this and is already necessarily different from what it is in a non parliamentary system. Moreover, solidarity (or fusion) is also, in principle, political: the cabinet brought to power by the support from a majority is normally made up of components of that majority, and maintains close ties with it at all times (this intimacy varying with party-political situations). The principle of the cabinet’s political accountability to parliament makes it possible to maintain not just the unity of views between the two organs (failing which a defeated minister must stand down – possibly after a dissolution) but above all conveys a bilateral dynamic allowing the cabinet to guide and direct the majority and the majority to demand to be heard in return for its support. In short, the logic engendered by fusion tends to unite the government and parliament (at least a more or less fixed majority within parliament) politically and, from that point on, the ‘executive’ (lato sensu) and legislative functions are ‘virtually in the same hands’ (as Earl Henry Grey put it in 1858).
This principle of unity, organised legally in very variable ways from one country to another and practised in forms that may vary markedly with time and place, radically alters the rationale of checks and balances initially conceived by liberal theory: the confrontation between organs (at least the government and the assembly returned by universal suffrage) is no longer the driving force behind the constitutional dynamic since the government depends on the parliament (while monitoring it) in an entirely different way from its US or Swiss counterparts.
This dynamic of the parliamentary system is difficult to understand via the classical legal approach because all the components of interpenetration between parliament and government have been grafted onto a formal architecture that predates the parliamentary period (roughly, a system of balance of powers) and live in articulation with it. This framework has not always adapted itself explicitly. Some illustrations shall be provided below.
Moreover, the rationale behind the parliamentary system – if examined carefully – rests even less than its competing models on formal legal norms. In some instances it operates in an entirely conventional or ‘customary’ way (this was the case of most European monarchies in the early twentieth century and remains so in some countries like the Netherlands, Canada and Australia). In other instances (now the most widespread) some of its principles and/or rules of organisation are stated in constitutional texts, although they are not sufficient to explain ‘everything necessary to make it work’. In both cases, as seen above, the principle of parliamentary government requires, in addition to the rules of constitutional law, a set of intellectual representations, practices, behaviours and more or less informal arrangements between the institutions and the forces animating them. It is not so much some shortcoming in the precision of legal rules that is at issue here (rules that are too developed seldom solve all difficulties and may throw up fresh ones – again Benjamin Constant had sensed as much), but rather the fact that the parliamentary system of government cannot be reduced to formal mechanisms. It contains some part that is irreducibly recalcitrant to being captured by norms (which does not mean that it is supposedly not ‘legal’; all told, it can be said that it reflects rather its nature of political law). This character necessarily has some effect on the work the courts are sometimes asked to perform.
III. The irreducibly finite character of court control with respect to the system of government
The expansion of constitutional justice should not make us lose sight of the fact that there are many instances in which a constitutional question is removed from the outset from the control of the courts. Admittedly, it is exceptional that a constitutional text should explicitly exclude as much. But more often than not, the side-lining of the court is implicit or arises from its own refusal to examine a motion, as shall be seen.
When called upon to intervene in relations between organs of government, the constitutional court may sometimes be able to determine without excessive difficulty what positive law is on some particular point (law of the constitution), that is, to specify some or other capacity to act, some competence of an organ, some procedural requisite. But it may also run up against a bigger problem: how to appraise relations between organs that are not unequivocally determined by legal norms or that are only legally formalised in a superficial manner if at all.
A. The ‘invisible constitution’ or the existence of a judge-free area of constitutional law
1) First, it is important to emphasise this essential point, that is so obvious that it is generally overlooked, namely that the advent of a system of government, strictly speaking, owes nothing or almost nothing to the constitutional court. It is the other institutions (assemblies, government, head of state) that collectively shape the system by making the first interpretations of the statements of the formal constitution, by supplementing these interpretations by secondary rules (organic laws in some countries, ordinary statutes, rules of deliberative assemblies and other texts on the cabinet and ministers, countless rules and regulations), decisions as to persons, appointments and various practices that progressively materialise the framework outlined by the constitutional document. It may well be that the constitutional court is called on very early to specify some point or other (this has been seen, for example, with decisions of France’s Conseil constitutionnel on the rules of the assemblies in 1959). But whatever importance they may take on, such court decisions remain sporadic and cannot be considered decisive for constructing the system of government as a whole. Conversely, the decisions and behaviour of other actors (institutions proper but also para-constitutional actors such as political parties or more simply private individuals) are always decisive.
The ‘presidential’ system of the Fifth Republic, for example, was not shaped at all by the Constitutional Council, no more than the Karlsruhe Court (set up only in 1951) played a role in the advent of the system of parliamentary government energetically led by the chancellor in Germany or the Constitutional Court (set up only in 1955) in post-war Italy’s ‘partitocratic’ parliamentary government. Similarly, the changes that a system of government sometimes undergoes unfold independently of the work of the constitutional court: French-style ‘cohabitation’ was the outcome of combined action by the governing institutions, without the Constitutional Council having any part in it, no more so than its Italian counterpart in the recurring phases of presidential cabinets since 1992.
Emphasising this fundamental point does not detract in any way from the fact that, mostly, the presence of constitutional instances of justice really does influence the overall balance of the democratic system and is an important factor of the system in its totality. But such instances intervene as supplementary components, generally with a moderating effect on the system of government in the narrow sense, which, at bottom, would exist anyway independently of those instances.
2) Second, one of the irreducible limits to court supervision applied to the system of government arises from the fact that the system is riddled with informal practices, that unfold on the margins of official and formalised legal processes. The jurist (even the simple observer – and all the more so the actor, whether it be a judge called on to rule on a constitutional issue or the lawmaker ‘legi-constituent’ when looking to ‘correct’ some or other element of the system) is generally ill-at-ease with this dimension of constitutional life over which he generally has no hold. And yet, it is both undeniable and inevitable. A seasoned positivist like Georg Jellinek even went so far as to consider that ‘real political forces move by their own laws, which act independently of any legal form’. There is probably no need to go so far systematically. The actors of the system of government, especially those in official positions (ministers and parliamentarians) ordinarily act largely within institutions. But it is certain that not all interactions among institutions and among actors come ‘within the reach of law’; much of their behaviour unfolds without being subject to any formalised legal rule. Constitutional law or political law is a large swath of a body of law that endeavours to frame behaviour but it cannot determine it in full, nor always directly.
This fundamental difficulty can be illustrated by an easy-to-understand example. France’s 1946 and 1958 Constitutions sought to remedy the governmental instability that had been typical of French parliamentary government since 1875 mainly using procedures ‘rationalising’ the ways in which parliament could express its mistrust or withdraw its confidence in the government. It is common knowledge how this mechanism was blithely circumvented in practice under the Fourth Republic (‘double investiture’, unofficial question of confidence, especially). It was largely for this reason that the Constitutional Council, not without audacity, prohibited parliamentary resolutions in 1959. But the Constitutional Council would be powerless if the government decided of its own initiative to unofficially raise a question of confidence in relation to a text, without resorting to the procedure of article 49(3)(C). It seems that this is what Prime Minister Lionel Jospin threatened to do over the finance bill before the Communist group in October 2000. Similarly in Germany, where the court does not systematically supervise the Bundestag’s regulations as in France, it is accepted that the Diet may pass motions blaming a minister, at the risk of ruining the consistency of the arrangement in the Basic Law that concentrates the question of legitimising the federal government on the Chancellor (art. 63, 67 and 68 Basic Law).
A further point can be emphasised about the weight of informal practice in the life of a system of government. Parliamentary control is never reduced in practice to official operations by deliberative assemblies and the formations within them. And while the opposition is generally and logically the main driving force behind such operations, we should not overlook the ‘internal’ supervision by the backbenchers with regard to the government they support. This ‘regulating’ supervision is that much more effective when it goes on out of the public eye and it is certainly more important in practice for correcting government action. It is all too familiar how the US system operates via the ‘corridor parliamentarism' on the margins of the official procedures and by which the system of sharp separation between president and congress operates more or less without systematically fouling up. There again, it is as though there was an ‘invisible’ part of the constitution, inherent in any system of government. It would be absurd to relegate this to the domain of pure fact, to challenge its legal character on the grounds that it escapes from the courts completely.
B. The court’s occasional admission of its powerlessness
It is probably because they are sometimes aware of the barely normative character of the system of government that judges, in some cases, debar themselves when called on to decide a question relating to that system. The US Supreme Court, as is well known, holds discretionary power to select the matters it will deal with and does not shy away from using it to avoid adjudicating on questions it considers too delicate. In other countries, examination of a motion may be delayed until the political context has changed and the controversy died down. This is the meaning of the ‘political question doctrine’ of the US Supreme Court, the doctrine of government acts in France, or other decisions of lack of jurisdiction and more generally the problem of ‘non justiciable acts’.
More interesting still are the cases of restricted control because the court does not exactly disqualify itself but nevertheless runs up against the limits of its office because of the particular character of the subject matter before it. An exemplary case occurred twice in Germany, that of the dissolution of the Bundestag. This matter is of particular relevance to the problem under discussion here. Not that the law of dissolution is essential to the workings of the parliamentary system (contrary to what popular belief constantly trots out about it being one of the decisive criteria), but it is still an important mechanism and its use, which is closely framed by constitutional law, went to the core of understanding of the constitution. The case was also especially interesting because it involved one of the most strong-willed and active constitutional courts in the family of western democracies, a prestigious and proud institution, that does not readily abdicate its role.
It is known that the German Basic Law of 1949 seeks to rationalise the use of dissolution, by providing that it can only be employed, apart from the specific instance of it being impossible to elect a chancellor commanding a majority (art. 63(4) BL), in the event that an incumbent chancellor is explicitly denied the Bundestag’s confidence by an absolute majority of its members (art. 68 BL). In both instances, the Chancellor (Helmut Kohl in 1982, Gerhard Schröder in 2005) called for a vote of confidence (although for different reasons and in different political contexts) so as to lose the vote and open up the path to a dissolution. In both instances, a part of the chancellor’s majority abstained so that the absolute majority was not reached. In this respect, there was, if not an abuse of procedure, in any event some sleight of hand by the parliamentary majority to ‘help’ the chancellor secure a dissolution. Twice a few members of parliament and citizens brought the matter before the Federal Constitutional Court. Twice the court dismissed the motions and ‘validated’ the dissolution de facto. But rather than merely finding that the procedure of article 68 BL had been strictly observed and dismissing the motion, it set about what might be characterised as a substantive analysis of the matter and detailed its assessment at great length. Its line of argument changed a little from one decision to the other (we will stick to what is essential). Confronted with a formal arrangement (article 68 BL), full of implied presuppositions, the Court sought to reformulate the meaning (Sinn) of the provision within the rationale of the system of government. In 1983, the Court explicitly acknowledged that article 68(1), sentence 1 of the BL was an ‘open norm’ that ‘can and must be materialised’. Such materialisation, it added, is incumbent not just on the court but also on the supreme constitutional organs, it being posited that the materialisation of the Constitution as a fundamental legal order (rechtliche Grundordnung) requires consensus among the constitutional organs.
The Court then asserted the existence of an ‘unwritten material criterion’ (ungeschriebenes sachliches Tatbestandsmerkmal) of article 68(I)(1) BL: the chancellor shall not resort to the procedure of article 68 unless he thinks he cannot continue to govern the current Bundestag; the political relations with the Parliament are such that they impede or paralyse his capacity to act, that a ‘policy sustained constantly by a majority cannot decently be followed’. For the Court, an interpretation accepting that the chancellor pose the question of confidence from the outset just to secure the dissolution would not comply with the meaning of article 68 BL.
Whereby it can be seen how very, very far the court went to seek out the ‘legal norm’, a norm of pith and moment: it thought itself authorised to resort to elements outside the letter of article 68, to rewrite a ‘logical’ schema, that of a precise parliamentary system of government whose character it dissects and into which dissolution fits logically. It was sticking its neck out. The Court attempted to give substance to article 68 BL, whereas the only ‘objective’ component of the article is a formal one: an absolute majority vote or not, within a minimum of 48 hours.
It can be seen through this example that the Karlsruhe Court has a certain idea of what the parliamentary system of government should be, it constructs the system very freely and claims it is unique. This view of things is probably entirely defensible but it may equally well be criticised: strict law (the law of the constitution) does not ‘prescribe’ the government should govern with a stable majority. It could choose to govern with alternative ad hoc majorities on the ideas to be passed. And it could be added that the resort to early elections is a qualitative factor of democracy.
Anyway, after having pretended to set a very high standard, the Karlsruhe Court suddenly backtracked finally exercising (admittedly very wisely) limited supervision. It emphasises that by providing for the intervention of three constitutional organs (Chancellor, Bundestag and Federal President) in the dissolution process, each with their own assessment of the actual political situation, the Basic Law has ‘reduced the possibilities of court verification more than in the domain of legislation and enforcement’ and has thus ‘in the front line trusted in this system of reciprocal control’ organised by article 68 BL itself. Under the circumstances, it admits that it is difficult for it to substitute its own assessment of the concrete political situation (the state of the Liberal Party, which had left the government coalition to ally with the opposition, or the individual state of mind of each member of parliament not having voted for the motion of confidence) for the concordant assessment of the other three organs. Accordingly it ended up by validating the dissolution.
In 2005, the Constitutional Court, without completely abandoning the ‘substantive’ interpretation of article 68 BL, which was even more flexibly argued, gave precedence from the outset to the convergent views of the situation of the Chancellor, the Bundestag and the Federal President and insisted even more than in 1982 on the importance of the political appraisal of the situation by those involved and the limits of the Court’s assessment of affairs.
C. When court intervention is counter-productive
Given the complexity of the internal structure of the system of government and its articulation with the rules of law, it is not surprising that, in some instances, court intervention is quite simply counter-productive, not to say wrong-headed. Several examples drawn from the case of France can illustrate this.
In 2006, the Constitutional Council censured a modification of the National Assembly’s rules that recognised in parliamentary law the notions of ‘majority’ and ‘opposition’ and conferred embryonic privileges on the latter. More specifically, the proposed reform forced groups to declare they belonged either to the majority or the opposition, with the bureau of the Assembly (supplemented, for the occasion, by the party presidents (chief whips)) ruling in the event of contention. In addition, it was provided that the groups that had declared they were in the opposition should obtain as of right for their members the presentation of reports on the enforcement of statutes and the office of chair or rapporteur within commissions of inquiry and review boards. Now, without any serious argument, the Constitutional Council censured the whole of this. First on the ground that the recognition of the ideas of majority and opposition would breach article 4 of the Constitution (freedom of formation of political parties and groups), which is to say the least a rash inference from the letter of the article; then, on the ground that the advantages bestowed on members of the opposition would ‘effectively introduce an unwarranted difference of treatment among groups’. As if the existence of a regular political opposition was not one of the ferments of the parliamentary system of government! This – plainly quite scandalous – case law somewhat hypocritically gave precedence to a degree of formalism that the court claims to find in the law of the constitution (the principle of equality among members of parliament) over the overall logic of the system of government that worked in favour of introducing this positive discrimination in favour of the opposition. Now, it is well known that on other occasions the Constitutional Council has not hesitated to go beyond the letter of the constitutional text to imagine rules that it does not contain. The prohibition of parliamentary resolutions in 1959 is the best example of this.
Another significant illustration of the counter-productive effect resulting from court intervention can be found in connection with parliamentary supervision, with decision no 581-DC of the Constitutional Council of 25 June 2009 once again on the amendment of the National Assembly rules. The Constitutional Council censured (cons. 58), virtually without any justification, adversarial debate with the ‘administrative officials of public policy’ in the context of the presentation of reports from rapporteurs of the Commission for Evaluation and Supervision of Public Policies that had been newly created by the Assembly (art. 146-3 RAN). While apparently innocuous, this censorship further revealed something deeper in the issue of court control of the objects of political law.
The Constitutional Council makes a stark distinction between evaluation (a notion that made its way into the formal Constitution in 2008 in the new drafting of article 24(1)(C)) and monitoring. The two notions, the Council argued, were not complementary but alternatives. Monitoring ‘has as its corollary the holding to account of the authority monitored’ (i.e. the government). Accordingly, the Constitutional Council inferred that the Committee of evaluation and control of public policy, newly created by the Assembly rules, had only ‘a simple informational role’ and censured the principle of adversarial debate evoked above.
It shall be observed first, with regard to the method, that the court must first make statements ‘speak out’, that is, it must explain the notions that the constitutional text itself cannot spell out (there is therefore necessarily a ‘constitution behind the constitution’, an implicit notional constitution, as said above), but that it does so poorly both because it is too laconic and because it instrumentalises a key concept of constitutional law by distorting it. Basically, then, it can be seen that the Constitutional Council here merely repeats what it said in 1959, that permanent commissions merely fulfil an informational role to enable the Assembly to exercise … its control over government policy in the conditions provided for by the Constitution (i.e. art. 49(C)). This reasoning that could be justified in 1959 to ‘clean up’ French-style parliamentary government was anachronistic in 2009. It revealed that the Constitutional Council was incapable of thinking deeply about parliamentary control in contemporary circumstances (which are obviously not specific to France for that matter), control that could not formally be limited to implicating the responsibility of the government via the motion of censorship. Pierre Avril observes that this is case law that ‘thwarts the modernisation of Parliament called for by the constituent and paralyses efforts to promote parliamentary control “worthy of a modern democracy” (in the terms of the 2007 Balladur committee report)’. The 2008 addition on the evaluation of public policies did indeed aim to legitimise parliamentary intervention in the assessment of resources implemented and their suitability for the objectives pursued. Apart from a policy, that cannot easily be justified, of resistance to the manifest wishes of the constituent legislature (and contradictory attitude with the limits usually invoked by the court itself as to its office as ‘guardian of the constitution’), this affair again reveals the difficulties of court intervention when major concepts of systems of government are at issue.
Fundamentally, as Pierre Avril puts it very well, ‘this difficulty goes to the core of understanding the constitution, because the fundamental concepts that it deals with cannot be reduced to their legal definition alone. This is true of representation and accountability with which control is related. (…) The stringent court-based approach has brought out the limits of an exclusively legal approach to the constitution by accentuating the dissociation of two dimensions which although related were previously too often treated as the easy opposition of rule (law) and practice (fact). The constitution must be read as a double-entry system, one static and the other dynamic’.
D. The indirect effects of case law
A number of court decisions relating to legal relations between assemblies and government or head of state are, strictly speaking, indifferent to the question of the system of government proper, since, as we have seen, its main outline is built by the organs of government themselves. Probably the Supreme Court decisions refine the respective spheres of intervention of each organ in the US system, but when all is said and done they only specify the system’s character (checks and balances) without really affecting its comparatively stable nature.
Court interventions seldom have much more impact in parliamentary systems since they only exceptionally concern the core of the relation between government and parliament, that is legally hard to pin down, namely the ‘fusion’ of powers.
Even so we can mention one decision by India’s Supreme Court in 2008, which – obviously with no textual basis – set a limit to the number of ministers (the 1950 Indian Constitution does not stop at authorising compatibility between the functions of minister and parliamentary mandate; it requires ministers to be members of parliament), probably to avoid the risk of excessive instrumentalisation of the majority by the government, a paralysis of control. The ‘legi-constituent’ finally followed the Court. Such case law has as its reason for being to temper somewhat the fusion between government and parliament (parliamentary majority). But determining whether the objective has truly been achieved in this way is another matter.
In France, case law of the Constitutional Council has quite plainly helped the divorce with parliamentary sovereignty called for by the executive, but that divorce cannot be said to have been achieved thanks to the Council.
In Germany, the two decisions of the Court on the dissolution of the Bundestag (discussed above) have specified a point by enshrining the discretionary right of the Federal President to decline or accept the application for dissolution made by the Chancellor. Now, in the monistic logic of the parliamentary system (which has prevailed in deeds and in minds since 1949), resting essentially upon the relationship between the cabinet and the lower chamber, it would have been perfectly defendable to have the chancellor’s choice prevail. This did not happen then by virtue of the Court’s ‘authentic’ interpretation. Here, as often when it comes to competences, the Court gives precedence to the letter of the Basic Law, at the risk of complicating the two-way interplay between a chancellor and the chamber and, additionally, leaving the question of presidential accountability unresolved for so political an issue.
In any event, it is not wrong to consider that the intervention of the constitutional court may, marginally at least, attenuate the fusion of government and parliament since the level on which the court organ can intervene is essentially that which is formalised and it therefore maintains the autonomy of the organs and their formal competences. This does not mean that, at bottom, such an intervention changes much in the actual and material political equilibrium between the executive and the assemblies (because it does not come down to formalised mechanisms), but one of the functions of law is, among others, to draw barriers and boundaries for institutions.
IV. Application of the formal constitution or partial regulation of the system of government?
The question of the role of the constitutional court with respect to relations among organs of government is therefore necessarily – unless we wish to miss what is essential – the question of its role with respect to the system of government properly understood. The question has until now been if not ignored at any rate dealt with approximately. The contemporary paradigm of juridicisation (an incontrovertible although, as seen, less intense phenomenon than it seems) and jurisdictionalisation of constitutional orders inasmuch as they establish a ‘scheme of government’, provides a ready incentive to observers to consider that the ‘constitution’ is ‘protected’ by the court, an organ that sees that it is duly ‘upheld’, as with any other court in all other branches of law. The satisfaction the jurist may feel with this presentation is understandable but is it not to good to be true?
One of the lessons from the analysis proposed here may be that it goes to show how inadequate the dominant way of presenting and justifying constitutional justice is.
A. The judge and the implicit constitution
It has been seen above that the court, in the vast majority of cases concerning relations between the organs of government, was forced to imagine and think up a sort of overall blueprint, that we called an ‘implicit notional constitution’, to be able to explain a statement drawn from the formal constitutional text. It seems to us that this is an essential prior operation to give meaning to (most jurists would even say ‘discover the meaning of’) the statements, even the most straightforward and unambiguous ones, insofar as they contain concepts that the constitutional text cannot itself explain (as with the notions of ‘executive power’, ‘general policy’, ‘monitoring’, ‘confidence’, or ‘accountability’). It is in a sense a matter of guessing about the immersed part of the concept, while the law of the constitution concerns just the tip of the iceberg.
It is a task of an equivalent kind that the court must set about when, in the absence of an obviously usable statement, it must ‘seek out’ the rule of constitutional law to resolve the legal problem brought before it.
It generally does this by resort to the ‘principles’ that are either unwritten or merely stated but not explained in the constitutional texts. Sometimes it extrapolates from a technical statement or simple mention. Thus the German constitutional court using article 38(1) BL: almost all of its case law on the ‘reserves of Parliament’ (right of participation of the Bundestag) is based on this mention of the representation for which the court abundantly develops a whole series of implications. The same goes for the principle of the distinction or separation of powers, which is frequently invoked by the courts, which the French Constitutional Council does too. It is in point of fact an argumentative process, intended to legitimise the formulation (one might say the invention) of the rule of positive law that the court is going to ask, while this — oh so complex – ‘principle’ cannot be reduced to just a simple and unequivocal technical recipe. Moreover, knowing that in doing this the French Constitutional Council rests upon article 16 of the 1789 Declaration of the Rights of Man, which all the indications are was only a very general negative principle (M. Troper), one can gauge, incidentally, what artifice there is in this case in speaking of ‘textual constitutional basis of the principle’.
In any case, examples of construction of the implicit constitution are countless. They can easily be found in the courts whose decisions are argued at length, especially the US Supreme Court, Italy’s Constitutional Court or the Karlsruhe Court. We can take up again the example of the speaking time of Bundestag members and groups: in one of its decisions, the German Court develops at length the idea that the government is more than the simple extension (Exponent) of the parliamentary majority, because it represents the power of the state, which, the court argues, justifies that speaking time not being strictly shared between majority and opposition. It thereby reconciles maintaining the traditional organic divides (Executive/Parliament) along with the new political divide (majority/opposition). Germany’s highest court regularly hands down major lessons of constitutional law, that is, it outlines each time this implicit notional constitution (its discourse necessarily varies, according to the period and the changes in its composition) that serves as a basis for its reasoning.
France’s Constitutional Council is no different, except that it gives little official weight to its conception. This has been seen with the attitude it struck over parliamentary resolutions in 1959 then 1992 and 2009: it is by (intellectual) reconstruction of what it judges to flow implicitly from a logic resulting from combining articles 20, 49 and 50(C) that it considers that resolutions of a general character should be prohibited.
Since the court can almost never mechanically ‘apply’ a supposedly unequivocal, objectively clear, normative statement, it must continually ‘tinker around’ (if we are allowed to put it that way), reformulate, reason by inserting or resituating the statement of the law of the constitution within the broader scheme of the implicit notional constitution (whatever it may be), which alone can impart meaning. It should be specified that its decision and the grounds for it serve to strengthen the rationale of the system of government already practised but may equally well thwart that rationale. And that would come as no surprise since this rationale itself is not entirely unambiguous. Accordingly, the more openly the court lays it bare, the more it risks coming in for criticism.
There remains the problem of what legal value is to be attributed to the court’s reasoning or at least to the grounds that ‘necessarily underpin’ the operative part of a decision. Through its line of argument, the court seeks not merely to justify its final decision (validate an act or not), but also, in some sense, to give legal effect to its ‘thinking’. And yet, the implicit constitution envisioned by the court is no more ‘true’ or ‘false’ than that envisioned by other actors of the constitutional order. But legal doctrine, and even those who govern, will obviously be inclined to recognise it carries a certain weight, not necessarily because of its intrinsically convincing character, but because this thinking in advance of the decision generally makes it possible to anticipate the court’s reaction in another case. From this standpoint, although largely deprived of any legal value, the court’s conception has every chance of impressing itself on people’s minds and becoming considered as the mainstream or even the only ‘right’ thinking.
B. Application or regulation?
It is obvious therefore that the constitutional court hardly ever just applies pre-established and objectively knowable legal rules (apart perhaps from in very straightforward cases, but such are rare, or when it exactly reiterates its clearly established precedent), but it almost invariably engages in a form of arbitration, weighing up several equally possible solutions. As such, the term regulation is more suited to what the constitutional judge does; it can hardly reasonably be characterised as application.
Paradoxically, in its famous decision of 6 November 1962, by which it validated (without censuring) what was to say the least the controversial use of the referendum for revising the constitution by the president of the republic and his government, the Constitutional Council boldly characterised itself: ‘it results from the spirit of the Constitution that has made the Constitutional Council an organ to regulate the activity of the public authorities …’. The expression is sometimes used in legal scholarship. Louis Favoreu once proposed correcting this equivocation: by clarifying that it was the normative activity of the public authorities and not all of their activity that was concerned, that ‘the regulation of the “political” activity of the public authorities’ was alien to the Constitutional Council’s mission, and that the ‘Council does not “regularise” political interplay’, he duly reduced the scope of the expression used by the Constitutional Council. In truth, by the same token, Favoreu introduced in turn another equivocation by using in a blurred way to say the least the notion of ‘political’, an ambiguity he maintained later with the overly fortunate title of his work La politique saisie par le droit. Being simplistic, the expression gave rise to illusions about the role of the Constitutional Council, but above all it maintained more generally a degree of conceptual confusion about the role of the court in terms of a country’s system of government.
In any event, the court performs two construction jobs first intellectually and then practically by laying down new rules to supplement the structural legal framework that was only ever outlined by the formal constitutional text. This construction work, which is sometimes cautious, sometimes (and usually) intrepid, at times hectic, at times hazardous, is at least as important (and perhaps even more so) in terms of the regulation of the system of government as it is in terms of the protection of rights and freedoms, since the former can be reduced even less than the second to formal legal rules and obeys other ends (exercise of the power to conduct a state’s policy).
The foregoing analyses tend to show that it is irrelevant to affirm that contemporary liberal-democratic constitutional systems have supposedly shifted from the paradigm of ‘the balance of powers to that of the supremacy of the (formal) constitution’, because the two terms, as I see it, are not opposed to one another. One can speak of the supremacy of the formal constitution in the sense that most constitutional texts today lay down norms about the relations between deliberative assemblies and the organs of executive power more than they did in the nineteenth century. But that does not mean that this law of the constitution is in itself enough to render operative or even effectively frame that part of the constitutional order that is a system of government, nor that when they attempt to do so, constitutional courts are always able to impose their view as to the observance of legal rules.
Fundamentally, constitutional orders have therefore remained, as before, primarily systems of organs, institutions, whose combined action, sometimes through consensus, sometimes through conflict, remains essentially decisory, sometimes despite (or without) the judge. One of the finest definitions of constitution (in the sense of constitutional order) is, in my opinion, that proposed by Bolingbroke, in his Dissertation upon Parties (1733): ‘By constitution, we mean, whenever we speak with propriety and exactness, that assemblage of laws, institutions and customs, derived from certain fixed principles of reason, directed to certain fixed objects of public good, that compose the general system, according to which the community hath agreed to be governed.’ If here we leave aside the problems of ‘fixed principles of reason’, which smacks of the Enlightenment, three points of this definition are to be highlighted for their originality:
‘Assemblage’: every constitutional order is a compound of several elements, of which the texts are only a part.
‘Institutions’: they are a key component, because, from a moderate realist point of view, it is rather the institutions that make the norms than the opposite.
‘General system’: the sundry materials of this assemblage form a system and this encompassing view of things is crucial.
The constitutional court is a party to this system. Like it or not, it is a political institution (in that sense) – inevitably so – even if it takes on the trappings of objectivity and political neutrality (and it can be agreed that it is probably often better so for the legitimacy of the democratic system). Even if it does so with specific methods, it too is therefore an actor in the ‘separation of powers’, a ‘governing power to temper’ (the organs of government) to take up the expression Montesquieu intended for the ‘body of nobles’ (that is, the upper chamber), in short a regulatory political institution. However, although its presence within a constitutional order is generally important nowadays, the institution is not in itself necessarily ‘more democratic’, since such an appreciation depends on the definition of democracy used and the conception we hold of it. It is quite possible then, subjectively, to praise the role of the constitutional court… just as it is possible to criticise it. But, in any event, doctrine would do better to recognise the irreducible limits of this instance of partial regulation of the thing that forms the heart of the constitutional order, the system of government.
 The term ‘fusion’ (popularised, as is now notorious, by Bagehot in 1865) refers to parliamentary systems of government – which are by far the most widespread type in liberal democracies – and judiciously so because it accounts better than the also overused expression ‘flexible separation’ for the legal and political dovetailing between assemblies and government (see below). That is why it is given precedence here. But the developments that follow may be valid, by extension and mutatis mutandis, for non-parliamentary systems.
 It was even said that ‘In becoming a law of precedent and an applied law, constitutional law is becoming “judge-made” law again. It is no longer a catalogue of vaguely mandatory political recipes’. Louis Favoreu and Loïc Philip, Le Conseil constitutionnel (P.U.F., « Que sais-je ? », 1978) p. 120.
 A. Le Divellec, ‘Les prémices de la justice constitutionnelle en Allemagne avant 1945’, in D. Chagnollaud (ed.), Aux origines du contrôle de constitutionnalité XVIIIe– XXe siècle (Paris, Ed. Panthéon-Assas, 2003), pp. 103-140.
 A. Le Divellec, ‘Des conflits constitutionnels dans un “Etat constitutionnel” : le mécanisme des “litiges entre organes” devant la Cour constitutionnelle fédérale d'Allemagne’, in J. Hummel (ed.), Les conflits constitutionnels, (Rennes, P.U. de Rennes, 2010) pp. 101-121. The mechanism of disputes among organs plays a non-negligible role in Germany, but remains seldom used in Italy (at least between the executive and the parliament, it comes into play more with respect to judicial power) and has virtually no role in Spain. Nonetheless, it was largely institutionalized in the constitutions of countries of central and eastern Europe in the 1990s and in South Africa (1996).
 This should therefore include, for France, the State Council (Conseil d’Etat) and (although to a lesser degree) the supreme court (Cour de cassation).
 Notably, in addition to the United Kingdom and its dominions, the Benelux countries, the Nordic countries, Italy and France under the charters. Norway did not explicitly posit in its fundamental law the principle of political accountability of the government before the parliament until 2007, although it had practised the system of parliamentary government since 1884. Even the United Kingdom, mater parliamentorum, ended up giving it a legislative basis (see A. Le Divellec, ‘Un tournant de la culture constitutionnelle britannique : le Fixed-Term Parliaments Act 2011 et l’amorce inédite de rationalisation du système parlementaire de gouvernement au Royaume-Uni’, Jus Politicum, n° 7, March 2012 and Jus Politicum, Dalloz, vol. V, 2013, p. 269-280). Today only Canada, Australia, New Zealand, and the Netherlands continue to do without an explicit legal foundation for the principle of parliamentary government.
 D. Baranger and A. Le Divellec, ‘Régime parlementaire’, in M. Troper, D. Chagnollaud (eds), Traité international de droit constitutionnel, (Paris, Dalloz, t. 2, 2012) pp. 159-193.
 S.C.C. 7/1996 of 18 January 1996.
 See P. Lauvaux and A. Le Divellec, Les grandes démocraties contemporaines, PUF, 2015, n°196 and 198.
 See the many examples in the thesis by Anne Jussiaume, Le juge et la Constitution dans les systèmes britannique, canadien et israëlien. Contribution à une théorie de la valeur de l’écrit en droit constitutionnel, Université de Paris II, 2002. The court generally respects the principle of the sovereignty of parliament, but parliament is subject to moderating influences.
 Speech to the Conseil d’Etat, 27 August 1958.
 P. Avril, ‘La jurisprudence du Conseil constitutionnel est-elle créatrice de droit ?’, Archives de philosophie du droit, t. 50, 2006, p. 33-39.
 Decision no 59-2 DC of 17, 18 and 24 June 1959. The Constitutional Council took up word for word the specifications about the resolutions with respect to the Senate (decision no 59-3 DC of 24–25 June 1959). P. Lauvaux, ‘Sur une décision fondatrice du Conseil constitutionnel’, Cahiers du Conseil constitutionnel, n°25, 2008, p. 55-57.
 Thus, for example, the Constitutional Council systematically prohibits anything that looks to it like ‘injunctions’ from the parliamentary legislature towards the governments (e.g. decision no 76-73 DC of 28 December 1976). On the other hand, it does not hesitate to consider that the ‘programme’ mentioned in art. 38(C) is not that of art. 49(1)(C) (decision no 76-72 DC of 12 January 1977). It is also worth mentioning its stringency in preventing any relaxation of the rule of incompatibility of ministers’ functions with the parliamentary mandate (esp. decisions no 71-46 DC of 20 January 1972 and 77-80/81 DC of 5 July 1977).
 For example in decision no 2009-577 DC of 3 March 2009 and the critical commentary by Agnès Roblot-Troizier, in M. Verpeaux et al., Droit constitutionnel. Les grandes décisions de la jurisprudence, P.U.F., « Thémis », 2011, p. 115-123. See also O. Beaud, ‘Le Conseil constitutionnel et le traitement du président de la République : une hérésie constitutionnelle’, Jus Politicum, vol. V, 2013, pp. 181-221.
 Without counting the decisions made by panels of three judges examining the admissibility of direct constitutional appeals (Verfassungsbeschwerde).
 See the lengthy art. 93 of the 1949 Basic Law, supplemented by the Federal Statute on the Court of 12 March 1951. l
 BVerfGE 67, 100 of 17 July 1984 and BVerfGE 105, 197 of 8 April 2002.
 2 BvE 4/07 of 10 June 2014.
 Esp. BVerfGE 84, 304 of 16 July 1991.
 Respectively BVerfGE 106, 253 of 3 December 2002 and BverfGE 112, 118 of 8 December 2004, as well as BVerfGE 72, 175 of 13 May 1986.
 BVerfGE 106, 310 of 18 December 2002.
 2 BvE 2/09 and 2/10 of 10 June 2014.
 Esp. BVerfGE 10, 4 of 14 July 1959.
 E.g. BVerfGE 1, 372 of 29 July 1952 (Franco-German economic agreement): ‘The Bundestag neither governs nor administers itself but it oversees the government. If it disapproves of its policy, it can express its lack of confidence in the Chancellor (art. 67) and so overthrow the government. But it cannot dictate policy itself.’
 BVerfGE 49, 89 of 8 August 1978 (Kalkar I).
 BVerfGE 68, 1 of 18 December 1984 (Pershing).
 BVerfGE 40, 237 (248-250) of 28 October 1975. This may be seen in a way as the equivalent of the principle of the legislature’s ‘incompétence négative’ as defended by France’s Constitutional Council.
 Starting with the judgment BVerfGE 90, 286 of 17 July 1994, and followed by many others along the same lines.
 Esp. BVerfGE 89, 155 of 12 October 1993 on the Maastricht Treaty in which the Court put forward the idea of the ‘necessary chain of democratic legitimation’ through voting in the Bundestag. BVerfGE 123, 267 of 30 June 2009, by which the Court recognises the Lisbon Treaty is compatible with the Basic Law but censors the accompanying statute on extension and reinforcement of the rights of the Bundestag and Bundesrat that fail to give these two organs sufficient say in conducting the policy of European integration. BVerfGE 130, 318 of 28 February 2012 (European stability mechanism) and BVerfGE 131, 152 of 19 June 2012 and BVerfGE 132, 195 of 12 September 2012.
 Later we shall touch upon the question of grounds that ‘necessarily underpin’ the operative wording of a decision.
 Die Verfassung hinter der Verfassung, Paderborn, Schöningh, 2009.
 An Introduction to the Study of the Law of the Constitution, London, Macmillan, 1885.
 French doctrine usually prefers to speak in this regard of ‘political regime’. This expression does not seem judicious here for two reasons. First because it has been used, since ancient thinkers, to designate more widely the forms of a state and/or its government (especially the threesome of monarchy, aristocracy and democracy) and continues to be used as such in political philosophy today. Second, it readily slips, in contemporary thought, towards broader considerations that extend beyond the specific issue of constitutional law (with commentators using without distinction the term ‘political system’).
 Des réactions politiques, Paris, An V (1797), chap. IX (brochure recensée dans Le Moniteur, n°236 du 26 Floréal et reprise in B. Constant, Cours de politique constitutionnelle, Plancher, 1819, rééd. par Laboulaye, Guillaumin, 1861, t. II, p 124). The expression is quoted by P. Avril in Conventions de la constitution, PUF, 1997, p. 4. It has seldom been observed that the Sénatus-consulte of 16 Thermidor An X (4 August 1802) stated, in its article 54 that ‘the Senate settles by an organic sénatus-consulte: (…) 2nd whatever has not been provided for by the Constitution and that is necessary to its working’. The Constitution of 14 January 1852 adopted the expression word for word (art. 27(2)).
 Among many other analogous examples, Roger Pinto wrote: ‘A constitution defines a political regime. It expresses it not just in the letter of the written texts, but in the origins and practices of power’ (‘La logique juridique des régimes politiques’, Le Monde, 5 July 1960, reprinted in Au service du droit, Publications de la Sorbonne, 1984, p 25). As Pierre Avril observed, he anticipated de Gaulle’s famous wording in his press conference of 31 January 1964: ‘a Constitution is a spirit, institutions, a practice’.
 The history of the theoretical doctrinal discourse on systems of government is still to be written. Here we can cite by way of example M. Duverger in his 1955 textbook Droit constitutionnel et institutions politiques, (P.U.F.): ‘A set of political institutions operating in a given country at a given time constitutes a “political regime”: in some sense, political regimes are constellations of which the political institutions are the stars’. Political regimes, he goes on, ‘are coherent and coordinated sets of institutions, the various elements of which it is artificial to separate. All the institutions of one and the same regime are closely interdependent’ (p. 11). All told, the image is not a bad one.
 ‘De quelques doctrines de l’ordre juridique’, Droits, n°33, 2001, p 19-39 (20).
 Especially in Germany where the term Verfassungsmässige Ordnung occurs three times in the Basic Law of 1949, articles 2(1), 9(2) and 20(3), not to mention the expression ‘fundamental liberal and democratic order’ (freiheitliche demokratische Grundordnung) especially in articles 10, 11, 18 and 21 of the Basic Law. The Karlsruhe Court makes abundant use of it from the outset. The Constitution of the Czech Republic of 16 December 1992 also mentions it (art. 3), as does the 1993 Russian Constitution (title 1).
 See a similar formulation in the French Constitution of 4 November 1848 (art. 43).
 See especially Jean-Marie Denquin, ‘Sur le respect de la Constitution’, in Les 50 ans de la Constitution, textes réunis par D. Chagnollaud, Litec, 2008, p. 113-123. Similarly, articles 56(1) and 97 of the 1978 Spanish Constitution about the King and the Government, respectively.
 While it can be easily inferred from the dominant conceptions (to avoid saying ‘intentions’, which is known to be a problematic notion) in American thought since the origins, and is reflected in particular by there being separate elections for the President and for Congress, the position is a little more complex in the Swiss directorial system since the Federal Council is elected by the Federal Assembly, a mechanism that could easily have given rise to a political connection of trust and entailed relations between executive and parliament of a type comparable to parliamentary systems. This was not the case for complex reasons.
 Which does not mean that the ‘separation’ of powers there is ‘strict’ or ‘rigid’, contrary to commonly held beliefs in France; on the contrary, it is ‘flexible’ because each organ has the means to influence others.
 On the various solutions depending on country and period, readers will forgive me for referring them to my paper ‘Le gouvernement, portion dirigeante du Parlement. Quelques aspects de la réception juridique hésitante du modèle de Westminster dans les Etats européens’, Jus Politicum, vol. n°1, Dalloz, 2009, p 185-225 (for the reference to Guizot, see p 194).
 And also, after recalling the right the British executive of the time had to dissolve the lower house at its discretion: ‘the English system, therefore, is not an absorption of the executive power by the legislative power, it is a fusion of the two’ (The English Constitution, London, Fontana, 1963, p 65 and 69). That many jurists, even eminent ones (since Esmein and Hauriou) have substituted confusion for fusion attests to a far-reaching misunderstanding, one might even go so far as to say intellectual… confusion! See the present author’s ‘Adhémar Esmein et les théories du gouvernement parlementaire’, in S. Pinon, P.-H. Prélot (eds), Le droit constitutionnel d’Adhémar Esmein, Montchrestien, 2009, p 149-182.
 Significantly, from the beginning of the eighteenth century, the principal minister in Great Britain bore the unofficial title of Leader of the House (that of Prime Minister was no more official at the time). It became more affirmed with time. Nowadays (in fact since 1942), the prime minister, the leader of the majority party, confers the office of Leader of the House on another minister, who deals essentially with questions of procedure, but the prime minister remains the true ‘head’ of the House.
 Apart from the articulation between assemblies and government, the formal constitutional framework of parliamentary government has difficulty managing the tension between the individual representative mandate of the member of parliament and the existence of political groups, the place of parties or the recognition of the opposition.
 ‘Be careful not to institute so narrow a constitution that it impedes all movement required by circumstances. It must circumscribe them, not hamper them, mark out their bounds, and not compress them.’ (Des réactions politiques, loc. cit.)
 Notice incidentally that some constitutional texts explicitly evoke the system of parliamentary government as a ‘literary principle’. The Constitution of Luxembourg is a case in point: ‘The Grand Duchy of Luxembourg is placed under a regime of parliamentary democracy’ (art. 51(1), introduced in 1948). The same goes for Sweden (art. 1(2) of the 1974 Constitution), Greece (art. 1 of the 1975 Constitution) and several others. See the present author’s contribution ‘Ecrire le système parlementaire de gouvernement ?’, to the meeting of the Groupe d’étude sieyesienne sur l’écriture constitutionnelle, Paris, May 2012, forthcoming.
 Saying that the core of parliamentary government is ‘political’ does not solve much, particularly as ‘legal’ and ‘political’ are not pure essences and not categories that are irreducibly opposed.
 This is notoriously the case of para. 9 of the British Bill of Rights of 13 February 1689: ‘the freedom of speech and debates or proceedings in Parliament ought not to be impeached or questioned in any court or place out of Parliament’. A more recent notable case is to be found with the 1950 Constitution of India, in article 74(2) about the advice ministers give to the head of state: ‘The question whether any, and if so what, advice was tendered by ministers to the President shall not be inquired into in any court’.
 And not the least of them. Is it sufficiently realised that the systematic choice of French political leaders to (attempt to) stand for the presidential election rather than aim for the position of prime minister is decisive in continuing de Gaulle’s conception of presidential primacy under the Fifth Republic? This choice entails capital consequences for the actual balance between the president, prime minister and cabinet within the two-headed executive in the constitutional system and so, by repercussion, has implications on the exercise of their respective legal competencies (withholding signatures, determining the cabinet meeting agenda, etc.).
 Particularly as the political primacy of the head of state is not exercised by formal deeds. It is rather through a form of ‘capture’ of the cogs of majority parliamentary government that the president exerts a hold over the government and, indirectly, parliament. See the present author’s ‘ Le Prince inapprivoisé. De l’indétermination structurelle de la Présidence de la Ve République’, Droits, n°44, 2007, p. 101-137.
 The same can be said of the US Supreme Court in the nineteenth century (even if it promoted the development of the powers of the Federation to the detriment of the federated states) then during the slide from ‘congressional’ towards ‘presidential’ government in the twentieth century. Its case law certainly had more impact than in parliamentary systems but even so it was the successive presidents and congresses who built the actual system of government.
 Verfassungsänderung und Verfassungswandel, Berlin, O. Häring, 1906, p. 72.
 A. Le Divellec, Le gouvernement parlementaire en Allemagne, L.G.D.J., 2004, esp., p. 368-272.
 See A. Le Divellec, op. cit., esp. pp. 307-324 and ‘Des effets du contrôle parlementaire’, Pouvoirs, n°134, 2010, pp. 123-138.
 E. Zoller, ‘Le pouvoir discrétionnaire de juger de la Cour suprême des Etats-Unis’, Pouvoirs, n°84, 1998, pp. 163-175.
 This technique is used especially by the German federal constitutional court.
 See the famous decision n°62-20 DC of 6 November 1962 of the Constitutional Council on the (constitutional) law on referenda.
 BVerfGE 62, 1 of 16 February 1983 and BVerfGE 114, 121 of 25 August 2005.
 Article 68 BL states: ‘(1) If a motion of the Federal Chancellor for a vote of confidence is not supported by the majority of the Members of the Bundestag, the Federal President, upon the proposal of the Federal Chancellor, may dissolve the Bundestag within twenty-one days. The right of dissolution shall lapse as soon as the Bundestag elects another Federal Chancellor by the vote of a majority of its Members. (2) Forty-eight hours shall elapse between the motion and the vote.’
 In 1982, Kohl had come to power by the overthrow of an alliance of the Liberal Party which had voted a constructive censure motion with the opposition. In 2005, Schröder felt that his majority (re-elected in 2002) had been weakened by electoral defeats in the Länder and was doubtful about the policy he was conducting.
 The first decision takes up more than 60 pages of the collected decisions (not including separate opinions) and the second close to 50 pages.
 It reasserts especially that the dissolution is only due and proper if it complies not just with procedural requirements but also the ‘purpose’, the reason for being (Zweck) of article 68 BL.
 It should besides be considered that the ‘abuse of procedure’ was less shocking in 2005 because the Chancellor’s majority had shown real signs of division.
 We might also cite the case law of the Constitutional Council on parliamentarians’ right of amendment, that was long too liberal, that facilitated obstruction before the ‘Amendement Séguin’ of 23 January 1987 sought clumsily to put an end to it. It has been corrected since (see P. Avril, J. Gicquel, J.-E. Gicquel, Droit parlementaire, Montchrestien, 5th edn 2014).
 Decision no 2006-537 DC of 22 June 2006.
 For further criticism, see P. Avril, ‘L’improbable “statut de l'opposition” ’, Les Petites Affiches, n°138, 12 juillet 2006, pp. 7-9.
 See P. Avril, ‘L’introuvable contrôle parlementaire’, Les Petites Affiches, 14-15 juillet 2009 and ‘Le statut de l’opposition ou l’introuvable discrimination positive’, Mélanges Yves Guchet, Bruylant, 2008, pp. 3-14.
 N.B.: it seems to me ‘formal’ would be more appropriate here.
 ‘L’introuvable contrôle parlementaire (après la révision constitutionnelle française de 2008)’, Jus Politicum, n°3 online and Dalloz, vol. II, 2010, pp. 133-139 (137 and 138-139).
 Art. 75(5): ‘A Minister who for any period of six consecutive months is not a member of either House of Parliament shall at the expiration of that period cease to be a Minister.’
 Amendment 91, adopted on 1 January 2004 thus supplements article 75(1) of the Constitution: ‘(1-a) The total number of Ministers, including the Prime Minister, in the Council of Ministers, shall not exceed fifteen per cent of the total number of members of the House of the People’.
 Besides the control achieved through examination of the rules of the assemblies, already referred to, we can cite the case law of the Constitutional Council on parliamentary incompatibilities (Decisions no 71-46 DC of 20 January 1972 and no 77-80/81 DC of 5 July 1977) that prevented relaxation of the rule of incompatibility between the functions and parliamentary mandate and after the 2008 reform worked in the opposite direction (dec. no 2008-572 DC of 8 January 2009).
 Unlike that he can exercise should it prove impossible to elect a majority chancellor by the procedure of article 63(4) BL, presidential dissolution under article 68 is not dispensed from being countersigned (by the Chancellor or a minister) by the terms of article 58 BL, a contrario.
 ‘And so constitutional law becomes true law whereas it was long wondered whether it was a form of law like others, sanctioned by a court’, wrote Jean Rivéro (Rapport de synthèse, in Cours constitutionnelles et droits fondamentaux, Economica, 1982, p 528-529).
 Or even more lightly, this expression of the State Council, dismissing an action against a recommendation of the broadcasting authority (Conseil supérieur de l’audiovisuel) deciding not to count the utterances of the President of the Republic in the referendum campaign: ‘Whereas, because of the position, pursuant to republican tradition, of the head of state in the constitutional organisation of public authorities, the President of the Republic does not speak in the name of a party or political group (…)’ (C.E., 13 May 2005, Hoffer).
 ‘Members of the German Bundestag shall be elected in general, direct, free, equal and secret elections. They shall be representatives of the whole people, not bound by orders or instructions, and responsible only to their conscience.’
 La séparation des pouvoirs dans l’histoire constitutionnelle française, L.G.D.J., 1973.
 BVerfGE 10, 4 of 14 July 1959.
 The question could also be asked for the ‘reserves of interpretation’ and the German equivalent, the technique of consistent interpretation (verfassungskonforme Auslegung).
 Cons. const., decision 62-20 DC of 6 November 1962, 2nd recital (emphasis added).
 P. Avril and J. Gicquel in their book Le Conseil constitutionnel, Montchrestien, « Clefs », 6th edn. 2011, p. 113. Further on (p. 149), they affirm that the Constitutional Council is ‘the court-system regulating the exercise of power’.
 ‘Le Conseil constitutionnel, régulateur de l’activité normative des pouvoirs publics’, RDP, 1967, p. 5-120.
 His main argument was that ‘the Constitution tasked the Council with regulating the legislative function of Parliament and not its political function of supervision of the government’ (p. 15). But this distinction is untenable when being called on automatically to monitor the rules of the assemblies, the Constitutional Council must also examine parliamentary acts and procedures falling under the monitoring function (title III of the National Assembly rules).
 Economica, 1988.
 The expression had been notoriously used by Michel Troper (in conclusion to an analysis that was significantly different from that adopted here) in his foundational article ‘Le problème de l’interprétation et la théorie de la supralégalité constitutionnelle’, Mélanges Eisenmann, 1974, taken up again in his collected papers Pour une théorie juridique de l’Etat, PUF, ‘Léviathan’, 1994, pp. 293-315 (306).
 A Dissertation upon Parties (1733-1734), in Works of Bolingbroke, edn 1809, t. 3 p. 157.