History of the Concept
At the latest since the work of Carl Schmitt we interpret this history in a movement from empirical concepts of political regime structure, to normative legal concepts. While it is important to avoid anachronism, the concept of politeia and its types in Plato and Aristotle represents the precise origin of one concept of the constitution that survives to our day, the empirical one. All states in this sense have a constitution, though the Greeks (inconsistently) included under politeia only the forms of the polis, excluding “despotic” empires. That omission was definitely corrected by Bodin and Montesquieu. By the 17th century, in both popular and inter-elite struggles in England and France a normative not yet legal concept was introduced: the so-called “historical constitution” used as a critical standard by which existing political orders could be measured, and to which revolutions understood as parts of cyclical patterns aimed to return. Here the role of constitution was analogous to and sometimes combined with the idea of ius naturalis, or natural law that was even earlier considered a limit on absolutist forms of the monarchy. The historical constitution was a defensive reaction to state building, as represented by the concept of sovereignty that the new concept sought to tame or “civilize”. It was in the American and French revolutions, in the context of actual constitution making efforts that two new ideas emerged: constitution as a process of constituting an order, and constitution as a legally codified and potentially enforceable document that established the powers of government (division and separation of powers) and their limits (fundamental rights). This very move indicated the survival of empirical and normative-philosophical concepts. The powers of government to be established were derived from earlier supposedly empirical descriptions such as Montesquieu’s. The fundamental rights were remnants from both natural law and the historical constitution. The link between the two domains was the separation of powers. In a new normative, trans legal version as stated by the 1789 Declaration of the Rights of Man and Citizen, a country without fundamental rights and the separation of powers did not have a constitution. The same claim was however sometimes made in a narrower version: only a country with a written constitution had one. Finally in an intermediate version, a constitution could be distinguished from a mere treaty only if it had origins in popular constitution making.
Dimensions of the Concept
The history is of course a clue to the contemporary significance of constitution, in which past dimensions are sedimented. The use of constitution for empirical structure of regimes has become somewhat obsolete, and, according to the professional separation between legal theory and political science, it is supposed to be the latter that deals with empirical structure and regularities, or how a legal-normative framework actually works. Legal science or jurisprudence thus is left with constitution in the legal sense, even if empirical outcomes cannot be and are not disregarded even by constitutional court judges. Empirical analysis also does not avoid at least implicit normative judgments among which the contrast legal-illegal often plays an important role.
Once the concept takes a legal turn, the empirical-normative contrast is transformed into that between the material and the formal constitution, best discussed by Kelsen. Constitution in the material sense is now a kind of meta-law regulating the process of law making (and more broadly: making, judging and enforcing of law). The distinction was later developed by Hart’s contrast of secondary and primary rules, no longer based in the authority of an imputed basic norm. In Kelsen’s version constitution in the formal sense is “the solemn document” called the constitution. In order to fully exist, not only the written character, but its entrenchment against ordinary rule making by amendment rules, and constitutional enforcement are also stressed. Not all the material constitution needs to be or even can be in the formal version, e.g. in the United States judicial review is not part of the document or its amendments. All states have constitutions in the material sense; but not all do or need to have formal constitutions. Nevertheless, the formal’s very purpose is to guard and protect the material against easy alteration, and by constitutional review powers or suitable alternatives. This last step raises the question of what exactly in a material constitution needs to be protected, if constitutional development should not exclude both formal and informal alteration.
The most obvious answer lies in the rule of law. The formal constitution accordingly is the pinnacle of the rule of law, bringing the rulers themselves under rules. The formal constitution serves above all this purpose by being openly codified, entrenched and judiciable. But the formal constitution empirically contains much that is incidental, often quite ordinary legislation put in to utilize the entrenchment provisions for the defense of non-constitutional political purpose. To deal with this issue the basic structure doctrine was introduced by Indian judges. Here the idea is that some parts of the written constitution are materially so important that they should be protected even against the amendment rule of the constitution. However important it is notoriously difficult to distinguish between elements of the basic structure and other features of the material constitution, as visible in the earliest attempt in this direction: Schmitt’s distinction between the constitution and mere constitutional laws. He tried to base the distinction on the mythological agency of the original popular pouvoir constituant. It is however more suitable to defend the important conception of the basic structure if not by eternity clauses as in the German Federal Republic and Turkey then by codifying multi track versions of constitutions, in particular by the makers of the current South African constitution.
A converging approach focuses on the normative constitution in the 18th century sense. Even the basic structure could be interpreted as the empirical regime. But not all regimes ought to be defended. Focusing on classical democratic revolutionary desiderata, only constitutions with fundamental rights and separation of powers are worthy of defense, and within constitutions it is these features that constitute the basic structure. This was already the trend of Indian constitutional review, as well as South Africa’s codification. There is however a tendency to further insist on the link of the basic structure to a democratic at times revolutionary constituent power. This idea was introduced by Schmitt in a political theological version, but has played a role in Turkish as well as Indian jurisprudence. The idea is already in the Federalist Papers, namely that a polity can only have a constitution if foundational documents are enacted by “the people” themselves, rather than governments as in the case of treaties. A more sophisticated if lawyerly version of this idea has been recently applied to the European Union. Accordingly unless a popular collective constituent power based on all the citizens can be imputed or ascribed to an order it does not have a constitution. The argument has been used to disqualify the European treaties as formal constitutions (Grimm), and equally implausibly to exactly so qualify them. (Habermas)
Constituent Power and Constitution Making
The strong Schmittian definition of the essence of the constitution as the decision of the constituent power of a unified entity, people or monarch, has left interpreters with largely paradoxical options. They could follow Schmitt where he never went and insist on a genuinely democratic form of constitution making as the only legitimate form of the constituent power. This is the option of Andreas Kalyvas and Joel Colon-Rios. The cost is either myth making, or overburdening the concept of constitution, since empirical constitutions rarely if ever live up to this standard. Or the whole idea of democratic constituent power can be abandoned, with constitutional legitimacy being derived from contents alone. This is the option of David Dyzenhaus. An in between solution is to rely on the concept of imputation whether on Kelsenian (Lindahl) or Rawlsian (Habermas) grounds. In both of these cases the actual problems of constitution making would be disregarded, along with the legitimation needs and options of such a process. Yet as both American and British traditions assume, it is implausible to legitimately entrench constitutions against the legislative will, and thus empower unelected courts, unless the constitution itself has been made according to especially high standards of legitimacy in a special process or processes.
How a non-mythological notion of the constituent power with increased legitimacy can be conceived we learn from historical processes of constitution making neglected by imputation theorists. Constitutional history since the American and French Revolution has been dominated by four forms:
1. Governmental constitution making that includes both imposition by executives and legislation of parliaments
2. Constitutional conventions as finalized in form by the U.S. Federal Convention of 1787
3. Constituent Assemblies with the plenitude of power as in France in 1789 and 1793-5
4. Round Table negotiations followed by elected assemblies from Poland in 1989 and after, to South Africa in the 1990s.
I can only sum of the results of my work elsewhere. Governmental constitution making or constitutional reform is legitimate only under strict assumptions of the continuity of a basic structure, or when they are merely responses to a variety of popular pressures. Conventions that merely recommend are linked to multi stage processes that can address the needs of legitimation, but they run the risk of dual power unless stabilized by federal arrangements. One of their advantages is the separation of the power of established institutions that are mobilized, and the authority of new, democratic bodies used in ratification. Constituent assemblies assume that the identity of the power holder is such that it produces authority through the normative power of that fact alone. But the identification of the people as that subject is mythological, especially since a coherent meaning of the people merges only through the constitution itself. This model is often only a cover for what Schmitt already called sovereign dictatorship, of the assembly or a provisional government. Finally, the round table is the most synthetic paradigm and assumes a multi stage process like the American convention. In its developed form it entails the making of two constitutions, one interim, the other “final”. The interim constitution brings the whole process, and not only the result under the heading of constitutionalism. This is how it was possible in South Africa to include the new Constitutional Court in its second stage. Finally, the round table model strongly differentiates authority and power and is at least as efficient producer of legitimacy as the Convention, without however the danger of dual power. Its second stage moreover is dominated by a freely elected assembly, that unlike the American ratifying conventions is within some limits free to make the final constitution.
The forms of constitution making are generally path determined. Only intact regimes will use reformist governmental forms. Revolutions involving legal rupture are likely to have sovereign constituent assemblies. What has been called revolutionary reform for the American case has an elective affinity to Conventions that need to be co-ordinated with intact institutions to have a stable result short of dual power. Finally, the round table assumes a political conflict between regime and opposition where change is nevertheless necessary, yet one side is not capable of reform while the other is unable to carry out a revolution. Given the legitimation advantages of two of these forms, the question today is often whether convention or round table can be superimposed on revolutionary or reformist processes for the sake of legitimacy.
Constitutional Interpretation and Judging
Ever since the democratic revolutions the question of constitutional enforcement has been linked to alteration and development. In revolutionary Pennsylvania these linked desiderata produced the censors, while in France there were plans of Condorcet and Sieyes to develop other forms of supervision and enforcement, e.g. the famous constitutional jury of Sieyes. With the exception of the judicial review already known in colonial situations and developed in America most famously by Hamilton (Federalist 78) and John Marshall (Marbury v. Madison) no such option was established till Kelsen’s successful efforts in Austria after World War I. After the Second World War constitutional review became a more general phenomenon under liberal democracy, in all regions of the world, in quite spectacular fashion in India and South Africa.
Giving the power of review to a court or an independent tribunal however raised the problem of the “counter-majoritarian difficulty” from the very beginning. This became an even greater problem when supreme and constitutional courts took on the problem of interpreting and indeed developing fundamental rights as in the substantive due process jurisprudence of the United States. Today much more tame efforts often come under the strictures of judicial activism, gouvernement des juges, “juristocracy” or “kritocracy”. Yet no plausible model of constitutionalism has emerged without significant role for judges, including countries without (UK) or with very partial (Israel) formal constitutions.
Thus the doctrines judges and jurist choose to explain and legitimate judicial or constitutional review is very important. I can only focus on the oldest and longest practice here, that of the U.S. Supreme Court. The doctrine of original intent would be the most consistent with a formal separation of powers where legislators legislate and judges judge. Unfortunately, it is hermeneutic nonsense, since no text is self interpreting, and genuine interpretation is based on the fusion of horizons. (Gadamer). That concept is also forgotten by advocates of the living constitution, that put the focus on contemporary meanings and needs. Often supported by an exaggerated realist (or: critical legal studies) understanding of the complete indeterminacy of the text, the approach can easily give rise to voluntarism and the confusion of law and politics. Thus it creates an opening for the opponents of active judicial review, generally but not only on the conservative side. Finally, focusing on the text and structure of the constitution, the most contemporary approach, allows taking the text, judicial precedents and new meanings equally seriously. When practiced under the supervision of a professional public of jurists, as well as a politically aware citizenry, this approach has the best chance to overcome the counter majoritarian difficulty. Operating in a framework where formal constitutional change, preferably on multiple tracks, is more possible than in the United States would further remove the burden from judges who would be forced to operate “in the shadow” of a flexible, democratic amendment rule with respect to many if not all constitutional contents. Such a combination of amendment rule and constitutional jurisprudence, would also have the advantage of helping to avoid the informal track of change that often leads to abuses like the packing of courts and the alteration of their jurisdiction as recently in Turkey and Hungary.
Elkins, Ginsburg, Melton The Endurance of National Constitutions (Cambridge, 2009)
 Hannah Arendt On Revolution (New York, 1965)
On the fictional status of popular sovereignty see E. Morgan Inventing the People. (New York, 1988)On its political theological status see Lefort “The Permanence of the Theologico-Political?” in Democracy and Political Theory (Minneapolis,1988)
Verfassungslehre (Berlin, 1928). In this tradition see the essays by Dieter Grimm in Die Zukunft der Verfassung (Frankfurt, 1991) and Wolfgang Böckenförde Staat, Verfassung, Demokratie (Frankfurt, 1992)
Most recently and interestingly: Foucault Society Must be Defended. ( New York, 2003) As always: E. Morgan cited above
T. Paine Common Sense (New York, 1986)
The Federalist Papers (Middletown, CT, 1961) that in other sections uses “direct effect” and “supremacy” to make the same distinction
General Theory of State and Law (Cambridge, Mass. 1945)
The Concept of Law (Oxford, 1961)
See Jellinek Verfassungsänderung and Verfassungswandlung (Berlin1906), recently stressed by Julian Arato for international law: “Constitutionality and Constitutionalism Beyond the State: Two Perspectives on the Material Constitution of the United Nations” International Journal of Constitutional Law (2012).
Arato, “Multi-Track Constitutionalism Beyond Carl Schmitt” Constellations (2012)Volume 18, Issue 3, pages 324–351
D. Grimm “Integration by Constitution“ in International Journal of Constitutional Law (2005) v.3 numbers 2 and 3CON; Habermas The Crisis of the European Union: Cambridge, UK: Polity, 2012. The distinction between treaty and constitution remains important, though these should not be seen as exclusive alternative as the term “constitutional treaty” shows. According to the second major argument of Grimm, more plausible than the first, the main forms of state power cannot be outside the constitution. Yet as the American practice of presidential prerogative shows (see especially Curtiss Wright v. United States, 1936), and thinkers as different as John Locke and Carl Schmitt always insisted, executive power in war and emergency cannot be subsumed under rules. I disagree with the “cannot”, but such a subsumption presupposes the transformation of international law, that Habermas believes has occurred already, but that is still a future project.
See his Weak Constitutionalism (London, 2012)
“Constituent Power and Reflexive Identity” in Loughlin and Walker eds. The Paradox of Constitutionalism (Oxford, 2007)
This is the meaning of the two track structure insisted on in America from Hamilton and Marshall, to Hayek and Ackerman. See especially the latter’s We the People (Cambridge, Mass. 1991 and 1998) I and II. The point is all the more important for a multi track structure with or without absolute entrenchment or eternity clauses. See again my “Multi-track Constitutionalism”
“Conventions, Constituent Assemblies and Round Tables” in Global Constitutionalism v1 # 1 (2012)
Arendt has stressed the importance of this differentiation between power and authority, while Schmitt who knew it often disregarded it.
See Arato and Tombus “Learning from Success, Learning From Failure: South Africa, Hungary, Turkey and Egypt” forthcoming in Philosophy and Social Criticism
Dicey The Law of the Constitution (Indianapolis, 1985)
See Hirschl TOWARDS JURISTOCRACY: THE ORIGINS AND CONSEQUENCES OF THE NEW CONSTITUTIONALISM, by Ran Hirschl. Cambridge, Mass, 2004); Ozbudun and Genckaya Democratization and the Politics of Constitution Making in Turkey (Budapest, 2009)
Professor Arato is the Dorothy Hart Hirshon Professor in Political and Social Theory, has taught at Ecole des hautes etudes, and Sciences Po in Paris, and the Central European University in Budapest, had a Fulbright teaching grant to Montevideo in 1991, and was Distinguished Fulbright Professor at the Goethe University in Frankfurt/M,Germany. Professor Arato has served as a consultant for the Hungarian Parliament on constitutional issues: 1996-1997, and as U.S. State Department Democracy Lecturer and Consultant (on Constitutional issues) Nepal 2007. He has been re-appointed by the State Department in the same capacity for Zimbabwe, during November of 2010 where he had discussions with civil society activists and political leaders in charge of the constitution making process. He was invited Professor at the College de France, Spring 2012. Go to his profile on The New School’s website.