Эрик Миллард, профессор университета Западный Париж (Франция), бывший вице-президент Международной ассоциации философии права и социальной философии (IVR), почетный член Университетского института Франции
Введение. Толкование конституции не является научной интерпретацией текста или текстов, известных как конституция. Толкование конституции – это нормотворческий процесс; процесс, который учреждает конституцию как норму. До интерпретации не существует нормы и, соответственно, конституции. Более того, это означает, что толкователи могут создавать конституцию как смысл, как имеющий юридическую силу источник, также как они создают иерархию норм, вытекающих из конституции. В данной научной статье сравниваются три подхода к этой проблематике, а именно формалистический (конституционализм), условно антиформалистический (неоконституционализм) и истинно антиформалистический (реализм). Обсуждается политический вопрос о том, какая теория обеспечивает наибольшие гарантии демократии и прав человека, и анализируется судебный активизм с этой точки зрения. В статье доказывается, что, используя так называемую «разумную» аргументацию основаную на демократической легитимности учрежденных конституцией, может решать не только вопрос о полномочиях конституционных органов власти в рамках конституции, но и о том, что им (в том числе официально избранным демократическим путем) необходимо сделать для принятия новой конституции – установить положение о вневременном существовании действующей конституции, несовместимое с любым представлением о демократии.
Ключевые слова: толкование конституции, смысл правовых норм, нормотворчество, судебное усмотрение, неореализм, неоконституционализм.
1 – It is generally accepted that people who believe in fundamental values, such as democracy, the rule of law or human rights, should also have confidence in a written constitution or in a constitutional court. The first approach rests in classical formalism, which perceives the constitution as a text that binds all authorities which are constituted by it. The second school of thought draws on the idea that the strong values in which we believe are more important than the text of a constitution, and that, using reason, judges, can decide what constitutes a good constitution as our highest norm.
It therefore seems that we are now far removed from the bad nightmare that annoyed Hart for instance [4; 5; 8]. Or, using the scientific method, we can understand what constitutes a constitutional norm by reading the constitution. Or, this knowledge is not important because we can trust judges to use moral reasoning to interpret the constitutional text. Even if we consider the most complex and persuasive theories relating to these questions, such as Kelsenian normativism for formalism  or Atienza’s or Alexy’s versions of what is now known in Latin countries as neo- constitutionalism [1-3; 9; 10], we have to choose between two noble dreams. The nightmare has disappeared.
2 – For many reasons, I do not feel comfortable with these two available choices. Certainly the main concern arises from the defence of values, such as democracy or so-called human rights. The available choices are not practical mechanisms for effectively implementing these values. There is no necessary contradiction between these two conceptions in the ethical position; it depends on metaethics. On the other hand, there is no necessary link between a liberal position in ethics and neo-constitutionalism. In fact, there is likely an existing contradiction. Why do I have to trust a judge, and even if I do, how is this consistent with the idea of democracy? Why do I have to ignore Kelsen’s warning that legal (normative) science can only let us know how a law ought to be, not how it is? These questions continue to be important, and I think that these values, at least for me, are too important to be confined to theory. I am concerned by the defence of these values in practice, not in theory. And in order to defend these values in practice, we need a theory that does not lead us to misunderstand what law is and how it functions.
In a democracy, it is important that rules adopted for and by democracy are effective. But to state that this is important does not inform our understanding as to how this point can be effective. A noble dream, whatever it is, has no utility if you cannot realize it. You can only realize a dream when the environment provides the opportunity for it to take place; our world, material or mental, does not suit these dreams. But, of course, it is possible that, even without being effective, these dreams have some effect on our world, for instance having a bearing on important topics, such as legitimacy or persuasion. But they do not automatically produce the effects they set out, which are to guarantee human rights and/or democracy. Of course, to be analytically consistent we can employ new definitions. For instance, we can decide that the word democracy refers not to the decision of the majority of the voters, but only to what these voters decide according to certain values, and even maybe to the values that they have to decide (and if so is it still necessary to maintain the participation of the voters if there is nothing more to decide?); or that unjust law is not law. But once we have provided new definitions, we rest in peace with our theories, but totally disconnected from the most obvious and numerous states of affairs in our world. Definitions are free, and our concepts are mostly stipulated; but changing a definition or a concept does not erase the existing state of affairs; it just changes the title we use to refer to it.
My concern is not law and constitutional interpretation as they should be, but rather what happens when constitutional interpretation takes place – that is, what effects are produced as a consequence of constitutional interpretation by courts and other legal bodies. I believe that understanding and accepting constitutional interpretation is a major issue for those who really want to support democracy and human rights. But I do not believe it will help to reconstruct it as a noble dream, which we cannot attain. Nor will it help redefine democracy and human rights to the ends of theoretical consistency or stick to an idealist conception of the law.
3 – From this point of view, things appear very differently: constitutional interpretation does not refer to a scientific interpretation of a text or a set of texts called “the constitution”. Constitutional interpretation is a norm-creation process, one which creates the constitution as a norm. Before interpretation, no norm exists; in other words, there is no constitution. And we can never be sure that this norm creation will become an interpretation of a text or set of texts known as the constitution, and we cannot be sure that this norm creation process is or will be reasonable, and will support democracy, human rights, or justice. I do not think we are finished with the nightmare; and, perhaps, this is not a nightmare, but in fact our real life.
This approach, known as legal realism, is not a new one and is, in fact, one of the most known and discussed approaches. Legal realism symbolizes the bad dream for anyone who believes in any form of formalism, or reason in an absolute sense. But legal realism is a form of scepticism. And I suspect that it is scepticism which is not accepted by the so-called neo-constitutionalism: according to neo-constitutionalism, reason can protect us from judges’ discretion, because judges, guided by reason, will know what constitutes a true or correct constitution and a true or correct norm.
The anti-formalism of an approach such as neo-constitutionalism is only apparent. This approach is as formalist as any other approach that pretends that we are able to know what a norm is, except that we know not from the text, but from Reason (something even more abstract than a text). The reconstruction of anti-formalism as a noble dream is only possible through the affirmation of non-positivistic normative knowledge: a modern form of natural law.
4 – On the other hand, legal realism is a very confusing concept. It is usually used in legal theory to indicate that we have a conception of the law as part of reality. Law is not just an idea.
But what does it mean? Does it mean that law really exists? Or that we can call law only what we can observe, that which exists? Or does it mean something else? Of course, a good strategy would be to say that realism is a theory designed by those known as legal realists. But who are they, and do they have a theory, or a specific theory?
For many people, legal realists do have a theory, which simply states that courts do not apply (formal) law, but rather decide according to their individual preferences. And, the reasons for their decisions are undecidable, legal or not. But in that sense, legal realism is just a description. Logically, it does not imply that judges do not decide according to (formal) law. The theory should logically accept that judges can be formal, and can decide in a formalistic way. If they are formal. And maybe are they, if not truly, at least mainly. That is just a contingent issue and this issue depends on mere facts: what are doing courts.
On the other hand, regardless of whether judges ignore the formally conceived law or not, what they do should always be considered the empirically conceived law. I conceive of the law as being based on judges’ rulings: what I call law is what judges make . The role of formalism in a such approach is not to establish whether judges ought to decide according to formal law, but to establish that they can decide the law, according to a formal interpretation of a text, or reasoning, or whatever else. A statement, in its last part, that even Kelsen should accept as far as it refers to what judges really do in practice, not what they should do. But maybe not the part of neo- constitutionalism which supports judicial activism.
Judicial decisions, obviously, appear to always be grounded in law: based on precedents or texts. Realists state that this is only an appearance: strategically or not, it gives the impression that decisions are formally derived from previous legal materials, and not the expression of individual preferences. These decisions are not discretions and they are not creations. But for realists, the facts are different. Formal or not (and because they can be formal or informal), judges always decide with discretion and, thus, create law. And this is the precondition for trusting judges, who are guided by reason, but it is not a guarantee that the outcome will be reasonable, as far as we can determine what is reasonable.
From this point of view, realism leads to two different approaches: American realism and the so-called Scandinavian realism. I do not intend to explore here what they are, but will just stress certain differences in relation to the idea of constitutional interpretation as a norm creation process.
5 – Initially, American legal realism was conceived as a mood, a non-theoretical approach to law. It was conceived as a tool box, which anyone could adopt freely according to their own premises. American legal realism was more of a practical approach than an epistemological one. And that is probably the reason for its success in the US and its extensive influence there and in other countries. It remains a descriptive approach, but it very quickly surpassed description (theory) and appealed to other things, that only pre- deconstruction and description of facts allowed.
This movement to deconstruct the law allowed for various reconstructions, grounded in the idea that law in general, and constitutional norms in particular, serve as the forum for political confrontation, and that interpretation, foremost constitutional interpretation, is a political choice between opposing political views. Furthermore, the constant confrontation with a more formalistic approach derives less from a confrontation between two theories of legal knowledge, and more from the confrontation between two conceptions of legitimacy – between two political doctrines of legitimacy, notably the legitimacy of what comprises a constitution.
The phrase “creating a norm” is, therefore, a vague statement, which suggests that a norm is a product of a judge’s work that has been grounded in various elements, including the political element (and including it maybe as the most important). Formalism supports the idea that a constitution is a political pact or consensus that cannot be changed by interpretation. Realism supports the one that the constitution has to be adapted constantly by interpretation, according to our world evolution.
There is no question about what a norm is, or how we conceptualise norm. Or the text has to be applied, or the court should decide the norm.
6 – The so-called Scandinavian realism, which these days is no longer considered Scandinavian, on the contrary, derives from an epistemological critical school of though. It is not a tool box, but rather a theory that sets out to be consistent. A theory of cognition: how can we assert that we know what law is? The four main theses supported by this theory are as follows:
a) There is a distinction between science and its object; in this case, the distinction is between legal science and law.
b) A scientific process is seen as politically neutral, that is, the law is not identified as it should be but as it is.
c) Science describes its object as an objective reality, that is the law of legal science must be reduced to a reality, something that exists.
d) Reality is a set of facts but a set of mere facts: legal science must be an empirical science, which describes facts and this sometimes leads to the idea of a test of verification.
This Scandinavian theory of realism can easily share with American legal realism the idea that norms, if law is a set of norms, are created by judges. When a judge state that this is Law, we have a social fact (the fact of stating : linguistic entities). And asserting that what judges say is law allows, when it’s interpreted as a prediction of what judges will say, a test to verify that it is true, in a scientific sense, that this the law.
But, on the other hand, this theory must explain a bit further its concept of a norm – why it should be considered consistent to perceive interpretation as a norm creation activity. American legal realism can avoid this treatment for two reasons: it does not intend to be a theory and it does not intend to (only) describe reality, but rather act on it. In other words, it does not describe interpretation, but rather proposes, supports, criticizes or imposes (fair) interpretations. That does not mean that American realism ignores these questions, but providing an answer is not a necessity, as it is in Scandinavian realism.
7 – So when we state that constitutional interpretation is a norm creation process, we need to provide precise definitions for all relevant concepts in that statement: constitution, constitutional interpretation, norm, and norm creation.
a) We have many conceptions of a constitution. Most of them are not legal conceptions. For instance, we use the word constitution to designate a political program, a national pact, a moral document, etc. These conceptions are plausible, and it is a fact that, historically, constitutions fit these conceptualisations (some continue to do so even to this day).
In legal thought, constitution is seen as a norm, as far as we accept the idea that law is made of norms. But not all documents known as constitutions are norms, and not all constitutional norms are texts or something based on texts. On the other hand, all texts known as constitutions do not have a sole normative function. For these reasons, constitutional interpretation is important: we need norms, and we sometimes, but not always, have texts that do not intend or do not solely intend to be normative. Constitutions as texts are not designed (only) by and for jurists.
b) Norms. If we abandon the source of the noblest dream – the weak idea that we do not need interpretation because constitutional texts are clear, endowed with a sole normative function and a sole meaning – we can usually define a norm as a prescriptive meaning. A prescriptive meaning of a text, or just a prescriptive meaning (disconnected formally or substantially from any text). A constitutional norm is therefore the prescriptive meaning of a constitutional value.
Of course not all constitutional meanings are valid, but that is another point, much more complicated: constitution as a norm is a meaning of a certain (constitutional) value. Here, I only look at the concept of a norm, or more precisely concepts of norm. Indeed, we can have two conceptions of a norm as a prescriptive meaning.
First, we can use an ideal conceptualisation, in which a norm is an acceptable meaning according to certain points of view. When using a scientific approach, such as the one adopted by Scandinavian realism, these points of view are strictly connected with the linguistic analysis of texts, rather than with political preferences of the interpreters of the text. But a linguistic analysis of a text rarely provides us with a singular meaning. We can only determine a framework of more or less broad meanings, whereby we can only distinguish between several meanings that can be considered correct and some that are not considered correct. It is well known that this kind of activity is not perceived as norm creation activity, but rather as knowledge activity. On the other hand, this kind of activity does not offer a solution to someone who has to decide a case according to the law: even a formalist judge would have to decide within a framework what is the (valid or effective) norm; and, in turn, the judge could decide the (valid or effective) norm as something that is not correct according framework.
So we also need an expressive conceptualisation of a norm: a meaning that has been in fact adopted by certain authorities under certain circumstances. And for an empirical conception of legal science, this norm is a valid (effective) norm, this norm exists and can be described as law as it is. And only this norm. It doesn’t matter if it is an explicit norm (it is a correct meaning according to linguistic analysis) or an implicit one (it is an incorrect meaning according to linguistic analysis or it is a meaning disconnected from any text according to linguistic analysis).
c) Constitutional interpretation is therefore a process which provides us with (valid or effective) constitutional norms. In other words, constitutional interpretation offers constitutional prescriptive meaning, which jurists refer to. But there are as many conceptions of constitutional interpretation as there are conceptions of norms.
First, we can consider constitutional interpretation as a scientific process designed to reveal the ideal meanings of a constitution as a text, held as valid (constitutional sources). Importantly, we cannot avoid a scientific interpretation of such texts, if any, because there is no clear text that can be understood without scientific (linguistic) methods ; but we cannot establish here one and only one meaning.
Second, we can consider constitutional interpretation as a decisive process: the one that gives or has given constitutional prescriptive meaning to a text, or maybe constitutional prescriptive meaning outside a text.
d) Creation norm activity. For the same reasons as above, we can have two conceptions of a norm creation process – one strong and the other weak.
In the weak sense, we can say that any interpreting body, even a court which sincerely wants to stick with formalism and respects what has been written in a constitutional text, has to decide what meaning is the norm, within a framework of the text’s possible meanings. No judge can avoid making a decision, but if he decides within the framework, the norm creation process is not the outcome of the sole judge’s decision. The norm creation process includes constitutional power, which adopted the text, and the interpreting body which selects one possible meaning for the text from a linguistic point of view. Most constitutions offer vague texts, allowing the interpreter’s choice to be broad (and the vaguer it is, the broader is the choice). Of course, the interpretation of the constitution as having one meaning is a political phenomenon, as it expresses the personal preferences and emotions of the person who interprets the constitution. Even from a formalist or idealist point of view, the interpretation of a constitution as a text in practice is partially political, and all those who interpret it take part in the norm creation process. This kind of process is may be desirable, and it is also possible; but there is no guarantee that the interpreting bodies limit themselves to this process.
So, applying the strong conceptualisation, we can say that the enacting authority only provides a text – one of the sources of law (official, if not real) – but never provides a norm (that is the element of law, not the source). As a consequence, we never know how effective (real) this (official) source will be. Sole the norm decided by an interpret in a case is law, and the only person who has norm creation power is the interpreter. This role is primarily performed by courts, but other bodies, such as the executive administration, or constitutional authorities can also take on that role. If we adopt this point of view, we do not need to know if the norm meaning is chosen within or from outside a given framework, even if such knowledge would help develop typologies to the ends of understanding how much weight is given to the text and how much to its interpreter. But each choice is a pure decision, and all choices are expressions of complete power over the decision of what constitutes a norm, even within the framework. Norm creation is the sole power of the interpreters.
When realists say that constitutional interpretation is a norm creation process, they can refer to either the weak or the strong conception. In any case, if they are truly realists, in that they base their assumptions on facts, they assume that constitutional interpretation is always a norm creation process, just because there is no constitutional norm before interpretation. Constitutional interpretation creates all implicit norms, and all explicit norms, even if in the latter case man can say that this creation is not the result of the sole interpretation, and in the first that this is not interpretation, but pure autonomous creation.
8 – It is true that we can find facts that allow such a conception. But if we want to go a bit further, we should pay a little more attention to something else. What are we saying exactly when we say that constitutional interpretation creates a constitution? Obviously, we imply that this process creates a constitutional norm. But, in what sense?
We can distinguish at least four meanings:
a) First, constitutional interpretation establishes which prescriptive meaning counts for constitution. This is the most common meaning and I will not expand further on it.
b) Second, sometimes a text, which is or is not called a constitution, is an official, or legitimate, source of law. In this instance, constitutional interpretation creates not only the norm, but also the source of the norm. Or, if we prefer, constitutional interpretation establishes which text is to be considered the source of the constitutional norm. This is the condition for an explicit norm.
c) Third, there is a prescriptive meaning that counts as constitution without any text as its source. This presents the possibility of an implicit norm.
d) And fourth, some texts that are considered as sources of law, but are not constitutional law, are now considered as official, or legitimate, sources of constitutional law. This is known as the creation of normative hierarchy.
Let me offer some examples to illustrate the latter three meanings.
In the US, the admission of judicial review in Marbury vs Madison is certainly the most obvious example of the second meaning – assigning the text the status of a source of law. Chief Justice Marshall was the first to establish the US constitution as a source of law, providing an interpretation of a text known as the “constitution”. This was certainly not the most obvious possible way of understanding the text, which transformed a political pact into a legal source. And it was the precondition for the decision of the prescriptive meaning of that legal source: the constitution as a norm.
In Israel, Justice Barak did the same, but in a context without any constitutional text. This is the third meaning of the norm creation process. In 1995, in the case of the United Misrahi Bank, an Israeli court limited Knesset’s power to adopt a statute, while there was no enacted text counting as a constitutional source, which could be used as a starting point for imposing power limitations on parliament. In other words, the constitution had to be established as a norm in the absence of a constitution as a source.
And, finally, in France, we can find illustrations of the fourth meaning: the changing of the hierarchical status of an official source of law or the creation of the hierarchy of sources. In 1971, the Conseil Constitutionnel decided that some fundamental principles included in republican legislative statutes were part of the constitution and could not be modified by another legislative statute. In this instance, a French court interpreted the preamble of the constitution, referring to the “Principes fondamentaux reconnus par les Lois de la Republique” firstly as a normative statement (exactly as Justice Marshall did in the first example), and secondly as a statement allowing the court to decide using judicial review what counts as a legislative provision and what counts as a constitutional source in statutes previously adopted by parliament. Of course, the same court will have to decide what, in these statutes, is a principle of constitutional value, and what is the meaning of this principle.
As we can see, interpreters, who in this instance mainly comprise constitutional courts, create constitutions in all possible senses of the word as is prescribed in legal philosophy. Interpreters can establish the sources of constitutional law; they can decide the hierarchy of law sources; they can decide the prescriptive meaning of the constitution, regardless of whether it is derived from a written source or not. Constitutional interpretation, which is the name usually given in legal philosophy to that kind of activities of constitutional courts, even if it’s not interpretation strictly, or if it not what courts should do, goes beyond the formalist idea of an a priori knowledge of a text.
9 – If we now come back to neo- constitutionalism and judicial activism, we can see that we need an important dose of optimism in order to have confidence in courts and in their reasonable reasoning to protect human rights or democracy.
It is a contingent problem, not a necessary connection, between court decisions and political quality of norms. When we support judicial activism because a certain court decides in a certain manner, we often forget that a different court could have decided differently, in a way which we potentially would not support politically.
Examples drawn from Latin American experiences illustrate this. In Colombia, for instance, the judicial activism of the constitutional court is narrowly connected to the country’s political situation and the weakness of parliament. This allows the court to act and to appear as having legitimacy to act. Judicial activism in this case implies a living conception of the constitution as a political document. On the other hand, a norm creation process is usually seen as a constraint: it is embodied in decisions with formal argumentation, through a justification of the decision, whatever that justification is.
For strategic or just pragmatic reasons, decisions and justifications are often disconnected from each other in the Colombian experience. For instance, the court can make decisions through press communiques, which become effective immediately (accion de tutela), and then delay for months or years the formal publication of the statement, which outlines the justifications. And it’s not rare that the statement, whilst addressing the political or doctrinal reactions to the decision, is modified in a way that does not always appear to provide logical reasons for the previous and effective decision. At that point, the court does not really create a norm; it decides without any norm in any sense.
We may support the decision politically, as, most of those decisions really support values that we call human rights. But a different court (with different members, who have different preferences) could decide differently, and we would have no useful constraints of argumentation limiting their power.
With the statement called juizio de substitucion, the court shows the real nature of judicial activism. The court decided whether there is a difference between amending the constitution and adopting a new one. Therefore, the amendment of the constitution could only be done through the process established by the constitution and it falls under the powers of this court to control that process, according to its ^le of protection given by the constitution: a procedural constitutionally grounded control. But the same also decided to control the process to enact a new constitution: to perform a political, but not constitutionally grounded control.
Logically, if a court exists according to a certain constitution and controls all attempts to amend this constitution, the process of enacting a new constitution, by definition, is disconnected from the previous constitution. In other words, the distinction established by the court between amending and enacting a new constitution implies that enacting a new constitution does not fall under its competence. The first argument used by the court to deny that difference was to refuse to interpret the reform as an amendment : the enacting process was not valid in the first instance according to the enacted constitution because it amended a crucial part of that enacted constitution, and thus cannot be seen as an amendment, but rather as a process to change the whole constitution. The second argument used by the court was that the enacting process, when interpreting the same reform as an enactment of a new constitution, was not democratic enough to be perceived as a legitimate means to change the whole constitution (to enact a new constitution). These two arguments may seduce from a political point of view (even if in the case of Colombia, courts changed their political position about the issue: changing the process and possibility of re-electing the President). But from a logical point of view, it shows only that a court created by a constitution can estimate, through a so-called reasonable reasoning grounded in democratic legitimacy, not only what constitutional powers can do within the constitution, but also what powers are required to enact a new constitution (outside any constitution). That is to create a clause of eternity of the valid enacted constitution: an implicit constitutional norm created by the sole court, that includes a sole possibility of derogation under the conditions of its appreciation (its authorization) by the Court, according to its appreciation of legitimacy.
I am not here to complain about or support this point. I just want to demonstrate to what judicial activism can be applied, as well as show what it means and implies.
10 – Neo-constitutionalism is not a new version of constitutionalism; it is something entirely different, as it does not perceive the constitution as a political or legal text that is able to guarantee liberalism. Neo-constitutionalism places its trust in constitutional courts.
It is possible that neither constitutionalism nor neo-constitutionalism can provide strong, definitive and efficacious guarantees. Constitutions can be changed or not respected, as accepted by constitutionalism, while courts can decide anything for any reason. Stating this, of course, does not provide us with practical solutions. But to deny it is no more efficacious.
At least, a sceptical position addresses the fact that we have a problem, a position not taken by the neo-constitutionalist school of thought. Noble dreams are good for those who sleep. Nightmares have at least one virtue: they awake us and open our eyes. It is very possible that the defence of values, such as democracy and human rights need, much more than law, or whatever we call law.
1. Alexy R. The Argument from Injustice: A Reply to Legal Positivism. Oxford, Oxford University Press, 2003.
2. Alexy R. Theorie der Grundrechte. Suhrkamp Verlag, 1986.
3. Atienza M. El sentido del Derecho. Ariel, 2001.
4. Hart H.L.A. American Jurisprudence through English Eyes: the Nightmare and the Noble Dream, 11, Georgia Law Review, 1977, no. 5.
5. Hart H.L.A. The Concept of Law. 2nd ed. Oxford, Clarendon Press, 1994.
6. Holmes O.W. The Path of the Law. Harvard Law Review, 1897, vol. X, pp. 457-478.
7. Kelsen H. Pure Theory of Law. 2nd ed. University of California Press, 1967.
8. Lacey N. A Life of H.L.A. Hart: The Nightmare and the Noble Dream. Oxford, Oxford University Press, 2006.
9. Pozzolo S., ed. Neoconstitucionalismo, Derechosy derechos. Lima, Palestra, 2011, pp. 13-14.
10. Pozzolo S. Neoconstitucionalismo y especificidad de la interpretaciyn constitutional. Doxa, 1998, vol. II, no. 21, pp. 339-353.
Вестник ВолГУ. Правовая парадигма. 2017. Т. 16. № 4