1.º Introduction

In his paper, Maimom Schwarzschild addresses the “political question doctrine” (“PQD”), according to which there are some questions—due to its political nature—that cannot be adjudicated by courts.² According to him, even if courts sometimes invoke this doctrine to avoid adjudication or to adjudicate in favour of whatever the elected government has done, on the one side, the doctrine imposes little real restraint on the courts’ power, and, on the other side, the questions covered by the doctrine are not necessarily the most important for the American life. This seems to allow American courts to assume an increasingly political role, deciding social controversies that should be left to legislatures. He explains the PQD as the American courts apply it, showing how the doctrine coexists with growing judicial activism.³

From his jurisprudential analysis of the PQD and respective cases’ description,⁴ a metatheoretical question immediately arises to my mind: is there truly a PQD or are judges and scholars just talking about already known phenomena? In my commentary I will try to briefly address this question and I will do it from an analytic jurisprudence perspective.

2.º In search of the lost political question doctrine

The analysis of the PQD requires to start by analyzing the concept of “political question”. And this entails to surpass a first difficulty: what do we mean by “political”? In a first approximation, in this context, we can say that the term “politics” refers to the set of activities associated with the governance of a country or a region, and it involves making decisions that apply to members of a group.⁵ Therefore, a political question would be related to any of those governance activities and/or to any decisions took by political bodies, such as the legislator or the government.

This broad definition immediately shows a problem that affects the PQD: how can one distinguish a “political question” from a merely constitutional or legal question? Which property allows us to distinguish between political and legal questions? Such a distinction presupposes formal or material criteria. Let us look at some candidates.

Regarding formal criteria, a strong candidate would result from the cases in which constitutions expressly define which questions are simply political. This criterion allows us to distinguish, with certainty, which are the political questions and, therefore, in which cases courts are not competent to decide them. However, the existence of such constitutional norms is contingent, which means that there can be no universal “PQD”⁶. In addition, the problem is that, on the one hand, in most cases constitutions do not establish a clear differentiation between political and constitutional questions, and, on the other hand, even if it expressly tried to, there would always be cases of penumbra about the qualification of questions as political.⁷ For example, when a State decides, for reasons of internal or external security, to approve measures allowing access to citizens’ metadata, which clearly restricts the fundamental right to privacy, is it a mere political issue and therefore immune to judicial review or is it constitutional and therefore amenable to be judicially reviewed? ⁸

Regarding the material criteria, I can immediately think of a possible candidate: the politicity level of the question.⁹ If a certain normative question has a minimum level of politicity, then we can qualify it as “political”. However, on the one side, it seems obvious that we still need to define politicity and, on the other side, it does not seem to allow us to draw a firm line between the political and non-political questions. A second material criterion may focus not on the question itself but on the act produced to address the question, and therefore may be the distinction between deontic actions (institutional actions that change the legal system, such as norms, administrative acts or judicial decisions) and non-deontic actions (such as the drafting of a legislative impact report or a speech by the Prime Minister)¹⁰. However, it does not seem to be a good criterion, since the creation of norms is usually pointed as a token of a political act.

Therefore, it seems to me that even if one could resort to other criteria, the truth is that it is highly doubtful that there is any universal criterion that allows us to distinguish between political and non-political questions, and one of the reasons lie in the fact that all constitutional questions, in a broad sense, may be considered to some extent political.

2.2. Separation of powers and the distinction between creation and application of law

In addition to the obvious difficulties in distinguishing between political and non-political issues, the doctrine seems to presuppose a clear-cut distinction between creation and application of law. The idea that in modern constitutional systems there are some questions to be resolved and decisions to be made by the political branches and not by the courts seems obvious from a separation of powers perspective.¹¹ Nevertheless, the proposition that there are non-judiciable political questions naturally cannot ignore the existence of norms of competence giving jurisdiction to courts to review the constitutionality of the political branches’ acts, whether or not they have a political content. Moreover, such a perspective cannot result in a constitutionally unauthorized bottleneck of issues that may or may not be judicially controlled, in accordance with the subjective criteria of each political theorist.¹²

It is the principle of the separation of powers—as an imposition of dispersion of powers by various bodies—that must solve the legal problems raised by the allocation of powers in each legal system.¹³ Courts have competence to exercise judicial power, deciding cases and controversies arising under the law of each legal system precisely created by the political branches. Therefore, if courts are asked to make “law” or to extend it beyond the linguistic limits of the law’s statements, they must deny the claim on the basis that it is a legislative competence and may argue that is a “political question”. As courts have said, they can only assess whether the political branches have exceeded constitutional limitations; as long as they act within their constitutional competence, “whether they have done wisely or well is a «political question» which is not for the courts to consider.”¹⁴

The problem is that, in addition to the interpretative problems raised, due to its structure this principle tends to conflict with other norms, which leads to difficulties in its application to some legal cases—the hard ones.¹⁵ This create doubts, for example, as whether a court can review certain choices made by the legislature, under its jurisdiction conferred by the Constitution.

4. The supra mentioned concept of “politicity” relates to the autonomy to evaluate political conditions or the merit of the activity of other branches, within a framework of prognosis about the respective conditions for the pursuit of constitutional purposes.¹⁸ Usually, legal systems let the exercise of the political branch in its narrow sense bound only to political accountability. But again, this is a contingent matter.

The degree of politicity is linked, on the one hand, to the extent of the discretion conferred on the choice of means and ends in satisfying the collectivity necessities, as well as on the freedom granted to democratically legitimized legislative branch to interpret constitutional provision regarding the ends of the state. Therefore, for example, political acts such as the appointment of members of government or the impeachment of the President are endowed with a high level of politicity. Legislative acts also are, in greater or lesser way, political acts, but have a less degree of politicity. However, contrary to the latter, they are judicially reviewable even in a substantive way, namely by the violation of norms enshrining fundamental rights. Differently, the politicity of judicial decisions is almost null. There are exceptions: the constitutional courts’ powers to declare the unconstitutionality with general binding force of norms created by legislators. In this case, as a negative legislator, the Court’s rulings have a considerable level of politicity, which is visible within the margin of free interpretation of the Constitution, as well as the competence to authoritatively decide the related disputes.¹⁹

This means that judicial decisions can also involve more or less politicity. And, from the normative point of view, this is not a major problem if it was the legal system itself to allow it.

5. Let us return to the principle of separation of powers. The traditional idea of this principle (since the French Revolution) consisted in the qualitative distinction between legislative power as the creator of law and the judiciary as an enforcer of law—such an idea presupposes a strong distinction between creation and application of law, and that the law contains a single correct answer to each legal question and that legal systems must be complete and consistent.²⁰

When we talk about the “political conduction of a country”, it is clear that this function comprehends the creation of norms. The problem is that all functions presuppose the creation of law (which is the result of any deontic action). Contrary to what one might think, on the one hand, the creation of law also means application of law, insofar as the legislator when creates norms is doing so by applying higher norms—namely and at least those attributing its competence;²¹ on the other hand, regarding the legalist-formalist positivism is dead for a century, as Kelsen or some realists point out (albeit exaggeratedly) the application of a norm also implies the creation of law regarding the respective external justification: either because of the necessity to interpret the normative provisions created by the legislator, or by the need to choose the definite norm to apply to the case, when several are in conflict, operations which are to some extent constitutive.²² In addition to the necessity to interpret and to solve normative conflicts, not infrequently, the resolution of specific cases also presupposes the attribution of discretion to the judiciary. And the cases of constitutional norms’ conflicts, in which is necessary to balance, involve a huge amount of discretion, as we will see.

In conclusion, as Kelsen warned, the distinction between creation and application of law is therefore much weaker than initially assumed, and it is clear that there is a confluence of these activities both in the legislative and in the judicial function.²³

6. However, what has just been described regarding the separation of powers indicates what Henkin had already guessed: “[o]ne needs no special doctrine to describe the ordinary respect of the courts for the political domain. If a political question is one which the Constitution commits to the political branches, our political life is full of them. The courts may sometimes have occasion to decide whether a question was in fact constitutionally committed to the political branches, but that, too, needs no special doctrine suggesting a quality of “nonjusticiability” with connotations that the courts must dismiss for lack of jurisdiction and authority without reaching the merits.”²⁴

And this is explained because the issues in question may give rise to more or less hard cases, but their solution always occurs within the legal system. Thus, analytically, it would only make sense to isolate a PQD if and only if there were cases “in which prima facie and by usual criteria would seem to be for the courts, will not be decided by them but, extra-ordinarily, left for political decision.”²⁵

2.3. Is there a political question doctrine? Three possibilities

In a different fashion, Seidman thinks there may be some specific cases that justify the doctrine. He distinguishes three types of cases: (i) cases where the legislative or executive power possesses a constitutional or legal discretion; (ii) cases where the Constitution vests in the political branches final interpretive authority as to the meaning of some constitutional provisions; and (iii) cases where even if there is an right or determined answer to a constitutional question, courts must (politically) decide whether they should abide by that answer.²⁷

Let's look at each of these cases.

8. The first type of cases—the “discretion cases”—are characterized by the fact that a court may conclude that there is no violation of the Constitution because the legislator or government acted based on its constitutional competence, which means that the case is decided on the merits and there is no need to resort to a preliminary, political question analysis.

The concept of discretion is a normative one, and therefore discretion is conferred by norms or by the respective dispositions, and it translates in autonomy of the competent authority to choose between alternatives of action in the creation of norms or of decisions. However, discretion is also a phenomenon governed by norms which affect the prima facie autonomy conferred, reducing it, either by imposing to choose a certain alternative or by reducing or removing some of the possible alternatives. Therefore, there are norms conferring discretion at a preliminary stage (“prima facie discretion”). But discretion may still be reduced by other norms applicable to the situation in question; only after the examination of all the converging norms on discretion is it possible determine the real amount of autonomy (“all things considered discretion”)²⁸.

In cases of discretion, the use of PQD is no more than a purely rhetorical-argumentative but legally innocuous exercise, especially if we have in mind that there are legal resources available to respond to it—this is a discretionary situation. Thus, the use of the doctrine in these cases does not stand up to Ockam's razor.

9. The second kind of cases—the “interpretive authority cases”—are characterized by the fact that the final interpretive authority would be in a branch of government rather than the judiciary. More specifically, a court might conclude that the Constitution entitled the plaintiff to relief, but nonetheless consider that the political branches should have the final “word” when there is disagreement about the meaning to ascribe to constitutional provisions.²⁹ According to Henkin, even if this was a theoretical possibility, an accurate analysis of Supreme Court jurisprudence seemed to point to the inexistence of this “doctrine”.³⁰ But Seidman argued that Henkin was looking to the wrong side—he was focusing too much on the expression “political question”, while the Court was implicitly using the interpretative version of the PQD, as it did in the Katzenbach v. McClung case. Moreover, despite not having used nowhere the expression “political question”, “[t]he Court's test makes sense only if one supposes that when there is disagreement about the substance-about what is or is not necessary to protect commerce-Congress has final interpretive authority so long as its judgment is «rational».”³¹

In support of this thesis, authors such as John Hart Ely and Jesse Choper precisely emphasized the ubiquity of the interpretive authority-like political questions.³² These scholars suggested that much of ordinary constitutional jurisprudence could best be understood by viewing them through the lens of political question.^33/34 If we focus on Ely’s opinion, and assuming Seidman’s view on it is correct, one can say that he argued that, even if courts have final authority concerning constitutional interpretation, they should interpret the Constitution as granting substantive discretion to the political branches in the absence of a process defect.³⁵ And in principle other interpretive authority positions would end up standing for the same thing. But if this is correct, just like Henkin sensed, we are dealing again with cases of discretion.

Let me say some things about this second version of the PQD. First of all, these authors are talking about interpretation in an ambiguous way: sometimes they use the term interpretation to denote the activity of assigning meaning to a constitutional provision, other times the term is clearly used for situations concerning the resolution of normative conflicts.³⁶ Second, when one says that the Court's test makes sense only if one supposes that when there is substantive disagreement and that Congress has final interpretive authority so long as its judgment is “rational”, there is at stake a prima facie discretion case; if the judgement is irrational, the court has a legal parameter—such as the “European” proportionality test—to assess the judgement. If we are invariably returning to the phenomenon of discretion, perhaps the problem simply lies in the misunderstanding of this legal concept.

It is important to stress that interpreters are all norm’s addresses.³⁷ The question is who have the final word? Who determines who has the last interpretative word (in the broad sense) is of course the legal system itself. And which constitutional principle could determine the political branch to have the last word? Obviously the democratic principle. But in what concrete situations in which, for example, the legislator intends to restrict fundamental rights to pursue some public interest does the legal system gives him the last word? It seems to me that this only happens in situations of epistemic uncertainty.³⁸ In other words, the conditions for judicial deference seem to lie precisely in the verification of a certain level of epistemic uncertainty, which must be weighed against the intensity of the restriction of the fundamental right intended to be carried out. From the balancing process, therefore, may result the prevalence of the political competence over the judicial review competence. But as I see it, in Henkin’s fashion, in this case the conclusion is that the political measure intended to be carried out by the political power it is under the political discretionary autonomy and therefore does not violate the Constitution.

Differently from what we have seen previously, Bickel—one of the most prominent defendants of the PQD—considered the doctrine involved far more than the mere recognition that the political branches had acted within the discretion the Constitution granted to them. In his own words, there is “something different about [the doctrine], in kind not in degree; something greatly more flexible, something of prudence, not construction and not principle. And it is something that cannot exist within the four comers of Marbury u. Madison.”⁴⁰ For Bickel, the PQD is the mechanism by which, when principle and expedience conflicts, courts give expedience its due. Therefore, his thesis could be interpreted in the sense that the doctrine reflects the idea that constitutional adjudication has limits and that courts must inevitably make judgments about whether to apply constitutional law. In some cases, therefore, the Court simply should not obey constitutional commands. However, Bickel’s PQD it only provided reasons for inaction, but not reasons for action.⁴¹

Going even further than Bickel, Seidman argues that what he calls the secret PQD takes seriously the fact that no normative principle can establish its own legitimacy. Hence, even if the answer to a constitutional question is clear, courts must always decide whether they should abide by that answer. And he concludes that cases such as stare decisis, constitutional remedies and doctrine elaboration, which are usually carried out by the Supreme Court, are outside the constitutional law.⁴² But I think he is partially wrong.

First of all, I think there is some problems with his concept of constitutional law; it is each legal system’s rule of recognition that decides what are the sources of constitutional law: they are invariably the Constitution and the constitutional custom (which usually includes judicial precedent). And it is equally perfectly normal to have also implicit constitutional norms expressly (of customary root).⁴³ This immediately shows that, for example, the stare decisis is usually imposed by the legal certainty principle, and the alleged judicial doctrine is usually permitted by legal systems, being within court’s competence of interpretation and application of law, from which one begins to create jurisprudential customs. But if this is so, as it seems to me, we continue to find ourselves within the law, that is, within the legal system.

Is there, then, any “truth” in Seidman’s thesis? That is, is there any legal phenomenon that, by its peculiarity, can justify the alleged autonomy of this third type of PQD? I think there is: it is the phenomenon of defeasibility,⁴⁴ which is especially important within constitutional principles’ conflicts.⁴⁵

Although defeasibility is not uniquely pointed to a specific area, I consider it as a (necessary) property of norms—as the norm’s “sensitivity” to the “factual and legal context”⁴⁶—according to which the fulfillment of the application conditions of a particular norm does not necessarily mean that this norm will be, all things considered, applied to the case. Whenever more than one norm is applicable to a case, the definitive applicability of these norms depends on the result of the normative conflict; all things considered, only the prevailing norm will be applied, and the other will be defeated. Since all norms may enter into normative conflicts—regardless of whether they are rules or principles—all norms are prima facie applicable, because if they are specifically defeated they do not apply (all things considered) to the case.

Although it is possible to distinguish between “undercutting defeasibility”, according to which one norm prevails over another due to a third prevalence norm, of particular interest in this context is the “rebutting defeasibility”, according to which one norm may prevail over another in the context of a balancing process—which applies precisely in cases where no other normative conflict resolution norms resolve the antinomy.⁴⁷

The conflicts between constitutional norms—though not necessarily—usually (i) belong to the partial-partial type,⁴⁸ (ii) are composed of norms of principle⁴⁹ and (iii) the 1st degree norms of conflicts do not apply in most cases to these types of conflicts.⁵⁰ This is because the norms in question have the same hierarchy, were adopted at the same time and are not in a “specialty” relationship. It is precisely for the cases where none of the other conflict resolution norms resolves the antinomy that it is necessary to resort to the so-called balancing.⁵¹

The bigger problem in the balancing cases—which resembles Seidman’s thesis— lies in the fact that, since the legal system does not directly solve this type of normative conflicts, the balancing occurs, to a certain extent, outside the legal system,⁵² and it is necessary to resort to value judgments about the concrete case. In different words, the balancing operation simply means giving the applier discretion to choose between the conflicting rules.⁵³ Nevertheless, the all things considered discretion in this case is smaller than it might seem, because the balancing operation is carried out under the balancing regulating norms⁵⁴ existing in each legal system, such as the proportionality norm, that somehow tell us how to balance.⁵⁵

The conclusion is now clear: in these cases, although we are not dealing with the creation of general norms in the narrow sense, the flank is opened to a manifestly creative judicial activity and can also serve as a precedent, which will be more or less persuasive. However, the resort to balancing is allowed by the legal system, which forbids non-liquet scenarios.

11. From what I have been arguing, it is axiomatic that there are some “political questions” to be addressed and decided by the legislative or government branches and not by courts in typical rule of law systems, characterized precisely by separation of powers.⁵⁶ However, it does not seem necessary to “construct” a PQD because, in a normative perspective, “political questions” seem to be mere spaces of political autonomy conferred to political branches—they are spaces of political discretion.⁵⁷ This means that the legislator and the government, and only them, have the prima facie competence to assess and decide these issues. And it is clear that from this jurisdiction distribution—which is typical in modern rule of law systems—courts have no jurisdiction to control the merits of the political options, as it is imposed by the principle of separation of powers.

This simplistic image, however, is more complicated because, in many cases, the norms conferring competence to the political branches conflict in concreto with the norms conferring competence to courts, for example, to assess the constitutionality of laws. This kind of normative conflicts is peculiar because its resolution necessarily presupposes the operation of balancing, which also entails judicial discretion in assessing which competence should prevail in each case.

That said, I think it is possible to distinguish two hypotheses: (i) it is the constitution that—contingently—forbids the judicial adjudication of certain acts such as the political ones, at least in a substantive way; or (ii) we are facing a case of prima facie discretion, whose extension, however, can only be all things considered determined after the analysis of other constitutional norms also applicable to the case, such as the principles of proportionality, equality, legal certainty, etc.

This means that the legislator and the government does not have full discretion regarding political questions—the extent of the discretion conferred depends on the norms regulating such discretion. In practice, such norms imply a reduction of such discretion and respective margin of autonomy. The problem then lies in determining the extent of discretion and who ought to proceed to such a determination—usually, it is precisely for the courts to determine authoritatively whether the choices made are within or outside the political autonomy. Even if one may disagree of this enormous power conferred to courts—the so-called and feared “judicial supremacy”—, as long as they act within the competence conferred by legal systems, eventual criticisms of being political actors will be normative and non-descriptive.⁵⁸ And in the end, be it good or bad, their supremacy derives from constitutions, which confer authority to courts to authoritatively interpret them.⁵⁹

Therefore, as stated by the Maimom, the point is not that courts have to refrain from deciding, but that judicial decisions will have to recognize that if constitutions allow making certain choices, then they are not being violated. Differently, in cases where there was no discretion, since the choice sought violated the Constitution, courts may oppose the will of the legislature. Moreover, the various examples mentioned in his paper seem to point precisely to my conclusion—there are cases regarding some questions the court somehow refrains to adjudicate, but then, regarding the same questions but in more extreme cases (in which is clear there is a Constitution’s violation), the Court considers itself competent to adjudicate the question.

In conclusion, whenever one concludes that there is no judicial competence to adjudicate a certain decision because it is based on political discretion, and therefore it is a political question, this shows that political questions are after all the result of the assessment of the case and not a criterion for determining what is judicially reviewable.