Harvard Law Review Harvard Law Review Harvard Law Review

Harvard Law Review Forum

Comparing Constitutional Bad Faith


David Pozen’s Constitutional Bad Faith1×1. David E. Pozen, Constitutional Bad Faith, 129 Harv. L. Rev. 885 (2016). is a rich, insightful, and highly original article identifying, and then exploring the causes and consequences of, two gaps in American constitutional culture. The first is between the scarcity of good faith principles in constitutional doctrine and their near-ubiquity in nonconstitutional doctrine, especially in private and international law. The second is between constitutional doctrine and constitutional discourse outside the courts, where allegations of bad faith are legion. In this brief Response, after raising a question about his emphasis on judicial underenforcement, I mostly supplement Pozen’s analysis by considering each of these gaps in turn from a variety of comparative perspectives.

I. The Constitutional/Nonconstitutional Law Gap

Throughout his discussion of the first, doctrinal deficit, Pozen posits that good faith norms in constitutional law are judicially underenforced. Indeed, in express reference to Lawrence Sager’s seminal work on the phenomenon, he suggests that the norm against constitutional bad faith may be the ultimate underenforced norm in the American legal system.2×2. Id. at 897 (citing Lawrence Gene Sager, Fair Measure: The Legal Status of Underenforced Constitutional Norms, 91 Harv. L. Rev. 1212, 1221, 1235 (1978)). For Sager, judicially underenforced norms are still law and so are legally binding on other public officials.3×3. See Sager, supra note 2. But even putting this potentially complicating issue aside, is good faith a judicially underenforced norm in constitutional law or is there (mostly) no such norm in the first place? Why presume that constitutional law is just like private and international law, except that judges have distinctively strong reasons to ignore or underenforce good faith norms in the former? Can one reason from the few specific norms of this kind — the Oath, Take Care, and Full Faith and Credit Clauses — to the existence of a general one? Can one infer from the absence of x that x is underenforced? After all, Sager was discussing such express and indisputable constitutional norms as the Fourteenth Amendment’s guarantee of equal protection and the Fifth Amendment’s prohibition against taking property without just compensation.

In fact, there may be affirmative reason to question the presumption that, in norm-existence terms, constitutional law is just like private and international law. Pozen deploys a spotlight to search for explanations of the doctrinal deficit in the internal workings and logic of constitutional law itself, presenting a range of specific institutional and self-interested reasons why judges are, and perhaps should be, reluctant to enforce good faith norms in this area.4×4. Pozen, supra note 1, at 909–18. But might a floodlight also reveal a broader, more basic, or “architectural” explanation rooted in the general natures of, and comparison between, constitutional law on the one hand and private and international law on the other? Although constitutional and international law are both “law for states” rather than “law by states,”5×5. See generally Jack Goldsmith & Daryl Levinson, Law for States: International Law, Constitutional Law, Public Law, 122 Harv. L. Rev. 1791 (2009). private and international law have in common that they mostly (or at least classically) regulate horizontal relationships among formally equal actors: private individuals and sovereign states, respectively. Moreover, in such regulation, significant roles are played by consent, custom, and bilateralism. This common regulatory posture seems particularly apposite for duties of good faith, as part of treating other actors as equals. Even the special types of horizontal relationships that give rise to fiduciary duties in private law — agent and principal, trustee and beneficiary, director and corporation — can be seen as ways of requiring that (the interests of) the other party be treated on at least an equal basis, in situations of asymmetric information and control between two actors.

By contrast, constitutional law significantly (though, of course, not exclusively) regulates vertical relationships, between the state and its citizens or a federal government and constituent units. These are not relationships between formal, or actual, equals and the regulatory roles of consent and bilateralism tend to be weak, if not non-existent. Accordingly, as a general matter this may well be less fertile terrain for notions of good faith, independent of issues about judicial enforcement. Interestingly, in those specific areas in which constitutional law does regulate more horizontal relationships — among the “coequal” branches of a national government, among constituent federal units inter se, or where constitutional rights bind individuals — one might expect a greater role for legal duties of good faith or their equivalents, such as “comity” or “full faith and credit.” To be clear, this point is not that good faith obligations are categorically excluded from vertical relationships or apply only to actors who are on an equal footing, but that overall, one might expect them to be fewer in constitutional law than in private and international law, to be more the exception than the general background rule, and for those that do exist to be quite specific. For example, liberal democracy is widely thought to require that governments treat their citizens with equal dignity and respect, which arguably gives rise to a particular duty of good faith or its near equivalent in the constitutional-rights jurisprudence of such regimes, as explained and illustrated below.

This all said, it is unclear how much in the article really turns on the underenforcement thesis per se, despite the emphasis on it. The two gaps would still exist, indeed would be greater, if duties of good faith were simply rarer, rather than underenforced, in constitutional doctrine. I suspect its major appeal to Pozen is more of a rhetorical one in that, as in Sager’s account, it suggests the slack will/should be taken up elsewhere, outside the courts, which foreshadows the second gap between constitutional doctrine and discourse.

Let me now turn to the comparative question that Pozen identifies in footnote 115 as worthy of careful study but, “tentative hypothesis” aside, beyond the scope of the article.6×6. Pozen, supra note 1, at 909 n.115. Is this marginal status of bad faith distinctive to American constitutional law, or is it a general feature of constitutional adjudication, perhaps in part for the “architectural” reason just stated? Moreover, to the extent it is distinctive, what might explain this new third gap, between U.S. and non-U.S. constitutional doctrine?

Within rights jurisprudence, one difference seems to be that while for the most part constitutional courts around the world are at least as reluctant to question or review governmental ends, and especially motives, as their U.S. counterpart, they have increasingly subjected governmental means to what is arguably a broader “functional equivalent” of a good faith test through the employment of proportionality analysis. This analysis is part of a structure of constitutional and human rights in which limitations are permissible only if premised on a demonstrable public justification, and failing any of the standard three prongs of the proportionality inquiry — the rationality, necessity, and proportionality in the strict sense of the means used — is tantamount to a finding that the asserted justification was not made in constitutional good faith. That is, an important part of what it means to have a constitutional right is that it can be limited by the government only if acting in good faith for the public interest, and the point of the structured proportionality test is to concretize this requirement, itself stemming from the broader obligation within liberal democracies for governments to treat all citizens with equal dignity and respect. So, for example, in a well-known case, the Israeli Supreme Court held that the rights to property and freedom of movement of Palestinian villagers were disproportionately, and therefore unjustifiably, infringed by the army’s location of a security fence, even though it had first found that the army’s objective was valid, that the fence promoted it (“rationality”), and that there was no alternate, less injurious route with the same security benefits (“necessity”).7×7. HCJ 2056/04 Beit Sourik Vill. Council v. Gov’t of Isr. 58(5) PD 807 (2004) (Isr.). In other words, by imposing excessive burdens on the villagers’ rights relative to the marginal gains in security, the military decisionmakers failed to treat them with equal dignity and respect, and in this sense failed something akin to a constitutional good faith test.

Now, it could be (indeed, has been) argued that U.S. tiers-of-scrutiny jurisprudence reflects a similar basic structure of constitutional rights, even without either the general label or the final prong of “proportionality,” so that any difference is more one of degree than kind. As one who has made this argument,8×8. See Stephen Gardbaum, The Myth and the Reality of American Constitutional Exceptionalism, 107 Mich. L. Rev. 391, 416–31 (2008). not surprisingly I think this is largely right, although the extent to which non-U.S. courts (aided by the large amount of recent normative scholarship on proportionality) increasingly recognize and understand their task in terms of this equal dignity/good faith framework as well as the specific added contribution to it of the final prong, as in the above example, mean that this difference of degree is not negligible. On structural issues, as Pozen notes, constitutional courts in several other federal systems, most notably Germany and the European Union, have created and enforce a more explicit constitutional doctrine of good faith both between federal institutions and the states, and among the states inter se: Bundestreue or the duty of federal loyalty.9×9. See Pozen, supra note 1, at 906 n.101 (citing Daniel Halberstam, Of Power and Responsibility: The Political Morality of Federal Systems, 90 Va. L. Rev. 731, 732–34 (2004)).

So, to the extent there is such a comparative doctrinal gap, what explains it? Here I think Pozen’s discussion of the Madisonian model of opportunistic behavior by politicians as not deviant but rather “the engine of checks and balances”10×10. Id. at 914. is highly pertinent, as it forms the basis for a useful contrast with other systems. For within the general constraints of its role in constitutional law suggested above, I believe one important explanatory factor is the distinctively deep and abiding anti-governmentalism of U.S. political culture, which means that “good faith” is not something presumed to exist within political institutions or among public officials and so is not a factor on which governmental duties are based or named, even aspirationally. As with Madison’s “enlightened statesmen,”11×11. The Federalist No. 10, at 75 (James Madison) (Clinton Rossiter ed., 2003) (“Enlightened statesmen will not always be at the helm.”). good faith in government might be nice, but it would be foolish to rely on it. Far more important in the real world is to create a system that looks more like Adam Smith’s “invisible hand,” in which the public interest is served not through the “benevolence” of political leaders, any more than dinner is through that of the baker and butcher, but rather through the aggregate effects of their pursuit of self-interest.12×12. See generally Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (1776). Interestingly, The Wealth of Nations, published in the same year as the Declaration of Independence, is not cited or referenced in either The Records of the Federal Convention of 1787 or The Federalist Papers. By contrast, in the land of their publication, Smith’s principles that so transformed economics and economic policy were never applied to the structure of government itself. In Great Britain, as elsewhere in Europe, belief in the general benevolence of rulers — hereditary or elected — persisted, so that a political system in which officials and institutions were mostly assumed and expected to act in good faith for the public interest did not seem (to mix philosophical metaphors) self-evident nonsense upon stilts. Within this broad political culture, parliamentary systems without the checks and balances of U.S. presidentialism became the norm in the nineteenth and twentieth centuries.

II. The Constitutional Doctrine/Culture Gap

Pozen characterizes the second gap he identifies, between the rarity of good faith principles in constitutional doctrine and the ubiquity of bad faith talk in constitutional discourse outside the courts, as a “hallmark of American constitutionalism.”13×13. Pozen, supra note 1, at 919. It seems to me that one major explanation for this gap and the relatively large amount of constitutional bad faith talk outside the courts is itself doctrinal: namely, U.S. standing rules. Especially (although not only) with respect to structural issues, and separation of powers disputes in particular, it is comparatively more difficult for politicians and public officials in the United States to litigate their bad faith claims or functional equivalents, many of which are made in doctrinal terms (for example, usurping legislative/executive/state powers, “animus,” and so on), than in many other countries due to restrictive standing rules and a host of other jurisdictional policies and principles — case or controversy, certiorari, political question — that often in practice mean there is no judicial outlet for them. By contrast, politicians and political institutions are often privileged claimants elsewhere. So, typically, only political actors, not citizens, may bring abstract judicial review proceedings to challenge newly enacted statutes and, until 2010, no other type of constitutional review of legislation existed in France.14×14. See Organic Law 2009-1523 of December 10, 2009 on the Application of Article 61-1 of the Constitution, Official Gazette of France, Dec. 11, 2009, p. 21379. This difference in turn reflects alternative conceptions and justifications of judicial review: as part of the ordinary judicial function of deciding whether the plaintiff or defendant wins a litigated case15×15. See Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803). or as an extraordinary, quasi-political, and public function entrusted to a special — and specialist — judicial body.16×16. Hans Kelsen, La garantie juridictionnelle de la Constitution [The Jurisdictional Protection of the Constitution], 45 Revue du Droit Public 197 (1928) (Fr.). To this extent, the sharp distinction between American constitutional doctrine and discourse is an artificial creation of the former, rather than reflecting any inherently extrajudicial nature of bad faith claims or preference by politicians for airing them outside the courts.

Another important explanation is, as Pozen indicates, the comparatively rare but inflammatory U.S. combination of the sparseness, relative difficulty of formal amendment, and sacralization of the constitutional text.17×17. Pozen, supra note 1, at 944–45. I would add to this list the “majestic”18×18. See W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624, 639 (1943) (referring to “the majestic generalities of the Bill of Rights”); see also Tsvi Kahana, Majestic Constitutionalism? The Notwithstanding Mechanism in Israel, in Israeli Constitutional Law in the Making 73 (Gideon Sapir, Daphne Barak-Erez & Aharon Barak eds., 2013). style of significant parts of the original Constitution as well as the Fourteenth Amendment. Even without tough formal amendment rules, a majestic, sparsely worded, relatively enigmatic constitution requires interpretation more than a mundane, prolix one and is also more likely to be sacralized, as it reads like a religious text, oracular rather than profane in style. Sacralization itself makes amendment of the text practically more difficult and adds to the centrality of interpretation — the texts of the Torah, New Testament, and Koran are “unamendable,” though often reinterpreted. Japan’s at least semisacralized constitution has never been amended despite quite liberal formal amendment rules, which is the immediate context for the current widespread allegations of constitutional bad faith against Prime Minister Abe’s “reinterpretation” of Article 9, broadening the concept of self-defense to permit deployment of Japanese troops abroad to aid an ally under attack.19×19. See Bruce Ackerman & Tokujin Matsudaira, Dishonest Abe, Foreign Pol’y (June 24, 2014), http://foreignpolicy.com/2014/06/24/dishonest-abe [https://perma.cc/5UPD-CPEQ]. As in this example, when interpretation becomes the central focus of constitutional discourse and the near-exclusive mode of constitutional change, claims of interpretive bad faith are likely to follow.

Two additional comparative factors that I believe play at least some role here are form of government and judicial appointment/style. American-style separation between legislative and executive branches largely creates the issue of “usurpation” that fuels much constitutional bad faith talk in a way that is rarer in more “blended” parliamentary systems, at least on specific issues. This is maximized when different political parties control the two branches, often leading to the sort of paralysis and frustration that results in the blame game and its raised-stakes version of bad faith talk. A second “inflammatory” combination in the United States is that of (1) life tenure and (2) political appointment for federal judges with (3) the difficulty of amending the Constitution. For, especially at the Supreme Court level, this double no-exit scenario raises the stakes of each relatively rare vacancy and the interpretive philosophies of the various candidates. The U.S. system is therefore more likely to create a breeding ground for suspicion and claims of betrayal resulting in allegations of bad faith than either the fixed term/rolling changeovers of many constitutional courts operating under more easily amended constitutions or the more independent, nonpolitical appointment process of several others. Similarly, the attributable, individual opinions of U.S. (and other common law) judges, compared to the traditional anonymous and collective opinion of the court elsewhere, provide the public exhibits for bad faith talk directed at the judiciary.

Turning to a different comparison, Pozen suggests that this second gap is distinctive to constitutional bad faith and does not apply to private and international law.20×20. “Relative to their role in other areas of law, charges of bad faith are both less important (within the courts) and more important (outside the courts) in the constitutional domain.” Pozen, supra note 1, at 887. That is, there is either no extrajudicial bad faith discourse in these fields or far less of it; almost “all the action” as it were takes place doctrinally, in the courts. While this is probably accurate for private law, as even the most sacralized of civil codes such as the French and German ones do not regularly provide much fodder for extrajudicial bad faith jousting by public officials, it is more questionable for international law. Bad faith talk appears to be very common outside courts in international law/relations, even if not the most tactful or effective of diplomatic modes. Especially given the typically limited jurisdictions and enforcement powers of international tribunals relative to domestic constitutional ones, the gap between charges of bad faith pursued inside and outside the courtroom would seem to be large. So, just as Prime Minister Abe is being widely accused of bad faith inside Japan for his “reinterpretation” of Article 9, the external political allegations of bad faith — in terms of international use-of-force norms and masking his real intentions — coming from Japan’s neighbors, alarmed at what a more liberated Japanese military means for them, are at least as great. Accordingly, in some cases the doctrine/discourse gap may reflect the interplay of legal norms and their judicial enforceability to a greater extent than is true of the constitutional law/nonconstitutional law gap.

Let me conclude this Response with a potential implication of Constitutional Bad Faith for the general debate about judicial review. Pozen shrewdly argues that there may be certain benefits of a bad faith talk surplus to set against its more obvious costs.21×21. Id. at 951–54. One of the slightly more subtle, and perhaps ironic, of these costs to emerge from the article is that a consequence of constitutionalization per se, of legal constitutionalism, may be the futility of containment. Once courts are in the business of judicial review, talk of constitutional bad faith is all but certain to spread from them to the political institutions, even if the size of the second gap may vary based on some of the factors discussed above. Accordingly, not least among the article’s many contributions is that the seeming inevitability of such discourse — with its associated costs and benefits — should be taken into account by both normative theorists and the remaining handful of holdouts from the global trend toward judicial review if and when they ever reconsider their positions.



* MacArthur Foundation Professor of International Justice and Human Rights, UCLA School of Law.