Rights are more than mere interests, but they are not absolute. And so two competing frames have emerged for adjudicating conflicts over rights. Under the first frame, rights are absolute but for the exceptional circumstances in which they may be limited. Constitutional adjudication within this frame is primarily an interpretive exercise fixed on identifying the substance and reach of any constitutional rights at issue. Under the second frame, rights are limited but for the exceptional circumstances in which they are absolute. Adjudication within this frame is primarily an empirical exercise fixed on testing the government’s justification for its action. In one frame, the paradigm cases of rights infringement arise as the consequences of governing poorly. In the other, the paradigm cases arise as the costs of governing well.
The first frame describes the approach of the U.S. Supreme Court over roughly the last half century. The second frame describes the approach of the rest of the developed world over the same period. Neither frame is perfect; many of their flaws track the inherent limits of judicial review in a democracy. The two frames might indeed produce similar results in particular cases. But across time and space, the choice of frame has profound consequences for constitutional law and for its subjects. In particular, the first frame, the one that dominates U.S. courts, has special pathologies that ill prepare its practitioners to referee the paradigmatic conflicts of a modern, pluralistic political order.
To wit, two men, Charlie Craig and Dave Mullins, wanted a cake for their wedding.1×1. Masterpiece Cakeshop, Ltd. v. Colo. Civil Rights Comm’n, 138 S. Ct. 1719, 1724 (2018). They visited a bakery, Masterpiece Cakeshop, in Lakewood, Colorado, where they were told by the owner and chef, Jack Phillips, that his Christian religious beliefs prevented him from baking custom cakes for same-sex weddings.2×2. Id. Discrimination on the basis of sexual orientation violates Colorado’s public accommodation law,3×3. Colo. Rev. Stat. § 24-34-601(2)(a) (2017). and Craig and Mullins sued to require that the law be enforced against Phillips.4×4. Masterpiece Cakeshop, 138 S. Ct. at 1725. Phillips answered that, for him, designing cakes was a form of artistic expression, and so requiring him to do so for a same-sex wedding violated his constitutional rights not just to freedom of religion but to freedom of speech as well.5×5. Id. at 1726.
In the artistic spirit, consider two portraits of the dispute that became Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission.6×6. 138 S. Ct. 1719. Within one frame we observe a dispute among friends, citizens who share a core set of values but who are quibbling — vigorously, to be sure — over how those values apply to a particular set of facts. That core includes the view that art is a protected form of expression that the state may not compel an artist to produce against his will.7×7. See Transcript of Oral Argument at 63, 78, Masterpiece Cakeshop, 138 S. Ct. 1719 (No. 16-111), https://www.supremecourt.gov/oral_arguments/argument_transcripts/2017/16-111_f314.pdf [https://perma.cc/Y4HB-ZM6M]. What’s more, conditioning Phillips’s ability to operate a bakery on the state’s controlling the content of his art would count, prima facie, as a forbidden compulsion.8×8. See id. at 78. Both sides likewise agree that a free speech defense would not permit Phillips to refuse to sell cakes to Craig and Mullins simply because they are gay men.9×9. See Reply Brief for Petitioners at 14, Masterpiece Cakeshop, 138 S. Ct. 1719 (No. 16-111); Brief for Respondents Charlie Craig and David Mullins at 2–3, Masterpiece Cakeshop, 138 S. Ct. 1719 (No. 16-111); Brief for Respondent Colorado Civil Rights Commission at 3, Masterpiece Cakeshop, 138 S. Ct. 1719 (No. 16-111). Phillips’s argument was not grounded in a generalized right of association or autonomy that would allow him to engage consensually with whomever he chooses and on whatever terms. His was a subtler claim, a pinpoint in the canvas, and the couple’s rebuttal was equally so. Phillips has his rights, so do Craig and Mullins, and, crucially, so do the people of the State of Colorado in whose name its public accommodation law speaks. This is a portrait of rights on all sides, reconcilable only at retail, if at all.
Within the other frame rests a darker portrait, a legal Guernica cluttered with slippery slopes, law school hypotheticals, and assorted horribles on parade. At oral argument, Justice Gorsuch asked David Cole, representing the couple for the ACLU, whether the Colorado law would require a baker to sell a cake with a cross on it to a member of the Ku Klux Klan.10×10. Transcript of Oral Argument, supra note 7, at 87. Justice Alito asked the State’s Solicitor General whether the law could force the baker to provide a cake honoring the anniversary of Kristallnacht.11×11. Id. at 70. Justice Breyer and Justice Kagan both compared Masterpiece Cakeshop to Ollie’s Barbecue, the “whites only” Alabama restaurant that challenged the public accommodation provisions of the Civil Rights Act of 1964.12×12. Id. at 18–19, 37; see also Katzenbach v. McClung, 379 U.S. 294, 296–98 (1964) (describing the discriminatory practice at Ollie’s Barbecue). One does not align another with Nazis and segregationists to be friendly or subtle. Rather, these pathological cases signal the outlandishness of the position they are recruited to support.13×13. See Vincent Blasi, The Pathological Perspective and the First Amendment, 85 Colum. L. Rev. 449, 462 (1985) (defining “pathologies” to which Blasi argues First Amendment doctrine ought to respond). This is a portrait of rights on one side, bad faith on the other, and powerful disagreement about which is which. This conflict is reconcilable only at wholesale, and without mercy to the loser.
This categorical, zero-sum frame reflects a noble instinct. Professor Ronald Dworkin gave the most articulate expression to the idea that rights are best conceived as “trumps.”14×14. See Ronald Dworkin, Taking Rights Seriously xi (1977). Dworkin argued that to subject rights to balancing against the public good is to deny them altogether.15×15. Id. at 192. But one consequence of rarely subjecting rights to balancing is that the rights themselves must be articulated with care and specificity. The line demarcating those who hold rights and those who do not becomes a momentous one that merits the attention of lawmakers, citizens, and adjudicators. Dworkin emphasized that, inasmuch as it is concerned with rights, the Constitution’s attention should be rewarding; it follows that its attention must likewise be precious.
This frame creates many problems for constitutional law. For one thing, it is ill-equipped to address the core conflicts that populate the constitutional dockets of U.S. courts. Within a mature rights culture, the typical cases that reach deep into the appellate courts and up to the Supreme Court do not arise from the wholesale denials of citizenship that preoccupied Dworkin but rather from workaday acts of governance from which individuals seek retail exemption: a zealous licensing scheme, a questionable automobile search or dog sniff, a novel or annoying time, place, or manner restriction on speech or gun possession. The paradigm cases that might once have been thought to justify judicial review — racial segregation, McCarthyism, and the like — are no longer paradigmatic, if they ever were.
Less momentous cases sit uneasily within a rights-as-trumps frame. The frame induces our identification of rights to track the categories judges are able to access, articulate, and delimit rather than the moral, political, or even constitutional justice the rights mean to promote. And so Americans have a right to market pharmaceuticals to doctors16×16. See Sorrell v. IMS Health Inc., 564 U.S. 552, 557 (2011). or to parlay the corporate form into electioneering expenditures,17×17. See Citizens United v. FEC, 558 U.S. 310, 339–40 (2010). both of which the Court categorizes as “speech,”18×18. Sorrell, 564 U.S. at 552; Citizens United, 558 U.S. at 310. but no federal constitutional right to food, shelter, or education, which are harder to corral. The Court’s two major partisan gerrymandering cases this Term model the constitutional distortion rights as trumps produces. Judges fear holding that they can adjudicate partisan gerrymandering claims not because such claims are fallacious or frivolous — most members of the Court seem to agree that partisan gerrymandering is antidemocratic — but rather because such claims are intuitive and inviting.
The tension between judges’ intuitions about justice and their understandings of the bounds of inherited categories leads to a second problem: typically tacit (and therefore baffling) distortions of the categories themselves. We all have our favorite examples of the Court pretending to apply rational basis review but instead applying a heightened form of scrutiny,19×19. See, e.g., Romer v. Evans, 517 U.S. 620, 631–35 (1996); id. at 640–43 (Scalia, J., dissenting); City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 447–50 (1985); id. at 456 (Marshall, J., concurring in the judgment in part and dissenting in part); Plyler v. Doe, 457 U.S. 202, 227–30 (1982); id. at 244 (Burger, C.J., dissenting); Reed v. Reed, 404 U.S. 71, 76–77 (1971). or vice versa.20×20. See, e.g., Fisher v. Univ. of Tex. at Austin (Fisher II), 136 S. Ct. 2198, 2208, 2211–14 (2016); id. at 2215 (Alito, J., dissenting); Grutter v. Bollinger, 539 U.S. 306, 327–43 (2003); id. at 350 (Thomas, J., concurring in part and dissenting in part). When an ex ante choice of category largely determines the ex post decision, manipulation of that choice is to be expected: to deny a rights claim within this framework is to say the right does not exist.21×21. See Richard H. Fallon, Jr., The Linkage Between Justiciability and Remedies — And Their Connections to Substantive Rights, 92 Va. L. Rev. 633, 635–37 (2006). And so these cases do not reflect lawlessness tout court, a standard accusation,22×22. See, e.g., Suzanne B. Goldberg, Equality Without Tiers, 77 S. Cal. L. Rev. 481, 481–84 (2004). so much as a breakdown in legal form, not so unlike resort to equity to surmount the limits of common law pleading. Still, lack of transparency about the basis for decision is a rule-of-law problem that the rights-as-trump frame invites.
A third problem has been less recognized but might be most damaging. The costs of the rights-as-trumps frame extend beyond substantive constitutional justice and legal form and into a relational register. Constitutional law is not just a set of foundational rules and standards that govern the structure of the constituted government and the behavior of its actors. It is also a style of — a grammar for — political argument.23×23. Philip Bobbitt, Constitutional Fate 5–6 (1982); J.M. Balkin & Sanford Levinson, Constitutional Grammar, 72 Tex. L. Rev. 1771, 1782–84 (1994). Every case begins, by Article III hypothesis, as a narrow one between adversaries.24×24. U.S. Const. art. III, § 2. A legal claim socializes their enmity and, by submitting it to public scrutiny, generalizes it as well. Forcing rights into prefabricated modules flattens litigants’ gazes and encourages them to tie each other to the most pathological case. Because the rights-as-trumps frame cannot accommodate conflicts of rights, it forces us to deny that our opponents have them. When rights are trumps, they favor rhetoric over judgment, simplicity over context, homogeneity over diversity. The frame requires us to formulate constitutional politics as a battle between those who are of constitutional concern and those who are not. It coarsens us, and by leaving us farther apart at the end of a dispute than we were at the beginning, it diminishes us.25×25. See Mary Ann Glendon, Rights Talk: The Impoverishment of Political Discourse 14 (1991).
It is sometimes said that a special, if not unique, feature of U.S. constitutional law is that it constitutes us, not just as a nation, but as a people as well.26×26. See Akhil Reed Amar, America’s Constitution 5 (2005). Ours is not an ethno-national or religious project but a political one, dedicated to the audacious idea that liberalism and pluralism are mutually constituted. Needless to say, it has not always been so. The early history of the United States required reconciling an unprecedented commitment to liberty for and equality among whites with an equally impressive commitment to chattel slavery for Africans and their descendants and domination of indigenous people.27×27. See Aziz Rana, The Two Faces of American Freedom 5–7 (2010); Patrick Wolfe, Settler Colonialism and the Elimination of the Native, 8 J. Genocide Res. 387, 394 (2006). The civil rights movement coincided with a rights explosion that has challenged the movement’s priority of place within the constitutional culture. The paradigmatic rights conflict of the twenty-first century has involved multiple principles that must be jointly maximized or else selectively abandoned. Respect for the nation’s complexity requires that rights recognition be a jurisgenerative rather than a jurispathic process. Our rights culture cannot constitute us unless all rights count, and all rights cannot count if all rights are absolute.
This Foreword charts a path forward. The claim that the rights-as-trumps, or categorical, frame describes U.S. constitutional law will encounter resistance, and so Part II defines the terms of reference and makes the positive case for fitting the Court’s jurisprudence into those terms. The U.S. Supreme Court balances pervasively, and what categories it maintains are riddled with exceptions. What is distinctive about the American position, and what aligns it with Dworkin, comes into view only in comparison to the dominant alternative: proportion-ality. Categoricalism and proportionality reflect different orientations toward what Professors Moshe Cohen-Eliya and Iddo Porat describe as the dyad of legal authority versus legal justification.28×28. See Moshe Cohen-Eliya & Iddo Porat, Proportionality and Constitutional Culture 111–12 (2013); see also Etienne Mureinik, A Bridge to Where?: Introducing the Interim Bill of Rights, 10 S. Afr. J. on Hum. Rts. 31 (1994). Whether balancing is viewed as exceptional or instead as inherent in the task of rights adjudication affects the constitutional imagination in ways that the Term’s rights cases make vivid.
What follows from the U.S. approach to constitutional adjudication forms the subject of Part III. Categorical adjudication is rule-like in its orientation, and rules by their nature distort the underlying norm they are designed to implement. The frame deprives constitutional decisionmakers of the resources necessary to adjudicate conflicts of rights, as in abortion, affirmative action, or religious exemption cases, and obscures the constitutional dimension of governmental interests, which derive from a constitutionally protected right of political participation. The benefit of rule-like decisional norms lies in the promise of transparency and predictability, but in a society in which rights claims are both ubiquitous and reasonable, this benefit turns out to be elusive. We disagree — reasonably — about the rights that we have,29×29. See Jeremy Waldron, Law and Disagreement 12 (1999). and so a categorical frame burdens the categories with more pressure than they can bear. The result is the worst of both worlds: a dogmatic but capricious devotion to categorical rules.
To be sure, that this state of affairs is problematic does not mean remedies avail themselves, but the Court’s approach is less autochthonous than we might suppose. Part IV corrects the common but misplaced view that conceiving of rights as trumps is some kind of American birthright or is baked into U.S. constitutional arrangements. It is true that the U.S. Constitution does not typically attach limitations clauses to rights, but that fact neither explains the emergence of the U.S. approach nor justifies its continuation. Proportionality analysis is more congenial to the way the lawyers and statesmen of the Founding generation understood rights than the presumptive absolutism that characterizes the modern frame. That frame is rather an artifact of the second half of the twentieth century, when U.S. constitutional lawyers constructed the categorical frame as a way of reconciling the post-Lochner30×30. Lochner v. New York, 198 U.S. 45 (1905). regime of deference to government actors with the unique place of race in the American constitutional order. Countries whose constitutional courts lack the historical baggage and the inherited doctrinal architecture of the U.S. Supreme Court have structured their adjudicative frameworks to match the fecundity of modern rights claims.
Part V applies the Foreword’s insights to some of the Term’s First Amendment, Fourth Amendment, and partisan gerrymandering cases, with special reference to the lessons of foreign experience in these areas. The constitutional jurisprudence of other countries is of course peculiar to their own histories, constitutional structures, and institutional arrangements. But the considered views of foreign legal experts confronting similar problems and drawing on analogous conceptual frames nonetheless have some power to persuade, if lacking power to control.31×31. See Skidmore v. Swift & Co., 323 U.S. 134, 140 (1944); Vicki C. Jackson, Constitutional Engagement in a Transnational Era 71–72 (2010).
The Court views freedom of speech as a classic trump, inviting a good lawyer to formulate a state requirement to bake a cake for a same-sex wedding,32×32. Masterpiece Cakeshop, Ltd. v. Colo. Civil Rights Comm’n, 138 S. Ct. 1719, 1728 (2018). to pay union dues,33×33. Janus v. AFSCME, Council 31, 138 S. Ct. 2448, 2462 (2018). to give women accurate information about available health services,34×34. Nat’l Inst. of Family & Life Advocates v. Becerra, 138 S. Ct. 2361, 2368–70 (2018). or even to engage in partisan gerrymandering as speech infringements.35×35. See Brief for Appellees at 36, Gill v. Whitford, 138 S. Ct. 1916 (2018) (No. 16-1161); see also Gill, 138 S. Ct. at 1934, 1937–38 (Kagan, J., concurring). The tidiness of a speech frame pulled the Court into the case, but Masterpiece Cakeshop was never really about freedom of expression. What made Phillips’s claim constitutionally interesting was not that he bakes especially awesome cakes but rather that his motivation for refusing to make one for Craig and Mullins was grounded in his religious beliefs. A “disparate impact” freedom of religion claim was off the table in the case because the Court — fearing the need for balancing — had declared such claims to be beneath constitutional concern.36×36. Emp’t Div. v. Smith, 494 U.S. 872, 882–83 (1990). Phillips’s claim of intentional religious discrimination remained in the case and eventually proved decisive. See Masterpiece Cakeshop, 138 S. Ct at 1730–31. Other jurisdictions handle religious discrimination claims with subtler instruments.
Unlike the apparently absolute language of the First Amendment — “Congress shall make no law”37×37. U.S. Const. amend. I. — the Fourth Amendment’s text seems to invite an adjudicator precisely to evaluate the reasonableness of a search at retail, on a case-by-case basis.38×38. See Akhil Reed Amar, Fourth Amendment First Principles, 107 Harv. L. Rev. 757, 759 (1994). And yet, because most searches must be conducted pursuant to warrants supported by probable cause, the weight of analysis in many Fourth Amendment cases resembles an ontological inquiry into the nature of a search.39×39. See id. at 762–85; Anthony G. Amsterdam, Perspectives on the Fourth Amendment, 58 Minn. L. Rev. 349, 358–60 (1974). The resulting doctrinal Byzantium requires the Court’s constant care, as rules built against the limits of human attention and cognition are jerry-rigged onto panoptic digital technology, omniscient DNA evidence, and countless other unforeseen circumstances. Carpenter v. United States40×40. 138 S. Ct. 2206 (2018). well illustrates the costs of sidelining reasonableness as the Fourth Amendment’s über-value.
In partisan gerrymandering cases, the Court has searched in vain for a categorical frame to save it from the social science needed to assess the effects of a politically biased districting map.41×41. See Vieth v. Jubelirer, 541 U.S. 267, 314 (2004) (Kennedy, J., concurring in the judgment). At oral argument, Chief Justice Roberts referred to the statistical evidence offered to resolve such cases as “sociological gobbledygook.”42×42. Transcript of Oral Argument at 40, Gill v. Whitford, 138 S. Ct. 1916 (2018) (No. 16-1161), https://www.supremecourt.gov/oral_arguments/argument_transcripts/2017/16-1161_mjn0.pdf [https://perma.cc/S9PW-2CBE]. The rights-as-trumps frame takes as a premise that judges are better suited to resolve interpretive disagreement over the stylized content of the law rather than empirical uncertainty about facts in the world. Along with the Court’s decision upholding the President’s ban on travel from a set of mostly Muslim-majority countries,43×43. See Trump v. Hawaii, 138 S. Ct. 2392, 2423 (2018). the gerrymandering cases show how the categorical frame’s fixation on policing the borders of political authority can deny the protection of rights at just the point when protection is most urgent.
More broadly, the partisan gerrymandering cases offer a poignant example of what can be gained in reimagining the relationship between constitutional law and constitutional politics. Professor William Eskridge has identified “lowering the stakes of politics” as the overriding end of judicial review within a pluralist democracy.44×44. See William N. Eskridge, Jr., Pluralism and Distrust: How Courts Can Support Democracy by Lowering the Stakes of Politics, 114 Yale L.J. 1279, 1293–94 (2005). A rights-as-trumps frame makes hash of that goal, for it saps the losing side in constitutional disputes of the leverage to deliberate toward political consensus. The states not only urged a hands-off approach to partisan gerrymanders but also insisted that allowing the controlling political party to insulate itself from political competition is uniquely consistent with the judicial role.45×45. See Brief for Appellants at 39–40, Gill, 138 S. Ct. 1916 (No. 16-1161); Brief of Amici Curiae States of Michigan et al. in Support of Appellees at 3–8, Benisek v. Lamone, 138 S. Ct. 1942 (2018) (No. 17-333).
It is in fact quite the opposite. The best justification for judicial review in a pluralist democracy with a mature rights culture is that judges have the unique capacity to call partisans to the table, to enable them to see the dignity in each other’s commitments. That trusty footnote in United States v. Carolene Products Co.,46×46. 304 U.S. 144, 152 n.4 (1938). especially as Professor John Hart Ely famously envisioned it, sought to reserve judicial review for instances in which the ordinary political process was unworthy of trust.47×47. See John Hart Ely, Democracy and Distrust 75–77 (1980). If the measure of such a politics is a systematic disregard of the interests of, and a refusal to negotiate with, those not currently holding power, then courts today should be very busy indeed.48×48. Cf. Bowers v. Hardwick, 478 U.S. 186, 196 (1986) (“[If] all laws representing essentially moral choices are to be invalidated under the Due Process Clause, the courts will be very busy indeed.”).
Rights are constantly at stake. And while we take rights seriously enough, we do not do so reasonably enough. Therein lies the path to rebuilding American politics, a feat that is, if I may, worthy of Hercules.
* Dwight Professor of Law, Columbia Law School. For helpful comments and criticism, I thank Jessie Allen, Akhil Reed Amar, Nicholas Bagley, Randy Barnett, Seyla Benhabib, Evan Bernick, Deborah Brake, Samuel Issacharoff, Vicki Jackson, Jeremy Kessler, Suzanne Kim, Gillian Metzger, Henry Monaghan, David Pozen, Jedediah Purdy, Jack Rakove, Russell Rob-inson, Fred Schauer, Sarah Seo, Reva Siegel, Yvonne Tew, students at the Georgetown Advanced Constitutional Law symposium, participants at workshops at Columbia Law School and the University of Pittsburgh School of Law, and the editors of the Harvard Law Review. I owe special thanks to Katherine Yon Ebright, Michael Lemanski, Zach Piaker, Swara Saraiya, and Sophie Schuit for indispensable research assistance.