The Constitution’s First Declared War: The Northwestern Confederacy War of 1790–95

What counts as the first presidential war—the practice of Presidents waging war without prior congressional sanction? In the wake of President Donald Trump’s attacks on Syria, the Office of Legal Counsel opined that unilateral presidential war-making dates back 230 years, to George Washington. The Office claimed that the first President waged war against Native American tribes in the Northwest Territory without first securing congressional authorization. If true, executive war-making has a pedigree as old as the Constitution itself. Grounded in a systematic review of congressional laws, executive correspondence, and rich context of the era, this Article evaluates the claim that our first President waged war in reliance upon his constitutional authority. In fact, there is little that supports the bold claim. Congress authorized war against Northwestern tribes raiding frontier settlements. In other words, Congress exercised its power to declare war and did, in fact, declare war, albeit without using that phrase. Moreover, Washington and his cabinet repeatedly disclaimed any constitutional power to wage war without congressional sanction, making it exceedingly unlikely that he waged war of his own accord or in sole reliance on his constitutional powers. Washington’s abjurations of power should make executive-branch lawyers blush, for the Commander in Chief and his celebrated advisors, including Alexander Hamilton, Thomas Jefferson, and Henry Knox, consistently observed that Presidents could not take the nation to war and, therefore, could not sanction offensive measures, including attacks. The Constitution’s First War was a congressional war through and through, just as the Constitution requires. It was not a presidential war and cannot be cited as a long-lost precedent for presidential wars in Korea, Libya, or Iran.


In January of 2020, the United States killed Qasem Soleimani.1.Merrit Kennedy & Jackie Northam, Was It Legal for the U.S. To Kill a Top Iranian Military Leader?, NPR (Jan. 4, 2020), []; Oona A. Hathaway, The Soleimani Strike Defied the U.S. Constitution, Atlantic (Jan. 4, 2020), [https://pe­].Show More Soleimani was Iran’s second-most powerful leader and responsible for killing many American military personnel. The drone strike touched off praise and censure, including doubts about its constitutionality.2.Rand Paul (@RandPaul), Twitter (Jan. 3, 2020, 9:02 AM),­RandPaul/status/1213098238573723649 [].Show More Could the President kill a foreign leader with no congressional authorization? Senator Rand Paul insisted that “[i]f we are to go to war [with] Iran the Constitution dictates that we declare war.”3.Risch Says Soleimani Was ‘Ratcheting Up’ Attacks on the U.S., PBS NewsHour (Jan. 3, 2020), [] (statement of Sen. James Risch). Although the Senator also cited the War Powers Act, the Act conveys no authority to order attacks. Id.; War Powers Resolution of 1973, Pub. L. No. 93-148, § 2, 87 Stat. 555.Show More Senator James Risch disagreed, arguing that “the president . . . has [war] powers under Article 2 of the Constitution.”4.Paul Kane & Mike DeBonis, Trump’s Order To Strike Iranian Commander Sparks Fresh Debate in Congress over War Powers, Wash. Post(Jan. 3, 2020), https://www.washington [].Show More He further noted that “[t]his debate [over war powers] started under George Washington.”5.April 2018 Airstrikes Against Syrian Chemical-Weapon Facilities, 42 Op. O.L.C. 1, 1 (May 31, 2018),­31/2018-05-31-syrian-airstrikes_1.pdf [].Show More

The audacious attack was hardly unprecedented. In 2018, the United States launched a missile strike against Syrian chemical weapons facilities.6.Michael R. Gordon, Helene Cooper & Michael D. Shear, Dozens of U.S. Missiles Hit Air Base in Syria, N.Y. Times (Apr. 6, 2017),­middleeast/us-said-to-weigh-military-responses-to-syrian-chemical-attack.html [].Show More And the year before, the military attacked a Syrian air base with targeted airstrikes.7.April 2018 Airstrikes Against Syrian Chemical-Weapon Facilities, supra note 6, at 1.Show More Again, no federal law sanctioned any of these earlier strikes. Rather, President Donald J. Trump relied upon his constitutional powers.

In the wake of the 2018 Syrian strikes, the Department of Justice’s Office of Legal Counsel (“OLC”) opined that President Trump had constitutional authority to attack other 7 (quoting Presidential Authority To Permit Incursion into Communist Sanctuaries in the Cambodia-Vietnam Border Area, 1 Op. O.L.C. Supp. 313, 331 (1970)).Show More The OLC stressed that “[the President] as Commander in Chief, is authorized to commit . . . hostilities, without prior congressional approval.” 3.Show More Although the OLC opinion briefly gestured towards constitutional provisions, it actually relied almost entirely on practice. The claim was that President Trump could order the strikes because his predecessors on “dozens of occasions over the course of 230 years” had done the same.10 10.The administration provided a rather brief legal justification for the Soleimani strike, arguing that under Article II, Presidents could use force to, among other things, “protect important national interests.” White House, Notice on the Legal and Policy Frameworks Guiding the United States’ Use of Military Force and Related National Security Operations (2020),­05722/6E1A0F30F9204E380A7AD0C84EC572EC.doc148.pdf []. In a call with reporters, National Security Advisor Robert O’Brien cited both the President’s constitutional authority and the 2002 Iraqi Authorization for Use of Military Force (“AUMF”). Maggie Haberman & Catie Edmondson, White House Notifies Congress of Suleimani Strike Under War Powers Act, N.Y. Times (Jan. 4, 2020),­01/04/us/politics/white-house-war-powers-resolution.html [].Show More In short, longstanding practices, not specific statutory authorization, set the metes and bounds of presidential war powers.11 11.This Article uses the terms “Native American” and “Indian” interchangeably. This is to acknowledge and respect the preferences that different indigenous people have. See Samantha Vincenty, Should You Use Native American or American Indian? That Depends on Who You Ask, Oprah Mag. (Oct. 30, 2020), []; Native Knowledge 360°: Frequently Asked Questions, Nat’l Museum of the Am. Indian, https://american­­tive,preferred%20by%20many%20Native%20people [] (last visited Feb. 10, 2021).Show More

This confident claim, that Presidents have waged war on their own authority since the Constitution’s earliest days, rests on an unjustly obscure conflict: the Northwestern Confederacy War (or First War) conducted against several Native American12 12.The war goes by many names, including the “Northwest Indian War,” the “Little Turtle War,” and “President Washington’s Indian War.” In this Article, we will use either “Northwestern Confederacy War” or “First War.” We delve more deeply into the events infra Part II.Show More tribes north of the Ohio River.13 13.April 2018 Airstrikes Against Syrian Chemical-Weapon Facilities, supra note 6, at 6.Show More According to the OLC, “Presidents have exercised their authority to [wage war] without congressional authorization since the earliest days of the Republic.”14 14.Id.Show More Specifically, “President Washington [ordered] offensive operations against the Wabash Indians in 1790.”15 15.A number of scholars have helped establish the dominant view that the original Constitution left the decision to go to war to Congress, to be exercised by bicameralism and presentment. Here is a partial list: Michael D. Ramsey, The Constitution’s Text in Foreign Affairs, ch. 11 (2007); Louis Fisher, Presidential War Power 6–7 (2d ed. 2004); John Hart Ely, War and Responsibility: Constitutional Lessons of Vietnam and Its Aftermath 3–4 (1993); Michael J. Glennon, Constitutional Diplomacy 80–84 (1990); Francis D. Wormuth & Edwin B. Firmage, To Chain the Dog of War: The War Power of Congress in History and Law 17–18 (2d ed. 1989); Saikrishna Prakash, Unleashing the Dogs of War: What the Constitution Means by “Declare War,” 93 Cornell L. Rev. 45, 48 (2007); William Michael Treanor, Fame, the Founding, and the Power To Declare War, 82 Cornell L. Rev. 695, 699 (1997); Raoul Berger, War-Making by the President, 121 U. Pa. L. Rev. 29, 36 (1972); Charles A. Lofgren, War-Making Under the Constitution: The Original Understanding, 81 Yale L.J. 672, 679 (1972).Show More And because Presidents since George Washington have authorized military attacks without legislative sanction, modern Presidents likewise enjoy the power to wage war without congressional approval.

If our first President waged war without congressional authorization, that fact undermines a common constitutional assertion—that Presidents cannot take the nation to war.16 16.Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 61011 (1952) (Frankfurter, J., concurring).Show More Although many modern scholars and legislators insist that Presidents cannot wage war without congressional authorization, Washington apparently committed the very act that they regard as constitutionally verboten. Further, one might suppose that what was true for Washington must be no less true for Harry Truman, Barack Obama, and Donald Trump. Hence, as a matter of constitutional law, Presidents can wage war as they please against North Korea, Libya, Syria, or, for that matter, Canada.

The OLC’s argument could be understood in two different ways. First, the OLC could be asserting that because Presidents have enjoyed the power to wage war from the Constitution’s inception, this practice sheds light on the original meaning of “executive power,” “Commander in Chief,” and “declare war.” Second, the OLC could be advancing a different claim, namely that despite the original meaning of these phrases, practice from the government’s earliest days has layered a “gloss” on them,17 17.John C. Yoo & Robert J. Delahunty, Authority for Use of Military Force To Combat Terrorist Activities Within the United States 10 n.15 (Oct. 23, 2001), https://nsarch­ [­4QEN-AF­WP]; Authorization for Continuing Hostilities in Kosovo, 24 Op. O.L.C. 327, 333 (2000).Show More meaning that whatever the original scheme, Presidents today enjoy the power to wage war. We believe the 2018 OLC opinion makes the first sort of claim. After all, dutiful and upright Washington would never deliberately violate the Constitution. If he took the nation to war, it would seem that, notwithstanding Congress’s power to declare war, the original Constitution truly sanctioned presidential wars. And it follows that the conventional view about war powers is misguided because Washington’s war refutes it.

The OLC’s recent invocation of the Northwestern Confederacy War is not exceptional. Other OLC opinions have cited the war, although none have given it the prominence and weight that the 2018 opinion does.18 18.Abraham D. Sofaer, The Power Over War, 50 U. Miami L. Rev. 33, 38–41 (1995); John Yoo, George Washington and the Executive Power, 5 U. St. Thomas J.L. & Pub. Pol’y 1, 19–20 (2010).Show More These opinions relied upon the work of scholars, most notably Abraham Sofaer and John Yoo, who drew constitutional lessons from the war.19 19.See Yoo & Delahunty, supra note 18, at 10 n.15; Authorization for Continuing Hostilities in Kosovo,supra note 18, at 333.Show More

Because the OLC has repeatedly cited the First War to justify the executive’s unilateral use of military force abroad,20 20.Federal Legislature, Phila. Gen. Advertiser, Jan. 4, 1793, at 3 (comments of Rep. Wadsworth).Show More it is necessary to carefully assess it. There is a considerable risk that an incomplete or mistaken understanding of the war may become embedded in the historical narrative and mislead politicians and scholars. The First War may become the sturdy keystone for a view that Presidents can take the nation to war because that is what Washington supposedly did only a year after the Constitution’s inception.

The OLC’s opinions, and the underlying scholarship, while rigorous in many respects, rely on incomplete evidence and fail to properly situate the conflict in its historical context. The historical record demonstrates that Congress in fact authorized Washington to start the Northwestern Confederacy War and repeatedly approved the war’s continuation. Far from inaugurating the practice of presidential wars, the First War marked the earliest exercise of Congress’s power to “declare war.”

Consequently, Washington laid no novel gloss on the “executive power” or the “Commander in Chief” Clauses. Claims to the contrary tether the first President to a flawed and anachronistic proposition he never once entertained—that Presidents enjoy constitutional authority to start wars. As we demonstrate, George Washington in fact publicly proclaimed exactly the opposite. He forcefully insisted that Commanders in Chief could not wage war unilaterally. He endorsed this principle categorically, applying it even in the wake of declarations of war issued by other nations. On this point, his cabinet fully agreed. The claim that Presidents could lawfully take the nation to war was so outside the mainstream that neither Washington nor anyone else voiced it, even to reject it. At the time, no one read the Constitution as the executive branch (mis)reads it today. The debate we have today simply did not exist during the Washington administration because no one at the time claimed that the Constitution authorized Presidents to start wars.

Resting on the first in-depth evaluation of primary materials, this Article corrects the record and sheds new light on the original War Constitution. In our telling, America’s First War teaches a number of vital lessons. First, Congress’s power to declare war encompassed authority to sanction military expeditions, including the power to authorize offensive measures. Second, despite serving as Commander in Chief and enjoying the “executive power,” the President clearly lacked such power. Third, Congress could exercise its authority to “declare war” without using the precise phrase or a formal declaration. Fourth, via its decisions over the army’s size and the delegation of authority to summon state militias, Congress regulated the President’s conduct of the First War.

The Northwestern Confederacy War witnessed a remarkable number of “firsts.” The war marked the first exercise of Congress’s power to declare war. As one critic said, it was “the war of the legislature.”21 21.By Particular Desire, Phila. Gen. Advertiser, Jan. 7, 1792, at 2.Show More As another detractor put it, the new government found the Indians in the Northwest “in a state of disquietude” and “declare[d] war against them, as a display of power.”22 22.Prakash, supra note 16, at 96, 105.Show More The war also marked the first major interplay between the Commander in Chief and Congress, with the latter guiding the former and the former acting under the auspices of legislative decisions. The Commander in Chief was under the command of Congress.

Part I reviews existing treatments of the Northwestern Confederacy War and recounts the First War. Part II discusses the power to declare war and what the Founders said of that power prior to 1789. Part III recounts the statutes that Congress passed to authorize and support the First War. Part IV discusses what Washington and his cabinet said about presidential power to wage war without congressional authorization. Part V draws concluding lessons from America’s first war.

Conflict Avoidance in Constitutional Law

­­­­Hard cases present a dilemma at the heart of constitutional law. Courts have a duty to decide them—to vindicate rights, to clarify law—but doing so leads to errors (judges do not know the “right answer”) and strains the credibility of courts as impartial decision makers. Theories of constitutional adjudication tend to embrace one horn of this dilemma. We explore a principle for deciding hard cases that appreciates both. We argue that courts should decide hard cases against the party who could have more easily avoided the conflict in the first place. This is the conflict-avoidance principle. The principle builds on and systematizes “least cost avoidance” in private law and myriad constitutional doctrines. We apply the principle to several cases, generating insights into discrimination, affirmative action, religion, and so on. The principle represents a form of common-law constitutionalism, and it reveals connections between rights, markets, and State power. It also invites objections, to which we respond. Conflict avoidance is not “value-neutral,” and it cannot resolve every hard case. But it can resolve many in a practical way.

Take any demand, however slight, which any creature, however weak, may make. Ought it not, for its own sole sake, to be satisfied? If not, prove why not? The only possible kind of proof you could adduce would be the exhibition of another creature who should make a demand that ran the other way.

William James (1891)1.William James, The Moral Philosopher and the Moral Life, 1 Int’l J. Ethics 330, 339 (1891).Show More


How should courts resolve hard constitutional cases?2.See Ronald Dworkin, Hard Cases, 88 Harv. L. Rev. 1057, 1060 (1975) (defining hard cases as ones where “no settled rule dictates a decision either way”).Show More On the one hand, deciding them on the merits strains courts’ credibility as impartial decision makers, especially when they engage in judicial review of legislation where the constitutional text is vague and the interests at stake essentially political.3.See T. Alexander Aleinikoff, Constitutional Law in the Age of Balancing, 96 Yale L.J. 943, 971–73, 977–78 (1987) (criticizing interest balancing). For a thorough and optimistic account of the capacity of courts to balance interests optimally, see generally Robert Alexy, A Theory of Constitutional Rights (Julian Rivers trans., Oxford Univ. Press 2002) (1986) (offering an account of constitutional rights that connects the analytical, empirical, and normative dimensions of legal doctrine); Robert Alexy, Constitutional Rights, Balancing, and Rationality, 16 Ratio Juris. 131 (2003) (arguing that there is a rational structure within balancing). Roughly speaking, interest balancing focuses on which party (or possibly which group) can bear a loss in court more easily. Are the losses to this side (or to this principle) outweighed by the gains to the other? Our enterprise is quite different. We focus on which party could have avoided more easily the conflict that led to the hard case in the first place.Show More On the other hand, courts are constitutionally charged with deciding such cases. A refusal to decide them amounts to shirking that responsibility.4.U.S. Const. art. III, § 2 (“The judicial Power shall extend to all Cases, in Law and Equity, arising under this Constitution . . . .”); Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803) (“It is emphatically the province and duty of the judicial department to say what the law is.”); Alexander M. Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics 70 (1962) (explaining that not deciding cases must “be justified as compatible with the Court’s role as defender of the faith”).Show More Theories of constitutional adjudication often embrace one horn of this dilemma.5.Theories of judicial deference embrace the first horn by treating most constitutional issues as political ones appropriately decided by the political branches. See, e.g., James B. Thayer, The Origin and Scope of the American Doctrine of Constitutional Law, 7 Harv. L. Rev. 129, 129–39 (1893) (explaining origins of judicial review); see also Larry D. Kramer, The People Themselves: Popular Constitutionalism and Judicial Review 7–8 (2004) (comparing early judicial review to contemporary practice). Originalist theories, “moral” interpretations, and “living constitutionalism” tend to treat constitutional questions as essentially legal questions with which the Court is properly tasked with deciding, thereby embracing the second horn. See, e.g., Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law 45–46 (1997) (discussing issues with living constitutionalist interpretations); James E. Fleming, Living Originalism and Living Constitutionalism as Moral Readings of the American Constitution, 92 B.U. L. Rev. 1171, 1172–73 (2012) (offering a “complete, ecumenical approach to constitutional interpretation”); David A. Strauss, The Living Constitution 41–44 (2010) (arguing that judges and lawyers are not properly equipped for originalist interpretation). Process theory and prudential approaches attempt to reconcile the two. See, e.g., John Hart Ely, Democracy and Distrust: A Theory of Judicial Review 4–5 (1980) (arguing that judicial review is best justified when it can be understood as a mechanism for improving the democratic process); Bickel, supra note 4, at 64 (“No good society can be unprincipled; and no viable society can be principle-ridden.”). Our approach draws on elements of both the process and prudential traditions.Show More This Article explores a principle that appreciates the force of both horns: courts should decide hard cases against the party who could have more easily avoided the constitutional conflict in the first place. We call this the conflict-avoidance principle.

To preview the principle, consider an example. Suppose a student wears a Confederate flag shirt to school, in violation of the dress code, and gets disciplined. She argues that this violates her free speech rights, and the school responds that it has the authority to ensure a conducive learning environment.6.Cf. Castorina ex rel. Rewt v. Madison Cnty. Sch. Bd., 246 F.3d 536, 538, 548 (6th Cir. 2001) (noting that a “disruption-free educational environment is a substantial government interest”); Defoe v. Spiva, 625 F.3d 324, 335 (6th Cir. 2010) (holding that the school officials’ concern that displays of the Confederate flag would be disruptive was reasonable).Show More For the sake of argument, assume the case is hard (we will say more about “hard cases” below). A court applying conflict avoidance would compare the relative costs to the parties of avoiding the conflict in the first place. Could the student have expressed herself in another way? Could she have transferred to a school with a more permissive dress code? Could the school have ensured a conducive environment without banning the flag? Whoever could have avoided the conflict more easily would lose.

This is a simple example, but the principle applies the same way in real, controversial cases like Masterpiece Cakeshop, Our Lady of Guadalupe School, Fisher, and Janus.7.Masterpiece Cakeshop, Ltd. v. Colo. C.R. Comm’n, 138 S. Ct. 1719 (2018) (on discrimination and religion); Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049 (2020) (same); Fisher v. Univ. of Tex. at Austin, 136 S. Ct. 2198 (2016) (on affirmative action); Janus v. Am. Fed’n of State, Cnty., & Mun. Emps., Council 31, 138 S. Ct. 2448 (2018) (on speech).Show More We will examine these cases and others below.

Applying the conflict-avoidance principle has several advantages. For one thing, it requires courts to decide cases instead of deflecting or delaying judgment.8.Cf. Bickel, supra note 4, at 71 (approving Justice Brandeis’s statement that “[t]he most important thing we do . . . is not doing” and observing that Brandeis “had in mind all the techniques . . . for staying the Court’s hand”).Show More Second, and more important, applying the conflict-avoidance principle requires courts to decide cases by looking to relatively concrete facts and considerations, rather than to abstract political values. Such an approach not only plays to courts’ institutional strengths; it may also produce a pattern of decisions that vindicate the relevant values where they are needed most. That, at least, is the theory of the common law.9.Frederick Schauer, Do Cases Make Bad Law?, 73 U. Chi. L. Rev. 883, 883 (2006) (“Treating the resolution of concrete disputes as the preferred context in which to make law . . . is the hallmark of the common law approach.”).Show More As Oliver Wendell Holmes famously said, “[i]t is the merit of the common law that it decides the case first and determines the principle afterwards.”10 10.Codes, and the Arrangement of the Law, 5 Am. L. Rev. 1, 1 (1870) (unsigned article by Oliver Wendell Holmes).Show More

Finally, the conflict-avoidance principle encourages parties to avoid the sorts of conflicts that produce hard cases. Deciding such cases imposes real costs. In addition to financial costs, such cases can undercut the legitimacy of courts as judicial institutions, especially when the political stakes are high.11 11.Precisely that concern underlies the Supreme Court’s practice of treating some politically charged issues as “political questions,” incapable of impartial judicial resolution. See Baker v. Carr, 369 U.S. 186, 217 (1962) (stating that cases lacking “judicially discoverable and manageable standards” or requiring a “policy determination of a kind clearly for nonjudicial discretion” involve political questions).Show More Furthermore, deciding hard cases can lead to errors in the sense that judges do not know the “correct” answer (if they did, the case would not be hard). We think reducing the incidence of hard cases is itself a benefit.12 12.We relax the assumption that deciding hard cases imposes more costs than benefits. See infra Part V.Show More

The conflict-avoidance principle has roots in private and public law. It relates to least cost avoidance, which Guido Calabresi identified and developed in tort law.13 13.Guido Calabresi, The Costs of Accidents: A Legal and Economic Analysis 135 (1970) (advocating placing liability on “those acts or activities . . . which could avoid the accident costs most cheaply”).Show More It also resonates with various constitutional doctrines—such as time, place, and manner doctrines in First Amendment law—that inquire into the alternative courses of action available to the parties to a dispute.14 14.See infra Part IV.Show More Also, some scholars have advanced proposals that sound in cost avoidance.15 15.The clearest example would appear to come from Professor Tang, who has two papers in draft form. See Aaron Tang, The Costs of Supreme Court Decisions: Towards a Best Cost-Avoider Theory of Constitutional Law (Sept. 27, 2019) (unpublished manuscript), [­WW] [hereinafter Tang, Cost-Avoider]; Aaron Tang, Constitutional Law After Mazars, Vance, & June Medical: The Case for Harm-Avoider Constitutionalism, 109 Calif. L. Rev. (forthcoming 2021) (on file with authors) [hereinafter Tang, Harm-Avoider]. Professor Tang’s work and ours, which developed simultaneously and independently, are quite different. In brief, we aim to minimize conflicts by placing the onus on the party who could have avoided the dispute at lowest cost, whereas Professor Tang aims to minimize the “costs” of judicial decisions by placing the onus on the group that could bear the loss most easily. See infra note 46. Professor Tang’s work relates more closely to interest balancing, covering, or mitigation (i.e., bearing loss after the fact) than to a conventional understanding of avoidance (preventing the loss from occurring).For other scholarship that sounds in conflict avoidance, see, for example, Douglas Laycock, The Broader Implications of Masterpiece Cakeshop, 2019 BYU L. Rev. 167, 193 (arguing that free exercise claims by service providers should not prevail over non-discrimination claims by LGBT customers in communities where “discrimination is still widespread”); J.H. Verkerke, Is the ADA Efficient?, 50 UCLA L. Rev. 903, 941 (2003) (applying least cost avoidance to disability law in the workplace); Robert D. Cooter, The Strategic Constitution 129–32 (2000) (connecting rights to mobility costs); Frank I. Michelman, Pollution as a Tort: A Non-Accidental Perspective on Calabresi’s Costs, 80 Yale L.J. 647, 666–86 (1971) (book review) (applying least cost avoidance to pollution).We note that conflict avoidance can be seen as a distinct kind of “minimalist” theory of adjudication. See, e.g., Cass Sunstein, Burkean Minimalism, 105 Mich. L. Rev. 353, 355–56 (2006). Minimalist theories direct judges to concentrate on the facts of the case. See id. at 376 (describing as non-minimalist an approach that is “not limited to the facts of particular cases”). Conflict avoidance directs judges to focus on a particular subset of facts, namely on who could have avoided the conflict more easily.Finally, we note that our argument is consistent with a broader, emerging approach to constitutional law. See generally Robert D. Cooter & Michael D. Gilbert, Constitutional Law and Economics, in Research Methods in Constitutional Law: A Handbook (Malcolm Langford & David S. Law eds., forthcoming 2021) (discussing the emergence of economic theory as applied to constitutional law).Show More Thus, we do not offer a radically new approach to constitutional adjudication. Rather, we collect strands of reasoning that already permeate law and legal scholarship and show how, once systematized, they yield a promising and innovative approach to hard cases.

Why hasn’t anyone systematized these ideas before? Why haven’t judges and scholars, many of whom are familiar with least cost avoidance, already applied these ideas to constitutional law? Here is one explanation. Constitutional adjudication often proceeds “top-down.”16 16.Richard A. Posner, Legal Reasoning from the Top Down and from the Bottom Up: The Question of Unenumerated Constitutional Rights, 59 U. Chi. L. Rev. 433, 433 (1992) (defining top-down reasoning as when “the judge or other legal analyst invents or adopts a theory about an area of law—perhaps about all law—and uses it to organize, criticize, accept or reject, explain or explain away, distinguish or amplify the existing decisions to make them conform to the theory and generate an outcome in each new case as it arises that will be consistent with the theory”).Show More The constitutional principles at stake loom large, sweeping away particular case facts. In contrast, least cost avoidance proceeds in a “bottom-up,” context-sensitive fashion.17 17.For an analysis of the formal difference between bottom-up and top-down reasoning, see Charles L. Barzun, Justice Souter’s Common Law, 104 Va. L. Rev. 655, 708–13 (2018) (explaining that, whereas under top-down reasoning, courts apply a fixed major premise (or rule) to the minor premise (or facts) in order to deduce a conclusion, with bottom-up forms of reasoning, the judge aims to let the facts of the case themselves be the guide to the proper outcome).Show More Courts concentrate on the facts (who could have avoided the crash more easily?), rather than on how to best apply the relevant legal principles. Applying least cost avoidance to the Constitution requires taking a bottom-up approach to a subject dominated by top-down reasoning.18 18.Of course, our approach is top-down in the sense that it involves applying the conflict-avoidance principle to many different cases. But the point is that it is a meta-principle that directs courts to focus on the sort of factual nuances that bottom-up approaches consider critical.Show More

Gesturing at least cost avoidance and “bottom-up” reasoning is easy. The hard part is translating it to constitutional law. We take the main contribution of our project to lie in showing what the translation requires.

The conflict-avoidance principle is not a panacea; nor does it claim “value-neutrality.” But it does offer a fresh way of thinking about how to resolve hard cases. Rather than seeing constitutional conflicts as brute clashes of values—liberty vs. equality, positive liberty vs. negative liberty, substantive equality vs. formal equality—courts might make more progress by looking at the concrete difference that vindicating those values would have made in parties’ actual lives. The goal is to see what work rights claims are doing in social and political life.

We develop our argument in five Parts. Part I clarifies the scope of the principle: we confine its use to hard cases, where “hard cases” has a specific meaning that we will explain. Part II briefly reviews least cost avoidance in private law, drawing out a key distinction between avoiding costs and bearing them. Part III operationalizes the conflict-avoidance principle by developing a doctrinal test for its application. Part IV applies the test to real cases, including recent, controversial cases before the Supreme Court. In Part V, we respond to various objections. The Conclusion develops a broader point. Although the conflict-avoidance principle requires no special commitment to private ordering or negative liberty, it does illuminate a connection between markets, rights, and State power.

Settled Law

Settled law” appears frequently in judicial opinions—sometimes to refer to binding precedent, sometimes to denote precedent that has acquired a more mystical permanence, and sometimes as a substantive part of legal doctrine. During judicial confirmation hearings, the term is bandied about as Senators, advocacy groups, and nominees discuss judicial philosophy and deeper ideological commitments. But its varying and often contradictory uses have given rise to a concern that settled law is simply a repository for hopelessly disparate ideas. Without definitional precision, it risks becoming nothing more than empty jargon.

We contend that settled law is actually a meaningful concept, even though it does not embody any single, unified idea. First, we argue that controlling law, which essentially corresponds to binding precedent, is a fundamentally distinct concept that is neither synonymous with nor a subset of settled law. Second, we draw on seminal jurisprudential theories to build a taxonomy of five frameworks that capture how legal actors can invoke settled law, both rhetorically and doctrinally. Third, we demonstrate how a clearer understanding of settled law can make doctrine more coherent and administrable. Situating certain doctrines within the appropriate frameworks, and not conflating controlling law and settled law, would resolve myriad doctrinal anomalies. Moreover, greater conceptual precision can improve political rhetoric during the confirmation process by promoting clearer dialogue and discouraging legal actors from talking past one another.


What does it mean to say that Roe v. Wade1.410 U.S. 113 (1973).Show More is “settled law”? Or Citizens United v. FEC?2.558 U.S. 310 (2010).Show More Or even Brown v. Board of Education?3.347 U.S. 483 (1954).Show More

The idea of settled law has played a pivotal role in Supreme Court confirmation hearings for more than thirty years, and it has animated myriad legal doctrines as far back as the eighteenth century.4.See, e.g., Penhallow v. Doane’s Adm’rs, 3 U.S. (3 Dall.) 54, 118 (1795) (Cushing, J.) (describing as “settled law and usage” the idea that “courts of Admiralty can carry into execution decrees of foreign Admiralties”).Show More Yet the meaning of settled law has proved stubbornly elusive. Does it refer simply to the idea that the Supreme Court has decided a particular issue, or does it connote something more enduring about particular precedents? Does it imply that a precedent is somehow “right”? Which courts (or other legal actors) have the power to settle the law? And how exactly does that happen?

Even though settled law had come up during earlier confirmation hearings,5.See, e.g., Nomination of Justice William Hubbs Rehnquist: Hearings Before the S. Comm. on the Judiciary, 99th Cong. 356 (1986) (statement of Rehnquist, J.) (declaring that the incorporation of the right to a speedy trial through the Fourteenth Amendment “is settled law, and [his] opinions reflect it”); Nomination of Judge Antonin Scalia: Hearings Before the S. Comm. on the Judiciary, 99th Cong. 83 (1986) (statement of Sen. Specter) (asking whether Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803), “is a settled issue”); id. at 104 (statement of Sen. Biden) (“If it’s on the books, if it is settled constitutional law for an extended period of time, and the argument to overturn that settled constitutional principle does not in fact meet the test of on its face being consistent with what the correct constitutional principle is, do you have to stick with what the settled law is?”).Show More it first took center stage in the political arena during the Bork hearings. Under intense scrutiny about his academic writings, Judge Bork repeatedly tried to parry criticism of his controversial views by promising over and over that he would respect “settled law,” even if he disagreed with it.6.See, e.g., Nomination of Robert H. Bork To Be Associate Justice of the Supreme Court of the United States: Hearings Before the S. Comm. on the Judiciary, 100th Cong. 279 (1987) (statement of Bork, J.) (repeatedly calling Brandenburg v. Ohio, 395 U.S. 444 (1969),“settled”); id. at 327 (declaring that “I certainly have no desire to go running around trying to upset settled bodies of law”); id. at 423 (“It seems to me that the settled law is now that the person writing the book does not have to prove that it is political or any way connected to politics. The settled law is the Government has to prove it is obscene.”); id. at 428 (“I am not changing my criticism of [Brandenburg]. I just accept it as settled law.”); id. at 434 (“It’s settled law. . . . I have said that I accept that body of precedent and will apply it. That’s all I’ve said.”); id. at 438 (declaring that “some things are absolutely settled in the law” and that “[a]ny judge understands that you don’t tear those things up”); id. at 587 (“I accept them as settled law. I have not said that I agree with all of those opinions now, but they are settled law and as a judge that does it for me.”); id. at 667 (“I have repeatedly said there are some things that are too settled to be overturned.”).Show More And notably, Judge Bork used that term to mean something quite distinct from the familiar principles of stare decisis.7.See id. at 989 (statement of Sen. Specter) (“[Judge Bork] flatly made a commitment to accept settled law. On the privacy cases he has not made that commitment. He has talked about various considerations of reliance and stare decisis, but he has made no commitment on privacy . . . .”).Show More Since then, every Supreme Court nominee has faced questions about settled law, even as the term’s ambiguity has grown increasingly apparent.8.See infra notes 278–85 and accompanying text.Show More

Discussions of settled law have become even more prominent in recent years as President Trump’s judicial nominees faced pointed questions about whether they agreed with certain precedents or, at a minimum, regarded them as settled.9.SeeLaura Meckler & Robert Barnes, Trump Judicial Nominees Decline To EndorseBrown v. Board Under Senate Questioning, Wash. Post (May 16, 2019, 7:28 PM),­rse-brown-v-board-under-senate-questioning/2019/05/16/d5409d58-7732-11e9-b7ae-390de­4259661_story.html [] (describing how Sen. Blumenthal frequently asks whether nominees regard Brown as correct); Marcia Coyle, Revisiting Amy Coney Barrett Statements About Abortion Rights, Nat’l L.J. (Sept. 25, 2020, 3:12 PM),­ts-about-abortion-rights/?slreturn=20200908080816 [] (des­cribing how Sen. Blumenthal asked then-nominee Amy Coney Barrett if she “think[s] Roe v. Wade was correctly decided”).Show More Sometimes nominees have refused to engage.10 10.See Meckler & Barnes, supra note 9 (describing nominees who refused to directly answer Sen. Blumenthal’s question about Brown); see alsoAriane de Vogue, Judicial Nominees Are Changing Their Approach to the ‘Brown v. Board’ Question at Senate Hearings, CNN (Feb. 10, 2019),­inees/index.html [] (noting that in response to Sen. Blumenthal’s question about Brown, now-Judge Neomi Rao described the case as “longstanding precedent of the Supreme Court,” declared that it was “not appropriate” to comment on the “correctness of particular precedents,” but argued that “it’s hard for me to imagine a circumstance in which Brown v. Board would be overruled by the Supreme Court”).Show More On other occasions, Senators and nominees have appeared to use “settled law” in conspicuously different ways,11 11.For example, in 2010, then-Senator Sessions suggested that “settled law” connoted “a more firm acknowledgment of the power of that ruling” than mere “precedent” and asked then-U.S. Solicitor General Elena Kagan whether she was using “settled law” and “precedent” interchangeably. The Nomination of Elena Kagan To Be an Associate Justice of the Supreme Court of the United States: Hearing Before the S. Comm. on the Judiciary, 111th Cong. 231 (2010) (statement of Sen. Sessions). She responded: “I don’t mean any difference.” Id. (statement of Elena Kagan, Solicitor General of the United States).Show More a phenomenon brought into stark relief during Justice Kavanaugh’s confirmation hearing. Responding to a question from Senator Feinstein, the future Justice declared that Roe v. Wade was “settled as a precedent of the Supreme Court, entitled to respect under principles of stare decisis.”12 12.C-SPAN, Supreme Court Nominee Brett Kavanaugh Confirmation Hearing, Day 2, Part 1, C-SPAN (Sept. 5, 2018),­inee-brett-kavanaugh-confirmation-hearing-day-2-part-1 [­MC] (relevant exchange occurring from 48:25 to 49:10).Show More The next day, The New York Times published a previously confidential e-mail from 2003 in which Kavanaugh had written: “I am not sure that all legal scholars refer to Roe as the settled law of the land at the Supreme Court level since [the] Court can always overrule its precedent, and three current Justices on the Court would do so.”13 13.Charlie Savage, Leaked Kavanaugh Documents Discuss Abortion and Affirmative Action, N.Y. Times (Sept. 6, 2018),­kavanaugh-leaked-documents.html [].Show More These statements provided grist for some to call him disingenuous.14 14.See, e.g.,Igor Bobic, Susan Collins Downplays Brett Kavanaugh Email About Abortion Rights and ‘Settled Law’, HuffPost (Sept. 6, 2018, 4:58 PM),­entry/brett-kavanaugh-susan-collins-roe-v-wade_n_5b9165b1e4b0511db3e04121 [https://p­] (quoting Sen. Blumenthal urging undecided Republicans to “read this [email] and then tell [him] Judge Kavanaugh has been candid with [them]”).Show More Others argued that he was making distinct and mutually consistent claims—a prediction of whether the Court would revisit the abortion precedents versus an assessment of whether those precedents should stand undisturbed.15 15.See, e.g., id. (quoting Sen. Collins saying that Kavanaugh “was merely stating a fact, which is that three [Justices] on the [C]ourt were anti-Roe,” and “[i]f that’s the case and he was not expressing his view, then [she was] not sure what the point of it [was]”).Show More

Settled law is far more than an enigmatic buzzword that gets bandied about during confirmation hearings, though; it also serves an important structural role and has profound doctrinal implications. For example, lower-court judges often speak about their duty to follow the settled law of superior courts.16 16.See infra notes 26–27 and accompanying text.Show More Most surprisingly, an array of doctrines depend substantively on whether the law is “settled.” In the realm of constitutional torts, for instance, a plaintiff attempting to bring a Section 1983 claim usually must overcome the defendant’s qualified immunity by showing that the defendant violated a constitutional rule that was “clearly established” under “settled law.”17 17.District of Columbia v. Wesby, 138 S. Ct. 577, 589 (2018); see also id. at 591 (“The rule applied by [the court below] was not clearly established because it was not ‘settled law.’” (quoting Hunter v. Bryant, 502 U.S. 224, 228 (1991))).Show More So, too, settled law undergirds the circumstances when post-conviction relief is available,18 18.E.g., In re Jones, 226 F.3d 328, 333–34 (4th Cir. 2000) (holding that whether the settled law established the legality of a conviction is part of the Fourth Circuit’s three-prong test to determine the availability of a writ of habeas corpus).Show More lawyers’ ethical obligations under Rule 11,19 19.E.g., Pro. Mgmt. Assocs. v. KPMG LLP, 345 F.3d 1030, 1033 (8th Cir. 2003) (remanding and ordering the lower court to impose a Rule 11 sanction to a plaintiff’s counsel for ignoring the “well-settled law” of res judicata under the circumstances of the case).Show More standards of review,20 20.E.g., United States v. Gary, 954 F.3d 194, 202 (4th Cir. 2020) (quoting United States v. Ramirez-Castillo, 748 F.3d 205, 215 (4th Cir. 2014)) (holding that “if the settled law of the Supreme Court or this circuit establishes that an error has occurred,” the error satisfies the plain error standard of review).Show More and a host of other doctrines.21 21.See, e.g., Hunter v. Philip Morris USA, 582 F.3d 1039, 1043 (9th Cir. 2009) (quoting Hamilton Materials, Inc. v. Dow Chem. Corp., 494 F.3d 1203, 1206 (9th Cir. 2007)) (fraudulent joinder).Show More Across these contexts, though, a firm understanding of what counts as settled law has proved chimerical.

Given the definitional morass, one might conclude that “‘settled law’ is just a euphemism.”22 22.Ilya Somin, Why “Settled Law” Isn’t Really Settled—and Why That’s Often a Good Thing, Reason: The Volokh Conspiracy (Sept. 9, 2018, 3:57 PM),­2018/09/09/why-settled-law-isnt-really-settled-and/ [­SU-3N4A].Show More On this view, the term is so capacious as to become meaningless, conveying nothing useful about the weight that precedent deserves or the conditions (if any) under which a court should overrule it.

Our principal goal is to show that settled law does coherent and powerful work, even though it resists a single, overarching definition. In fact, settled law makes sense only when one appreciates that it comprises several distinct notions that do not share a common attribute. A more precise understanding of this hydra-like term has the power to clarify doctrine and improve political rhetoric. What seem like conceptual oddities in a number of doctrines actually make good theoretical sense when viewed through the lens of settled law. Moreover, settled law can play a meaningful role in confirmation hearings, but only if legal actors fully grasp its multifaceted nature. It offers a productive way to explore how politicians, judicial nominees, and the general public understand the judicial role, including how the obligations of Supreme Court Justices differ from those of lower-court judges.

We begin in Part I by differentiating between two concepts that we call controlling law and settled law. Controlling law essentially refers to the concept of binding precedent, including in its most conspicuous manifestation: an inferior court’s duty to follow the precedents of superior courts. Although one might think of controlling law as a species of settled law, we argue that the two are actually distinct ideas that address very different questions and are, at most, only tangentially related. Much of the confusion about settled law, in fact, stems from conflating these concepts. Not allowing discussions of settled law to revert into the familiar language of controlling law is thus a critical first step.

Part II demonstrates that settled law is not just an empty euphemism, even though it doesn’t embrace a single idea. In fact, settled law makes sense only when one appreciates that it comprises several notions that do not share a common attribute.

On an intuitive level, the starkest divide lies between normative and descriptive claims about settled law. For example, someone might classify Brown as settled law, normatively, because it achieved the right substantive result. Or, irrespective of Brown’s fundamental correctness, one might view it as descriptively settled because everyone recognizes that it’s here to stay. Even within these broad categories, though, variation abounds. For example, calling Brown normatively settled could mean that the decision was consonant with the original meaning of the Fourteenth Amendment or, alternatively, that it achieved a socially desirable outcome by advancing the cause of racial justice. Calling Brown descriptively settled could mean that the Supreme Court has left the precedent undisturbed for more than fifty years, that a future Court is unlikely to overrule it, that principles of stare decisis have effectively entrenched it, or that it has achieved wide popular acceptance.

We bring theoretical rigor to this intuition about the descriptive-normative divide by overlaying it with seminal jurisprudential theories: formalism, realism, and legal process theory. Based on these theories, we develop a taxonomy of five concepts that “settled law” can embrace.

The first two concepts derive from legal formalism.23 23.See, e.g., Warren Sandmann, The Argumentative Creation of Individual Liberty, 23 Hastings Const. L.Q. 637, 645 (1996) (“Legal formalism . . . is in its many guises one of the more dominant approaches to judicial decisionmaking.”).Show More As a normative matter, a formalist insists that law is settled when it has achieved the demonstrably “right” result based on the law’s internal logic.24 24.Thomas C. Grey, Langdell’s Orthodoxy, 45 U. Pitt. L. Rev. 1, 8 (1983).Show More But from a descriptive perspective, a formalist might accept that law is settled—even if it has not reached the objectively correct result—when the concerns of stare decisis, such as reliance, predictability, and basic fairness, are paramount.25 25.See Randy J. Kozel, Settled Versus Right: Constitutional Method and the Path of Precedent, 91 Tex. L. Rev. 1843, 1874 (2013) (citing Lawrence B. Solum, The Supreme Court in Bondage: Constitutional Stare Decisis, Legal Formalism, and the Future of Unenumerated Rights, 9 U. Pa. J. Const. L. 155, 186, 192–95 (2006)) (describing a “neoformalist” model of stare decisis in which even judicial “mistakes” can and should create binding precedents, if they are decided through a formalistic process of reasoning); Amy Coney Barrett, Originalism and Stare Decisis, 92 Notre Dame L. Rev. 1921 (2017) (describing Justice Scalia’s approach to the tension between the value of stare decisis and a formalistic, originalist reading of the Constitution).Show More

The next two concepts of settled law draw on the legal realist school.26 26.SeeFrederick Schauer, Legal Realism Untamed, 91 Tex. L. Rev. 749, 749 (2013) (“Legal Realism is conventionally understood, in part, to question legal doctrine’s determinacy and positive law’s causal effect on judicial decisions.”).Show More Descriptively, a realist regards law as settled when it faces no material threat of reversal.27 27.As we build out below, it essentially constitutes an exercise in Holmesian Prediction Theory. SeeO.W. Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 460–61 (1897).Show More Normatively, a realist will insist that law is settled only when it has achieved the “right” result, but she understands that idea very differently than a formalist does. The correct result for a legal realist corresponds to some external frame of reference, such as utility, efficiency, or social justice.28 28.For the foundational realist works encouraging an interdisciplinary approach to law, see Felix S. Cohen, Transcendental Nonsense and the Functional Approach, 35 Colum. L. Rev. 809 (1935); Jerome Frank, Law & the Modern Mind (Transaction Publishers 2009) (1930); Karl N. Llewellyn, Law and the Social Sciences—Especially Sociology, 62 Harv. L. Rev. 1286 (1949); Roscoe Pound, The Scope and Purpose of Sociological Jurisprudence (pts. 1 & 3), 24 Harv. L. Rev. 591 (1911), 25 Harv. L. Rev. 489 (1912).Show More

The fifth and final concept of settled law draws on legal process theory, which focuses on a legal decision’s methodological process rather than its substantive outcome.29 29.The seminal tome of the legal process school is Henry M. Hart, Jr. & Albert M. Sacks, The Legal Process: Basic Problems in the Making and Application of Law (William N. Eskridge, Jr. & Philip P. Frickey eds., 1994). For an overview of the related concept of procedural justice, see Lawrence B. Solum, Procedural Justice, 78 S. Cal. L. Rev. 181 (2004).Show More For the legal process theorist, law is settled if and when a duly constituted court reaches a decision through an appropriate methodology, and within this framework the descriptive and normative perspectives essentially become inseparable.

A simple illustration might help reify these five theoretical concepts. Consider the question: “Is Marbury v. Madison settled law?” Nearly everyone would say “yes,” but Table 1 identifies more precisely the five different ideas that someone could intend to communicate when asserting that Marbury is settled.

Table 1: The Taxonomy of Settled Law


Marbury v. Madison is settled law.

Normative Formalism

Marbury arrived at the objectively correct understanding of constitutional law.

Descriptive Formalism

Principles of stare decisis require continued adherence to Marbury.

Descriptive Realism

There is no material chance that the Supreme Court will overrule Marbury in the near future.

Normative Realism

Marbury achieved a desirable outcome in light of its intra- and extra-legal consequences.

Legal Process

Marbury merits continued adherence because it was issued by a duly constituted court employing an appropriate methodology.

In Part III, we show why developing a clearer understanding of settled law is far more than an academic exercise. At the intensely practical level, settled law suffuses a diverse array of doctrines, and failing to appreciate how it functions has led to pervasive confusion and mistakes. Our principal example comes from the qualified immunity context. Although courts often cast the relevant inquiry in terms of controlling law—whether binding precedent has clearly established that a particular right exists—this approach has invited a host of anomalies and errors. Instead, we argue that viewing qualified immunity through the lens of settled law makes much more sense doctrinally and normatively. Moreover, understanding qualified immunity as turning on settled law—specifically, two of the taxonomy’s five concepts—alleviates nearly all of the current conceptual problems and has the potential to refocus courts on the heart of the inquiry.

Finally, we argue that a more nuanced understanding of settled law can enhance legal dialogue, particularly the conversation about judicial nominations. Too often legal actors talk past one another because they use “settled law” to convey different ideas, and that in turn can lead to unfounded allegations of bad faith. On this level, the taxonomy is not a panacea; far from it. But greater conceptual clarity about settled law can train attention on the debates that truly matter rather than a bewitching semantic game.