Congress’s Constitution:

Legislative Authority and the Separation of Powers.

By Josh Chafetz. New Haven: Yale University Press, 2017. Pp. 449. $45.00.

Contemporary scholarship and jurisprudence concerning the Constitution’s separation of powers is characterized by sharp disagree­ment about general theory and specific outcomes. The leading theories diverge on how to model the motives of institutional actors; on how to weigh text, history, doctrine, and norms; and on whether to characterize the separation of powers system as abiding in a stable equilibrium or as enthralled by transformative convulsions. Congress’s Constitution—an important contribution to theorizing on the separation of powers—provides a platform to step back and isolate these important, if not al­ways candidly recognized, disputes about the empirical and normative predicates of separation of powers theory—predicates that can be use­fully grouped under the rubric of “separation of powers metatheory.” Unlike much other work in the field, Congress’s Constitution directly identifies and addresses the three key metatheoretical questions in play when the separation of powers is theorized. This Review analyzes how it grapples with those profound challenges and tries to articulate a de­scriptively well fitted and normatively compelling account of our federal government. Evaluating Congress’s Constitution from this vantage point offers valuable opportunities for considering the state and direc­tion of academic theorizing on the separation of powers more broadly.


    1. Separation Models
    2. Balance Models
    3. Exogenous Models
    1. The Theory of Constitutional Politics
    2. From Theory to Case Study
    1. The Question of Motive
    2. The Sources of Law
    3. The Choice Between Equilibrium and Change Models
    4. The Uses of Metatheory
    1. Motive
      1. Motives in Time
      2. Public Competition for Power
      3. Public-Facing Competition and Judicious Outcomes
    2. Sources of Law
      1. How Law Arises
      2. Law, Politics, and Conventions
    3. Equilibrium and Change



Contemporary academic writing on the Constitution’s separation of powers 1 The separation of powers comprises bilateral interactions between Congress, the executive, and the federal judiciary, in addition to more complex dynamics pairing two branches against one. References to the separation of powers are often vague in the sense of failing to specify which of these interactions is picked out. This Review, consistent with the scope of the book under consideration, focuses upon the bilateral Congress–executive dynamic. is thick on the ground, but prolixity has not precipitated much by way of consensus. Issues that divided jurists in the early twentieth cen­tury—such as how much control the president must have over subordi­nate officials; when Congress can delegate policy decisions; and whether and how federal jurisdiction can be scrapped—continue to elicit heated division among scholars as well as jurists. The resulting cacophony comes not only from sharp local controversies but also from the fact that schol­ars disagree about the basic terms of the debate. They are at odds, that is, over a number of rather basic theoretical premises of the separation of powers. One school of scholarly theory emphasizes the categorical sep­aration and autonomous functioning of each branch, chiefly by relying on historical materials to precisely fence off each branch’s domain. 2 See, e.g., Steven G. Calabresi & Kevin H. Rhodes, The Structural Constitution: Unitary Executive, Plural Judiciary, 105 Harv. L. Rev. 1153, 1165–68 (1992) (“Unitary executive theorists read [the Article II Vesting Clause], together with the Take Care Clause, as creating a hierarchical, unified executive department under the direct control of the President. They conclude that the President alone possesses all of the executive power . . . .” (footnotes omitted)); Gary Lawson, The Return of the King: The Unsavory Origins of Administrative Law, 93 Tex. L. Rev. 1521, 1538 (2015) (reviewing Philip Hamburger, Is Administrative Law Unlawful? (2014)) (“The Constitution, however, vests ‘[t]he’ executive power—meaning all of the executive power—in the President. There is no executive power remaining to be vested in anyone else.” (alteration in original) (footnote omitted)); Martin H. Redish & Elizabeth J. Cisar, “If Angels Were to Govern”: The Need for Pragmatic Formalism in Separation of Powers Theory, 41 Duke L.J. 449, 454 (1991) (“[T]he Court’s role in separa­tion of powers cases should be limited to determining whether the challenged branch action falls within the definition of that branch’s constitutionally derived powers—executive, leg­islative, or judicial.”). Another school of thought, more heavily influenced by political theories of mixed government, celebrates instead a more disorderly, dynamic interaction within and between branches that is said to alchemize a beneficial check­ing and balancing. 3 See, e.g., Laura S. Fitzgerald, Cadenced Power: The Kinetic Constitution, 46 Duke L.J. 679, 779 (1997) (characterizing the separation of powers principle “less [as] a device for policing institutional boundaries and more like a standard for preserving in that debate the qualities of participation and accountability” and “mark[ing] the cadence for the People’s argument among themselves over time”); Jon D. Michaels, Of Constitutional Custodians and Regulatory Rivals: An Account of the Old and New Separation of Powers, 91 N.Y.U. L. Rev. 227, 227 (2016) [hereinafter Michaels, Of Constitutional Custodians] (underscoring “the multidimensional nature of administrative control in which the consti­tutional branches . . . and the administrative rivals . . . all compete with one another to influence administrative governance”). A third approach seems to step out­side the law entirely to predict the optimal assignment of institutional author­ity between the branches through an abacus-rattling calculus of how to maximize social welfare. 4 See, e.g., Eric A. Posner & Adrian Vermeule, The Executive Unbound: After the Madisonian Republic 18 (2010) [hereinafter Posner & Vermeule, Executive Unbound] (arguing that separation of powers has “collapsed” because of Madison’s incorrect assumption that “individual ambitions of government officials would cause them to support the power of the institutions they occupy”). And these are only the most prominent ap­proaches; yet other, more retail contributions about many discrete, local questions of law abound.

As below, so above. For the Supreme Court’s treatment of legislative–executive relationships has also “cycled” between fidelities to diverse and incompatible theories since the 1920s, generating an unstable, unpre­dictable, and oft-criticized jurisprudence. 5 For a comprehensive account of the relevant jurisprudence that emphasizes the Court’s movement between different theories of the separation of powers, see generally Aziz Z. Huq & Jon D. Michaels, The Cycles of Separation-of-Powers Jurisprudence, 126 Yale L.J. 346 (2016). For a more general account of prescriptive legal theories that offers inter­esting hypotheses about the causes and inevitability of instability, see generally Jeremy K. Kessler & David E. Pozen, Working Themselves Impure: A Life Cycle Theory of Legal Theories, 83 U. Chi. L. Rev. 1819 (2016). What has been endless fuel for the dark mills of academic disputation has yielded little benefit for a public concerned with orderly government under constant rules. It would be one thing if these disputes were localized within one facet of the separation of powers. But they are not. All three relationships that link Congress, the presidency, and the federal courts have seen their por­tion of fitful disagreement. In what follows, I focus largely on congressional–executive relationships—in part because it’s what the book reviewed herein concerns, and so is properly my focus too—but the other margins of the separation of powers have not been starved for controversy. The judicial construction of Article III boundaries, for example, still oscillates between a categorical approach that emphasizes separation and a more fluid approach that celebrates balance and checking effects. 6 Compare Stern v. Marshall, 564 U.S. 462, 475, 484 (2011) (holding that assign­ment “counterclaims by the estate against persons filing claims against the estate” in an Article I bankruptcy proceeding violated Article III), with Wellness Int’l Network, Ltd. v. Sharif, 135 S. Ct. 1932, 1954–55 (2015) (allowing litigant consent to cure an Article III defect in an Article I bankruptcy adjudication). See also Exec. Benefits Ins. Agency v. Arkison, 134 S. Ct. 2165, 2172–73 (2014) (holding that an Article III flaw is cured if the bankruptcy judge’s ruling is treated as a set of proposed findings of fact and conclusions of law to be evaluated de novo by a district court). For a comprehensive analysis of the Article III ques­tion in the context of bankruptcy, see generally Anthony J. Casey & Aziz Z. Huq, The Article III Problem in Bankruptcy, 82 U. Chi. L. Rev. 1155 (2015).

There are a number of competing theories of the separation of powers. These theories are reasonably explicit in the scholarship. They are some­what more submerged in the jurisprudence. Like theories formulated for other constitutional domains, 7 For an excellent and self-conscious example, see Silas J. Wasserstrom & Louis Michael Seidman, The Fourth Amendment as Constitutional Theory, 77 Geo. L.J. 19, 21 n.6 (1988) (cataloging “attempts to formulate a general theory of the fourth amendment in light of modern constitutional thought” before their own). they purport to offer prescriptive guidance across plural strands of doctrine. Yet they also seem continuous with a larger domain of “constitutional theory.” This raises the question of whether the task of theory building for the separation of powers is mean­ingfully distinct from the enterprise of constitutional theory writ large. I think there is good reason to indeed treat it as a distinct subject—and want to explain why this is so before going any further.

A good place to start thinking about the relationship of separation of powers theory to constitutional theory more generally is Professor David Strauss’s useful definition of constitutional theory as “an effort to justify a set of prescriptions about how certain controversial constitutional issues should be decided.” 8 David A. Strauss, What Is Constitutional Theory?, 87 Calif. L. Rev. 581, 582–83 (1999) [hereinafter Strauss, Constitutional Theory]. One might reasonably object that Strauss is far too optimistic about the extent and nature of disagreement in constitutional theory writ large, but such an argument is orthogonal to my point here. There are other defini­tions of “constitutional theory,” many of which focus on the general question of how the Constitution as a whole should be glossed and applied by courts. See, e.g., Richard H. Fallon, Jr., How to Choose a Constitutional Theory, 87 Calif. L. Rev. 535, 537 (1999) [hereinafter Fallon, How to Choose] (describing constitutional theories as concerning “the nature of the United States Constitution and how judges should interpret and apply it”); Richard A. Posner, Against Constitutional Theory, 73 N.Y.U. L. Rev. 1, 1 (1998) (“Con­stitutional theory, as I shall use the term, is the effort to develop a generally accepted theory to guide the interpretation of the Constitution of the United States.”). Strauss explains that such a theory will necessarily “dra[w] on the bases of agreement that exist within the legal culture and try[] to extend those agreed-upon principles to decide the cases or issues on which people disagree.” 9 Strauss, Constitutional Theory, supra note 8, at 582. But disagreement in the separation of powers context does not quite fit Strauss’s description. Theoretical disagreement in this domain remains lively—or perhaps hell­ishly Sisyphean—precisely because there is pervasive and deep disagree­ment about specific “bases” and “principles” upon which any theoretical account with prescriptive force must necessarily rest. And while constitutional theory writ large provides some of the basic terms of this theoretical discord, the separation of powers has its own bedeviling and disagreement-fostering problems.

Indeed, immediate debate about specific separation of powers issues can be scraped away, like so many coats of thick-lathered varnish, to uncover a deeper layer of roiling disagreement about the empirical and normative premises that support different theories. I call this underlying layer of disagreements “metatheoretical” by analogy to the philosophical field of metaethics. The latter field “is the attempt to understand the metaphysical, epistemological, semantic, and psychological[] presupposi­tions and commitments of moral thought, talk, and practice.” 10 Geoff Sayre-McCord, Metaethics, Stanford Encyclopedia of Philosophy (Jan. 26, 2012), (on file with the Columbia Law Review). Corre­spondingly, a metatheoretical disagreement in the law turns on the meaning, analytic robustness, normative salience, or empirical validity of a term that appears in one or more theories of the separation of powers. Theory predicts or directs outcomes in specific cases. Metatheory, in contrast, does not directly specify the normative framework tendered for the analysis and resolution of specific legal disputes. It is instead con­cerned with the detailing of necessary and foundational terms used within, and shared across, one or more theories. 11 A handful of scholars have employed the term “metatheory” in the constitutional context, albeit in a different sense than I do. Michael Dorf describes questions about how to evaluate constitutional theories as metatheoretical. Michael C. Dorf, Create Your Own Constitutional Theory, 87 Calif. L. Rev. 593, 597 (1999). Garrick Pursley uses the term to refer rather vaguely to “the question of theory assessment in legal theory.” Garrick B. Pursley, Metatheory, 47 Loy. U. Chi. L.J. 1333, 1337 (2016); see also Joseph William Singer, The Player and the Cards: Nihilism and Legal Theory, 94 Yale L.J. 1, 16 (1984) (suggesting that metatheories are needed to “tell us where to draw the line” within legal theories). I use the term in a simpler fashion to capture debate about terms or ideas shared across differ­ent jurisprudential theories; unlike Dorf, I do not use it to refer to procedures or criteria for choosing between such theories. Even if we disagree in theory, at least we can articulate the terms of that disagreement—or at least so I hope.

A central claim advanced in this Review is that separation of powers metatheory is characterized by a small number of predictable and persistent questions. One way to understand and adjudicate between the seemingly interminable debates about the separation of powers is to notice that theorists’ disagreements in fact hinge upon a series of deep divergences about basic descriptive or normative terms. This metatheo­retical level of disagreement is often obscured by superficial accounts of theories as (say) formalist or functionalist. We would do better, I suggest, to ignore the surface-level disagreements and concentrate on how best to answer metatheoretical questions. A secondary claim that flows from the first one is that by accounting for such metatheoretical disputes, we can get a better grasp on the positions and contributions of new scholarship.

I see three metatheoretical questions running through most current debates on the separation of powers—sometimes recognized, sometimes working sub silentio. 12 I should emphasize that my typology is based on a reading of separation of powers scholarship. Others reading the same scholarship might come to different conclusions, but since part of my purpose here is to stimulate thinking about the metatheoretical predi­cates of this scholarship, I am not especially troubled by that possibility. I call these disputes about motives, sources of law, and equilibrium conditions. Divergent theories of the separation of powers can be usefully sorted, in my view, by focusing on how they answer these three deep questions.

First, there is dispute about how the motives of relevant institutional actors should be modeled (if at all) for the purpose of defining horizon­tal relationships between the branches. Motive is often imagined to count only when individual rights, rather than constitutional structure, are at stake. But this is quite wrong. It matters an awful lot whether judges, legislators, and bureaucrats are moved by law, naked partisan ambition, loyalty to their institution, or the wish to defend their turf. Indeed, the motive question in structural constitutional law is quite different in character from the motive question in most rights analysis. The two should not be conflated. And without an understanding of officials’ mo­tives, it’s hard to see how any theory of the institutions those officials control gets off the ground.

Second, there is a question of what sources of law—text, preratification history, postratification practice, judicial precedent, or first-order normative reasoning—exist and also how they should be weighed and prioritized. The sources-of-law debate in the separation of powers, I should concede, has strong continuities with constitutional theory debates more generally. 13 For a synoptic and fair-minded view of the relevant approaches, see Sotirios A. Barber & James Fleming, Constitutional Interpretation: The Basic Questions 59–65 (2007). But the identification of the sources of law in separation of powers cases implicates distinc­tive problems not observed elsewhere. The hermeneutic heft of historical “gloss” comprising the past acts of officials wielding state power, 14 For a definition of historical gloss, see Curtis A. Bradley & Trevor W. Morrison, Historical Gloss and the Separation of Powers, 126 Harv. L. Rev. 411, 418–19 (2012) [hereinafter Bradley & Morrison, Historical Gloss] (discussing the use of “historical gloss” in the separation of powers context). for instance, is of a differ­ent magnitude in the separation of powers context from the weighing of prior state practice in, say, debates about whether Brown v. Board of Education 15 347 U.S. 483 (1954). was rightly decided.

Finally, a metatheory of the separation of powers will also need to articulate an idea about how institutions either remain at equilibrium over time or how they change within constitutional contours. It needs, that is, a theory of change and stability over time. Unlike the rights-related provi­sions of the Constitution, the separation of powers describes a complex and novel system of interlocking institutions. The national government has been conflicted with new legal difficulties, new geostrategic opportu­nities, and new governance challenges. 16 For example, the Louisiana Purchase itself presented an early, and quite serious, constitutional question of the separation of powers—one that, in a striking harbinger of later applications of the political question doctrine to structural constitutional questions, was never adjudicated by a federal court. Sanford Levinson & Bartholomew H. Sparrow, Introduction, in The Louisiana Purchase and American Expansion, 1803–1898, at 1, 3 (Sanford Levinson & Bartholomew H. Sparrow eds., 2005). On the internal political challenges created by conflicts over slavery’s expansion, see Barry R. Weingast, Designing Constitutional Stability, in Democratic Constitutional Design and Public Policy: Analysis and Evidence 343, 357–58 (Roger D. Congleton & Birgitta Swedenborg eds., 2006); Aziz Z. Huq, The Function of Article V, 162 U. Pa. L. Rev. 1165, 1185 (2014). As early as the first decade of the nineteenth century, the prospect of radical change in the scope of the nation—and hence the operation of the national legislature and the electoral college—was squarely in view. 17 Justifying Lewis and Clark’s literal path-marking expedition west, Thomas Jefferson told Congress on January 18, 1803, that “[t]he interests of commerce place the principal object [of the expedition] within the constitutional powers and care of Congress . . . that it should incidentally advance the geographical knowledge of our own continent, cannot but be an additional gratification.” Thomas Jefferson, Message to Congress (Jan. 18, 1803), http:// [].
The branches today are not the branches of 1789. They are a complex accretion of reforms, interbranch conflicts, aftershocks of military and economic shocks, and more. Any theory in this domain has to reconcile these changes with the assumption that this is a unitary phenomenon—the separation of powers, in the singular not the plural—that persists over time. Why do we assume that the relations of the three branches of the federal government can be cap­tured and explained in a single conceptual breath? Why do we think that a network of complex and evolving institutional relationships is governed by a single covering law? These difficult questions, I should add, cannot be re­solved merely by changing the level of generality at which the constitu­tional text is construed. 18 Laurence H. Tribe & Michael C. Dorf, Levels of Generality in the Definition of Rights, 57 U. Chi. L. Rev. 1057, 1057–58 (1990) (“[A]t what level of generality should the Court describe the right previously protected and the right currently claimed? The more abstractly one states the already-protected right, the more likely it becomes that the claimed right will fall within its protection.”). The role of time in the separation of powers cannot be accommodated, that is, simply by ratcheting up the abstraction with which one views the text—at least not if one wants an analytically useful answer.

The publication of Professor Josh Chafetz’s monograph Congress’s Constitution: Legislative Authority and the Separation of Powers is the immediate occasion for elaborating a meta­theoretical lens onto the separation of powers and for applying that lens. Congress’s Constitution makes two main contributions to the separation of powers scholarship. The book first tenders its own theory of how the political branches interact, called the “constitutional politics” account (p. 16). In brief, the constitutional-politics account posits that formal, constitutional rules settle relatively little about the scope of relative insti­tutional powers. Instead, Chafetz argues, the au­thority of officials inhabiting national institutions is a function of public-focused claim-making and a residue of previous historical practice. Hence, “politics”—in the sense of debating, arguing, and bargaining before the public—and historical experience infuse and inform the scope of institu­tional authority and channel the results of institutional confrontations.

The book’s second contribution is, commensurate with its title, a recapitulation of historical conflicts over the national Congress’s powers. A central theme is that “there is nothing inherent in Congress’s constitu­tional place that dooms it to play second (or third) fiddle” (p. 42)—an observation that would surely not have surprised those who located Congress’s powers in Article I of the Constitution, even if that aperçu pur­ports to flout fashion today. Unlike the mine run of constitutional scholar­ship, however, Chafetz does not focus on Congress’s Article I, Section 8, powers to enact substantive, regulatory policy. Rather, he trains upon a suite of less studied instruments of investigation, deliberation, speech, and self-government. In a sequence of chapters each focused on a distinct legislative power, Chafetz provides rich portrayals of those powers’ development. Each is an enjoyable read. Of particular interest is the manner in which he explores the English parliamentary forbearers to legislative powers such as Speech and Debate immunity (pp. 203–25), authority over internal disciplinary matters (pp. 235–59), and the legisla­tive power of contempt (pp. 158–71). 19 The latter is extensively discussed in an insightful precursor article. See Josh Chafetz, Executive Branch Contempt of Congress, 76 U. Chi. L. Rev. 1083 (2009). Chafetz is adept at covering decades, even centuries, of institutional ground in a narrow compass of pages. Few readers will finish the book without profiting from his synoptic vision and detailed knowledge of legislative and constitutional history. As a book designed for a reasonably informed professional audience, on an important but oft-neglected element of constitutional law, Congress’s Constitution is to be applauded. My disagreement with its theoretical premises and meta­theoretical implications should not be confused for a more general indict­ment of the quality of Chafetz’s analytic work.

Historically inclined, steeped with doctrinal intricacies, and intimate with much of the relevant political-science literature, Chafetz is in other words a sophisticated guide to separation of powers dynamics. He focuses on Congress–executive relations, which is the locus of much current scholarly preoccupation, 20 For an earlier generation of scholars who came of age in an era of relative détente between the president and Congress, the question of legislative ability to eliminate federal court jurisdiction loomed larger. See Richard H. Fallon, Jr. et al., Hart and Wechsler’s The Federal Courts and the Federal System 295–411 (7th ed. 2015) (summarizing various permutations of the debate). A decade ago, Congress’s effort to eliminate federal court jurisdiction over habeas corpus petitions filed by alleged enemy combatants intermittently rekindled the debate. Richard H. Fallon, Jr., Jurisdiction-Stripping Reconsidered, 96 Va. L. Rev. 1043, 1046–47 (2010) [hereinafter Fallon, Jurisdiction-Stripping]. but finds fresh angles to explore. His attention to Article I institutions is welcome. With a handful of exceptions, 21 See, for example, the essays in The Least Examined Branch: The Role of Legislatures in the Constitutional State (Richard W. Bauman & Tsvi Kahana eds., 2006), and Elizabeth Garrett & Adrian Vermeule, Institutional Design of a Thayerian Congress, 50 Duke L.J. 1277, 1279 (2001), for recommendations intended to improve Congress’s deliberations concerning constitutional issues. separation-of-powers scholars tend to be too beguiled by the mysteries of the executive branch. 22 Representative examples (without endeavoring to be exhaustive) include Posner & Vermeule, Executive Unbound, supra note 4; Curtis A. Bradley & Trevor W. Morrison, Presidential Power, Historical Practice, and Legal Constraint, 113 Colum. L. Rev. 1097 (2013) [hereinafter Bradley & Morrison, Presidential Power]; Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245 (2001). It is worth noting that there is, of course, a large literature concerning how Congress works in the political science domain and a growing literature on congressional dynamics by scholars of the legislative process. For a recent example in these pages, see Abbe R. Gluck, Anne Joseph O’Connell, and Rose Po. Unorthodox Lawmaking, Unorthodox Rulemaking, 115 Colum. L Rev. 1789 (2015). To the extent that Chafetz is read to say that no legal scholars have taken Congress seriously, his claim is plainly an un­tenable one. But he can be read more narrowly to say more mod­estly that theorists of the separation of powers writ large have not taken Congress suffi­ciently seriously as a constitu­tional actor. (It is, perhaps, no coincidence that more law profes­sors spend time in the latter than in Congress before relocating to the academy.) This scholarly asymmetry tracks a more general drift of effectual governance authority from Article I to Article II actors across the twentieth century. 23 One might suspect it also has something to do with the fact that more legal scholars have experience in the executive branch rather than Congress. A less charitable reading might be that more legal scholars expect employment in the executive branch—or a nomi­nation nod by someone in the executive branch. In an era of skepticism of legislatures, and perhaps of deliberative democracy more generally, Chafetz usefully rows against the zeitgeist by taking Congress seriously as a constitutional actor. Moreover, his account of the separation of powers warrants close and careful scrutiny—both for the commendable ambition to explicate its metatheory systematically and also for its specific content. And even where he goes astray, as I think he does at moments, Chafetz remains a judicious and careful analyst. The rewards of a close reading of Congress’s Constitution, in other words, are ample.

My Review has four Parts, two about theoretical debates and two concerning metatheoretical debates. The first Part provides context by offering a simplified topography of current theoretical debates over the separation of powers. The second locates Chafetz in this analytic topol­ogy. Part III posits three axes of metatheoretical debate that striate the current literature. The final Part applies this typology to Congress’s Constitution to offer a case study of why a focus upon separation of powers metatheory, as well as theory, is useful. (Hence, readers most interested in engaging with Chafetz’s work can focus on Parts II and IV, while those interested in the more general idea of metatheory for the separation of powers can look to Parts I and III.) The exercise also generates a series of insights into the strengths, weaknesses, and lacunae in Chafetz’s work and about the future of separation of powers theory.

I. The Separation of Powers (In Theory)

It is a truth grudgingly acknowledged by all 1L Con Law students that judges and scholars have no common sense of what the phrase “sep­aration of powers” means. A simplified cartography of disagreements that bedevil students and judges alike is useful here for two reasons. First, by mapping the landscape of theoretical disagreement, it is possible to better grasp its underlying wellsprings—that is, the pivotal questions of separation of powers metatheory. Second, with accounts of both theory and metatheory in hand, it is possible to position Congress’s Constitution in the scholarly landscape and thereby to ascertain the precise nature of its contribution by considering how it tracks, or deviates from, earlier accounts of the separation of powers.

Scholars typically adopt one of three general accounts of the inter­actions between the three branches of the federal government. 24 This summary focuses on U.S. constitutional scholars and excludes comparative constitutional scholarship. For a pathbreaking comparative piece that identifies “the model of constrained parliamentarianism as the most promising framework for future develop­ment of the separation of powers,” see Bruce Ackerman, The New Separation of Powers, 113 Harv. L. Rev. 633, 640 (2000). These can be labeled the separation, balance, and exogenous models. This Review offers necessarily stylized accounts, which generalize away from particulars and try to grasp shared themes. Whereas these models also inflect the case law, I focus here on scholarly presentations of the relationship between the executive and Congress with the aim of zeroing in on Congress’s Constitution’s contribution. 25 The first two of these models are sometimes labeled “formalist” and “functionalist.” See Peter L. Strauss, Formal and Functional Approaches to Separation-of-Powers Questions—A Foolish Inconsistency?, 72 Cornell L. Rev. 488, 489 (1987). I avoid these labels because I want to juxtapose both with a third, even more functionalist, model. Further, as others have observed, the labels of formalism and functionalism are somewhat misleading insofar as so-called formalists often have a functional account of their separation principle. V. F. Nourse, Toward a New Constitutional Anatomy, 56 Stan. L. Rev. 835, 899 (2004).

A. Separation Models

Separation models of the separation of powers doctrine understand each branch of government as a distinctive and stand-alone entity wield­ing a defined, delimited set of powers. 26 Readers should be aware that the first paragraph of each of the following sections comprises my own synthesis and summary of the three different positions. These opening paragraphs are light on citations because they are synthetic and generalizing. The following paragraphs then set out particular positions in more detail. They offer a Newtonian model of branches as discrete zones of authority. These interact much as billiard balls on the blaze do. In its simplest and most elegant exposition, the model holds that “Congress’ grants of legislative powers must enable it to legislate, the President’s grant of the executive power must enable him to execute all federal laws, and the federal judiciary’s grant of the judicial power must enable the federal courts to decide certain cases and controversies.” 27 Steven G. Calabresi & Saikrishna B. Prakash, The President’s Power to Execute the Laws, 104 Yale L.J. 541, 544 (1994). Such models are often derived from textual exegesis of the original meaning or understanding of the first three Articles of the Constitution, and from historical practice. 28 See, e.g., id. at 546 (starting with “the original meaning of the words of the con­stitutional text that the Framers actually wrote”); Steven G. Calabresi, The Vesting Clauses as Power Grants, 88 Nw. U. L. Rev. 1377, 1381 (1994) (noting a similar analysis of the term “vest”). But they need not be so rooted. 29 See, e.g., Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 Colum. L. Rev. 1, 4 (1994) (defending a robust account of presidential authority based on “the best reading of the framers’ structure translated into the current, and radically transformed, context,” and explicitly considering and rejecting original meaning as a source).

Leading work in this field takes what political scientists once deri­sively termed a “literary” approach. 30 See, e.g., Richard E. Neustadt, Presidential Power and the Modern Presidents 37 (1990) (“The probabilities of power do not derive from the literary theory of the Consti­tution.” (emphasis omitted)). One of its most voluble and cogent exponents, Professor Saikrishna Prakash, begins with eighteenth-century dictionary definitions of the powers vested by the first Articles of the Constitution as a foundation for specifying powers granted to each branch. 31 Saikrishna Prakash, The Essential Meaning of Executive Power, 2003 U. Ill. L. Rev. 701, 716 [hereinafter Prakash, Essential Meaning] (quoting 1 Samuel Johnson, A Dictionary of the English Language 684 (4th ed. 1773)); see also Saikrishna Bangalore Prakash, Imperial from the Beginning: The Constitution of the Original Executive 1–11 (2015) [hereinafter Prakash, The Constitution of the Original Executive] (exploring the avail­ability of a lin­guistically rooted approach to the exegesis of various government powers). The executive, he posits, possesses “all of those rights, powers, and privileges commonly associated with a chief executive vested with the executive power, subject to the many exceptions and limitations enumer­ated in Article II and elsewhere in the Constitution.” 32 Prakash, Essential Meaning, supra note 31, at 714. For a similar analysis of foreign affairs, see Saikrishna B. Prakash & Michael D. Ramsey, The Executive Power over Foreign Affairs, 111 Yale L.J. 231, 253 (2001) (“Executive power, as commonly understood in the eighteenth century, included foreign affairs powers.”). But cf. John Hart Ely, On Constitutional Ground 149 (1996) (“The Constitution gives the president no general right to make foreign policy. Quite the contrary[:] . . . virtually every substantive constitutional power touching on foreign affairs is vested in Congress.”). Prakash recently applied the same approach to Congress. In an article on military powers, he categorized the legislature’s war and emergency powers as “sweeping” through a close examination of the “basic structure” of the Constitution, the parallel between that structure and precursor models among the early American states, and the extensive textual specifications in Article I of “powers to declare war, raise an army and navy, and regulate both, . . . [all of which] implied that Congress could continue to pass laws needed to defeat foreign enemies.” 33 Saikrishna Bangalore Prakash, The Sweeping Domestic War Powers of Congress, 113 Mich. L. Rev. 1337, 1367 (2015) [hereinafter Prakash, Sweeping Domestic War Powers]; cf. Prakash & Ramsey, supra note 32, at 241 (“Congress enjoys only certain specific foreign affairs powers outlined in Article I, Section 8.”). Writing about the courts, Professor Caleb Nelson strikes a kindred separationist chord when he proposes that “Article III . . . strongly implies that neither Congress nor entities within the executive branch can exercise ‘[t]he judicial Power of the United States,’” but vests it “in true federal courts.” 34 Caleb Nelson, Adjudication in the Political Branches, 107 Colum. L. Rev. 559, 565 (2007) (alteration in original); id. at 571–72 (defining the “‘judicial’ power” in terms of “the kinds of legal inter­ests that were at stake,” and distinguishing public and private rights). At the logical limit of this approach is Dean John Manning’s textualist reduction—an account that strips most of the conceptual and historical meat from the separation of powers and urges instead attention to the denuded bones of constitu­tional text. 35 John F. Manning, Separation of Powers as Ordinary Interpretation, 124 Harv. L. Rev. 1939, 1948–49 (2011) [hereinafter Manning, Separation] (arguing for “a clause-centered approach” that rejects any “freestanding separation of powers doctrine”); accord John F. Manning, Foreword: The Means of Constitutional Power, 128 Harv. L. Rev. 1, 55 (2014) (“[T]o the extent one can discern the purposes underlying federalism and separa­tion of powers, those purposes are vague, numerous, unranked, and often self-contradictory. Because neither doctrine provides firm answers in the abstract, the particulars of each almost invariably require the creation, rather than the excavation, of constitutional meaning.”). Manning styles his intervention as diverging from the approaches of both functionalists and formalists. Although he and the scholars he denominates as formalists may differ as to the level of generality, both he and the formalists work from a textual unit of a clause of an Article rather than reasoning down from a systemic property such as balance. Hence, he is appropriately labeled a separation theorist for my purposes.

Another prominent strand of this kind of separationist reasoning, associated often with Professor Steven Calabresi, focuses on the extent of hierarchical control over personnel decisions within the executive branch. Advancing a separation model of interbranch relations, Calabresi and other so-called “[u]nitary executive theorists claim that all federal officers exercising executive power must be subject to the direct control of the President,” while resisting the countervailing argument that Congress can limit such direct control. 36 Calabresi & Rhodes, supra note 2, at 1158; see also Steven G. Calabresi & Nicholas Terrell, The Fatally Flawed Theory of the Unbundled Executive, 93 Minn. L. Rev. 1696, 1697 (2009) (defining unitary executive theory in terms of the “presidential power to remove all subordinates in the executive branch for policy reasons”). Their argument is buttressed by textual evidence that congressional impingement on executive control of admin­istration is impermissible 37 Steven G. Calabresi & Christopher S. Yoo, The Unitary Executive: Presidential Power from Washington to Bush 30–38 (2008) (summarizing textual arguments). and also through a voluminous marshaling of historical evidence and a suite of functional arguments. 38 Id. at 39–41 (recounting how the concept of a unitary executive power arose as a response to the history of the country’s founding). For functional arguments, see, e.g., Steven G. Calabresi, The Virtues of Presidential Government: Why Professor Ackerman Is Wrong to Prefer the German to the U.S. Constitution, 18 Const. Comment. 51, 52–53 (2001) (“The existence of presidentialism and of the separation of powers in our Consti­tution is a praiseworthy feature of the document that should be emulated abroad.”). Notwithstanding its historical depth and rhetorical sophistication, I find that their work doesn’t quite shake the sense of an a priori commitment to robust presidential control. Given that they are works that purport to be back­ward looking and historical, their contemporary relevance and harmony with positions in current political debates is rather hard to ignore.

The analytic key to the unitary executive theory is its identification of a specific branch’s powers, commonly based on text, preratification interpretive conventions, 39 See Caleb Nelson, Originalism and Interpretive Conventions, 70 U. Chi. L. Rev. 519, 549 (2003) [hereinafter Nelson, Originalism] (offering an account of the hermeneutical role of “the linguistic and legal principles that formed the background for the Constitution”). and postratification practice. Given the lin­guistic separation and variegation of Articles I, II, and III—each of which starts with a Vesting Clause referring to a distinct and different power—this approach will naturally tend to end in an account of the branches as separate and distinct entities. 40 Gary Wills, Explaining America: The Federalist 119 (1981) (“Checks and balances do not arise from separation theory, but are at odds with it. Checks and balances have to do with corrective invasion of the separated powers . . . .”). Overlap between branches, to be sure, is conceptually possible. 41 Saikrishna Bangalore Prakash, The Separation and Overlap of War and Military Powers, 87 Tex. L. Rev. 299, 308 (2008) (recognizing “the possibility of separation and overlap” and arguing that such a possibility should be analyzed “clause by clause”). But it does not often play a motivating role in the theory. 42 It is telling that Prakash begins his analysis of war powers by specifying “exclusive” executive and congressional powers, and only then considering the Constitution’s “system of concurrent powers.” Id. at 351. But see also Saikrishna Prakash, Removal and Tenure in Office, 92 Va. L. Rev. 1779, 1785 (2006) (contending that “the Constitution establishes a system of shared removal powers”).

A separation model is in tension with observed practice in contem­porary government in many ways. Congress exercises influence over the content of regulation and the operation of administrative agencies through its oversight process. The agencies themselves blend together the speci­fication of first-order conduct rules, the investigation of past violations, and (at times) adjudication of such acts. Faced with what to them seems to be the wholesale defenestration of constitutional norms, advocates of the separation model frankly recognize the root and branch character of their preferred reforms. 43 See, e.g., Steven G. Calabresi, Mark E. Berghausen & Skylar Albertson, The Rise and Fall of the Separation of Powers, 106 Nw. U. L. Rev. 527, 545–48 (2012) (proposing dramatic reforms to the operation of all three branches, such as random assignment of Congressional members to committees, a general sunset law that would sunset most fed­eral statutes after twenty years, and life tenure for administrative law judges). Its advocates look to the courts to restore appropriate separation through vigorous superintendence of institutional walls. 44 See, e.g., Martin H. Redish, The Constitution as Political Structure 101 (1995) (“[T]he Court’s role in separation-of-powers cases is to be limited to determining whether the challenged branch action falls within the definition of that branch’s constitutionally derived powers . . . .”). Indeed, it is possible to gloss the literature on the separation model as a sub rosa brief to the federal judiciary soliciting urgent and extensive intervention to change the shape of the federal government and vindicate a perceived original design.

B. Balance Models

Balance models of the separation of powers are premised on a rejection of the possibility of deriving from either the Constitution’s text or history a delimited and determinate set of powers for assignment to each branch. To the contrary, the leading work in this vein finds the text inescapably ambivalent. 45 Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573, 603 (1984) (finding “imprecision inherent in the definition and separation of the three governmental powers”); see also Curtis A. Bradley & Martin S. Flaherty, Executive Power Essentialism and Foreign Affairs, 102 Mich. L. Rev. 545, 553 (2004) (“[T]he textual arguments in support of the Vesting Clause Thesis are, at best, indeterminate.”); Lessig & Sunstein, supra note 29, at 48 n.195 (“[T]he [Article II] Vesting Clause does nothing more than show who . . . is to exercise the execu­tive power, and not what that power is.”). Separation theorists, however, acknowledge this difficulty. Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231, 1238 n.45 (1994) [hereinafter Lawson, Rise and Rise] (“The problem of distinguishing the three functions of government has long been, and continues to be, one of the most intractable puzzles in constitutional law.”). Such work instead situates the Constitution in what is described as a fluid, contested, and unstable eighteenth-century debate about the appropriate internal organization of government, a debate infused with the ideas of Montesquieu, Locke, and civic Republican theorists. 46 Martin S. Flaherty, The Most Dangerous Branch, 105 Yale L.J. 1725, 1755 (1996) (“[T]he complex, messy, and at times contradictory ferment in constitutional thinking renders it unlikely at best that, by 1787, Americans had reached a consensus on the doctrine in anything like the precise, thoroughgoing manner that modern formalists prescribe.”). Rather than finding uniform evidence of a unitary executive acoustically walled off from quotidian congressional control, for example, advocates of the balance model find more variation and ambiguity in late eighteenth-century state constitutional treatment of the same issue. 47 See, e.g., Peter M. Shane, The Originalist Myth of the Unitary Executive, 19 U. Pa. J. Const. L. 323, 338, 363 (2016) (finding that “despite executive power vesting clauses,” each of sixteen state constitutions contemporaneous with the Constitution’s ratification “contemplates either a mandatory or permissive legislative role in the appointment of officials involved in public administration”). In consequence, they decline to draw a strong conclusion from the Consti­tution’s text, preratification practice, or Founding-era interpretative conventions about the precise contours of each branch’s authority.

Rather than starting with semantics, the proponents of balance models focus on the general purposes of the Constitution’s design. Their style can be usefully described as architectural and not literary. Their focus is upon the net effect of interactions between the branches rather than on matching specific powers to particular state entities. Hence, on Professor Martin Flaherty’s influential historical account, the Constitution’s tripartite design of branches was intended to ensure that “both the basic division of powers, as well as their frequent mixture, merely served the more fundamental goals of ensuring that the branches of government would remain balanced, extending accountability throughout government, and making government more efficient than it had been in recent memory.” 48 Flaherty, supra note 46, at 1784 (emphasis added). For Flaherty, “the goal of balance” was “the Founding’s most important separation of powers value.” 49 Id. at 1816; see also Cynthia R. Farina, Statutory Interpretation and the Balance of Power in the Administrative State, 89 Colum. L. Rev. 452, 496 (1989) (describing the functionalist premise that “through the carefully orchestrated disposition and sharing of authority, restraint would be found in power counterbalancing power”). In a similar vein, Professor Abner Greene argues that “a system of checks and balances, rather than a system in which each branch exercises power on its own, ensures against the inflation of power in any one branch.” 50 Abner S. Greene, Checks and Balances in an Era of Presidential Lawmaking, 61 U. Chi. L. Rev. 123, 132 n.23 (1994). More recently, Professor Jon Michaels reads the Constitution to embody “an enduring, evolving com­mitment to separating and checking power.” 51 Jon D. Michaels, An Enduring, Evolving Separation of Powers, 115 Colum. L. Rev. 515, 520 (2015) [hereinafter Michaels, Enduring, Evolving]. Building on a literature about a so-called “internal separation of powers,” 52 For accounts of “internal” checks and balances, see, e.g., Gillian E. Metzger, Ordinary Administrative Law as Constitutional Law, 110 Colum. L. Rev. 479, 498 (2010) (arguing that they are ways of enhancing legality and rights); accord Neal Kumar Katyal, Internal Separation of Powers: Checking Today’s Most Dangerous Branch from Within, 115 Yale L.J. 2314, 2319 (2006) (“[T]his Essay takes a case study, the war on terror, and uses the collapse of external checks and balances to demonstrate the need for internal ones.”). Michaels looks not to inter­branch relations but rather to “subconstitutional, rivalrous counter­weights [as a means to] constrain the political leadership atop admin­istrative agencies in ways more reliable and immediate than anything the legis­lature or courts could regularly do.” 53 Michaels, Enduring, Evolving, supra note 51, at 534; see also Huq & Michaels, supra note 5, at 391 (positing a “thick political surround” as “a complex ecosystem of intrabranch and entirely external actors not traditionally accounted for in the separation-of-powers literature that do a lot of the work pushing and pulling, advancing prized values, and jockeying with one another”); Michaels, Of Constitutional Custodians, supra note 3, at 229 (describing, and finding normative significance in, “horizontal power dynamics between and among . . . administrative rivals” as a means of legitimizing the administrative state).

Balance models of this ilk focus on system-level qualities, such as the relative legal or political powers of one branch as against another. They resist the call for crisply demarcated fences. Instead, they excavate com­modious common spaces 54 Rebecca L. Brown, Separated Powers and Ordered Liberty, 139 U. Pa. L. Rev. 1513, 1527–28 (1991) (describing an approach to the separation of powers that stresses “the interdependence of the branches”). or explore the complexities of observed interbranch interactions. 55 See, e.g., David E. Pozen, Self-Help and the Separation of Powers, 124 Yale L.J. 2, 4 (2014) (focusing on the “informal norms that determine how [institutional] authorities are wielded and disputes about them settled”). Pozen’s close study of information leaks also probes what in effect is another element of these informal norms. David E. Pozen, The Leaky Leviathan: Why the Government Condemns and Condones Unlawful Disclosures of Information, 127 Harv. L. Rev. 512, 624 (2013). They appeal not to the legalistic touchstones of the separation model—which is the alignment between textual allocation and institutional practice—but to the practical consequences of institutional design as a whole. Balance, most importantly, is thought to generate systemic legitimacy and to promote individual liberty interests. 56 See, e.g., Greene, supra note 50, at 176 (“[T]he framers’ core checks and balances value was to ensure a balance of power among the branches—to prevent the tyranny of any one branch (and thus help preserve individual liberty) . . . .”); David H. Moore, Taking Cues from Congress: Judicial Review, Congressional Authorization, and the Expansion of Presidential Power, 90 Notre Dame L. Rev. 1019, 1028 (2015) (positing that the “checks and balances” should be honored so as “better to preserve individual liberty”); Cass R. Sunstein, Constitutionalism After the New Deal, 101 Harv. L. Rev. 421, 443 (1987) (describing “the system of checks and balances as a necessary safeguard of private property and liberty against factionalism and self-interested representation”).

Lacking the end-of-days tone of some separationist accounts, balance theorists nonetheless disagree as to whether observed institutional dynamics reflect a constitutionally desirable equilibrium. Flaherty, for instance, views the executive branch as excessively powerful and urges recalibration of interbranch relations to cut it down to size. 57 Flaherty, supra note 46, at 1816–17 (discussing the “inexorable” increase in executive power). Greene bemoans the large number of regulatory delegations, which in his view has left the balance of institutional powers quite lopsided. 58 Greene, supra note 50, at 183 (complaining that “the nondelegation doctrine has been underenforced for the past fifty years”). In contrast, Michaels is optimistic about the current administrative state, which he sees as an institutional locus in which the checking and balancing dynam­ics that exclusively occurred among the branches can now be “renew[ed]” and “updat[ed].” 59 Michaels, Enduring, Evolving, supra note 51, at 522. To Michaels, the impinging tide of privatization poses a fresh challenge to the post-New Deal reconstruction of the separation of powers—but one for which history provides adequate responsive tools. 60 Id. at 522–23. Despite these tonal differences, the vector of reform emerging from the balance models tends to be at odds with the recommendations typically issuing from the separation model. Where the separation model slides into a backward-looking Burkean cri de coeur about the follies of heedless institutional change, the balance model keeps its eyes on a dis­tant future horizon and presses its spurs more deeply into the Constitution’s flanks. It embraces a headlong pursuit of institutional adaption and change in a rapidly evolving world.

One element of this transformative commission is a relative skep­ticism about the wisdom of judicial review of separation of powers questions. The idea of balance provides no reliable benchmark for judi­cial application, since judges are unlikely to be well positioned to eval­uate the effects of institutional change on the interbranch status quo. 61 For an extended unpacking of arguments for this position, see Aziz Z. Huq, The Negotiated Structural Constitution, 114 Colum. L. Rev. 1595, 1674–83 (2014) [hereinafter Huq, Negotiated Structural Constitution]. For a parallel critique by an avowed formalist, see Stephen L. Carter, Constitutional Improprieties: Reflections on Mistretta, Morrison, and Administrative Government, 57 U. Chi. L. Rev. 357, 375–76 (1990). And if balance is elusive, especially from the vantage point of the bench, then why make it judicially enforceable? The prospect of second-guessing from the bench, after all, raises the expected costs of needful institu­tional recalibration. With the balance model’s eye on the future and its default setting being one of acceleration, judicial review of separation of powers issues comes to be seen as an otiose encumbrance of a bygone era. So it is no surprise that scholars such as Flaherty have a dim view of the justiciability of separation of powers questions. 62 See Flaherty, supra note 46, at 1828 (“The Supreme Court should rarely inter­vene in separation of powers conflicts. When it does, it should do so principally when faced with a compelling violation of one of the basic values of balance, joint accountability, or sufficient energy.”). Greene, in contrast, generally approves of the doctrine’s outcomes. See Greene, supra note 50, at 196. Michaels and I have coauthored a piece that offers a sympathetic reconstruction of the Court’s jurisprudence in this field as an effort to adapt and respond to the complex institutional environment in which the branches operate. See Huq & Michaels, supra note 5, at 416–35. The article’s conclusion carefully hedges any normative implication from the doctrinal reconstruction because the authors’ views of the merits of the justiciability question diverged. Id. at 437. While not a con­sistent tenet of the balance model, this perspective is a natural infer­ence from its logical roots—and yields another sharp contrast to the separation model.

To this point, I have presented the balance and the separation models as crisply distinct. But these are useful categories only insofar as they help roughly slice up the jurisprudential territory. In practice, the difference is one of degree not kind. It is hence possible to start from an assumption of separation and to work in small steps toward an account of balanced, necessarily multivariate government action. Such accounts embarrass my crude attempt at categorization. But I think they are better thought of as derivatives of the balance model. These approaches start with the idea of distinct institution authorities but then suggest that government action that impinges on individual freedoms should always involve diverse branches. They hence imagine legitimate government action as a compo­site in which each discrete branch’s distinctive function works in graceful counterpoise to create a harmonious whole. Their emphasis on the necessity of a totality, and upon the finely calibrated equilibrium between branches, aligns them with the balance rather than with the separation model.

Two prominent examples are associated with NYU School of Law. Professor Rachel Barkow has characterized the criminal justice system as demanding the input of all three branches prior to the deprivation of a person’s liberty. 63 Rachel E. Barkow, Separation of Powers and the Criminal Law, 58 Stan. L. Rev. 989, 1012 (2006) (making an “argument for strict separation of powers when the state uses its criminal powers”). Her account hits formalist notes, but these blend together to yield a holistic, functional chord. Professor Jeremy Waldron has simi­larly suggested that the essence of the separation of powers lies in its demand for the “ordinary sequence” of legislative, executive, and judicial actions. 64 Jeremy Waldron, Separation of Powers in Thought and Practice?, 54 B.C. L. Rev. 433, 434–35 (2013) (“[A]nything we do to X or about X must be preceded by an exercise of legislative power that lays down a general rule applying to everyone . . . and a judicial pro­ceeding that makes a determination that X’s conduct in particular falls within the ambit of that rule, and so on.”). Although Waldron’s account draws on logic familiar to the separation model insofar as it insists upon distinguishing the branches, it better aligns with the balance model insofar as he insists not on each power in isolation but rather views them together as a “general articu­lated scheme of governance.” 65 Id. at 466. Rebecca Brown offers a somewhat similar claim that the separation of powers should be “understood as a concern for protecting individual rights against encroach­ment by a tyrannical majority.” Brown, supra note 54, at 1516. In Waldron’s work, formalism about insti­tutional functions has an archaistic yet utilitarian affect. That is, it works as a stilted but necessary artifice for redeeming necessary institutional reticulation, which in turn enables meaningfully civil democratic self-government. 66 This part of Waldron’s argument is better developed in Jeremy Waldron, Political Political Theory: Essays on Institutions 111 (2016).

C. Exogenous Models

Exogenous models of the separation of powers renounce the task of allocating powers to distinct branches and shrug off the task of finding ways to maintain a balance across the federal government. Instead, they locate benchmarks for constitutional evaluation beyond the document.

The exogenous model starts from the proposition that it is not possible to identify ex ante a specific government action as legislative, executive, or judicial because there is commonly an observational equiva­lence between those forms of state action. Scholars working in this vein are skeptical of any approach that treats the “relevant constitutional language . . . as a set of descriptive labels, a set of terms like ‘executive,’ ‘state,’ or ‘judicial’ (terms that seem ripe for definition or drawing boundaries), [such that] texts are then matched against the challenged practice under review.” 67 Nourse, supra note 25, at 841; accord David Orentlicher, Conflicts of Interest and the Constitution, 59 Wash. & Lee L. Rev. 713, 726 (2002) (“Even if we wanted to follow a formalist approach, it is impossible to do so. We cannot develop independent definitions of executive, judicial, and legislative action and assign those actions to their corresponding branch of government.”); see also M. Elizabeth Magill, Beyond Powers and Branches in Separation of Powers Law, 150 U. Pa. L. Rev. 603, 604 (2001) [hereinafter Magill, Beyond Powers and Branches] (“The effort to identify and separate governmental powers fails because, in the contested cases, there is no principled way to distinguish between the rele­vant powers.”). They are skeptical that the “branch” is the truly relevant unit of analysis if one wishes to understand and predict the actions of official actors. 68 Magill, Beyond Powers and Branches, supra note 67, at 606 (arguing that “government authority cannot be parceled neatly into three categories, and government actors cannot be understood solely as members of a branch of government”); accord Huq & Michaels, supra note 5, at 391– 407 (discussing “the institutional context in which com­peting and sometimes conflicting values are reconciled”).

In a leading critical statement, Dean Elizabeth Magill argued that even if functional distinctions can be identified, there is still “[n]o reason” to expect that “functional dispersal of power creates and maintains tension and competition among the departments.” 69 M. Elizabeth Magill, The Real Separation in Separation of Powers Law, 86 Va. L. Rev. 1127, 1170 (2000) [hereinafter Magill, Real Separation]. Her former col­league at the University of Virginia School of Law, 70 The identification of institutional affiliations is not merely a heuristic for organizing the taxonomy here. Rather, the institutional context in which scholarship is produced mat­ters in the sense that physical proximity makes the diffusion of intentions easier and the alignment of ideas easier. Hence, it should not surprise that leading law schools, such as NYU and Virginia, should evince some elements of discrete, hard-edged intellectual formations. Professor Daryl Levinson, has in similar terms challenged the implicit assumption that officials within a branch will pursue the latter’s interests faithfully. 71 Daryl J. Levinson, Parchment and Politics: The Positive Puzzle of Constitutional Commitment, 124 Harv. L. Rev. 657, 670 (2011) [hereinafter Levinson, Parchment and Politics] (“Madison never explained why the branches of government, or the state and federal governments, would reliably have political incentives at odds with one another—why they would tend to compete rather than cooperate or collude.”).

The exogenous model also turns away from the aspiration of balance. Again, Magill’s work provides the basic building blocks for a skeptical turn. Not only is the idea of “power” in the analysis of interbranch relations ambiguous and underspecified, 72 Magill, Real Separation, supra note 69, at 1195–96 (developing ambiguities in the relevant conception of power). Magill observes, but rarely is any explanation offered for how a balance between the branches is to be achieved. 73 Id. at 1173 (arguing that the notion that “dispersal of governmental functions mysteriously leads to balance” rests on “obscure” assumptions). Magill is equally critical of the hybrid logic of the sort Barkow and Waldron develop, which blends separation and balance. Id. at 1174 (“[A]ttempting to merge the distinct conceptions into a coherent set of ideas is a fruitless enterprise.”). Magill’s argument is not a defense of judicial capacity to measure balance: It is instead a striking and comprehensive criticism of the concept. Its implication is not simply that courts should decline to enforce constitutional common law prohibitions founded on an assump­tion of balance. Its core entailment is rather that the enterprise of identi­fying any kind of architectural equilibrium between the branches is a fool’s errand in the first instance. 74 Id. at 1194 (“We do not know what ‘balance’ means, and we do not know how it is achieved or maintained.”); see also Magill, Beyond Powers and Branches, supra note 67, at 605 (“We have not come close to articulating a vision of what an ideal balance would look like.”).

An implication of these criticisms is that the separation of powers cannot be evaluated on the basis of intrinsic criteria—that is, normative values implicit in the Constitution’s design. And if the literary and the architectural accounts of the separation of powers fail, the path is clear for a consideration of alternative explanations and arguments that look beyond the Constitution’s text to evaluate institutional arrangements.

Exogenous models that emerge from the critical ground-clearing in Magill’s and others’ work can take either a positive or a normative flavor. One largely positive pathway dispenses almost wholly with the “law” of the separation of powers and looks alternatively to its politics. If there is no firm functional contour to a branch’s action and no interbranch equi­librium to be maintained, what is left is the raw force of partisan politics. On this account, associated now with Professors Daryl Levinson and Richard Pildes, the development of a national political-party system, although unforeseen by the Constitution’s Framers, has “tied the power and political fortunes of government officials to issues and elections,” and thereby fostered “a set of incentives that rendered these officials largely indifferent to the powers and interests of the branches per se.” 75 Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 Harv. L. Rev. 2311, 2323 (2006). As a result, “when government is unified and the engine of party competition is removed from the internal structure of government, we should expect interbranch competition to dissipate.” 76 Id. at 2329; accord Bradley & Morrison, Historical Gloss, supra note 14, at 443 (“[T]he Madisonian model of interbranch rivalry is especially inaccurate during times of unified government.”). The result is a Beardian 77 See generally Charles A. Beard, An Economic Interpretation of the Constitution of the United States (1913) (contending that the 1787 Constitution was designed to facilitate the retention of wealth by, and the transfer of wealth to, a small minority of prop­ertied men). account of constitutional law in which “the interests of the politically powerful” dominate. 78 Jack Goldsmith & Daryl Levinson, Law for States: International Law, Constitutional Law, Public Law, 122 Harv. L. Rev. 1791, 1833 (2009) (“Constitutional law is pervasively shaped by the same political forces that it purports to regulate.”). Note that Goldsmith and Levinson’s account does not specify a definition of “power” or “political forces.” They point to federal actors, state governments, and social movements as examples of the forces shaping constitutional law. Id. at 1811–12 (suggesting an understanding of politics that is capacious and not limited to partisan forces). The strength of these exogenous models is largely descriptive rather than normative. It is hard to see the allure in a normative consti­tutional theory that directs the strong to take what they can and the weak to bear what they must. 79 Writing in a more normative vein, Levinson and Pildes offer a set of recommen­dations “for preventing strongly unified party government from taking hold.” Levinson & Pildes, supra note 75, at 2380. For a similar turn toward a “minimally functional democ­racy” as a normative objective, see Daryl J. Levinson, Empire-Building Government in Constitutional Law, 118 Harv. L. Rev. 915, 972 (2005). To the extent their reform agenda accepts and builds upon the institutional dominance of the two main political parties, it is not at all clear that its fundamentally quiescent orientation toward elite-dominated constitutional law and politics is meaningfully inflected.

If one believes that the concepts of separation and balance are flawed beyond redemption, the other available pathway is to appeal directly to social welfare as a desirable maximand. The leading scholarship on this point, by Professors Eric Posner and Adrian Vermeule, identifies the executive branch as the optimal instrument of public policy. Posner and Vermeule assert that “major constraints on the executive . . . do not arise from law or from the separation-of-powers framework” because legislators and courts are necessarily “reactive and marginal” in comparison to the presidency. 80 Posner & Vermeule, Executive Unbound, supra note 4, at 4–5 (emphasis added). They contend that this is a normatively desirable state of affairs. On their view, the executive will typically (or perhaps inevitably) select the better policy by dint of institutional specialization and better access to information. 81 Eric A. Posner & Adrian Vermeule, Crisis Governance in the Administrative State: 9/11 and the Financial Meltdown of 2008, 76 U. Chi. L. Rev. 1613, 1614 (2009) (“In the modern administrative state, it is practically inevitable that legislators, judges, and the public will entrust the executive branch with sweeping power to manage serious crises of this sort.”); accord Eric A. Posner & Cass R. Sunstein, Chevronizing Foreign Relations Law, 116 Yale L.J. 1170, 1204 (2007) (arguing that the executive is best placed to resolve difficult foreign affairs questions requiring judgments of policy and principle and that the judiciary should defer to the executive based on its foreign policy expertise). A “rational and well informed” executive acting without constraints will always outperform the same body constrained by other branches or noisome rights. 82 Eric A. Posner & Adrian Vermeule, Emergencies and Democratic Failure, 92 Va. L. Rev. 1091, 1099 (2006) (“There is no general reason to think that judges can do better than government at balancing security and liberty.”). Concerns about excessive concentra­tions of governmental authority in the executive’s hands are unwarranted because chief executives are necessarily subject to electoral checks. 83 Posner & Vermeule, Executive Unbound, supra note 4, at 5 (asserting that political checks will “at least block the most lurid forms of executive abuse”). Compliance with the “law and executive practice” are important not because of an intrinsic interest in legality but because they “allow a well-motivated executive to send a credible signal of his motivations, com­mitting to use increased discretion in public-spirited ways.” 84 Eric A. Posner & Adrian Vermeule, The Credible Executive, 74 U. Chi. L. Rev. 865, 867 (2007); see also Posner & Vermeule, Executive Unbound, supra note 4, at 151 (asserting vaguely that credibility will lead “voters and legislators . . . to confer authority”). The result is a happy gloss on the modern dominance of the executive as a benign Leviathan, duly attentive to the interests of its citizenry thanks to elec­tions and effortlessly selecting among available policies with Panglossian deftness. Given this relentless optimism, the Posner–Vermeule iteration of the separation of powers unsurprisingly leaves little space for a judicial role enforcing the structural constitution.

The theoretical and empirical premises of this last version of the exogenous model have been thoroughly challenged. 85 For full-length criticisms of both the empirical and theoretical coordinates of the claims presented in the main text, see generally Aziz Z. Huq, Binding the Executive (by Law or by Politics), 79 U. Chi. L. Rev. 777 (2012) [hereinafter Huq, Binding the Executive]; Richard H. Pildes, Law and the President, 125 Harv. L. Rev. 1381 (2012) (book review). A central problem in the Posner–Vermeule analysis is that it is commonly constructed on the basis of an apples-to-oranges comparison between the executive branch and other branches. Posner and Vermeule consistently model an ideal “rational and well informed” executive while looking at actual Congress and the courts. Attention to the actual construction and performance of executive branch institutions leads to very different conclusions. See Aziz Z. Huq, Structural Constitutionalism as Counterterrorism, 100 Calif. L. Rev. 887, 908 (2012) (identifying path-dependent institutional development and institutional blind spots as sig­nificant drags on the efficiency of executive branch action). It persists as a touchstone for analysis despite the fact that it does not accurately capture empirical regularities in the behavior of official actors or embody a per­suasive account of the relation of institutional design to social welfare. Perhaps this is because it is admirably parsimonious, boldly stated, and provocative. In contrast, alternative exogenous approaches to the problem of interbranch interaction tend to be less compelling because, as Magill explained, they “seek to understand the incentives of the actors who will exercise that power in a pointed enough way that it helps us comprehend how those powers will be exercised.” 86 Magill, Beyond Powers and Branches, supra note 67, at 659. Such granular analysis of situated and specific dynamics provides a less breathtakingly synoptic view of legal and constitutional institutions—that is, it yields less by way of high theory—but may provide better insight about how the separation of powers works in practice. 87 It is appropriate and perhaps necessary here to disclose where to locate in this typology my own contributions to this field. In addition to criticisms of the Posner–Vermeule account, I have advanced two main points. First, from an exogenous perspective, I have argued on welfarist grounds for a narrow view of the justiciability of separation of powers questions in favor of broader latitude among political officials to negotiate new institutional arrange­ments. See Huq, Negotiated Structural Constitution, supra note 61, at 1674–83; Aziz Z. Huq, Standing for the Structural Constitution, 99 Va. L. Rev. 1435, 1490–514 (2013) [hereinafter Huq, Standing]. Second, in harmony with Magill’s endorsement of more granular studies, I have developed more fine-grained, multimethod accounts of specific separation of powers dynamics concerning detainee policy and remedies. On detainees, see generally Aziz Z. Huq, The Predicates of Military Detention at Guantánamo: The Role of Individual Acts and Affiliations, 13 J. Empirical L. Stud. 567 (2016); Aziz Z. Huq, The President and the Detainees, 165 U. Pa. L. Rev. 499 (2017) [hereinafter Huq, President and Detainees]. On constitutional remedies, see generally Aziz Huq & Genevieve Lakier, Apparent Fault, 131 Harv. L. Rev. (forthcoming 2018); Aziz Z. Huq, Judicial Independence and the Rationing of Constitutional Remedies, 65 Duke L.J. 1 (2015) [hereinafter Huq, Judicial Independence].

II. Congress’s Constitution as Separation of Powers Theory

This Part considers Congress’s Constitution in terms of the tripartite schema of separation of powers theory that I just discussed. Because his six detailed case studies of congressional authorities are a substantial part of Chafetz’s contribution, I also examine here the book’s selection and elaboration of case studies.

A. The Theory of Constitutional Politics

As noted earlier, Congress’s Constitution showcases two main contri­butions. 88 See supra text accompanying notes 19–22. The first is a theory of the separation of powers with special attention to the legislature’s role. The second is the series of case studies of three “hard” and three “soft” congressional powers: appropriations, executive branch personnel control, contempt, speech and debate immun­ities, internal disciplinary devices, and intracameral rules. Let’s consider the theory before thinking through how it does or does not connect to Chafetz’s case studies.

The theoretical contribution of Congress’s Constitution has three core elements. 89 Chafetz offers a different four-part breakdown of his claim (pp. 18–19). This account focuses on the implications rather than the causal core of the claim. Isolating them facilitates the comparison of Chafetz’s theoret­ical armature—call it the constitutional-politics account of the separation of powers—with the extant theories charted in Part I. That account con­tains three elements: a causal claim about the production of law, an argu­ment about the nature of politics, and an evaluative judgment about the sort of politics that the American system will produce.

First, the constitutional-politics account offers a causal claim. I read Chafetz to assume the existence of a causal vector that runs first from “politics,” which comprises of “political behaviors and interactions,” to “political power.” From there, this vector leads to the resolution of legal questions concerning “the distribution of governmental authority within a political community” (p. 16). A negative implication of this assertion is that questions about interbranch relations are “not answerable by reference to the normal tools of constitutional interpretation,” such as “[t]ext, history, structure, and precedent,” since the latter are insufficiently determinative (p. 16). 90 For a discussion of whether the case studies follow through on this approach, see infra text accompanying notes 224–225. At the same time, because politics is a matter of “historical construction” (p. 4), it constrains the range of options avail­able to official institutional actors at any given time. All this is said to leave “significant room for the play of contingency and agency” (p. 4). Chafetz also adds that the settlement of interbranch disputes is “nonhier­archical” and indeterminate in duration, in the sense that there is no entity with power to issue dispositive settlements and hence no logical stopping point to disputes (pp. 18–19). This in turn suggests that what is viewed by most scholars and jurists as “law,” Chafetz views as largely a fragile and contingent political settlement.

The second element of the constitutional-politics account concerns the nature of the “institutional actions and interactions” (p. 4) that are at the root of the causal chain that ultimately yields outcomes to separation of powers disputes. In brief, Chafetz characterizes the law of separation of powers in paradoxical terms. On the one hand, it has an epiphenomenal character, since it is a mere byproduct of interbranch battles over power. On the other hand, this byproduct has the oddly durable effect of freeing one branch or another from the binding constraints of diffuse public judgment. This second claim, as just formulated, is particularly opaque and needs some unpacking before the relationship between the two claims can be properly considered.

According to Chafetz, interbranch disputes are resolved “locally and contingently as they arise, often via compromise or negotiation, and without binding implications for the future” (p. 20). 91 This is parallel to a central claim in Huq, Negotiated Structural Constitution, supra note 61, at 1601 (“Both states and branches engage in such bargaining routinely, notwith­standing scholarly inattention to the practice.”). Such settlements are reached in the “public sphere” (p. 195). They are part of a complex, mechanistically varied, and perhaps evolving domain in which “dialogic and highly mediated” processes of “public political interaction” occur (p. 21). 92 See also p. 24 (“Political trust—and with it political power—arises out of . . . con­versations between political elites and the public.”). The net result of this public process—and seemingly the goal of political actors—is “trust,” which is manifested in persistent public sup­port for an individual official that can survive observed decisions by that official inconsistent with the first-order policy preferences of members of the public (pp. 21–22). The more trust an institution accrues, Chafetz implies, the more leeway the public will give that institution and hence, “the more power that institution has” (p. 22). A notoriously slippery concept, power is implicitly defined here in Congress’s Constitution as the extent of relational discretion obtained between a political institution and a diffuse democratic public. 93 For competing approaches to understanding the term, compare Jean Bethke Elshtain, Power Trips and Other Journeys: Essays in Feminism as Civic Discourse 136 (1990) (“Power is a form of compulsion exerted by the already (relatively) powerful upon one another within official political institutions designed to promote the aims and interests of competing groups. It is of, by, and for elites.”), with Michel Foucault, Afterword: The Subject and Power, in Hubert L. Dreyfus & Paul Rabinow, Michel Foucault: Beyond Structuralism and Hermeneutics 208, 217 (2d ed. 1983) (“[P]ower as such does not exist.”).

The third element of the constitutional-politics account is an opti­mistic toleration for nondisputive and democratically credentialed change hardwired into Chafetz’s theory. Given institutional actors’ dependence upon public trust, their strategies for claiming institutional prerogative and for making countermoves are necessarily bounded. They must be “judicious” and not “maximal” (p. 24; see also p. 308). The system as a whole is “dependent . . . on the will of the people” (p. 34), which in the final analysis determines its actions.

It is rather hard to read these passages as anything other than an approving and optimistic account of the tempering effect of officials’ competition for public trust. This is so for at least two reasons. To begin, Chafetz’s argument aims for a reassuringly democratic pedigree for struc­tural arrangements and the concrete policy outcomes flowing from them. The constitutional-politics account imagines institutional change as bounded by a sensible and sensitizing popular will. 94 For instance, the “feasibility and success” of constitutional innovation at an insti­tutional level are said to “depend on how successful [officials] are at engaging in the public sphere” (p. 38). It treats each branch as vocalizing a different element of the polity in ways that “enhance the quality of public deliberation” (p. 310). Even if beset by conflict, “noise, and clamour” (p. 314), it seems that the resulting ca­cophony is liable to generate a stable, representative, and nontyrannical form of government (p. 303). It is this model, I think, that animates Chafetz’s optimistic conclusion that the American separation of powers system is “more fully representative, more deliberation promoting, and more resistant to assertions of tyrannical or autocratic power” than other common constitutional systems (p. 303). These passages simply do not read as neutral and descriptive. Given the shared commitment to demo­c­racy among constitutional scholars and jurists, it is not possible to treat this argument as anything other than an endorsement of the separa­tion of powers system as a whole, albeit with some space reserved for criticisms of discrete outcomes at the margin.

The second reason for taking Chafetz’s theory as both normative and descriptive in ambition is that, even as the constitutional-politics model explicitly recognizes and embraces the inevitability of institutional evolution (for examples of this recognition, see pp. 25, 42, 312–13), there is no suggestion that this process will ever get out of hand. 95 Chafetz also endorses specific instances of institutional transformation, such as the move from fees and bounties to salaries in the eighteenth-century federal bureaucracy (p. 73). To be sure, the envisaged process of institutional change has “no logical stop­ping point” (p. 19), and Chafetz says that the separation of powers system did not calcify into a “fixed” pose circa 1791. But even as he characterizes official conduct as necessarily “judicious,” Chafetz does not say anything about the possibility of destabilizing conflict or suggest that constitutional politics will produce anything more than locally subopti­mal outcomes. There is no exploration of how the separation of powers system might generate substantial social injustices, such as the ante­bellum preservation of slavery or the Japanese internment, or structural exclusions and subnational authoritarian regimes, 96 For an excellent historical account of subnational authoritarianism in the United States, see generally Robert Mickey, Paths Out of Dixie: The Democratization of Authoritarian Enclaves in America’s Deep South, 1944–1972 (2015). although I have no doubt that Chafetz deplores these outcomes. More generally, there is no hint in the book of what Thoreau called “the unhandselled globe” in its raw and bloody glory; there is rather an eminently habitable, if somewhat chaotic, “garden.” 97 Henry D. Thoreau, The Maine Woods: A Fully Annotated Edition 63 (Jeffrey S. Cramer ed., 2009). In short, the book’s silence about such enduring negative outcomes of the American political system necessarily nudges the reader toward a positive view of that system.

Finally, there is an internal tension torqueing this third element of the constitutional-politics account—one that will help us locate it within the topography of extant theoretical accounts. Simply put, the mechanism of institutional change Chafetz posits is not self-stabilizing. In the constitutional-politics account, interbranch conflict persists over time, and “the branch that most successfully engages the public will accrete power over time” (p. 14). But notice that this descriptive claim appears to undermine two other key elements of Chafetz’s argument. First, if one branch accretes asymmetrically high levels of power in relation to the other, at some point interbranch conflict will end because the powerful branch will always or almost always win. Chafetz, however, seems to assume that conflict itself is a stable-state equilibrium in which all branches continue to play a meaningful role. 98 It is otherwise hard to know why he insists that there is “no a priori determinable rule” for resolving interbranch conflicts (p. 305). If one branch kept accreting power until it was hegemonic, the rule of decision for such conflicts would, in fact, be quite simple. Second and relatedly, the bulk of the book is devoted to the proposition that Congress has the capacity to be a player equal to the executive (see, e.g., p. 42). Yet at the same time, its central theory implies that if the executive has gained the upper hand today, this is because it has played its cards better than Congress has. The theory then provides no reason to disturb this result—as Chafetz explic­itly seeks to do.

These puzzles can be solved only by smuggling a normative notion of balance into the theory. And Chafetz does precisely this when he predicts that “any actor attempting to aggrandize her powers beyond tolerable limits can expect to find herself opposed by other actors” (p. 312). 99 Notice again the normatively freighted and reassuring quality of this conclusion. By focusing on the ambiguity of the adjective “tolerable” in this formulation, it quickly becomes clear this dictum bears more than a passing resemblance to a familiar strand of balance theory. 100 To be clear, Chafetz does not disavow the balance label, but he also does not align himself with past theoretical claims. That is, public conceptions of “tolerable” conflict are implicitly taken to have a stabilizing, balancing effect upon interbranch dynamics—even though the constitutional-politics account itself provides no a priori reason for thinking that the public will hold or act upon such preferences. Hence, the theory silently smuggles in a set of quite controversial assumptions in respect to the preferences of the democratic public about the nature of interbranch conflict. 101 See infra text accompanying notes 214–216.

How then is the theory animating Congress’s Constitution new and distinct from the three general approaches sketched in Part I? 102 I should also add that this account of Chafetz’s theory slights certain parts of Congress’s Constitution. For example, I have not highlighted or analyzed Chafetz’s assertion that his theoretical account is distinctive insofar as it is “multiplicity based” because it “aims to highlight the ways in which claims of authority multiply and overlap in a non­hierarchical constitutional order” (p. 18). See also Josh Chafetz, A Fourth Way? Bringing Politics Back into Recess Appointments (and the Rest of the Separation of Powers, Too), 64 Duke L.J. Online 161, 162–63 (2015), (on file with the Columbia Law Review) (“The label ‘multiplicity-based’ is meant to emphasize the ways in which claims of institu­tional authority multiply, overlap, and interact in a non-hierarchical constitutional order.”); Josh Chafetz, Multiplicity in Federalism and the Separation of Powers, 120 Yale L.J. 1084, 1112–28 (2011) (“By multiplicity, I mean not simply the now-familiar observation that all three branches . . . engage in constitutional decisionmaking [but that] there is (sometimes) affirmative value in promoting the means for interbranch . . . conflict without any . . . superior body that can articulate a global, principled, final, and binding decision on the matter.”). But the propositions that (1) the separation of powers comprises a plurality of actors, and (2) those actors all make consti­tutional claims that overlap and sometimes conflict, are both commonplace in theoretical work on the separation of powers. Hence, it is not its basis in multiplicity that renders the constitutional-politics account a distinctive and new contribution to the literature. I have chosen to accent what is truly new in the theory. In brief, the constitutional-politics account might well be characterized as an ambitious hybrid of the balance and exogenous models. Its greatest dis­tance, in intellectual terms, is from separation models. For the constitutional-politics account is at odds with the methodological premises of most separation theorists insofar as Chafetz, while conceding that some parts of the written text are “static” irrespective of politics (p. 17), emphasizes the endogeneity of constitutional rules to transient partisan contestation (e.g., pp. 20, 25–26). In this sense, Chafetz’s theory has a kinship with exogenous models, although his approach yields a discernably more optimistic trajectory for interbranch relations. Instead, the constitutional-politics account intertwines elements of both the balance and the exogenous models into a patterning that bodes to be novel.

Moreover, as noted above, the constitutional-politics account assumes a stable state of interbranch conflict in which no one branch becomes hegemonic (p. 305) and implies a kind of self-stabilizing mechanism that kicks into gear whenever one branch gets too big for its britches (p. 312). It seems that public opinion of a certain sort must fuel that mechanism. At the same time, the theory does not rest on any claim that officials within each branch will always or often act upon the basis of institutional loyalties (cf. pp. 30–31, 311). Rather, it locates official actors within an explicitly political field in the sense of rooting power in the context of their relations with the public.

One rather glib way of summarizing Chafetz’s theory—surely an excessively reductionist one that does not do justice to its ambition or sophistication—is to say that it is a retooling of the balance model (in which dynamic interactions between the branches yield healthful sparring over policy and collaterally prevent any one branch from overreaching) with the exogenous model’s insights into the motives and operative con­text of government bolted uneasily to its side.

B. From Theory to Case Study

An important contribution of Congress’s Constitution is the six case studies of discrete legislative powers created by Article I of the Constitution. Because these historical analyses constitute a substantial portion of the book, I would derogate from my obligations as reviewer if I said nothing about them. As it happens, Chafetz’s selection of case studies helpfully illuminates his theoretical project—and its limits—in interesting ways.

To begin with, readers familiar with conventional debates about Congress’s powers will be immediately struck by the fact that Chafetz’s particular choice of case studies deliberately omits Congress’s authority to promulgate substantive rules pursuant to various provisions of Articles I, III, and IV. 103 See U.S. Const. art. I, § 4 (specifying Congress’s power to alter state electoral rules); § 8, cls. 3–18 (identifying enumerated powers); § 10, cl. 2 (dealing with regulation of state taxation); id. art. III, §§ 1–2 (describing regulation of lower federal courts and Supreme Court appellate jurisdiction); id. art. IV, § 1 (detailing full faith and credit regulation); § 3 (describing territorial regulation). At first blush, this selection of topics raises obvious questions. There is a close relation between the powers he discusses and the ones he omits. Appropriations, for example, might be a close substi­tute for pure regulation, in particular when conditions are attached to an expenditure. 104 Todd D. Peterson, Controlling the Federal Courts Through the Appropriations Process, 1998 Wis. L. Rev. 993, 1008 (“Through the use of explicitly targeted restrictions on appropriations, Congress can regulate the conduct of the other branches and fre­quently can impose its will in spite of executive or judicial opposition.”). The scope of Congress’s power to use conditions on federal spending to achieve policy ends arises most often when states are recipients of those funds. Samuel R. Bagenstos, The Anti-Leveraging Principle and the Spending Clause After NFIB, 101 Geo. L.J. 861, 864 (2013) (noting that “[a]n enormous amount of the New Deal/Great Society state is built on conditional spending statutes” in which states are recipients of federal funds and “[t]hrough the years, states have challenged a number of these laws”). The internal disciplinary choices of Congress interact with its power to promulgate ethics and lobbying rules without running afoul of the First Amendment. 105 For an argument that the current structure of lobbying violates the Petition Clause of the First Amendment, see Maggie McKinley, Lobbying and the Petition Clause, 68 Stan. L. Rev. 1131, 1133–42 (2016). And it is hard to see how the power to appropriate can be disentangled from Congress’s authority under the first two clauses of Article I, Section 8, to tax and borrow money. 106 U.S. Const. art. I, § 8, cls. 1–2. On the interaction of these clauses, see Neil H. Buchanan & Michael C. Dorf, How to Choose the Least Unconstitutional Option: Lessons for the President (and Others) from the Debt Ceiling Standoff, 112 Colum. L. Rev. 1175, 1197–202 (2012) (analyzing the relationship between the powers to tax, spend, and borrow). Even if their borders were crisper, the six heterogeneous powers Chafetz selects would not snap into immediate focus as a logical and obvious whole that invites a singular analytic lens. 107 Chafetz asserts that he is considering only “specific constitutional tools that Congress has at its disposal in its interactions with the other branches” (p. 3). But this is not quite right. Certainly, he begins with an example of interbranch conflict involving the judiciary (p. 9). But as a whole, his book has rather little to say about “Congress’s power to withdraw jurisdiction from the federal courts as a means of shielding questions about the legality of official conduct from judicial review.” Fallon, Jurisdiction-Stripping, supra note 20, at 1048. Where the courts do enter the picture, it is to criticize their meddling in Congress–executive relations (e.g., pp. 182–90). Readers hence may be saddled with a nagging concern that other relevant congressional author­ities are playing out at the periphery of the book’s vision.

There is, though, a justification for Chafetz’s selection of cases. His selection, I believe, flows from his larger normative project rather than a functional homology among the chosen topics. Perhaps the best way to understand his subject matter is in simple terms of Chafetz’s wish to offer an account of “congressional power” (p. 27) not simply in the abstract, but in relation to the executive branch, which dominates contemporary commentary. The six case studies of congressional power concern those instruments most relevant to Congress’s interactions with its Article II counterpart. 108 In contrast, the difficult question of how and when the states can influence con­gressional behavior, and thus when congressional actions raise federalism concerns, is outside the book’s scope. For a skeptical view of the leading general theories, see Aziz Z. Huq, Does the Logic of Collective Action Explain Federalism Doctrine?, 66 Stan. L. Rev. 217, 218–25 (2014). The case studies on this view are instrumental to, and perhaps analytically subordinate to, Chafetz’s claim that the “separation-of-powers system privileges judicious, rather than maximal, combative­ness” between the political branches (p. 311). 109 This is in accord with recent findings that (for example) “investigations offer Congress a check on presidential aggrandizement that is often more effective than that provided by its legislating function.” Douglas L. Kriner & Eric Schickler, Investigating the President: Congressional Checks on Presidential Power 9 (2017).

Correlatively, to the extent a theory emerges from his analysis, it is best understood as being nested in, and potentially reflective of, an abiding concern with interactions between Congress and the executive. It is by no means clear, however, that general propositions about that political-branch binary would also be descriptively accurate or predictively valid as applied to other interbranch dyads. 110 For instance, in other contexts, the influence of one branch (say, Congress) on another (say, the courts) could be mediated by the independent actions of the third branch. For an account along these lines, see Tara Leigh Grove, The Article II Safeguards of Federal Jurisdiction, 112 Colum. L. Rev. 250, 252 (2012) (arguing that “the executive branch has a strong incentive to use this constitutional authority to oppose efforts to curb federal jurisdiction”). Nevertheless, with this caveat in mind, the six case studies do comprise a unified whole with shared points of ref­erence. Four points of analytic commonality across the case studies are worth flagging here.

First, each of the six case studies is organized as a brisk historical narrative, each of which “begin[s] in earnest in England around the turn of the seventeenth century” (p. 4). This history matters, Chafetz assumes, even if it is not consciously recalled by contemporary political actors. (Today, I suspect, anyone who can distinguish their Wolseys from their Mores is able to do so only because she has perused her Mantels. 111 If this seems obscure, that’s the point. No richer, or more delightfully sordid and brutal, an account of the Tudor politics can be imagined than Hillary Mantel, Wolf Hall (2009), and Hillary Mantel, Bring Up the Bodies (2012). ) So why does it matter? Although the justification for the particular historical frame is not extensively developed, it seems fair to say that Chafetz envisages a “developmental” process in which later iterations of insti­tutions are in some fashion “deeply rooted” in earlier versions beginning in the Tudor period (pp. 4–5). Both the historical frame and the specific language employed to explain that frame suggest that historical experi­ence is posited as implicitly relevant to the resolution of subsequent and contemporary disputes. 112 I am not a historian but should flag a methodological concern here likely to resonate most among historians. I think it is fair to say that Chafetz’s accounts are framed as fairly conventional narratives of political conflict in which one side wins, the other loses, and as a result some fairly clear understanding of institutional power arises. Much as I was taught Tudor English as a schoolboy in London, historical battles seem to be a sequence of soccer matches with crisp scores (“Wolsey—one, More—nil”). But are understandings of institu­tional settlement or the lessons drawn from institutional conflicts really so ambiguous? Reviewing a spate of historical books on executive power as it was understood at the time of Founding, John Fabian Witt flags that such matters were “rarely settled and almost always hotly contested.” John Fabian Witt, Anglo-American Empire and the Crisis of the Legal Frame (Will the Real British Empire Please Stand Up?), 120 Harv. L. Rev. 754, 766 (2007) (book review). While the schoolboy in me appreciated the limber narrative Chafetz offered, the historian manqué lurking in me wondered about its shallowness. In particular, the very dif­ferent nature of parliamentary democracy before the Great Reform Act of 1832 receives insufficient attention if the aim is to generate a persuasive account of path-dependent historical development. For a detailed study of the period at the center of Chafetz’s narrative, see generally David L. Smith & Graham E. Seel, Crown and Parliaments, 1558–1689 (2001). This is in some tension, obviously, with Chafetz’s claim that specific institutional settlements are “without binding implica­tions for the future” (p. 20). The two positions can be reconciled only if one posits that history’s relevance is not “binding.” But then, some account must be offered of how and why ancient and partially recollected institutional settlements have the potential to influence contemporary controversies—a question to which I will return in Part IV. 113 See infra text accompanying notes 222–226.

Second, with some important exceptions addressed below, the case studies have an implicitly teleological flavor. Thus, while Chafetz acknowl­edges some back-and-forth in institutional development, there is never­theless an overall impression of positive evolution toward more and more desirable institutional forms. For example, Chafetz tells of the struggle between the King and Parliament over the latter’s procedural workings and then portrays state constitutions as embodying lessons learned from that history (p. 278) and the U.S. Constitution as “evinc[ing] a similar desire” (p. 279). That is, British history holds lessons—and those lessons were absorbed and successfully acted upon in the new world. More generally, the tenor of his accounts is consistent with a gradual unfolding of an immanent rationality of institutional design—a story with more than a trace of a teleological rationality. The impression of learning and improvement though history infuses the constitutional-politics account with a generally optimistic tone. 114 See supra text accompanying notes 94–97.

Third—and in some tension with the previous observation—Chafetz does not limit himself to describing the positions reached in the context of specific, “local[]” disputes (p. 20) over the course of his historical narratives. Rather, he repeatedly demurs to the manner in which specific local disputes have been resolved (see, e.g., pp. 120, 182, 255). 115 This is to be distinguished, however, from Chafetz’s implicit view that the system as a whole yields beneficial and legitimate outcomes rather than catastrophe. See supra text accompanying note 97. Hence, in discussing Congress’s decisions in the late twentieth century to seek the support of the federal judiciary in extracting information from the executive notwithstanding claims of executive privilege, he notes critically that “Congress thus simultaneously diminish[es] its own standing in the public sphere and enhance[s] the courts’ standing” (p. 195). When discussing judicial precedent respecting the scope of the President’s neces­sary Article II authority to terminate officials, Chafetz criticizes formalist decisions by the Supreme Court limiting Congress’s power to structure the federal bureaucracy on the ground that they “rob Congress of central elements of its ability to structure and monitor government offices” (p. 147). 116 An example of such a formalist decision is Myers v. United States, 272 U.S. 52, 116 (1926); see also Bowsher v. Synar, 478 U.S. 714, 722 (1986) (invalidating budgeting lock-box mechanisms designed to control budgetary excesses). Consistent with Barkow and Waldron, see supra text accompanying notes 63–64, it is worth asking whether formalism on behalf of some other branch’s power (e.g., the judiciary’s) merits a different result. It is not clear from Chafetz’s analysis what he would say on this score. An implication of these passages is that the theory being brought to bear to generate the case studies does not hold that “political behav­iors and interactions” are the sole determinants of desirable outcomes in separation of powers conflicts (pp. 16–17). 117 Neither of the aforementioned condemnations, moreover, can be explained by labeling the relevant branch’s behavior “injudicious.” Some outcomes, even if sought by Congress, are normatively unmoored. An implication of this is that there must be an exogenously given normative benchmark against which local and contingent outcomes are measured—a benchmark that does not emerge immediately on superficial examination of the constitutional-politics account.

Fourth, notice that the two aforementioned examples of institutional teleology gone awry, both of which concern the Supreme Court, hinge on the inapt interventions by the federal courts. So when should judges intervene, consistent with the constitutional-politics account? It is hard to say. There is ambiguity in Congress’s Constitution as to how, and indeed whether, courts should resolve separation of powers disputes. At times, Chafetz is fiercely critical of the judicial role and presses against justiciability (see, e.g., p. 182). At other times, he seems to recognize that some legal questions must inevitably come before a judge. For instance, he concedes that courts have a “perhaps greater” judicial role in respect to Speech and Debate Clause questions because “members are most likely to be ‘questioned’ in a courtroom” (p. 226). 118 See also pp. 258–59 (writing with approval of the manner in which a judicial chal­lenge to the exclusion of Rep. Adam Clayton Powell was handled). And he contends that Congress “should consider leaning more heavily” on its authority to directly enforce its contempt power through arrest and imprisonment (p. 198). Presumably, the ensuing imprisonments would be amenable to a court’s review via a habeas corpus action of the kind used to challenge earlier imprisonments by the sergeant-in-arms (see, e.g., p. 178).

Aside from such ad hoc judgments, however, there is no general account of when separation of powers questions should be justiciable before an Article III bench. To be sure, Chafetz could fairly say that this is simply not his project. He doesn’t quite do this, though. Rather, he resists the need for an internal account of the separation of powers juris­prudence on the ground that no judicial opinion is ever final because “there is no separation-of-powers equivalent of the Supremacy Clause” such that “massively iterated” conflict over a legal question can continue even if there has been a putatively legal settlement (p. 19). I do not find this a compelling answer. If judges are important actors in the separation of powers system, as Chafetz doesn’t deny, it would seem needful to have some account of their behavior. But it is not plausible to assume that judges are “political” in the same way as legislators or presidents. 119 I do not read Chafetz as making this implausible claim; rather, I am pressing on the undernourished quality of his account of the federal judiciary. Something more needs to be said about why and how they reach their decisions.

Moreover, in my view, most contemporary decisions by the Supreme Court—except those that track abiding and emotional fractures of public opinion—are treated by most participants in American politics as legally dispositive. It is rarely worth anyone’s time to exert the necessary political effort to alter them. Even if these opinions are undermined by resistance from other branches or even lower court judges, they are not typically “massively iterated” thereby through continuing conflict in the public sphere except for the occasional effort to reopen an issue before the Supreme Court. 120 For a case study of a Supreme Court opinion on the scope of constitutionally mandated habeas corpus relief under Article I, Section 9’s Suspension Clause that was largely negated in practice—but through lower-court defiance and executive foot-dragging—and yet remains a binding rule of law, see Huq, President and Detainees, supra note 87, at 501–11. There is, for example, far less contestation now over recess appointments since the Supreme Court has addressed the ques­tion. 121 See NLRB v. Noel Canning, 134 S. Ct. 2550, 2560 (2014) (addressing the “scope of the phrase ‘the recess of the Senate’” (emphasis in original)). In consequence, I take judicial supremacy to be more settled than Chafetz suggests. 122 There is no generally accepted metric of judicial supremacy, and anecdotal evi­dence of judicial supremacy (or its absence) would not be decisive. Rather, I suspect that Chafetz and I simply have a reasonable disagreement about the normative heft and rooted­ness of judicial supremacy today. Although I agree with Chafetz that judicial supremacy is on balance less desirable than many think, I recognize that my view is a minority position. I also believe, just as a descriptive matter, that there is far more judicial settlement of separation of powers questions than he seems to allow and, in particular, that the Court’s opinions on the separation of powers exercise a far greater gravitational influence on the deliberations of non-judicial actors that he allows. 123 This position is consistent with the now increasingly common claim that the Court’s “constitutional decision-making is inextricably intertwined with the will of the people, channeling the views of political and popular majorities in numerous ways.” Corinna Barrett Lain, Soft Supremacy, 58 Wm. & Mary L. Rev. 1609, 1612 (2017). Indeed, the interaction of judicial and popular preferences helps explain the durability of judicial settlement in the first instance. Perhaps this is unwise; I think it is, and I infer that Chafetz thinks the same. But without a systematic challenge to the glacial grasp of judicial supremacy on the American constitutional imagination, Chafetz’s shrugging off of Article III settlement is not descriptively compelling.

Congress’s Constitution also skirts the Article III standing question raised when an individual files suit on the ground that a government action causing a cognizable harm should be enjoined not because the harm cannot lawfully be inflicted on her but rather because the action violates some separation of powers protocol. The Court has answered this question in the affirmative. 124 See Bond v. United States, 564 U.S. 211, 223 (2011) (stating that “individuals . . . are protected by the operations of separation of powers”). An eccentric minority of scholars has objected. 125 See, e.g., Huq, Standing, supra note 87, at 1514–23 (recommending reforms to Article III standing doctrine). Because it offers no account of when courts should intervene in this crucial swathe of cases, the constitutional-politics account leaves ambiguous the question of which of the legal issues at stake would be resolved through the give-and-take of informal political tussling and which would be resolved through the more formal, canalizing instrument of judicial review.

The “when” of judicial review matters in part because it helps clarify thinking about the “how” of judicial review. The constitutional-politics account offers some recommendations as to when courts should step aside (p. 182). 126 For instance, I read Chafetz as suggesting that courts should play a much less aggressive role enforcing presidential control over subordinates and that they should not be gatekeepers for legislative requests for information from the executive. For my argu­ment in favor of these same outcomes on institutional competence grounds, see Huq, Negotiated Structural Constitution, supra note 61, at 1595–600. But it is framed as a description, albeit a normatively freighted one, of how separation of powers disputes are resolved from what Professor H.L.A. Hart called the “external . . . point of view” of “an observer who does not himself accept them.” 127 H.L.A. Hart, The Concept of Law 89 (1994). More precisely, Chafetz takes a “hermeneutic” point of view, which Scott Shapiro defines as “seek[ing] to describe the law . . . by reference to the insider’s point of view,” in particular by “attending to the attitudes of the members of the group.” Scott J. Shapiro, What Is the Internal Point of View?, 75 Fordham L. Rev. 1157, 1160 (2006). Yet judges do not decide cases from an external point of view. They rather understand themselves as “member[s] of the group which accepts and uses [legal rules] as guides to conduct.” 128 Hart, supra note 127, at 89. It is not clear whether Chafetz thinks that judges should be using the same analytic approach to separation of powers questions as he endorses. And it is far from clear that they could or should. In general, there are powerful reasons why the operation of constitutional canons and analytic methods would vary by institutional context that Congress’s Constitution does not address. 129 See, e.g., Aziz Huq, Enforcing (but Not Defending) “Unconstitutional” Laws, 98 Va. L. Rev. 1001, 1001–05 (2012) (distinguishing between executive branch obligations in respect to individual rights and structural constitutional questions); Trevor W. Morrison, Constitutional Avoidance in the Executive Branch, 106 Colum. L. Rev. 1189, 1239–58 (2006) (suggesting different treatment of the avoidance canon in the executive branch context).

This absence of systematic answers to the “when” and “how” questions of judicial review is puzzling. The constitutional-politics account explains how “nonhierarchical interbranch contestation,” when handled “judiciously,” leads to desirable outcomes (pp. 18–19) but says little about when “judicious” also means “judicial.” So long as courts deploy different sources of law from the ones Chafetz endorses, the theory will be off-kilter. Thus, even if courts rely on interbranch practice in resolving dis­putes, they additionally turn to “[t]ext, history, structure, and precedent” (p. 16) in ways that the constitutional-politics account declines to endorse. In consequence, where judicial review begins (wherever that may be), the theory falls silent, or at least sings with a muted tongue.

III. Metatheory for the Separation of Powers

How do we adjudicate between different theories of the separation of powers? And how do we determine where and how Chafetz’s theory of constitutional politics is a persuasive addition to their ranks? Both tasks require a perspective that sits beyond or above any one individual theory from which multiple theories can be evaluated. This sort of metatheoreti­cal evaluative exercise—“meta” in the sense that it is an exercise in theorizing about theories—might be pursued from one of two perspec­tives. First, metatheory might proceed by establishing an independent vantage point from which various theories could be compared and contrasted. 130 For the use of metatheory in this sense, see Dorf, supra note 11, at 597–98; Pursley, supra note 11, at 1337. A 1990 student note uses the framing device of “metatheor[ies] of separation of powers” but does so to refer to what I have called theories of the separation of powers. Edward Susolik, Note, Separation of Powers and Liberty: The Appointments Clause, Morrison v. Olson, and Rule of Law, 63 S. Cal. L. Rev. 1515, 1527 (1990). The sense of metatheory here is also somewhat distinct from what Brian Leiter calls “theoretical disagreement,” which involves “disagreeing about what most legal philosophers call the criteria of legal validity.” Brian Leiter, Explaining Theoretical Disagreement, 76 U. Chi. L. Rev. 1215, 1216 (2009). The second species of metatheoretical disagreement I adumbrate below, concerning the sources of law, has this character. But my sense of the term “meta­theory” is broader. In the separation of powers context, however, it is not at all clear there is shared ground of this kind. Theorists disagreements might be both deep (in the sense of arising from conflicting first principles) and wide (in the sense of reaching many discrete questions of law). The second, more modest form of metatheoretical inquiry asks instead if there are empirical or normative concepts that are common to one or more schools of thought. Rather than attempting a synoptic judgment of the validity of any given theory of the separation of powers, this sort of metatheoretical inquiry focuses on the more granular and manageable task of adjudicating between the different ways in which a given empiri­cal or normative question is resolved across the different theories. That is, by isolating the core questions of fact or value that are motivating disagreement, a metatheoretical analysis tries to cut to the quick of a constitutional question.

As noted, there is an analogy to the field of metaethics in philosophy, which “is concerned to answer second-order non-moral questions, includ­ing (but not restricted to) questions about the semantics, metaphysics, and epistemology of moral thought and discourse.” 131 Terry Horgan and Mark Timmons, Introduction, in Metaethics After Moore 1 (Terry Horgan & Mark Timmons eds., 2006). Exemplary questions in metaethics, for example, include whether there are such things as moral facts or whether knowledge of such facts is possible. Thankfully, the metatheoretical questions raised by the separation of powers debate are a bit more tractable. Indeed, isolating these cross-cutting concepts—and considering the different ways in which they are specified—provides new purchase on the evaluation of different theoreti­cal claims in Congress’s Constitution. This is not to say that metatheoretical analysis will obviate the need for normative choice. How one answers a given metatheory question will often turn at least in part on how one frames the normative ambitions of constitutional law, in addition to one’s more general her­meneutic tendencies. But even if it is not possible to reach a conclusive judgment on which theory is superior given the persis­tence of larger normative debates in respect to constitutional law, a focus on metatheory has the potential to at least narrow the scope of dis­agreement about the separation of powers.

I perceive three important metatheoretical questions in the thicket of separation of powers theories canvassed above. These concern the model’s assumptions about the motives of official actors, the relevant sources of law, and the model’s assumptions about dynamics of equilib­rium or change. The first of these is a fundamentally empirical dispute, even if some authors characterize it in nonempirical, theoretical terms. 132 Perhaps the most influential example of a bare motivational assumption is found in Justice Scalia’s famous dissent in Morrison v. Olson, in which he complains that the majority’s decision to permit legislative limits on the president’s removal authority shows that “the Court does not understand what the separation of powers, what ‘[a]mbition . . . counter­act[ing] ambition’ is all about, if it does not expect Congress to try them.” 487 U.S. 654, 726 (1988) (citation omitted) (quoting The Federalist No. 51 (James Madison)). The second is profoundly normative. The final metatheoretical question rests on an empirical prediction about the way a complex governmental system will behave. I discuss each in turn.

A. The Question of Motive

Motive is often thought to be a question that looms large only in the individual constitutional rights context. This is a mistake. Any plausible theory of the separation of powers needs an account of official motives, whether or not it admits as much.

All agree that theories of separation of powers set out to describe, or to offer prescriptions for, a set of ongoing institutions. All also agree that these institutions are comprised of individual officers. No institution acts except through its individual officers, and when we talk of “Congress” or “the presidency” as a reified, singular noun, we lose sight of this basic fact. A theory of government institutions such as the separation of powers is therefore a theory of the behavior of the discrete set of individuals who assume the roles of legislators, presidents, bureaucrats, judges, and the like. A theory of the separation of powers that has nothing to say about what motivates these individuals—their dispositions, preferences, beliefs, or ambitions—is thus a theory with a hole where its heart should be.

The central role of structural constitutional law is to “create or define new forms of behavior” in the form of new institutions—and it is only secondarily that such law can “regulate a pre-existing activity, an activity whose existence is logically independent of the [relevant constitutional] rules.” 133 John R. Searle, Speech Acts 33–34 (1969); accord Christopher Cherry, Regulative Rules and Constitutive Rules, 23 Phil. Q. 301, 303 (1973) (criticizing Searle’s approach as “unsatisfactory” due to “oscillating between [regulative rules and constitutive rules]”). The centrality of this task of delineating actual, formal institu­tions distinguishes the separation of powers from many other domains of constitutional law—although arguably not federalism. Unlike the separation of powers, many other constitutional doctrines are plainly not constitutive of legal institutions in the first instance. The First Amendment’s speech and religion clauses, for example, presuppose the existence of an institutional press and organized churches. 134 For an analysis of the role of institutions in First Amendment law, see Frederick Schauer, Is There a Right to Academic Freedom?, 77 U. Colo. L. Rev. 907, 925 (2006) (“An institutional understanding of the First Amendment is structured around the principle that certain institutions play special roles in serving the kinds of values that the First Amend­ment is most plausibly understood to protect.”); Frederick Schauer, Towards an Institutional First Amendment, 89 Minn. L. Rev. 1256, 1268–69 (2005) (criticizing framing of the First Amendment as an individual entitlement). Similarly, whereas the Fourth Amend­ment’s prohibition on unreasonable searches and seizures may primarily regulate the police, it does not create the police. 135 Indeed, the urban police force postdates the Bill of Rights in the United States by some four decades. David A. Sklansky, The Private Police, 46 UCLA L. Rev. 1165, 1202–08 (1999) (discussing historical origins of the police). In defining institu­tional roles and authorities, a theory of the separation of powers must tell us something about why officials hew to those roles and how they use those authorities.

More concretely, the Constitution’s first three Articles articulate a series of roles into which specific individuals slot. The efficacy of such institutional articulation depends on the behavioral fact that specific individuals indeed do assume the roles limned in those Articles and then assert attendant powers. The Constitution, in other words, is akin to a play: It provides official actors with a bare script. It does not compel a particular interpretation of that script. Moreover, like any great perform­ance text, the Constitution leaves a good deal of room for many inter­pretations. It allows for Garricks and Oliviers, as well as Tommy Wiseaus. The motives of a specific individual determine how she responds to the role and powers constituted by one of these roles. They determine how powers will be deployed to combat or cooperate with other branches. And they shape how legal constraints will be embraced or resisted.

Rough generalizations can be offered to describe the assumptions about official motives that animate the separation, balance, and exogenous models. For instance, separation models implicitly assume that the law permits and limits official behavior within each branch. The implication is not crisply spelled out, but it would make no sense to offer propositions about the extent of congressional war powers, 136 Prakash, Sweeping Domestic War Powers, supra note 33, at 1367. or the directive authority of the president in relation to principal officers, 137 Calabresi & Prakash, supra note 27, at 544. in the absence of an expectation that the specific individuals were motivated in some substantial measure by the law—that is, motivated to exercise their duties and fulfill their roles as specified by the relevant legal material. The balance model, by contrast, often rests on the Madisonian assump­tion that “[t]he interests of the man must be connected with the consti­tutional rights of the place,” 138 The Federalist No. 51, at 319–20 (James Madison) (Isaac Kramnick ed., 1987) (advocating “giving to those who administer each department the necessary constitutional means and personal motives to resist encroachments of the others”). such that individuals staffing each branch should be expected to act on the basis of the interests of their home institution. Finally, the exogenous models discussed above 139 See supra text accompanying notes 68–87. take the view that officials act on political or partisan motives, which are framed as distinct and different from legal motives. The precise quality of a “political” motive can be rendered in different ways. For example, Professors Jack Goldsmith and Daryl Levinson predict the exploitation of the weak by the strong. 140 Goldsmith & Levinson, supra note 78, at 1833. Others seem to gesture at a more conventional, but more optimistic, account of electoral responsiveness consistent with the canonical accounts of political representation in the political science literature. 141 For a classic model of electoral responsiveness, see David R. Mayhew, Congress: The Electoral Connection 12–14 (1974). Yet other scholarship evinces skepticism about the effect of legal and con­stitutional bonds as a general matter. 142 See Frederick Schauer, The Political Risks (If Any) of Breaking the Law, 4 J. Legal Analysis 83, 91 (2012).

The role of motive in separation of powers jurisprudence, to reiter­ate, is to a large extent distinct from its function in other domains of constitutional law. 143 Note the qualifying phrase. I am cognizant that Akhil Amar, among others, has insisted on the structural quality of the Bill of Rights. But his account of the Bill of Rights does not accent the branches per se but rather a host of distinct, extrinsic actors: “states’ rights and majority rights alongside individual and minority rights; and protection of various intermediate associations—church, militia, and jury—designed to create an educated and virtuous electorate.” Akhil Reed Amar, The Bill of Rights as a Constitution, 100 Yale L.J. 1131, 1132 (1991). If Amar is correct (and I think he is to some extent), it may be that some of the metatheoretical questions raised here deserve airing more broadly in constitutional theory. Consider the prohibitions found in the Bill of Rights and Fourteenth Amendment. Several provisions in that part of the Constitution make motive, or intent, a touchstone of constitutional infir­mity. 144 See Aziz Z. Huq, What Is Discriminatory Intent?, 103 Cornell L. Rev. (forthcoming 2018) (analyzing the role of intent in the context of the Religion Clauses and the Fourteenth Amendment). In effect, such prohibitions pick out a species of motivation familiar from daily experience and treat it as a taint fatal to state action. Such rules exclude some actions on the basis of motive but otherwise have no implications for the range of legitimate official motivations. In other regulative contexts, motive does not bear on the validity of government action but pertains to the availability of a judicial remedy. For example, Justice Elena Kagan famously argued that First Amendment doctrine “comprises a series of tools to flush out illicit motives and to invalidate actions infected with them.” 145 Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 414 (1996). In contrast, in Fourth Amendment jurisprudence, the motivation of a state agent is not relevant to the constitutionality of their action. 146 Heien v. North Carolina, 135 S. Ct. 530, 539 (2014) (stating that a court “do[es] not examine the subjective understanding of the particular officer involved” in a challenged policing action). Rather, questions of intentionality become relevant when the Court turns to the question of whether a remedy is available, such as the exclusion of evidence or damages for a constitutional tort notwithstanding qualified immunity. 147 See Huq, Judicial Independence, supra note 87, at 20–40 (documenting the pervasive use of a “fault” standard of objective culpability across constitutional remedies). To the extent that courts engage “remedial equilibration,” whereby the costs and benefits of various remedies shape the contours of individual rights, it is with a measure of embarrassment and obfuscation. 148 See Daryl J. Levinson, Rights Essentialism and Remedial Equilibration, 99 Colum. L. Rev. 857, 873 (1999) (defining “remedial equilibration” as a process in which “constitutional rights are inevitably shaped by, and incorporate, remedial concerns”). For reasons Professor Richard Fallon has forcefully outlined, this view is “unduly reductionist” and inconsistent with “widely shared assumptions of the legal profession.” Richard H. Fallon, Jr., Judicially Manageable Standards and Constitutional Meaning, 119 Harv. L. Rev. 1274, 1314–15 (2006). In consequence, the domains of rights and remedies remain meaningfully distinct and not just a nice lawyer’s fiction.

It is only in the separation of powers context—where the causal effects of legal design upon official motives is central rather than inci­dental to the enterprise 149 Federalism, in contrast, generally takes the states as they are and considers how the federal government then acts upon them. —that the law plays the role of a careful and diligent gardener of souls, seeding and cultivating the springs of official action with the ambition of realizing a healthful and vigorous institu­tional ecosystem.

B. The Sources of Law

Separation of powers theorizing varies widely in the choice of sources from which claims about the law are derived, as well as the hierarchical scale along which such claims should be ordered. These divisions reflect disagreements in constitutional theory writ large. Famously, Professor Philip Bobbitt offered a quasi-canonical list of six “modalities” of argument in constitutional law that operate as “the grammar of law, that system of logical constraints that the practices of legal activities have developed in our particular culture.” 150 Philip Bobbitt, A Constitutional Interpretation 12–13, 24 (1991). Each of Bobbitt’s modalities turns on different sorts of evidence. Some are controversial. 151 Consider, for example, the status of precedent. Compare Henry Paul Monaghan, Stare Decisis and Constitutional Adjudication, 88 Colum. L. Rev. 723, 748 (1988) (“Precedent is, of course, part of our understanding of what law is.”), with Gary Lawson, The Constitutional Case Against Precedent, 17 Harv. J.L. & Pub. Pol’y 23, 24 (1994) (“In [some] circum­stances . . . the practice of following precedent is not merely nonobligatory or a bad idea; it is affirmatively inconsistent with the federal Constitution.”). Separation of powers jurisprudence is hardly quarantined from these disputes. The tides, undercurrents, and squalls that afflict the wider scholarship and jurisprudence of constitutional interpretation hence upset the stability of separation of powers theory and law too. Still, debate over the sources of law in the separation of powers context has some bespoke qualities.

To see this, consider an official assigned a novel role and new powers, whether as a result of an election or an appointment. The nature of her new roles and powers necessarily depends, in part, on the existence of a shared understanding of the force of the Constitution’s first three Articles. A person with a wholly idiosyncratic understanding of their role—for example, a president who believed that she had authority to exile or execute political opponents—would not be cognizable (at least for now) as a constitutionally legible actor. 152 This is not to say that all claims of legal authority or prohibition that fall beyond shared understanding of a constitutional rule set are illegible within the constitutional system or irrelevant to its operation. To the contrary, claim-making and behavior at the boundary of the legal system can shape subsequent understandings of the law’s context. See generally Eric A. Posner & Adrian Vermeule, Inside or Outside the System?, 80 U. Chi. L. Rev. 1743, 1766–75 (2013) (explaining this phenomenon and its consequences). Law plays a central role in that determination, and its perceived sources therefore matter. Even if there is a repertoire of shared understandings about what a legislator or a president or a judge can or should do, the ensuing bundles of rules and powers is also likely to be alternatively incomplete, ambiguous in scope, or overtly contested in places. 153 There is a useful analogy to the sociological concept of “habitus,” which is the “system of dispositions which acts as a mediation between structures and practice,” Pierre Bourdieu, Cultural Reproduction and Social Reproduction, in 3 Culture: Critical Concepts in Sociology 63, 64 (Chris Jenks ed., 2003), or to the related ideas of “culture” as a “tool kit” for certain “strategies of action” and that “both the influence and the fate of cultural meanings depend on the strategies of action they support.” Ann Swidler, Culture in Action: Symbols and Strategies, 51 Am. Soc. Rev. 273, 273, 284 (1986). In the separation of powers context, when discretionary powers dominate specific ministerial obligations, these notions provide useful guides for thinking about the relation of law to observed behavior. Again, law provides the terms upon which such ambiguities are resolved. Its sources matter. Given that officials are always asking themselves under conditions of uncertainty whether they have legal authority to take an action and whether they are prohib­ited by law from an action, their understanding of the relevant sources of law is likely to be an important determinant of their actions. 154 Perhaps the best example of this is the debate over the first Bank of the United States. See, e.g., 1 Annals of Cong. 1940–2012 (1931) (Joseph Gales ed., 1834) (reporting the debate preceding enactment of the An Act to Incorporate the Subscribers to the Bank of the United States, ch. 10, 1 Stat. 191 (1791)).

A key question in practice is how shared understandings of legal constraints and authorities change. Here, the question of sources of law intersects with the question of official motives in the following way. On the account of law associated with Hart (which I think is largely correct), the litmus test to ascertain whether a norm is a rule of law as opposed to say, a convention, a norm of political life, or coincidental empirical regularity, is whether it is understood and accepted as law by a relevant group of legal officials. 155 Cf. Jules L. Coleman, Rules and Social Facts, 14 Harv. J.L. & Pub. Pol’y 703, 722 (1991) (“[A] particular rule of recognition that is the rule of recognition in a particular community is a social fact about that community.”). I take no position here on Professor Joseph Raz’s more demanding “sources thesis,” which pertains to the permissible content of a rule of recog­nition. See Joseph Raz, The Authority of Law: Essays on Law and Morality 212 (1979). It is, therefore, a matter of social practice whether the official can appeal to one or another source of law. 156 This is a controversial proposition. Scholars who offer strong, narrowing claims about the appropriate sources of law are likely to resist the idea that the validity of their claims is not a matter of objective facts but of coherence with a social practice that they are actively trying to shift or curate. Their view is a legitimate internal perspective on the law and not a matter of false consciousness. Appealing to new or different sources of law is an obviously important way of changing shared understandings of law’s content. Because social practice is neither static nor absolutely constraining, an official newly invested in one of the three branches has a degree of freedom about whether to hew closely to the already-inscribed contours of a role based on extant and uncontro­versial sources of law, or whether to kick against the pricks of convention by asserting newfangled interpretations of novel sources of law. 157 Cf. Acts 26:14 (King James) (“Saul, Saul, why persecutes thou me? It is hard for thee to kick against the pricks.”). Her choice among these options, in turn, depends on her motivations.

There are some specific ways in which disputes about the sources of law in the separation of powers context diverge from such debates more generally. To begin with, consider the range of sources that validly inform a constitutional determination. 158 See, e.g., Cass R. Sunstein & Adrian Vermeule, Interpretation and Institutions, 101 Mich. L. Rev. 885, 886 (2003) (describing interpretive techniques and suggesting that more focus should be placed on the role of institutional capacities in legal interpretation). It is quite possible, even probable, that separation of powers decisions are systematically decided by actors who are not deciding other constitutional questions. In particular, it is likely that nonjudicial actors as opposed to judges will play a larger role because separation of powers questions are, all else being equal, less likely to be justiciable than the analogue questions of constitutional law. It is also likely that relevant actors systematically differ in the sources of law they deploy and the manner in which they prioritize them. As a result of these vari­ances, separa­tion of powers jurisprudence will likely be characterized by a different distribution of source materials than other domains of consti­tutional law.

More generally, there is a wide margin of disagreement about the extent to which the text (perhaps situated within its original Founding Era context) can provide determinative answers. 159 See Adam M. Samaha, Levels of Generality, Constitutional Comedy, and Legal Design, 2013 U. Ill. L. Rev. 1733, 1739 (noting that “the list of relevant constitutional sources is not fully settled” and “[a]greement is solid for many source categories (such as clauses, history, and cases), but thoughtful people disagree over what else should be on the list (such as contemporary understandings, foreign law, and acquiescence)” (footnotes omitted)). Such disagreements, to be sure, exist in other domains of law. There are exceptionally creative scholars and judges who dicker with the original understanding of “free speech” as a prohibition on prior restraints and instead brocade a magnificent, extensive lattice of limits on both ex ante and ex post speech regulations out of the thin warp of original text and practice. 160 See, e.g., Akhil Reed Amar, Foreword: The Document and the Doctrine, 114 Harv. L. Rev. 26, 55 (2000) (rejecting the “honest reading of the Constitution [that] leaves us with a First Amendment preventing prior restraint but providing no other protection of free expression”). But disagreement is especially profound in the separation of powers domain. To scholars working within the separation tradition, the resolving power of the first three Articles’ fine-spun and careful arrangement is crystal clear; to those in the balance tradition, it speaks in more muffled tones. 161 Compare Calabresi & Rhodes, supra note 2, at 1167–68 (advancing a textual case for a unitary executive at length), with Lessig & Sunstein, supra note 29, at 2 (describing that case as “just plain myth”). To my mind, there is a qualitative difference between this dis­pute about the resolving power of text and its original meaning and parallel disputes in other fields. By and large, my reading of contempo­rary scholarship leaves me with the view that free speech, Fourth Amendment, and equal protection jurisprudences are by comparison far less preoccupied by the meaning of words in their eighteenth-century context than the literature on the separation of powers, although it is worth underscoring that this is a difference in degree and not in kind. 162 And the scholarship may change in the future, perhaps in response to changes in the perceived preferences of the Justices.

In addition, relevant historical practice—a source that plays a large if ambiguous role in Chafetz’s account—takes a different form and has a distinct valence in the separation of powers context from other constitu­tional domains. Historical practice can be a probative instantiation (and hence affirmance) of sound constitutional meaning that, at the limit, works as a definitive, even immobile “fixation” of that meaning. 163 Curtis A. Bradley & Neil S. Siegel, After Recess: Historical Practice, Textual Ambiguity, and Constitutional Adverse Possession, 2014 Sup. Ct. Rev. 1, 41 (discussing “[t]he idea of fixation through long-standing acceptance of a practice”); Nelson, Originalism, supra note 39, at 535 (“Although Madison conceded that the words used in the Constitution might well fall out of favor or acquire new shades of meaning in later usage, he was sug­gesting that their meaning in the Constitution would not change; once that meaning was ‘fixed,’ it should endure.”). Alter­natively, historical practice can operate as an aversive precedent, a dark mirror from which current constitutional practice properly recoils. 164 Cf. Kim Lane Scheppele, Aspirational and Aversive Constitutionalism: The Case for Studying Cross-Constitutional Influence Through Negative Models, 1 Int’l J. Const. L. 296, 312–17 (2003) (describing how totalitarian, fascist, and Nazi governments functioned as aversive precedent for the U.S. Supreme Court in developing the modern constitutional law of criminal procedure). Other axes of variation include timing (for example, pre- or post­ratification), institutional location (for example, executive branch practice, versus sequential legislative action, versus judicial precedent), and stability (that is, whether a practice was contested and whether that contestation, in the courts or otherwise, proved successful).

The kind of historical practice relevant to the separation of powers will differ from the kind of practice relevant in other domains. Outside of structural constitutional law, current constitutional law is replete with aversive historical precedent that orients current law by providing an example of what government should not do. 165 For a discussion of the aversive precedent in the criminal procedure context, see William J. Stuntz, The Substantive Origins of Criminal Procedure, 105 Yale L.J. 393, 396–403 (1995). Aversive or negative prec­edent often illuminates the prohibitive force of regulative constitutional provisions such as individual rights. In contrast, it is less useful for glossing the constitutive rules that make up the bulk of the separation of powers. Rather, as a recent flurry of academic interest attests, the idea of historical practice of some sort, or “gloss,” plays a particularly important role in the separation of powers context. 166 See, e.g., Bradley & Morrison, Historical Gloss, supra note 14, at 418–19. To be sure, the term “historical gloss” is imprecise insofar as it might refer to “custom, tradi­tion, prescription, or something else.” 167 Alison L. LaCroix, Response, Historical Gloss: A Primer, 126 Harv. L. Rev. F. 75, 77 (2013), [
MAR2-8Z42]; see also Curtis A. Bradley & Neil S. Siegel, Historical Gloss, Constitutional Conventions, and the Judicial Separation of Powers, 105 Geo. L.J. 255, 262 (2017) (noting problems of threshold categorization, level of generality, and the sufficiency of interbranch acquies­cence in the use of historical gloss).
But even acknowledging that imprecision, it remains the case that positive historical practice plays a larger role in determining the distribution of powers between branches than in other constitutional domains.

Again, the three different theories mapped in Part I have different metatheoretical predicates when it comes to the relevant sources of law. Separation theorists tend to find more meaning embedded in the text’s seams. 168 See supra text accompanying notes 26–34. They are also inclined to find in the practice of the early Republic a reference point that allows them to resolve the ambiguity of a constitutional term. This approach to the early Republic as a storehouse of dispositive examples can rest on an implicit assumption that politicians and officials in the early Republic tended to be morally more uncompli­cated, as well as more legalistic, than politicians today. Such heroic assumptions are not necessarily well grounded in historical fact. In contrast, balance theorists are less willing to find a dispositive textual settlement. Typically, they are more concerned with later practice. 169 See supra text accompanying notes 49–62. They are also more likely to embrace more recent policy innovations as appropriate responses to long-term trends. In contrast, exogenous models that hinge on politics of various forms generally attribute relatively little effect to law, sidelining the sources-of-law question. 170 See supra text accompanying notes 67–84. Yet law implicitly remains central to their accounts insofar as they model officials as occupying and acting on the basis of roles defined by law as opposed to charismatic or traditional sources of authority.

Along other metrics, the sources of law used in separation of powers law are not distinctive from those typically observed in other parts of con­stitu­tional law. For instance, practice-based claims abound across the constitu­tional waterfront based on preratification state, colonial, or English practice. 171 Prakash & Ramsey, supra note 32, at 265–78 (examining preratification materials). Arguments are also often grounded in the behavior of officials in the early Republic. 172 See, e.g., Saikrishna Prakash, New Light on the Decision of 1789, 91 Cornell L. Rev. 1021, 1022 (2006) (describing Congress’s decisions about how to structure depart­ments under the President for the first time in 1789 as one of “the most significant yet less-well-known constitutional law decisions”). And just as practice-based settlements—or their judicial analogue, written precedents—can prove to be of variable endur­ance in the separation of powers context, 173 Huq & Michaels, supra note 5, at 357–77 (describing doctrinal instability in the separation of powers domain). so, too, can legal settlements by judicial precedent or practice outside that domain be of varying quality and robustness. These cross-cutting continuities in debates about the sources of law unite most of constitutional law and should not be ignored.

C. The Choice Between Equilibrium and Change Models

An account of the separation of powers requires a theory to predict institutional change or its inverse, a stable institutional equilibrium. Al­though the resolution of constitutional questions outside the domain of separation of powers necessarily raises some questions of chronological situation and interrelation, the role of time—and change over time—in the separation of powers context is distinctive. It is properly analyzed sep­arately from the consideration of time in other fields of constitutional law.

The central temporal question in many fields of constitutional law concerns the level of generality at which a constitutional principle is understood. 174 See Tribe & Dorf, supra note 18, at 1058 (making this claim in the fundamental rights context). That is not to say that the idea of levels of generality is a simple one. Cf. Samaha, supra note 159, at 1743 (flagging three ways in which the level of generality can be calibrated: “abstractness, breadth, and dynamism”). That is, the Constitution can be interpreted as a relatively specific settlement of contestable institutional design questions. Alterna­tively, it could be viewed as “an initial framework for governance that sets politics in motion and must be filled out over time.” 175 Jack M. Balkin, Framework Originalism and the Living Constitution, 103 Nw. U. L. Rev. 549, 550 (2009). Temporality has little role in constitutional interpretation if a relevant legal concept is defined at a low level of generality: Specific rules stand fast against the relentless erosions of time. In contrast, when the Constitution is under­stood as a relatively general settlement picking out relatively abstract values, inevitably questions arise as to how best to evince “fidelity” to those principles over time. 176 For the canonical account of “fidelity” as a concept that requires changing appli­cation over time, see Lawrence Lessig, Understanding Changed Readings: Fidelity and Theory, 47 Stan. L. Rev. 395, 396–401 (1995). To give just one example, some scholars have argued that the fundamental rights protected under the Equal Protection Clause should be defined in relatively abstract terms, which will tend to result in new and perhaps unanticipated applications of the right over time. 177 See, e.g., Tribe & Dorf, supra note 18, at 1058 (“The more abstractly one states the already-protected right, the more likely it becomes that the claimed right will fall within its protection.”). Others, predictably, disagree. 178 See, e.g., John Safranek & Stephen Safranek, Finding Rights Specifically, 111 Penn St. L. Rev. 945, 979 (2007) (“The Supreme Court must eschew general rights and seek refuge in principled specification.”).

In one respect, the same question of levels of generality does come up in the context of the separation of powers. For instance, Manning has characterized the central question of all separation of powers jurispru­dence as simply a matter of which level of generality is appropriate in reading the Constitution’s text. 179 Manning, Separation, supra note 35, at 2005–07. Consistent with Manning’s framing, many (but hardly all) separation of powers disputes can be redescribed as questions about whether to understand an issue at a specific or an abstract level of generality. Opting for abstraction forces the question of how to translate an eighteenth-century idea two-plus centuries forward in time. 180 See Lessig & Sunstein, supra note 29, at 4–5 (arguing that the unitary executive is justified based on the application of the Framers’ structure of government translated into the modern era). This is, importantly, not a problem distinctive to the separation of powers. For instance, in defining a fundamental right under the Equal Protection Clause or defining one of the rights in the 1791 Amendments, an interpreter might opt for a specific or an abstract understanding of the relevant interest.

But the separation of powers implicates the choice between equilibrium and change in yet another distinct way. Specifically, the separation of powers compels the following question: How (if at all) does the arrangement of institutions created by the Constitution’s first three Articles maintain its stability over time? And when instead is the “time out of joint,” 181 William Shakespeare, Hamlet act 1, sc. 5 (“The time is out of joint; O cursèd spite!/ That ever I was born to set it right!”). and so in need of repair? What conjoins the separation of powers circa 1789 and the separation of powers circa 2018? The separa­tion of powers constitutes a series of formal institutions inhabited by specific individuals acting on diverse motives over time. Given that, there is inevitably a question of whether those individuals’ cumulative actions will push institutional arrangements out of their initial roles, or out of alignment, over time. Further, there is a question of whether the prospect of such disequilibration should provoke consternation or alternatively be embraced.

A simple model of the separation of powers can illuminate how the question of equilibrium arises over time. A person vested with a role or powers by dint of the Constitution’s first three Articles must decide how to deploy her newfound authority. Depending on her motives and under­standing of valid sources of law, this person will either hew to past understandings of roles and the appropriate deployment of powers or, alternatively, innovate in her conception of the role and her deployments of official authority. Her choice might create the risk of changing, and hence durably destabilizing, existing institutional arrangements. Predicating the effects of such choice, moreover, is complicated by the fact that background social, economic, and political conditions are themselves unstable. Hewing to a historical model of appropriate behavior given a changing external context might itself be destabilizing. Forgetting the lesson of Heraclitus’s river, one might accidently oust the old in favor of an unknown new via an obdurate fidelity to obsolete behaviors. 182 Flaherty makes such an argument to support the legislative veto, stating:
“Given that balance was a primary purpose for dividing government author­ity, and given further that the executive has supplanted the legislature as the branch posing the greatest threat to this balance, it follows that any jurist faithful to the past should applaud, not deride, legislative attempts to maintain that balance, especially when those attempts appear in part of a package delegating still more power to the executive.”
Flaherty, supra note 47, at 1834.

The question whether the separation of powers should be understood as a stable equilibrium or as a dynamic system is squarely acknowledged and confronted in the political science literature. It is largely ignored in the theoretically inclined legal scholarship canvassed above. As I discuss below, 183 See infra text accompanying notes 240–248. political scientists and historians have explicitly diverged over whether the constitutional system should be characterized as a stable equilibrium that persists over time or whether it should be seen as an evolving composite of institutional formation moving at different temporal registers. 184 Compare Sonia Mittal & Barry R. Weingast, Self-Enforcing Constitutions: With an Application to Democratic Stability in America’s First Century, 29 J.L. Econ. & Org. 278, 278–79 (2011) [hereinafter Mittal & Weingast, Self-Enforcing Constitutions] (modeling the constitutional system as a stable “self-enforcing” system), and Barry R. Weingast, The Political Foundations of Democracy and the Rule of Law, 91 Am. Pol. Sci. Rev. 245, 246–48 (1997) (same), with Karen Orren & Stephen Skowronek, The Search for American Political Development 108–18 (2004) [hereinafter Orren & Skowronek, Search for American] (describing the “intercurrence” of varying political regimes). Equilibrium models are also suited to describe single institutions such as legislatures. See Kenneth A. Shepsle, Institutional Arrangements and Equilibrium in Multidimensional Voting Models, 23 Am. J. Pol. Sci. 27, 27 (1979) (offering a model of legislative behavior that results in “equilibrium”). No parallel formulation can be located in the legal scholarship.

Nevertheless, to offer another rough generalization, separation theo­rists tend to assert that the stability of legal forms in the separation of powers is desirable. They despair when that stability is perceived as not only abandoned but also beyond plausible grasp. 185 See, e.g., Lawson, Rise and Rise, supra note 45, at 1254 (concluding, morosely, that “one must choose between the Constitution and the administrative state”). In contrast, balance theorists perceive a need for shifting specific institutional arrangements as a means to preserve the systemic, architectural qualities they value. 186 See, e.g., Greene, supra note 50, at 128 (arguing in favor of permitting the legis­lative veto once more as a way to “restore a proper balance of powers consonant with the framers’ view of checks and balances”). Finally, exogenous theorists such as Posner and Vermeule tend to embrace change—and in particular the metastatic growth of the execu­tive branch—as a welcome adaption to the complexity of governance in the contemporary United States. 187 Posner & Vermeule, Executive Unbound, supra note 4, at 16 (praising the increasing power of the executive in light of “the rapidity of change in the policy-making environment”).

Varying judgments over the extent of separation of powers stability, finally, are likely to correlate with differences in judgments about rele­vant sources of law. The latter can be “characterized as relatively dynamic or relatively static,” depending on whether they “require or forbid rethinking [the] proper specification or application.” 188 Samaha, supra note 159, at 1753. In this way, the different metatheoretical questions are yet again entangled with each other.

D. The Uses of Metatheory

I have proposed three metatheoretical categories of interest: matters of motive, sources of law, and equilibrium versus change. In isolating these three cross-cutting variables, I’ve tried to identify questions that any theory of the separation of powers must answer and that extant theories do answer, whether implicitly or explicitly. I have not, though, tried to nail down the doctrine’s “purpose” in the sense of identifying a specific normative value immanent in the constitutional scheme as a touchstone of analysis. Possible “purposes” of the separation of powers include efficiency, the prevention of tyranny, democratic accountability, and deliberation. Purpose-based arguments are common. 189 See, e.g., Jessica Korn, The Power of Separation 14–26 (1996) (emphasizing effec­tive governance and prevention of tyranny as goals of separation of powers); Jack M. Balkin & Sanford Levinson, Understanding the Constitutional Revolution, 87 Va. L. Rev. 1045, 1059 (2001) (observing “[t]hat the separation of powers . . . may have the specific purpose of promoting a dialogue among different voices even with regard to foreign policy issues”); Sunstein, supra note 56, at 432–33 (describing efficiency and prevention of tyranny as two chief purposes of separation of powers). But they are also singularly unfruitful as starting coordinates for mapping theoretical disagreement about the separation of powers. This is because a theory of the separation of powers will rarely be bottomed on the claim that there is one, and only one, value that explains all separation of powers controversies. Rather, plausible theories of the separation of powers need to acknowledge the fact that the design of national government institutions reflects many normative ends pursued simultaneously. 190 See Huq & Michaels, supra note 5, at 382–91 (describing the “normative pluralism” of the separation of powers). The multiplicity of ends and the absence of any obvious way to make trade-offs between them means that merely invoking the “purposes” of the separation of powers rarely does much meaningful analytic work. Instead, separation, balance, and exogenous mod­els—as well as the constitutional-politics model of Congress’s Constitution—are better understood as analytic devices for working through the neces­sary trade-offs between the normative ends implicated in separation of powers law.

To be clear, an analytic quarantining of metatheoretical claims generates no quick and direct solution for the interminable disputes over the proper specification of the separation of powers. Certainly, it yields no immediate resolution of intractable first-order issues. One reason for this is that the three metatheoretical questions identified here differ in tractability. There is a fact of the matter, however difficult to extract, about the motives of discrete government officials. Those motives might also change over time. 191 See infra text accompanying notes 195–202. It is less clear whether motive then becomes a fact of the matter about the sources of law at a given historical moment. If one follows Hart in positing that the ultimate “rule of recognition,” a rule that picks out valid sources of law, is a convergent social practice among officials, 192 Hart, supra note 127, at 116 (characterizing the “rules of recognition as specifying the criteria of legal validity and its rules of change within a system of laws”); Kenneth Einar Himma, Situating Dworkin: The Logical Space Between Legal Positivism and Natural Law Theory, 27 Okla. City U. L. Rev. 41, 68–69 (2002) (glossing Hart to hold that “the behavioral element that gives rise to the rule of recognition is convergent behavior on the part of the officials in making, changing, and adjudicating law”). then careful empirical inquiry might unpack the precise crite­ria used to identify the law. Nevertheless, at least in the American context, it seems tolerably clear that there is a high degree of scholarly and juristic disagreement about both the potentially relevant sources of constitu­tional law and also the appropriate rule of priority when evaluating competing pieces of evidence. 193 Fallon, How to Choose, supra note 8, at 537 (noting the “large and proliferating number of constitutional theories” available to jurists and scholars). Legal scholars often participate in that disagreement even as they purport to objectively describe it. 194 For one example of a consideration of how “some of the benefits of theory” can be translated “to practice,” see Neal Kumar Katyal, Hamdan v. Rumsfeld: The Legal Academy Goes to Practice, 120 Harv. L. Rev. 65, 67 (2006). The most comprehensive, if eclectic, consideration of the fraught relation of the judiciary to the academy is Richard A. Posner, Divergent Paths: The Academy and the Judiciary (2016). Finally, the descriptive claim that a constitutional system is in equilibrium implicates a mix of both empirical and normative judgments. At minimum, it entails a normative specification of the relevant criteria along which stability is defined and the bounds of permissible fluctuation within homeostasis. That is, one must say what kind of change counts—and why.

Complicating this analysis yet a bit more, my three proposed meta­theoretical substrates interact in complex ways. For instance, the quality of officials’ motives is likely to inform their attitude toward the law. As a result, it will likely inflect their judgment of what constitutes a valid source of law. Motive and the sources of law are hence endogenous. Moreover, the extent of constitutional equilibrium will mold officials’ incentives and the plausible sources of law. Where institutional change as a consequence of exogenous partisan or social pressures is perceived as proceeding at a rapid rather than gentle clip, for example, officials may be disinclined to invest deeply in legal knowledge. Older sources of law become less trenchant. Motives, in sum, change, and perhaps with them the relevant sources of law.

Nevertheless, attention to these metatheoretical questions is a way of gaining better purchase on precisely why and how scholars disagree about the separation of powers. It is also a pathway to better understanding and appreciating new contributions to the scholarly literature such as Congress’s Constitution.

IV. Metatheory for Congress, and its Constitution

Is the constitutional-politics account founded upon a set of con­sistent and defensible presuppositions in respect to the empirical facts of institutional behavior and the normative questions raised by observed regularities in official conduct? This Part applies the tripartite framework of metatheoretical inquiry to the constitutional-politics account. Its ancil­lary ambition is to demonstrate the utility of an analysis that focuses upon the predicate assumptions of separation of powers theory. This task is aided by the fact that Chafetz, unlike the other leading separation of powers theorists, is admirably candid and clear about his methodological presuppositions. The very fact of such candor is in itself one of Chafetz’s contributions to the quality of jurisprudential debates in this domain. In this endeavor, it makes most sense here to begin with the question of what motivates official actors before turning to questions of sources of law, and then the choice between equilibrium and change as a baseline assumption.

A. Motive

The question of official motive in the theorization of the separation of powers is a difficult and subtle one. In my view, the constitutional-politics account runs into three difficulties respecting such motives. First, Chafetz does not address how official motives are likely to vary substan­tially by time and institutional locus. Second, the question of what counts as “power” in Chafetz’s account interacts with his understanding of official motive in ways that undermine the latter’s plausibility. Finally, I consider the way in which Chafetz links public-facing motives with desirable, “judicious” outcomes (p. 311)—and suggest that the linkages are unsub­stantiated and untenable.

1. Motives in Time. — It is helpful to begin with a simple, if destabilizing, observation: The motives and preferences of official actors in a branch of government do not remain static over time. Nor can it be assumed that those motives are constant between branches. Consider for example the narrow question of whether members of Congress are likely to engage in debate about, and be influenced by, legal and especially constitutional matters. On the one hand, Professor David Currie’s sedu­lous historical work has demonstrated that early congresses seriously debated a “breathtaking variety of constitutional issues great and small.” 195 David P. Currie, The Constitution in Congress: The Federalist Period, 1789–1801, at 296 (1997). More recently, a former representative and federal judge concluded that “legislative debate does not explore the constitutional implications of pending legislation; and, at best, Congress does an uneven job of consid­ering the constitutionality of the statutes it adopts.” 196 Abner J. Mikva, How Well Does Congress Support and Defend the Constitution?, 61 N.C. L. Rev. 587, 587–88, 610 (1983). The leading work in political science also suggests that legislators pursue a variety of ends simultaneously, sometimes trading ends off against each other. 197 The classic work is Richard F. Fenno, Jr., Congressmen in Committees (1973). For a rare example of legal scholars that consciously adopt Fenno’s insights, see Elizabeth Garrett & Adrian Vermeule, Institutional Design of a Thayerian Congress, 50 Duke L.J. 1277, 1288 (2001) (treating “legislators as maximizing a complex utility function, in which constitutional considerations are one argument”). Consti­tutional concerns do not take consistent priority, and the strength of transient, nonlegal concerns will ebb and flow over time.

The manner and extent to which members of Congress attend to various legal and nonlegal ends, moreover, changes over time. For the legislative institution itself develops in ways that change the incentives and resources of its members. 198 Changes to the external electoral environment—in the form of different forms of campaigning, campaign financing, the varying effect of partisanship with the choice of primary structure—all add further complications to an effort to predict legislators’ motives. See R. Douglas Arnold, The Electoral Connection, Age 40, in Governing in a Polarized Age: Elections, Parties, and Political Representation in America 29 (Alan S. Gerber & Eric Schickler eds., 2017) (emphasizing that “[w]hat has changed—and what will continue to change—is the turbulence of the political waters through which legislators navigate their careers” and noting several new forms of complexity in the electoral environment). Given the complexity already identified, I bracket these considerations henceforth. As Professor Michael Gerhardt has explained, the range of opportunities that members of Congress have for exercising judgments in respect to constitutional questions depends on the institutional structures of congressional deliberation. Gerhardt un­derscores the role of not just committees but also “informal practices, norms, and traditions.” 199 Michael J. Gerhardt, Non-Judicial Precedent, 61 Vand. L. Rev. 713, 739 (2008). These channels for deliberation on legal and constitutional issues have shifted over time. As Chafetz explores in some detail, “committees were an important part of cameral ordering from the beginning” but have undergone dramatic changes since 1789 (p. 282). 200 See also Fenno, supra note 197, at 94–114 (exploring the historical importance of congressional committees). More generally, the “industrial organization of Congress” reflects a dynamic “environment within which legislators bargain with one another in order to facilitate their individual and col­lective goals.” 201 Daniel B. Rodriguez & Barry R. Weingast, The Positive Political Theory of Legislative History: New Perspectives on the 1964 Civil Rights Act and Its Interpretation, 151 U. Pa. L. Rev. 1417, 1431–32 (2003); see also Barry R. Weingast & William J. Marshall, The Industrial Organization of Congress; or, Why Legislatures, Like Firms, Are Not Organized as Markets, 96 J. Pol. Econ. 132, 132–37 (1988) (providing a theory of legislatures based on the theory of firms and contractual institutions). This environment does not stay still. In the words of Professor David Mayhew, examining “two and a quarter centuries” of congressional operation, what “impresse[s]” is “the system’s flexibility, its variety, its capacity to turn on a dime.” 202 David R. Mayhew, The Imprint of Congress 99 (2017) [hereinafter Mayhew, Imprint of Congress].

Substantially parallel points could be made about the judiciary and the executive. Neither has remained static over time. Changes in their internal operations modulate the strength of legalistic compulsions in comparison to other motivational drives. This is obviously true with respect to the executive but also true of the judiciary. Hence, even in the absence of a “separation-of-powers equivalent to the Supremacy Clause” (p. 19), the Supreme Court has ebbed and flowed in the extent to which it has asserted a final authority to settle disputed questions of constitutional law. 203 On the historical genealogy of these claims and their frequency within the Court’s docket, see Aziz Z. Huq, When Was Judicial Self-Restraint?, 100 Calif. L. Rev. 579, 584–86 (2012) (summarizing the path of judicial review from 1800 to 2000). The institutional apparatus, in terms of administrative operations, lobbying power, and sheer, on-the-ground infrastructure available to the judiciary has dramatically changed over time. 204 For a synoptic historical account, see generally Justin Crowe, Building the Judiciary: Law, Courts, and the Politics of Institutional Development (2012). This has changed its capacity to take the Constitution seriously. It has also altered the range of voices that can influence its decisionmaking process. 205 The most high-profile form of this expanding openness is the increase, beginning in the 1970s, in the number of amicus briefs filed with the Court. Donald R. Songer & Reginald S. Sheehan, Interest Group Success in the Courts: Amicus Participation in the Supreme Court, 2 Pol. Res. Q. 339, 339–40 (1993). Amici influence the Court in part by providing information the parties would not otherwise present and by signaling the range and depth of interest-group engagement on an issue. Paul M. Collins Jr., Friends of the Court: Examining the Influence of Amicus Curiae Participation in U.S. Supreme Court Litigation, 38 Law & Soc’y Rev. 807, 807–08 (2004).

A theory of the separation of powers that seeks to move from a descriptive account of how the branches do behave to a prescriptive vision about how they should behave needs to come to terms with the mutability and endogeneity of official motives. To his credit, Chafetz does not claim that there is one single kind of motive that explains official action across time and between branches. 206 This is not to say Chafetz says nothing about official motives. For example, Chafetz rejects the claim tendered by Levinson and Pildes that partisanship dominates the calculations of members of Congress (pp. 28–35). He instead asserts that “members of Congress do, on some occasions, care about their chamber’s power, per se” (pp. 30–31), even as they are shaped by “[t]he motivating power of partisanship” (p. 35). While the latter claim can fit nicely within Chafetz’s dialogic account of constitutional politics, his recognition that institutional loyalties do occasionally bite is less easily nestled within his account. It is unclear why constitutional politics would ever generate such dispositions. For a discussion of institutional motives, see David Fontana & Aziz Z. Huq, Institutional Loyalties in Constitutional Law, 85 U. Chi. L. Rev. 1 (2018). But he also does not explain how motives change over different institutional contexts. Nor does he address the question of motivational change from a dynamic and endog­enous perspective. 207 But it is worth noting that some of the historical narratives bespeak not “constitutional politics” at work but a relatively narrow-gauge version of individual self-interest. For example, when Member of Parliament Richard Strode persuaded his colleagues to enact Strode’s Act, annulling several money judgments against him and barring actions that “vexed or troubled” a member, we can be fairly confident mere public spiritedness was not at work (p. 203). What results from this is a gap in his constitutional-politics account of separation of powers.

2. Public Competition for Power. — To the extent there is a singular motive that seems to animate the inhabitants of constitutional politics, it seems to involve the quest for public trust as a means to “give the institu­tion more power” (p. 21). That is, Chafetz centrally posits political actors as engaged in a dialogic process of building and expending public trust, leading and being led, by their constituents (see, e.g., p. 22). Officials, on this view, are motivated by the desire to win the trust of the public insofar as the latter enlarges their discretionary policymaking authority (id.).

But the notion of political power that is central to Chafetz’s analysis is, to say the least, an unorthodox one. Power is commonly understood as the capacity to exercise influence over another so that she takes actions that would otherwise not be undertaken. 208 See Steven Lukes, Introduction, in Power: Readings in Social and Political Theory 1, 2 (Steven Lukes ed., 1986) (citing Max Weber’s definition of power as “the probability that an actor in a social relationship will be in a position to carry out his own will despite resistance, regardless of the basis on which that probability rests”); id. (citing Robert Dahl’s idea that “A has power over B to the extent that he can get B to do something that B would not otherwise do”); cf. Elshtain, supra note 93, at 136 (“X has power over Y if he can get Y to do something Y would not otherwise do.”). As I understand him, Chafetz conceptualizes power as a sort of license to exercise discretion in relation to a baseline state of compulsory compliance with popular opinion. That is, power is conceptualized as freedom from public opinion, even as Chafetz acknowledges that political leaders also shape public opinion. At the very least, this definition of power is orthogonal to—and thus ignores—the extent to which the institutional design of a branch either enables or inhibits actions. For example, on this definition of power, the growth in federal infrastructure—in the sheer number of officers, personnel, and funds—over the twentieth century is not itself a source of power but a result of power. Likewise, the Supreme Court’s authority to define its own certiorari docket 209 See Evarts Act, ch. 517, § 6, 26 Stat. 826, 828 (1891) (“And excepting also that in any such case as is hereinbefore made final in the circuit court of appeals it shall be competent for the Supreme Court to require, by certiorari or otherwise, any such case to be certified to the Supreme Court for its review and determination . . . .”); see also Act of Dec. 23, 1914, ch. 2, 38 Stat. 790; Judiciary Act of 1925, Pub. L. No. 68-415, §§ 239, 240, 43 Stat. 936, 938–39. That authority, moreover, is but one element in a long series of institutional transformations that have augmented the federal judiciary’s freedom of maneuver since the late eighteenth century. See Huq, Judicial Independence, supra note 87, at 53–55 (summarizing that history in terms of comparative institutional power). —and hence to exercise control over the range of issues the Justices resolve or avoid—is apparently not a source of “power.” 210 See Aziz Z. Huq, The Constitutional Law of Agenda Control, 104 Calif. L. Rev. 1401, 1433–38 (2016) (discussing the effect of the judiciary’s agenda-control powers). This seems quite implausible. 211 Cf. p. 41 (“[T]he American judiciary is undeniably powerful.”); id. at 333 n.88 (describing the judiciary as “at the apex of . . . the entire American political system”).

Contra Chafetz, it seems to me that not all “power” (as that term is commonly understood) flows from a relationship with the demo­cratic public. There are many elements of institutional design, including sheer institutional size, and subtle agenda-control tools, that meaning­fully change power relationships. One implication of the point is this: If Chafetz is correct that official actors take decisions that substantially reflect a wish to accrue power, and if power is not exclusively pursued in the public sphere, then it follows that their strategies and behaviors cannot be understood in a way that is wholly “public focused” (p. 20). Rather, it would be necessary to attend to a far more complex “thick political surround” 212 Huq & Michaels, supra note 5, at 391–407. within and outside government to understand the ways in which institutional actors seek, acquire, and expend power. And if the sources of official power are not exhausted by public-facing dynamics, then the outcomes of institutional conflict will not be determined by popular democratic forces alone. The sources of power, therefore, cannot be limited in terms of their popular pedigree.

3. Public-Facing Competition and Judicious Outcomes. — Finally, a central claim of Congress’s Constitution is that officials’ reliance on public “opinion” (p. 22) will lead to “judicious, rather than maximal,” inter­branch contestation (p. 311). Again, it is important to stress that it is clear from context that “judicious” here is a term of normative approval, suggesting wise and objectively well-tempered behavior rather than behavior simply calibrated to win over the public. After all, it may be that “maximal” combativeness is, as an empirical matter, most likely to succeed in securing approval. Nevertheless, it is posited as distinct from “judicious” behavior. Similarly, when Chafetz states that “constitutional actors must be willing to press—albeit judiciously—their distinct representational claims” (p. 308), he is most plausibly (and indeed, only plausibly) under­stood to mean that constitutional politics will generate morally sound and well-tempered politics.

This asserted causal thread raises two concerns. The threshold difficulty is that it is not at all clear how often, or how powerfully, public opinion shapes official action in practice. Congress’s Constitution presents no empirical data on this point. It relies on what might charitably be called case studies. For instance, Chafetz employs the Senate hearings on the nomination of Judge Robert Bork to a seat on the Supreme Court to illustrate dynamic interactions between the public and officials over institutional roles (pp. 22–23). But it is telling that the subsequent case studies are not overflowing with parallel examples of overt public contes­tation over institutional power. Nor does Chafetz provide evidence that absolute levels of public opinion decisively influence official behavior. This is a point of some importance given recent work suggesting that changes in levels of public support, and not their absolute levels, influence official decisions. 213 See James A. Stimsom, Tides of Consent: How Public Opinion Shapes American Politics, at xx (2015) (“It is movement that matters. Politicians ask, ‘How will the public respond?’ . . . .” (emphasis omitted)). If these studies hold true, there is some cause for skepticism regarding the claim that free-floating public opinion will oper­ate as a stabilizing ballast to institutional behavior. If legislators care merely about avoiding downticks in public opinion, they gain considerable leeway in separation of powers interactions simply by supporting popular taxa­tion or spending measures to offset such variation. 214 In addition, one might draw attention to the large volume of political science scholarship that problematizes the idea of a “public” to which Congress is faithful, substituting in its place a critical portrait of a distinctively regressive pattern of democratic responsive­ness. See, e.g., Larry M. Bartels, Unequal Democracy: The Political Economy of the New Gilded Age 2 (2008); Martin Gilens, Affluence and Influence: Economic Inequality and Political Power in America 194–98 (2012); Jacob S. Hacker & Paul Pierson, Winner-Take-All Politics: How Washington Made the Rich Richer–and Turned Its Back on the Middle Class 73–91 (2010); Kay Lehman Schlozman, Sidney Verba & Henry E. Brady, The Unheavenly Chorus: Unequal Political Voice and the Broken Promise of American Democracy 580 (2012). Were I to further credit Chafetz’s theory of public-facing contestation, this would be the basis of a distinct objection.

The second difficulty abides in Chafetz’s assertion that our “separation-of-powers system privileges judicious, rather than maximal, combative­ness” (p. 311). Simply put, why should this be so? Nothing in Congress’s Constitution explains why the public should demand sensible forms of contestation and condemn hazardous ones (however defined). Chafetz’s account strongly implies that the public has some kind of innate sense of what counts as appropriate interbranch conduct, and thus catastrophic outcomes will rarely ensue. 215 See supra text accompanying note 94. Available empirical evidence, however, suggests that the public may value outcomes over institutional process. 216 See David Fontana & Donald Braman, Judicial Backlash or Just Backlash? Evidence from a National Experiment, 112 Colum. L. Rev. 731, 734 (2012) (concluding, based on national experimental evidence, that variance in normative judgments “can be explained by whether a given citizen views the Court’s decision or Congress’s legislation as threatening or privileging her core worldview”). Today, more than most instances in recent political history, claims that moderation will inevitably triumph over extremism ring very hollow. No “invisible hand” guarantees the wisdom or judiciousness of official action guided by public sentiment—at least unless one believes against the odds that the public is comprised of reasonably minded law professors. 217 For a definition of invisible-hand arguments, see Adrian Vermeule, The System of the Constitution 70, 73–80 (2011) (defining an invisible-hand argument as one in which an “overall system . . . produces a good that none of its components can individually produce, and that none of its components may even intend to produce,” and critiquing their use in legal scholarship). I hasten to add that I make no claim that law professors are all reasonable, let alone wise or judicious. This is, after all, not a work of fiction.

B. Sources of Law

The constitutional-politics account rests upon a novel and provocative account of the sources of law informing the separation of powers. At first blush, Chafetz’s account seems to be that the separation of powers is not accurately characterized as law. On his view, traditional sources of law “substantially underdetermine” the separation of powers, and “the most important” questions must be answered through “constitutional politics” (p. 16). The latter, in turn, are shaped by “public focused” competition among federal actors for “trust” (pp. 20–22). They are also decisively influenced by history in particular—hence the case studies that make up the body of the book—which is said to enable some options while foreclosing others (p. 4). All this seems refreshingly novel—even a touch radical—in its willingness to dispense with the rigidities of law and to embrace the quiddities of politics, even if historically infused. Chafetz, I think, wants to jettison the fusty antiquarianism of separation theorists, enfold the hard glint of realism so beloved of exogenous theorists, and yet find some measure of predictability and constraint.

This is an admirable effort to recognize the fact that both transient political considerations and legalistic concerns, to varying degrees at varying moments in time, seem to shape institutional behavior. That politics and law imbricate each other when it comes to the separation of powers, that is, seems to me correct. 218 I agree with the general proposition that both legalistic and political considerations influence official behavior. See Huq, Binding the Executive, supra note 85, at 789–804. But there are several points at which Chafetz’s specific formulation of how rigid legal rules and transi­ent politics interact is unsatisfactory. I will focus on two elements of the theory of constitutional politics that stick in my gullet: the process by which law becomes determinate in the absence of a precise text, and the role of political norms and conventions. By picking at these elements, I show that the theory of “constitutional politics” does not move us far beyond the quite general and familiar point that both law and politics in some form matter.

1. How Law Arises. — I have focused above on passages in Congress’s Constitution that evince skepticism about the stability and binding force of law. But it is not the case that Chafetz categorically rejects the possibility that coordinates for institutional contestation are fixed by law. Indeed, he recognizes a number of “relatively static [and, seemingly, determinate] separation-of-powers provisions in the Constitution” that are “most assuredly not subject to the play of politics” (p. 17). Elsewhere, he writes of politi­cal forces as “constituted and constrained by constitutional institutions” (p. 26). The observed political world, in short, is made up of patches of legal regulations and also patches in which “constitutional politics” has free play.

Contrary to his early and prominent protestations (see, e.g., p. 17), moreover, Chafetz does not in fact circumscribe narrowly the domain of law in favor of a broad dominion for constitutional politics. Even on Chafetz’s own book-length account, it is not the case that only “precise and rule-like” elements of the Constitution produce stable law in the separation of powers domain. Consider, for example, the tolerably clear legal rule recognized almost fifty years ago that the President benefits from an evidentiary privilege from certain forms of discovery—a privilege that finds no home in the constitutional text and yet is hardly contro­versial in its central applications today. 219 United States v. Nixon, 418 U.S. 683, 713 (1974) (holding that the President benefits from executive privilege, but this privilege does not supersede a criminal defendant’s con­frontation rights). Chafetz canvasses a familiar history of the initial creation of government departments and Tenure in Office Acts of 1820 and 1867 (pp. 100–22) as a basis to draw a general conclusion about Congress’s power to “structure and monitor government offices” (p. 147). Judges, he tells us, “have no explicit protection from being questioned in Congress,” and there is “no con­vincing” reason why they should (p. 197). And in the teeth of contrary views of the executive branch and the judiciary, Chafetz asserts that as a matter of law Congress has the authority to “arrest” executive branch officials for contempt of Congress and “hold them in . . . custody” (pp. 185–93). Congress’s Constitution, in short, is brimful of claims about what the law does and does not condone far beyond the limited terms of the constitutional text.

Furthermore, Chafetz’s case studies are trimmed with all the accou­terments of standard-form argumentation familiar to all constitutional scholars in a fashion that is quite at odds with Chafetz’s avowed abjuration of the traditional ways of searching for and fixing law (p. 16). On one page of Congress’s Constitution, for example, a reader finds in quick succession excerpts not only from the Constitution but also from the Articles of Confederation, James Wilson’s lectures, and Thomas Jefferson’s missives (p. 211; see also p. 241). The Federalist Papers, of course, play a comfortably familiar supporting role (see, e.g., pp. 57, 66, 99–100, 145, 279)—the very Statler and Waldorf of constitutional analysis. Select judi­cial precedent, when it fits Chafetz’s purposes, is described as resting on “forcible” ground (p. 215), even as other rulings are distinguished and dickered away (p. 120). 220 See also supra section II.B. (noting the force of history and the selection between endorsed and disfavored precedent). The result is that the case studies in particular look an awful lot like standard-form legal argumentation in the separation of powers tradition.

And yet if law arises beyond the perimeter of “precise and rule-like” elements of the Constitution, how is it that the ordinarily “underdeter­minate” (p. 16) legal materials upon which Chafetz relies generate hard-edged, and quite clearly determinate, answers? This is a critical question. But Chafetz provides no account of how the facts embodied in historical examples, consequential considerations, and current political contestation precipitate into law. That is, he does not explain the relationship between “Faktizität und Geltung,” or facticity and validity. 221 This is the German title of Jürgen Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (William H. Rehg trans., 1996). Habermas draws attention to multiple ways in which the factual order and the juridical normative one relate or come apart. See, e.g., id. at 39 (identifying an “external relation between facticity and validity” as the problem of “the facticity of legally uncontrolled social power that penetrates law from the outside”); see also id. at 82 (identifying the “external tension between the normative claims of constitutional democracies and the facticity of their actual functioning”). Habermas’s second specification is closest to the relevant sense of the distinction here. A theory that takes subsequent institutional practice as salient must offer some account of how the “fact” of an institutional settlement or stalemate is transformed into a normative postulate within the system of constitutional law that then exercises a binding force on intertemporally distant actors. It must answer the question of how “the realms of ‘real’ and ‘right’ interact in a social or political order over time.” 222 Peter L. Lindseth, Between the ‘Real’ and the ‘Right’: Explorations Along the Institutional–Constitutional Frontier, in Constitutionalism and the Rule of Law: Bridging Idealism and Realism 60, 60 (Maurice Adams et al. eds., 2017).

The question is, moreover, one that arises and must be answered separately for judicial precedent, 223 See, e.g., Randy J. Kozel, Precedent and Speech, 115 Mich. L. Rev. 439, 443–45 (2017) (summarizing justifications for precedent’s force). for historical gloss, 224 See, e.g., Bradley & Morrison, Historical Gloss, supra note 14, at 424–25 (exploring justifications for relying on gloss). and even for the normative force of “tradition” (in which the latter is defined as a pattern of historical behavior by governmental and nongovernmental actors). 225 See, e.g., Kim Forde-Mazrui, Tradition as Justification: The Case of Opposite-Sex Marriage, 78 U. Chi. L. Rev. 281, 281 (2011) (exploring the question in the equal protection context). Consider the relevance of historical settlements (or gloss, if you prefer) to law. On the one hand, Chafetz’s theory suggests that separ­ation of powers law is a seething and unstable broth of volatile com­pounds (pp. 18–19). The field of strategic action is in “constant flux” (p. 25). “[U]ncertainty, instability, and unpredictability” are the order of the day (pp. 312–14). Solutions, to the extent they are realized, exist only “locally and contingently” and “without binding implications for the future” (p. 20). This is the separation of powers as a quantum universe, in which perceptions collapse institutional wave functions into fixed par­ticulars and spooky (public) action at a distance takes the place of the smooth, Cartesian action of a Newtonian atomic universe. 226 For a strikingly different view of Congress that emphasizes “stability,” see David R. Mayhew, America’s Congress: Actions in the Public Sphere, James Madison Through Newt Gingrich 216–17 (2000).

But a famous property of quantum theory is that its rather uncanny effects—its instabilities, paradoxes, and inversion of intuitive causation—all vanish as one moves from the subatomic to the world of discernable clumps of atoms and molecules. 227 For an elegant explanation, see Brian Cox & Jeff Forshaw, The Quantum Universe: (And Why Anything that Can Happen, Does) 4–13 (2012). So it is too with a constitutional-politics approach as one moves from theory to observable studies of history’s salience. For Congress’s Constitution often treats history as evidence of the existence of a lawful authority pursuant to Article I. Its case studies are solidly upholstered, Victorian manses in comparison to the churning mire predicted by the theory. In discussing the contempt power, the historical proposition that “[a]lmost from the beginning the houses of Congress have . . . punished nonmembers” (p. 172) segues by imperceptible steps into the normative proposition that such power is “at least as essential for the houses of Congress” as for the courts (p. 180). It is hard to read this passage as doing anything other than deriving normativity from historical traces. And talk of how the English Bill of Rights “enshrined a strong concep­tion of the speech or debate privilege,” a conception “picked up” by American constitutional designers (p. 210), is superficially a claim about constitutional backdrops, but in substance a claim that history’s lessons have normative force today by dint of their role in motivating specific constitutional texts. All of these passages are in some tension with Chafetz’s early claim that antecedent settlements of institutional conflict are “with­out binding implications for the future” (p. 20). 228 There are moments when Chafetz might be glossed as migrating, widdershins, to the position held by Manning—to the effect that there is no law once one proceeds beyond the strict and narrow confines of specific text. Compare p. 12 (stating that “no doubt there are a number of . . . relatively stable separation-of-powers provisions . . . [b]ut a lot more is up for grabs than we commonly realize”), with Manning, Separation, supra note 35, at 1948–49 (contending for “a clause-centered approach” that rejects any “freestanding separation of powers doctrine”). But these moments are not consistently sustained.

Adding to the complexity of the problem, the question of “Faktizität und Geltung” can be framed in one of two, albeit somewhat overlapping, senses. On the one hand, the fact–law problem can be pondered from an internal perspective of normative justification: When should an institution’s decisions or judgment be entitled to a normative complexion? Why should a court’s decisions be viewed as normatively freighted in a way that, say, the decisions of many executive branch agents are not? 229 A domain in which this sort of question is encountered and grappled with directly concerns the boundaries of judicial deference to administrative agency decisions taken to have the force of law. See, e.g., United States v. Mead Corp., 533 U.S. 218, 230–32 (2001) (discussing when Chevron deference would be appropriate). A recently contested question of fact–norm mediation concerns the scope of Auer v. Robbins, 519 U.S. 452 (1997), and Bowles v. Seminole Rock & Sand Co., 325 U.S. 410 (1945), both of which concern agency interpretations of their own regulations. See, e.g., Christopher v. SmithKline Beecham Corp., 132 S. Ct. 2156, 2166 (2012) (finding Auer deference inappropriate “when the agency’s interpretation is ‘plainly erroneous or inconsistent with the regulation’” or “when there is reason to suspect that the agency’s interpretation ‘does not reflect the agency’s fair and considered judgment on the matter in question’” (quoting Auer, 519 U.S. at 461–62)); see also Perez v. Mortg. Bankers Ass’n, 135 S. Ct. 1199, 1208 n.4 (2015) (“Auer deference is not an inexorable command in all cases.” (citing Christopher, 132 S. Ct. at 2166)). But the prob­lem is quite general in character. On the other hand, inquiry might instead turn on the “sociological” question of when a specific institution’s decisions are accorded wider normative significance as evidence of how constitutional questions should be settled. 230 See, e.g., Richard H. Fallon, Jr., Legitimacy and the Constitution, 118 Harv. L. Rev. 1787, 1795 (2005) (“[A] constitutional regime, governmental institution, or official decision possesses [sociological] legitimacy in a strong sense insofar as the relevant public regards it as justified, appropriate, or otherwise deserving of support for reasons beyond fear of sanctions or mere hope for personal reward.”). The latter sociological inquiry necessarily turns on judgments about the normative dispositions and attitudes of participants in the legal system. But it is easily entangled in (or confused with) the first purely normative version of the inquiry.

In sum, Chafetz’s case studies implicitly reflect the application of a normative criterion to determine when history (and more) is relevant to law. But his account does not answer the question of how to move from historical fact to legal normativity in a constitutional system in which textual sources play a relatively modest role and in which there is sharp scholarly and public disagreement about how to rank and apply the shared norms of political life. 231 Cf. David A. Strauss, Foreword: Does the Constitution Mean What It Says?, 129 Harv. L. Rev. 1, 3 (2015) (“If we read the text of the Constitution in a straightforward way, American constitutional law ‘contradicts’ the text of the Constitution more often than one might think.”). Yet this (unanswered) question is crucial in delimiting the legal sources of Congress’s Constitution. 232 A related question that has not received as much attention as I think it deserves is whether and how the protocol for determining when historical facts about one branch’s behavior should be presumed to be identical to, or different from, the protocol for a different branch. One might, for example, ask whether the reception conditions for historical gloss and judicial precedent should diverge in some instances. Cf. Bradley & Morrison, Historical Gloss, supra note 14, at 427–28 (arguing that many of the rationales for adhering to judicial precedent “can support referring to nonjudicial precedent as well,” but treating judicial precedent and historical gloss separately).

2. Law, Politics, and Conventions. — The constitutional-politics account also raises a question regarding how best to conceptualize the boundaries of law and politics. Framing the matter of sources of law as a binary one is an intuitive choice: Either something counts as law or it does not. But there is no a priori reason such an impermeable boundary should be taken as given. British constitutional theorists have long employed a con­cept of “convention” to capture a more fluid idea of law’s boundaries. Hence, Judge A.V. Dicey identified “conventions, understandings, habits, or practices which, though they may regulate the conduct of the several members of the sovereign power . . . are not in reality laws at all since they are not enforced by the Courts.” 233 A.V. Dicey, Introduction to the Study of the Law of the Constitution, at cxl–cxli (8th ed. 1915). Conventions, as matters of “constitutional morality,” nevertheless played a central structural role in the British constitution system by determining “the way in which the pre­rogative powers of the Crown (and therefore of ministers) were exercised in practice.” 234 Geoffrey Marshall, What Are Constitutional Conventions?, 38 Parliamentary Aff. 33, 33 (1985). American legal scholars have begun of late to consider the role that such an analogous conception might play in U.S. constitutional law, albeit without the structural functionality of conventions in the British context. 235 See Adrian Vermeule, Conventions of Agency Independence, 113 Colum. L. Rev. 1163, 1184 (2013) [hereinafter Vermeule, Conventions] (noting, in the context of a more wide-ranging discussion, that “conventions may supply crucial context for the interpretation of written laws, and should thus be incorporated into that interpretation”); Keith E. Whittington, The Status of Unwritten Constitutional Conventions in the United States, 2013 U. Ill. L. Rev. 1847, 1855 (“Constitutional conventions are one mode of construction.”).

One way of parsing the ambiguity of the constitutional-politics account as to the status of past institutional settlements—and perhaps also a way to navigate the path from facticity to normativity—is to view the outcomes of past institutional conflicts as conventions that are capable of supporting a normative inference under the right conditions. Such con­ventions lack the binding force of law but are distinct from the epiphe­nomenal spume of daily partisan conflict. They further reflect a shallow, primitive judgment about the valence of some historical facts based on the practical morality of daily practice. Navigating the transition from historical fact to conventionality presents questions that are similar to, but easier than, questions surrounding the passage from fact to law. The questions are similar because in both instances the theorist seeks to derive an “ought” from an “is.” The questions diverge insofar as the weaker, peripheral nature of conventions arguably lowers the burden of justification needful to making that transition. Appeals to consequentialist justifications for lending precedential effect to past practice on the basis of their ability to operate as coordinating focal points may be more appealing than they would be in a strictly legal context. 236 Cf. Vermeule, Conventions, supra note 235, at 1192 (“In game-theoretic terms, judicial recognition of a convention may provide a focal point on which politicians may converge in an ongoing game with a coordination component, including sequential or iterated Prisoners’ Dilemma games.”).

C. Equilibrium and Change

In the constitutional-politics account, is the separation of powers in equilibrium, or is it an unstable and dynamic system? More colloquially, is it a theory of institutional change over time, or one of institutional stability? Separation and balance theories most clearly diverge over this question. Whereas separation theories tend to favor stability and the preservation of long-standing understandings of law, balance theories are more tolerant of institutional transformation and hybridization with an eye toward the vindication of more abstract, systemic values.

Again, the position of the constitutional-politics account is ambiguous depending on whether theory or case study is prioritized. As noted, the theory underscores and even celebrates “uncertainty, instability, and unpredictability” (pp. 312–14). 237 At its inception, the account also invokes the idea of “intercurrence” associated with Professors Karen Orren and Stephen Skowroneck (p. 4). But the critical element of that theory is the existence of institutional orders created at different historical moments that overlap onto each other, leading to a complex accretion of institutional forms at any one given moment of time. Karen Orren & Stephen Skowronek, Institutions and Intercurrence: Theory Building in the Fullness of Time, 38 Nomos 111, 138 (1996) [hereinafter Orren & Skowronek, Institutions]. Ideas of overlap and intermingling, however, do not figure prom­inently in the theoretical account offered in Congress’s Constitution. At the same time, the case studies, to a noteworthy extent, are fairly legible by the interpretive lights of ordinary constitutional jurisprudence as rather conventional accounts of structural constitutional law, replete with traditional trimming in terms of both sources and attendant normative judgments of a reassuringly democratic flavor. 238 See supra text accompanying notes 227–228. To my sublunary lawyer’s ear, the case studies provide the more compelling descriptive account. For it would be quite surprising if official actors were observed constantly renegotiating the rules of institutional engagement rather than relying on historical precedent to supply rules of the road. While imperfect, even considering their effects in medium term, the latter is inevitably cheaper than constant renegotiation. 239 For a similar point, see Levinson, Parchment and Politics, supra note 71, at 695 (“The coordination advantages of bundling multiple (probabilistic) policy decisions into a single institutional decisionmaking process are obvious.”).

To reconcile these themes by saying only that constitutional history is a mixture of stability and change would be to oust theory with unen­lightening platitude. One needs to say more about what remains constant, what mutates over time, and what forces drive the selection of institu­tional margins into one of these two categories.

An account of equilibrium versus change in the separation of powers context might therefore profitably begin by specifying the sense in which, and level of generality at which, the term “equilibrium” is deployed. The term is, in operation, ambiguous and requires a measure of clarification. Political scientists and political historians have engaged in useful polemics over the appropriate role of equilibrium concepts in theorizing the sepa­ration of powers, and legal scholars have much to learn from the distinc­tions that have emerged from this work. 240 For a useful account of the disputes, see Orren & Skowronek, Institutions, supra note 237, at 124–37. A brief summary of the use of equilibrium concepts to answer structural constitutional questions about institutional design shows that they are inadequate to explain the doc­trine in this area. Such concepts can be useful analogies but are not well suited to direct transplantation into the structural constitutional context.

Within the political-economy literature, two concepts of equilibrium are common: “preference-induced equilibrium” and “structure-induced equilibrium.” 241 Kenneth A. Shepsle, Institutional Equilibrium and Equilibrium Institutions, in Political Science: The Science of Politics 51, 52 (Herbert F. Weisberg ed., 1986). The former arises when “[a]n outcome, x, is [stable] if there exist[s] no y preferred to it by a decisive coalition of agents.” 242 Kenneth A. Shepsle, Studying Institutions: Some Lessons from the Rational Choice Approach, 1 J. Theoretical Pol. 131, 136 (1989). By contrast, the latter is an arrangement “that is invulnerable in the sense that no other alternative, allowed by the rules of procedure, is preferred by all the individuals, structural units, and coalitions that possess distinctive veto or voting power.” 243 Id. at 137. At its core, the first focuses directly on prefer­ences in isolation, whereas the second analyzes preferences within a given institutional context. In both theories, change to the institutional circumstances of choice is modeled as “episodic and homeostatic—a momentary transition between stable states.” 244 Orren & Skowronek, Institutions, supra note 237, at 124. In this fashion, they “effectively remove[] time” from the analysis. 245 Id. at 125. For the kind of separ­ation of powers theory that constitutional scholars are interested in developing, which is aimed, inter alia, at specifying the terms and limits of acceptable change, this is an unhelpful, even debilitating, limit on the analysis. 246 One exception is Mittal & Weingast, Self-Enforcing Constitutions, supra note 184. Squarely confronting the intertemporal question, Mittal and Weingast argue that a constitution will be self-enforcing if “at a given moment, all actors . . . find it in their interests to adhere to the constitutional rules, and as circumstances change, they must be able to adapt policies and institutions to maintain that system over time.” Id. at 282. Constitutions create “focal procedural and substantive limits” in order to enable popular enforcement, but also create “incentives [for officials] to search for and create solutions to new and pressing prob­lems.” Id. at 286. Whereas these conditions might be understood as sufficiently plausible general and abstract statements of the circumstances of democratic constitutional stability, it is hard to see how they facilitate the specification of particular structural design margins. Rather, I read Mittal and Weingast to instruct constitution-makers to create some provisions that don’t change, which are acceptable to enough people, and also to enable other provisions to change if and when needed. All well and good—but also easier said than done.

Scholars of institutional evolution, most notably political scientists working in the American Political Development (APD) stream that Chafetz briefly invokes (p. 4), repudiate static models of institutional interaction. Time, in their view, is instead “a construct of the intercurrence of institutions” in which “collisions and combinations, the changes and cycles, of institutions in their various . . . external relations” must be situated. 247 Orren & Skowronek, Institutions, supra note 237, at 141, 143. Institutions, on APD’s account, have a purpose or mandate, establish norms and rules, assign roles, operate within boundaries, and develop “norms and values [that] affect their members,” even as the institution itself remains “subject to innovation, redirection, disruption, and to all manner of personal motives of individuals.” 248 Orren & Skowronek, Search for American, supra note 184, at 82. For a helpful gloss on this passage, see Pamela Brandwein, Law and American Political Development, 7 Ann. Rev. L. & Soc. Sci. 187, 188–89 (2011) (discussing Orren and Skowronek, and the “project of American Political Development”).

An acute, and accurate, implication of such models is the futility of strictly static equilibrium analysis of the several branches. But it is hard to see how this strand of theorizing can provide a firm foundation for normative theorizing about whether a given status quo is desirable in comparison to imaginable alternatives. It is hard to see, that is, how it can yield prescriptions about what is and is not allowed. Nor is it clear how the theory can be used to sort among different historical conceptions of institutional form, labeling some as licit and others as undesirable. Absent any guidance on when a status quo is desirable, such theories merely predict change without attaching positive or negative valence to its various forms. As I have suggested in reference to the constitutional-politics account, APD supplies no basis for judgment about the present or potential futures. 249 Since historians do not aspire to such normative judgments as outputs from their work, this cannot be taken as a criticism. To the contrary, one might worry that any judgment about the practical entailments of an institutional change could be made only in the longue durée, one in which we still await a final reckoning on the French Revolution as well as its sister revolt on the other side of the Atlantic.

Legal scholars have successfully deployed APD’s account of institu­tions as a means of understanding historical periods of institutional con­flict. It is less clear how a normative theory of the separation of powers can be articulated on its basis. Some exogenous criterion must be invoked as support for normative judgments about which historical settlements are legitimate and which should be abandoned—as Congress’s Constitution, perhaps inadvertently, demonstrates. The theoretical compass that can draw a “circle just” 250 Cf. John Donne, A Valediction: Forbidding Mourning in Selected Poems 54–55 (John Hayward ed., 1950) (“Such wilt thou be to mee, who must/ Like th’other foot, obliquely runne;/ Thy firmness drawes my circle just,/And makes me end, where I begunne.”). to knit together in perpetuity the divergent arcs of stability and healthy institutional adaption has yet to be found.


Theorizing the separation of powers entails an effort to reason from observed facts and accepted norms to prescriptions about how state power should be parceled out across the federal government. Attention to metatheoretical questions in separation of powers law—that is, the theoretical assumptions about empirical regularities and normative judg­ments that serve as needful scaffolding to the theory—helps flush out questions of lingering difficulty in this enterprise. If the constitutional-politics account offered by Chafetz does not fully surmount those diffi­culties—and I don’t think it does—it is not for want of sophistication or ambition. To the contrary, the fact that his account grapples explicitly with the core difficulties of theorizing the separation of powers is to its large credit.

The first metatheoretical problem to pose a difficulty in separation of powers theory generally is the question of motives. How “branches” as composites behave depends on the motives of institutionally pivotal actors. 251 Note that in the legislative context in particular, collective action problems arise and are typically overcome by cameral rules, which concentrate agenda-setting and deliberation-structuring authority in a small number of actors. Kenneth A. Shepsle, Congress Is a “They,” Not an “It”: Legislative Intent as Oxymoron, 12 Int’l J.L. & Econ. 239, 245–48 (1992). Yet motives are mutative over time and across branches, as well as being endogenous to institutional design decisions. Assumptions that one particular kind of motivation (for example, ideological or reelection-focused) dominates are implausible as a ground for prescription. But—as the limits of Chafetz’s account demonstrates—it is quite difficult to build a theory on the shifting sands of motivational agnosticism. If, for example, congressional behavior is best characterized by “its flexibility, its variety, its capacity to turn on a dime,” 252 Mayhew, Imprint of Congress, supra note 202, at 99. the challenge of normative theorization seeking to identify optimal institutional bounds, stable interbranch equi­libria, or welfare-maximizing arrangements only grows. 253 The research agenda suggested by Magill—which would resist the urge to trans­historical and cross-institutional generalization—escapes this problem. See Magill, Beyond Powers and Branches, supra note 67, at 659. I view much of my work as belonging to this tradition.

The second metatheoretical inquiry implicates the choice of sources of law in a domain in which text is (perhaps as usual) indeterminate while evidence of historical practice overflows. I have suggested that a theory of the separation of powers must engage with several questions, beginning with a clarification of the extent to which the label of “law” as opposed to “conventions” or “politics” is appropriate. To the extent either law or convention has been identified on the basis of historical fact (for example, past practice or a specific decision), the further question arises of criteria for navigating from fact to legal norm. Different species of historical facts (for example, a judicial decision, a statute, a congressional com­mittee report, an executive decision to launch an overseas military action) might be subject to different criteria to ascertain their normative weight—but then some account of the ensuing interbranch differences in law-creating capacity is needed. At best, the cream of current scholar­ship offers a series of partial equilibrium models on these points. No Walrusian general equilibrium is in sight.

The final metatheoretical question concerns the extent of stability versus change in interbranch relations. Separation and balance theories are distinguished by the different levels of generality at which they answer this question. They are united insofar as neither explains why the institutions they sketch would honor those equilibriums once they are set off and running in the live space of political contestation. The constitutional-politics account instead draws fruitfully on APD’s richer toolkit of institu­tional intercurrence. But whatever the gain in descriptive accuracy, there is a loss in normative traction. Ultimately, efforts to theorize the separation of powers must grapple with the difficult question of how the law can create a set of institutions for the conduct of national politics that simul­taneously remain stable but also responsive to the inevitable efforts of different political coalitions to refashion the inchoate and underspecified terms of political engagement.

It must do so, moreover, while remaining alive to the possibility that past stability is an unreliable guide to future performance. In constitu­tional law, like the political life it regulates, there are no firm guarantees. Theories of the separation of powers that embed optimistic metatheoreti­cal assumptions about the extent of systemic stability, the certainty and force of law, and the predominance of benign motives merit especially careful and critical handling at a moment when the trajectory of our chief national institutions appears uncertain and up for grabs. 254 For a more plenary rehearsal of worries on this front, see generally Tom Ginsburg & Aziz Z. Huq, How to Lose a Constitutional Democracy, 65 UCLA L. Rev. (forthcoming 2018). In Congress’s Constitution, citizens and scholars who are committed to main­taining the best historical legacies of our Constitution as a vehicle for decent and civilized political conversation and settlement have an important resource. For that alone, Chafetz’s accomplishment deserves resounding applause.