This page has been accessed times since Nov. 13, 1996.

Originally published at: 88 Nw. U. L. Rev. 1420 (1994).

*1420

STILL NAKED AFTER ALL THESE WORDS

Copyright (c) 1994 A. Michael Froomkin

A. Michael Froomkin<*>

This Reply concentrates on three points which are emblematic of more general problems in The Structural Constitution,<1> A Structure Without Foundation,<2> and The Vesting Clauses as Power Grants.<3> First, Professor Calabresi and Mr. Rhodes seem to believe that their method of interpretation is apolitical and precise, when in fact it is overly formalistic, inconsistent at times, and vague at a critical point. Second, while Professor Calabresi's latest formulation of the "executive Power" is more detailed and perhaps less expansive than The Structural Constitution's, it is also circular, allowing courts and commentators unfettered discretion to define the "executive Power" at will. Third, the authors indulge in rhetorical excess, exaggerating the dangers of rooting federal jurisdiction in Article III, Section 2 and misstating the arguments in The Imperial Presidency's New Vestments.<4> Space constraints and concern for the reader's patience limit this Reply to but a few examples of each of these problems. New Vestments must in most cases be left to speak for itself. I *1421 am confident that it does.<5>

I. Excessive Formalism

Professor Calabresi and Mr. Rhodes seem to believe that armed with the right sort of dictionary, the Constitution can be read to provide code-like answers to an enormously wide range of difficult constitutional questions.<6> Like Justice Scalia, they write as if the right dictionary were the philosophers' stone of law, a device to turn conflicted terrain into neutral principles. Professor Calabresi's claim that Blackian structuralism is peculiarly subject to the danger of unconscious importation of political views is illustrative.<7> On the contrary, one of the many virtues of Blackian structural analysis is that it tends to make us more likely to consciously articulate the sources of our beliefs for all to see--something textual analysis too often tends to obscure.<8>

Similarly, Mr. Rhodes seeks to prove that the Vesting Clause of Article *1422 III is the only reasonable textual source for the federal judiciary's authority to act, by finding a distinction between the "judicial Power" (conferred by the Vesting Clause in Article III) and jurisdiction (delineated in Article III, Section 2).<9> The two terms are not inevitably synonymous, although the Supreme Court's usage suggests that Mr. Rhodes has it backwards,<10> or the terms are interchangeable.<11> Even if Mr. Rhodes's distinction exists, it cannot carry anywhere near the freight required to *1423 disprove the claim that Article III, Section 2 is a fully adequate textual hook for the federal judiciary's authority to decide cases: Mr. Rhodes's nonjurisdictional "judicial Power" is the power to abstain from decision.<12> In fact, a federal court has the jurisdiction (judicial power) to determine its own jurisdiction, but without jurisdiction a court has no "judicial Power." The "judicial Power" that Mr. Rhodes describes is jurisdiction, or else it flows from it.<13> If a court has jurisdiction, it has the power to decide; apparently it also has the power to decide not to decide, which from the litigants' perspective amounts to a decision as well.<14>

I do not believe that legal texts are usefully seen as inherently indeterminate; some legal questions do have answers that are beyond productive dispute so long as words have any meaning. If, however, this debate proves anything it is that people can disagree wildly on fundamental structural constitutional issues. Text is the agreed starting point, but The Structural Constitution and New Vestments propose quite different ways to read the text.

Professor Calabresi turns to his dictionary and finds that "vest" *1424 means to give power, but "judicial Power shall extend" defines the scope of a power without granting it because "extend" explains only the size of the power.<15> How this distinction works for a court's jurisdiction, which is granted when its scope is defined, is still unclear.<16> I readily confess that I have not made an extensive search of the dictionaries in our library, but I do not think it any more of a strain on the English language to read "shall extend to" to mean something like "shall consist of jurisdiction over" than it would be to follow Professor Calabresi and render "shall extend to" as "heretofore granted shall be restricted to."<17>

If Professor Calabresi is absolutely wedded to the view that "vest" in the Constitution is a signal that power is granted, it should follow that the Necessary and Proper Clause proves that "power" is "vested" in inferior executive officers. That clause provides that Congress may "make all Laws which shall be necessary and proper for carrying into Execution . . . all other Powers vested by the Constitution in the Government of the United States, or in any Department or Officer thereof."<18> Professor Calabresi's unwillingness to adopt this conclusion illustrates a general inconsistency in The Structural Constitution. Having decided that text is the major focus of their inquiry, the authors nevertheless ignore every bit of the text which runs counter to their thesis, rather than seek a more nuanced view, which might accommodate those portions of the text that run counter to their view.<19>

In New Vestments, I suggested that the Vesting Clauses are "performatives"-- that they create the branches.<20> On reflection, this seems functionally identical to the view that the Vesting Clauses are "power grants" completely defined in the articles that follow. In either case, one must rely on the text of the articles that follow for our first approximation of the branches' powers, and in either case one must sometimes leave the text. Professor Calabresi accepts that the enumerations in Articles I and III define the powers vested, while Mr. Rhodes, I gather, concedes this for Article I only.<21> Both assert (based on the absence of words similar to "herein granted" or "shall extend to") that the powers "vested" in Article II go beyond what can fairly be implied from the powers enumerated in that Article. Though the absence of these *1425 words is surely Professor Calabresi and Mr. Rhodes's strongest textual argument,

the "legislative history" of the difference in language between the legislative power "herein granted" and "The executive Power" provides no basis for ascribing any importance to this difference. That discrepancy occurred late in the Convention, on September 12, 1787, as a result of a Report on the Committee on Style, which had narrowed Congress' legislative powers to those "herein granted," but left unchanged "The executive Power." This change seemed designed only to reflect the limits of federalism on national regulatory power, not to ratify or to recognize substantive executive power.<22>

If the Vesting Clauses are power grants independent of what follows, we need to know how to figure out what powers are vested.<23> I faulted The Structural Constitution for leaving the text--and thus seeming to rely so heavily on tradition<24> to define the "executive Power"--while at the same time giving so little attention to the extrinsic sources. The omission is particularly glaring because it seems difficult to decide which extrinsic sources are relevant to the definition of the "executive Power" in the Constitution, and so much of The Structural Constitution's argument regarding Article II seems to turn on the result.<25> The burden of coming forward with relevant sources surely lies on the proponents of a construction of the "executive Power" that draws meaning from outside the text so as to go beyond what is fairly implied from the text *1426 that follows. The burden will not be easy.<26>

II. Fuzzy Vesting

In Power Grants , Professor Calabresi again tells us that Articles II and III work in exactly the reverse manner from that traditionally supposed. Rather than creating departments with enumerated powers spelled out in the text of those articles, the Constitution "vests" sweeping and " 'somewhat nebulous' "<27> powers in the first sentence of Articles II and III--and then takes it back in varying degrees in the clauses that follow.<28> Professor Calabresi now adds that the Vesting Clauses "gain most of their content from the specifications that follow," and are clauses "deriving almost the whole of their discernible meaning either from (1) the subsequent language of the Articles in which they appear or from (2) the other Vesting Clauses and the language that follows them."<29> This reads like a retreat from the suggestion that the Vesting Clause of Article II is a broad grant of unenumerated powers (among which we find the removal power)<30> to the view that the Vesting Clause is a "limited and unusual"<31> grant of unenumerated powers (which includes the removal power).

Professor Calabresi's attempt to keep his vision of the Vesting Clause of Article II from eating the Constitution begins with the assertion that one branch's power ends where that of the next begins; there can be no overlap.<32> This runs counter to relevant precedents, which are *1427 far more friendly to partly overlapping powers.<33> In his formulation, extra powers are determined not by enumeration and implication from enumeration, but by the procedure by which a power is exercised. If the power is exercised through litigation, it is judicial; if through passing a bill, it is legislative. "Any governmental action taken where these two prerequisites are absent must be an executive action which the President can control."<34> Suddenly, the executive is the Constitution's residuary beneficiary; any legitimate federal power not belonging to another branch belongs to it.<35>

The question begged is whether Congress can constrain powers such as removals. The key issue is not whether Congress can exercise such powers itself (it cannot), nor even whether they can be assigned to someone else in the executive branch, but rather, to what extent Congress can constrain the President. Professor Calabresi glosses over the difference between the exercise of a power (executive, subject to contrary enumeration) and the power to define the limits of what may be exercised (legislative, subject to contrary enumeration). I take it that, as a consequence of his view that each branch's powers are hermetically distinct from the others, he denies that Congress can enact any limits to enumerated and *1428 unenumerated presidential powers, even when Congress has an enumerated power of its own to deploy. Worse, the assertion that the President can control every executive action has been tacked on without explanation or justification other than reference to the Vesting Clause.

Suppose Congress passes a bill purporting to allow the President power to remove the Governor of the Federal Reserve Board only if there is "cause" for removal.<36> Is this constitutional? The answer to this question turns on the nature of the President's removal power. Professor Calabresi's Vesting Clause analysis instructs us that the removal power is executive because it is exercised in a fashion that is neither legislative nor judicial; so far, so good.<37> Since it comes from the Vesting Clause in Article II, and since powers cannot be shared between the branches, he says that Congress cannot diminish it. But, as both Professor Calabresi and Mr. Rhodes have been careful to note, presidential powers are not despotic; the President may not "disregard duly enacted, constitutionally valid statutes."<38> Without this caveat, Professor Calabresi would have to argue that impoundments are legal; rather than endorse the Nixonian view of presidential power, he says impoundments are unconstitutional.<39> Thus, it cannot be that identifying removals as procedurally executive ends the inquiry: Professor Calabresi agrees that not all nonlegislative and nonjudicial actions the President might take or command are inherently constitutional. We still need to know whether the Federal Reserve statute is constitutional to know whether the President's "Vesting Clause" power has been unduly infringed. This puts us back where we started.<40>

Professor Calabresi clearly does not see this as bootstrapping. Both he and Mr. Rhodes believe that once a power is identified as executive (which they equate with presidential<41>), it is perfect and complete and *1429 immune from congressional diminution. The task of identifying the executive power vested may be difficult, but they are confident that it is possible.<42> In Professor Calabresi's hands, this produces a list of powers that includes removals and control over all executive officials, but does not include presidential impoundments of funds; in different hands, it could produce a very different list.<43> The President's powers are not limitless since it is axiomatic that he is not a despot or even a George III. Although the limits are unclear, Article II, Section 2 can "help to define, limit, and give content to this otherwise vast grant of power . . . ."<44> But the "modest list" that appears in Article II is only an " e.g., exemplary list of the kinds of powers that it was expected that the President would have."<45> To quote this assertion is, I believe, to demonstrate what is wrong with it. The sole constraint on the President's unspecified powers is that they are limited to whatever the fair-minded reader, or the Supreme Court, terms "executive," and the authors provide no principle other than the rejection of despotism by which to sort through the various possible meanings of this term.<46> Since enumerated powers are commonly held to imply unenumerated powers,<47> a focus on enumeration *1430 does not eliminate the need to go outside the text. Unlike the vesting thesis, however, a focus on enumeration constrains what can fairly be imported.

III. Rhetorical Excess

A. The Judicial Power

Professor Calabresi and Mr. Rhodes assert that rooting the judicial power in the Heads of Jurisdiction would allow Congress to vest federal judicial powers in Article I courts or in executive branch agencies.<48> Professor Calabresi then criticizes this "disastrous and absurd construction" which I did not put forward.<49> To say, as I did, that Congress can choose to let state courts exercise jurisdiction over some substantial classes of cases does not imply that senators, attorney-advisors, or even administrative law judges can exercise Article III powers. If this position causes discomfort because the combination of state courts and Article I courts gives Congress too many alternatives to the original jurisdiction of the federal judiciary, the solution is to rethink the status and powers of Article I courts, not to erase Congress's power to structure the executive and judicial branches.<50>

Similarly, Professor Calabresi suggests that rooting the judicial power in the enumeration of Article III threatens the concurrent jurisdiction of the state courts because the Constitution would extend the judicial power to, for example, " all Cases in Law and Equity."<51> No one familiar with the dual sovereignty thesis propounded by the Federalists<52> would make such a claim, and of course I do not. Nor is there any danger that the textual road I mapped out could carry anyone so astray. Even if the word "all" in Article III, Section 2 means "every single one," as suggested by some commentators, this is a far cry from requiring exclusive jurisdiction since the state courts derive their powers from a separate delegation of power from "We the People." Thus, there is no logical *1431 incongruity in a federal court having jurisdiction over "every single one" of a class of cases and state courts having jurisdiction over "every single one" of the same class of cases.<53> The federal government has the tools to ensure that it will win in any clash where federal and state courts have concurrent jurisdiction.<54>

Mr. Rhodes maintains that adoption of Professor Amar's "two-tiered" jurisdiction thesis compels the unitarian vision of the executive.<55> It does not. Mr. Rhodes is correct that Professor Amar treats the Vesting Clause of Article III as a power grant, and that his view would require the same treatment of Article II.<56> Professor Amar, however, defines the powers vested in Article III by carefully examining the enumerations in that Article. Thus, to say that Article II deserves precisely the same treatment is not inconsistent with Professor Amar's view.

B. A Plain Meaning

Professor Calabresi again misreads me to argue that "because Congress has the power to create and structure every post in the executive department but two, it must therefore also have" the lesser power to give every top executive post protection from presidential dismissal.<57> My claim is far more modest and relies on the history of administrative practice and the powers enumerated in Articles I and II to conclude that:

The civilian side of the executive branch is less monolithically hierarchical than the armed services, and . . . Congress can legitimately give some inferior officers and heads of departments at least limited protection from dismissal and countermand by the President.

. . . Focusing on the political branches' enumerated powers provides a principled way to begin to define precisely which powers Congress can assign to executive agents shielded from the presidential removal at will. . . . For some executive officers, Congress can institute a requirement that the President have "cause" before exercising the removal power. I accept that a reasonable reading of the Take Care Clause may require that the President retain the power to dismiss for "cause" over all persons in the executive branch and further recognize that precisely what constitutes "cause" *1432 in the constitutional sense remains incompletely defined.<58>

While precise line-drawing is hard, some general lines can be discerned.<63> In some areas, such as foreign affairs, the President has the *1433 power to remove at will regardless of what Congress may legislate; in other areas, where the enumerated powers are more closely matched, Congress has the power to require "cause" for dismissal but perhaps no more.<64> Not only does this proposal have a historical pedigree<65> and fit well with the Supreme Court's separation of powers decisions,<66> but it makes sense. Plurality in the executive is suspect when it diminishes democratic accountability.<67> A "cause" standard does not create plurality in the executive as it does not give any subordinate official absolute tenure in office. Instead, it requires that the President give reasons for a dismissal that can then be examined either by Congress and the people or by a court.<68> In so doing, it creates a weak form of tenure, but even more *1434 importantly, it imposes political consequences for unwarranted dismissals in sensitive domestic offices. This political component is the key, as it puts the President on notice that Congress and the people will expect a reasoned explanation for personnel changes (and ensuing policy swerves) in critical offices.<69>

IV. Conclusion

We are not about to agree; indeed we are at times talking at cross- purposes. The Structural Constitution's defense of the unitarian vision of the Presidency stands or falls on its reliance on the Vesting Clauses of the first three articles of the Constitution as substantive grants of power to the three great departments of government--grants which, at least in the case of Article II and perhaps Article III as well, are supposed to have meaning independently from what follows. In my note, I tried to show why a different strategy is appropriate for Article II. In New Vestments, I sought to show why the vesting-as-power-grants argument is unpersuasive *1435 in the context of Article III in order to underline its irrelevance to Article II. I also suggested that this whole debate had a pretextual feel to it, and that we should be talking directly about more fundamental issues.<70>

Even on its own terms, The Structural Constitution's highly formalist vision of the executive power requires some identification of the external sources on which we can draw so as to fairly delimit the executive power. It seems more sensible to look first to the words that follow the grant; indeed, in Power Grants Professor Calabresi seems to move in that direction, although not far enough. Focusing on the relatively detailed enumerated powers--not the undefined and perhaps undefinable "vesting" of "executive power"--remains the best way to shape the Supreme Court's discretion when adjudicating disputes between Congress and the President and reduce, albeit not eliminate, the need to go outside the text. This focus suggests that Congress may constrain the President's removal power over officers charged with executing matters within Congress's enumerated domestic powers. Not only does this keep us close to the text, but it allows due scope for other important structural and semistructural concerns such as balance between the branches.<71> Whether this is a fair reading of text and structure, balance or disequilibrium, a good or bad political choice, a beautiful or an ugly aesthetic, or perhaps simple error, I leave for the reader to judge.

FOOTNOTES

<*>Associate Professor, University of Miami School of Law. I am grateful to the editors of the Northwestern University Law Review for letting me have the last word (for now). Their generosity came coupled with tight deadlines-- making it more important than usual for me to emphasize that the inevitable errors in this Reply are my responsibility, despite help and advice from Akhil Amar, Caroline Bradley, Fran Hill, Sarah Lyke, Michael Seidman, and Steven Winter, and research assistance from Julie Owen and Greg Reed.

In an effort to be fair, I made only trivial changes to the draft of New Vestments that was sent to Professor Calabresi and Mr. Rhodes for their comments some months ago.

<1>Steven G. Calabresi & Kevin H. Rhodes, The Structural Constitution: Unitary Executive, Plural Judiciary, 105 Harv. L. Rev. 1155 (1992) (hereinafter The Structural Constitution).

<2>Kevin H. Rhodes, A Structure Without Foundation: A Reply to Professor Froomkin, 88 Nw. U. L. Rev. 1406 (1994) (hereinafter Without Foundation). Mr. Rhodes suggests that in restricting our debate to Congress's power to limit the President's removal power, I have unfairly or unwisely focused on a narrow fragment of the grand themes in The Structural Constitution. See id. at 1407- 08. I agree with Mr. Rhodes that The Structural Constitution valuably directs attention to the relationships between Article II and Article III. Indeed, that is one of the things I tried to do in my student note. Nevertheless, I think that Article II is a reasonable place to focus this discussion if only because that is where The Structural Constitution's claims most deviate from what the law is or should be, and where the unitarian argument is most likely to matter in the short run.

<3>Steven G. Calabresi, The Vesting Clauses as Power Grants, 88 Nw. U. L. Rev. 1377 (1994) (hereinafter Power Grants). My failure to respond to many of Professor Calabresi's assertions, and especially his interpretations of The Imperial Presidency's New Vestments, should not be read as acquiescence to them.

<4>A. Michael Froomkin, The Imperial Presidency's New Vestments, 88 Nw. U. L. Rev. 1346 (1994) (hereinafter New Vestments).

<5>Two points may need more emphasis in light of legitimate questions raised by Mr. Rhodes. First, the reason that New Vestments fails to address the important historical arguments is that I felt I had little useful to add regarding the historical evidence, beyond harping on the precedent of the Bank of the United States, to the excellent historical (and structural) analysis that I knew was forthcoming in Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 Colum. L. Rev. 1 (1994). Second, my attempt to conduct a proper structural analysis of one part of the Constitution can be found in my Note, In Defense of Administrative Agency Autonomy, 96 Yale L.J. 787 (1987), little of which was repeated in New Vestments.

<6>Compare what Chief Justice Marshall said in McCulloch v. Maryland:

A constitution, to contain an accurate detail of all the subdivisions of which its great powers will admit ... would partake of the prolixity of a legal code, and could scarcely be embraced by the human mind.... Its nature, therefore, requires that only its great outlines should be marked, its important objects designated, and the minor ingredients which compose those objects be deduced from the nature of the objects themselves.... We must never forget that it is a constitution we are expounding.

17 U.S. (4 Wheat.) 316, 407 (1819).

<7>See Power Grants, supra note 3, at 1401. According to Professor Calabresi, the true focus of The Structural Constitution, despite the title, was "textual distributions of power made by the first three Articles of the Constitution" and not structural analysis in the Charles Blackian sense. See id.; compare id. with Without Foundation, supra note 2, at 1407-08. At one level this is fair enough. After all, despite its title this essay is not about nudism. But at another level this disclaimer should make readers pause. While it is reasonable to attack a problem like the nature of the removal power by focusing first on the text, it is far less reasonable to stop there.

I prefer Mr. Rhodes's more nuanced view that no "rigid distinction should be drawn between textual and structural analysis." Without Foundation, supra note 2 at 1406 n.3. In particular, when confronted with a plausible textual argument for a different reading, it makes sense to check one's answer by whatever other means are available. The type of structural analysis championed by Charles Black is one means I happen to favor. I agree with Professor Calabresi that this technique carries with it some danger of unconscious importation of personal political views, see Power Grants , supra note 3, at 1402, although I think that this criticism applies with equal force to all other methods.

<8>The Blackian structuralism to which we refer is most clearly exhibited in Charles L. Black, Jr., Structure and Relationship in Constitutional Law (1969), although it permeates a great life's work. I challenge any reader of Structure and Relationship to say that its political premises are anything but conscious.

<9>See Without Foundation, supra note 2, at 1409 -10, 1412; see also Power Grants, supra note 3, at 1394 - 95. In New Vestments, supra note 4, I take seriously--without endorsing--the concern (which, perhaps erroneously, I took to be the Calabresi-Rhodes view) that in the absence of a very robust reading of the Vesting Clause in Article III, s 2, that Article might be read to enumerate the federal judicial power without specifying whether that power, or portions of it, would be exercised by state or federal courts. I do not share this concern because I think it is answered by reference to the ratification debates. The federalists argued that both state and national governments were logically possible because "We the People" were sovereign and could choose to delegate powers to more than one agent--effectively dual sovereigns. See The Federalist Nos. 82, 84 (Alexander Hamilton); Gordon S. Wood, The Creation of the American Republic 1776 -1787 , 519 - 64 (1969). In the particular case of the courts, the Federalist Papers treat Article III as empowering federal courts and rejects arguments that the powers at issue should have been given to state courts. See The Federalist Nos. 81, 82 (Alexander Hamilton). In this light, and in the absence of explicit contrary language, the federal constitution is most reasonably read to constitute an organ of the agent to which that particular delegation was directed: the federal government. Admittedly, the Framers themselves may have been less than clear in their own minds on this point. See Wood , supra.

Somehow, Professor Calabresi sees my attempt to respond to this concern as a concession which determines the "whole ballgame." Power Grants, supra note 3, at 1383 - 84. He argues that a clause which (he thinks) fails to specify whom it empowers "cannot be a power grant because it does not grant power ' to ' anybody or anything." Id. at 1383 (footnote omitted). He omits mention of the subsequent two paragraphs of my article, in which I discuss how Professor Amar's "two-tiered" approach itself offers one method of resolving which judicial powers are purely federal, which are potentially concurrent, and which are within Congress's discretion to allow sole jurisdiction in the state courts. He also omits mention of my argument that the claimed deficiency is absent because even if one read that section potentially to allow federal judicial powers to be exercised by state courts, it was up to Congress, a federal institution, to determine which powers it reserved for the federal judiciary. I argued that the Framers would reasonably have anticipated that Congress would legislate along the lines of the first Judiciary Act, which it did, and that they therefore had little reason to be concerned as to which courts would actually have this power.

Saikrishna Bangalore Prakash offers a fascinating discussion of how the Framers may have viewed the relationship between state and federal courts and administrators in Saikrishna B. Prakash, Field Office Federalism, 79 Va. L. Rev. 1957 (1993). He suggests that the Framers not only believed that state court judges might sit as federal court judges, but that they expected the federal government to be able to call on state executive officers to execute federal laws. This belief as to state judges may explain why the recipient of the federal judicial power is not as clear as Professor Calabresi and Mr. Rhodes desire.

<10>The Supreme Court has defined jurisdiction as the power to hear and determine the subject matter in controversy between parties to a suit, United States v. Arrendono, 31 U.S. (6 Pet.) 691, 709, 729 (1832), and as the "power to entertain the suit, consider the merits and render a binding decision thereon." General Inv. Co. v. New York Cent. R.R., 271 U.S. 228, 230 (1926). Jurisdiction is thus a prerequisite to the exercise of judicial power. Cf. Ex parte Robinson, 86 U.S. (19 Wall.) 505, 505-11 (1874) (stating that contempt power of federal courts inhered "the moment the courts ... were called into existence and invested with jurisdiction over any subject").

<11>e.g., Liner v. Jafco, Inc., 375 U.S. 301, 306 n.3 (1964) ("Our lack of jurisdiction to review moot cases derives from the requirement of Article III of the Constitution under which the exercise of judicial power depends upon the existence of a case or controversy."); followed by DeFunis v. Odegarrd, 416 U.S. 312, 316 (1974) (per curiam); North Carolina v. Rice, 404 U.S. 244, 246 (1971).

<12>Mr. Rhodes defines "judicial Power" as "the power to adjudicate and decide cases on behalf of the nation" and defines "jurisdiction" as the "categories of 'Cases' and 'Controversies' to which that power ... 'shall extend' ... ." Without Foundation, supra , note 2, at 1409 -10. The definitions could just as easily be reversed. Mr. Rhodes also argues that the Supreme Court's power to refuse to exercise jurisdiction "exemplifies the distinction between the judicial Power and jurisdiction," as do justiciability doctrines such as mootness, ripeness, standing, political question, and other judge-made abstention doctrines. Id. at 1409-10 n.22. In short, he argues that the power not to decide is an example of "judicial Power" outside the set of powers we call "jurisdiction." Modern standing doctrine is a particularly poor example of a distinction which is supposed to have existed since the framing. See Steven L. Winter, The Metaphor of Standing and the Problem of Self-Governance, 40 Stan. L. Rev. 1371 (1988) (demonstrating that modern standing doctrine is a modern, ahistorical construct).

I wonder whether Mr. Rhodes is prepared to argue that the federal courts misread the Constitution if, when faced with a case in the mandatory jurisdiction categories (whatever he finds those to be), they decline to decide it on a theory of mootness, ripeness, standing, or political question. That is certainly where his distinction leads. See Evan T. Lee, Deconstitutionalizing Justiciability: The Example of Mootness, 105 Harv. L. Rev. 603, 641-43 (1992).

Mr. Rhodes also argues that the distinction between "jurisdiction" and "judicial Power" is visible in the relationship between Article I courts (which have jurisdiction) and those Article III courts which review them (which have "judicial Power"). Without Foundation, supra note 2, at 1411. Does an appellate court have something other than jurisdiction when it hears an appeal from a trial court?

<13>See supra note 10 (citing Supreme Court's usage). Mr. Rhodes's strongest argument might have been that the federal courts have used their "judicial Power" to broaden their own jurisdiction by inventing doctrines such as pendent and ancillary jurisdiction, but these can be seen as somewhat extreme cases of a court declaring its own jurisdiction.

<14>See Louis Henkin, Is There a "Political Question" Doctrine?, 85 Yale L.J. 597 (1976); Winter, supra note 12, at 1373, 1392-93 (discussing this view); Geoffrey R. Stone et al., Constitutional Law 114 (2d ed. 1991).

<15>Power Grants, supra note 3, at 1380 - 81.

<16>Cf. supra text accompanying note 9 (discussing Mr. Rhodes's related argument).

<17>5 The Oxford English Dictionary 594-95 (J.A. Simpson & E.S.C. Weiner eds., 2d ed. 1989) gives among its several definitions of extend, " Metaph. Used in passive with generalized sense: To possess 'extension' or spatial magnitude" and "to write out (a legal instrument) in proper form," and "to cover an area; to stretch out in various directions .... Usage example from 1559: The parliament, which I acknowledge to be of great strengthe in matters whereunto it extendethe."

<18>U.S. Const. art. I, s 8, cl. 18 (emphasis added).

<19>For other examples of the same phenomenon, see New Vestments, supra note 4, at 1372 n.141.

<20>New Vestments, supra note 4, at 1363.

<21>See supra text following note 9.

<22>Henry P. Monaghan, The Protective Power of the Presidency, 93 Colum. L. Rev. 1, 22 (1993) (footnotes omitted). If, despite this history, one lines up the three Vesting Clauses side by side and attempts to deduce the most that one can from them ahistorically, one must confront the puzzle of the words "of the United States." These words modify the power vested in Article III, but are absent in Article I and are clearly part of the President's title, not a modifier of the "executive Power," in Article II. The Structural Constitution suggests that the phrase "of the United States" is a qualifier in Article III, i.e., that it makes the powers vested less than they otherwise would be. See The Structural Constitution, supra note 1, at 1195 - 96 n.214, 1196. The problem with this view is that if the words "of the United States" are a qualifier, then the absence of that "qualifier" in Article II means that the President's grant of power is somehow greater. It is difficult to see how this could be: what power "not of the United States" could the President possibly have? And if the answer is "none," then we have an excellent example of the dangers that come from an over-literal mindedness.

<23>"The powers of the President are not as particularized as are those of Congress. But unenumerated powers do not mean undefined powers." Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 610 (1952) (Frankfurter, J., concurring).

<24>I have never denied the potential relevance of tradition, or any other extrinsic aid, to interpretations of the Constitution, although I recognize that they create problems of their own. In fact, it is Professor Calabresi who seems suspicious of appeals to tradition. See Power Grants, supra note 3, at 1391 n.54, 1404.

<25>We all agree this is one case where the English model supplies only a negative precedent; I would go further and argue that the President was supposed to be a wholly new institution of government. Thus, Mr. Rhodes's confidence that "defining 'executive Power' should be no more difficult than defining numerous other constitutional provisions," Without Foundation, supra note 2, at 1412 n.34, strikes me as misplaced.

<26>Cf. Myers v. United States, 272 U.S. 52, 247-48 (1926) (Brandeis, J., dissenting) (noting that in 1787, no state constitution gave the executive illimitable removal powers).

<27>Power Grants, supra note 3, at 1394 (quoting The Structural Constitution , supra note 1, at 1195). The Structural Constitution stated that the clauses which follow the first sentence in Article II "are not offered to us as an exclusive list," The Structural Constitution, supra note 1, at 1196, but said little more. Mr. Rhodes is not troubled by this because his concept of what it means to be a chief executive is fundamentally extrinsic to the Constitution. That alone does not mean it is wrong, but it does mean that we have a right to ask where it comes from.

I am not suggesting that we should close our minds to any plausible account of extrinsic materials that could reasonably help define "executive." I am suggesting, first, that the burden of proof is on anyone who wants to use extrinsic materials to suggest a meaning that goes far beyond what is fairly implied from the text that follows and, second, that prior to Power Grants Professor Calabresi and Mr. Rhodes had yet to address this problem, much less solve it.

<28>See Power Grants, supra note 3, at 1396 - 97 (executive power); id. at 1394 - 95 (judicial power). Contrast the Federalists' own (if somewhat suspect) account of the limited and enumerated nature of the grants of constitutional power in, e.g., The Federalist No. 84 (Alexander Hamilton).

<29>Power Grants , supra note 3, at 1389 - 90. He also states that Art. II, s 2 is "a kind of, e.g., exemplary list" of the President's more general powers. Id. at 1397.

<30>The Structural Constitution, supra note 1, at 1167-68, 1177-78, 1186, 1193, 1196.

<31>Power Grants, supra note 3, at 1397.

<32>Id. at 1390 n.45. I assume that Professor Calabresi means what he says in the text of his article, although there is a tension between the statement that no power can belong to two branches at the same time, id., and the statement that "certain kinds of actions ... can be undertaken by the executive ... and also by Congress or by the Article III federal courts," id. at 1390 n.46. I have not had the advantage of the forthcoming article to which Professor Calabresi makes reference, The Trinity of Powers and the Lessig/Sunstein Heresy, see id. at 1390 n.43, so I do not know to what extent Professor Calabresi endorses the nondelegation doctrine. It will also be interesting to see whether Professor Calabresi assigns the power to interpret the Constitution to one branch or finds that all three branches have this duty.

<33>See INS v. Chadha, 462 U.S. 919, 951 (1983); Dames & Moore v. Regan, 453 U.S. 654, 669 (1981); Buckley v. Valeo, 424 U.S. 1, 121 (1976); Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 640-41 (1952) (Jackson, J., concurring). It rejects the clear admission, if not celebration, of overlap in The Federalist No. 47 (James Madison) (stating Montesquieu "did not mean that these departments ought to have no partial agency in, or no control over, the acts of each other" but only that no department should wholly control the other); id. No. 48 (James Madison) ("Unless these departments be so far connected and blended as to give to each a constitutional control over the others, the degree of separation which the maxim requires, as essential to a free government, can never in practice be duly maintained.").

<34>Power Grants, supra note 3, at 1391 (emphasis added). Professor Calabresi neglects to note the judicial functions of the legislature ( e.g., impeachment); the executive functions of the legislature ( e.g., advice and consent); the executive powers of the courts ( e.g., interbranch appointments, see Morrison v. Olson, 487 U.S. 654 (1988)); and the President's legislative power (the veto) and judicial duties ( see, e.g., Ex parte Randolph, 20 F. Cas. 242, 254 (C.C.D. Va. 1833) (Marshall, J.) (noting that executive branch is called upon to act in a "judicial" capacity). And all three branches have the power and duty to interpret statutes and, I would argue, the Constitution itself.

<35>If there is a residual beneficiary in the Constitution--an unhelpful concept--Congress seems the best candidate since Congress is given the power "to make all Laws which shall be necessary and proper for carrying into Execution ... all other Powers vested by this Constitution in the Government of the United States, or in any Department or Officer thereof." U.S. Const . art. I, s 8, cl. 18; see E. Donald Elliott, INS v. Chadha : The Administrative Constitution, the Constitution, and the Legislative Veto, 1983 Sup. Ct. Rev. 125, 142, 175; William W. Van Alstyne, The Role of Congress in Determining the Incidental Powers of the President and of the Federal Courts: A Comment on the Horizontal Effect of the Sweeping Clause, Law & Contemp. Probs. , Spring 1976, at 102; see also Charles L. Black, Jr., The Working Balance of the American Political Departments, 1 Hastings Const. L.Q . 13, 15-16 (1974).

<36>See 12 U.S.C. s 242 (1988) (President may remove members of the Federal Reserve Board for "cause"); see also Froomkin, supra note 5, at 810 n.146 (describing "murky" legal status of Federal Reserve Board Chairman).

<37>One could also argue that the power to legislate the terms of an office holder's tenure is by definition legislative.

<38>Power Grants, supra note 3, at 1393; see also Without Foundation, supra note 2, at 1412 n.36; The Structural Constitution, supra note 1, at 1197 n.219.

<39>Professor Calabresi agrees that presidential impoundments of duly appropriated funds are unconstitutional, even though the actual expenditure of the funds is unequivocally executive in nature. See Power Grants, supra note 3, at 1391 n.52. Interestingly, this (reasonable) conclusion relies on tradition, despite Professor Calabresi's contention that he "did no such thing in The Structural Constitution." Id. at 1404.

<40>Compare the thorough and sensitive examination of the nature of the "executive Power" in the absence of legislation found in Monaghan, supra note 22, passim .

<41>For an example of recent court decisions recognizing the presidential- executive distinction, see Franklin v. Massachusetts, 112 S. Ct. 2767, 2775 (1992) (plurality opinion) (stating that unique constitutional position of President means that his conduct is not "agency action" under Administrative Procedures Act); Armstrong v. Executive Office of the President, 924 F.2d 282, 288 (D.C. Cir. 1991) (construing APA to exclude President from definition of agency); Mail Order Ass'n. of America v. United States Postal Serv., 986 F.2d 509 (D.C. Cir. 1993) (allowing Board of Governors of Postal Service to represent itself when its position is at odds with Justice Department and President). Statutes also recognize the distinction. For example, The Presidential Records Act, 44 U.S.C. s 2201(2) (1988), distinguishes between "staff, or a unit or individual of the Executive Office of the President whose function is to advise and assist the President, in the course of conducting activities which relate to or have an effect upon the carrying out of the constitutional, statutory, or other official or ceremonial duties of the President" and "any documentary materials that are ... official records of an agency," as the term "agency" is defined in the Freedom of Information Act, 5 U.S.C. s 552(f) (1988). See Armstrong v. Executive Office of the President, 1 F.3d 1274, 1290 (D.C. Cir. 1993) (noting distinction).

<42>Power Grants, supra note 3, at 1390 - 91 n.45; Without Foundation, supra note 2, at 1412 n.36. I am not criticizing Professor Calabresi for proposing a method of separation of powers adjudication that lacks bright lines. Contra Power Grants, supra note 3, at 1392. That would be a case of the pot calling the kettle black. I am criticizing the theory for lacking any useful definition of what constitutes executive power other than ad hoc determinations by the Supreme Court. As I hope I have demonstrated above, Professor Calabresi's procedural limitation is inadequate. It would, however, be fair to say that I am criticizing Mr. Rhodes for drawing bright lines in the wrong places.

<43>For example, based on an examination of the Vesting Clauses and many contemporaneous documents, William Crosskey suggested that all three Vesting Clauses (coupled with the Necessary and Proper Clause) should be read as grants of sweeping federal power. See 1 William W. Crosskey, Politics and the Constitution in the History of the United States 385-408 (1953).

<44>Power Grants, supra note 3, at 1397.

<45>Id. Professor Calabresi does not read the duties imposed on the President in Article II, s 3 as implicit power grants. I would be far more comfortable with reading the duties as implied powers than with the Vesting Clause as a blank check subject to constraints by analogy.

<46>See New Vestments, supra note 4, at 1363-65.

<47>I think it is most reasonable to follow the traditional approach and view the removal power as an unenumerated power derived from the appointment power. "It is the starting point of all judicial analysis in this area that the President's power to remove, however much it may be restricted, derives from the constitutional grant of his power to appoint." Synar v. United States, 626 F. Supp. 1374, 1402 (D.D.C.) (per curiam) (three-judge court including then-Judge Scalia) (citation omitted), aff'd sub nom. Bowsher v. Synar, 478 U.S. 714 (1986); see also, e.g., Shurtleff v. United States, 189 U.S. 311, 318 (1903); Ex parte Hennen, 38 U.S. (13 Pet.) 230, 259 (1839). Viewing the removal power as an unenumerated power emanating from the Take Care Clause changes little. In either case, I agree that the Take Care Clause has something to do with the problem and that it limits Congress's power to protect policy-making officers from presidential removal.

<48>Power Grants, supra note 3, at 1383; Without Foundation, supra note 2, at 1410 -11.

<49>Power Grants, supra note 3, at 1384. I invite the interested reader to review the far less exciting statements I actually made in New Vestments, supra note 4, at 1355-56.

<50>Were it not for review by an Article III court, there would be merit to Professor Calabresi's argument that Congress's power to structure the other branches creates a danger that it will circumvent the federal courts. I suspect that the three of us share an intuition that there is something slightly suspect about Article I courts. I find comfort in review by Article III courts, perhaps because I am persuaded by Richard F. Fallon, Jr., Of Legislative Courts, Administrative Agencies, and Article III, 101 Harv. L. Rev. 916 (1988).

<51>Power Grants, supra note 3, at 1385 (emphasis in original).

<52>See, e.g., The Federalist No. 84 (Alexander Hamilton). See generally Wood, supra note 9, at 524 -32, 600 - 01.

<53>See The Federalist No. 82 (Alexander Hamilton).

<54>See generally Prakash, supra note 9. To the extent that the Supremacy Clause (which speaks of federal laws and the Constitution, not judicial decisions) does not directly define a relationship between state and federal courts, the relationship can be derived from the Supreme Court's appellate jurisdiction. Failing that, it at least remains within the power of Congress, operating under the Necessary and Proper Clause, to legislate in support of federal court power over the states where the jurisdiction is concurrent.

<55>See Without Foundation, supra note 2, at 1416.

<56>See, e.g., Akhil Reed Amar, A Neo-Federalist View of Article III: Separating the Two Tiers of Federal Jurisdiction, 65 B.U. L. Rev. 205, 239 (1985). I think one can reach Amar's conclusions without treating the Vesting Clauses as substantive grants of power. But, as I argued supra text following note 20, reading the Vesting Clauses as substantive grants of power only raises the issue of how one decides what powers are vested.

<57>Power Grants, supra note 3, at 1400 (emphasis added).

<58>New Vestments, supra note 4, at 1373 -74 (footnotes omitted); see also Froomkin, supra note 5, at 805-08 (discussing early history of administrative practice).

<59>A far more thorough attack on the same problem has produced a remarkably similar conclusion. See Lessig & Sunstein, supra note 5, at 106 -14.

<60>See supra note 47.

<61>Professor Calabresi barely addresses this issue, except to repeat that this amounts to rewriting the Vesting Clause of Article II. Power Grants, supra note 3, at 1404. Professor Calabresi explains that my position amounts to allowing Congress to strip the President of his "Vesting Clause powers." Id. Since I deny there is such a thing as an unenumerated Vesting Clause power for any branch of government and therefore see this as a null set, I can hardly protest this characterization.

<62>Mr. Rhodes suggests that the President's power to run the executive branch (and thus to dismiss executive officers at will) is an enumerated power given to the President either in the Vesting Clause of Article II or in the Take Care Clause. See Without Foundation, supra note 2, at 1409 n.17. But so long as Mr. Rhodes remains unable to explain how one should define "executive," see id. at 1412 n.34 ("a substantive exploration of the meaning of the term 'executive Power' is not possible here"), his suggestion that the words "the executive Power shall be vested" constitute an enumeration of the President's powers robs the concept of enumeration of any meaning.

Professor Calabresi denigrates the Take Care Clause as a possible source of presidential power. Power Grants, supra note 3, at 1398 n.87. By contrast (as the quote in the text demonstrates), I agree that the Take Care Clause can reasonably be read as an enumerated presidential power to police the executive branch and to monitor the performance of all executive officers. Thus, I agree with the general thrust of Without Foundation, supra note 2, at 1413 (but not id. at 1414 n.44). In particular, I agree that many important, even necessary, presidential powers are either implied powers or derive from the duty to take care that the laws be faithfully executed. I have never suggested that "Congress can automatically trump" this power. Compare id. at 1414 n.44 with New Vestments, supra note 4, at 1372-74.

The difference between us here--and it is not small--is that Mr. Rhodes wants to say that the removal power is enumerated rather than implied (although it appears nowhere in the Constitution). The removal power would be enumerated if Article II, s 2, para. 1 of the Constitution stated that the President "shall have Power by and with the Advice and Consent of the Senate, to ... appoint ... all ... Officers of the United States and to remove them at will." It doesn't.

<63>See Froomkin, supra note 5, at 812-14. Mr. Rhodes protests that a focus on enumeration or cause could require lengthy, messy, case-by-case adjudication and impose costs in energy and uncertainty. See Without Foundation, supra note 2, at 1412 n.34. This will be true in some, but by no means all, cases. Line drawing is often messy, but the neatness which accompanies erroneous a priori decisions can be worse. It may be that the litmus for the world view that divides us is our reaction to Justice Jackson's Youngstown opinion. See Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 640 (1952) (Jackson, J., concurring). Professor Calabresi, I gather, wishes to reject it. See Power Grants, supra note 3, at 1390 - 92. Mr. Rhodes too wishes to formulate a bright-line system of separation of powers. I think Justice Jackson's opinion in Youngstown describes and reflects a continuing problem that every generation should do as little as possible to resolve so as to preserve the maximum flexibility for its successors.

<64>See Froomkin, supra note 5, at 812-14.

<65>See Lessig & Sunstein, supra note 5, at 22-32, 38-61; Froomkin, supra note 5, at 804-08.

<66>See New Vestments, supra note 4, at 1366 - 69.

<67>I find it ironic to be accused by Professor Calabresi of reading my personal policy preferences into the Constitution, see Power Grants, supra note 3, at 1402, when in fact I have serious doubts about the wisdom and effectiveness of most so-called independent agencies, see Froomkin, supra note 5, at 789 nn.6, 8.

<68>A critical issue in this context is whether the federal judicial power extends to enjoining the President from removing an officer who claims that the President lacks the requisite "cause." Despite the ruling in Mississippi v. Johnson, 71 U.S. (4 Wall.) 475 (1867) (refusing to hear suit seeking to enjoin President on grounds that courts lacked power to issue such injunction), the extent of the President's immunity to injunctions remains open to debate after Franklin v. Massachusetts, 112 S.Ct. 2767 (1992) (plurality opinion). In Franklin, the court refused to grant injunctive relief against the President, but the plurality left open the general issue, stating, "For purposes of establishing standing, however, we need not decide whether injunctive relief against the President was appropriate...." Id. at 2777; see also United States v. Nixon, 418 U.S. 683 (1974) (holding court has jurisdiction to issue subpoena to President for use in criminal prosecution); United States ex rel. McLennan v. Wilbur, 283 U.S. 414, 420 (1931) (holding mandamus "will issue only where the duty to be performed is ministerial and the obligation to act peremptory, and plainly defined. The law must not only authorize the demanded action, but require it; the duty must be clear and indisputable."); Wilbur v. United States ex rel. Kadrie, 281 U.S. 206, 218 (1930) (holding mandamus may be used "to compel action, when refused, in matters involving judgment and discretion, but not to direct the exercise of judgment or discretion in a particular way nor to direct the retraction or reversal of action already taken in the exercise of either."); United States ex rel. Louisville Cement Co. v. ICC, 246 U.S. 638, 642-43 (1918); ICC v. United States ex rel. Humboldt Steamship Co., 224 U.S. 474, 484 (1912); Commissioner of Patents v. Whiteley, 71 U.S. (4 Wall.) 522 (1866); National Treasury Employees' Union v. Nixon, 492 F.2d 587, 616 (D.C. Cir. 1974) (holding court has authority to issue mandamus to President requiring that he implement statutorily required governmental employee pay increase). See generally Laura Krugman Ray, From Prerogative to Accountability: the Amenability of the President to Suit, 80 Ky. L.J. 739 (1991/92). Cf. Mackie v. Bush, 809 F. Supp. 144 (D.D.C. 1993) (enjoining President Bush from carrying out threat to dismiss members of Board of Governors of Postal Service Board), vacated as moot sub nom. Mackie v. Clinton, No. 93-5001, 1993 U.S. App. LEXIS 33466 (D.C. Cir. Oct. 27, 1993); Mackie v. Clinton, 827 F.Supp. 56 (D.D.C. 1993) (declaring recess appointment to office held by hold-over appointee invalid); Bush Defies Judge and Names New Member to Postal Board, N.Y. Times , Jan. 9, 1993, s 1, at 10.

<69>I believe that the limited congressional power to structure the executive proposed in my Note and in New Vestments fully comports with the original intent of the Framers and ratifiers (to the extent that we can reconstruct it) and the practice of the earliest Congresses and administrations. See Lessig & Sunstein, supra note 5, at 22-32; see also Froomkin, supra note 5, at 804-07. The Framers' intentions on this subject seem to have been less than fully formed, see Leonard D. White, The Federalists 26 (1948) (stating Framers "displayed a notable lack of interest in the organization of the executive branch"), but even if the Framers had reached a consensus that the President should have an unfettered removal power, we would still need to understand this consensus in the context of the Framers' other assumptions about how policy- making power would be shared between the President and Congress. See Jan G. Deutsch, Precedent and Adjudication, 83 Yale L.J. 1553 (1974) (arguing that choosing relevant level of abstraction determines result); Frank H. Easterbrook, Abstraction and Authority, 59 U. Chi. L. Rev. 349 (1992) (same). The Framers, I suspect, would have been shocked to learn that Presidents argue they can sign laws and then choose not to enforce the parts they consider unconstitutional. See Michael B. Rappaport, The President's Veto and the Constitution, 87 Nw. U. L. Rev. 735, 771-76 (1993); Michael T. Brady, Note, Executive Discretion and the Congressional Defense of Statutes, 92 Yale L.J. 970 (1983). We cannot resolve these disputes about constitutional interpretation without first paying more attention to method. See, e.g., David L. Faigman, Madisonian Balancing: A Theory of Constitutional Adjudication, 88 Nw. U. L. Rev. 641 (1994); Richard H. Fallon, Jr., A Constructivist Coherence Theory of Constitutional Interpretation, 100 Harv. L. Rev. 1189 (1987), for some recent examples. We could be discussing which intent has a greater claim on our fidelity: our best reconstruction of the Framer's vision of a relationship between the branches of government, our best reconstruction of their answer on the matter of removals, or our best reconstruction of their intent as to how their successors should approach constitutional problems. See H. Jefferson Powell, The Modern Misunderstanding of Original Intent, 54 U. Chi. L. Rev. 1513 (1987); H. Jefferson Powell, The Original Understanding of Original Intent, 98 Harv. L. Rev. 885 (1985). Contra Raoul Berger, The Founders' Views--According to Jefferson Powell, 67 Tex. L. Rev. 1033 (1989); Raoul Berger, "Original Intention" in Historical Perspective, 54 Geo. Wash. L. Rev. 296 (1986). Or we could talk about whether we should attempt any of these exercises. See, e.g., Hans W. Baade, "Original Intent" in Historical Perspective: Some Critical Glosses, 69 Tex. L. Rev. 1001 (1991) (questioning continuing relevance of "original intent").

<70>See New Vestments, supra note 4, at 1374 -75; see also supra note 69. Neither Mr. Rhodes nor Professor Calabresi accepted this invitation, and in my latest attempt to meet them on their own grounds I am again guilty of responding in kind. I imposed this restriction on myself in the hopes of being more persuasive to readers attracted by textual and formal arguments, but it is not without cost.

<71>See New Vestments, supra note 4, at 1374 -75.