The New Unitary Executive and Democratic Theory: The Problem of Alexander Hamilton

Abstract

Central to the recent argument from the “unitary executive” is the claim that the unitary executive is consistent with the text and history of the Constitution. But because this veracity and importance of this claim is contested, unitarians also argue that the unitary executive is consistent with democratic theory. This article examines that argument by addressing a question in the political thought of Alexander Hamilton. Although Hamilton was an important defender of an energetic executive, and is associated with an expansive interpretation of executive power, he wrote in The Federalist that the president and Senate would share the removal power. In contrast with existing scholarship, which either overlooks Hamilton's statement on removals or dismisses it as a careless error, this article argues that Hamilton's statement limiting presidential removals illuminates his larger argument about executive energy. By showing how “duration” would check “unity,” this article clarifies Hamilton's political thought and offers an important critique of the modern argument from the unitary executive.

The executive branch shall construe Title X in Division A of the Act, relating to detainees, in a manner consistent with the constitutional authority of the President to supervise the unitary executive branch and as Commander in Chief and consistent with the constitutional limitations on the judicial power, which will assist in achieving the shared objective of the Congress and the President, evidenced in Title X, of protecting the American people from further terrorist attacks.

—George W. Bush, Statement on Signing of H.R. 2863, December 30, 2005

One of the most important developments under the Bush administration is the argument from the “unitary executive.” Broadly, the argument from the unitary executive recommends a powerful president on the grounds that the unitary executive is compatible with constitutional design and democratic theory. Although the origins of the term are unknown, it seems to have arisen in its present form in the Reagan Justice Department. In the past twenty years, the term has been debated in countless law reviews, and it has been employed by at least four justices on the Supreme Court. More famously, George W. Bush has used the term in numerous signing statements, including the opening selection from his remarks on the McCain Detainee Amendment, and it figured prominently in John Yoo's memorandum of September 25, 2001, which argued for presidential authority to respond to the attacks of September 11 (U.S. Department of Justice, office of the Legal Counsel 2001). It is possible that the argument played a role in Bush's novel interpretations of international agreements, departure from the Foreign Intelligence Surveillance Act, and the recent firing of U.S. attorneys.¹

It would be beyond the scope of this article to chart the subtleties in the development of the strands of the argument from the unitary executive, but it is easy to notice that the argument has two forms—domestic and foreign.² According to the domestic argument from the unitary executive, the oneness of the executive power, the fact that the Constitution vests the executive power in a president means that Congress may not limit the ability of the president to control executive branch officials. Carried to its fullest extension, this argument holds that Congress may not limit the president's removal authority over executive officers by creating independent commissions or special prosecutors and may not subpoena executive officers to testify. With regard to foreign policy, John Yoo has made the most extensive, and infamous, case for a unitary executive.³ According to Yoo, the “text, structure, and history” of the Constitution reveal that the document does not set up a precise legal method or “fixed process” for waging war, but is rather quite “flexible” and can therefore accommodate twentieth-century practice, in which presidents assume that they may change the state of the nation from peace to war with or without congressional approval. Under Yoo's Constitution, the president's authority over decision making in foreign affairs is limited only by the requirement that Congress must fund wars the president initiates and that the Senate is required to ratify treaties he negotiates. Subject only to these two express requirements that he share the executive power, the president is free to commit the military to war and change the state of peace by interpreting and even terminating treaties.

Because defenders of the unitary executive rely on originalist arguments to make their case, it is no surprise that a common characteristic is that they rely on Alexander Hamilton. Hamilton, after all, is perhaps the most famous American defender of executive power. In the Constitutional Convention of 1787, his proposed constitution included an executive who would serve for “good behaviour” (i.e., for life unless impeached) (Madison [1840] 1987, 138), and, as George Washington's Treasury Secretary and confidant, he defended the first president when James Madison and Thomas Jefferson founded a party opposed to Washington's use of power.⁴ More important, in The Federalist, Hamilton defended “energy” in the executive, and, in that argument, wrote that “unity in the executive” was necessary both for energy and for democratic accountability. It is this argument that has been of most use to modern unitarians. However, as we will see, Hamilton's making unity in the executive compatible with, and even necessary to, the republican form points to a larger and underappreciated question in Hamilton's political thought.⁵

Before we turn to Hamilton, it is important to point out that even as argument of the unitary executive makes legal claims with obvious partisan implications, it also raises a question that has long been a concern for political science. Namely, what is the best way to arrange and organize the executive power within a constitutional framework? Beginning with Clinton L. Rossiter (1948), John P. Roche (1952), and Edwin S. Corwin (1957), political scientists have noted this difficulty of constitutional design: executives must confine themselves to written law even as they are expected to meet the emergencies and opportunities of political life. Following their lead, Louis Fisher (1978), Richard Pious (1979), Harvey Mansfield, Jr. (1989), and Robert Scigliano (1989) have in different ways charted how constitutionalism has attempted to tame “prerogative.” In the past decade, political scientists have returned to the history of political thought and American political development to address the problem of executive power, particularly regarding the constitutionality of prerogative (Arnold 2007; Bailey 2004; Corbett 2006; Fatovic 2004; Kleinerman 2005; Thomas 2000; Ward 2005) and the ability of the people to discern and judge it (Feldman 2008; Kleinerman 2007). Given the democratic aspirations of the modern presidency, this latter question regarding the relationship between the people and the unitary executive takes on a new urgency and therefore another reason to reconsider Hamilton's political thought.

TEXT AND HISTORY

Unitarians insist that the text and history of the Constitution support their understanding of the unitary executive, but the problem is that the text is not always clear. There is, for instance, a parallel difficulty in the Constitution. With regard to appointments and treaties, Article II gives the Senate a share of the positive power to act (to give advice and consent on appointments as well as treaties), but is silent with regard to the power to change the original act (by removing an officer or terminating a treaty).

Unitarians try to resolve this lack of clarity in the text with a structural argument arising from the vesting clauses of Articles I, II, and III. An important example is Justice Scalia's dissenting opinion in Morrison v. Olson (1988). One of the questions in this case was whether a provision of the Ethics in Government Act that limited the Attorney General's power to remove an independent counsel to instances of “good cause” encroached on the president's authority to remove officers whose duties were neither legislative nor judicial, but “purely executive.” Writing for seven of the eight participating justices, Chief Justice Rehnquist argued that the act did not interfere with the president's duty to execute the laws; that is, he concluded that the formal category of the independent prosecutor mattered less than the functional result (Farber 2006). So, instead of asking whether a prosecutor is by nature an executive, legislative, or judicial officer, Rehnquist argued that the real way to resolve the separation of powers question was to ask whether the act interfered with “the role of the Executive Branch” (Morrison v. Olson 1988, 693). In dissent, Scalia argued that the majority's opinion had violated separation of powers by announcing “open season upon the President's removal power” (Morrison v. Olson 1988, 727). As he put it, separation of powers includes “fortifications,” such as the veto power, as well as structural arrangements:

But in addition to providing fortification, the Founders conspicuously and very consciously declined to sap the Executive's strength in the same way they weakened the Legislature: by dividing the executive power. Proposals to have multiple executives, or a council of advisers with separate authority were rejected. Thus, while “all legislative Powers herein granted shall be vested in a Congress of the United States, which shall consist of Senate and a House of Representatives,” “the executive Power shall be vested in a President of the United States” (698–99, emphasis Scalia's)

As Scalia read it, the difference in vesting clauses suggests that the Framers envisioned an executive who would be able to control other executive officers, which means that Congress may not limit the president's power to remove executive officials. Put simply, the vesting clause “does not mean some of the executive power, but all of the executive power” (705, emphasis Scalia's). Scalia's argument has since been defended and extended by Steven G. Calabresi and Kevin H. Rhodes (1992), who point to the vesting clause of Article III to enlist defenders of judicial independence to the cause of the unitary executive. In their reading, scholars who find in the Article III vesting clause a “substantive grant of power,” that is, a power that cannot be lessened by congressional tinkering with jurisdiction, must also accept a similar reading of Article II, which would mean that the vesting clause in Article II also gives a substantive grant of power to “a president.”

The problem with this argument from the vesting clause, however, is that it simply pushes the question back because it requires that we first determine whether a power is executive in nature. There is no better example than the famous debate between Hamilton and James Madison over the Neutrality Proclamation of 1793. The question then was whether the president had the power to declare neutrality when the United States was obliged by a previous treaty to aid France in the event that France went to war with England.⁶ Hamilton took the pseudonym Pacificus to argue that the president must be able to interpret treaties, and even decide whether they apply to present circumstances, in order to execute them (Hamilton 1961–87 [hereafter PAH] 15: 33–43). Like modern unitarians, Hamilton pointed to the “different mode of expression” in the vesting clauses (15: 39) to show that the vesting clause in Article II gave a “comprehensive grant” and that exceptions to that grant “are to be construed strictly” (15: 42). Because the powers of foreign relations were by nature executive powers, and because the Constitution did not specifically limit the president's power to proclaim neutrality as it did with the power to make treaties and declare war, it must “of necessity” belong to the executive (15: 38). Madison, however, took the name Helvidius to dispute Hamilton's claim that the powers of “making war and treaties, are in their nature, executive” (Madison 1962–91 [hereafter PJM], 15: 67). As he put it, a treaty is “not an execution of laws,” but rather is more like a law in that it has the “force of law” and needs to be executed (15: 69), which would mean that it is more legislative in nature. Hamilton's reading, Madison charged, relied less on nature than it did on British history and practice, a history and practice that had been rejected as insufficiently republican (15: 72). If the text and structure of the Constitution is unclear, as suggested by the exchange between Madison and Hamilton, what then?

To be sure, it could be argued that the difference in 1793 between the former collaborators arose not out of constitutional interpretation but rather partisan maneuvering. Madison, after all, had previously made a proexecutive argument from the vesting clause. In what is often called the Decision of 1789, when the First Congress created the executive departments, the question arose as to whether the Constitution granted the president the power to remove high executive officials.⁷ In the House, some members argued that Congress could delegate the power as it saw fit, whereas others argued that executive officers could be removed only by impeachment. Madison, whose position ultimately won, argued that the removal power was executive. Surprisingly, there were resemblances between Madison's argument and Hamilton's later defense of the Neutrality Proclamation, and even to that of modern unitarians. He pointed to the difference between the vesting clauses to assert the rule that there is a broad grant of executive power to the president (Annals of Congress 1789, 481). Then, he argued that the “association of the Senate” in appointing was an exception and “exceptions to general rules, I conceive, are ever to be taken strictly” (516). Paving the way for unitarians, Madison claimed that this principle is made clearer by the example of the judiciary, in which it is obvious that Congress may not locate the judicial power outside the judiciary. If Congress may not change the president's executive authority,

The question now resolves itself into this, Is the power of displacing an executive power? I conceive that if any power whatsoever is in its nature executive, it is the power of appointing, overseeing, and controlling those who execute the laws. (481)

As we will see, Madison made this case in a way that distinguishes him from Hamilton and from modern unitarians with regard to unity in the executive; however, the point for now is that Madison considered the removal power, but not the treaty power, to be executive in nature.

This fact leads to more difficulties. First, it is important to note that Madison was in a complex rhetorical position because the opponents of presidential removal powers in 1789 wielded a passage from The Federalist as evidence. In No. 77, Hamilton wrote, “the consent of that body [the Senate] would be necessary to displace as well as to appoint” (Hamilton, Jay, and Madison [1787–88] 2000 [hereafter FP] 489). Although Hamilton was publicly silent during the removal power debate, his earlier comment in The Federalist, with Madison's apparent flip-flop in 1793, adds to the textual dilemma. Why would Hamilton, the great defender of energy in the executive, take a narrow view of executive removal powers in 1788? Why did Madison, the 1793 critic of broad executive powers in foreign relations, offer a broad defense of executive powers in 1789? Second, Madison and Hamilton were not the only early Americans to disagree about what powers were granted by Article II's vesting clause. In fact, Madison's claim that the president should possess the power to remove without seeking the advice and consent of the Senate was by no means uncontested. In the Senate, presidential removal powers were secured only after Vice President John Adams broke the tie. In each house, the debate pitted signers of the Constitution against each other, and, from Chief Justice William Howard Taft's arithmetic in Myers v. United States (1925, 114–15), we know that of the 18 members of Congress who were also members of the Convention of 1787, 12 supported presidential removals—but 6 did not. Like Hamilton's evolution and Madison's inconsistency, the diversity of views among the Founders regarding the removal power poses a problem for arguments based on the history of the early republic.

Normally, unitarians appeal to their methodology to resolve these problems. Calabresi and Prakash (1994, 553), for example, lay out their method of originalism with the following set of rules, in order of significance:

1. 1. Consider the “plain meaning” of the words of text; if necessary, use a dictionary and grammar book from the time

2. 2. If the meaning is still ambiguous, consult other public statements made at the time of ratification

3. 3. If the meaning is still ambiguous, consider private statements made prior to, or at the time of, ratification

4. 4. If the meaning is still ambiguous, consult statements or practice after ratification.

Central to their method is an emphasis on preenactment history over postenactment history. So, for example, a statement by Hamilton or Madison in the Constitutional Convention of 1787 or in the ratification debates would be more authoritative than a statement by Hamilton or Madison from 1800 because the former would be free from partisan motivation or institutional loyalty. Or, in a somewhat different way, unitarians might follow John Yoo and emphasize the distinction between “original understanding” and “original intent” (2005, 27–29). Yoo appeals to the original understanding to move away from “pure intellectual history” of the important figures of the early republic and instead focuses attention on the “theory and practice” of the British Constitution, the state constitutions, Articles of Confederation, and arguments at the state ratifying conventions. Either approach, according to this argument, preserves the virtues of originalism but avoids mistaking one Founder's partisanship for a constitutional principle.

The problem is that Hamilton's essays on the executive in The Federalist complicate these neat methodologies. On the one hand, The Federalist meets either methodological test because it is both a preratification source untarnished by postratification partisanship and an important source of information for those ratifying the Constitution in state conventions. However, on the otherhand, The Federalist defies these characteristics. As we have seen, Hamilton in that work wrote that the Senate would share the removal power. Other problems abound. In The Federalist No. 75, Hamilton concluded that the treaty power was “more of the legislative than of the executive character” (FP 479), but, as we have seen, in 1793, as Pacificus, Hamilton wrote that the treaty power was executive in character. Likewise, in The Federalist No. 69, Hamilton assured readers that the power to receive ambassadors “is more a matter of dignity than of authority” (FP 444); yet, as Pacificus, Hamilton relied on the power to make the point that the president would have to determine whether to observe a treaty with a nation that has just undergone revolution (PAH 15: 41). One way to explain these difficulties would be to say that Hamilton was deceiving his readers in order to sell the Constitution to those worried about executive power, as John Yoo (2005, 122–25) suggests of Hamilton's treatment of the treaty power in The Federalist No. 69. Perhaps difficulties such as these explain why for all their insistence on an original Constitution, Calabresi and his allies have written a book-length examination of the postratification history of the removal power (Calabresi and C. Yoo 2008; C. Yoo and Calabresi 1997, 2003, 2004; and C. Yoo, Calabresi, and Nee 2004).

THE IMPORTANCE OF HAMILTON TO THE UNITARIAN PROJECT

However, there is a more important way in which unitarians make their case. Even if the arguments from the text and the history of the Constitution were conclusive, they would still need to demonstrate the continuing worth of the unitary executive. Accordingly, unitarians make the normative claim that the unitary executive is conducive to democratic theory. Consider Scalia's dissent in Morrison. More than merely formal, as one account (Farber 2006) characterizes it, Scalia added the functional benefits of the unitary executive. As he put it, the unitary executive would “preserve individual freedom” (Morrison v. Olson 1988, 728) and “achieve a more uniform application of the law” (732). More important, at the heart of Scalia's functional argument is his claim that the Founders believed a unitary executive was essential for democratic accountability. Under Scalia's reading, the difference between the vesting clauses of Articles One and Two “is the difference that the Founders envisioned when they established a single Chief Executive accountable to the people: the blame can be assigned to someone who can be punished” (731). Put more directly, the “primary check against prosecutorial abuse is a political one” in that presidential elections ensure that the “unfairness will come home to roost in the Oval Office” (729). By creating an executive officer who can “erod[e] the public support” of the president and who is irremovable by the president except for standards created by Congress, the act violated separation of powers while striking at the democratic reservoir of the president's power. As Scalia put it, executive unity and democratic accountability go hand in hand. Following Scalia, other unitarians have made the argument from democratic theory to add heft to their arguments from the vesting clause. Calabresi (1995) made a normative case for the unitary executive on the grounds that the modern president can solve “the Congressional Redistributive Collective Action Problem.” As he put it, the president's national electoral basis can serve as the antidote for the regional pressures placed on Congress by its selection process. So, according to this third claim, democratic theory requires responsibility, and unity in the executive is the best—maybe the only—way to ensure that administration is accountable to the people. If Congress were able to direct an executive officer to contravene the will of the president, then the people would be unable to judge whether the president had been faithful in executing the laws.

Again, Hamilton is central to this account. In The Federalist Nos. 70–72, Hamilton explained that unity rather than plurality in the executive would be safe in the “republican sense” because it would allow the people to extend blame and credit to their presidents. Hamilton's explication and defense of unity in the executive seems to provide both the evidence and the logic for the unitarian argument. According to McKenzie (2008, 25), the first occurrence of the term unitary executive in a federal court opinion was in re Sealed (838 F. 2d 476), the independent prosecutor case building up to Morrison. In his majority opinion, Judge Laurence Silberman leaned heavily on Hamilton's No. 70 to make a case for the unitary executive, citing Hamilton eight times. On appeal, Solicitor General Charles Fried pointed to Hamilton in his brief before the Supreme Court and repeated the term (MacKenzie 2008, 22), later explaining (Fried 1991, 153–54) that Hamilton “craved clarity” because clarity was a precondition for liberty. Following Silberman and Fried, Scalia enlisted Hamilton to make a case on the merits in addition to that of intent:

As Alexander Hamilton put it, “[t]he ingredients which constitute safety in the republican sense are a due dependence on the people, and a due responsibility.” The President is directly dependent on the people, and since there is only one president, he is responsible. (Morrison v. Olson 1988, 729)

Later, pointing to Scalia's opinion, Calabresi and Prakash (1994, 597) predicted that 30 years of historical research would likely support a “Hamiltonian reading” of the relevant texts of the Constitution. Indeed, it would be impossible to read Calabresi and Prakash's seminal law review without noticing the importance of Hamilton to the unitarian project. Like Scalia, Calabresi and Prakash (1994, 614) rely on Hamilton's defense of unity as an ingredient of energy in the executive (No. 70) as well as Hamilton's claim that that administrative officers would need to be under the president's “superintendence” (No. 72). So, too, with Calabresi's self-described normative argument:

Hamilton asserted that a unitary executive would both cause power and energy to accrue to the office and facilitate public accountability for and control over how that power and energy was exercised. Thus, whereas a plural executive would both dilute executive energy and popular accountability and control, a unitary executive would lead to the opposite result. Executive energy would be enhanced and so would the likelihood that it would be used in conformity with the interests of the nation. A unitary executive would thus be an accountable executive: a “tamed prince” whose actions would promote the general welfare. (1995, 44–45)

Likewise, in their ongoing historical examination, Calabresi and Christopher Yoo assert that “Hamilton was one of the strongest defenders of executive power, energy, and unity during the founding era” in Yoo and Calabresi (1997, 1481). Specifically, from Hamilton's loyalty to George Washington during Hamilton's tenure as secretary of the Treasury, and from his “sophisticated textual argument for presidential power in the foreign policy context and over removals” (1486), they find in Hamilton an early, and the most important, defender of the domestic unitary executive.

In a similar way, unitarians in foreign policy rely on Hamilton to make the democratic case for the unitary executive. In the minority report of the congressional committee investigating the Iran Contra Affair, which was at least coauthored by Dick Cheney and is regarded by Jack Goldsmith (2007, 88) as a founding document of the new defense of executive power, analysis of Hamilton's writings in The Federalist comprises at least one-half of the evidence for the chapter presenting the “Framers’ Intentions” (U.S. Congress 1987, 457–60). Pointing to Hamilton's No. 70, the minority concluded that foreign policy “decisions in the hands of Congress was considered [by the Framers] to be less democratic than giving them to the President, because there would be no way for the people to hold any one person accountable for a legislative decision” (460). Hamilton also loomed large John Yoo's “influential” (Goldsmith 2007, 97–8) memorandum of September 25, 2001, which argued that the president had inherent constitutional authority to respond to the attacks of September 11. There, John Yoo, who has since described Hamilton as his “role model” (2006, xii), relied heavily on Hamilton's writings in The Federalist in the section making the case from the Constitution's text and structure (U.S. Department of Justice, office of the Legal Counsel 2001). In another work, John Yoo linked at least some of the powers of the unitary executive to its intended democratic virtues: a “prominent theme in the federal Constitution,” he wrote, is that the president “is seen as representative and protector of the people” (2005, 71). Finally, in his dissent in Hamdi v. Rumsfeld (2004), Justice Thomas cited Hamilton's No. 70 to assert that “the Founders intended that the President have primary responsibility—along with the necessary power—to protect the national security and to conduct the Nation's foreign relations. They did so principally because the structural advantages of a unitary Executive are essential in these domains” (580).

To summarize, Hamilton is important to the argument from the unitary executive for two reasons. First, unlike Madison, Hamilton seems to offer a consistently proexecutive body of writings throughout his career, thereby offering an opportunity to use postratification practice to illustrate preratification understanding. Second, his defense of unity in the executive offers a democratic logic that can transcend history and speak to modern political practice, even as it offers a way to understand that history. That is, the normative case for the unitary executive is so important because it points away from the problems with the text and history of the Constitution, or, put differently, it softens questions like “which Founder is more important” by offering a way for Hamilton and Madison to agree. By appealing to democratic theory, it pulls the unitary argument into order.

THE PROBLEM OF ALEXANDER HAMILTON AND THE REMOVAL POWER

However, a question remains. If Hamilton is the father of the unitary executive, why did he write in The Federalist that the president would share the removal power with the Senate? One possibility is that Hamilton's comment in The Federalist was not his last word on removals. According to Jack N. Rakove (1996, 287), Hamilton “simply did not think through the implications” of his statement in The Federalist and then “repudiated” it a few years later. Alternatively, Hamilton may have deliberately understated executive powers in order to sell the Constitution. There is some support for this interpretation. In his opinion on the constitutionality of the bank, after Congress had decided in favor of presidential removal powers in 1789, Hamilton approvingly cited the removal power decision as an example of the necessity of “constructive powers” (PAH 8: 106). Somewhat more publicly, as Pacificus in 1793, Hamilton again pointed to the 1789 decision in Congress to support his own argument for broad executive treaty powers:

With these exceptions the EXECUTIVE POWER of the Union is completely lodged in the President. This mode of construing the Constitution has indeed been recognized by Congress in formal acts, upon full consideration and debate. The power of removal from office is an important instance. (PAH 15: 40)

As we might expect, unitarians such as Calabresi and Christopher Yoo conclude that the Pacificus passage indicates that the great defender of unity of the executive was at first wrong about the domestic unitary executive or perhaps had been less than frank: “It appears that by the time Hamilton wrote his Pacificus letters, he had completely disavowed the views expressed in The Federalist No. 77 and had fully embraced both presidential removal power and implicitly a power to control all exercises of law execution as well,” Yoo and Calabresi (1997, 1488). However, this claim of Hamilton's change of heart suffers one major problem: in these passages, Hamilton is endorsing ways to interpret the Constitution, not presidential removal powers. In fact, Hamilton did not in the 1793 Pacificus passage, or in any other, renounce what he had written in The Federalist, nor did he ever explicitly endorse presidential removal powers.

There were four occasions when Hamilton might have done so. The first opportunity was the Decision of 1789. Recall that it was Madison, and not Hamilton, who argued for presidential removal powers and whose coalition eventually carried the day. This opportunity was arguably the greatest because it was prior to the rise of the Republican opposition and therefore not partisan on its face, and because Hamilton's own argument against presidential removal powers from The Federalist No. 77 was cited in the debates by those arguing for Senate participation in the removal power (Annals 1789, 474–75). On the one hand, if Hamilton had recognized his error in The Federalist, or was willing to be more forthright about the scope of executive power now that the Constitution had been ratified, this would have been the time to go on record as changing his mind. On the other hand, Hamilton was also known at the time to be a candidate for secretary of the Treasury, so perhaps he had reason to remain silent. The second opportunity arose when John Adams fired members of his cabinet. Hamilton might have defended Adams against his critics, but instead distributed a pamphlet criticizing Adams' judgment (PAH 25: 169–234). Again, the problem is that Hamilton was personally connected because the men fired were cronies of Hamilton. The third opportunity came when Jefferson as president removed Federalists from office and offered a sweeping defense of presidential removal powers. As we might expect, Hamilton did not come to Jefferson's aid with a defense of executive power; instead, it is possible that he commissioned a surrogate to criticize presidential removals on the grounds that they would weaken administration by making administrative office less attractive to qualified men.⁸ In each case, Hamilton chose to remain on the sidelines in the debates on the removal power, perhaps for partisan or personal reasons.

The fourth chance came when Jefferson and his party proposed the repeal of the Judiciary Act. This time, Hamilton entered the fray by writing a series of essays called The Examination, and, significantly, he left evidence suggesting his suspicions about presidential removal powers. At first glance, Hamilton appeared to do what he had not yet done—criticize the policy but concede the power. In No. 12, he explained,

There are two modes known to the Constitution, in which the tenure of office may be affected—one the abolition of the office; the other the removal of the officer. The first is a legislative act, and operates by removing the office from the person—the last is an Executive act and operates by removing the person from the office. Both equally cause the tenure, enjoyment, or holding of the office to cease. (PAH 25: 540)

So, it would seem that Hamilton here, and for the first time, recognized removals as a presidential power; however, he continued and complicated this discussion a few essays later in No. 17, where he rejected the theory that “all offices are holden of the president” (PAH 25: 570). In this essay, Hamilton reasoned his way back to his position that the Senate would participate in removals just as it did in appointments: “The appointment is indeed confined to a particular organ, and in instances in which it is not otherwise provided by the Constitution or the Laws, the removal of the officer is left to the pleasure or discretion of that organ” (PAH 25: 570). Having endorsed the principle that the power to remove is incident to the power to appoint, he then explained how removals could be directed under the legislative power to provide for the general welfare:

A further topic of argument is that our doctrine would equally restrain the legislature from abolishing offices held during pleasure. But that is not true. The two things stand on different ground. First, the Executive has such an agency in the enacting of laws, that as a general rule, the displacement of the officer cannot happen against his pleasure. Second, the pleasure of the President, in all cases not particularly excepted, is understood to be subject to the direction of the law. Third, an officer during pleasure, having merely a revocable interest, the abolition of his office is no infringement of his right. In substance he is a tenant at the will of the government, liable to be discontinued by the Executive Organ, in the form of a removal; by the Legislative in the form of an abolition of the office. These different considerations reconcile the legislative authority to abolish, with the prerogative of the Chief Magistrate to remove, and with the temporary right of individuals to hold. And therefore, there is no reason against the exercise of such an authority; nothing to form an exception to the general competency of the legislative power to provide for the public welfare. Very different is the case as to the judges. (PAH 25: 573)

Even as he seemed to grant a presidential “prerogative” in removal powers, Hamilton qualified it by noting that the president's pleasure “is understood to be subject to the direction of the law” as well as to the “general competency of the legislative power to provide for the public welfare” (PAH 25: 573). This is to say that the president's power to remove executive officers at “pleasure” could be subject to legislative direction or reclassification. In what would be his last statement on removal powers, and his only clear statement on them since 1793, Hamilton attempted to cabin executive removal powers rather than expand them.

To be sure, each of these four missed opportunities can be explained with reference to some other objective of Hamilton, so none offers a clear case for Hamilton's position on removals. However, the same can be said for Hamilton's alleged 1793 turnaround as Pacificus. According to Harvey Flaumenhaft (1992, 292–323), Hamilton in 1793 might have conceded the point about removals in order to achieve the larger and more important point that the executive power, minus explicit “exceptions,” is “completely lodged” in the president. If Flaumenhaft is correct, then Hamilton was willing to sacrifice his position on removals in order to win the power to direct foreign policy. But this only returns us to the original question: why would Hamilton, the great defender of executive power, be reluctant to concede the point about removals? To answer this question, we must reexamine Hamilton's argument against presidential removals in The Federalist to ask if it was necessarily connected to his larger argument for energy in the executive. Lest we mistake a central point for a minor one, the seeming inconsistency requires a second look at the larger theory.

UNITY AND RESPONSIBILITY IN THE FEDERALIST

A closer look at Hamilton's presentation of energy in The Federalist points to a consistency in Hamilton's writings that reveals him as less than enthusiastic about strong presidential removal powers. As is well known, Hamilton argued that “unity” in the executive would be conducive to energy in the executive while making executive energy safe in “the republican sense” (FP 70: 447–48). Hamilton's first argument about the relationship between unity in the executive and energy in the executive is straightforward and only needs to be briefly summarized. Unity was an ingredient of energy in that it, as opposed to plurality, would allow “decision, activity, secrecy, and despatch” (FP 70: 449). Although deliberation would be a virtue in legislative operations, deliberation would embarrass the nation by crippling it during wartime or even by encouraging factious division during routine administration (FP 70: 449–52). As he put it, both history and common sense confirmed that effective administration would be more likely from a singular rather than a plural executive.

However, Hamilton's second and what he described as an especially weighty argument about safety in the republican sense deserves more attention. In The Federalist Nos. 70–72, Hamilton showed how unity would provide the accountability required by republican principles: as opposed to plurality, which is a “clog” on an executive's “good intentions” and “a cloak to his faults,” unity allows republicans to bestow praise and blame with relative ease. Alone, however, unity could not render energy safe for republican liberty because praise and blame would be mere words, like barriers of parchment, without some way for praise and blame to be credited. Thus, Hamilton explained that unity would need to be implemented by an institutional hitch, or “duration,” to make the case that energy was safe in addition to being essential.⁹

For Hamilton, duration consisted of two parts: term of office and eligibility for reelection. Counter to republican theory at the time (Federal Farmer [1788] 1981, 2:310–14; Montesquieu [1748] 1989), Hamilton showed that eligibility for reelection could work with unity to use accountability to appeal to the ambition or interest of the sitting president, because “it is a general principle of human nature, that a man will be interested in whatever he possesses, in proportion to the firmness or precariousness of the tenure by which he holds it” (FP 71:457). So, in addition to giving the president an adequate amount of time in office, it was essential to give the president an opportunity to serve additional terms. Like unity, eligibility for reelection was essential to republican safety because it would connect the interest of presidents, those strange and “noble” men whose minds were ruled by a passion for fame (FP 72: 464), to the public good. By giving the president indefinite prospect for reelection, the president would be willing to take on extensive and arduous projects, and the people would be able to reward successful presidents with another term in office, thus making long-term administration both possible and responsible (FP 72: 463–68).

However, Hamilton complicated this otherwise straightforward presentation. Before showing how eligibility for reelection was part of duration, Hamilton signaled that he would back away from his argument for unity in the executive. In the first paragraphs of The Federalist No. 72, Hamilton wrote of the “intimate connection between the duration of the executive magistrate in office and the stability of the system of administration” to warn against the “ruinous mutability in the administration of the government” that would result from frequent rotation of the “men who fill the subordinate stations.” Specifically, Hamilton mentioned the likelihood of removals as another reason to prepare the way for his argument for perpetual eligibility as one of the two aspects of duration:

To undo what has been done by a predecessor is often considered by a successor as the best proof he can give of his own capacity and desert; and in addition to this propensity, where the alteration has been the result of public choice, the person substituted is warranted in supposing that the dismission of his predecessor has proceeded from a dislike to his measures; and that the less he resembles him, the more he will recommend himself to the favor of his constituents.

Remarkably, Hamilton here went out of his way to show that elections would add to ordinary human pride in encouraging new presidents to remove officers appointed by prior presidents. Put another way, responsibility—the very thing that makes unity safe for republican government—would itself need to be checked. Because a change of chief magistrate would increase the likelihood of removals, it was better to limit the frequency of the change of magistrate by allowing the president to serve unlimited terms. Fewer presidents, according to this logic, would beget fewer removals. Even though unity could meet the demands of republican theory, unity and duration combined would bring the “advantage of permanency” to republican government (FP 72:462–63).

Hamilton's later mention in The Federalist No. 77 of the shared removal powers was thus more than an off-hand remark because it completed his larger discussion about the dangers incident to presidential selection. As he explained in that essay, which would be his last in The Federalist on the executive power,

A change of the Chief Magistrate, therefore, would not occasion so violent or so general a revolution in the officers of the government as might be expected, if he were the sole disposer of offices. Where a man in any station had given satisfactory evidence of his fitness for it, a new President would be restrained from attempting a change in favor of a person more agreeable to him, by the apprehension that a discountenance of the Senate might frustrate the attempt, and bring some degree of discredit upon himself. (FP 77: 489)

In his emphasis on the demonstrated fitness of the officers of the government, Hamilton redirected the focus from electoral responsibility to administrative expertise. As a result, he continued the thread he had started in The Federalist No. 72, where he warned about the relation between stability in the law and the duration of the president, but now he explicitly revealed how the Senate's stability would be a check on executive unity:

Those who can best estimate the value of a steady administration will be most disposed to prize a provision which connects the official existence of public men with the approbation or disapprobation of that body which, from the greater permanency of its own composition, will in all probability be less subject to inconstancy than any other member of the government. (FP 77: 489)

In this striking appeal to those men who know the value of connecting the will of public men—men ruled by interest rather than the passion for fame?—to the most stable department of government, Hamilton completed his presentation of two different executives. Because “existence,” which is another word for “will,” was inseparable from the approval and disapproval of the Senate, executive officers would be more stable than their presidents who, being free from Senate control, would be more subject to “inconstancy” than their executives (FP 77: 489). In this light, Hamilton's discussion in No. 72 of the first “ill effect” of a limit on the president's eligibility might have been written with one eye on these executive officials: even the “love of fame” would be insufficient to persuade a qualified man to serve in an administration if he “foresaw that he must quit the scene” whenever the administration changed hands (FP 72: 463–64). Forrest McDonald (1979, 131) was thus correct to speculate, “Had Hamilton's interpretation [in No. 77] been sustained, the door would have been opened to a permanent ministry independent of the president—or, as in the parliamentary system, one responsible to the legislative as well as the executive.” Anticipating the Progressives, Hamilton tried to steady administration by protecting it from the presidential electoral cycle.

To summarize, Hamilton's initial argument for energy in the executive, even with its surprising praise of the Roman dictator, suggested that unity could complement—and even perfect—republican liberty. However, in his argument connecting unity to duration, Hamilton went out of his way to warn against the republican tendencies of a president who would have a fixed term and be eligible for reelection. Because each president would want to curry favor with “his constituents” by removing executive officials associated with the repudiated ex-president, unity and republican principles would conspire against stability and thus steer the republican government toward traditional republican vices. In his notes of the debates at the Federal Convention, Hamilton recorded, “At the period which terminates the duration of the Executive there will be always an awful crisis—in the National situation” (PAH 4: 164). For Hamilton, Senate participation in the removal power, it seems, would be a way to prevent such a crisis. Later, when presented with the opportunity, he did not write that the president must have the power to remove executive officers in order to make execution “responsible” or even efficient. Instead, he left that argument to be made by those who would become his partisan and philosophic opponents.

JAMES MADISON AND THE CASE FOR RESPONSIBILITY

It was Madison, not Hamilton, who asserted the president's power to remove in the Decision of 1789. However, this is not to say that Madison was a Hamiltonian in 1789, but rather that Madison's argument brings into focus Hamilton's fear that unity in the executive could undermine stability by subjecting those who are charged with the “actual conduct” of government to the electoral cycle (FP 72: 462). Although Madison did say that the removal power was executive by nature and thus granted by the vesting clause, he also argued that the executive was the branch most “responsible” to the people. It is this political term, which Madison “apparently coined” (Mansfield 1989, 270), that provides a clue to understanding the heart of the debate between Hamilton and Madison.

During the first debate in the House, on May 19, Madison explained that “one of the most prominent features of the constitution, a principle that pervades the whole system,” is “that there should be the highest possible degree of responsibility in all the executive officers.” Any measure “which tends to lessen this responsibility” and is not “saddled upon us expressly by the letter of that work” is contrary to its “spirit and intention.” Responsibility in the executive power would be undermined by the Senate's sharing the removal power:

If . . . he shall not be displaced, but by and with the advice of the Senate, the President is no longer answerable for the conduct of the officer; all will depend on the Senate. You here destroy a real responsibility without obtaining even the shadow; for no gentleman will pretend to say, the responsibility of the Senate can be of such a nature as to afford substantial security. (Annals 1789, 395)

Madison's argument rested on two claims, that sharing the removal power would destroy executive responsibility and that the Senate could not offer any responsibility to make up for the loss.

The first claim is straightforward. The participation of the legislative branch would undermine responsibility by destroying unity. As Hamilton had noted in The Federalist, a council of appointments in New York, comprised of the governor and three to five members of the Senate, had resulted in a system in which everyone agreed that the appointments were “improper,” but nobody agreed as to who was to blame (FP 77: 491–92). Madison applied the same logic to removals. By giving the Senate a share of the removal power, “you make the executive a two-headed monster . . . you destroy the great principle of responsibility, and perhaps have the creature divided in its will, defeating the very purposes for which a unity in the executive was instituted” (Annals 1789, 519). In 1834, to counter Whig efforts to give the Senate a share of the removal power, Madison explained that the Senate would be able to force a “continuance in office” of those in “a state of open hostility” toward the president and thereby “double the danger of throwing the Executive machinery out of gear, and thus arresting the march of the Govt. althogether” (Madison 1910, 9: 534–36, 560–63).

The second claim, however, requires more elaboration, and if true, would shed light on the first. There are several explanations for this more difficult assertion that there is a difference in the quality of responsibility between the Senate and the president. One possibility is the Senate's mode of election. By being more distant from the people—elected by state legislatures and holding staggered terms of six years—Senators would be less concerned about public opinion and therefore less responsible. As Madison put it, the Senate was a “permanent body,” which by its staggered or “particular mode of election” would be “in reality existing for ever” (Annals 1789, 519). Appealing to lingering anti-Federalist sentiment, Madison said that the Senate possessed a portion of “aristocratic power” and was therefore less trustworthy than the president. Or, to put it in the vocabulary of The Federalist No. 51, the Senate's “mode of election” resulted in a particular “principle of action,” one that was unsuitable for the removal power (FP 51: 332).

There is another explanation for Madison's claim that the executive is the most responsible department, and it clarifies the preceding one. It is possible that Madison believed that the state equality of representation in the Senate made it less fit than the president to oversee national administration. Madison alluded to this character fault of the Senate in a letter to Edmund Pendleton (PJM 12: 252). In this report of the 1789 debate on the removal power, Madison summarized the “four constructive doctrines” articulated in Congress and provided the arguments for and against each position. Taking up the argument that the power to remove was incident to the power to appoint (Hamilton's position in The Federalist), Madison explained that it would have encouraged executive officers to find a party in the Senate and make themselves irremovable. More broadly,

It transfers the trust in fact from the President who being at all times impeachable as well as every 4th year eligible by the people at large, may be deemed the most responsible member of the Government, to the Senate who from the nature of that institution, is and was meant after the Judiciary & in some respects with[ou]t. that exception to be the most unresponsible branch of the Government. (PJM 12: 252)

In this remarkable passage, Madison indicated both that the president is the most responsible officer and that the Senate is in some respects less responsible than even judges, who hold their office for good behavior. This is to say that responsibility cannot be merely length of term and mode of selection because the judiciary would clearly be less responsible than the Senate, and the House more responsible than the president. Instead, if responsibility includes the breadth of election, then the fact that the president represents the people “at large,” rather than the people in their state or local capacities, means that the president would be the most responsible. Senators, however, owed their existence to the “federal” compromise and thus could be considered as less responsible than judges nominated by the president. Or, as he put it in the debate, “Shall we trust the Senate, responsible to individual legislatures, rather than the person who is responsible to the whole community?” (Annals 1789, 519). In The Federalist No. 37, Madison wrote that the compromise between large and small states gave rise to a “fresh struggle” in the convention and influenced later debates (FP 37: 227). It is possible that in 1789, with some delegates seeking to interpret the Constitution according to its impact on state size, Madison was still fighting the good fight against state equality of representation.¹⁰

However, more than an attempt to keep the partisans of state equality from expanding the power of the Senate, Madison's defense of presidential removal powers suggests that his understanding of responsibility in the executive included an account of the president as a national officer. In the debate on June 16, Madison connected responsibility to presidential selection in order to explain why removals were executive in nature. To lay the groundwork for the constitutional argument, he first offered a principle: “I believe no principle is more clearly laid down in the constitution than that of responsibility. After premising this, I will proceed to an investigation of the merits of the question upon a constitutional ground” (Annals 1789, 480). But this time, to explain what might otherwise be too abstract, Madison prefaced his remarks with the details of presidential selection. He agreed with one representative's admonition that legislators should be guided by the “merit of the men” who will be president in the “ordinary course of things,” rather than by the “splendor of the character” of Washington, because the “power here declared is a high one, and, in some respects, a dangerous one” (479). With that note of caution, Madison explained why there would be reason to trust ordinary presidents:

When we consider that the First Magistrate is to be appointed at present by the suffrages of three millions of people, and in all human probability in a few years’ time by double that number, it is not to be presumed that a vicious or bad character will be selected. If the Government of any country on the face of the earth was ever effectually guarded against the election of ambitious or designing characters to the first office of the State, I think it may with truth be said to be the case under the constitution of the United States. (479–80)

Madison's appeal to the suffrage of three million people is striking for at least two reasons. First, taken in the context of the argument for responsibility, it comes pretty close to the claim of a mandate constructed by later presidents. Second, it seems to contradict other, earlier statements by Madison. In his 1787 report of the Constitution to Jefferson, Madison pointed to the president as an example of the general government deriving some of its authority from the subordinate governments: “The President also derives his appointment from the States, and is periodically accountable to them” (PJM 10: 211). In The Federalist, Madison had described the president as both national and federal, in that presidential selection would take into account both the people and the states (FP 39: 244). However, in the removal debate, Madison seemed to envision a national popular majority, even a popular majority that could be trusted. Although he added that the electoral college had corrected the “infirmities” of “popular selection,” and that the possibilities of impeachment and the prospect of reelection would be a check on presidents, Madison explained that he “was not afraid” to place his “confidence” in a fellow citizen whom the people had chosen. Even though a national majority of millions of people could not guarantee a Washington in the ordinary course of things, it would usually find presidents who could be trusted with what would otherwise be a dangerous power.

This curious appeal to the national popular majority as a safeguard against presidential tyranny might have been a rhetorical device to sweeten the coming argument for presidential power, but it was also central to his argument from responsibility and his larger attempt to instruct others how to determine whether a power was executive by nature. On the following day, Madison explained how removals would fit with presidential selection:

If the President should possess alone the power of removal from office, those who are employed in the execution of the law will be in their proper situation, and the chain of dependence be preserved; the lowest officers, the middle grade, and the highest will depend, as they ought, on the President, and the President on the community. The chain of dependence therefore terminates in the supreme body, namely, in the people; who will possess, besides, in aid of their original power, the decisive engine of impeachment. (Annals 1789, 518)

Unlike members of the Senate, the president would be able to claim a direct “chain of dependence” to the people and, thus, subject administration to the people's control. Madison's The Federalist No. 37 casts some light on Madison's curious appeal to a presidential electoral majority. In that essay, which is the most straightforwardly philosophic of The Federalist and served as the introduction to the second half of the essays as a whole, Madison explained that one difficulty faced by the delegates to the Convention of 1787, and perhaps even one imperfection in the Constitution itself, was combining two properties necessary in any government—energy and stability—with the “genius of republican liberty” (FP 37: 223–25). The two “requisite” ingredients demanded that power be lodged in the hands of the one (energy) and the few (stability) for long amounts of time, but the additional ingredient of republican liberty demanded that power be exercised by the many for a short of amount of time. The problem was that republicans feared the very thing about which they had no choice.

THE DIFFERENCE BETWEEN HAMILTON AND MADISON

This account of Madison's defense of presidential removal powers illustrates how Madison and Hamilton differed in their solutions to the problem of republican demands. Madison seems to have believed that the national basis for presidential selection would ensure that the removal power would be used safely and in accord with the national will. In this, he seemed to rely on his famous argument that the extended republic would solve the problem of faction by rendering a national majority, the only truly dangerous faction under majority voting rules, so weak that it would be unable to trample minority rights or seriously damage the public good (FP 10: 57–63). At the same time, Madison seemed to prefer public opinion, by way of an electoral chain of responsibility, to administration by experts freed from politics. Although Hamilton admired Madison's work at the convention enough to enlist him in writing The Federalist, Hamilton may not have been persuaded by Madison's grand thesis. From Hamilton's notes of Madison's June 6 speech at the convention, when Madison presented his extended republic argument to the convention, we know that Hamilton believed that majority faction would survive the extended republic because representatives would be just as susceptible to the same causes of faction as citizens (PAH 4: 165). For Hamilton, the better solution was a well-composed Senate, which could check faction (Madison [1840] 1987, 194). This trust in the Senate carried over into his understanding of executive energy, at least with regard to executing the law. To introduce his study of law enforcement, Hamilton mocked the “enlightened well wishers of republican government” and argued that they would have to revise republican government to accommodate energy and stability (FP 70: 447). At the same time, he worried that republican excess would undermine the stability he believed executive administration would need. By the time Hamilton finished his examination of executive energy, having warned about the dangers of having too frequent a change in presidents and having given the Senate a share of the removal power, he had explained that there was also stability in the executive department. More bluntly, Hamilton would not have accepted Madison's claim that responsibility was more important than stability.

In this context, Madison's otherwise hyperbolic characterization of the removal power debate, which book-ended the debate in the House on the Bill of Rights, makes more sense. To introduce his third and final speech, Madison said that the question of removals involved “fundamental principles” and “liberty itself” because it would decide whether the government would “retain that equilibrium which the constitution intended, or take a direction toward aristocracy or anarchy among the members of the Government” (Annals 1789, 514). It is no coincidence, then, that Madison defended presidential removal powers in the 1789 House debate by praising responsibility over stability. Madison must have been worried that Hamilton's case for stability included an executive arm comprised of men who would be formally appointed and removed by the president with the advice and consent of the Senate, but actually would have little to fear from national elections and thus serve for life. As he put it, in the absence of presidential removal powers, “Every individual, in the long chain which extends from the highest to the lowest link of the Executive Magistracy, would find a security in his situation which would relax his fidelity and promptitude in the discharge of his duty” (515–16). If Hamilton had, as Madison ([1840] 1987, 136) reported, proposed lifetime tenure (i.e., during good behavior) for the president and senators at the Constitutional Convention of 1787, then the removal debate would have been an opportunity for Madison to connect the dots.

IMPLICATIONS FOR THE UNITARY EXECUTIVE

Although it is easy to determine that the delegates to the Constitutional Convention of 1787 came to agree that a single executive was to be preferred over a plural executive, it is not so easy to determine how that decision can shed light on the more difficult question as to whether Congress or the president possesses a particular power. It is no surprise, then, that debates about separation of powers under the Constitution require theories about how to read the Constitution. This article has taken up one such theory—the unitary executive—and evaluated it by one of its own terms, namely, that Hamilton's normative argument for the unitary executive can illuminate what the text and history of the Constitution leave dark.

Although Hamilton is rightly remembered as the great defender of unity in the executive, he was no advocate for presidential removal powers. Hamilton wrote in The Federalist that the Senate should wield a veto over presidential removals and then never fully repudiated this view. Instead of a careless error, Hamilton's discussion of the removal powers fits so nicely with his discussion of duration—indeed, it seems to be central to it—that it would be difficult to remove Hamilton's discussion of removals and preserve Hamilton's discussion of energy. This complication, I have argued, grew out of Hamilton's fear that the presidential electoral clock would work to undermine executive administration. To make elections less revolutionary, and to make administration more stable, Hamilton argued against and tried to contain presidential removal powers. To put it another way, at the level of democratic theory, modern unitarians owe more to Madison's argument for presidential accountability, and less to Hamilton's case for unity in the executive, than they know. This is to say that the dispute between Hamilton and Madison concerning separation of powers can be seen not only as a fight between partisans of the executive and legislative departments, but also as a contest about the relative principles of stability and responsibility. More practically, Hamilton's case for stability and expert administration would leave room and even prepare the way for independent commissions and executives.

This is not to say that unitarians simply have the wrong Founder because Madison's executive might complicate the unitarian argument in one of two ways. The first and most likely problem for unitarians is that Madison's executive is too modest. For example, Madison's 1793 critique of Hamilton's writings as Pacificus and his handling of the War of 1812 (J. Yoo 2008, 454–57) suggest that Madison's executive would not have the latitude in foreign policy desired by those who make foreign policy arguments from the unitary executive. A similar complication for unitarians arises in Madison's argument for presidential removal powers. As we have seen from the Decision of 1789, Madison's argument was not merely about gathering executive powers under the broad contours of the vesting clause of Article II, but was also an attempt to find a way to determine whether a power was executive by nature. Central to this attempt was the claim that the president was the department most responsible to the people. In pointing to the president's unique connection to the people as a source of control on the president, Madison's departure from Hamilton's argument from rule based on expertise and stability reveals how unity might make the unitary executive more dependent on a will outside itself.¹¹

However, in a different way, a second possible problem for unitarians is that Madison's defense of responsibility is too democratic for their brand of constitutionalism. Because Madison's argument for responsibility included the claim that the president is the most responsible officer, and therefore conferred more power on the president, it laid the groundwork for what I have argued elsewhere (Bailey 2007) to be Jefferson's transformation of executive power. In direct contradistinction to Hamilton, Jefferson aimed to ground the legitimacy of execution on consent, and for this reason, avoided legalistic assertions of the constitutionality of his vigorous use of executive power (Bailey 2004; Fatovic 2004; Schlesinger 1973). As a result, Jefferson aimed at constructing an executive based on opinion instead of law: it was Jefferson who went further than Madison by, for the first time, making the case that the president should be able to remove executive officers and replace them with those from his party on the grounds that administration should reflect the majority will, as registered in presidential elections (Bailey 2007, 151–70). Whether Madison would follow Jefferson in looking to a majority will for a new kind of executive energy is a question for another study, but the point is that Madison's embrace of responsibility suggests a possible tension with textual arguments based primarily on the vesting clause. Once harnessed to a predictable election cycle, the democratic argument from responsibility rests uneasily with the constitutional argument for inherent executive powers.

And why does Hamilton matter? Set against the modern backdrop of the democratic presidency, Hamilton's account of unity in the executive in The Federalist reveals the limits of the normative argument for the unitary executive. Although energy in the executive can be made compatible with republican principles, energy is not the only necessary character of executive power. Rather, the administration of the laws also requires stability. The problem, as Hamilton perceived in his treatment of duration, is that republican principles might at times be too compatible with energy. With each praise and blame of a single president taking the form of an election, new presidents would find reason to regard themselves as repudiators of previous principles of administration. Thus, the removal power, used in conjunction with the power to appoint, would become an instrument of permanent electoral revolution. With popular majorities and new presidents conspiring against administration, leading citizens would be less likely to serve as department head. This is to say that Hamilton showed the need for not one, but two executives—one chained to electoral accountability and one freed from it. The one would use elections to make republican theory work with the requirements of energy. The other would be staffed by qualified men who would be attracted by the permanence of the office as well as its insulation from public opinion. Freed from republican theory, this second executive would be able to moderate the more republican, more unitary, president.