Northwestern University Law Review : Colloquy : 2009 : Tillman
Senate Termination of Presidential Recess Appointments
Seth Barrett Tillman[*]
[Eds. Note: In light of the interest in our presidential transitions colloquy and the continuing debate over recess appointments, we would like to highlight a debate between Seth Barrett Tillman and Brian C. Kalt that occurred in the first volume of the Colloquy.]
To a fair-minded person assessing the broad ramparts of American constitutional scholarship, it is striking how little attention legal scholars and public intellectuals pay to the text of the United States Constitution. The document is a mystery to many of them. Mostly they concern themselves with prior judicial decisions. If a clause is not litigated, it is, as Judge Posner put it, off their "radar screen." Even where a particular constitutional provision is the subject of litigation, the judicial opinions almost invariably lead to scholarship assessing the propriety of the decision, as opposed to still-open aspects of the clause's meaning. Thus the history of our legal scholarship leads to an ever narrowing legal imagination, with the rhetoric of debate escalating over constantly declining intellectual stakes.
But as citizens and lawyers interested in policy reform, including the institutional reform of our governing bodies, we should not limit ourselves to these narrow disputes. The whole of the Constitution is within our grasp, if we would but seize it. And, even today, the Constitution remains chock-full of unused (and therefore judicially untested and unapproved) powers whose potential might be used, or, at least, explored.
One such power is the power of a Senate majority to terminate a presidential recess appointment. The Recess Appointments Clause, Article II, Section 2, Clause 3, states:
Certain aspects of this clause have been adjudicated and are much discussed in the academic literature. For example, it has been questioned whether this power extends to the President's making appointments to the Article III courts, whether the President can make a recess appointment after the Senate has returned from its recess to an office which had become vacant during (or prior to) the recess, and whether the President can make a recess appointment during a recess to a statutory office that had never before been filled. These debates uniformly go to the limits of the President's power to appoint under the terms of the clause. There is no discussion of any concomitant removal power—it is just assumed that the removal power (wherever it is vested) with regard to recess appointments is coextensive with the removal power generally. This short article seeks to test that assumption.
As stated, the text of the Recess Appointments Clause provides that such appointments last until the end of the Senate's next session. The Executive Branch has also made this clear. I suggest that, after the President makes a valid recess appointment, the Senate could convene, immediately terminate its session, and then reconvene instantly. The Senate would enter its adjournment order on its journal and notify the President by message of their action. Admittedly, there is a certain "fictional" quality to this sequence. But it is precisely on such fictions that the Executive Branch has justified contentious prior appointments. For example, President Theodore Roosevelt took the position that even when the prior Congress met until the last moment of its constitutional term and the successor Congress convened immediately thereafter, that interregnum of one moment was a "recess" permitting the President to make a constitutionally valid recess appointment.
The termination of the Senate's (post-recess) session should terminate the appointment—along with any other recess appointments made by the President over the course of the recess. The Senate could not pick and choose which individual appointees would "survive" termination—it would have to choose between terminating all the recess appointments or none of them. And where the Senate objects to intrasession appointments as a matter of constitutional principle, then the Senate should not be permitted, as a normative matter, to pick and choose which appointments survive termination and which do not. On the other hand, where only one appointment was made during the recess, the Senate would have no need to pick and choose.
There are a few potential problems with the proposed stratagem.
First, it is new. Thus, although it seems consistent with the Constitution's text, it does not (yet) have judicial approval. But then again, Marbury v. Madison occupies more than forty pages in the official reporter, and Marbury only has a lone citation to a prior (foreign) judicial decision, and for an obscure point, not central to its celebrated holding relating to judicial review. My point is only that sometimes, when the text is reasonably clear, simple textual interpretation, even absent supporting precedent, is a legitimate method of constitutional interpretation.
Second, it could be argued that the word "Session" in the Recess Appointments Clause refers not to whatever the Senate chooses to designate as a Session, but rather to a year-long period in which the Senate sits, regardless of adjournments, recesses, or other breaks. If this were true, then the Senate could not decide that a "Session" was over solely by taking a vote. However, although the term session is associated in the popular mind with annual terms, it is simply not hardwired into the Constitution. Nowhere does the Constitution prescribe or even assume fixed annual sessions, although the Constitution commands that the Congress should meet at least once annually. Thus the Senate is free to depart from the expectation of distinct annual sessions. Historically, some years have had multiple sessions and arguably a session can last multiple years—at least up to two years for reasons of comity with the House's election cycle. In other words, there is no such thing as a "full year's session" unless the Senate stays in session a full year.
Third, it is not clear that the Senate, acting alone, can terminate its session and then instantly reconvene, at least absent concurrence from the House. As a textual matter it appears that the decision is one for the Senate alone to make. There are some good reasons to believe that where, as here, the adjournment is for less than three days, each house controls its sessional agenda and its journal independent of the other house. Moreover, one recent commentator on legislative procedure has taken the position that the Constitution embodies cameral autonomy as a structural norm. There are also historical and textual arguments for believing the President simply plays no role (or next to no role) in decision-making involving the Senate's decision to recess and to reconvene. For example, the Orders, Resolutions, and Votes ("ORV") Clause provides:
The traditional view of the ORV Clause is that it exempts the President from participating in (by signing or by vetoing) bicameral congressional adjournment resolutions. If this view is correct, then that is some additional and powerful reason to believe that the President does not participate in bicameral decisions to go into recess. Whether the Senate can terminate and reconvene on its own (as I believe it can), or whether such a move requires concurrence of the House, the President is not part of such decision-making. Thus, either the Senate acting alone or the two houses acting collectively can terminate a presidential recess appointment. If either view is correct, that would represent a sea-change in our current recess appointment practices. And, for reasons that I explain below, this change would be beneficial although it admittedly might upset current expectations and reliance interests.
It is generally believed that the purpose of the Recess Appointments Clause is to see to it that the administration of government does not suffer "due to a vacancy in office left unfilled while the Senate is dispersed and unavailable during its recess." Thus, the Recess Appointments Clause is a second best compelled by circumstances, i.e., an unexpected vacancy that needs to be filled at a time the Senate is unavailable. The best solution is, of course, presidential nomination in tandem with Senate advice and consent. The procedure outlined in this article would give the President a strong incentive to make recess appointments that would withstand Senate advice and consent, even if the President does not put the candidate forward on a "permanent" basis following the reassembly of the Senate. If the President during a recess appoints a person that is not supported by a majority in the Senate, the Senate could terminate the appointment when it reassembles. (And, where empowered under preexisting rules, the majority leader might call an earlier than scheduled meeting of the Senate to displace an appointee the majority actively opposes.)
One might believe that this policy reform weakens the hand of the President vis-à-vis the Senate. It would certainly temper and moderate their choices, and whether or not that is a good thing will depend largely on who is elected to the Senate and who is elected to the presidency. But the more important effect of this policy will be to make the Senate responsible to the electorate. The Senate would no longer be able to shift responsibility for appointments onto the President merely by going into recess. Similarly, the Senate could not escape responsibility to the public by saying the President's appointment is final until the Senate's next fixed or scheduled session ends. The Senate would no longer be able to play a role by mere inaction—as they have too often done in the advice and consent process. Rather, to turn out the President's nominee, the Senate will have to affirmatively vote to terminate their session upon reassembling at the end of the recess. If they refuse to exercise this power, they could be held accountable by the voters.
Thus, the Senate has the constitutional power to terminate presidential recess appointments by a majority vote, although perhaps the consent of the House of Representatives is also needed, at least in some cases. The suggested procedural innovation might have a beneficial policy result: increasing the Senate's responsibility to the electorate. I admit that this latter prediction on my part is just a prediction, and that such policy considerations certainly do not control the original public meaning of the Constitution's text. But perhaps this sort of argument offers some comfort to those afraid of institutional change. In other words, the winner here is not the Senate or the President, but it may be us.
*. Mr. Tillman is an associate of Richards, Layton & Finger, P.A., Wilmington, Delaware, and is a member of the Delaware bar. The views expressed do not necessarily represent the views of the firm or its clients. I thank Professor Aaron-Andrew P. Bruhl, Richard Albert, Stephen E. Sachs, the Federal Judicial Center's History Office, and the U.S. Senate Historical Office.
1. Richard A. Posner, Breaking the Deadlock: The 2000 Election, the Constitution, and the Courts 205 (2001) ("[A] constitutional problem, however serious, that does not leave its traces in the pages of the United States Reports is unlikely to appear on the academic radar screen.").
3. In this brief article I do not address the interplay between the proposed procedural reform and the Senate filibuster rule which requires a supermajority to terminate debate. See Senate Rule XXII (2007), available at http://rules.senate.gov/senaterules/rule22.php (link).
4. See William Ty Mayton, Recess Appointments and an Independent Judiciary, 20 Const. Comment. 515, 542, 552-53 (2004) (generally arguing that the power of the President to make a temporary judicial recess appointment should give way to the right of litigants to be free from appointments subject to domination by the political branches) (link).
6. See Letter from Alexander Hamilton to James McHenry (May 3, 1799), in 23 The Papers of Alexander Hamilton 94 (Harold C. Syrett ed., 1976) (noting that "[v]acancy is a relative term, and presupposes that the Office has been once filled") (emphasis omitted).
7. See Gaillard Hunt, The History of the Department of State, 5 Am. J. of Int'l L. 414, 424-25 (1911) (reprinting President George Washington's commission of Supreme Court Justice Thomas Johnson, which stated that the recess appointment lasted "during his good behaviour, and until the end of the next session of the Senate of the United States, and no longer"); fax from Federal Judicial Center to Seth Barrett Tillman (Jan. 3, 2007) (reproducing President George W. Bush's commission of Judge William H. Pryor Jr., which stated that the recess appointment lasted "until the end of the next session of the Senate of the United States and no longer; subject to the provisions of law") (on-file with Colloquy) (link).
8. This strategy will only work for an intersession recess appointment. An intrasession recess appointment, i.e., an appointment made during an adjournment within a given session, lasts the remainder of that session and additionally for the life of the next session. Thus, if the President has made an intrasession recess appointment, then the Senate will have to convene and terminate two "sessions" back-to-back in order to terminate the President's intrasession appointment.
9. See Edward A. Hartnett, Recess Appointments of Article III Judges: Three Constitutional Questions, 26 Cardozo L. Rev. 377, 416-18 (2005) (link) (describing President Roosevelt's contested appointment); see also Michael A. Carrier, Note, When is the Senate in Recess for Purposes of the Recess Appointments Clause?, 92 Mich. L. Rev. 2204, 2228, 2247 (1994) (arguing that presidential authority extends to intersession recesses, not to "fictional" or "constructive" recesses between back-to-back meetings of the Senate).
11. Id. at 168-69 (quoting Rex v. Barker, 3 Burr. 1265, 1266 (1762), although erroneously giving the case name as "Baker").
12. See Rappaport, supra note 5, at 1565 (acknowledging that Congress could have multiple sessions in a calendar year); but see Carrier, supra note 9, at 2218, 2223 (arguing that the founders "anticipated" one session and one intersession recess each year). Carrier does not argue that the expectation was embodied in the text.
14. See infra note 20.
15. Indeed, given the modern practice of carrying over sessional business between sessions within a given two-year Congress, arguably there is only one session (in the constitutional sense) every two years. See Senate Rule XVIII (2007), available at http://rules.senate.gov/senaterules/rule18.php (link); Cong. Globe , 30th Cong., 1st Sess. 1085 (Aug. 14, 1848) (link) (passing the progenitor of current Senate Rule XVIII that was proposed by Senator Webster); Thomas Jefferson, A Manual of Parliamentary Practice § 51 (2d ed. 1818) (explaining that all matters "before Parliament were discontinued by the determination of the session").
16. See U.S. Const. art. II, § 2, cl. 3 (link) (permitting recess appointments during the "Recess of the Senate," not the "recess of the Congress"); but see id. at art. I, § 5, cl. 4 (link) (referring to the "Session of Congress").
20. For example, the First Congress over the course of its two years had three regular sessions, exclusive of any sessions called by the President and of any Senate executive sessions. All the relevant decisions, i.e., when to adjourn and when to reconvene, were made by concurrent resolutions, absent presidential participation, impliedly establishing that the number of sessions is within Congress' exclusive control. See, e.g., 1 Journal of the House of Representatives of the United States 130 (Washington, Gales & Seaton 1826) (Sept. 29, 1789) (link) (adjourning first session to a day certain opening second session); id. at 298 (Aug. 12, 1790) (link) (adjourning second session to day certain opening third session). Although this goes some way to establish that the President plays no role, it does not establish that the two houses must act in concert. Bicameral action may have been taken for reasons of comity or only because each recess was for more than three days. See supra note 17.
21. See U.S. Const. art. I, § 4, cl. 2 (link) (making first Monday in December the default date for the annual meeting of Congress unless changed by statute), amended by id. at amend. XX (link) (changing date to January 3). The fact that Congress must meet on January 3 every year does not mandate, at least as a textual matter, Congress taking a recess prior to that date or starting a new session on that date.
23. See, e.g., S. Rep. No. 54-1335, at 1-2, 8 (1897) (link) (taking the position that the subject matter of the ORV Clause is bicameral resolutions); see also Rappaport, supra note 5, at 1558-59 (taking the position that the ORV Clause's parenthetical is tied to Article I, Section 5, Clause 4, and thus both refer to bicameral adjournment resolutions); but see Seth Barrett Tillman, A Textualist Defense of Article I, Section 7, Clause 3: Why Hollingsworth v. Virginia Was Rightly Decided, and Why INS v. Chadha Was Wrongly Reasoned, 83 Tex. L. Rev. 1265, 1347-49 (2005) (link) (urging a contrary view, albeit one that, like the traditional view, supports the position taken in the text of this article).
24. See supra note 20; cf. Rappaport, supra note 5, at 1569 ("Congress can choose to have recesses during a session of whatever length it determines. Moreover, Congress can end a session at its discretion . . . .").
26. As in baseball, where ties go to the runner, Senate ties (absent unlikely vice-presidential intervention) will leave the President's appointee in place.
27. Of course, an incumbent majority could automate this process by rule, so that after returning from each and every recess, the Senate automatically terminates its session and instantly reassembles.
Copyright 2009 Northwestern University
Cite As: 103 Nw. U. L. Rev. Colloquy 286 (2009), http://www.law.northwestern.edu/lawreview/Colloquy/2009/1/.
Persistent URL: http://www.law.northwestern.edu/lawreview/colloquy/2009/1/