In a recent post on the Recess Appointments Clause, I argued that the current method (the multi-session recess appointment theory) for determining the tenure of a recess appointee is based on “precedent” which does not amount to a hill of beans. It is time now to consider afresh the question of how one ought to determine the end of the Senate’s “next session” for purposes of the RAC.
The Reciprocal Meaning of Recess and Session
Discussion of the term “recess” has generally agreed that it may refer, both today and at the time the Constitution was written, to suspensions of business of any duration. With regard to today’s usage, Professor Hartnett notes “as anyone who has ever attended elementary school, a committee meeting, or a trial can attest, a ‘recess’ is quite frequently a rather short break.” With regard to usage in the late 18th and early 19th centuries, the 11th Circuit states that “recess” was “a general term for the suspension of business,” pointing to Samuel Johnson’s “A Dictionary of the English Language” (1755) and Noah Webster’s “An American Dictionary of the English Language” (1828).
The term “session,” as used today, can refer to “a single continuous sitting, or period of sitting, of persons so assembled,” or “a continuous series of sittings or meetings of a court, legislature, or the like.” Thus, we might refer to “the afternoon session,” or “today’s session,” or the “fall session.” The degree of continuity is relative; saying that the Senate is “in session” might mean that Senators are currently on the floor conducting business, or that the Senate is conducting business today, or that the Senate is “in town” and meeting from time to time.
Although both “recess” and “session” can be used in different senses, one sense is as mutually exclusive statuses of a legislative body. As Senator Gore explained in 1814, “[t]he time of the Senate consists of two periods, viz: their session and their recess.”
This usage of the terms “session” and “recess,” which Hartnett terms the “reciprocal” usage, was recognized in the 18th century. Thus, Johnson’s dictionary defines “session” as “an assembly of magistrates or senators” or “the space for which an assembly sits without intermission or recess.” Put another way, a “recess” would break the continuity of a “session” and thereby end it. An “intra-session recess” would therefore be oxymoronic.
As an example of the reciprocal usage, when the Continental Congress proposed to establish an Office of the Secretary of Congress, it provided that the Secretary “shall attend Congress during their session, and in their recess, the committee of the states.” Journals of the Continental Congress (Proceedings of March 31, 1785). Similarly, the First Congress provided, in the Act of Sept. 22, 1789, that the Senate’s engrossing clerk, would be paid “two dollars per day during the session, with the like compensation to such clerk while he shall be necessarily employed in the recess.”
Not only do these examples show that “session” and “recess” could be used in the reciprocal sense, but, as I discussed in a prior post, the usage of “recess” and “session” as mutually exclusive underlies every discussion of the RAC up until Comptroller General Warren’s opinion in 1948. Prior to that time, there is not a single analysis of the RAC suggesting that the Senate could be in recess and session at the same time, unless one counts Attorney General Knox’s opinion that indirectly considers, but rejects, the possibility.
Does the RAC Use Recess and Session in the Reciprocal Sense?
Hartnett argues that the Constitution cannot be using recess and session in the reciprocal sense because it “plainly contemplates the possibility of recesses that do not end the session, by providing that neither house may adjourn without the consent of the other for more than three days during a session of Congress.”
I must say that I do not understand this argument. The constitutional provision to which Hartnett refers, Article I, section 5, cl. 4, does not use the term “recess” at all. It provides that “[n]either House, during the Session of Congress shall, without the Consent of the other, adjourn for more than three days . . . .” This clause doesn’t say anything about whether the “adjournments” should be considered “recesses” or whether such adjournments will or will not end the “session.”
Hartnett’s position seems to be that all adjournments are “recesses” and, since all adjournments cannot end the “Session of Congress,” some recesses must run concurrently with the Senate’s “Session” referred to in the RAC. But there are two problems with this position.
First, it is not at all clear that all adjournments are “recesses.” For this proposition, Hartnett relies on the fact that one contemporary usage of the term “recess” would encompass adjournments or breaks of even a very short duration. But he provides no evidence that this was the usage intended in the RAC. To the contrary, he acknowledges that the Framers could not possibly have intended that recess appointments be made during ordinary day-to-day adjournments. Accordingly, he suggests that adjournments of three days or less (i.e., those which each house may take without the consent of the other) should be considered “de minimis” recesses and therefore not subject to the RAC.
Second, Hartnett’s position would mean that while “recess” is being used in the colloquial sense of any break in the conduct of business, “session” is not being used the reciprocal sense of the time when business is being conducted. Instead, Hartnett assumes that the “session” is the period of time designated as such for parliamentary purposes, regardless of whether business is being conducted.
It makes little sense, however, to interpret the RAC in this fashion. Consider Hamilton’s explanation of the RAC in Federalist No. 67:
The ordinary power of appointment is confided to the President and Senate jointly, and can therefore only be exercised during the session of the Senate; but as it would have been improper to oblige this body to be continuously in session for the appointment of officers, and as vacancies might happen in their recess, which it might be necessary for the public service to fill without delay, the succeeding clause is evidently intended to authorize the President, singly, to make temporary appointments “during the recess of the Senate, by granting commissions which shall expire at the end of their next session.”
Hamilton is clearly using “session” and “recess” in the reciprocal sense. He presents two options (1) oblige the Senate to be “continuously in session” or (2) provide a mechanism for handling vacancies that “might happen in their recess.” If the Senate could be in session and recess at the same time, it would make no sense to present these as alternatives.
What did Hamilton mean by the Senate’s “session” or the Senate being “in session”? He was not referring to the time when Senators are actually on the floor conducting legislative business. It would not be necessary for Senators to remain continuously on the floor in order to act on nominations, nor would it be feasible for them to do so. Nor could he be referring simply to a parliamentary status unrelated to the Senate’s ability to act on nominations.
Hamilton’s meaning should not be mysterious. Recall that the Constitution requires Congress to “assemble at least once in every year” and sets the default date for this “meeting.” The President, moreover, is given the power to “convene” Congress on extraordinary occasions.
According to Johnson’s dictionary, to “assemble” means “to bring together into one place.” To “convene” means to “call together” or to “assemble.” A “meeting” is an “assembly.” And, as mentioned earlier, a “session” is “an assembly of magistrates or senators” or “the space for which an assembly sits without intermission or recess.”
All of these terms, therefore, refer to the members of a legislative body being brought together or assembled in one place, namely the seat of government. When Hamilton refers to the Senate’s “session” or its being “in session,” he is referring to the period of time when Senators are assembled at the seat of government, and therefore are not in their home states. The reason that it would be “improper” to require the Senate to be in session “continuously” is that this would prevent the Senators from returning to their states, which, as Professor Rappaport explains, would have been contrary to the prevailing view that “it was important for the legislature to return to the people for significant periods.”
The “recess” of the Senate is the period when Senators are not assembled at the seat of government and the “next session” (or, as Hamilton puts it, the “ensuing session”) is the next time that the Senate assembles following that recess. The end of a session begins a recess, the end of a recess begins the next session, and the end of the next session begins another recess. It really couldn’t be any simpler.
Of course, when Hamilton wrote Federalist No. 67, it was not possible for Senators to travel back and forth from their home states in a matter of hours or even a few days. Modern technology makes it trickier to determine when the Senate is in recess and when it is in session. But it doesn’t change the relationship between the recess and the session.
The reciprocal meaning of “session” and “recess” is the only one that squares with the purpose of the RAC. As Justice Story explained regarding the RAC:
The propriety of this grant is so obvious, that it can require no elucidation. There was but one of two courses to be adopted; either, that the senate should be perpetually in session, in order to provide for the appointment of officers; or, that the president should be authorized to make temporary appointments during the recess, which should expire, when the senate should have had an opportunity to act on the subject. The former course would have been at once burthensome to the senate, and expensive to the public. The latter combines convenience, promptitude of action, and general security.
J. Story, Constitution of the United States §1551 (emphasis added). Similarly, Attorney General Wirt noted that the RAC ensures that “the President shall have the power of filling [a vacancy] by an appointment to continue only until the Senate shall have passed upon it; or, in the language of the constitution, till the end of the next session.”
Under the multi-session recess appointment theory, these purposes are thwarted. The commission does not expire when the Senate has an opportunity to act on the subject or pass on a nomination. To the contrary, the President may have no reason to give the Senate the opportunity to consider a nomination when it next assembles after a recess appointment.
For example, President Obama recess appointed Richard Cordray as head of the Consumer Financial Protection Bureau on January 4, 2012. Although the Senate has been available to pass on a nomination to fill that position since at least late January, no nomination has been submitted. If Cordray’s commission does not expire until the end of the session that will begin in January 2013, as claimed by the mult-session recess appointment theory, there is little incentive for the President to seek the advice and consent of the current Senate, though it is available to give it. The multi-session recess appointment theory thus undermines one of the key purposes of the RAC, facilitating the exercise of advice and consent at the earliest opportunity.
Hartnett’s Structural Argument
Only one final justification for the multi-session recess theory remains to be considered. Hartnett argues that “[b]ecause the Constitution leaves the process for controlling when recesses begin in different hands than it leaves the process for controlling when sessions begin, it must contemplate that they are not inherently reciprocal.” His argument is premised on the fact that the Constitution sets a default date on which Congress must assemble each year, and permits this date to be changed only by “law.” Because acting by “law” means the involvement of the both Congress and the President, while “recesses” are determined by Congress alone, Hartnett argues that Congress cannot start a new “session” simply by ending a recess.
This argument suffers from a significant degree of circularity. To determine how a “session” may be started, Hartnett is looking at how Congress has identified sessions for parliamentary purposes. But if the “session” referred to in the RAC is the period during which the Senate is actually assembled, as argued above, then the process for controlling when both sessions and recesses begin are in the same hands (namely the Congress), and the problem identified by Hartnett disappears.
It may be helpful here to consider why Congress divides its meetings into “sessions” in the first place. The original parliamentary purpose of a “session” was as a self-contained unit for legislative business. As Jefferson explains in his Manual of Parliamentary Practice, the end of a session means that “all matters depending before them are discontinued, and at their next meeting are to be taken up de novo, if taken up at all.”
This, however, has long since ceased to be the function of a “session” in the Congress. “At first the Congress attempted to follow the rule of the British Parliament that business unfinished in one session should begin anew in the next, but in 1818 a rule was adopted that bills should be continued at the next session of the same Congress after six days, except bills referred to committees.” William McKay & Charles Johnson, Parliament and Congress 32 (2010). Current practice in both the House and Senate is that all pending legislative business carries over from one session to the next within the same Congress. Id.
It is true that under Senate rules nominations do not carry over from one session to the next. Clearly, however, this is a matter determined by the Senate, and it would be free to provide differently if it wished. Moreover, as noted before, Senate rules also provide that nominations are to be returned to the President whenever the Senate adjourns or recesses for more than 30 days. If the RAC “session” were determined by the parliamentary consequences of an adjournment, it would make the most sense that any adjournment of more than 30 days would end a session for purposes of the RAC.
At one time Congress also seemed to believe that ending a “session” terminated Congress’s ability to re-assemble until the default meeting date (or a substitute date provided by law) the following year. Thus, in 1947 and 1948 Congress did not adjourn sine die when its legislative work was completed, but remained technically in session so that the congressional leadership could re-convene their respective houses if necessary. Congress, however, has apparently abandoned that theory because it now adjourns sine die while reserving the right for congressional leadership to recall their members if necessary. McKay & Johnson, 81-82.
Thus, as things stand now, a congressional “session” appears to be little more than a label to identify when a particular legislative action was taken. There is nothing in the Constitution that suggests Congress is required to label its meetings on a one session per year basis. If anything, the requirement that Congress meet at least once a year suggests the contrary. Nor does the fact that the Congress must act by law to change the constitutional default date suggest that it must act by law if it wishes to hold more than one meeting a year.
For purposes of argument, however, lets assume that Hartnett is correct that Congress can only act by law if it wishes to hold more than session in a year and, if it fails to do so, all of its meetings in a particular year are to be deemed part of one session. This assumption is not necessarily inconsistent with the reciprocal meaning of recess and session because it could still be that any periods of adjournment within this session are not “recesses” within the meaning of the RAC.
Moreover, even if one were to reject the reciprocal meaning, it doesn’t follow that the multi-session recess appointment theory should be adopted. No one disputes that when the Framers wrote the RAC, they expected that the “next session” of the Senate would begin immediately following the recess during which the temporary appointment was made. If one grants that, for reasons unanticipated by the Framers, there can be “intra-session recesses” during which appointments could be made, it still would make the most sense to interpret the RAC as terminating the commissions at the next end of session (as opposed to the end of the next session). This takes a little liberty with the text of the RAC, but rather less than the executive branch has long taken with the clause in other respects.
There is little if anything to recommend the multi-session recess appointment theory. When a valid recess appointment is made, the Senate’s “next session” begins when that body re-assembles, and it ends when it next disperses or upon the constitutional default date the following year.