by Jay Wexler
The procedure for appointing principal officers—nomination by the president and confirmation by the Senate—works just fine most of the time. If the Senate is in session, it can easily vote to confirm or not confirm the president’s nomination to a top post. But what if the Senate is not in session, and the president wants to appoint someone to a very important position in the government? Should the president have to wait until the Senate gets back? The question is important, because the Senate is by no means always in session. Each Congress (e.g., the 103rd Congress, the 97th Congress, etc.) lasts for two years and generally consists of two sessions and one intersession recess. The length of both the sessions and the intersession recess has varied over the years. Prior to the Civil War, the sessions lasted somewhere between three and six months, with the intersession recess lasting between six and nine months. These days, however, the sessions are longer, and the intersession recesses are shorter; the recesses now last somewhere between one and three months. Just as an example, the first session of the 108th Congress ran from January 7, 2003, until December 9 of the same year, and the second session ran from January 20, 2004, through December 8. The intersession recess between the two sessions, then, lasted about six weeks.
Even when the Senate is in session, though, it’s not like the senators spend their whole lives together, eating from a communal soup bowl and having pajama parties every night on the floor of the Senate chamber. The senators break for lunch and go home at night and ordinarily have holidays and weekends off. They also enjoy a number of intrasession breaks (I’d call them “recesses” except that whether they deserve to be called “recesses” is kind of the whole problem). In the early days of the Republic, these breaks were rare, but now each session might include anywhere between half a dozen and a dozen breaks, some of which can last a while. During the second session of the 108th Congress, for instance, Congress took nine breaks, most of which were about ten days long, but a couple of them lasted over a month, not much shorter than the intersession recess that divided the first and second sessions of the Congress.
The framers of the Constitution knew that the Senate would not always be in session, and they worried about what might happen to important presidential appointments if the senators who needed to confirm them were back in their home states or vacationing at Niagara Falls or something. Especially because getting from place to place was not nearly as easy as it is these days, when we can just hop on a plane and sit on the runway for six hours before zipping off to our destination, the framers were concerned that the nation might be without a secretary of state or other important official for months at a time. Thus, the recess-appointments clause: The President shall have Power to fill up all Vacancies that may happen during the Recess of the Senate, by granting Commissions which shall expire at the End of their next Session. If the Senate is in recess, and the president needs to fill an important post, he can do it without getting Senate confirmation, although the person appointed to the post can only remain in that position until the end of the session after she takes office (unless the Senate comes back and confirms her). The framers thought that the clause was important, but there’s no evidence that they thought it would be used particularly often. As Alexander Hamilton wrote in The Federalist Papers, the recess-appointments clause was meant “to be nothing more than a supplement to the [appointments clause] for the purpose of establishing an auxiliary method of appointment, in cases to which the general method was inadequate.”
The framers might not have envisioned that presidents would use the recess-appointments clause a whole lot, but from the very beginning of the Republic, they have. George Washington appointed John Rutledge to be the chief justice of the Supreme Court by using the recess-appointment power. James Madison used the power to appoint John Quincy Adams, among others, as an envoy to negotiate the peace treaty that ended the War of 1812. Abraham Lincoln used the power hundreds of times, as did Andrew Jackson. Modern presidents, too, have relied heavily on the recess-appointment clause. Ronald Reagan made 243 recess appointments. Bill Clinton made 139, once using the clause to appoint the first openly gay ambassador in the nation’s history. George W. Bush used the power to appoint 171 officials, including John Bolton, his unpopular and reportedly insufferable nominee to be ambassador to the United Nations. Bolton, probably one of the only United Nations ambassadors ever to publicly proclaim that “there is no such thing as the United Nations,” stepped down from his position in December 2006 when it became clear that the Senate was not going to confirm him.
Some recess appointments are uncontroversial. If the secretary of state were to die on the third day of an intersession recess, and the president felt the need to get a new secretary of state in there before the Senate returned in a month, surely nobody would object. If the same thing happened at the very beginning of a month-long intrasession break, again most likely the president’s decision to unilaterally appoint a new secretary wouldn’t raise many eyebrows. But what if instead the president wants to make an appointment during a weeklong intrasession break? What if the president wants to wait until the very last day of such a break to make the appointment? The idea that the president has to act immediately because the Senate isn’t in session is a complete fiction; the senators will be back tomorrow, or at least by the next week. Moreover, if the president is allowed to make a recess appointment at the end of a weeklong intrasession break, why wouldn’t it be possible to make a recess appointment after the Senate goes home for the evening, or when it steps out for a coffee break? Can the president really use the recess appointment power in this way? These are some tricky questions, and they bring us back to the case of William H. Pryor.
If there is any office for which the recess-appointment power never needs to be used, it’s the office of federal judge. The country is not going to fall apart if some court happens to have a vacancy or two for a month while our senators get some rest and relaxation in the Bahamas. And indeed, although early presidents did use the recess-appointment power to appoint judges, such recess appointments have been rare in modern times. Between 1980 and 2000, for example, not a single federal judge was put on the bench during a Senate recess or break. In recent years, though, as the country has become more bitterly divided politically, fights over federal judicial nominations have become ferocious, with Democrats refusing to confirm Republican nominees and vice versa. So perhaps it is no surprise that presidents, facing a series of Senate stonewalls, have begun using the recess-appointment power to appoint judges. Clinton put a judge on the federal court of appeals during the Christmas break right before his term ended, and George W. Bush made two such recess appointments—Pryor and the cross-burner-helper guy.
It is also not surprising that these controversial moves have been met with aggressive countertactics, including lawsuits challenging various recess appointments. In the suits challenging Judge Pryor’s appointment, challengers made two main arguments. First, they argued that because the position for which Pryor was appointed did not become vacant while the Senate was in recess, that vacancy did not happen during a recess, and therefore the recess-appointment clause did not give the president the power to appoint someone to fill it. Second, they contended that when the Constitution says “recess,” it means only the recess between sessions of Congress, and not any little break that happens during one of those sessions.
So, what about these arguments? Does the position have to become vacant during the recess for it to have “happened” during the recess? What does the word “happen” mean, anyway? On the one hand, the word does usually mean something like “occur.” When we ask when an accident “happened,” for instance, we want to know when it occurred. On the other hand, it is plausible to read “happen” as meaning something like “exist,” or “be going on,” at least some of the time. One commentator points to things like “a vacation, an illness, a sabbatical, a leave of absence, a war,” which he says “are generally understood to ‘happen’ over an extended period of time.” “You went
to North Dakota for vacation?” we might ask, shocked, upon learning that a colleague had recently spent a week in Bismarck: “When did that happen, and did you at least bring me back a lousy T-shirt?” Okay, fair enough. Still, though, I think the most natural reading of “happen” is more like “occur” then like “exist.”
The problem with this reading of “happen” in the context of the recess-appointments clause, however, is that it leads to results that are inconsistent with the purpose of the clause. If we read “happen” in the clause to mean “occur,” then the president would not be able to fill a vacancy during a recess if the position had become vacant before the recess, no matter how long the recess was supposed to last. If the secretary of state, for instance, quit or died the day before a two-month break, the president would be stuck with running a government without a replacement until the Senate returned. But the whole point of the recess-appointment clause is to give the president the power to fill important positions when the Senate is not around to confirm them. It shouldn’t be surprising, then, that the federal appeals court in the Pryor case joined several other courts which have also held that “happen” basically means “exists” for purposes of the recess-appointments clause, even though that’s not the most natural reading of the word.
The intrasession versus intersession issue is harder, in part because neither interpretation is clearly better given the purpose of the recess-appointments clause. Some have argued that because the main purpose of the clause is to allow the president to appoint officials unilaterally during long Senate absences, and because in the nation’s early years the Senate’s intersession recesses were much longer than its occasional intrasession breaks, therefore the clause should be read only to apply to intersession recesses. The problem, though, is that this length difference disappeared, to a large extent, after the Civil War. Indeed, in the twentieth century, we’ve had plenty of long intrasession breaks, including at least eight that lasted over a month in the last twenty years, not to mention a two-month Reagan-era break and one in 1948 that lasted over four months. Likewise, we’ve had some short intersession recesses, including at least one that lasted exactly zero seconds. In 1903, after a single gavel slam ended one Senate session and began a new one, Teddy Roosevelt appointed 160 officers in what he described as the “constructive” split-second recess that must have existed between the two sessions, an action that the New York Times subsequently lambasted, in an editorial entitled “The Infinitesimal Recess,” as “preposterous.”
What about the text of the clause—does it resolve the issue any better than the purpose? Not really. People who argue that the recess-appointments clause does not apply to intrasession breaks argue that the word “recess” refers to the main break taken during any activity; that the word “the” before “recess” shows that the framers envisioned that only a single recess counted for purposes of the clause; and that short intrasession breaks are referred to elsewhere in the Constitution as “adjournments” rather than “recesses.” On this last point, supporters of the intersession-recesses-only argument point to the adjournment clause of Article I, Section 5, which says that “Neither House, during the Session of Congress, shall, without the Consent of the other, adjourn for more than three days.”
Those on the intrasession-breaks-count-too side have answers for all of these arguments. They respond that “recess” means any short break (think of a judge telling a jury that there will be a “fifteen-minute recess”); that “the” can refer to a state or condition rather than a single occurrence (as one court put it, quoting a dictionary from 1965, the word “the” can be “used to mark a noun as being used generically” as in “the dog is a quadruped”); and that the Constitution makes no neat and clear distinction between “adjournments” and “recesses.” For this last point, the intrasession-breaks-count-too crowd likes to point to the pocket-veto clause of Article I, Section 7, which says that if the president fails to either sign or veto a bill within ten days, it becomes a law, “unless the Congress by their adjournment prevent its return, in which case it shall not be a law.” Since everybody who has ever thought about it for more than two seconds has concluded that the word “adjournment” in this clause must include the intersession recess as well as any intrasession breaks, the argument follows that the terms “adjourn” and “recess” are overlapping terms in the Constitution.
Because the text and purpose of the clause do not resolve the issue, it is no surprise that the executive branch has wavered in how it has interpreted “recess” over the years. In 1901 Attorney General Philander Knox issued an opinion concluding that the president could not make a recess appointment during an intrasession break, finding it “irresistible” that the word “recess” meant “the period after the final adjournment of Congress for the session, and before the next session begins.” But then twenty years later, Attorney General Harry Daugherty wrote an opinion saying that the president can make an intrasession-recess appointment whenever, “in a practical sense,” the Senate is on a long enough break that its consent cannot be obtained. While Daugherty didn’t say exactly how long such a break had to be, he did make it clear that a really short break like two or three or maybe even ten days wouldn’t count. For the next sixty or so years, presidents following this advice made a number of intrasession appointments, but almost never during breaks of less than a month. Then, in a brief filed in 1993, the Justice Department changed course once again and said that the president can make a recess appointment during any Senate break, no matter how short. Subsequently, Presidents Clinton and Bush made lots of recess appointments during extremely short breaks, often within only six days of the Senate’s return to business.
As it turned out, the court of appeals in the Pryor case found the government’s arguments more persuasive, and Judge Pryor got to stay in office. In June of 2005, more than two years after President Bush first put him on the bench, the full Senate confirmed Judge Pryor to a lifetime appointment. Three years later, Pryor wrote the majority opinion in Pelphrey v. Cobb County, a 2–1 decision holding that it was okay for public commissions to start their meetings by praying to specific religious figures like Jesus and Mohammed.
At least at this point in our nation’s history, then, the president can make a recess appointment either between the sessions of the Senate or when the Senate takes a break during a session. What if the Senate doesn’t like the president’s recess appointment though? Is there anything it can do to punish the president or to deter the president from making another controversial recess appointment?
Enter once again Seth Barrett Tillman, the scholar who suggested, back in chapter 1, that a senator could remain in the Senate even after ascending to the office of president. In a recent article, Tillman suggests that although most people have assumed that the Senate is stuck with the recess appointee until the natural end of its next session, in fact nothing in the Constitution prevents the Senate from getting rid of a recess appointee by reconvening, immediately adjourning its session, and then starting a new session. Or, alternatively, if the recess appointment in question occurred during an intrasession break, the Senate could reconvene, adjourn its session, begin a new session, adjourn that session, and then start a new session, all with a couple of swings of the gavel. In support of his so-called Tillman Adjournment, Tillman argues that the procedure would give the president an incentive to make recess appointments who are amenable to the Senate and would ensure that the Senate remains accountable to the people by giving it an active role in deciding whether to keep the president’s recess appointments.
Tillman’s proposal occasioned a response from Brian Kalt, a Michigan State University professor who is one of the country’s preeminent experts on the Constitution’s odd clauses. A few years back, Professor Kalt gained some much-deserved fame for a brilliant article in which he pointed out that as a result of the interaction between a couple of constitutional provisions and a weird statute, there’s a tiny area of land in the Idaho portion of Yellowstone
National Park where the government cannot constitutionally prosecute anyone for committing a crime (note to those who would like to try to harass a black-footed ferret: this may be the place for you). Kalt raises a number of legal and practical concerns with Tillman’s proposal, pointing out, for instance, that there are much easier ways for the Senate to deal with an overreaching president, including using its power over government money to refuse to pay the salary of questionable recess appointments, something the Senate has in fact already done.
Kalt also argues that if the Senate did what Tillman suggests, the president would fight back. Specifically, Kalt says that the president could reappoint his recess appointments in the constructive recess between the Senate’s constructive adjournment and constructive reconvening, and then use his authority under Article II, Section 3, to convene an extraordinary session of the Senate which only the president could adjourn, thus ensuring that his recess appointments would remain in office until he decided to adjourn his special session. I presume that if the president tried such a maneuver, the Senate might respond by saying that it had constructively reconvened before the president had a chance to convene his special session and that therefore his special session never really existed. If this happened, I have no idea what would take place next. Perhaps someone would bring a lawsuit, and an irate Supreme Court would send both the president and the Senate to bed without supper. This, of course, is all quite absurd, but it’s in fact not too different from the kind of standoff that occurred at the end of the Bush administration, when Senate Majority Leader Harry Reid kept the Senate in pretty much continuous pro forma session to keep Bush from making any more recess appointments.
When we as citizens think about how the Constitution should be interpreted, usually it’s in the context of some highly emotionally charged issue on which we already have very strong feelings. It is inevitable that our views on the underlying issue will affect how we think the Constitution should be interpreted. It is hard, for example, to think dispassionately about the proper method of constitutional interpretation when abortion rights or affirmative action or religious freedom are at stake. I would guess that most people who strongly believe that the government should not interfere with the reproductive choices of women are also inclined to argue that the Constitution prohibits the government from making abortion illegal. But there is a big difference between a policy conclusion that the government shouldn’t do something and a constitutional conclusion that the government can’t do something, and it is important to keep these two inquiries separate when thinking about what principles courts should use to interpret the Constitution. After all, principles of constitutional interpretation apply to all cases, not just the one you’re thinking about, and if you conclude that the Supreme Court should read text like the due process clause (no state may “deprive any person of life, liberty, or property, without due process of law”) very broadly to stop the government from prohibiting abortion, then in the next case some judge who doesn’t share your policy views might use your “read the text broadly” principle to conclude that the takings clause of the Fifth Amendment (no state shall “take private property” without “just compensation”) prohibits the government from passing environmental leg islation that will interfere with the rights of private property owners.