The D.C. Circuit’s recent decision regarding the Recess Appointments Clause (Noel Canning v. National Labor Relations Board) bills itself as an exercise in Heller-style textualism: “When interpreting a constitutional provision, we must look at the natural meaning of the text as it would have been understood at the time of the ratification of the Constitution. District of Columbia v. Heller, 128 S. Ct. 2783, 2788 (2008).” As a result, much of the decision is devoted to linguistic issues.
I’m going to take a look at how the court handled some of those issues—at the conclusion it reached and the reasoning it used to get there.
The verdict: the Recess Appointments Clause is a lot less clear than the D.C. Circuit makes it out to be, and the court’s reasoning isn’t very good.
Ordinarily, the president’s power to appoint high-level officials is subject to the requirement that his choices be confirmed by the Senate. But because the Senate isn’t always open for business, the Constitution provides that the president “shall have the 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.” One of the questions raised by this provision is what exactly is meant by the phrase the Recess of the Senate. That’s the issue I will discuss in this post.
To put the question into context, here’s some quick background. Every two years, when the most recently-elected members of the House of Representatives start their terms, a new term of Congress begins. Each new term is referred to as a separate Congress; the current Congress is the 113th. Since the Constitution requires Congress to meet at least once a year, each two-year Congress is divided into at least two officially-designated sessions. Currently, the 113th Congress is in its first session.
In between its formally-designated sessions, Congress is in recess. Everyone agrees that these breaks count as “recesses” for purposes of the Recess Appointments Clause. Congress also takes breaks periodically during each officially-designated session. During those breaks, Congress is simultaneously in session and not in session. It is in session because the officially-designated session is still in progress, but it is out of session because it has temporarily stopped conducting business.
The question is whether these breaks within an officially-designated session constitute “recesses” such that the president can exercise his recess-appointment power. According to the D.C. Circuit, the answer is no.
The court’s reasoning
Here is a summary of the linguistic reasons that the D.C. Circuit gave for its decision.
1. The Constitution uses the phrase the Recess, not a Recess. The word the is “‘[an] article noting a particular thing’” (quoting Johnson’s dictionary; emphasis by the court). “Unlike ‘a’ or ‘an,’ that definite article suggests specificity. As a matter of cold, unadorned logic, it makes no sense to adopt the [NLRB’s] proposition that when the Framers said ‘the Recess,’ what they really meant was ‘a recess.’
2. The noun in the Recess is singular.
3. The Constitution uses the phrase the Recess twice, whereas it uses the verb adjourn or the noun adjournment six times “to refer to breaks in the proceedings of one or both Houses of Congress.” Moreover, none of the uses of adjournment were preceded by the definite article. “All this points to the inescapable conclusion that the Framers intended something specific by the term ‘the Recess,’ and that it was something different than a generic break in proceedings.”
4. The Recess Appointments Clause provides that appointments made during “the Recess” will expire “at the End of [the Senate’s] next Session.” This sets up a dichotomy between “the Recess” and the “Session”. “It is universally accepted that ‘Session’ here refers to the usually two or sometimes three [officially-designated] sessions per Congress. therefore, ‘the Recess’ should be taken to mean only times when the Senate is not in one of those Sessions.”
In the remainder of this post, I’ll discuss, in turn, the D.C. Circuit’s reliance on the use of the word the, its interpretation of the phrase the Recess, and its discussion of the Constitution’s use of recess on the one hand and adjourn and adjournment on the other.
The first problem with the court’s analysis—and it’s a big one—is its reliance on what it thinks is the meaning of the.
As I’ve said, the court relied on Samuel Johnson’s Dictionary of the English Language, which defined the as an “[t]he article noting a particular thing.” While the D.C. Circuit treats that definition as authoritative, Johnson himself was under no such illusions. “Dictionaries are like watches, he once wrote, “the worst is better than none, and the best cannot be expected to go quite true.”
For present purposes, the major problem with Johnson’s definition of the is that it’s too narrow. While the can of course be used to mark the noun as referring to “a particular thing” (e.g., I want you to meet the guy I told you about yesterday), it can also be used in other ways. For example, it can be used with nouns denoting types of things, as in The sun comes up in the morning.
Also worth noting, though it’s less directly relevant here, is that the definition doesn’t give any basis for distinguishing the definite article the from the indefinite article a, which can also be used in referring to a particular thing (I met a guy yesterday who I want you to meet). In fact, both articles can be used as parts of noun phrases that refer to the exact same particular thing:
I met a guy yesterday when I was downtown. I talked with him for a little while and I finally realized that he was the guy you told me about last week.
Even apart from the flaws in Johnson’s definition, the definition doesn’t really support the D.C. Circuit’s interpretation of the Recess Appointments Clause. Specifically, even under the D.C. Circuit’s interpretation, the the in the Recess doesn’t mark the phrase as referring to “a particular thing.”
The D.C. Circuit did not interpret the Recess as referring to a particular recess (e.g., the one between the first and second sessions of the 101st Congress). Rather, the court interpreted it as referring to recesses generally (with recesses meaning the breaks between formally designated sessions). Thus, under the D.C. Circuit’s interpretation, the president has the power to make recess appointments every time such a recess occurs.
This suggests that the D.C. Circuit was wrong to rely on the fact that the Recess Appointments Clause says the Recess rather than Recesses. While the Recess is grammatically singular, it refers to indefinitely many separate recesses. During each two-year election cycle, there are at least two recesses (under the court’s interpretation): one between the first and second sessions of that Congress, and one between the last session of that Congress and the first session of the next one. Since the 113th Congress just began, there have so far been at least 224 breaks that count as recesses under the D.C. Circuit’s interpretation. (I say “at least” 224 because during some Congresses there were three formally designated sessions.)
In contrast to the D.C. Circuit, the U.S. Court of Appeals for the Eleventh Circuit had, in a 2004 decision regarding the Recess Appointments Clause, recognized the reality that definite noun phrases can refer to types of things as well as particular things:
We do not agree that the Framers’ use of the term “the” unambiguously points to the single recess that comes at the end of a Session. Instead, we accept that “the Recess,” originally and through today, could just as properly refer generically to any one—intrasession or intersession—of the Senate’s acts of recessing, that is, taking a break. See The Random House Dictionary of the English Language 1965 (2d ed.1987) (1966) (defining “the” as “used to mark a noun as being used generically: the dog is a quadruped”); 17 The Oxford English Dictionary, 879 (2d ed.1989) (1928) (defining “the” as “referring to a term used generically or universally” and providing examples of such usage from the 18th Century).
The D.C. Circuit criticized that interpretation for “relying on twentieth-century dictionaries” rather than following Samuel Johnson’s treatment of the “as ‘noting a particular thing’” (emphasis added by the court).
That makes no sense. Sure, the meanings of words can change over time, and a modern dictionary might not accurately reflect how a word was used more than two hundred years ago. But lexicographers have learned a thing or two about language since Samuel Johnson’s day. And the Oxford English Dictionary, which the D.C. Circuit stuck up its nose at, is a historical dictionary. Its purpose is not simply to reflect current usage, but to trace the meanings of words beginning with their earliest uses. When the OED says that the has been used in noun phrases that refer to types of things (i.e., that have “generic reference”), it backs that statement up with nine quotations predating the Constitution, the earliest one dating back a thousand years.
And if that’s not good enough for you, consider this: The Constitution itself includes definite noun phrases that have generic (or at least nonspecific) reference:
The Senate shall chuse their other Officers, and also a President pro tempore, in the Absence of the Vice President, or when he shall exercise the Office of President of the United States.
[The President] shall from time to time give to the Congress Information of the State of the Union, and recommend to their Consideration such Measures as he shall judge necessary and expedient; he may, on extraordinary Occasions, convene both Houses, or either of them, and in Case of Disagreement between them, with Respect to the Time of Adjournment, he may adjourn them to such Time as he shall think proper….
The Trial of all Crimes, except in Cases of Impeachment, shall be by Jury….
Congress shall make no law…abridging the freedom of speech, or of the press…
…nor shall any person be subject for the same offence to be twice put in jeopardy of life or limb….
And at the risk of piling on, here’s a nonlegal example from 1771 that I can’t resist:
We may likewise observe, that the lion is not a cruel animal; he kills rather from necessity than choice, never destroying more than he eats, and whenever his appetite is satisfied he is mild and peaceable….All the passions of the lion, the soft passion of love not excepted, are excessive; the love of offspring is extreme: the lionness is naturally weaker, less bold, and more gentle than the lion; but she becomes perfectly rapacious and terrible when she has young….The roaring of the lion is so strong and loud, that it resembles the rumbling of distant thunder. [Source.]
So to sum up, the D.C. Circuit got the meaning of the wrong.
While the D.C. Circuit followed Johnson’s Dictionary regarding the meaning of the, it didn’t follow Johnson regarding the meaning of recess, which Johnson defines as “remission and suspension of any procedure.” The Court explained, “[T]he natural meaning of ‘the Recess’ is more limited than the broad dictionary definition of ‘recess.’ In context, ‘the Recess’ refers to a specific state of the legislature, so sources other than general dictionaries are more helpful in elucidating the term’s original public meaning.”
I have no problem at all with going beyond dictionary definitions; courts give more weight to dictionaries than they should. But the D.C. Circuit overlooked evidence in the Constitution and in 18th-century usage more generally that doesn’t support its conclusion.
Let’s start with the phrase during the Recess of the Senate. I don’t know about you, but to me that phrase has a strange ring to it. There’s something about it that makes it sound weird—something beyond the fact that the Constitution is written in a somewhat archaic style. This makes me wonder whether the word recess had shades of meaning in 1787 that it subsequently lost, so that relying on our 21st-century intuitions might not be the best idea.
This conjecture is supported by the fact that the Constitution also contains a similar (and semantically related) construction that sounds strange in precisely the same way as during the Recess of the Senate:
Neither House, during the Session of Congress, shall, without the Consent of the other, adjourn for more than three days, nor to any other Place than that in which the two Houses shall be sitting.
Now consider a third construction, which follows the same grammatical pattern as the previous two, but that sounds perfectly fine to me:
The Senate shall chuse their other Officers, and also a President pro tempore, in the Absence of the Vice President, or when he shall exercise the Office of President of the United States.
The phrase the absence of the Vice President can be paraphrased as the state of the Vice President being absent. Maybe the semantics of the constructions using Recess and Session are similar. Maybe the Recess of the Senate and the Session of Congress could have been understood in 1787 as meaning the state of the Senate being in recess and the State of Congress being in session. If so, that wouldn’t resolve the question of what the Recess Appointments Clause means, but I do think that it would weaken the argument in favor of the D.C. Circuit’s interpretation.
This sense of recess doesn’t appear in the OED, which obviously doesn’t help my hypothesis, but I don’t think it necessarily rules the hypothesis out, either. To test the hypothesis, I’ve looked for other uses of the recess from the late 18th century. While I haven’t found anything that provides a clear answer, I have found usages that are at least consistent with the hypothesis.
The first one is in the Constitution itself. The Constitution originally provided that each state’s senators would be chosen by the state’s legislature; the issue of vacancies in the Senate was dealt with in the Senate Vacancies Clause:
[I]f Vacancies happen by Resignation, or otherwise, during the Recess of the Legislature of any State, the Executive thereof may make temporary Appointments until the next Meeting of the Legislature, which shall then fill such Vacancies.
Whereas the D.C. Circuit read the Recess Appointments Clause as setting out a dichotomy between the Recess and the Session (and understood session to mean ‘formally designated session’), the Senate Vacancies Clause distinguishes between the Recess of the Legislature and the next Meeting of the Legislature. The latter dichotomy arguably invites the interpretation that the Recess refers to any period when the legislature is not meeting.
To go further than that in trying to understand what the Recess of the Legislature means, it seems to me that it’s necessary to know more about the functioning of each state’s legislature and about each legislature’s practices regarding what kind of breaks counted as recesses. Just as the D.C. Circuit’s interpretation of the Recess Appointments Clause was informed by the court’s knowledge of how Congress operates, the interpretation of the similar language in the Senate Vacancies Clause has to be informed by knowledge about how the state legislatures operated. I certainly don’t have the necessary knowledge, and I doubt that judges in Noel Canning do, either.
Continuing our look at how the recess was used in the late-18th century, here these examples are from the 20 years or so before the Constitution was ratified
1765—From the Journals of the Massachusetts House of Representatives
The Answer of the honorable Board to the Remonstrance of the House, gives us a very sensible Concern, as it contains not only a Justification of the measures complained of, but is in Effect a Declaration of the honorable Board, that in the Recess of the General Court all the Powers of Government are devolved on, and vested in the Governor and Counsil [sic], at their Discretion to be exercised and executed. [Source]
There is nothing more certain than that the Governor and Council of this Province, in the Recess of the General Court, can rightfully claim only Executive Powers[.] [Source]
1777—Order of the Pennsylvania War-Office
And whereas by the recess of the Legislature of this State, no law, however necessary, can at this time be enacted…. [Source]
1780—Constitution of Massachusetts
The governor, with advice of council, shall have full power and authority, during the session of the general court to adjourn or prorogue the same to any time the two houses shall desire; and to dissolve the same on the day next preceding the last Wednesday in May; and, in the recess of the said court, to prorogue the same from time to time, not exceeding ninety days in any one recess; and to call it together sooner than the time to which it may be adjourned or prorogued, if the welfare of the commonwealth shall require the same…. [Source]
Like the Senate Vacancies Clause, these examples seem consistent with the hypothesis that the recess of the Senate could have meant ‘the state of the Senate being in recess.’ And as with the Senate Vacancies clause, we can’t really be sure how recess is used in these passages without knowing more about the functioning of the legislatures in question.
Finally, there’s an example from the Federalist Papers. The Federalist Papers, in case you skipped civics class the day they were covered, are essays that were published in 1787–88 by Alexander Hamilton, James Madison, and John Jay, urging that the Constitution be ratified. Our example comes from the Federalist No. 18 (attributed to Hamilton and Madison, writing as “Publius”), which discussed the “insufficiency of [the Articles of Confederation] to preserve the union.” Hamilton and Madison made their case by comparing the United States under the Articles of Confederation to two ill-fated confederations of republics in ancient Greece. In describing the government of one of the Greek confederations, Hamilton and Madison referred to the chief executive as “administer[ing] the government in the recess of the senate[.]”
Once again: This usage is consistent with my hypothesis about the Recess of the Senate, but without further historical information, it’s impossible to be sure of how the recess is being used. And presumably the necessary historical information is presumably even less accessible (by far) than the information about state legislatures in 18th century.
Where does that leave us? These examples obviously have the potential to shed light on the meaning of the Recess Appointments Clause, but that potential can’t be realized without additional historical information. Without that information, these examples are a source of uncertainty rather than enlightenment.
“Recess” versus “adjournment”
The D.C. Circuit relied on differences in how the Constitution uses the words recess on the one hand and adjourn and adjournment on the other:
Six times the Constitution uses some form of the verb “adjourn” or the noun “adjournment” to refer to breaks in the proceedings of one or both Houses of Congress. Twice, it uses the term “the Recess”: once in the Recess Appointments Clause and once in the Senate Vacancies Clause, U.S. Const. art. I, § 3, cl. 2. Not only did the Framers use a different word, but none of the “adjournment” usages is preceded by the definite article. All this points to the inescapable conclusion that the Framers intended something specific by the term “the Recess,” and that it was something different than a generic break in proceedings.
Although the Court isn’t fully explicit in its reasoning, I think that this argument assumes that the use in the Constitution of the words adjourn and adjournment represented a choice to use those words rather than recess. And that assumption, in turn, relies on what I’ll call the Assumption of Substitutability—the assumption that in the English language of 1787, the word recess could have been more or less freely substituted for adjourn or adjournment in the provisions the court refers to. For if recess was not used as a verb in the late-18th century, the Constitution’s use of adjourn didn’t represent a choice of one near-synonym over another.
I think that the Assumption of Substitutability is unjustified and therefore that the Court’s argument based on it is unjustified as well.
The starting point is to look at the six uses of adjourn or adjournment that the court refers to; they are set out below. As you read each one, consider whether it deals with (a) the act of adjourning or (b) the period during which Congress is adjourned.
 [A] Majority of each [house of Congress] shall constitute a Quorum to do Business; but a smaller Number may adjourn from day to day….
 Neither House, during the Session of Congress, shall, without the Consent of the other, adjourn for more than three days, nor to any other Place than that in which the two Houses shall be sitting.
 If any Bill shall not be returned by the President within ten Days (Sundays excepted) after it shall have been presented to him, the Same shall be a Law, in like Manner as if he had signed it, unless the Congress by their Adjournment prevent its Return, in which Case it shall not be a Law.
 Every Order, Resolution, or Vote to which the Concurrence of the Senate and House of Representatives may be necessary (except on a question of Adjournment) shall be presented to the President of the United States…
 [The President] may, on extraordinary Occasions, convene both Houses, or either of them, and in Case of Disagreement between them, with Respect to the Time of Adjournment, he may adjourn them to such Time as he shall think proper….
It seems clear to me that in provisions 1, 4, and 5, the instances of adjourn or adjournment denote the act of adjourning and not the ensuing period of being closed for business. In provisions 2 and 3, they also denote the act of adjourning, but might be understood as also encompassing the closed-for-business period.
Thus, for the Assumption of Substitutability to be valid, the vocabulary of the English language in 1787 had to make the word recess available to denote the act of interrupting legislative proceedings. To what extent was that the case?
In the case of recess used as a verb, the answer seems to be, very little, if at all. The earliest citation in the OED for the use of recess (in the relevant sense) as a verb is from 1844. That doesn’t necessarily mean that the word wasn’t used as a verb before then, but it’s a data point. And it’s supported by other data points.
In an effort to compare the relative frequencies of the use of recess and adjourn as verbs during the 35+ years before the Constitution was ratified, I ran some searches using two of the corpus-based resources that are available on the Web: the Corpus of Historical American English (COHA) and the Google Books Ngram Viewer. The results of those searches are consistent with what the OED suggests: in the late 18th century, recess was rarely used as a verb.
COHA’s coverage only goes back to 1810, so it can’t provide any direct evidence about usage patterns at the time the Constitution was drafted and ratified. However, usage early in the 19th century is relevant, I think. The earliest use in COHA of the words recess, recesses, recessed, or recessing in the legislative context was in 1925, and in COHA’s first century of coverage, there were only nine verbal uses of the word in any of its senses. (Search results here.) In contrast, the various forms of adjourn (adjourn, adjourns, adjourned, adjourning) were used as a verb 43 times in the period 1810–1829, and even more frequently in each subsequent decade. (Search results here.)
My Google Ngram results (which covered the period 1750–1788) were similar: recess in its various forms was used as a verb much less commonly than was adjourn, and in some if not all of those uses the word may have been used in an entirely different sense (e.g., ‘To make a recess or recesses in (a structure or object); to cut into, so as to form a recess.’ or ‘To set (a part of a wall, or other structure) in a recess.’ (Definitions from the OED)). (Search results: all English; American English.)
Further evidence that recess was not used as a verb when the Constitution was adopted (or at least was not often used as a verb) comes from examples like these, in which the verb adjourn is used to denote the action of putting Congress into a “recess”:
1775—Journal of the Continental Congress
On motion Resolved, That when the Congress adjourns for recess, it be adjourned to meet at Philadelphia. (Source.)
1784—Journal of the U.S. Congress (under the Articles of Confederation)
A motion was made by Mr. Howell, seconded by Mr. Gerry, That the president be, and he hereby is authorized and directed, to adjourn Congress on the third day of June next, to meet on the thirtieth of October next at Trenton, for the dispatch of public business; and that a committee of the states shall be appointed to sit in the recess of Congress. (Source.)
1789—Proceedings of the House of Representatives
Mr. Benson introduced a resolution to this purport: that a committee be appointed, to join a committee of the senate, to consider and report when it will be convenient for congress to adjourn ; also, to report what business now before congress must necessarily be attended to previous to a recess, and what will be proper to postpone till the next session—laid on the table. (Source.)
These examples are cases of dogs that didn’t bark. If it had been common to use recess as a verb meaning ‘to go into recess,’ the statements above would have provided obvious opportunities to use the word that way.
While this evidence doesn’t definitively answer the question, I think it provides significant reason to doubt the Assumption of Substitutability with respect to the use of recess as a verb. At a minimum, it puts the burden on defenders of the D.C. Circuit’s opinion to provide some evidence supporting a different conclusion.
Now let’s look at the Constitution’s use of Adjournment to denote the act of adjourning. To what extent was recess used as a noun to similarly denote the act of adjourning for a recess?
The OED has no entry that the word as ever been used that way (whereas it does have an entry for adjournment as a noun meaning “The action or an act of adjourning”). That’s not conclusive evidence against the Assumption of Substitutability, but it obviously provides no evidence in favor of the assumption, either. I’d like to tell you that I’ve found evidence one way or the other in the corpus resources I’ve been discussing. Unfortunately, I can’t think of a search strategy that would produce such evidence, short of brute force—looking at lots and lots of uses of recess.
But let me offer this speculation. If it’s true that recess was seldom if ever used as a verb in the late 18th century, one wouldn’t expect that recess would be used as a noun denote the act of going into recess—especially given that adjournment was available for that purpose. I can’t point to anything backing that up, but it makes sense to me. In any event, there is as far as I can tell no evidence that one could point to in support of the Assumption of Substitutability as it relates to the word adjournment.
The bottom line is that the D.C. Circuit puts more weight on the distinction between the words recess and adjourn(ment) than the usage evidence justifies.