Corpora and the Second Amendment: “bear arms” (part 1), plus a look at “the people”

« previous post | next post »

An introduction and guide to my series of posts “Corpora and the Second Amendment” is available here. The corpus data that is discussed can be downloaded here. That link will take you to a shared folder in Dropbox. Important: Use the "Download" button at the top right of the screen.

New URL for COFEA and COEME:

This is the first of what will be three posts on bear arms; it will be devoted to critiquing the Supreme Court’s discussion of bear arms in District of Columbia v. Heller. My examination of the corpus data on bear arms will appear in my next two posts. In the meantime, if you're interested, you can read discussions of the data by Dennis Baron (“Corpus Evidence Illuminates the Meaning of Bear Arms,” in the Hastings Constitutional Law Quarterly) and by Josh Blackman & James C. Phillips (“Corpus Linguistics and the Second Amendment,” in the Harvard Law Review Blog), both of which reach conclusions consistent with mine. (The piece by Blackman & Phillips is especially noteworthy, given that they are both gun-rights advocates.)

My focus in this post will be on the Supreme Court’s conclusion that at the time the Second Amendment was proposed and ratified, bear arms unambiguously meant ‘carry weapons, for purposes of being prepared for a confrontation,’ without regard to whether the carrying was in connection with military service. What I conclude is that even without taking account of how bear arms was actually used, the court’s arguments don’t hold up. Assuming for the sake of argument that bear arms could reasonably have been understood to mean what the court said it meant, the court didn’t show that it unambiguously meant that.

That’s not to say that I think bear arms was ambiguous. As I’ll discuss in the next two posts, the corpus evidence points toward the conclusion that bear arms unambiguously conveyed the military meaning that the Supreme Court rejected: “to serve as a soldier, do military service, fight” or “to wage war.” But even if the evidence were equivocal, the absence of evidence unambiguously supporting the court’s interpretation would still be important.

That’s because the court’s analysis in Heller depends crucially on its conclusion that bear arms was unambiguous. It was that conclusion that enabled the court to interpret the Second Amendment’s operative clause (“the right of the people to keep and bear Arms, shall not be infringed”) without taking into consideration its prefatory clause (“A well regulated Militia, being necessary to the security of a free State”). In the court’s view, if the operative clause was unambiguous, the prefatory clause “does not limit or expand [its] scope.” So if court was wrong in thinking that the operative clause was unambiguous, it was wrong in refusing to consider whether the prefatory clause affected its meaning. And if the prefatory clause plays a role in interpreting the operative clause, the argument against the court’s interpretation is strengthened.

THE COURT IN HELLER started from the premise that in the late 18th century, bear meant ‘carry’ and arms meant ‘weapons.’ That was a reasonable view in 2008, when Heller was decided, given what was generally known about the relevant linguistic history. But in the decade since then, information has become available showing that the reality was more complicated. Those complications render Heller’s confident generalization about bear and arms untenable.

The newly-available information is discussed at length in my posts on bear and arms, so I’ll limit myself here to a very brief summary.

By 1789, when the Bill of Rights was approved by Congress and sent to the states for ratification, bear was no longer the verb that was generally used to denote the carrying of physical objects; that function had been taken over by carry. The effect of that change can be seen in the corpus data for the late 18th century, which reflects substantial differences in how the two words were used, with bear seldom being used in the way that the court in Heller thought. The data also shows that arms was (in addition to being used to mean ‘weapons) often used to convey a variety of figurative meanings having to do with warfare and the military.

While the court in Heller can’t be faulted for being unaware of information that wasn’t generally available, the fact is that its understanding of how bear and arms were used was inaccurate. And the court’s mistaken understanding on this score may was only one of a series of problems with its textual analysis.

The next problem arose when, after defining bear as ‘carry’ and arms as ‘weapons,’ the court immediately said that bear arms means something more than just ‘carry weapons’: “When used with ‘arms,’” the court said, “[bear] has a meaning that refers to carrying for a particular purpose—confrontation.” That, according to the court, was the phrase’s “natural meaning.”

That conclusion raises two questions. First, what is “natural meaning”? And second, how did  the court in Heller arrive at its conclusion about what bear arms’s natural meaning was?

To answer the first question: “Natural meaning” is a less common phrase in legal interpretation than is “ordinary meaning,” but the two phrases mean the same thing. In several cases in the 1990s, the Supreme Court  treated the two phrases as synonymous, as did. Justice Scalia, who wrote the opinion in Heller. During the same period, the court equated the “ordinary meaning” of a word or phrase with what it “ordinarily means,” and again, so did Scalia. (Citations for these propositions are provided at the end of the post.) Therefore, it’s reasonable to conclude that when the court in Heller adopted the Muscarello dissent’s formulation as a statement of the natural meaning of bear arms, it meant that the phrase ordinarily conveyed that meaning.

That brings us to the question of what the court relied on in interpreting bear arms. The answer is rather surprising, because the interpretation didn’t come from a dictionary definition of the phrase or from evidence of its actual use. Rather, the court relied on the dissenting opinion by Justice Ginsburg in Muscarello v. United States—a 1998 case that didn’t present the question of what bear arms means.

Muscarello raised the separate question of what carry a firearm means, and it was that issue that the dissent focused on. (Specifically, the  issue was whether driving somewhere with a gun in the trunk or glove compartment counts as “carrying” a firearm.)

Although the meaning of bear arms wasn’t at issue in Muscarello, the dissent invoked the Second Amendment:

Surely a most familiar meaning [of carry a firearm] is, as the Constitution’s Second Amendment (keep and bear Arms) and Black’s Law Dictionary . . .  indicate: “wear, bear, or carry . . . upon the person or in the clothing or in a pocket, for the purpose . . . of being armed and ready for offensive or defensive action in a case of conflict with another person.”

Other than the offhanded reference to the Second Amendment, this passage says nothing about the meaning of bear arms, much less what that phrase had meant in 1789. Similarly, the definition from Black’s Law Dictionary (6th ed. 1990), which the dissent quoted, was from the entry for carry arms or weapons, not for bear arms. That definition can’t be read as applying to bear arms, because bear arms had a separate entry of its own (albeit one that didn’t actually offer a definition, instead saying only, “The Second Amendment, U.S. Constitution, provides that the ‘right of the people to bear arms, shall not be infringed.’ This right has been restricted however by state and federal laws regulating the transportation, sale, use, and possession of weapons.”).

Nevertheless, the court in Heller declared, without explanation, that the passage I’ve quoted “accurately captured the natural meaning of ‘bear arms.’” The court added that bear arms “in no way connotes participation in a structured military organization”—a conclusion for which it cited no support independent of its extension of the Muscarello dissent’s interpretation to bear arms.

THE COURT WAS up to this point talking about the meaning of bear arms as of 2008, but it went on to say that bear arms had meant the same thing more than 200 years earlier: “From our review of founding-era sources, we conclude that this natural meaning was also the meaning that ‘bear arms’ had in the 18th century. In numerous instances, ‘bear arms’ was unambiguously used to refer to the carrying of weapons outside of an organized militia.” (Although that statement didn’t explicitly refer to the Muscarello dissent’s definition as the bear arms’s 18th-century“natural meaning,” it made that connection later in the decision, describing the idiomatic military sense of bear arms as being “significantly different from its natural meaning,” and saying that during the founding era bear arms “normally meant…what Justice Ginsburg’s opinion in Muscarello said.”)

What were the founding-era sources that the court relied on in applying the interpretation from the Muscarello dissent? They were provisions from nine state constitutions that were adopted during the period 1776 through 1820. The court described these provisions as protecting “a right of citizens to ‘bear arms in defence of themselves and the state’ or ‘bear arms in defence of himself and the state.’” According to the court, these provisions showed unambiguously “that ‘bear arms’ did not refer only to carrying a weapon in an organized military unit.”

The analysis that I’ve just described is flawed in several ways. The first can be summed up by a statement from Justice Scalia’s famous dissent in Smith v. United States, which Justice Stevens’s dissent in Heller quoted back to its author: “The Court does not appear to grasp the distinction between how a word can be used and how it ordinarily is used.” (Scalia’s emphasis.)

The second flaw in the court’s reliance on the state constitutional provisions is that the each of the provisions used bear arms in a context different from the context in which it appeared in the Second Amendment. In each of the provisions cited by the court, bear arms was modified by the prepositional phrase in defence/defense of themselves/himself, while in the Second Amendment bear arms is unmodified. That difference is important because when a verb phrase is modified, the phrase’s meaning in that context won’t necessarily be the same as its meaning without the modification. (I realize that that might sound counterintuitive, and if it does, I’d suggest you read my post “The semantics of sleeping in railway stations,” which shows that a prohibition on “sleeping in a railway station” isn’t necessarily violated by being asleep in a railway station.)

The upshot is that what bear arms means in a context in which the phrase is modified isn’t necessarily representative of what it means when it is unmodified. Therefore, the state constitutional provisions that the court relied on don’t, by themselves, provide an adequate basis for making a determination of how bear arms was ordinarily used.

Yet another flaw in Heller’s analysis arises from the fact that the court was mistaken in thinking that all of the state provisions it cited were unambiguous. As you’ll see, several of the provisions were ambiguous, and those were the ones that were the most directly relevant to the Second Amendment. In contrast, the provisions that weren’t ambiguous differed in important ways from the Second Amendment.

The provisions in questions were clauses in nine state constitutions protecting the right to bear arms.  Of those nine provisions, four of them shared the same phrasing, and three shared a different phrasing. The various formulations are set out below, grouped into three categories based on relevant similarities and differences in their wording. My focus here will be on categories 1 and 2 (category 3 being intermediate between those two).

  1. the people’s right . . . defence of themselves and the people have a right . . . defence of themselves:

“the people have a right to bear arms for the defence of themselves and the state” [four states; 1776, 1777, 1802, 1816]

“their [the people’s] right to bear arms in defence of themselves and of the State cannot be questioned” [one state; 1820]

  1. every citizen has a right . . . defense of himself:

“Every citizen  has a right to bear arms, in defense of himself and the state” [three states; adopted]

  1. the right of citizens . . . defence of themselves:

“the right of citizens to bear arms in defense of themselves and the State shall not be questioned” [one state; 1802]

The key language in these provisions is “defence/defense of themselves/himself”; that’s presumably what the court relied on in saying that these provisions were unambiguous in referring to the carrying of arms other than in a militia. I think that it’s reasonable to assume that the court saw a difference between the rights-holders’ bearing arms in defense of the state, which would have involved collective action as part of a militia, and bearing arms  in defense of “themselves” or of “himself,” which (in the court’s view) would not have involved participation in a militia.

That line of reasoning is valid as to the provisions in category 2 (every citizen has a right . . . defense of himself). The language of those provisions makes it clear that the right of self-defense that is protected belongs to each citizen individually, rather than belonging collectively to the citizenry as a whole. Such a reading is referred to in linguistics as a distributive reading, and it contrasts with a collective reading, in which the exercise of the right would involve collective action.

The fact that the provisions in category 2 are unambiguously distributive follows from the use of the grammatically singular phrase every citizen (The Cambridge Grammar of the English Language notes that every and each “explicitly indicate that the predication property applies distributively”), and it’s reinforced by the use of the singular pronoun himself, which refers to each citizen individually.

Although the court was right to read the provisions in category 2 distributively, those provisions differ from the Second Amendment in that they identify the holders of the right to bear arms as “every citizen,” while under the Second Amendment the right is held by “the People.” Of the provisions cited by the court, only those in category 1 do the same. (In categories 2 and 3 it’s “every citizen” and “citizens,” respectively.)

That difference matters, because the category-1 provisions are ambiguous between distributive and collective readings. In fact,it seems likely that these provisions were understood collectively (and as being associated with militia service) at least with respect to the right to bear arms in defense of the state. Defending the state is by its nature a collective endeavor, and it was one of the militia’s primary purposes.

Things are not as clear as to the people’s right to bear arms “in defense of themselves,” because such a right is by its nature more amenable to being understood distributively than is a right to bear arms in defense of the state. It could also be argued that if the people’s right to bear arms in defense of themselves is understood as having been collective, it wouldn’t add anything to the right to bear arms in defense of the state. But I’m not so sure that we can rule out the possibility that if there was thought to be a right of collective self-defense (or a collective right of self-defense), it was thought to extend to threats such as conflicts with Native Americans, which could have endangered one or more communities but not the state as a whole. And if there was thought to be such a right, it seems reasonable to think that it would have been associated with militia service.

Considering all of this, I think the court in Heller was too quick to assume that the state provisions in category 1 were unambiguous. (I say “assume” rather than “conclude” because there’s no indication that the court realized that this was an issue.

Although I think that so far the argument in favor of a collective reading is plausible, it’s not necessarily compelling. But there is an additional factor I haven’t mentioned yet, and it strengthens the argument. That factor is corpus data—in particular, corpus data for the relevant state-constitutional language: the people’s right and  the people have a right.

That data is dominated by uses that I interpret as being collective. Those uses represented more than half of the combined uses of the two phrases—63 out of 104 concordance lines (not counting one that I couldn’t categorize). Those uses involved rights such as these:

the right to say who shall be their ruler
the right by a convention, or otherwise, to change the existing government
the right to appoint such officers as they might think necessary
the right to alter the Constitution
the right to alter the government
the right to consult for the common good
the right to depose a bad king and set up a good one
the right to set up a civil government

Of the remaining 41 uses, I categorized 24 as being both collective and at least arguably distributive, and only 17 that I categorized as being purely distributive. Thus, times purely collective were almost four times a frequent as uses that were purely distributive.

I also looked at the data for the right of the people—the phrasing used in the Second Amendment—and it followed a similar pattern, but with a higher proportion of purely collective uses and a lower proportion of purely distributive uses. (For details about both searches and the compilation of the data, see the spreadsheets in which the data is set out.)

While the data isn’t conclusive, it lends weight to the argument that the right of self-defense protected by the state constitutional provisions in category 1 should be understood as a right of collective self-defense, and therefore as a right associated with service in a militia. That suggests that at a minimum, the Supreme Court was mistaken in saying that the right protected by those provisions was unambiguous. And the ambiguity of the provisions in category 2 doesn’t count for much in interpreting the Second Amendment, because those provisions didn’t refer to the right to bear arms as being held by “the people.”

UP TO THIS POINT, I’ve been examining the Supreme Court’s analysis as to the sense of bear arms that it regarded as the phrase’s natural (= ordinary) meaning. I’ll turn now to the court’s discussion of the argument that the ordinary meaning of bear arms was actually its military sense:

The phrase “bear Arms” also had at the time of the founding an idiomatic meaning that was significantly different from its natural meaning: “to serve as a soldier, do military service, fight” or “to wage war.” See Linguists’ Brief 18; post, at 11 (Stevens, J., dissenting). But it unequivocally bore that idiomatic meaning only when followed by the preposition “against,” which was in turn followed by the target of the hostilities. See 2 Oxford [English Dictionary] 21. (That is how, for example, our Declaration of Independence ¶ 28, used the phrase: “He has constrained our fellow Citizens taken Captive on the high Seas to bear Arms against their Country … .”) Every example given by petitioners’ amici for the idiomatic meaning of “bear arms” from the founding period either includes the preposition “against” or is not clearly idiomatic. See Linguists’ Brief 18–23. Without the preposition, “bear arms” normally meant (as it continues to mean today) what Justice Ginsburg’s opinion in Muscarello said.

Like Heller’s discussion of the natural meaning of bear arms, the passage above can be criticized on more than one basis. The first and most obvious of these is that the court offers no evidence or explanation to back up the conclusion announced in the  final sentence. The court simply invokes its extension to bear arms of the definition of carry a firearm from the dissent in Muscarello, and that invocation brings with it all the logical and empirical flaws that I’ve previously discussed. As my next two posts will show, the court’s conclusions can’t be squared in the face of the what the corpus data shows

The other major problem with this portion of Heller is that it was to a large extent an exercise in “Heads I win, tails you lose.”

When the court had described the state constitutional provisions that are discussed above as unambiguously protecting the carrying of weapons outside of an organized militia,” it implicitly relied on the fact that the provisions’ key language was bear arms in defense of themselves [or himself] and the state. Without those prepositional phrases, it wouldn’t have been possible for the court to argue that the provisions unambiguously used bear arms in a nonmilitary sense. Because no such modifying phrase appears in the Second Amendment, the decision presupposes that the presence of the prepositional phrase in the state provisions didn’t render them irrelevant.

Yet, as we’ve seen, when it came to dealing with bear arms against X, the court took a different tack. As to that issue, the presence of a modifying prepositional phrase, which hadn’t been a problem for bear arms in defense of Y, rendered uses of bear arms against X  irrelevant. And what makes that inconsistency especially strange is that to bear arms in defense of someone is necessarily to bear arms against whoever or whatever is attacking them. Bear arms in defense of X is therefore the yin to the yang of bear arms against X’s attacker.

The court didn’t try to explain or justify this inconsistency until three pages after the fact, when it discussed the amicus brief that had been filed by three linguists,  which had argued that in the founding era bear arms was ordinarily used in its military sense. These amici had argued that their position was not compromised by the existence of nonmilitary uses such as “bear arms . . . for the purposes of killing game” because those uses were “expressly qualified.” And when the court discussed that argument, it added that the dissent by Justice Stevens “uses the same excuse for dismissing the state constitutional provisions analogous to the Second Amendment that identify private-use purposes for which the individual right can be asserted.”

As you can tell from the description of Stevens’s analysis as an “excuse,” The majority didn’t think highly of the argument:

A purposive qualifying phrase that contradicts the word or phrase it modifies is unknown this side of the looking glass (except, apparently, in some courses on Linguistics). If “bear arms” means, as we think, simply the carrying of arms, a modifier can limit the purpose of the carriage (“for the purpose of self-defense” or “to make war against the King”). But if “bear arms” means, as the petitioners and the dissent think, the carrying of arms only for military purposes, one simply cannot add “for the purpose of killing game.” The right “to carry arms in the militia for the purpose of killing game” is worthy of the mad hatter. Thus, these purposive qualifying phrases positively establish that “to bear arms” is not limited to military use.

Stevens understandably objected that the court was taking inconsistent positions by basing its interpretation on one set of texts in which bear arms was modified by a purposive prepositional phrase but at the same time disregarding a different set of texts in which bear arms was similarly modified by a purposive prepositional phrase. Stevens had argued in effect that the military sense of bear arms was its default meaning, but that the phrase could mean something different if “the addition of a qualifying phrase signals that a different meaning is intended.” That argument was identical in its structure to the majority’s argument about bear arms against X; the arguments differed only with respect to which meaning (military or nonmilitary) was assumed to be the default, and which prepositional phrase (against X or in defense of Y) was assumed to fill the modifier slot.

The majority rejected the dissenters’ argument, and in doing so it drew two false dichotomies.

The first false dichotomy was between the majority’s argument and the dissent’s. According to the majority, the difference between the two arguments “is that we [the majority] do not maintain that “against” alters the meaning of “bear arms” but merely that it clarifies which of various meanings (one of which is military) is intended.” (Italics by the court.) I’ll assume for now that the distinction between “clarification” and “alteration” is valid, because even on that assumption, the distinction is irrelevant.

Contrary to what the majority implied, the dissent hadn’t argued that the effect of the against-phrase was to change the meaning of bear arms. As I read the relevant portion of the dissent (quoted below), its treatment of bear arms for purposes of killing game is perfectly congruent with the majority’s treatment of bear arms in defence/defence of themselves/himself: Under the dissent’s analysis, for purposes of killing game “clarifies which of various meanings [of bear arms] . . .  is intended” (italics removed):

The Court argues that a “qualifying phrase that contradicts the word or phrase it modifies is unknown this side of the looking glass.” Ante, at 15. But this fundamentally fails to grasp the point. The stand-alone phrase “bear arms” most naturally conveys a military meaning unless the addition of a qualifying phrase signals that a different meaning is intended. When, as in this case, there is no such qualifier, the most natural meaning is the military one; and, in the absence of any qualifier, it is all the more appropriate to look to the preamble to confirm the natural meaning of the text. The Court’s objection is particularly puzzling in light of its own contention that the addition of the modifier “against” changes the meaning of “bear arms.” Compare ante, at 10 (defining “bear arms” to mean “carrying [a weapon] for a particular purpose—confrontation”), with ante, at 12 (“The phrase ‘bear Arms’ also had at the time of the founding an idiomatic meaning that was significantly different from its natural meaning: to serve as a soldier, do military service, fight or to wage war. But it unequivocally bore that idiomatic meaning only when followed by the preposition ‘against.’ ” (citations and some internal quotation marks omitted)). [Footnote omitted.]

If after reading this passage you think that Scalia’s description was accurate, I hope you’ll post an explanation in the comments, because I just don’t see it.

The other false dichotomy drawn by the majority was the supposed distinction between a modifier “clarifying which of various meanings is intended” and a modifier “altering a word’s meaning.” As I see it, that is a distinction without a difference. The two formulations are just different ways of describing a single phenomenon: the sensitivity of word meaning to context.

As I like to put it, the meaning of a word in a particular context can be affected by the context. Words in isolation generally don’t have a determinate meaning; rather they have the potential to be used to convey a variety of different meanings, depending on the context in which they’re used. Despite the snark that Justice Scalia directed at linguistics, that proposition should be familiar to judges and lawyers, because it’s often expressed in judicial opinions. For example, in his majority opinion in Deal v. United States, Justice Scalia noted “[the] fundamental principle of statutory construction (and, indeed, of language itself) that the meaning of a word cannot be determined in isolation, but must be drawn from the context in which it is used.” (Although bear arms is two words, when it’s used as an idiom it functions to a large extent as a single semantic unit, and that’s how it was treated in the portion of Heller that I’m discussing.)

When the court in Heller said that the against in bear arms against “clarifies which of various meanings [of bear arms] . . . is intended,” it was implicitly invoking the “fundamental principle” of context-sensitivity that Justice Scalia referred to in Deal. The same is true as to the dissent’s statement that “‘bear arms’ most naturally conveys a military meaning unless the addition of a qualifying phrase signals that a different meaning is intended.” And although the majority described the dissent as arguing that the prepositional phrase in bear arms for purposes of killing game alters the meaning of bear arms, that argument could be turned back against the majority. That is to say, the majority could be described as contending that when bear arms is used without a modifier, it means ‘carry weapons for purposes of confrontation’ but that adding against X changes its meaning to ‘fight in a war.’

This isn’t to say that I think it’s accurate sense to describe the meaning of bear arms as being “altered” by the modifier. It’s not. It makes no sense to talk about the meaning of a phrase as being “altered” unless the phrase has a meaning that pre-exists the alteraion; it’s impossible to alter something that doesn’t exist. But as Justice Scalia recognized in his opinion for the court in Deal, “the meaning of a word cannot be determined in isolation, but must be drawn from the context in which it is used.” So when bear arms is used in a phrase such as bear arms for purposes of killing game, it is its use in that larger phrase that imbues it with meaning. There is therefore no preexisting  meaning to be altered, and as a result it makes no sense to say that the purposive phrase altered what bear arms meant.

What would make sense would be to say that when bear arms is used in bear arms for purposes of killing game, the presence of the modifier results in the “activation” or “accessing” of a potential meaning different from the one that would have been activated or accessed if the modifier had not been there. And Justice Stevens’s argument was framed in essentially those terms.

Much of what I’ve said in the last two paragraphs applies as well to the court’s argument that under the dissent’s interpretation of bear arms for purposes of killing game, the modifier (for purposes of killing game) “contradicts” the meaning of phrase it modifies (bear arms). Just as something that doesn’t exist can’t be altered, something that doesn’t exist can’t be contradicted. Given that bear arms has a determinate meaning only when it’s used in some context, what exactly is it that is contradicted  by the phrase bear arms for purposes of killing game? Some other phrase in which bear arms is used—just plain bear arms, perhaps, or bear arms against the King or bear arms in defense of ourselves? I don’t think so, but then I don’t even understand what it would mean to say that one of those phrases is or is not contradicted by one of the others. If anyone in the audience thinks they can offer elucidation, the comments are open.

TWO FINAL POINTS—not about flaws in Heller’s interpretation but about two consequences that would follow if Heller’s interpretation of bear arms were to be taken both literally and seriously (which so far isn’t how it’s been read). Both of these consequences will be surprising; one would, if it came to pass, be unwelcome from the point of view of gun-rights advocates; and the other would (I hope) be unwelcome by people on both sides of the issue.

First, the interpretation by its own terms precludes hunting from being treated as a variety of “bearing arms.” Recall that interpretation defines bearing arms as carrying weapons “for the purpose … of being armed and ready for offensive or defensive action in a case of conflict with another person.” That’s not how Heller has been applied, though, and as far as I can tell the issue has gone almost entirely unrecognized. (The main exception is an article by Yale law professor Akhil Amar, who wrote, “Justice Scalia limited recognition of an individual right to keep arms to situations of self-defense involving ‘confrontation’—that is, ‘conflict with another person’—as distinct from, say, hunting or recreation.”)

Second, and more important, if Heller’s definition of bearing arms is read for all it’s worth, it includes within its scope the carrying of arms for criminal purposes. That is at least strongly suggested by the fact that the definition refers to carrying weapons for the purpose of being ready for “offensive” as well as “defensive” action. And more importantly, remember that Muscarello had to do with a criminal prohibition on carrying firearms. Specifically, the statute at issue imposed a mandatory five-year sentence on anyone who carries a firearm “during and in relation to” a drug offense or crime of violence. So the dissent’s definition necessarily encompassed carrying firearms for criminal purposes. (In case you’re wondering, Justice Scalia had joined in the dissent in Muscarello.)

As to this issue, as with the hunting issue, Heller hasn’t been interpreted or applied in a way that’s consistent how the decision defined bearing arms. In Heller itself, the court said, “Like most rights, the right secured by the Second Amendment is not unlimited. From Blackstone through the 19th-century cases, commentators and courts routinely explained that the right was not a right to keep and carry any weapon whatsoever in any manner whatsoever and for whatever purpose.” And in the subsequent decision holding that Heller applies to state as well as federal law, the court said that Heller’s “central holding” was “that the Second Amendment protects a personal right to keep and bear arms for lawful purposes[.]” (my emphasis). But in neither case did the court recognize (much less try to resolve) the tension between these statements and Heller’s statement of the “natural meaning” of bear arms.

The result is that Heller is a decision at war with itself.

[Cross-posted on LAWnLinguistics.]



Cases regarding “natural meaning” and “ordinary meaning”

United States v. Bestfoods, 524 U.S. 51, 66 (1998):

This much is easy to say: the difficulty comes in defining actions sufficient to constitute direct parental “operation.” Here of course we may again rue the uselessness of CERCLA's definition of a facility's “operator” as “any person … operating” the facility, 42 U.S.C. § 9601(20)(A)(ii), which leaves us to do the best we can to give the term its “ordinary or natural meaning.Bailey v. United States, 516 U.S. 137, 145 (1995) (internal quotation marks omitted). In a mechanical sense, to “operate” ordinarily means “[t]o control the functioning of; run: operate a sewing machine.” American Heritage Dictionary 1268 (3d ed. 1992); see also Webster's New International Dictionary 1707 (2d ed. 1958) (“to work; as, to operate a machine”). And in the organizational sense more obviously intended by CERCLA, the word ordinarily means “[t]o conduct the affairs of; manage: operate a business.” American Heritage Dictionary, supra, at 1268; see also Webster's New International Dictionary, supra, at 1707 (“to manage”).

Dir., Office of Workers’ Comp. Programs, Dep’t of Labor  v. Greenwich Collieries, 512 U.S. 267, 272 (1994):

Because the term “burden of proof” is nowhere defined in the APA, our task is to construe it in accord with its ordinary or natural meaning. Smith v. United States, 508 U.S. 223, 228 (1993). It is easier to state this task than to accomplish it, for the meaning of words may change over time, and many words have several meanings even at a fixed point in time. Victor v. Nebraska, 511 U.S. 1, 13–14 (1994); see generally Cunningham, Levi, Green & Kaplan, Plain Meaning and Hard Cases, 103 Yale L.J. 1561 (1994). Here we must seek to ascertain the ordinary meaning of “burden of proof” in 1946, the year the APA was enacted.

F.D.I.C. v. Meyer, 510 U.S. 471, 476 (1994):

The first question, then, is whether Meyer's claim is “cognizable” under § 1346(b). The term “cognizable” is not defined in the Act. In the absence of such a definition, we construe a statutory term in accordance with its ordinary or natural meaning. Smith v. United States, 508 U.S. 223, 228 (1993).   Cognizable ordinarily means “[c]apable of being tried or examined before a designated tribunal; within [the] jurisdiction of [a] court or power given to [a] court to adjudicate [a] controversy.” Black's Law Dictionary 259 (6th ed. 1990).

Smith v. United States, 508 U.S. 223, 228  (1993):

When a word is not defined by statute, we normally construe it in accord with its ordinary or natural meaning. See Perrin v. United States, 444 U.S. 37, 42 (1979) (words not defined in statute should be given ordinary or common meaning). Accord, post, at 2061 (“In the search for statutory meaning, we give nontechnical words and phrases their ordinary meaning”).

Johnson v. United States, 529 U.S. 694, 715–16 (2000) (Scalia, J., dissenting) (footnote omitted):

The term “revoke” is not defined by the statute, and thus should be construed “in accordance with its ordinary or natural meaning.” FDIC v. Meyer, 510 U.S. 471, 476 (1994). As the Court recognizes, the ordinary meaning of “revoke” is “ ‘to annul by recalling or taking back.’ ” Ante, at 1803 (quoting Webster's Third New International Dictionary 1944 (1981)); see also American Heritage Dictionary 1545 (3d ed. *716 1992) (defining “revoke” as “[t]o void or annul by recalling, withdrawing, or reversing; cancel; rescind”). Under this reading, the “revoked” term of supervised release is simply canceled; and since there is no authorization for a new term of supervised release to replace the one that has been revoked, additional supervised release is unavailable.

The Court is not content with this natural reading, however, and proceeds to adopt what it calls an “unconventional” reading of “revoke,” ante, at 1803, as meaning “to call or summon back” without annulling, ante, at 1804. It thereby concludes that the revoked term of supervised release retains some effect, and thus that additional supervised release may be required after reimprisonment. The Court suggests that its abandonment of ordinary meaning is justified by the text, by congressional purpose, and by analogy to pre-Guidelines practice regarding nondetentive monitoring. None of the proffered reasons is convincing.


  1. neminem said,

    April 29, 2019 @ 3:14 pm

    Personally I prefer my own human arms, anyway.

  2. K. MacDonald said,

    April 29, 2019 @ 4:29 pm

    Here is how I understand the majority's comment about "bear arms in a militia for the purposes of killing game". The majority interprets the dissent as claiming that "bear arms" necessarily means "bear arms in a militia", even within a larger phrase. The majority infers from this claim that "bear arms for purposes of killing game" would have the same meaning as "bear arms in a militia for purposes of killing game". This meaning being ridiculous, the majority rejects the dissent's military interpretation of "bear arms".

  3. Stephen said,

    April 29, 2019 @ 4:33 pm

    Thanks for a great post. I'm still trying to process it all but enjoying the process of processing.

  4. Ted McClure said,

    April 29, 2019 @ 9:57 pm

    While I firmly agree with the poster's conclusions, I have three questions about the use of this corpus data in this situation.
    1. When did those who framed the Bill of Rights learn the use of words? Most of these people were of mature years (well, I'm not sure Franklin ever matured, but he was old). They would have picked up the meaning of "bear arms" some years before the Declaration of Independence, when they were in college or apprenticing for the law (again, with the exception of Franklin who apprenticed as a printer). Monroe was younger than the others, but I suspect he would have absorbed his elders' use of the term. Do we need to examine the use of "bear arms" in the third quarter of the 18th Century rather than the fourth? Would it make any difference?
    2. The Bill of Rights is a formal, legalistic document. Part of its overall context is "this is really, really important" and "this will be the foundation for the new nation's laws". Does it change our interpretation if we limit the corpus to similar formal legal documents? I suspect it strengthens the implication that the term was being used in a military, if not collective, context.
    3. What is the tone of the language used in the Bill of Rights and the Constitution as originally presented for ratification? From what time do the other legalistic terms come? Fourth quarter? Third quarter? Earlier? If the meanings of other terms are holdovers from Blackstone (royalist) or Coke (parliamentary), perhaps we need to look to earlier times for the meaning of "bear arms" in the Second Amendment.
    [Is there anything to the theory that these clever guys made it intentionally ambiguous, so that ratifiers in different states could assign their own meanings?]
    As a recovering attorney, I find that few words in English are truly unambiguous taken out of context. It is good practice to define and explain meaningful words in contracts, pleadings, regulations, etc.–unless of course ambiguity is intentional.

  5. Terry Hunt said,

    April 30, 2019 @ 11:23 am

    Reading this series of posts from the other side of the Atlantic, as I am, I have no personal stake in the overall political and historical context of the linguistic points being examined beyond abstract intellectual curiosity, so I am not seeking to grind any axes.
    That being established, a point has struck me that I don't remember being addressed before (forgive me if I am mistaken) which is something of a digression from the immediate subject. Most of the discussions have so far centered around the term "bear arms." What of the term "the people?"
    Much of the current political and legal furore around the Second Amendment assumes that the clause in question, arguably, gives any (adult) person the right to carry firearms. In
    modern American and British English, the wording one would expect in that case would surely be "the right of people to keep and bear Arms, shall not be infringed." Using the term "the people" implies to me (a modern British English speaker) a different meaning, namely that arms must be allowed to be carried by persons in some way acting on behalf of the people, such as "a well regulated Militia" or similar officially organised body. What it does not imply to (modern etc.) me is that this allowance should extend to all people, all or most of the time.
    I am wondering therefore if historical linguistic analysis can (or already has) shed any interesting light on what was, or might have been intended by "the people" in the Amendment. (I am aware that various State constitutions contain language that does imply the right to apply to (all) individuals rather than particular ones on behalf of the rest.)
    Apologies for the side-track, and if this point has already been analysed here without my noticing.

  6. Terry Hunt said,

    April 30, 2019 @ 11:26 am

    (Apologies for the Italics blunder in the post above: I probably messed up a 'close italics' code after the fourth "the" in the second paragraph.

  7. Kristian said,

    April 30, 2019 @ 12:07 pm

    You claim that the difference between these two meanings of "bear arms" is "unambiguous." In my opinion they overlap so much that they often can't be unambiguously distinguished. Just because a word usually appears in a military context does not mean it has a specific military meaning.

    Here is an example of "bear arms" "showing the military association" from Dennis Baron's essay,

    "The number of the Enemies that bear Arms, according to the truth, was about forty thousand more or less."

    Yes, the context is military. But how is the meaning of "bear Arms" here really distinct from the meaning "carry weapons"? It means, as far as I can tell, that the Enemies have forty thousand armed men (men carrying weapons), that this the strength of their army.

    Or am I to understand that this sentence might mean that the enemy army consists of about forty thousand men, who are "bearing arms" in the sense “to serve as a soldier, do military service, fight” or “to wage war" — but some non-trivial number of whom might not be armed?

  8. Neal Goldfarb said,

    April 30, 2019 @ 1:36 pm


    I'll address the question of whether bear arms really has two senses in a future post.

  9. Neal Goldfarb said,

    April 30, 2019 @ 7:24 pm

    In response to Ted McClure's questions:

    1 & 3. I think that using data from the 1760s-1790s is appropriate, because under the approach followed by the Supreme Court in Heller, and by the dominant strain of originalism, the focus is on how the text was likely to have been understood by members of the public (especially those who voted on ratification), not on what was intended or understood by the people we think of as the Framers (meaning those who participated in the process of deciding what the text would say).

    2. COFEA does include legal documents, but in any case the court in Heller framed the issue as being what the text's "natural" (=ordinary) meaning was (although they deviated from that approach when they relied on legal principles about the effect of the prefatory clause).

  10. Ted McClure said,

    May 1, 2019 @ 9:28 am

    Thank you.

  11. Peter Gerdes said,

    May 2, 2019 @ 6:03 pm

    Obviously the court's analysis is severely lacking in a number of ways (tho I think they were in a hard spot given their *fear* that the lack of balancing expert amici said more about linguist's ideology than evidence) but there are a number of issues I take with your analysis as well.

    First, just as the court tends to get things wrong when they try and opine on matters of linguists you also get things wrong when you try and step into primarily legal questions. For instance, your suggestion that the 2nd would protect criminal carrying is simply mistaken as a legal matter by the same reasoning that the first amendment allows criminalizing conspiracy. In many circumstances the court regards exercising a right as an integral part of violating a criminal statute covering unprotected conduct as subject to constitutional criminalization.

    Second, in your analysis you seem to blur the difference between a words connotation and it's meaning. Indeed, I don't even mean connotation here but something even weaker like affinity for use. For instance, if we were to look at a modern corpus of online discussion we would likely see that marginalized is (probably by a substantial margin) most frequently used in connection with the concept of a marginalized identity group but that association is in no way part of the meaning of marginalized. If I state, "Please don't marginalize anyone in this group" not modifier is required to communicated the fact that I don't want anyone being pushed aside or ignored even though it's not the most popular usage.

    Also, if we look closely we can see that marginalized is actually being used in two slightly different ways in the two cases in that homsexuals can be a marginalized group because their members (and not the group) are treated worse than a *comparable* member of society while, as applied to a conversation, it's about whether someone's contributions are being pushed aside in an absolute sense not relative to similarly situated individuals.

    We tolerate these small shifts in how a word applies so long as context is sufficient to work out a speaker's meaning. As such it's not a compelling argument to note that when I say "bear arms against" I'm expecting the totality of the phrase to convey a martial intent thus when raised in another context in the 2nd amendment it must also work in exactly the same way. That seems well within the degree of variation we are happy to fill in from context.

    More generally, reading your analysis the summary I would give it is: SCOTUS is full of shit in their arguments but the corpus analysis doesn't preclude either interpretation.

    A bit more to say but it is logically separate so goes in a followup comment.

  12. Peter Gerdes said,

    May 2, 2019 @ 6:55 pm

    Of course there is also the problem that we should probably be looking towards the understanding of the 2nd ammendment at the time of incorporation (passage of 14th) but putting that aside for the moment.

    I'd argue that, pragmatic considerations (if unbalanced) are the more important consideration and, absent a pure impossibility of using the word in the way one interpretation would have it, tend to trump (as they do in daily speech) the purely linguistic analysis

    In this case (unless I'm missing some historical detail) two considerations weigh in favor of the second amendment protecting individuals ability to carry guns in certain circumstances*.

    First, on the collective rights view the first amendment, at the time of adoption, would have protected the the right to keep firearms only insofar as they are within the scope of their state/local militia duties (which might include keeping the weapon at home etc..). However, that same effect could have been more clearly and simply carried out by stating that the federal government shall not infringe on the state's right (which it can devolve to subdivisions or other groups) to arm and maintain militia's as it sees fit.

    If the drafters had only meant to restrict the federal government's firearm regulation when it infringed on the state's power to create/allow militia's why didn't they just say that?

    To head off the argument you might be implicitly making I'd note that the use of "bear arms" rather than merely own, carry or use arms is explained by the desire to protect not only the ownership of unloaded weapons (as I believe some english laws had mandated in the past at that point) and the use of such weapons in hunting but also the ability to keep those weapons loaded and at hand. That's what I'd argue the confrontational connotation of bear arms is doing.

    Having said all this if I was on the court I wouldn't have read out the militia clause in interpreting the second amendment. The right interpretation, imo, is that it protects an individual right but only to own those weapons which would be in common usage in a militia. Thus, the government has to let people own M-16s and AK-47s but it can ban handguns, shotguns etc.. since assault rifles are the standard weapon of the infantry soldier. I actually think this would be good public policy but it has bad optics unfortunately.

RSS feed for comments on this post