This article first appeared in The Originalism Blog on March 1, 2021. It is somewhat technical, and I recommend it only for constitutional wonks.
I. The question
I recently wrote a post for the Federalist Society Blog, in which I examined the Constitution’s enumeration of who can be impeached and convicted. I concluded that the enumeration, coupled with the rules of construction the Founders intended to accompany the document, created a presumption that ex-officers could not be. But I added that this presumption was rebuttable by extraneous evidence.
Professor Michael Ramsey wrote a notice on the posting. He remarked that most originalist inquiries can proceed as I did—that is:
* examine the text and apply the rules of construction to it, thereby establishing a presumptive meaning, and then
* turn to outside evidence to see of it rebuts, qualifies, or reinforces the presumptive meaning.
I agree with Professor Ramsey. However, his comment got me thinking about the minority of cases where one cannot proceed in that manner. These are instances in which the text remains unclear even after applying the rules of construction. Moreover, sometimes after examining the text it even remains unclear whether that text is unclear. In fact, what seems obvious on initial examination may be entirely wrong.
In such cases, we must consult outside evidence to establish a presumptive meaning.
Fortunately, the impeachment enumeration I addressed in my FedSoc post was clear, at least as far as I needed to go. However, as explained below, on first reading another impeachment phrase—“high . . . Misdemeanors” —is not.
In this post I offer some guidance as to when an initial resort to extraneous evidence is necessary to establish the text’s presumptive meaning. For illustrations, I’ve mined my thirty or so articles and book chapters on the original meaning/understanding of specific constitutional clauses.
In some cases, I have been able to reach a presumptive meaning by examining 18th century dictionaries. Mind, though, that when consulting dictionaries, one must not stop with Samuel Johnson, because his definitions are sometimes archaic or idiosyncratic. I have about 25 18th century dictionaries in PDF form, not counting multiple editions. Some are specialized, such as dictionaries focusing on law, military terms, Latin, or Law French. Most are general-purpose. When writing on original meaning I consult all or most of them. I find I can learn much from how their definitions vary or track each other.
But dictionary searches can prove inconclusive. Sometimes dictionaries offer multiple definitions, so you have to examine extraneous evidence to determine which one the Constitution employs. Sometimes a term appears in a specialized dictionary, but not in others. Sometimes the definition employed by the Constitution does not appear in any dictionary at all.
Results like this compel resort to extrinsic evidence to determine presumptive meaning. Here are some typical scenarios. You will notice that some of them overlap.
III. Some Scenarios
Scenario #1: The dictionaries offer several common meanings of a word or phrase, but it is not clear which one the Constitution employs
I encountered an example when researching my article on the Coinage Clause. The Coinage Clause reads, “The Congress shall have Power . . . To coin Money, regulate the Value thereof, and of foreign Coin . . . ” As is true today, during the 18th century the most common use of the verb “to coin” was to strike metallic tokens. Libertarians long relied on the assumption that this was the Constitution’s meaning, and therefore claimed paper money is unconstitutional. Left-of-center commentators relied on the same assumption to illustrate the impracticality of originalism.
However, no one seems to have noticed a textual problem with interpreting coinage as referring only to metal. That interpretation would leave Congress in the unlikely position of enjoying power to regulate foreign metallic tokens, but bereft of power to regulate foreign paper currency.
The 18th century dictionaries reported a secondary meaning of “to coin:” to fabricate. Today that usage is exceedingly rare: a survival is the expression “to coin a phrase.” But a survey of 18th century databases revealed that the secondary meaning was then much more common. People spoke of “coining” paper money, leather money, and so forth.
Thus began a full-bore investigation into the history of money, especially the extraordinary monetary innovations undertaken in the colonies of British North America. After acquiring this background, my presumptive conclusion was that the Coinage Clause authorized Congress to issue and regulate “coin” in media other than metal. Ambiguous comments during the Constitutional Convention were insufficient to rebut this result, and the ratification-era debates only reinforced it.
When researching my early article statutory retroactivity, I encountered another problem with which the dictionaries offered no help. This was ambiguity of the phrase “ex post facto law.” During the 18th century, the term might refer only to criminal statutes, or it could include retroactive civil statutes as well. The framing and ratification records clarified that the prevalent understanding was the former. A search for contradictory evidence led only to confirmation: James Madison’s and John Lansing’s subsequent drive for Takings and Due Process clauses to limit the scope of civil retroactivity.
Scenario #2: A word or phrase with an ordinary meaning appears in the Constitution, but as a legal term of art
In this scenario, a colloquial meaning may appear in lay dictionaries, but the manner in which the Constitution uses the word appears in law dictionaries or other legal works.
Over the years, I’ve been impressed by how often a puzzling word—or even, at first blush, an unpuzzling one—turns out to be an 18th century legal term of art. It’s not so surprising, if you think about it. The chatter about how the Constitution is written in “plain language” is not quite the truth. The Constitution is a legal document and it was written by a group of people about two-thirds of whom had been practicing lawyers. Most of advocates who explained it to the public were lawyers, Madison and Tench Coxe being the most notable exceptions.
Nor were their explanations necessarily unintelligible to the involved American public, because that public was then unusually well educated in law: As Edmund Burke remarked when urging conciliation with America, “In no country perhaps in the world is the law so general a study.”
Of course, you expect a phrase like “habeas corpus” to embody a legal meaning. You don’t expect the same from a word like “necessary.” Yet in research on the Necessary and Proper Clause, I found that the Constitution uses “necessary” as a signal for incidental powers—a very common approach in 18th century legal documents.
Learning enough to establish a presumptive meaning for the Necessary and Proper Clause required deep initial research into contemporaneous law books, documents, and judicial cases.
Other examples of 18th century legal boilerplate reproduced in the Constitution include “Privilege” and “Privileges and Immunities.” A search through contemporaneous law and legal documents found that they denoted government-created entitlements, including very important ones such as trial by jury and “the Privilege of the Writ of Habeas Corpus.” But as the Constitution used those words, they did not comprehend rights established by nature and nature’s God. They were creations of government. Justice Bushrod Washington therefore got it largely wrong in his famous, and somewhat incoherent, passage in Corfield v. Coryell.
It probably comes as no surprise that “direct tax” was another legal term of art. What did surprise me is that the concept was far more defined and comprehensive than commonly supposed.
Both Britain and American states adopted omnibus tax statutes that provided for assessment of certain items and imposed levies upon them. In Britain and some states these were called “Land Tax” laws. But they levied on much more than real estate. They taxed human beings (capitations on both free and slave); status; wealth; professions (“faculties”) and other activities; wages, interest, profits and other kinds of income; household items, livestock, and other personal property. All the levies imposed by these omnibus statutes, and others like them, were called “direct taxes.” (Other statutes imposed indirect taxes, primarily on consumption or discrete events: excises and other “duties.”)
Thus, the Supreme Court’s much-abused case of Pollock v. Farmers’ Loan and Trust Company was correctly decided, for income taxes are direct taxes after all.
Obviously, arriving at a presumptive meaning of “direct tax” required a great deal of work with extrinsic evidence, including but not limited to 18th century tax statutes.
One final example: For decades commentators had argued over the Constitution’s term “high . . . Misdemeanors” as a ground for impeachment. Apparently only one researcher, Raoul Berger, had considered whether the phrase might be a legal term of art. But his investigation, undertaken without the benefit of modern word-search technology, was cursory and unsuccessful. Too late to prevent me from embarrassing myself in print with another hypothesized definition, I finally thought to check legal sources and certain lay sources (such as encyclopedias) explaining legal terms.
The answer was decisive: High misdemeanors were serious crimes not meriting the death penalty, such as bribery (which the Constitution mentions in this context) and assault. A mere breach of fiduciary duty was not, as I previously had thought, a high misdemeanor.
Scenario #3: The Constitution’s use of a term is not in dictionaries at all
Sometimes the framers employed words and phrases in ways that don’t appear in 18th century dictionaries or similar reference works. Instead, the constitutional meaning arose in a context the lexicographers had overlooked.
I encountered this situation when researching the Constitution’s use of the word “emolument.” The dictionary definitions were very broad: “profit” or “advantage.” Those broad definitions, I found, accurately reflected much lay usage, but for various reasons they made no sense in the constitutional text. Hence it became necessary to consult extraneous evidence to arrive at a presumptive meaning.
It turns out that in political discourse, “emoluments” carried any of three narrower meanings: (1) financial gain or, more commonly (2) gain from salary or wage plus associated fringe benefits, or (3) fringe benefits alone. Placing the Constitution’s emoluments clauses within the wider context of a then-current trans-Atlantic movement to reform how government officials were compensated enabled me to fix on what I believe is correct: An “emolument” as the Constitution uses the term refers to gain from a salary or wage plus associated fringe benefits.
Scenario #4: Some of the Constitution’s words are Americanisms
This is really a subset of Scenario #3, because these are usages that do not appear in the dictionaries, almost all of which were published in Britain.
For example, in researching my article on the Taxation Clause I found that “Eighteenth century British lay dictionaries defined ‘duty’ widely enough to include almost any financial exaction” and commercial dictionaries defined it more narrowly. But in American usage, a duty was “any financial exaction that did not qualify as a direct tax.” Thus, a duty could mean an indirect tax or a non-revenue-producing exaction to regulate commerce or other conduct. Other evidence reinforced this presumed meaning.
In 1787, the word “constitution” was an Americanism-in-development. In Britain it referred only to the political system, and that is how all contemporaneous dictionaries defined it. That was the American meaning when the Declaration of Independence was written (“a jurisdiction foreign to our Constitution”), and it was still employed that way in Virginia’s 1786 call for the 1787 federal convention. (Contrary to common belief, the convention was called in late 1786 by the Virginia legislature, not by Congress; Congress’s February 21, 1787 resolution merely endorsed the effort after seven states already had accepted Virginia’s invitation.)
However, the meaning was in flux. The “frames of government” adopted in the states often were called “constitutions.” Then came the U.S. Constitution, whose title sealed the change. A new Americanism was born.
Scenario #5: A word may have a constitutional meaning we don’t suspect because that meaning is now archaic
My favorite example of this—in fact, I think a perfect one—is the appearance of “perfect” in the Constitution’s Preamble. Modern Americans sometimes puzzle over how a Union can be made “more perfect.” This is because we almost always use this word to mean “without flaw.” How can something be “more without flaw?”
The answer is that the more common 18th century use of the term was the Latinate meaning of “complete” (Latin: perficere, to finish). The new Union was to be more complete—more tightly woven together—than the union formed by the Articles of Confederation.
Fortunately, there are not many non-legal archaisms in the Constitution—at least not many that sneak up on you. They are more likely to do that when you read other Founding-Era materials.
A good example was Leonard Levy’s mistaken conclusion that the Senate was to be the primary conductor of foreign policy because the framers sometimes referred to the president as the foreign policy “agent.” Levy didn’t realize that there was another, more Latinate, meaning of “agent” then common. Although an “agent” could mean a representative, it also could mean a “doer” or “driver” (from agere, to do or drive). The founders were using the latter sense: The president, not the Senate, was to be the primary driver of foreign policy.
Of course, you might not realize that a word is an archaism and therefore not think to consult a dictionary. That is one reason I recommend that constitutional scholars immerse themselves in 18th century writings and become familiar with the Latin language. (“But that’s time-consuming and hard!” Answer: “You bet it is.”) Latin was the Founders’ second language and significantly influenced 18th century English usage. If you have it, you see much more.
Professor Ramsey is correct to say that we usually can reach a presumptive meaning of a constitutional provision from the text alone. But there are many situations in which you have to consult extrinsic sources before arriving at a presumptive meaning.
If you find yourself in such a situation, I can offer two sources of comfort: First, much of the evidence you explore while seeking the presumptive meaning leads you to evidence that rebuts or reinforces that meaning. Second, the historical journey is usually fascinating.