My co-authors (Jacob Crump and Benjamin Lee) and I recently posted on SSRN a study we did of the original meaning of the phrase “Officers of the United States.” Our study was motivated by Jenn Mascott’s recent article in the Stanford Law Review. As part of her research, Professor Mascott performed “corpus-like analysis” on a small portion of the texts of the soon-to-be released Corpus of Founding-Era American English (COFEA, rhymes with “Sophia”). To be exact, she used as a resource approximately 7.6 million words from over 16,000 files obtained from the National Archive’s Founders Online project. These texts consisted primarily of letters written to and by six famous founders: George Washington, John Adams, Benjamin Franklin, Thomas Jefferson, James Madison, and Alexander Hamilton. Further, Professor Mascott limited her analysis to just the time frame of the Confederation: 1783-1789.
We wondered, then, whether what she found would be the same if one expanded the analysis in three ways. First, drawing on all of the analytical tools of a corpus: collocation, n-grams (or clusters), frequency data, and concordance lines. Second, expanding the analysis to texts falling under COFEA’s entire time period: 1760-1799. Finally, looking not only at these six founders’ papers, but also the two other sources of documents that make up COFEA: legal materials from Hein Online and pamphlets and books (often including other example of language usage, such as sermons) from the Evans Early American Imprint Series. In short, would applying the full analytical rigor of corpus linguistic tools to approximately 150 million words of a variety of texts and authors covering the last 40 years of the 18th Century provide the same results as Professor Mascott found with her narrower search? (Of note, while our corpus linguistic analysis was broader than Professor Mascott’s, her paper’s overall methodology was broader as she also drew on non-corpus linguistic originalist methodologies.)
Before turning to COFEA, we surveyed two dozen founding-era dictionaries, both general and legal, to get a sense (pun intended) of the range of meanings of the words officer and office. We found seven broad senses of the term officer, but most seemed irrelevant to the constitutional debate. So we just focused on the public/civil officer sense. We then, again from the founding-era dictionaries, came up with three potential sub-senses of that sense:
- appointed to or in a public post, business, charge, trust, office, or place
- performing any authorized public duty, service, function, or stewardship
- hired by the public to do something (of any nature or duration)
There is admittedly significant overlap between sub-senses 1 and 2. Finally, we noted that the term public employment (and its variations) was often used to define a public/civil officer. We thus posited four definitional scenarios:
- if one if publicly employed, i.e., hired by the government to do something, one is also considered an officer
- all officers are publicly employed, but not all who are publicly employed (i.e., hired by the government) are officers
- publicly employed is a term of art that means officer, and thus those hired by the state who are not officers are not publicly employed
- public employment is different than being an officer
Finally, with a better understanding of the linguistic lay of the land, we turned to COFEA. (As a side note, one could also take an approach of not delving into founding-era dictionaries so as to see the corpus data with fresh eyes, untainted by usually less than complete dictionaries; however, we opted not to take that route because we worried that our modern notions of English might bias our analysis. And to compensate, we left open the possibility that there were senses not included in our initial list.)
While the corpus has many tools and we used them all (see the paper for more details), arguably the most important for this type of inquiry is the hard work of slogging through instances of a word or phrase (found via concordance lines) and reading the text surrounding such to classify which sense is being used. We thus randomly sampled instances of the terms officer and public employ (and variations—see the paper for the various searches) from each mini-corpus of COFEA (Founders, Hein, and Evans) to see if we could find a dominant sense. This required reading the equivalent of a couple of Harry Potter novels of founding-era American English. Unfortunately, we struggled to distinguish some of the senses, leading to results that had a high level of ambiguity.
So we turned to a second-best approach: looking for the actual offices described when referring to an officer or office. This is the second-best approach because it has the danger of constructing a sense that is too narrow. Take the word bird, for example. Imagine we looked for which type of bird was most commonly referred to in a corpus of modern American English, and found that robins were the type of bird most often mentioned in near proximity of the word bird. To then determine that the meaning of the animal sense of the word bird did not include ostriches or penguins would be an error. Thus, when using factual frequency to get a handle on the scope of a sense, one has to take the exact opposite approach then when looking at sense frequency: expand the definition to include the various factual scenarios found.
In sum, our analysis found that when referring to office(s) or officer(s), those in the founding era appeared to take a broader view of those terms than the Supreme Court’s current definition: one who exercises significant government authority. Folks who we might not think of as officers today were referred to such in the last half of the 18th century, such as surveyors, public registers, impost collectors, notaries public, loan commissioners, land recorders, tax collectors, deputies and agents, postmasters, auditors, purveyors of the public supplies, assistant postmasters, assistants, stewards for the public affairs of the country, attorneys, accountants, personal secretaries to foreign diplomats, clerks (sometimes), geographers, etc. That our findings clash with the Supreme Court’s doctrinal definition is probably not surprising given that the Court’s definition does not pretend to be an attempt at divining the Constitution’s original meaning.
But it also doesn’t appear that everyone who worked for the government was necessarily an officer. Occasionally someone was referred to as working for the government but not being an officer, including some clerks. Given that we were not able to determine the definitive boundaries of the meaning of Officers of the United States, more research is needed, both of the COFEA data since we only took samples of the results, and also using other originalist methods. What we do know is that the Supreme Court’s four-decade-old definition is contradicted by the way people used language on this side of the Atlantic from 1760-1799. And thus at the founding, people probably understood the phrase Officers of the United States to be much broader than the Supreme Court has previously thought.
James Phillips is currently a constitutional law fellow with the Becket Fund for Religious Liberty and a PhD candidate in jurisprudence and social policy at UC-Berkeley. The views he expresses do not necessarily represent the Becket Fund or its clients.
This post is part of a symposium on Lucia v. SEC. All of the posts can be read here.