What Did They Really Mean by That?

Spoken language is a dynamic process in which the referents of words we use are influenced, and often limited, by the context in which speech takes place. Unless it is captured by some form of recording, speech does not survive the context in which it is produced. To the degree that speech lives on, it does so in memory only, and through the effects that it may produce. But for most practical purposes, speech has done its work as soon as it has ended.

Written language, on the other hand, has an afterlife that is practically unlimited. When we commit something to writing we make a record that far outlives the context in which it is produced. Like speech, writing produces an output — words arranged in coherent fashion to express meaning — but when we read, we are not in the context in which the words were formulated, nor do we have the facility that is sometimes present in speech to ask for clarification. And so we have to wonder as we read: do we really understand what is meant? Do we adequately know the context in which the words were produced to understand what their meaning and intention was?

The classic area in which such questions become pertinent is in the study of texts regarded as canonical — that is to say, constituting a canon, "a body of rules or principles generally established as valid and fundamental in a field." Government and religion are two domains in which such canons hold authority, and we deal with controversies about meaning in these two domains in very different ways. Let's save religion for another day and look at how meaning and intention are agreed upon (or not) in legal canons.

The best example of such a canon is a constitution. It's a great benefit for text interpreters that constitution-drafters do their work with an awareness that the force of their words will long outlive them: they choose their words very deliberately and they usually have the foresight to give their future interpreters a mechanism for revision of the canon.

The United States Constitution, now on the job for more than 225 years, has proven to be robust, largely owing the great care with which it was crafted and the provision in it for amendment. That provision has been invoked 27 times, so that the Constitution now consists of the original text plus the 27 amendments that either expand upon, clarify, or nullify the language of the original constitution. A few years ago I wrote here about a book by former Supreme Court Justice John Paul Stevens in which he proposed changes to the Constitution that he thought were still needed. None of these has been acted upon.

Perhaps because amending the Constitution is a fraught and time-consuming process that doesn't always result in success, it's a popular activity to argue about what certain provisions of the Constitution actually mean. The classic example here is the language of the Second Amendment. When I wrote about it here five years ago, the article became troll bait and then-editor Ben Zimmer chose to tone down the invective in the comments section. More recently, Language Log contributor Neal Goldfarb has written a series of posts that look at the language of the Second Amendment based on evidence in corpora surrounding the use of the phrase "bear arms" contemporaneously with the framing of the Constitution. This has also resulted in some rants in the Language Log comment section that resulted in editorial intervention.

Another recent example: Georgetown University professor John Mikhail has questioned the interpretation of emolument in the Constitution (paper here; video summary here) that the Department of Justice has put forward in defense of President Donald Trump's vast business empire and the profit he receives from it. Mikhail's scholarship has received some attention in the news; and where there's news coverage, there's often rancorous debate, and trolling.

In both of these cases, the upshot seems to be that if you are going to challenge a popular, long-established, or authoritatively asserted interpretation of the meaning of canonical language, even with the presentation of impressive evidence, you need to steel yourself against zealous defenders of the other view. This is surely because with a canon like the Constitution, only one interpretation can prevail. No American is free to live under the provisions of the Constitution as they interpret them personally; we must all accept the interpretation that legislators, and ultimately the Supreme Court, decides.

Scholars such as Goldfarb and Mikhail make the argument that if you want to know what a word meant at the time it was used (as opposed to what it means right now), you have to look at examples of its usage contemporaneous with the usage you are investigating. Goldfarb's use of corpora, and Mikhail's use of older dictionaries, are meant to get a fix on how words were used in the past. I wrote last year about COFEA, the Corpus of Founding Era American English, a repository of around 100 million words of text written between 1760 and 1799, taken from a variety of sources. COFEA is in fact one of the corpora that Neal Goldfarb has used in his research.

Such detailed historical research may be seen as a great benefit to understanding the original meaning of canonical works, or it may be viewed as a red herring. How you see it may depend on two things.

The worse of these two things is already being deeply invested in one view or the other about meaning. Research in the two cases noted above tends to support an idea that the popular interpretation of the Second Amendment, and the Justice Department's interpretation of emolument, are not accurate. So if you're a holder of the popular or asserted views, it may seem expedient to simply kill the messenger bearing an opposing view. If you oppose either or both of the current prevailing views, you are likely to say, "Three cheers for historical scholarship".

The better reason for having some reservations about the assertion of historical meanings is that it doesn't really matter exactly what the original writers had in mind, because we don't live in their world and there is no way, nor any reason, for us to recreate it, any more than there was a way for the Founders to anticipate the complications of our modern world. Perhaps it is not our job to painstakingly determine exactly what someone meant two centuries ago; rather, it's our job to find a realistic and workable interpretation of their ideas and apply it to the world we live in.