Is Originalism a Sandwich?

In the latest iteration of her "notable sandwiches" series, Talia Lavin tackles the age-old question "Is a hot dog a sandwich?" She gathered a host of experts from a range of different disciplines to give their take, and while there wasn't a consensus, it seemed to me (I didn't count) that more leaned against it being a sandwich. The general thrust of the argument that most resonated with me, from sociolinguistics professor Matt Garley, was to frame the question as "Do people commonly or regularly refer to a hot dog (outside of this particular debate) as a sandwich?" In that light, the answer seems to be generally "no", even if it seems to formally meet the dictionary definition of a sandwich ("two or more slices of bread or a split roll having a filling in between.").

Later in the post, Talia gets a quote from Jesse Sheidlower, a lexicographer and former editor of the Oxford English Dictionary, who gave some insight on how dictionaries themselves approach this problem. Contrary to (perhaps) popular belief, dictionaries are not in the business of trying to give precise definitions that perfectly include and exclude everything that descriptively falls within the category-type of a given word. I'll quote him at length:

The general thing to know about dictionaries is that you're usually not trying to capture the complete and exact description of something; you're trying to get a general picture of what something means. This is hard enough for concrete nouns that we more or less know, like "horse" or "sandwich"; it's impossible with abstract nouns like "freedom" or "beauty". One of the most famous definitions in lexicography is the one for "door" in Webster's Third of 1961:

"a movable piece of firm material or a structure supported usually along one side and swinging on pivots or hinges, sliding along a groove, rolling up and down, revolving as one of four leaves, or folding like an accordion by means of which an opening may be closed or kept open for passage into or out of a building, room, or other covered enclosure or a car, airplane, elevator, or other vehicle."

This is what happens when you try to be exact—you get something useless.

So most dictionaries, that are written for native speakers and that assume a good-faith effort to understand the definition, give a reasonably broad definition, that will include most things that should be included and exclude most things that should be excluded.

There are, conventionally, two main types of lexicographers: lumpers and splitters. Lumpers include as much as possible ('liquid food' for soup); splitters write a dozen super-narrow definitions, and when a new variant comes up, they write another one.

Dictionaries are generally more lumpy than splitty. A sandwich is a food with something inside a bready thing. Trying to be super-precise is only going to lead to frustration (or the "door" definition above): Most people feel that a meatball sub is a kind of a sandwich but a hot dog isn't, but that's very hard to explain, so unless you have a definition like "… or a split roll having a cold or hot filling (that is not a solid length of sausage)…", you're kind of stuck.

If I can turn serious for a moment—and this is very serious—the reason that this is genuinely important, and not just a parlor game, is that people sometimes put a lot of faith in dictionary definitions. In particular, courts use old dictionaries to try to determine what words meant at a time when laws were written. But that is very much not how dictionaries should be used. If it's this hard to determine what a "sandwich" is, what are we supposed to do about words like genocide, or to bear arms? Or woman in reference to a trans woman? People literally die because dictionaries are misused. There are ways to attempt to answer these questions—corpus linguistics, sociolinguistic interviews—but thinking that a dictionary is an exact map of reality is not a correct one of these.

I wasn't expecting to see this point made in a fun post about the concept of a hot dog, but here we are. And it did crystallize for me an objection I've been flagging recently about "vulgar" textualism or originalism; a practice of judicial interpretation that purports to distinguish itself by close and careful reading of texts, but actually is just very bad at reading texts. Many of the cases that take this approach begin with a very close parsing of dictionary definitions in order to fix textual meaning. But this from the jump misunderstands what dictionaries are even trying to do. Even at the moment they are written, dictionaries are an at best imperfect map onto actual public meaning (the idea being that even if we were looking at a dictionary published today to answer the question "is a hot dog a sandwich", we'd likely be heading off in the wrong direction). And that gap only grows wider as time passes, because the actual meaning of words depends on a host of agreed-upon implicit assumptions and cultural horizons that are constantly shifting and temporally-contingent. 

We run into this question when trying to figure out how to apply an old word ("search") to technology that hadn't been invented yet when the word was written ("heat scanning"). One way of answering "is heat scanning a search under the Fourth Amendment" is to look at the dictionary definition of "search" circa 1789 and figure out if it fits. But that actually wouldn't really be the accurate answer, because what we'd actually want to know is if the relevant interpretive community would have generally used "heat mapping" as falling under the category of search. And that question, in turn, is essentially incoherent unless we also import into that community a host of surrounding cultural and linguistic practices that make "heat mapping" a legible concept that could be part of a robust linguistic pattern to begin with (if you plop down a heat mapper into 1789 without all of that context, then it's going to be seen less as a "search" and more as "eldritch magical witchcraft"). So what we're really asking when trying to figure whether heat mapping qualifies as a search today is "how would the relevant class of interpreters understand the relationship between these words, if they had the full cultural and linguistic context that we have today -- and at that point, our "originalism" is essentially just living constitutionalism.