and what about gyros?

Last week a court case turned on the definition of a word - sandwich, to be exact. I find it interesting, yet a little disturbing, that a legal proceeding should rely on the normally arbitrary association between sound and meaning, as if it is a rock-solid and undisputable certainty, when instead we know that this association is not stable.

Here's the back story: Panera Bread Company has hundreds of sandwich shops, including one in a mall in Massachusetts. In this mall (as, no doubt, in others), Panera has an exclusive contract, preventing the mall from leasing to any other sandwich shop. The mall decided to lease to a burrito shop, Qdoba, and Panera objected on the grounds that a burrito is a sandwich. In court, the judge ruled in favour of Qdoba, saying that it indeed is not a sandwich.

Aside, I'd be happy with the decision myself, but for different reasons. I find it irksome to be in some enclosure (e.g. a mall or airline terminal) and have little or no choice of what to eat - expecially if the one choice is overpriced and of low quality. But the idea of ruling a burrito as "not a sandwich" is not such an simple task. Here's what the ruling looks to for its definition of sandwich:

The difference, the judge ruled, comes down to two slices of bread versus one tortilla. "A sandwich is not commonly understood to include burritos, tacos and quesadillas, which are typically made with a single tortilla and stuffed with a choice filling of meat, rice, and beans."

OK, using "not commonly understood" seems to rely on usage as a criterion - but this is empirically testable yet seems to have gone untested. Qdoba did call expert chefs as witnesses to agree that burritos aren't sandwiches, but I wouldn't generalize what elite chefs have to say about food to "a common understanding".

But two slices of bread vs. one tortilla? I guess subs, cheese steaks, and gyros aren't sandwiches. Nor are shawarmas, kebabs, or donairs/doners, because they all use one piece of bread. And I shouldn't have to point out that tortillas are bread - so if you end up allowing one-piece-of-bread as a criterion, you fail to exclude burritos.

The article does not reprint the entire ruling, so perhaps the judge in the case was more thorough than the article presents. But it seems that as soon as you introduce a set of criteria to distinguish sandwiches from other forms of food, one of two results follows: either the criteria exclude other more obviously sandwichy items along with burritos, or they fail to exclude burritos.

I'm not sure what the moral of the story is ... perhaps it's just a mildly amusing reminder that contract law depends so much on lexical semantics (in addition to scope no doubt), and is thus clearly vulnerable to mis- or re-interpretation.