A recent post referred to this week’s SCOTUS decision in FCC v. ATT (strictly speaking, FEDERAL COMMUNICATIONS COMMISSION ET AL. v. AT&T INC. ET AL.), a decision backed up by linguistic research into several electronic corpora (plural of corpus). A corpus is a large collection of texts, which can be analyzed for usage patterns, concordances, and patterns in language evolution.
SCOTUS has a long history of trying to ferret out the meaning of words; the approach predates the availability of electronic corpora. For example, in Gutierrez v. Ada, the court grappled with the meaning of the word election. For information management professionals, Gutierrez v. Ada is noteworthy because the initial ambiguity in the law would have been avoided if the original legislators had started with a conceptual data model.
To data modelers, there is an important difference between FCC v. ATT and Gutierrez v. Ada. In FCC v. ATT, the linguistic ambiguity concerns the meaning of an adjective (“personal”). In Gutierrez v. Ada, the ambiguity concerns the meaning of a noun (“election”). The semantic dispute of Gutierrez v. Ada could have been avoided with a conceptual data model, but the ambiguity in FCC v. ATT would not. Data models define nouns, not adjectives.
Here’s a snippet from the 8-0 decision in Gutierrez v. Ada:
Respondents' position boils down to the claim that the phrase "majority of the votes cast in any election" requires that a slate of candidates for Governor and Lieutenant Governor receive a majority of the total number of ballots cast in the general election, regardless of the number of votes for all gubernatorial slates by those casting ballots. If this is the correct reading of the phrase, the parties agree that a runoff was required. If, however, the phrase refers only to votes cast for gubernatorial slates, no runoff was in order, and petitioners were elected Governor and Lieutenant Governor.