If you’re like me, you occasionally find yourself poking around on Google Scholar for linguistic studies of patent law. And when you do, you find studies in two distinct clusters: computational studies of patent corpora for purposes of information retrieval and classification (generally published in computational-linguistics journals), and arguments about the applicability of various linguistic theories to the interpretation of patent claims (generally published in law reviews and written by legal scholars who are not linguists).
The two clusters don’t seem to be in conversation with one another. I haven’t done an in-depth analysis, but I’d be mildly surprised if a paper in either cluster has ever cited a paper in the other cluster.
Neither of these clusters is terribly surprising; they’re both fascinating areas of inquiry. But it’s striking that the broader “legal linguistics” movement seems to barely have touched on patent law at all. And that’s especially odd because of the unique areas of linguistic interest that patent law affords. For example, United States patent law offers the Markman order, perhaps the only distinct legal genre anywhere that deals exclusively with the meanings of words.
My suspicion here is that the barriers around patent law are high enough — and the available scholarly terrain within legal linguistics still open enough — that a proper linguistic characterization of the unique language games involved in patent law hasn’t yet seemed worth any scholar’s time.
But still, it’s rather odd. Perhaps the gaps will begin to be filled in as legal linguistics becomes a more mainstream field of inquiry and new scholars look for new fields to till.