Image

The Crazy Thing about Linguistic Research

Just when you think you have something nailed down, turns out you were holding a cherry tomato and the nail just made the whole thing explode.

I’m constantly figuring out how to be a better librarian to the disciplines I serve. I have pretty deep knowledge of the ways of literary research, since that was my own field, but the rest of it I’m still figuring out. And recently, the Linguistics department (finally a department in its own right, here!) has been ramping up the research requirements, and my involvement in those requirements. Which is great! And I have a lot to learn.

Today, for example, I had a student coming to me for help with a paper for his phonology class. He’d come yesterday, too, and we’d found a tiny smattering of research on his topic, but nothing that seemed like viable material for the foundations of a paper. I gave him my speech about sometimes needing to broaden out the search to related topics and apply what he learned from them to his current topic. A speech which went over about a well as it ever does, which is to say, not very. In a compressed term, that kind of research takes more time than most students (or professors) leave room for. We both pledged to do some more digging and scheduled a follow-up meeting for today.

In between yesterday and today, I remembered something I’d heard years ago but never really understood: that linguistic descriptions of individual languages are more like ethnographies than studies as far as the position they have in the field goes. They’re done once, and then that’s done. People propose tweaks, examine implications, explain why patterns exist the way they do, but a comprehensive description of Nepali phonology? That probably won’t get redone even once a half century. Like ethnography, the description from the 50s and 60s is probably still the description, no matter its gaps and flaws. Today we found him a whole collection of sources, now that we both knew to look for older things and to look for books.

And that’s where I find Linguistics research interesting, taken as a whole. On the one hand, it’s got a foot in ethnography, where the publication date hardly matters when deciding if the thing in hand is valid for study and citing. On the other hand, it’s got its foot in brain processing research, where publication matters a whole lot. When figuring out how people process and store words, imagine the difference between studies done before and after fMRI was prevalent. And that’s just the technology. What’s known about what, exactly, people see when they look at fMRI images is evolving day by day.

And yet again I remember the little “how to evaluate a source” check lists that I got in library school and how they are so terribly inadequate to describe the full scope of research values. In linguistics, date matters one moment, and not the next.

One thought on “The Crazy Thing about Linguistic Research

  1. I have SOOO enjoyed these last two posts! Linguistics is where I got my PhD, and its approach to data is partly why I left — and this is what you ran into here. Descriptions of grammars, phonologies, text-types, etc get written once, then indeed, small moves are subsequently made as new hypotheses about structures are emerge. When I was still doing linguistic research, I was using texts from the late 1800/early 1900s because that’s when someone catalogued the grammar. And in many cases, those analyses were written by non-linguists, especially a lot of missionaries. I never had a problem with this kind of data collection because it tended to be comprehensive. What drove me crazy is when a single sentence or small set of sentences is used as the empirical basis for a new idea because one native speaker (or maybe 6!) from a single region agree on the correctness of a given expression. This is where the anthropological history of linguistics fails to provide sufficient evidence for what are often cognitive claims. If appeals are made for using scientific criteria for evidence, then one can run afoul of linguistics profs who do not, by and large, feel the need for extensive data before hypothesizing wildly (Chomsky gets the blame for this — the practice is an extreme adaptation of his tenet that language is its own system, perfectly formed in each normal adult brain, hence a single native speaker is all you need to represent the whole system). You may need to categorize the linguists you’re working with into those who are traditionalists and those who are inter-disciplinary (meaning they look at connections to cognition and the brain). When I was a a student here, people who connected language to cognition belonged in psychology, not linguistics. And many of my profs were horrified when I started doing truly basic quantitative work (counting instances of a thing, for instance) — I think b/c it went against the Chomkian view of the perfect native speaker.

    Several years ago I put together a course on Writing in Linguistics. It went away after just 2 years b/c we don’t have enough linguistics undergrads (we have not yet made it from program into department!) to justify an outside-the-field upper-division class. I’ll be happy to clean up the links and send them your way if you think it may help! I am hoping eventually to transfer that class to wikidot just so that I can point to it. And typically of linguists, when I put together this class, they were all mystified — how dare I tell the language experts how to write??!!

    One piece of info that may be helpful if they ask you about writing. Linguistics paper don’t often use a straightforward thesis statement/research question. It causes too many fights before the writer has had the chance to build an argument. My boss didn’t believe me until I showed him a whole slew of publications, not one of which had anything resembling a scientifically-constructed thesis. The writing style is a blend of ethnography and philosophy. Very weird, but they are a contentious crowd.

Comments are closed.