Saturday, August 04, 2007

Public perceptions of linguistics

Much linguistic blogging focuses on public perceptions and representations of our field. That was certainly part of what got me into the game from the very beginning (here). Everybody from Language Log on down deals with it in various ways and the public side of the equation ranges from the BBC and professional mavens to the Comedy Channel. The punchline is usually that people feel free to publish about linguistics and language in the highest-profile place without knowing their larynx from a hole in the ground. A major theme is profoundly bad presentation of science (see here for one account of why we get so much of it), but often it's funny (here) and more often it's sad and stupid (here).

One of the reasons I started reading Polyglot Conspiracy regularly was precisely her perspective on this topic (e.g. here). Now PC has tackled one of the worst and ugliest examples, here, about the full cycle of stupidity from ignorant journalism to really ignorant blog posts, all about the work of Mary Bucholtz. (And don't miss the Language Log post by Mark Liberman if you haven't seen it.) If you haven't read the threads on this, do — I won't rehash them all here. The punch line is that linguistics is useless and unscientific (compared to what? Economics? String theory? Creation science?).

At the heart of this, as PC notes (in part implicitly), is a profound anti-intellectualism, and it's one our society suffers from at the very highest levels. People who don't believe in evolution shouldn't have access to medicines that are effective against new strains of infection. (Can't find the cartoon illustrating that.) Speech and hearing science (including stuff like cochlear implants), speech recognition and synthesis, other computational work, forensic linguistics, language teaching/learning, and so on is all done in collaboration with people from other fields, but linguistics is central to tons of it. Sure, you might say, but Bucholtz's work on language and identity doesn't involve any of that. Not true, actually (her CV shows that Bucholtz has done forensic work, like many sociolinguists), but I think the real issue for a lot of these people is that they don't understand the value of basic research generally — they're interested in engineering, not science. Part of the value of blogs and such is the chance to fight the kind of batter PC, LL and many others are taking up. But it might be a long one. A New Yorker cartoon has a judge talking to a humanoid-looking creature:
At your current rate of evolution, I can see no other choice but to give you ten to twenty million years.
Image from here.


Anonymous said...

One aspect of this whole discussion is the role of blogs and 'other new media' vis-à-vis the 'mainstream media': Over at LL, Liberman expressly puts his money on the former, as a better opportunity for getting good information out than science journalism.

In that regard, it's interesting to hear all the buzz about the Yearly Kos convention: This gathering of bloggers is where the real action is, etc., blah, blah. For linguistics, we've begun to achieve that, surely: Language Log alone gets 1,000s of hits per day and people into language/lx have mostly learned that it's the place to keep up on the topic.

The Ridger, FCD said...

Here's a link to Doonesbury you mentioned.

Frank Oswalt said...

I have pointed this out over at PC, but let me reiterate it here: I'm sure PC is right about the anti-intellectual undercurrent in the discussions she traces, but academics like Bucholtz are partly to blame for this.

PC assumes -- presumably correctly -- that most of the people discussing Bucholtz' ideas have not read her papers. Well, since they are all available on her web page and I had nothing better to do at the time, I have read them. So what did I find? Well, I found a lot of jargon and undefined uses of everyday words like "nerd" and "geek" and "white", as though these categories are a given and do not need to be discussed and operationalized. I also found a large number of assertions that were offered without any evidence whatsoever, sometimes backed up with quasi-anecdotal plausibility arguments (for example, the fact that "people" call Bill Gates a "nerd" shows that the word has negative connotations). The closest thing to actual evidence for her claims that Bucholtz offers is a number of passages from interviews with teenagers (less than ten different subjects in total) that she interprets to support a subset of her claims. Needless to say she does not lay out the criteria according to which her interpretation proceeds, nor is there any way of assessing how representative her interview partners or, indeed, the passages she selected from the interviews are. Oh yes, and then she presents a "Table" with the "scores of vowel fronting" for three of her subjects. This borders on cargo-cult science, as a results based on three subjects cannot possibly be meaningfully quantified, and she does not say what measurements were used to arrive at these scores, how the data were collected, or how the results differ from those of a control group. This complete lack of any of the characteristics of real research does not stop Bucholtz from referring to the passage containing this table as a "detailing" of the "phonological practices" of nerds.

I have never heard of Bucholtz before reading about her research in the NYT and on PC and although I'm, among other things, a linguist by training, I have never worked in academia, so I have no personal, professional or theoretical axe to grind with her. She may or may not be right in what she says. The point is, she certainly does not offer any operationalizations, methods or data that would allow an impartial reader to decide whether her ideas have any basis in fact or whether they are just plain stupid (having read three of her papers on the topic, I would lean towards the latter).

It is all good and well to chastise journalists for sloppy reporting and to chastise the public for relying on sloppy journalism rather than the "real science" behind it. However, in Bucholtz' case there is no real science behind the sloppy journalism and since she is hardly alone in this, esp. in the social sciences and the humanities, perhaps there is some justification in the public perception of academics.

Mr. Verb said...

Anybody who raises issues of identity is "partly to blame" for negative reactions, I suppose: It's tremendously controversial territory and the nature of that work is in frameworks that aren't yet very developed, surely. I haven't read much of her work, but remember it being very ethnographic, not quantitative at all. Beyond that, I need to read and reread her papers before I could comment.

Rosina said...

I've been trying to stay away from the blogosphere in order to concentrate on work, and see what comes of such resolutions. Somehow I missed Polyglot Conspiracy.

A couple points: 1) I haven't been following the MB story, and I haven't kept up with her work, so like Mr. Verb, I can't comment directly on the subject at hand.

However, I did train as a sociolinguist and I have to acknowledge that Frank is right. Sociolinguistics was, initially, very statistically oriented. SLs took sampling seriously, and knew what to do with an outlier (examine closely, it usually means something). Then in the usual way of things, the pendulum began to swing in the other direction and now there's a lot more anthropological-type linguistics out there under the Sociolinguistic banner. The thing is, linguistics and SL are as much a science as psychology and sociology -- basically, not a science in the strictest definition of the word. There is no way to quantify human behavior. Having said that, socio linguists need to adhere as closely as they can to scientific methods and be very careful when making generalizations.

If I have time to go and look at Mary's work, I'll do that. I've been out of the field for ten years and I am not so familiar with the current trends as I should be. Especially as CUP wants me to do a second edition of English with an Accent.

pc said...

Hey Verb, thanks for the links and the discussion. I am actually not 100% certain that "anti-intellectualism" is the right term for what's driving these kinds of comments, but my doubt is highly speculative (no numbers to back it up - so maybe invalid, eh?). The thing is, I'm guessing that most of the people whose comments I quoted in that post - most of the people reading Metafilter, Protein Wisdom, Sepia Mutiny, etc. - are actually very pro-intellectualism insofar as that means pro- having thoughs and being interested in things and being critical. They probably wouldn't be reading such blogs, much less commenting on them, if they weren't (although they might not call what they're doing a form of intellectualizing, since that has connotations of pretentiousness [which come to think of it probably supports your claim that we're dealing with "anti-intellectualism"! dagnabbit]). But this breaks down, for some reason, when they see people with institutional affiliations publishing and getting paid for work that challenges a) their beliefs about the world, society, culture; b) their beliefs about what constitutes professional or scientific or "intellectual" research. I think your comment about how people want engineering is spot-on; it seems to me like more that there's a current of anti- a specific kind of (faux-)intellectual product. Namely, one that doesn't have hard numbers to back up its claims, one that doesn't seem to "prove" anything or can't itself be "proven," one that deals with hard-to-operationalize concepts.

I DO think that in this country we have a widespread problem with anti-[public intellectual discourse], evident in the fact that reporters can't ask our president direct questions about serious issues and expect either an honest, thoughtful, satisfying, or not-snarky response. So when an academic publishes something provocative about a touchy subject - race, for instance - and dealing with their claims requires thought and conversation and a willingness to let go of some dearly-held presuppositions about The Way Things Are...that's when people seem to forget about the value of research in general in *challenging* us to think about things, not just telling us How Things Are, and when they stop believing that the intellectualism holds any value.

Or something. I don't know. Maybe it is all just anti-intellectualism, just flying under the radar.

A defense of Bucholtz's work itself is *really* not my concern here; it just provides a convenient point of entry to a larger discussion. But to respond briefly to Frank: you say that there is nothing solid to back up her analysis and that she doesn't have enough subjects to make claims about what they're doing, and that there's no way to know if she's "right" or "wrong." I think it's a mistake to think this is something subject to such judgments. This work is ethnographic and interpretive. Arguably, this is the necessary manner in which "meaning" or "social meaning" has to be studied. Social meaning does not come in neat packages that we can count the contents of (though unfortunately, I admit that sometimes academic work makes it seem like it does! because we have to package and term everything somehow, in the end). What a way of speaking means to a community or individual is not going to be revealed by a set of variant distributions. This kind of sociolinguistics is not really about proving things so much as trying to understand things at an initially local level, and then perhaps identify and clarify common processes by which things happen or come to be (as regards language, society, the individual, and culture). [Maybe people should look at that link, if they'd like a very brief way to grasp the goals/methods of this strain of research and its relation to the history of socio/linguistics/anthropology...and follow the links therein for more info.]

pc said...

Oh also I wanted to include something about how thinking that there's only one right way (a quantitative, "objectivity"-attempting, or proof-driven one) of investigating the world is a very unhealthy attitude for research in general. There are lots of ways to go about understanding things...hence different disciplines, different subfields, different theories, the requisite "methodology" section in every academic article. Oh and change. Change comes about because of trying new ways of looking, and very often good things emerge from that.

Frank Oswalt said...

PC, you say that

“This kind of sociolinguistics is not really about proving things so much as trying to understand things at an initially local level, and then perhaps identify and clarify common processes by which things happen or come to be.”

I can see the value of an exploratory and largely speculative approach to the subject in the initial phase of a research project. We need to come up with a set of hypotheses about the object of research before we can begin serious investigation. However, in much of the current sociolinguistic and anthropological research, that first phase is all we really see. Many researchers jump straight from initial exploration to jargon-driven theorizing. I think that is a real shame, as there are some really interesting research questions involved.

And you say “thinking that there's only one right way (a quantitative, "objectivity"-attempting, or proof-driven one) of investigating the world is a very unhealthy attitude for research in general”.

You are partly right about the quantitative part -- if you phrase your hypotheses in the right way, you don't always need quantitative evidence (think about grammaticality judgments -- they are not quantitative, but if you treat them carefully, they do constitute evidence). As soon as we're dealing with more complex aspects of human behavior, however, I do think that we need quantifyability and representativity. And I would strongly disagree with respect to your statements about “objectivity“ and “proof-drivenness“. These are an indispensable part of any research endeavor. If you give up these goals, you shouldn't be surprised if journalists and the general public stop taking you serious.

And: “Change comes about because of trying new ways of looking, and very often good things emerge from that.”

Absolutely. Karl Popper introduced a new way of looking, called science. Many good things have emerged from it.

Rosina, “Especially as CUP wants me to do a second edition of English with an Accent.”

That would be great. It's one of the few books I've held onto from my days in academia.