Linguistics and language-specific departments in general are at odds.

We need a better set of terms.

Transformational linguistics is a variety of formal linguistics. It says that there is some formal mechanisms that apply in and between "levels" of language--phonetics, phonology, morphology, syntax, semantics, pragmatics. Where to draw the lines, how many lines to draw, how complicated to make the mechanisms ... They vary. A lot.

Functional linguistics is more of the nature "language serves a function, so the actual processes and how language works depends on the function language serves in society and for communication *and* on how the brain and other parts of the speech apparatus work."

Formal linguistics is more likely to be mathy and try to be more data-driven than functional. This statement will cause functional linguists to try to track down my real identity so they can poison me, family, and my pets. However, in my experience formalists try to make a single principle or mechanism account not just for the data that they observe and fail to observe but to make predictions about what kinds of data will not be observed, and they try to produce the most compact, internally consistent mechanism that handles the widest range of data (and non-data). This can produce wildly insane analyses, but so be it. This was my training. I think that humans tend to systematize their language. I think they do a crappy job of it.

Functional linguistics, in my experience, is usually less mathy. While often data-driven, it has no trouble excluding data from an analysis or allowing the data sets to be split up and handled by different phenomena. It has no trouble with exceptions. It's okay with looking at metaphor as a driving force, or looking at how the brain processes information and trying to go from the fairly fuzzy knowledge we have about human cognition to how language is structured. I think that this drives a lot of language change, but can't account for the limits on how languages change. Language isn't nearly as neatly systematized as formal theories like; nor is it as open-ended as functional theories would like. (There, now everybody hates me.)

There's historical linguistics (oops ... "language change" is the most recent buzzword I know of for this). They often look at how languages change.

There's descriptive linguistics. These people describe and record things. Lexicographers. Language variation folk (whether geographical or social). They may be formal or functional, but ultimately value a good description over a brilliant analysis. Sometimes it's hard to tell them from historical linguists, because both deal with change.

There are anthropological linguists. They look closely at how language functions in society. Often they're better anthropologists than linguists, seldom very formal or functional (in the sense "functional linguistics&quot and most of the linguists in the preceding categories try to stay away from anthropologists. Dialectologists and esp. sociolinguists are exceptions, but even then opinion's decidedly mixed. It depends to what extent you're really into description or like wading into theory.



Language departments include language and literature. But a language at the university level may have a lot of dialects and varieties and a history, but it also has one (or several) literary norms with a literature. That's what language departments tend to do, lots of literature. They're also required to offer service courses, and to uphold, at some level, literary norms as a mode of educated communication, esp. in writing. Large ones allow for a lot of non-canonical literature or an expanded canon and tend to have specialists in specific varieties of languages, or focus on how a "literary language" is best described. (There's a whole set of theories on what a "literary language" is; sadly, they're not widely known in the US, and that allows new PhDs to re-invent as completely new stuff known abroad generations ago.)

Language departments tend to have perhaps a language variation linguist (historical, dialectal, social) or a pragmatist on board. They're often less theoretical, i.e., formal and functional, and much more descriptive. This makes sense because typically you need to know the language well to do this kind of work--it's not data analysis focused on a single phenomenon ("antecedents of reflexives in subordinated infinitival clauses&quot but on how language varies across or between communities or over time. You need to know the language, and know it very well. Language depts. are also good for quirky theories--so in Slavic there are specifically Russian or Czech varieties of linguistics (less now than there used to be), and that's where you'd find those kinds of linguistics. There are W. European varieties. Etc.


"Tension" is entirely the wrong word for the relations between some groups of linguists. It's like comparing Jew-killing Hamasites in Gaza with some Greater-Israel Kahanist that guns down Muslims at worship ... but worse, and that's on a good day. Often the poor grad student who fails to recognize how strident and deep-seated the animosity is between these groups will be sliced-and-diced, either at funding committee meetings or in their orals for crossing the wrong line. There are hecklers: I knew one language-specific linguist who delighted in going to formal linguistics talks just to provide a continuous stream of counter examples that undermined the speaker.

There's academic ill-will. So back in the early '60s an observation was made and picked up. It had no great formal or functional explanation. A grad student, CC, used it at the time in a conference paper that was published. A young professor (let's call him Noam C.) and one of his first grad students, B., were in the audience. They were all faculty or grad students in Cambridge. While CC had said it was a common observation and said she had no idea who innovated it--her advisor, RJ, had said it was known for years--her work was cited as the earliest reference in print. Twenty-five years later, Noam C. read a draft paper by B. in which he mentioned this insight and that it had found an explanation. At some point, it had grown explanations. Noam dubbed it "B's generalization" and it was then standard practice to cite B's paper from the late '80s, I believe, as the first mention of this important insight, and his generalization was all the rage for a while. It was B's real claim to fame. B, however, saw no reason to search outside formal linguistics work and he was richly rewarded for not doing so; when it was pointed out this was common knowledge when they were in grad school decades before he was dismissive. Was he there? Well, yes: He was one of the discussants for CC's paper, and the written summary he appended to CC's paper in the conference proceedings mentioned this generalization. I wrote a student paper that required I cite B's generalization and I cited CC instead, and mentioned that B later claimed credit for it. My professor was not pleased. Formalists don't like citing functionalists. Functionalists don't like citing formalists. And you don't ever point out the hypocrisy involved in this extreme tribalism.

There are tensions between general linguistics depts. and language-specific linguistics departments. For example, a professor I know gave a talk outlining a possible analysis for a linguistics problem of long standing. Another one in the audience pointed out a problem with it. Both went off irritated with the other. Both hit upon the same fix to the analysis within a week or so. Both wrote up their papers. Both submitted them to different journals the same month. Both were published, one month apart (because of journal turn-around time). Neither mentioned the other. Both became aware of each other's paper, and both claimed that the entire analysis was entirely their own. They didn't say more than a strained "hello" for years. (I left their social circles 12 years after the articles were published, and if they've mended fences in the 25 years since I moved I don't know. They both might be dead for all I know.)

I hardly know enough about linguistics to understand your reply, but i appreciate it nonetheless.

I have some background in math; no doubt that is why I am attracted to formal linguistics. I find the diagrams linguists and computer scientists draw more convincing than what I was taught in junior high school about how to "diagram" a sentence.

In fact, most of the public school curriculum in English grammar struck me as useless. The prescribed usage of "shall" vs. "will" was just plain wrong. Any exceptions to the rules were dismissed as "Idioms". When I pointed out that there were whole families of such "idioms", I was considered a troublemaker. The word "ago" is like a preposition, but it comes AFTER its object. This observation also was taboo, since the parts of speech presented to us did not include postpositionals. Particles were mistaken for adverbs for the same reason. Even though half of the verbs in colloquial speech are phrasal verbs, this topic was never mentioned.

I have taken quite a few English classes in my lifetime, but only one linguistics class, which was supposed to be on English syntax, although we spent a lot of time on the syntax of other languages, most of which were not even Indo-European. Of course, I had not taken the prerequisites for this class. I wanted to skip the elementary survey course and jump right into something interesting.

I find it interesting that linguistics grew out of English, as psychology grew out of philosophy, statistics out of math, etc. Computer science is unusual in that it grew out of math at some schools and out of engineering at other schools.

Computer scientists draw tree-like diagrams to parse statements in "languages" such as fortran, just as linguists do for natural languages. Syntax can be ambiguous, i.e., there can be more than one tree that fits a sentence, or even a single word (e.g., "unbuttonable&quot . This phenomenon is verboten in computer science but unavoidable in linguistics. That's because natural languages are messy and don't fit any set of rules very well. (I think you pointed that out already.)

English departments are Liberal Arts departments teaching proper use of the English language, English and American literature, and literary appreciation/interpretation.

Linguistics departments take scientific approaches to the study of the elements and dynamics of language in general, as well as in specific languages.

As an English professor, I can tell you that the two disciplines rarely cross and there is rarely interaction between them, much less tension. If there is interaction, it is usually between linguistics scholars interested in literature or literature scholars interested in linguistics. Those people usually get along in both departments.

and you disagree. Perhaps that's because you and he have had different experiences.

Since I have no expertise in either discipline, I try to keep an open mind.

Thanks for your input.