Although Ellen Goodman is often correct in her diagnoses of national trends, she has developed the irritating habit of making sweeping, cosmic generalizations from flimsy and highly subjective evidence. Thus, in "Speaking With an American Accent (op-ed, Dec. 16)," she concludes, after listening to a few voices from around the country on a late-night radio call-in show, that we Americans are losing our regional accents and that our language is becoming homogenized into a bland "National Speak." She attributes this phenomenon to increased mobility and the pervasiveness of the media. I think she is wrong.

While it is true that Americans move from one region of the country to another more frequently than they did in the past, no strong body of evidence indicates that in so doing large numbers of them lose their native regional accents. Even if a few of these transient Americans did modify their speech habits, they constitute too small a percentage of the entire population to be of much significance. As to the impact of television, while it is true that all the performers and journalists on the tube speak with a Middle American accent, there again is no evidence leading to the conclusion that the millions who watch television unconsciously lapse into a monkey-see, monkey-do syndrome so far as speech accents go. In my own frequent travels around the country, I have detected no individuals in dialect regions taking on the dulcet tones of TV anchormen.

Goodman's argument is based solely on historical linguistic phenomena. But other factors contribute to the diversity of the American language. Class, for example. There are still a lot of patrician types around--Averell Harriman, Claiborne Pell and George Plimpton, to name a few-- who speak with an inflection far removed from General American. And the language that Goodman would hear in a blue-collar bar is wonderfully colorful, but it ain't Radcliffe-Harvard prose. A few other pockets of linguistic diversity seem to have escaped Goodman's attention:

Black street lingo: filled with interesting grammatical inversions and imaginative folk metaphors.

Psycho-babble: the language of California sensitivity cults.

Jocko-Macho talk: "tough" male locker-room speech, typified by the Nixon tapes; a reaction against Women's Lib and the niceties of conventional English.

Academese: characterized by constant use of irony and paradox that is so totally removed from the language of the general population as to resemble a foreign tongue.

Teen-age talk: unique by its very inarticulateness.

There are scores of these separate "languages" in our culture; and I would urge Goodman to read Raymond Queneau's "Exercises in Style" in which one anecdote is told in 99 distinct prose styles. But one doesn't need to be French to pull off such a tour de force.

I fear that as Goodman reaches the status of a pundit she is relying too much on her intuition, not enough on the reality that is right under her nose. She could profit from taking a linguistic tour of America. A good place for her to start would be western Maine; from there she should plan to roam around the heart of Brooklyn; then on to the middle of Appalachia followed by a sojourn to the Deep South. If, after such a journey, she is still convinced that our speech is becoming homogenized, then I will abandon my current project of telling one anecdote in 99 distinct American styles and accents.