I recently objected to Louis Menand's assertion that "[P]rofessional linguists almost universally, do not believe that any naturally occurring changes in the language can be bad" ("Menand on linguistic morality", 10/22/2008). And I was quickly taken to task in the comments by Steve Dodson, who is the erudite and broad-minded author of the Language Hat blog. Hat (as he's called in the blogosphere) asserted that
I personally am happy to sign on to the Descriptivist position as "caricatured" and state that there is no such thing as bad language change. […] To say any form of language change is "bad" is to be ipso facto unscientific.
He also suggested that my acquaintances and I belong to "a small sample of linguists who have … weirdly quasi-prescriptivist views about language".
Despite these bracing sentiments, I think that Hat and I are not very far apart on this issue, in the end. But before we pass the peace pipe around, I'd like to take up the question of what bad means, and what language means, and to whom.
Prof. Menand clearly means bad in the OED's sense 2, "Lacking good or favourable qualities; unfavourable, unfortunate, untoward; that one does not like; not such as to be hoped for or desired". And by language he means not just the structure of spoken language, but the whole culture of communication, as indicated by these passages from his essay:
[Crystal's] conclusions are predictable: texting is not corrupting the language; people who send text messages that use emoticons, initialisms (“g2g,” “lol”), and other shorthands generally know how to spell perfectly well; and the history of language is filled with analogous examples of nonstandard usage. […]
A less obvious attraction of texting is that it uses a telephone to avoid what many people dread about face-to-face exchanges, and even about telephones—having to have a real, unscripted conversation. People don’t like to have to perform the amount of self-presentation that is required in a personal encounter. They don’t want to deal with the facial expressions, the body language, the obligation to be witty or interesting. They just want to say “flt is lte.” Texting is so formulaic that it is nearly anonymous. There is no penalty for using catchphrases, because that is the accepted glossary of texting. C. K. Ogden’s “Basic English” had a vocabulary of eight hundred and fifty words. Most texters probably make do with far fewer than that. And there is no penalty for abruptness in a text message. Shortest said, best said. […]
[T]exting has accelerated a tendency toward the Englishing of world languages. Under the constraints of the numeric-keypad technology, English has some advantages. The average English word has only five letters; the average Inuit word, for example, has fourteen. English has relatively few characters; Ethiopian has three hundred and forty-five symbols, which do not fit on most keypads. English rarely uses diacritical marks, and it is not heavily inflected. Languages with diacritical marks, such as Czech, almost always drop them in text messages. Portuguese texters often substitute “m” for the tilde. Some Chinese texters use Pinyin—that is, the practice of writing Chinese words using the Roman alphabet.
In this sense of bad and this sense of language, there's no question in my mind: most linguists, even when they're wearing their professional hats, believe that some real or hypothetical changes in "language" are or would be "bad". For example, the web page for the LSA's Committee on Endangered Languages and their Preservation clearly presupposes that language death is a Bad Thing. I know many linguists who work on literacy projects, and I'm confident that most others believe that illiteracy in modern societies is generally a Bad Thing. Linguists who work on discourse coherence universally believe that essays, narratives and conversations can be more coherent or less coherent, and I'm confident that most of them believe that reduced coherence is a Bad Thing.
Of course, that doesn't mean that most linguists would assent to Prof. Menand's particular views about the cultural and stylistic evaluation of texting. Some would, some wouldn't — and most, I believe, would respond in the way that David Crystal did, which is to ask, wait a minute, what are the facts here? That was the main point of my original post on this subject: "we believe that moral and aesthetic judgments about language should be based on facts, not on ignorant and solipsistic gut reactions".
But suppose it really were true, as Menand might be taken to suggest, that pre-adolescent texting causes a pathological inability to manage "the facial expressions, [and] the body language" required to have "real, unscripted conversation" in "face-to-face exchanges". In the face of credible evidence that texting leads to mass autism, it would be absurd to assert that as scientists, we must only describe the epidemic's progress, not evaluate its consequences and consider alternative measures for prevention or remediation. In fact, such an attitude would be not only foolish but also profoundly unscientific.
The thing is, though, examination of the facts shows that no such epidemic is underway. A central point of David Crystal's book was to debunk a whole list of similar sky-is-falling fears about texting, by looking at the facts of the case — and a central point of Louis Menand's review was to undermine Crystal's arguments on the grounds that those professional linguists, you know, they never met a language change they didn't like.
This leads us back to a rhetorical problem in the culture of modern linguistics. When one linguist says to another "Was that good for you?", at least in a professional context, the phrase usually means "was that sentence consistent with your personal grammatical norms?" We don't mean "was it clearly expressed?" or "was it appropriately punctuated for publication in Language?" or "what is your moral or political opinion of the people who characteristically use such expressions?", or "how would you suggest re-phrasing it?"
This is because linguists mostly use language in the narrower sense of "grammatical system" rather than in its various broader senses — "means of communication", "manner or style of expression", "vocabulary or phraseology", etc. There's nothing wrong with this choice, from a scientific point of view — speech really is different from writing, words really are different from gestures, syntax really is different from a set of word-association norms, and so on. But when a non-linguist uses changes in the language to include a decline in "the obligation to be witty or interesting", and uses bad to mean "I regret this, and invite others to join me in bemoaning it", then it's time to step out of the disciplinary box.
Depending on personal and professional opinions, linguists can choose from a large set of scientifically appropriate responses to (say) the claim that texting is making people less witty and interesting: "This is serious if true, and to check whether it's really happening, we should …"; or "OK, witty and interesting is good, but I seriously doubt that texting is actually making people more bland and boring, because …"; or "Actually, what you call 'witty and interesting' is mostly covert hostility, and society is better off with less of it, as is shown by …"; and so on. But among them is not: "We're scientists over here, and so as a matter of principle, we refuse to be bothered with your evaluative attitudes: whatever is, is right."
That's not the view that geophysicists take about climate change, for example, or the view that microbiologists and epidemiologists take about AIDS. Nor, in fact, is it the attitude that linguists take about language death, or literacy, or aphasia, or foreign language teaching. The problem with most so-called prescriptivists is not the presence of evaluative attitudes, but the absence of coherent analysis and accurate description. In some cases, the evaluative attitudes also express regional, class or ethnic prejudice; but again, this just means that they're the wrong evaluative attitudes, not that evaluation itself is out of bounds for scientists.
[Update: As I should have explained, there are senses of language and bad in which Hat's position is tautologically true, for everyone and not just for linguists. If language means "the lexical and grammatical norms of speech community C", and bad means "outside the lexical and grammatical norms of speech community C", then a "language change" could be "bad" only if the lexical and grammatical norms of speech community C changed in such a way as to become outside themselves.
However, this is not what people like Prof. Menand mean by these words. Linguists often focus on the question of whose norms are the basis of evaluation — obviously, for example, "kids these days" often develop ways of talking that are outside the lexical and grammatical norms of some old guy like me, and we need to keep teaching people that this is not, like, the burning of the library of Alexandria, but rather an inevitable side-effect of human creativity and freedom. But an equally common reason for misunderstanding is equivocation about what language is, as discussed above.]