I believe in descriptivism in English. If I understand it right, this means that you start with the entire corpus of the English language, and then determine the rules of grammar based on that. Am I wrong about this? Is there a better way of saying it?
I have no doubt that the reduction of the entirety of English to a single volume of syntax and morphology involves a great deal of work and skill. I am not asking about any details of this. What I would like to know is where linguists get the entirety of English to begin with. Do they read newspapers and magazines and books, and listen to radio and movies? Do they read personal communications and letters and emails, and listen to phone calls and public speeches and conversations? Anything else? And once they have all that, when two sources conflict, how do they distinguish mistakes from correct English?
As a bonus question, is there much disagreement in the linguistic community? Are there large controversies? Or are the rules agreed on, for the most part?
The debate between descriptivism and prescriptivism is long-running (perhaps eternal). Most grammarians are neither strictly prescriptivist nor strictly descriptivist, but fall somewhere along a spectrum between the two extremes:
I’m quite happy for the struggle between prescriptivism and descriptivism to continue, as allowing one or the other camp alone to define the language would lead to too much order in one case and too much chaos in the other. Sticking to the written rules until the weight of popular usage forces change is a reasonable middle ground, and the squabbling over when that point has been reached keeps things interesting. Squabble on, say I.
> As a bonus question, is there much disagreement in the
> linguistic community? Are there large controversies? Or are the
> rules agreed on, for the most part?
There’s two different questions that you could be asking here. One is whether when two linguists study English they always come up with the same rules or they find different rules. Well, yes, of course they can come up with different rules. Those are called dialects. There are many dialects of English and each of them has a slightly different description. But that sort of thing is expected by linguists, so they just note the location, social class, date of birth, race, etc. of the speaker who used some particular rule. Linguists know perfectly well that any language with a large number of speakers will have many dialects, so they try as well as possible in their descriptions of English to note where and when each particular rule of grammar, morphology, phonology, etc. is used.
The second possible question you could be asking is whether, even if you were describing a single dialect of English, there would be complete agreement about what the rules should look like. No, there isn’t. There are a number of schools and subschools of linguistics, and they differ about how the rules of a language should be formulated. I was going to give a link to a site that explains the various schools, but I couldn’t find one. I’m sorry about this, but all I can say is that you would need to read some linguistics textbooks to get a more complete description of the main schools of linguistics.
You also ask about what sources linguists use for determining the rules of grammar, etc. of a language. It depends. A lot of work on English was done by using linguists’ “native intuition” (i.e., they asked themselves whether a particular sentence was grammatical, since they were native speakers). There has been some work with large corpuses of text and speech. The early twentieth century linguists did a lot of the creation of grammars and dictionaries of obscure languages by living with the people for a year or two and recording as much about the language as they could in that time. Obviously, much of that work (even though it’s often all that’s available about a particular language, which may now have disappeared) is very sketchy.
Incidentally, by schools of linguistics, I’m not talking about the difference between prescriptivism and descriptivism. Linguists don’t even think about whether they should try to prescribe usage in a language. It’s hard enough to come up with a description of all the languages and dialects. The view of most linguists is roughly expressible as "Look, I’ve got my hands full trying to come up with a description for language. If you want to prescribe some particular use, don’t bother me about it. I’m not interested in that sort of thing, and I can’t offer you any basis for why any particular rule of grammar should be prescribed.
Prescriptive linguist is an oxymoron since a linguist’s task is to record, analyze, theorize, etc., about language and languages. There is just no place there for being prescriptive; that is the work of editors, English teachers and others. There are many schools within that group. Some see their job as recording dying languages in as much detail as possible. Others see their job as analyzing one language (often, not always, their own native language) minutely. Others may analyze classes of languages (such as languages whose fundamental word order is subject, verb, object, the so-called SVO languages) and make pronouncements such as, “In general, in SVO languages, adjectives follow nouns.” (Yes, there are exceptions, but this is the general tendencies.)
To get back to the OP, linguists who analyze English in great detail are often working at the margins with sentences that are not clearly grammatical and not clearly ungrammatical and they often find disagreement over judgments of grammaticality. There is considerable evidence that nowhere in our brain is a grammaticality tester and that the way we make such judgments is to try to parse it and if we succeed, we judge it grammatical and if we fail we say it is not. There is a well-known sentence that I once stared at for a day or so before I suddenly parsed it and now I cannot see what the problem was, while my wife parsed it immediately:
The horse raced past the barn fell.
One problem with answering the OP is the confusion between “grammar” and “usage.” These terms are continually misused, as can be seen by looking at almost any “grammar” thread on this board. the vast majority of them are about questions of usage.
Grammar is the structure of the language, the bones that hold it up - tense, gender, subject verb agreement and all those thorny points you learn in school. These change rarely. English grammar has stayed almost entirely fixed since modern English came into being.
Usage concerns connotation and denotation, the meanings and implications of words in speech and writing. Usage changes almost daily as new words get coined, old words and expression accrue new meanings, and solecisms get accepted into the language. Usage also has the gradients of formal speech or writing, colloquial, slang, informal, dialect, and substandard. Correct usage in formal speech is what the prescriptivists are normally referring to, but they also extend this incorrectly down to colloquial and informal speech making their arguments often seem unreal or nitpicky.
BTW, only the class I call the “illiterate pedants” still makes claims like “You can’t end a sentence with a preposition” or “Never split an infinitive.” Even the proscriptivist usage mavens stopped saying anything like that 100 years ago.
Part of the problem with citing style manuals is that by definition they’re seeking to impose uniformity, in many cases regardless of whether that uniformity is really needed or not. Remember, organizations like the BBC, the New York Times and the AP want to impose a common “look and feel” on what they publish, so the style manual as as much about marketing as it is about anything else. My far, far smaller employer has its own jealously guarded style book, which we see as the very core of our success. And it encourages begining sentences with conjunctions, because that construction is informal and friendly - in contrast to the more academic-sounding alternatives.
Other style manuals, like the Chicago or the Blue Book, are designed for fairly narrow audiences - academia for the former and lawyers for the latter. Their reasoning is often sensible, but ultimately not terribly relevant if your audience is significantly different. Fowler and Strunk & White (at least in their earlier editions) tend to be even more interesting and entertaining, but since individual authors inevitably have grammatical or stylistic hobbyhorses they, too, must be taken with quite a bit of salt.
So I think we’re best to dispense with any sort of Platonic ideal for language, despite the efforts of the Academie Francaise or the Academia Real Espanola. The trouble then becomes one of What Do We Tell the Children? That’s really where these wars come to the fore: everyone gets terribly upset when Teachers, the Sacred Guardians of our Children’s SAT Scores, use nonstandard grammar.
In those cases, I’m usually with the appalled parents. Academic English is privileged, so failing to teach it to our children often consigns them to lower status jobs. I’d like to see a lot more resources put into teaching academic English as a second language - my understanding is that programs using these methods have found some success, especially if they get kids in the early grades. This effort doesn’t usually require coming up with hard and fast answers to questions about multiple possessives, either.
Wendell Wagner: agreed. I think the nearest a descriptivist would come to saying something is “correct” would be in terms of a statistically dominant usage.
Thanks everyone for your replies. I’m beginning to get a better understanding. The first link, to the Bank of English, answered my main question, of what’s included in a Corpus Linguistics:
I’m still curious, though, are all these things given equal weight? I imagine that it’s up to the individual linguist to decide this. Or do they develop rules based on each medium, e.g., “This is what you say when you write a brochure, this is what you say in an interview, etc.”
Also, something that prompted this thread was hazel-rah’s comments in a previous thread. He or she seems to be saying that there are linguistically right and wrong ways of saying things.
If I understand what hazel-rah is saying correctly, he means that there are things that no one would ever say in English. Just because there are lots of dialects of English doesn’t mean that anything whatsoever will be accepted as a sentence in some dialect of English. Though the rules of the various dialects of English differ, they don’t differ that much. On any one grammatical form of English, we expect to find a small range of different possible rules.
Incidentally, I think you misunderstand something when you say, “. . . what’s included in a Corpus Linguistics . . .” A corpus of the language X is a large set of texts and transcriptions from speakers and writers of language X. Corpus Linguistics is the name for a branch of linguistics. It’s the study of languages using corpuses of languages. This is a relatively new thing. Most work formerly was done using native intuition and small samples of a language.
Linguists know that spoken and written language are different in some ways and are willing to say that some rules only hold in spoken language and some only hold in written language.
As Wendell Wagner says, there are usages that differ, but have the same underlying grammar. English sentences, for instance, are constructed as Subject-Verb-Object: in RP, say, “You were ill”, or in Estuary “You was ill”. But “You ill were” or “You ill was” would be incorrect grammar for English (unless in the dialect by Yoda spoken, hmm).
Part of the confusion is that people often call things “bad grammar” when they really mean non-standard or dialect usage. In reality, no native speakers use bad grammar.