I've been pondering language lately. Big surprise! Specifically, I've been wondering how normative grammars are created. Because we are not endowed at birth with a language, we must learn it from other speaking people in our environment. The language, we learn while sitting on our mother's lap, is rather different from the standard we are taught in school. Normative grammarians denigrate empirical linguists as simply recording speech without judging what is good or bad. My question is how do normative grammarians determine what is grammatically correct from what is not.

At first blush, it seems that traditional grammarians rely on criteria such as authority (great writers and speakers), logic, tradition (history), taste, and plain old fiat. But authorities can (and ought to) be ignored if they made a mistake. (How to tell if something is a mistake is left to the authority of the grammarian doing the correcting.) Logic is a dangerous thing to being into contact with grammar. The former admits of no exceptions while the latter is, at times, nothing but. Tradition seems to mean the period in which the grammarian grew up, archaisms and neologisms are simply bad. Taste does not seem like a good criterion, but as many are quick to point out de gustibus non est disputandum. This leaves us with fiat: it's wrong because I say it's wrong.

What are your criteria for determining grammaticality and can they be applied consistently and objectively?

Currently I like Merriam-Webster's Concise Dictionary of English Usage. They seem good at relying on usage only, not on logic or preference. They also describe the registers that the usage is more commonly found in, for instance if the usage is commonly found in fiction as opposed to sports writing or newspapers, they say so.

I don't know if a normative grammar can be consistent, since different registers and dialects have different usages.

Currently I like Merriam-Webster's Concise Dictionary of English Usage.

I, too, think it's the best current dictionary of usage, but others may not.

I don't know that source, but I will look at it.

I like the Chicago Manual of Style because it allows for various uses, unlike something like Strunk and White. I have to admit, though, that I used to be a prescriptivist, and I am probably not as consistent with my views as I should be. As with language, my views are evolving.

The nice thing about Merriam-Webster's Concise Dictionary of English Usage is that it gives detailed historical details about usages, especially where and when prescriptivists have gone awry. I've given some excerpts from it on this board when discussing certain usages. It always amazes and amuses me when I discover that some grammar maven, usually in medias rant, doesn't own or hasn't read a single English grammar. Three of my favorites are: (1) Otto Jespersen's A Modern English Grammar (1909ff., in 7 volumes), (2) Quirk, Greenbaum, Leech, and Svartvik's A Comprehensive Grammar Of The English Language (1985), and (3) Huddleston and Pullum's The Cambridge Grammar of the English Language (2002).

rotflmao at 'in medias rant', and trying to decide whether it should have been italicized as a foreign expression. Nice pun.