I happen to have drafted the following before John McIntyre posted a typically trenchant piece about rules on his blog. I hope the following will act as a complement (and as a compliment).
The term grammar rules is freely used and often without much thought about what it might mean. Typically it is used as a kind of guide to linguistic good manners. As Harry Ritchie writes in ‘English for the Natives’, grammar itself is assumed to be:
. . . a weird combination of finicky word usage and obscure social etiquette, like knowing how to address a viscount or where to place the sorbet spoons. The whole nebulous subject presided over by stern, scary men, who write books telling us we always get things wrong: ‘One should, of course, say “It is I, your viscountness”. All other forms are grievous errors. Sorbet spoons to the immediate left of fish-knives.’
Those who take such a view are reluctant to tell us where the rules come from, presumably because they don’t know. Those of us who have taken just a little more than cursory glance at the subject know that grammar rules describe how a language works, much as the laws of physics describe how the universe works. Newton didn’t make it up when he claimed that:
An object at rest stays at rest and an object in motion stays in motion with the same speed and in the same direction unless acted upon by an unbalanced force.
He based his statement on what he had observed. Similarly, if we notice that verbs like might, shall and can behave in a certain way, we can conclude certain things about them: they are are invariable, they form their interrogatives by inversion, and they are followed by an infinitive which is not preceded by the particle to. If we see that a word that takes its meaning from a second word in the same sentence cannot come before that second word if that second word is inside a subordinate clause, we can conclude, as the linguist Lanacker did in 1969, that there is a rule of English grammar that states that an anaphor may not both precede and command its antecedent.
By contrast, there is no rule that prevents us from ending sentences with a preposition, using they to refer to just one person, placing an adjunct between to and the verb or using coordinated I as the complement of a preposition. Native English speakers do these things all the time, and have done so for centuries. To say that they speak or write ungrammatically in doing so is as ludicrous as if Newton had said that an object at rest SHOULDN’T stay at rest and an object in motion SHOULDN’T stay in motion with the same speed and in the same direction unless acted upon by an unbalanced force.