One thing that bugs me about many prescriptivist claims (not all, just many) is that they just don't do much to prove that the prescription on which they are relying is indeed in any meaningful sense authoritative. Even if one accepts that "proper English" is to be defined by The Authorities, one needs to further show that The Authorities indeed frown on the particular usage. Instead, one often (again, not always, but often) gets a mix of the following:
1. Bare assertion: Someone says "forte," in the sense of "strength," as "for-tay." Wrong, someone else says, because the right way of saying it is "fort." But even a prescriptivist has to explain why this common pronunciation violates some particular prescriptive rule. Yet often prescriptivists don't point to any such rule — in this instance, for instance, the person who made the objection didn't point to the rule, and I doubt that she could have pointed to such a rule (judging by the sources I've consulted).
In conversation, of course there's no time to look up the term and to give a more complete argument. Some people, though, avoid that by not correcting other people's pronunciation — whether as a matter of good manners or just as a matter of caution, since there are lots of usage and pronunciation myths out there (and I mean myths that even many prescriptivists, I'd wager, would acknowledge as myths). Unless you're very sure of yourself, there's a good chance that the complaint you're about to make is unfounded even for prescriptivists.
In writing, there often is time to check one's claims, and find support for them. Why not do so? If you're going to argue that someone got something wrong, why not make sure that it is he who is wrong, and also find a source that you can cite supporting your claim? In some situations, the error may be so uncontroversial that just pointing out will lead the reader to realize that an error has been made. (I'm happy to get e-mails pointing to typos and other errors, since then I can correct the post; and the great majority of these messages point to mistakes that are entirely uncontroversially mistakes.) But if there's some controversy, it seems that it would be helpful to figure out for sure if one is right, again by prescriptivist standards.
2. My teacher said so: A commenter writes,
Though I recognize that English is a living therefore mutable language I find that I remain a prescriptivist. I had a strict 7th grade English teacher who would never have permitted, "...to carefully say...". Split infinitives warrented rewrights of entire paragraphs. Perhaps we need to asssign some linguistic norms to "cultural literacy" in addition to correct or acceptable grammar.
The trouble is that not everything your 7th grade English teacher told you is necessarily so. As it happens, many eminently respected usage commentators — such as Fowler, for instance — have challenged the claim that split infinitives are wrong, and have argued that there is no basis for thinking that "no split infinitives" is a legitimate prescription. Perhaps they are the ones who are mistaken; but your 7th grade English teacher's say-so shouldn't be the deciding factor. More broadly, keep in mind that your teacher (1) might have been trying to teach you what she thought was clear or elegant usage, even at the expense of mislabeling as "wrong" usages that were grammatically sound but unclear, (2) might have been oversimplifying complex rules for the purposes of teaching you more effectively, (3) might have been a partisan of one side of the debate, and not told you that the "rules" she was teaching you were controversial, (4) might have been repeating usage myths that were passed along to her by her own teachers, or (5) might have been misremembered by you. Certainly her say-so isn't enough to define "correct or acceptable grammar" or "cultural literacy."
3. I am a language maven and I say so: One commenter, who has firm views that the subjunctive remains obligatory in prescriptive English in a certain context, reports that he is "a language maven with a stack of style manuals on my desk, paid to edit medical and technical literature and to coach students in taking the SAT." His view is that those who disagree with him on this must therefore be "ignoran[t] of simple, straightforward language rules," and is "a challenged speaker of English who writes as if he had never studied, let alone mastered, another language" (or so judged by "the erudite," apparently a category in which the commenter fits). Oddly enough, the commenter does not actually quote the manuals from the stack on his desk.
4. Your style manual is descriptivist bunk: I actually like to look things up in style manuals, and quote them. Thus, for instance, when writing about whether the subjunctive remains obligatory in English, I looked it up in Merriam-Webster's Webster's Dictionary of English Usage. Ah, some object, but that authority is descriptivist. Very well; Fowler reported the subjunctive to be generally dying, and while it is alive for "if ... were clauses expressing a hypothesis," he does not report is as obligatory. Burchfield's third edition of Fowler reports that it has enjoyed a resurgence, but it too does not claim that it is obligatory. The American Heritage does say that "According to traditional rules, you use the subjunctive to describe an occurrence that you have presupposed to be contrary to fact," though this doesn't resolve whether the traditional rules still remain the rules — and, more importantly, it doesn't resolve whether the subjunctive is obligatory in the context in which I used it, where my point was that the presupposed occurrence was not contrary to fact, but rather that its accuracy was controversial. The Columbia Guide isn't clear on this, but it does report that there is "much more divided usage and much more argument about the appropriate usage" on the was/were subjunctive than on the "finite verb" subjunctive ("In conditions contrary to fact, for example, finite verbs such as arrive are rarely put into the subjunctive, except in the most careful Formal English.").
So what we have is disagreement among respected authorities, with at least a good deal of acknowledgment that there is "argument about the appropriate usage." Doubtless the descriptivist sources are the most accepting of a broad range of common usage. But either some prescriptivist sources do at least acknowledge the uncertainty, or the bulk of the authorities has become descriptivist. (Both might also be true.) What basis is there for those who look to the Authorities to simply disregard so many sources? Sure, if all the authorities become descriptivist, then that makes it hard for prescriptivists who care about the authorities. But that's part of the problem with prescriptivism: If you think the prescriptions of the authorities should define the language, you're in trouble when the authorities disagree with that view.
5. But we need rules! Some prescriptivists at this point turn to a general defense of prescriptivism, and of the importance of rules. Very well; at some point, even descriptivists agree with this — rules of grammar are important, and there are plenty of rules that actually do operate in the speech and writing of the great bulk of English speakers. You'll never hear someone saying "I have write to an e-mail" instead of "I have to write an e-mail." There surely is a rule of grammar related to infinitives in play here.
But that there should be rules — and even that there should be rules condemning common usage as well as descriptively highly deviant usage (a matter on prescriptivists and descriptivists do generally disagree) — is not much of an argument in favor of this particular rule. Accepting prescriptivism in general doesn't entail condemning split infinitives as wrong unless one can point so some binding prescription about split infinitives.
6. And we need clarity and precision: Some prescriptivists at time defend prescriptivism on the grounds that certain prescriptivist rules foster clarity and precision, and avoid ambiguity. But, first, not all unclear or imprecise usages are wrong, even if a clear or precise one is better. And, second, many of the often-condemned usages are not at all unclear or imprecise. "To boldly go where no man has gone before" is no less clear or precise than "Boldly to go where no man has gone before." And in fact, sometimes split infinitives are clearer and less ambiguous than the unsplit ones. "To [adverb] [verb]" makes clear that the adverb modifies the verb; in "[adverb] to [verb]" and "to [verb] [adverb]" the adverb might apply to the preceding or following verbs. Consider Garner's example, "she expects to more than double her profits next year," or Fowler's, "modifications intended to better equip successful candidates for careers in India," and see what happens when the "more than" or "better" gets moved outside the infinitive.
7. If you violate these rules, you'll annoy your prescriptivist readers: Finally, some readers argue that splitting infinitives, putting prepositions at the end of sentences, or violating various other supposed prescriptions simply annoys some readers, and makes them think the less of you. This is the argument I most sympathize with, because it's a descriptivist argument: Though a usage may track what many people, and maybe even almost all people, say and write, it may annoy the remainder so much that cautious writers ought to avoid it. And that's true even if the prescription is utter myth — for instance, if, as with the claim that it's wrong to put a preposition at the end of a sentence, virtually no major reference work accepts the prescription (or at least so my research has led me to conclude). So long as the myth is prevalent enough, one should take it into account, an odd analog of the descriptivist claim that once a usage is prevalent enough, we cannot condemn it as "wrong."
Yet while careful writers should think about this — in fact, I warn students whom I'm advising when they use words that arouse hostility, even if I think the hostility is unjustified — this is an argument about what's prudent, not about what's right. That people are hostile to prepositions at the end of sentences, and believe that there's a rule of grammar condemning them, doesn't make such placement of prepositions wrong as a grammatical matter.
So if you're a prescriptivist, I might not be able to persuade you to convert to descriptivism. But at least I hope that some prescriptivists may be convinced to be more careful about the prescriptivist claims they make. If they argue that the rules of good grammar should be set by authorities, they ought to explain which authorities support the rule they're invoking, and why those authorities — as opposed to whatever rival authorities there may be — are the ones that we should see as binding.