One of the first things you learn in Statistics class is that correlation does not imply causation. The field of linguistics has a tagline of its own: “descriptivism yay, prescriptivism nay”. (This is not the way my professor phrased it, but it’s been almost half a decade since I took Linguistics 101, so I may need to take some liberties here.)
Crudely stated, prescriptivism is concerned with how language should be used, the underlying idea being that usages can be correct and incorrect. A prescriptivist might say singular ‘they’, as in “someone forgot their wallet”, is wrong; it should be “someone forgot his or her wallet”. Descriptivism, on the other hand, is concerned with how language is actually used, in the real world, by real speakers. (I’m a huge fan of singular they. I use it all the time.) Prescriptivism has no place in the scientific – therefore objective – study of language, but it does have a place in aforementioned real world.
A digression on prescriptivism: British vs. American English
Take, for example, American English— a variety of English that is very dear to my heart. My only qualm is that I think manoeuvre looks more elegant than maneuver, but that’s about it. I love American colors, American pants, American organizations. Colours, trousers, and organisations are just not the same.
Not everyone on this side of the ocean would agree with me here, though. I’ve had conversations with British English speakers who are convinced that their variety is the superior variety. Some even reject the notion of American English being a variety of English altogether; mention American and British English in the same breath, and you run the risk of someone butting in with, “I’m sorry, did you mean incorrect English and proper English?”
The reason often given in support of this prescriptivist perspective is that American English arose from British English and is therefore somehow a lesser offshoot. American English is the illegitimate child looming in the shadows all the way in the back of the family photo. “Who, that? Oh, that’s… yeah, we kind of try to forget about that one.”
(Interestingly, neither Canadian nor Australian English seems to evoke the same sentiment. Maybe that’s because they’re barely in the picture at all, obliterated by the huge, ancient matriarch taking up the entire frame… or maybe it’s because the bastard kid is currently in the prime of his life, spreading unchecked across the world as the matriarch watches through narrowed eyes.)
But the equation of seniority with superiority is questionable, even more so in a linguistic context. Sure, American English arose from British English— but British English, in turn, is a melting pot of Celtic, Latin, Old Norse, Old Norman French, and I’m pretty sure I’m forgetting a few ancient languages. The reason why we’re not getting any Ancient Celts, Romans, Norsemen, or Old Normans complaining about issues comparable to the pronunciation of can’t as /kænt/ instead of /kɑːnt/ is because they’re all dead.
Back to descriptivism
Anyway, all of the above is a great (if somewhat extensive) example of how not to linguistics. Modern linguistics is about descriptivism— about observing and describing what happens in a language, and trying to explain why this happens, without dismissing in advance certain usages or varieties for reasons of purism, traditionalism, or personal taste. If someone decides to coin the phrase ‘how not to linguistics’, a linguist does not say, “No, Maud, that’s not how English works. You can only use the infinitive marker ‘to’ with a verb.” Instead, the linguist might file this phrase away as an idiosyncrasy, an unintentional error, or usage that sounds unnatural to a native speaker.
Upon further investigation, however, the linguist would discover that this phrase does occur more often, among a particular subgroup of English speakers. A Google search for ‘how not to cat’ yields 251,000 results; the search ‘forgot how to dog’ yields 56,9000 results (a considerable difference, but we all know the internet is for cats). The use of ‘how to’/‘how not to’ in combination with a word from a word class not ordinarily found in this construction, then, might be one of the many diverse features of internet slang.
Why on earth would someone say, or type, ‘my cat forgot how to cat’? In the realm of internet linguistics, two explanations often seem to go a long way: brevity and comedy. ‘My cat forgot how to cat’ is shorter than ‘my cat forgot how to behave like a cat’. It’s also funnier, possibly because it’s not conventional language usage; the human brain likes novelty, after all.
And, in the case of ‘how not to linguistics’, there’s an additional reason: gravity. ‘How not to linguistics’ doesn’t sound nearly as serious as ‘this is not how a linguist is supposed to behave’. Of course, I couldn’t possibly say that— it would be terribly prescriptivist of me.
(Note: according to Steven Pinker, “most writers who have given serious thought to language are neither kind of iptivist.” You can find his much more in-depth consideration of micro-level prescriptivism – as opposed to the macro-level “my variety is better than your variety” I discussed above – vs. descriptivism here.)