Descriptivism, Prescriptivism, and Assertionism

Our readers likely know that I have many disagreements with prescriptivists when it comes to English usage. But while I have philosophical disagreements with prescriptivists in general, my main practical disagreements are with people who might best be labeled “assertionists” — people who don’t just say that prescriptions set forth by some supposed authorities define what is “right” in English, but who simply assert a prescription even in the face of what those supposed authorities say. Usage X is wrong, they say. Why? Because it violates this rule. What’s your authority for the proposition that this is a rule? Well, it violates the rule.

The recent exchange about starting sentences with “and,” “but,” or “or” offers an excellent example. I pointed out that this is common usage — including among prominent authors, in leading newspapers such as the New York Times, and in leading legal sources such as the opinions of the U.S. Supreme Court. I also noted that the supposed usage authorities that I checked do not in fact condemn it.

Yet some people still argued that such a usage is wrong, at least in “formal” writing (though you’d think that Supreme Court opinions would be classified as pretty formal). When I inquired why this might be so, I got three basic replies:

1. “I learned basic grammar long ago. A conjunction joins two related thoughts in a sentence.” But this is mere assertion. One can equally say that conjunctions (or at least the kind we’re discussing) “link units of equal status”; indeed, this is precisely what the Oxford English Grammar (1996) says. This could mean two words in a sentence, two clauses in a sentence, two sentences in a paragraph, or two paragraphs in a work; and indeed the Oxford English Grammar gives an example of a sentence that starts with “But,” without any objection.

Now one could try to dismiss such an authority (and I’ve heard many assertionists do that in other contexts) by saying that the authority is merely descriptivist, and that therefore we shouldn’t trust its claims. I don’t buy that argument, because I’m a descriptivist; but let’s accept it for a moment. Even if one dismisses the Oxford English Grammar as an authority for the proposition that conjunctions “link units of equal status” as opposed to just “two related thoughts in a sentence,” one still needs an authority for the contrary proposition. What is that authority?

2. One commenter did try to point to such an authority, writing,

I have a Ph.D. in linguistics and I taught grammar at a university for 20 years – for what it is worth. It is indeed a rule in formal English that you cannot begin a sentence with a conjunction. See grammar texts by Azar.

Formal English = written English in a formal context = English that is meant to be read in a formal context, e.g. academic research.

In conversation, however, you can start a sentence with a conjunction. Blogging is often intended to be conversational, so we see conjunctions at the beginning of sentences a lot. I do it in my own blog.

To be formal, use ‘also’ for ‘and’; use ‘however’, for ‘so’. Make sure you use commas.

Please note: Language changes. This is one of those areas of English that we see changing before our eyes. My children will know how to use ‘however’ and ‘thus’, and know how to punctuate them, but I don’t expect that their public schooled peers will.

So I got a text by Azar (Betty Schrampfer Azar). As best I can tell, Azar generally writes schoolbooks, so I got Fundamentals of English Grammar, a “developmental skills text for lower-intermediate and intermediate students of English as a second or foreign language.” And, sure enough, Azar does say that “Except in very formal writing, a conjunction can also come at the beginning of a sentence,” though without (as best I can tell) explaining what writing qualifies as “very formal.”

But Azar doesn’t explain the basis for her assertion, so it’s basically her word against the Oxford English Grammar’s, Webster’s Dictionary of English Usage, the Harper Dictionary of Contemporary Usage, the Supreme Court, and others (at least unless “very formal writing” is such a vanishingly small category that none of the sources thought of noting the exception, and that Supreme Court opinions do not fit within it). Again, then, we have argument by assertion, with no explanation for why we should follow Azar rather than the others.

3. Finally, one of the commenters also argued, “My impression is that, in casual English, sentences beginning with conjunctions are usually incomplete sentences. However, they may contain complete thoughts with the missing words implied.” Now I think that functional arguments are often sound arguments for why we should avoid some usage (though not that the usage is “wrong”). But again this seems to me argument by assertion. The commenter thinks that sentences beginning with conjunctions (such as “but”) are usually incomplete sentences, but I don’t know of any evidence for that proposition. (Why would such sentences be incomplete more often than sentences that start with, say, “However” or “Moreover”?) Nor does the commenter offer argument in support of the more relevant proposition that we should avoid complete sentences beginning with “but” just because many such sentences are incomplete.

I’ve discussed this before, here, here, and here. But I thought it was worth noting, and worth breaking assertionism out as a separate category, and to call on people who make prescriptivist arguments to at least identify the supporting evidence or argument for their prescriptions — and to explain why this evidence or argument trumps the evidence or argument on the other side — rather than just relying on bare assertion.