How the Cold, Dead Hand of John Dryden Still Perpetuates Grammar Myths

A chain of happy research accidents starting at the 20th International Conference on English Historical Linguistics recently led me to the discovery that John Dryden (1631–1700) is responsible for not just one baseless modern prejudice about English grammar but two. He really has a lot to answer for.

Dryden famously invented the myth that sentence-ending prepositions are an error. Casting aspersions on a line from Ben Jonson’s Catiline, Dryden grumbles: “The preposition in the end of the sentence; a common fault with him” (see Mark Liberman’s “Hot Dryden-on-Jonson Action,” Language Log, 5/1/07). Though he admits immediately thereafter that his own prose shows the same feature; he actually knew that his own writings contained stranded prepositions, but he deprecated the construction anyway.

Nearly three and a half centuries later there are still American writers and academics who think stranding a preposition is something to feel guilty about.

But it seems Dryden was not done with his grammatical trouble-making. Two decades later William Walsh (1662–1708) sent Dryden the manuscript of his Dialogue Concerning Women, asking for comments. Dryden was complimentary but gave a few pieces of advice. Walsh should “avoid the words, don’t, can’t, shan’t, and the like abbreviations of syllables; which seem to me to savour of a little rusticity"; and of course (his views being unchanged from 1672) ending a sentence with a preposition “is not elegant.” But he also cited a new rule:

This appears to be the earliest source for the entirely false belief that a relative clause modifying a human-denoting noun must never begin with that.

Sarah Schwarz of Uppsala University gave a nice paper about this odd myth at the aforementioned ICEHL 20 meeting. Surveying usage guide authors, she found that some overtly reject this rule as spurious and virtually none endorse it. She found only one exception: the dreadful Publication Manual of the American Psychological Association, one of the bossiest of academic style guides, does reassert Dryden’s myth: “Use who for human beings; use that or which for nonhuman animals and for things.” With that exception, usage manuals provide no aid or comfort to the people who recoil at phrases like people that I’ve asked or a woman that I met.

And the evidence of published prose makes it overwhelmingly clear that such phrases are grammatically correct, and quite frequent. That-relatives with human head nouns were always grammatical and acceptable in English; Mark Liberman (“The Factual Impenetrability of Zombie Rules,” Language Log 4/9/17) illustrates with a man that had neuer seene an Elephant (1583, Sir Philip Sidney); some kinde of men, that put quarrells purposely on others in (c.1601, Shakespeare’s Twelfth Night); Men that all things learne; and nothing know (1624, George Chapman); and many more from subsequent centuries. Similar phrases are commonplace in the King James Bible (He that is born king of the Jews).

Yet on April 8 last year The New York Times published an opinion column by Frank Bruni that was almost foaming at the mouth about how terrible it was that people had started using that-relatives to modify human nouns. (It began in 2016, he thought, in speeches by Trump, Rubio, and Bush.) Who was “being exiled from its rightful habitat,” he moaned. We were being dehumanized: “My fear is that there’s a metaphor here: something about the age of automation, about the disappearing line between humans and machines. The robots are coming.”

How could The New York Times publish such nonsense? The answer is simple: In matters of language, fact-checking is not done.

Could Bruni possibly have believed what he was saying? Probably not. As Liberman points out, Bruni did enough research to discover that dictionaries and usage books don’t support his beliefs, and says so. He actually knew that he didn’t have a factual leg to stand on, but he published anyway. When it’s about language, you can just make stuff up.

Dryden had made up the same stuff much earlier, it turns out (though Bruni almost certainly didn’t know that). It’s sad to think of people being worried about violating an imaginary rule of grammar that has been occasionally and misguidedly bruited about for more than 300 years.

Sources: “Defence of the Epilogue” was published in 1672 as an addendum to a play, Almanzor and Almahide, or the Conquest of Granada; it is available in facsimile at Early English Books Online. (In the second edition a year later, curiously, the celebrated critique of preposition-stranding has been edited out.) The quotation from Dryden’s letter is from Charles E. Ward (1942), The Letters of John Dryden (Durham, N.C.: Duke University Press), Pages 33–34, cited by Joan M. Maling (1978), ‘The complementizer in Middle English appositives,’ Linguistic Inquiry 9(4), 719–725.