by

Beliefs About Grammar and Extraterrestrials

U.S.S.R. postage stamp of extraterrestrial satellite; grammar understanding and textbooks are similarly last century.

The vast number of books, pamphlets, articles, columns, blogs, and other study aids on English grammar are virtually all afflicted with a single problem. Using loose, mushy, meaning-related notions, they make statements that are (i) stated in essentially identical terms everywhere (it resembles mass plagiarism), and (ii) almost universally accepted, especially by educated people, yet (iii) patently false, as even a minute of reflection reveals.

The minority belief that intelligent aliens visit our planet is nowhere near as strange. (i) No real consensus emerges from the reports about aliens given by those few who claim to have encountered them; (ii) the minority of Americans who believe that aliens walk among us is tiny (something like 2 percent), college-educated people being less likely to believe; and (iii) the issue is epistemologically problematic. Positive existential claims like “There are aliens among us” can’t be refuted by any number of failures to observe aliens, so we have to settle for a sort of Bayesian epidemiological reasoning: The absence of widely observed outbreaks of alien appearances or activities reduces the plausibility of their being here to near zero, though not enough for the hypercredulous.

It’s much weirder to find claims being repeated verbatim in many sources and believed by almost all consumers when it is trivial to refute them. But that is what we see in the realm of English grammar.

I only have space for one example (though there are many). I’ve picked a hardy perennial: the passive. Here’s how Constance Hale’s latest article in The New York Times “Opinionator” series on writing explains the difference between active and passive voice:

In the active voice, the subject performs the action. In the passive voice, the subject is acted upon.

Ms. Hale isn’t an outlier: Popular grammar presentations always say this, often in exactly these terms. Yet it simply isn’t true. (And by the way, I don’t mean because of the conflation of subjects with what they denote—I’m ignoring that universally tolerated abuse.) Consider these active clauses:

Bob reluctantly submitted to an unpleasantly intrusive body search by TSA officers.
The infant underwent an 11-hour operation at the hands of a team of neurosurgeons.
The rabbit vanished when the magician waved his wand.
The council is receiving a blizzard of hate mail from angry residents.

The actions in the depicted situations are performed by TSA officers, neurosurgeons, a magician, and angry residents. No ordinary person viewing a video of the events described would say that the performers of the actions were Bob (motionless, legs apart, arms raised), or the infant (unconscious on a table), or the rabbit (tamely sitting where it has been trained to sit), or the council members (at home with their families while mail piles up at City Hall).

Now look at these passive clauses (subjects are underlined):

It has been proved that pi is an irrational number.
There aren’t believed to be any counterexamples.
Tabs are being kept on the movement’s ringleaders.
The letter Z is preceded by all the other letters.
Philology is included under linguistics in the statistical data.

Nobody could seriously maintain that it, there, tabs, the letter Z, or the discipline of philology are being acted upon by some force or agency in the situations these sentences describe. It, there, and tabs don’t really have meanings at all—they do not denote things that could conceivably have stuff done to them.

What’s gone wrong is that one special case—clauses describing animate actors doing things to people or objects—is being taken not just as prototypical but as the only case. That’s highly misleading.

Mushy, meaning-based talk of actors and actions and being “acted upon” is not what we need to characterize the concept of a passive construction. People are trying to squeeze syntactic blood out of the turnip of naive metaphysics. Not possible. To develop an adequate characterization (and yes, I’ve attempted one: see it here), we need syntactic notions like “participle” and “pre-auxiliary position” and “verb agreement” rather than semantic-role notions like “agent” or “action” or “affected entity.”

You’d expect that occasionally some texts might oversimplify like this. But to find that essentially all of them do is a startling indicator of the dire current state of English grammar as an academic subject. In texts on subjects like biology and cosmology, things have moved on a bit over the past century or two, but that isn’t true for grammar.

Return to Top