Learning Useless English Grammar in Japan

Have you ever struggled to understand a presentation by a scholar with a strong Japanese accent? Last week I wrestled with the speech of a brilliant young psychology lecturer visiting from Japan. For the word graph (IPA [gɹæf]) he produced what unfortunately sounded almost exactly like grape. I knew enough about Japanese phonology to guess what he probably intended by his pronunciation (IPA [gɻeːɸ]), but many audience members were baffled.

Many schools in Japan teach English from age 10, but the conversational ability actually obtained by some Japanese adults, including top-flight, internationally known academics, is often extremely poor. The many reasons include inadequate instruction in pronunciation; lack of meaningful opportunities to speak English; and the use of old-fashioned structure-drill grammar teaching.

A private English tutor in Japan, whom I’ll call Yuki, wrote to me recently to ask about the underlined relative clauses in sentences like these, which were used in actual examination questions and later published in high-school textbooks:

1. She said she didn’t like the film, which opinion surprised everyone.
2. The men wore kilts, which clothing I thought very interesting.
3. The doctor told her to take a few days’ rest, which advice she didn’t follow.
4. He spoke to me in Spanish, which language I have never studied.
5. The suspect didn’t drive his car on the day, which fact is important.
6. She favors equal pay, which idea I’m quite opposed to.

I was appalled (and not only by the politics of example 6!). The examples feature a nonrestrictive relative clause introduced by which plus a head noun. This is extremely unusual — nonexistent in conversation, vanishingly rare in modern sources.

Yuki has to help students with such material but reports that the sentences are rejected by native English speakers, who say that as far as they can remember, they have never heard or seen such sentences before.

Yet Japanese students are not just drilled on such sentences, they are examined on them in crucial university admission tests.

When Yuki wrote to a publisher to ask why they covered such odd sentences, the reply was that they would continue to do so because the test questions are genuine; and they added that the sentences are correct, citing Huddleston and Pullum (2002), The Cambridge Grammar of the English Language!

It is true that CGEL cites this sentence on Page 1,043:

I said that it might be more efficient to hold the meeting on Saturday morning, which suggestion they all enthusiastically endorsed.

But note that CGEL is a comprehensive scholarly reference grammar, not a textbook for teenage foreign learners, and needed the example to make the points that which cannot be separated from its head noun in such a case: Although which suggestion they all endorsed is marginally possible, *which they all endorsed suggestion is decidedly not.

Crucially, CGEL stresses that the construction is “quite rare and formal, verging on the archaic” (Page 1,044). Mark Liberman on Language Log discussed the construction in detail back in 2008, noting that it was dying by the end of the 19th century except in the prose of David Foster Wallace, who was infatuated with it.

The textbook authors who invented the six examples above clearly missed both the formality and the obsolescence. In fact half the examples have gross style clashes: Nos. 1 and 5 have didn’t, and No. 6 has I’m, unambiguous indications of casual or informal style that are incompatible with the formal, literary tone of the relative clause.

You could live a rich English-speaking life without ever hearing or reading an instance of this construction. It shocks me that any learner of English anywhere should spend time on it. But in Japan such utterly implausible examples are being used as the basis for exam questions. A published textbook reports that Jissen Women’s University gave an English entrance test containing this question:

And Hokusei Gakuen University set an exam with this question:

The study-aid book reprinting it says that should be changed to which. Notice, the sentence as it stands would be readily understood by a native speaker, though as punctuated it’s a comma-splice. Changing that to which makes it solely interpretable as an instance of a rare and archaic construction that typical English speakers reject, and you might well say that makes it worse. What kind of a test of English is this?

University-educated Japanese ultimately achieve remarkably good command of English syntax, despite the inadequacy of the old-fashioned methods of their early schooling. Better teaching, especially of pronunciation, is desperately needed. Teaching time must not be squandered. What disturbs me is that every minute that English teachers spend on the grammar of a rare relative clause type that no student will ever see again is a minute wasted.