In a post here last June I asked for editorial staff at The New Yorker to come forward, anonymously if they wish, and explain something to me. Why do they so resolutely refuse to employ subject-verb inversion with reporting frames, even when the policy drives them to print sentences that are not just inept but almost incomprehensible?
Chris Potts first documented the strange practice in one of the earliest Language Log posts back in 2003.) Nobody from the magazine came forward to explain, either then or last year. Instead, New Yorker staff redoubled their efforts to show that nothing could make them consider verb-subject order. On March 21 (p. 54, left column) they published what I think is the worst example yet, buried in the middle of an article by Dana Goodyear about Hollywood writer's-block therapists Barry Michels and Phil Stutz:
"We're like carnies, always out there trying to sell some idea," another writer, who sees Michels, and whose husband, also a writer, sees Stutz, told me.
I continue to wonder, what the hell is wrong with them that they could believe this is fine prose style?
What any sane writer or editor would have done with the above is to use inverted order of subject and a simple verb of saying:
"We're like carnies, always out there trying to sell some idea," said another writer, who sees Michels, and whose husband, also a writer, sees Stutz.
That gets the subject noun phrase, with its heavy load of comma-separated supplements, to the end of the sentence, where it can be processed more easily.
Alternatively, if "told me" rather than "said" is considered absolutely essential (I really don't see why it should be), the thing to do would be not to prepose the quotation:
Another writer, who sees Michels, and whose husband, also a writer, sees Stutz, told me: "We're like carnies, always out there trying to sell some idea."
That puts the quotation at the end.
But what you don't do, if you have any clue about how to write, is to interpolate five comma-suffixed phrases between quotation and verb of saying: (i) a noun phrase (another writer), (ii) the first coordinate of a supplementary relative clause modifying it (who sees Michels), (iii) a second coordinate beginning with a wh-phrase functioning as subject of a further supplementary relative clause (and whose husband), (iv) a supplementary predicative complement elaborating that wh-phrase (also a writer), and (v) the verb phrase of the second relative clause (sees Stutz).
Some fascist editor with a considerable amount of grammatical sophistication is still sitting there in The New Yorker's offices enforcing this awful style by banning subject postposing after preposed direct-speech complements, no matter what the cost in human misery and syntactic clunkiness.
It is possible, I realize, that they are doing it simply to perplex me. They're winding Language Log up. Well, I refuse to be perplexed or troubled. But I confess to a slight curiosity: Why would anybody insist on execrable style when a vastly better version can be readily constructed by a simple word order switch?
"Chris Potts and I, and the Language Log readership, would love to know why," Language Log, a group linguistic science blog, run out of the University of Pennsylvania, mainly by Mark Liberman, Trustee Professor of Phonetics, assisted by several other linguists, commented.
["Comments are closed," the author, Geoff Pullum, decreed; but New Yorker staff members who are ready to spill the beans about their magazine's stylistic perversions can write to email@example.com if they wish. We will keep their identity secret.]