This past week, the attacks on Sotomayor have turned from what she’s said to how she’s said it.  Conservatives began by hammering away at the “weird, unidiomatic constructions and errors of punctuation and grammar [in] her infamous 2001 ‘Wise Latina’ speech.”  Now, I advocate writing conference papers that “contain few expensive words and no Faulknerian feats of subordination” on the grounds that no human being—not even the academic ones—can parse grammatically complex arrangements of jargon on the fly, so I’m more attuned than most to the fact that what passes for grammatical in English as she is spoke doesn’t pass muster in English as she is wrote.  You can imagine, then, why I chafed at Heather McDonald’s criticism of Sotomayor’s unscripted speeches for containing errors endemic to spoken language.  Just because an unscripted speech is transcribed after the fact doesn’t make it written; or, as per my talk, just because it can be put to paper doesn’t mean it was meant to be read there.

Judging the quality of her prose from her speeches, as McDonald and fellow Bench Memos writer Ed Whelan did, is an intellectually dishonest exercise for the simple reason that nobody (outside of Henry Louis Gates, Jr.) speaks in paragraphs. Not that this stops the other member of the Bench Memos team, Matthew J. Franck, from claiming Sotomayor “writes” sentences that were clearly spoken aloud; or that those sentences “begin with a thought and trail off without saying much of anything after all, or double back and contradict themselves,” i.e. that the transcription of her unscripted speech transcribed sentences that were clearly spoken aloud.

When I taught literary journalism, I always included Mark Singer’s profile of Errol Morris on the syllabus because it works both as an introduction to New Yorker-style profiles and a meta-methodological essay on how to approach transcriptions critically and responsibly.  Here’s Singer transcribing Florence Rasmussen’s speech in Morris’s Gates of Heaven (1978):

If I could only get out. Drive my car. I’d get another car. Ya . . . and my son, if he was only better to me. After I bought him that car. He’s got a nice car. I bought it myself just a short time ago. I don’t know. These kids—the more you do for them . . . He’ s my grandson, but I raised him from two years old . . . I don’t see him very often. And he just got the car. I didn’t pay for all of it. I gave him four hundred dollars.

Singer responds to her outpouring by noting that

[w]ith an arresting instinct for symmetry, Florence Rasmussen manages to contradict most of what she has to say. It seems that she knows certain things, but then, in the next moment, she trots out contrary information[:] I’d like to drive my car; but I might not even have a car any longer, might have to buy a new one. I bought my son—O.K., he’s not my son, he’s my grandson—a new car; well, I didn’t pay for the whole thing, I gave him four hundred dollars, but anyway I want my money back.

As viewers of The Fog of War (2003) know, this technique works as powerfully on former Secretaries of Defense as elderly Floridians, not because either is particularly muddleheaded, but because Morris takes advantage of the infelicitous glitches that accompany the spontaneous production of spoken language.  McDonald, Whelan, and Franck seem not to understand how language works.  They scoured “virtually all [Sotomayor’s unscripted] speeches on the Senate website” and discovered damning evidence that they were, in fact, unscripted speeches; then, they lambasted those transcripts of her unscripted speeches for failing to meet the standards demanded of the written word, which proved to them that Sotomayor is “a mediocrity as a writer.” Whelan even suggests she’s a hypocrite for “present[ing] herself as a stickler for good grammar” when her speeches contain constructions that would be ungainly, if not outright ungrammatical, on the page.  How about we hoist Whelan by his own petard?

Well, it’s an unguarded moment where she says what folks on the left think which, their job is to use judicial robes to make sound policy and the law is largely a vessel for them to fill with their own preferences.

Sure, very much along the same lines, talking about what he calls the criterion but which selecting judges.

Well, how is that honoring people who put their lives at risk in public service and, look, at 9/11 we understood for a while what firefighters do.

Whelan clearly offends Franck’s standards: he “writes sentences that begin with a thought and trail off without saying much of anything after all.” The first sentence subordinates a clause about as clunkily as you can imagine; then you read the second sentence, observe that it doesn’t even include the subordinate clause it introduces, and marvel at the paucity of your imagination.  I suppose that second sentence meets Franck’s standard on a technicality—the sentence defies the laws of grammar and stops before it has a chance to trail off without saying much of anything—but that third sentence turns on a dime “and, look,” informs us that we know “what firefighters do.” Does that count as “much of anything”?

They do eventually get around to criticizing Sotomayor’s prose on the basis of what she’s actually written: they find that while it’s not “ungrammatical,” it is “tedious and ‘impenetrable.'” That link leads to Stephanie Mencimer’s article in Mother Jones, in which she pings Sotomayor’s prose for “rarely hit[ting a] sort of breezy cadence [because she] devotes the bulk of her legal analysis to quotes from statutes, regulations, and other opinions ad nauseam[.]” How she’ll legislate from the bench when she’s busy citing dull precedent is as mysterious to me as, for example, how a prose-scold could write “quote” instead of “quotation” in the middle of a complaint or believe Diane Wood’s summary of Merchant of Venice signifies much of anything in the Age of Wikipedia.

In the end, the case against Sotomayor basically amounts to this: on the one hand, her speeches betray all the ungrammatical tics common to spoken language, and if you treat those speeches as prose, you must conclude that she’s a poor writer; on the other, the prose of her legal opinions isn’t ungrammatical, but because it betrays the tedious tics common to lawyerly prose, you must conclude she’s a poor writer.  You see where this is headed.  If word leaked out that “Ed Whelan” was actually Sotomayor’s pseudonym, the Bench Memos team would argue that, because Whelanmayor did all that I documented above, you must conclude that s/he’s a poor writer.  Were it then revealed that Whelanmayor did Diane Wood one better and actually wrote Merchant of Venice, the Bench Memos team would claim that, in light of the difficulty of Shakespelanmayor’s prose, the Western Canon requires immediate revision and you must conclude s/he’s a poor writer . . .