November 16, 2006

Is, is

I just heard some talking head on NPR use the "is, is" construction again. That's a new (and rather irritating) piece of English syntax that reanalyzes pseudo-clefts and generates a spurious reduplication.

I suppose that, as a syntactician, I ought to greet such innovations in my native tongue with delight; but frankly this one irritates me even more than uses of intensifier so, or as far as.. without a bookend. Just getting curmudgeonly, I guess.

I hear it a lot on NPR, but mostly when I'm driving, it seems, so I can't write them down. Here are some examples collected last year [commas interpolated]:

(1) The difference is, is that it's not coming from the top. ('On the Media' 12-10-05)
(2) But the reality of it is, is that ... ('Marketplace' segment on 'Morning Edition' 11-25-05)
(3) The situation was, is that my mom died ... ('Smart Money' 11-6-05)
(4) The problem is, is that the traditional Haitian government ... ('All Things Considered' 11-25-05)

I've never encountered this in writing; it seems to be strictly oral, which is what you would expect for a nascent syntactic rule. I have a theory about how it developed -- i.e, I thought of an explanation, but I haven't made any attempt to search the literature, so I don't know if it's the same as or different from anybody else's explanation.

It seems to me that these come from the pseudo-cleft rule, or rather from a misunderstanding of the pseudo-cleft rule. You remember cleft and pseudo-cleft, right? As Fillmore
puts it, "Instances of cleavage have IT in front; instances of pseudo-cleavage have WHAT in front."

Any of the sentences above could have been pseudo-clefts with two is's in a row:

(1a) What the difference is is that it's not coming from the top.

This has one is isolated at the end of the What clause, before the trace formed by Wh-movement, followed immediately after the trace by a second is serving as the fulcrum of cleavage.

They could also have been simple uncleft sentences, with only one is:

(1b) The difference is that it's not coming from the top.

It seems to me that speakers that use the "is, is" construction re-interpret the double is as a constituent, a marker for some kind of cleavage (instead of a syntactic accident), and drop the What that is the real marker -- after all, it doesn't add any information to (1a), as (1b) shows. (I suppose someone who cared more than I do could use this as an argument against the reality of traces.)

Why keep the double is at all, then? For the same reason people use whom nowadays: to appear authoritative. I guess that's what irritates me about it.

Posted by jlawler at 09:58 AM | Comments (1)

November 11, 2006

The way to make a name in linguistics is to make a name in linguistics

I was reading some other linguistics blogs and found an interesting post from Noncompositional about names linguists (especially syntacticians) give to rules, phenomena, and papers.

Háj Ross is (quite correctly) blamed for many of the most egregious, but Robin Lakoff is -- as the post intimates, and as Háj will be among the first to agree -- the Onlie Begetter. Who can forget her

"The Logic of Politeness, or Minding Your P's and Q's"?
Not only a pun on logical notation, but also a takeoff on Grice at the same time, and both à propos her topic. Just brilliant.

Andy Rogers (always a nomenclatural extremist) went so far as to name the rule he wrote his dissertation about -- a strange species of Raising with certain verbs like seem and a complementizer like that copied instead of moved the downstairs Su, producing such sentences as

  1. Mary seems like she gets along with everybody.
-- as "Richard", on the principle that since all names are arbitrary he might as well name this one after Richard Nixon as anything else. (This was in the early 1970s, note.)

I'm guilty of a few of these, too, sometimes to excess (mea culpa, mea culpa, ...). The first few CLS papers I wrote were entitled

and then (I hang my head in shame) ...

Ah, youth. Ah, linguists.

Posted by jlawler at 09:35 AM | Comments (0) | TrackBack