In an interview on NPR’s Weekend Edition Sunday yesterday, Steve Kleinedler talked about the new edition (#5) of the American Heritage Dictionary, of which he is the executive editor. Partway through, he talked about the Usage Panel, saying:
(1) Every year, we send out the panel a ballot full of questions asking their opinions.
This is at the very edge of grammaticality for me (and many others), though there are some who find such examples acceptable, and they occur with modest frequency. What’s the problem?
The Dative Alternation. For transfer verbs like give and send (occurring with two non-subject NPs, one denoting the thing transferred and another denoting the recipient of the transfer), English allows two argument structures:
PP Dative: V + NP:THING-TRANSFERRED + PP[to NP: RECIPIENT] (We gave/sent money to Kim)
NP Dative: V + NP:RECIPIENT + NP:THING-TRANSFERRED (We gave/sent Kim money)
(“Dative” in the names is a syntactic label; English has no morphological dative case.) This is the Dative Alternation, and it’s been the subject of an absolutely gigantic literature. (The NP Dative is sometimes called the “Dative Movement” construction, from an old transformational analysis in which the PP Dative was converted to the NP Dative by “moving” the RECIPIENT NP into position as direct object, while “demoting” the THING-TRANSFERRED NP.)
Constraints on the NP Dative. The availability of the NP Dative is apparently constrained by a huge number of factors, having to do with the semantics of the V, the discourse prominence of the referents involved, the phonology of the V, the grammatical person of the NPs involved, the pronominal status of the NPs involved, and the particular V involved (with donate fine in the PP Dative but dubious in the NP Dative, for example).
The constraints on particular lexical items as the V in the NP Dative have often been taken to be largely idiosyncratic, just facts that speakers know about the availability of particular Vs in the NP Dative, but there are also potential generalizations having to do with properties of the Vs. One of these has to do with Vs that are V + Prt combinations (like give away and send out), which generally resist the NP Dative:
(2) PPDat: We sent out money to Kim / NPDat: ??We sent out Kim money [with V + Prt as a constituent]
(3) PPDat: We sent money out to Kim / NPDat: ??We sent Kim out money [with V and Prt separated by the object NP]
This is the constraint that’s operative in (1), though (1) has the extra wrinkle that the THING-TRANSFERRED NP (a ballot full of questions asking their opinions) is fairly long and complex. This allows an alternative to the PPDat in (3) in which the THING-TRANSFERRED NP can be “postponed” to the end of its clause:
(1a) We send out a ballot full of questions asking their opinions to the panel.
(1b) We send out to the panel a ballot full of questions asking their opinion.
The alternative (1b) is less awkward than (1a) — and it’s quite close in form to (1), a fact that might make (1) itself more acceptable to some speakers (who might see (1) as an abbreviated version of (1b)).
In any case, though things like (1) generally get low marks in collections of acceptability judgments, some people find them acceptable, and they occur with modest frequency. So they’re at the edges of acceptability.
November 14, 2011 at 4:02 pm |
How does “We sent Kim out some money” fit in here? And why do I like that better than “We sent Kim out money”?
November 14, 2011 at 4:49 pm |
“We sent Kim out some money” is a variant of NPDat (3), but with a more substantial second NP; the prosodic weight of the NPs is another significant variable here.
November 14, 2011 at 6:30 pm |
Believe it or not, this construction is part of my idiolect. Anymore I try to keep marked items in my idiolect at bay, but sometimes I get tired and they seep through.
November 15, 2011 at 6:28 am |
I suspected that it was indeed part of your idiolect. Most people who use it naturally probably don’t know it’s a marked item — like the positive anymore you threw into your comment — but you’re a linguist and are likely to be aware of its status.
November 28, 2011 at 9:17 am
(PS: I threw in the “anymore” there to be cute. This instance was completely intentional. :D)
November 15, 2011 at 7:52 am |
Kleinedler’s comment sounds like a straightforward rendering of Yiddish construction into English.
November 15, 2011 at 8:27 am |
There’s a big literature on “positive anymore” (Wikipedia page here). A Yiddish origin is hugely unlikely, though a Scots-Irish origin is possible.
November 15, 2011 at 9:13 am |
For some indication of how complex the conditions on the dative alternation are, look at “On the gradience of the dative alternation” (2003) by Joan Bresnan and Tatiana Nikitina (here) and “Is syntactic knowledge probabilistic? Experiments with the English dative alternation” (2006) by Joan Bresnan (here).
November 22, 2011 at 7:09 am |
[…] something to someone for give and other verbs involving the transfer of something to someone. He observes: The availability of the N[oun]P[hrase] Dative [i.e. give someone something] is apparently […]
November 22, 2011 at 7:14 am |
Nice post, esp. the Bresnan links. As it happens, there was some friendly discussion on dative constructions in the comments of one of my recent posts. The question: Is “Brynn will say who(m) she gave the cookies” grammatical? I tried to shed light on the issue here, with a link to this post and a CGEL cite.
January 25, 2012 at 5:56 pm |
[…] is what I said at the beginning of “NP dative on the edge” (on this very blog): This ["Every year, we send out the panel a ballot full of questions asking their opinions'] is […]
February 18, 2012 at 10:59 am |
[…] including the dative alternation Grimm alludes to (NP Dative vs. PP Dative; some discussion here). And of course alternations on a larger scale, like Grimm’s SSR case and the other […]