Date: Mon, 15 Apr 1996 13:51:48 EDT

From: Larry Horn LHORN[AT SYMBOL GOES HERE]YALEVM.CIS.YALE.EDU

Subject: the athlete/politician's third-person



I'd just been talking about an increasingly prevalent formation in my prag-

matics seminar when I got a phone call about the very same construction from a

reporter for the local newspaper investigating the same point (from a non-

theoretical point of view, to be sure). The construction is the one exempli-

fied in the following citations:



"I'm just going to do the things Derek Harper has done for 10 years,

and hopefully that will be enough." --Derek Harper, NY Knicks basketball

player

"I just want to go to a place where Howard Johnson is going to put up

some big numbers." --Howard Johnson, baseball player



"I gave Pittsburgh every opportunity to sign Neil O'Donnell", O'Donnell

said. [quarterback signed away from Pittsburgh by N. Y. Jets.]



"It didn't kill Mike Stanley", Stanley said. [ex-Yankee catcher, on

being dismissed and allowed to sign with Boston Red Sox]



"My view is that I'm going to talk about Bob Dole, and I've been doing

a little of that." [Dole on "Nightline"]



"Six of us grew up living in a basement apartment. That was Bob Dole's

early life, and I'm proud of it, because we learned a lot about values,

about honesty and decency and responsibility and integrity and self-

reliance and loving your God, your family, your church, your community...

[oops, almost drifted off there. Bob Dole, speech 3/14/96, Columbus OH]



Of course, this construction has now been notoriously linked to the one non-

athlete contributor above, but what I was wondering was how well established

this usage is (is there a paper, e.g. in American Speech or elsewhere, either

published or already in print, that deals with it?), and if not, whether any

other ADSers have some nice examples to contribute. My own hypotheses are

that it occurs primarily to mark a character-defining (rather than accidental

or "stage-level" property) of the individual in question [cf. the unlikelihood

of 'I'll have to take a break because Bob Dole needs to use the bathroom'] and

that the form of the name used corresponds to the way the speaker imagines

he (I have no instances of she) WOULD be referred to by a reporter, etc., in

the relevant context, but always FN + LN (Thus it's always "Bob Dole", not

"Dole" or "Robert Dole", "Mike Stanley", not "Stanley" or "Michael Stanley".)

Note too that some of these examples violate "Principle C" of Chomsky's

Binding Theory (names and other referring expressions can't be bound) and

others don't, but all seem to involve a condition Ellen Prince discusses in

"Assumed Familiarity" scale ("Towards a Taxonomy of Given-New Information", in

_Radical Pragmatics_, ed. by P. Cole, 1981) predicting that speakers will use

an "evoked" expression, in particular a 1st or 2d person pronoun, rather than a

name, if s/he is in a position to do so, and that the use of a name in these

circumstances will implicate (by the usual Gricean mechanism) that the speaker

was NOT in a position to use the deictic pronoun. ANyway, I'd be interested in

further references, discussion, history, etc.





Larry