In lieu of an abstract, here is a brief excerpt of the content:

  • The syntax of Southern American English personal datives:An anti-locality account
  • Youssef A. Haddad

1. An Overview

Southern American English licenses personal dative constructions (Horn 2008), also known as coreferential dative constructions (Al-Zahre and Boneh 2010). The sentences in (1) are a few examples; they contain nonsubcategorized pronominal arguments that are obligatorily coreferential with the subject. These unselected pronouns are dative case-marked in languages like Hebrew and Arabic—thus the name personal dative—but they are not overtly dative in Southern American English.

  1. 1.

    1. a. I'm gonna buy me a shot gun, just as long as I am tall.

(Jimmie Rodgers, "T for Texas")
  1. b. I'm gonna grab/catch me a freight train. (various songs)

  2. c. When I was a young girl, I had me a cowboy.

(John Prine, "Angel From Montgomery") (from Horn 2008:169-170, 2a-c)

Being unselected arguments means that personal datives (hereafter PDs) do not belong to the thematic grid of the predicate. Proof that this is the case comes from the fact that their deletion in (1) does not alter the truth conditions of the sentences. As Horn (2008:173) notes, PDs receive case but no theta role; in this sense "they do not represent true datives/recepients/goals". As (2) illustrates, a PD may be used in structures that already contain a recipient or a beneficiary.

  1. 2. He's gonna buy him/*himself a pickup for his son.

(from Horn 2008:172, (9a))

On the other hand, PDs do make a non-truth-conditional contribution to the content. Webelhuth and Dannenberg (2006) call them "pragmatic intensifiers [that] serve to underscore the agent's role in the action"; to the authors, the PDs in (3) (in original ex. 6) highlight the import of the accomplishments carried out by the subjects. Similarly, Horn (2008:181) holds that the use of PDs implies that "the speaker assumes [End Page 403] that the action expressed has or would have a positive effect on the subject, typically satisfying the subject's perceived intention or goals".

  1. 3.

    1. a. She bought her a house.

    2. b. They cut them some logs.

What is important for the purpose of this squib is that the PDs in (1-3) seem to violate Condition B of Binding Theory by being realized in positions where reflexive pronouns are expected. This means that sentence (3a) should be realized as (4).

  1. 4. She bought herself a house.

Although (4) is grammatical, it does not mean the same as (3a). The reflexive pronoun in (4), unlike the PD in (3a), indicates that the subject is necessarily a beneficiary. This is why if a sentence already contains a recipient, only a PD but not a reflexive pronoun is grammatical, as (5) illustrates. Also, compare (6a) and (6b). Only (6a) indicates that selling some toothbrushes will solve a financial problem that the subject faces.

  1. 5. She bought her/*herself a house for her son.

  1. 6.

    1. a. I only need to sell me a dozen more toothbrushes.

    2. b. I only need to sell myself a dozen more toothbrushes.

(from Conroy 2007:67, 13a-b)

The sentences in (3) through (6) seem to indicate that the distribution of PDs is different from the distribution of reflexive pronouns. In the following section, I explain why this is the case. More specifically, I explain why PDs are not realized as reflexive pronouns. The analysis is based on the movement approach to Binding Theory (Hornstein 2001, Kayne 2002, and Grohmann 2003) and Grohmann's (2003) Antilocality Hypothesis. The movement approach I adopt here argues that the suffix -self on reflexive pronouns is the result of movement and antilocality restrictions on the minimal distance an element must move. I show that PDs are not derivationally related to their antecedent via movement. Therefore, they are not subject to the same antilocality restrictions, which is why they may not surface as -self anaphors.

2. Why PDs are not reflexive pronouns

Condition B of Binding Theory states that a pronoun should be locally free. Therefore, by allowing a pronoun to be coreferential with a local, c-commanding antecedent, personal dative constructions violate Condition B. How can we account for this violation in a principled...

pdf

Share