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

  • The wrong two faces
  • Stephen Wechsler and Larisa Zlatić

1. Introduction.

Alsina and Arsenijević (2012, henceforth A&A) argue against the Wechsler & Zlatić 2000, 2003 account of hybrid agreement with certain Serbo-Croatian (S-C) nouns, claiming that their alternative account is simpler and more empirically adequate.1 But neither their analysis of these particular nouns nor their approach to agreement is viable. We defend the CONCORD-INDEX theory, arguing that it remains the better approach to agreement.

2. A&A's suppositions about agreement in S-C.

A&A's analysis of mixed agreement with deca-type nouns rests on the following suppositions about S-C grammar.

  1. 1.

    1. a. Supposition 1: Adjectives, possessives, and participles show syntactic agreement. (pp. 373, 378)

    2. b. Supposition 2: Finite forms agree both syntactically and semantically: 'A singular finite verb is compatible only with a subject that is both syntactically and semantically singular. Otherwise, a plural finite verb must be used.' (p. 373, ex. 4)

    3. c. Supposition 3: 'A singular personal pronoun in a case other than nominative cannot have a plural referent.' (p. 373, n. 9)

    4. d. Supposition 4: Pronouns can show either type of agreement. (p. 378)

Armed with these suppositions, together with the simple assumption that nouns like deca 'children' and braća 'brothers' are feminine singular, A&A claim to account for the hybrid agreement of these collective nouns.

But there are serious problems with A&A's proposal: suppositions 1 and 2 are falsified by much of the language, including very common, ordinary locutions, and supposition 3 is an unmotivated stipulation that obscures damaging counter evidence. A look at these problems helps illustrate the broader theoretical issues. Let us consider each of A&A's suppositions in turn.

According to A&A, adjectives, possessives, and participles show syntactic agreement (supposition 1). First- and second-person subjects, which are very common in speech, all falsify supposition 1. Adjective forms reflect the trigger's semantics, and not a syntactic feature.

  1. 2.

    1. a. Ja sam    gladn-a.
      I  AUX.1SG hungry-F.SG
      'I am hungry.' (spoken by a female)

    2. b. Ja sam    gladan.
      I  AUX.1SG hungry.M.SG
      'I am hungry.' (spoken by a male) [End Page 380]

  2. 3.

    1. a. Ti  si      gladn-a.
      you AUX.2SG hungry-F.SG
      'You are hungry.' (spoken to a female)

    2. b. Ti  si      gladan.
      you AUX.2SG hungry.M.SG
      'You are hungry.' (spoken to a male)

With participles, too, the inflectional form reflects the sex of a first- or second-person controller.

Adjectival modifiers of (nonnominative) personal pronouns also contradict A&A's first supposition, as shown by honorific uses of the second-person pronoun (Wechsler 2004:272, ex. 38).

  1. 4. [Vas        jadnu/jadnog/jadne]    niko    ne  poštuje.
    [you.ACC.PL poor.ACC.{F.SG/M.SG/PL} nobody NEG respect
    'Nobody respects poor you.'

    1. i. jadnu 'poor.ACC.F.SG': one female (formal) addressee

    2. ii. jadnog 'poor.ACC.M.SG': one male (formal) addressee

    3. iii. jadne 'poor.ACC.PL': multiple addressees

Gender and number on the adjective are semantically interpreted, as shown above.

Not only first- and second-person, but also many third-person subjects have this property. These include all proper names (examples in §4 below), as well as certain common nouns such as sudija 'judge' (Wechsler & Zlatić 2003:180, ex. 36).

  1. 5.

    1. a. Taj    stari  sudija je      dobro sudio.
      that.M old.M judge AUX.3SG well  judged.M
      'That old (male) judge judged well.'

    2. b. Ta    stara sudija je      dobro sudila.
      that.F old.F judge AUX.3SG well  judged.F
      'That old (female) judge judged well.'

The inflectional forms of the determiner, adjective, and participle all depend upon the sex of the judge, and not a syntactic gender feature.

Summarizing, many common varieties of S-C sentences contradict supposition 1, since the forms of possessives, adjectives, and participles are determined by the meaning and not the syntactic features of the trigger. At the same time, other triggers do determine syntactic agreement on these target types, as shown by some of A&A's examples.

Turning next to finite forms, A&A state in summary that they 'are sensitive to the...

pdf

Share