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

  • The Morphosemantics of Warlpiri Counterfactual Conditionals
  • Julie Anne Legate

Iatridou (2000) presents a morphosemantic theory of counterfactual conditionals in which past tense morphology is crucially implicated. Some subsequent work investigates how counterfactuality is realized in languages that lack tense (see Rackowski 1998 for Tagalog and Nevins 2002 for Chinese). In this squib I address a different issue: the morphosemantics of counterfactuality in Warlpiri, a language that has past morphology, but fails to use it in counterfactuals. I show that despite this prima facie challenge, the Warlpiri data provide additional support for Iatridou's theory.

Iatridou (2000) observes for a range of languages that counterfactuals necessarily display past tense morphology, which fails to contribute a past interpretation. This is illustrated in (1) for English future less vivid (FLV) conditionals, 1 present counterfactuals (PresCF), and past counterfactuals (PstCF).

(1)

  1. a. FLV
    (I don't think he will take my advice, but . . .)
    If he took my advice, he would get the job.

  2. b. PresCF
    (He isn't friendly, but . . .)
    If he were friendly, I would invite him.

  3. c. PstCF
    (I didn't have the car yesterday, but . . .)
    If I had had the car, I would have gone for a drive in the countryside.

Thus, (1a) receives a future interpretation and (1b) receives a present interpretation, despite the presence of past morphology in each.2 In addition, (1c) exhibits two layers of past morphologically, but only one layer temporally. Thus, although it has the morphology of the pluperfect, it is not interpreted as a past of the past; instead, it is interpreted as a simple past. Thus, the PstCF also involves past morphology that does not receive its usual past tense interpretation. Iatridou proposes that the past morpheme is better analyzed as an exclusion feature (ExclF)-that is, an underspecified morpheme that results in a past tense interpretation when it ranges over times and a counterfactual [End Page 155] interpretation when it ranges over worlds. ((2) is adapted from Iatridou 2000:246.)3

(2) ExclF = T(x) excludes C(x)
T(x) stands for "Topic(x)" (i.e.," the x that we are talking about"). C(x) stands for "the x that for all we know is the x of the speaker."

  1. a. ExclF(t) = the topic time excludes the time of utterance ("the time interval that we are talking about excludes the time interval that for all we know is the time of the speaker")

  2. b. ExclF(w) = the topic worlds exclude the actual world ("the worlds that we are talking about exclude the worlds that for all we know are the worlds of the speaker")

In FLV and PresCF conditionals, then, the past morphology is the realization of ExclF(w), yielding counterfactuality. In PstCF conditionals one layer of past morphology realizes ExclF (w), whereas the other realizes ExclF(t), in order to express both past tense and counter-factuality.

Iatridou further observes that in many languages imperfective aspectual morphology in counterfactuals also fails to receive its usual interpretation; in counterfactuals it is compatible with either a perfective or an imperfective interpretation. Iatridou proposes that "[w]hen the temporal coordinates of an eventuality are set with respect to the utterance time, aspectual morphology is real; when the temporal coordinates of an eventuality are not set with respect to the utterance time, morphology is always Imp [imperfective]" (p. 262).

Turning to Warlpiri, we observe that tense and aspect information appears in two separate positions in the clause: a second-position clitic cluster that also includes subject and object agreement, and a verbal suffix. The clitic position includes the present imperfective morpheme ka, (3a), the past imperfective morpheme lpa, (3b), and the perfective morpheme Ø, (3c). The suffixal slot on the verb may be filled by a greater range of elements, and the phonological realization of these elements varies according to the conjugation class of the verb. The morphemes relevant here are the nonpast (3a), past (3b), and irrealis (3c).4

(3)

  1. a. Ngaju ka-rna         wangka-mi.
    I       PRES.IMPF-1SG speak-NPST
    'I am speaking.'
    (Hale, Laughren, and Simpson 1995:1430) [End Page 156]

  2. b. Wati-lpa-lu        ya-nu.
    man-PST.IMPF-3PL go-PST
    'The men were leaving.'
    (Hale, Laughren...

pdf

Share