The role of discourse context in determining the argument structure of novel verbs with omitted arguments

Introduction The interpretation of verb meaning hinges in large part on the argument structure of the verb. In a series of experiments, young children have been found to use the number and type of arguments that appear with a verb in order to determine its meaning (e.g., Fisher 1996, 2002; Goldberg 2004; Landau & Gleitman 1985; Lidz, Gleitman & Gleitman 2003; Naigles 1990). For example, when hearing the novel verb blick in “She’s blicking her”, a child is likely to conclude that the verb refers to an act with two participants, one acting on the other, more often than when hearing the verb in “She’s blicking”. A verb’s subcategorization frame is thus a significant cue in the acquisition of verb meaning. However, languages vary with respect to how and when surface arguments are realized. There is a clear motivation from conversational pragmatics for leaving recoverable arguments unsaid, as expressed in Horn’s (1984a) R Principle or Grice’s (1967) Maxim of Quantity: say no more than you must. There is no need to utter arguments that are recoverable in the discourse context. As might be expected, then, many languages, perhaps the majority of languages in the world, routinely allow recoverable arguments, both subjects and objects, to be omitted. These languages include, for example, Hindi, Hungarian, Japanese, Korean, Laos, Russian, and Thai. Typical dialogues from Russian are shown in (1), Korean in (2) and Thai in (3):