Functional Variety of Toka as a Quotative Marker in Conversational Japanese

Previous studies of quotation in Japanese have mainly focussed on two types of quotative markers: to and tte. This paper demonstrates the function of another quotative marker toka (a combination of the quotative marker to and the question particle ka), as well as quotation that occurs without any marker at all (zero-marking), in conversational Japanese. Dictionaries define both to and tte as particles demonstrating quotation, while toka is defined as an indication of either uncertain imagination or hearsay. However, these differences in dictionary definition do not necessarily reflect how these markers are actually used. I will suggest that toka in actual conversations has two functions: 1) to show the informality of conversations; and 2) to show an event in a narrative, and contrast the event with the peak of the narrative. My analysis of 270 quotations from 19 conversations in Japanese reveals that toka is used 46.3% of the time, almost as often as tte (48.9%). The complementizer to, which is typical in constructed data, occurs far less often (2.2%). Following Hopper\u27s (1987) claim that grammar comes from and is shaped by discourse, I examine the functions of quotative markers with respect to their use in different types of discourse. Toka is frequently used in informal conversations, whereas to is only used in formal conversations. Use of toka is predominant in narrative type conversations, in which one speaker continues to keep the floor, but it is not used for the peak of such narratives. Instead, either tte or zero-marking is used to mark the peak of a narrative. This paper discusses the functions of toka in contrast with the functions of other quotative markers, based on analyses of spontaneous Japanese conversations