| Abstract: | In this paper we discuss some rather interesting tonal facts from Hakha Lai, a Tibeto-Burman language spoken in Burma and Mizoram State, India in which words are generally monosyllabic. In the first part of the talk, we show that a single conspiracy under lies all of the tonal alternations which occur in two-word sequences, which can be elegantly captured within optimality theory. In the second part of the talk we show how this "elegance" appears to dissipate once sequences of three or more words are taken into consideration: In particular, a serious problem arises in predicting the right-to-left direction of rule application, which produces opaque outputs violating the very conspiracy that motivates the tonal alternations in the language. In the last part of the paper we show how this problem is wholly dependent on the view one takes on how to represent the input-output relations in question. |
