Download AFLA8.pdf

Contents : Backward Control: Evidence from Malagasy* Maria Polinsky and Eric Potsdam University of California San Diego and the University of Florida 1. Introduction Control is an asymmetric interpretational dependency between two syntactic positions in which the referential properties of an overt one the CONTROLLER determine the referential properties of a non-overt one the CONTROLLEE (Bresnan 1982). In the canonical case of FORWARD CONTROL the controller is structurally superior to the controllee (represented atheoretically as ) (1a). However a second pattern which we will call BACKWARD CONTROL is also possible in which the controllee is structurally superior to the controller (1b). (1) a. b. The diver i tried i to hold his breath. i tried the diver i to hold his breath. (hypothetical) Backward Control has been reported in the literature for Japanese causatives and tokoro -clauses (Kuroda 1965 1978 Harada 1973) Brazilian Portuguese causatives (Farrell 1995) and Nakh-Daghestanian aspectual verbs (Kibrik 1999 Polinsky and Potsdam 2000 2001a). This paper contributes to the growing body of empirical evidence supporting the existence of Backward Control (BC) by documenting and analyzing the construction in Malagasy (see Polinsky and Potsdam 2001a b for a fuller treatment). The paper is organized as follows: Section 2 presents the Malagasy construction and lays out our structural proposal. Sections 3 and 4 argue against two analytical alternatives. Section 5 summarizes our empirical findings and sets the stage for possible theoretical analyses. To that end section 6 attempts to analyze BC within a PRO-based Principles and Parameters (P&P) framework and concludes that P&P quite generally rules out BC. Section 7 turns to a more recent Minimalist-based movement analysis of Control in which the controllercontrollee relationship is a result of movement. Our fundamental claim is that movement in Forward Control is overt while movement in the case of Backward Control is covert. BC is thus expected given widely recognized overt versus covert parametric variation in movement. * We would like to thank our Malagasy consultants Noro Brady and Solange Green for their help with the data. For numerous discussions we thank Stan Dubinsky Norbert Hornstein Ed Keenan Yuki Kuroda Ileana Paul Matt Pearson and the audiences at the Austronesian Formal Linguistic Association 8 at MIT and the Subtropical Summer Syntax Workshop at the University of Georgia. 1 2. Control in Malagasy Malagasy is an Austronesian language spoken on the island of Madagascar. Its basic word order is VOS but VSO order is also typically possible. A canonical Forward Control construction is given in (2) (Randriamasimanana 1986 Keenan 1976 1995 Law 1995 Paul and Ranaivoson 1998).1 Expectedly VSO word order is also possible (2b).2 (2) a. i Rabe i Rabe m-an-andrana Rabe i m-i-tondra ny fiara i PRES -ACT-try Rabe PRES -ACT-drive the car Rabe is trying to drive the car. b. m-an-andrana m-i-tondra PRES -ACT-try PRES -ACT-drive the car ny fiara For such examples and VOS order in general we assume a clause structure as in (3) following Guilfoyle Hung and Travis 1992. The predicate constitutes a VP and the clause-final subject occupies a rightward specifier of IP. 3 (3) IP 3 I DPi 3 Rabe I VP 3 V IP try 3 I i 3 I VP # drive the car The verbs manomboka begin mahavita accomplish and mitsahatra stop appear in superficially similar sentences (4a).4 Upon closer investigation however these verbs behave rather differently from Forward Control verbs. One difference is that they do not permit VSO order *(4b). 1 The exact analysis of such structures is a topic of current debate. Law 1995 argues that they do not instantiate canonical control. Note for example that the embedded clause is tensed Malagasy has no morphological infinitives. 2 We use the following abbreviations in glossing: A CC accusative ACT active CIRC circumstantial C O M P complementizer F U T future IMPER imperative NEG negation OBL oblique PASS passive PRES present Q question. 3 For alternative Malagasy clause structure see MacLaughlin 1995 Pensalfini 1995 Paul 1999 and Pearson 2001. We will call the clause-final DP the subject although nothing hinges on this terminology. 4 We illustrate only with manomboka begin . The other verbs behave similarly although some data on mitsahatra (Keenan 1995:195-196) suggest possible dialectal variation. 2 (4) a. m-an-omboka m-i-tondra fiara car *m-an-omboka Rabe m-i-tondra ny PRES -ACT-begin Rabe PRES -ACT-drive the Rabe is beginning to drive the car. PRES -ACT-begin PRES -ACT-drive b. ny the Rabe Rabe fiara car We will argue that this and other contrasts between m anandrana try and manomboka begin arise from distinct clausal organizations. We claim that begin accomplish and stop

INFO HASH : 9313ec6408d3d7ef40c1959434d4745251881993

Download now

File Size: 71.6 kb