Analyzing Ilokano Pseudoclefts

Several researchers have proposed that cleft constructions in many Austronesian languages are in fact concealed pseudoclefts (Chung, 1998 for Chamorro; Paul, 2001, 2008 for Malagasy; Georgopolous, 1991 for Palauan, among others). What this paper examines is the syntactic structure of pseudoclefts in Ilokano, a VSO Austronesian language spoken in the Northern Philippines. I argue that the language employs two types of pseudoclefts, both of which are biclausal. The first type (ti-type or null copula-type pseudocleft) utilizes a null copula between a focused constituent XP and a headless relative introduced by the determiner ti. Thus, we get a construction of the type XP < copula=ø < ti + wh-clause. Despite the lack of an overt wh-phrase, material after the determiner ti contains an operator-variable chain signaled by the 'trigger' morphology, creating a headless relative much like in English and other languages. Many Austronesian languages including Ilokano exhibit the famous 'trigger-only' restriction to A-bar movement (Keenan and Comrie, 1977; Aldridge, 2004), and thus the trigger morphology found on the verb in a headless relative marks the 'role' of the variable. The second type (ket-type pseudocleft) employs the topic particle ket with the word order ti + wh-clause < ket < XP. This time, the headless relative sits in a topicalization position and the constituent after the topic particle ket introduces the focused constituent XP. I argue that the ket-type of pseudocleft is in fact a TOPIC < COMMENT construction where the focus is a full IP subject to optional ellipsis, similar to a type of specificational pseudocleft found in English (cf. den Dikken et al., 2000).