Verb-Stranding Ellipsis: The View from Mayan

In recent years, there has been a growing debate around a type of ellipsis where the verb is the element surviving ellipsis with a complement going unpronounced akin to English Verb Phrase Ellipsis (VPE) under an auxiliary: John went to the store and Sally did too. This phenomenon has been dubbed “Verb Stranding VPE” (Goldberg 2005) but a line of inquiry has called into question whether VPE is actually involved in the derivation of verb-stranding phenomena cross-linguistically (Landau 2018, 2020a,b, 2021, 2023 cf. Gribanova 2020; Portelance 2020). The present study addresses this question from the lens of a Guatemalan Mayan language, K’iche’ and the Mayan language family at large, which has been notably absent from prior discussions of verb-stranding ellipsis. Using data from original fieldwork, I argue that K’iche’, and more broadly, the Mayan languages, constitute evidence against an approach that uses VPE to derive all instances of verb-stranding ellipsis (e.g. Goldberg 2005). The verb-stranding observed in Mayan must be derived via TPE with ellipsis conditioned by Polarity, a head high in the clausal spine whose complement contains the subject and the object (cf. McCloskey 2017 for Irish; Gribanova 2017 for Russian). However, there exists a second type of stranding ellipsis that occurs under a progressive aspectual marker. Here, the verb is elided in the general case but may be pronounced to the exclusion of other material. I propose that these cases instantiate a derivation closer to VPE, with ellipsis of VoiceP. The resulting picture is a novel view of K’iche’ clause structure and supports theories that take the syntactic representation of Polarity to be a key player in the derivation of ellipsis (e.g. Liptak 2012; Martins 2016; Gribanova 2017; Landau 2023).