Word formation is syntactic: Raising in nominalizations

Abstract

According to Chomsky (1970), raising to subject and raising to object may not take place inside nominalizations. This claim has largely been accepted as fact ever since. For instance, Newmeyer (2009) repeats the claim as crucial evidence for the Lexicalist Hypothesis, the view that word formation takes place in a component of the grammar separate from the phrasal syntax. This paper shows with attested examples and survey data that the claim is false: raising to subject and raising to object are both grammatical inside nominalizations. This argues for a purely syntactic model of word formation, and against Lexicalist accounts. Additionally, the paper shows that one argument against syntactic accounts of nominalization, that from coordination, does not go through, clearing the way for the most parsimonious type of theory: one with only one combinatorial component, not two distinct ones for phrases versus words.

Locative Inversion, PP Topicalization, and Weak Crossover in English

“…For one case involving reconstruction for Binding Principle C, see Adger et al (2016) and Bruening & Al Khalaf (2019). Other cases include multiple questions (Clifton, Fanselow & Frazier 2006), adjectival passives of raising to object verbs (Bruening 2014), and nominalizations of raising verbs (raising to subject and raising to object; see Bruening 2018). It is becoming increasingly important for linguists to support any judgments that they report in their work using multiple sources of evidence, such as corpus evidence and survey results.…”

Deverbal nominals possess no internal VP structure

“…A significant body of linguistic research carried out in the past several decades has argued that deverbal nominals with a morphological verbal base and PROCESS as their semantic character contain internal verbal phrase (VP) structure (Bruening, 2018a(Bruening, , 2018bBorer, 2003Borer, , 2013Roeper, 2005;Fu, Roeper, & Borer, 2001;Alexiadou & Grimshaw, 2008;Alexiadou, 2009, etc.). An illustration of this claim could be the following simplified syntactic tree with the deverbal nominal 'explanation', from an exemplary nominal phrase 'explanation of the question' ('the act of explaining the question').…”

“…The assumption underlying the latter, namely that argument structure for a deverbal nominal is made possible by its verbal base, is based on a comparison with nominals for which argument structure is (argued to be) either inadmissible or not obligatory (see 2.2, but also Borer, 2003Borer, , 2013Alexiadou, 2009;Alexiadou & Grimshaw, 2008). In the literature this claim is known to issue from "Distributed Morphology" (Bruening, 2018a(Bruening, , 2018bAlexiadou, 2009, etc. ) and "The Exoskeletal Model" (Borer, 2003(Borer, , 2013, two similar theories with certain theory-internal differences.…”

“…formed from a morphological verbal base and denoting a process belong rather in the lexical category of the verb than the noun (Alexiadou & Grimshaw, 2008, among others), while the second is that word formation takes place in deep-level syntax because of which there is no need to assume an independent language level called 'lexicon' (Bruening, 2018a;Borer, 2013, etc.). It was the questionable nature of the first of these two follow-up claims that led to the conception of the present paper.…”

“…They may be referred to also as 'complex event nominals', 'underived nominals' (e.g. Bruening, 2018a), or simply '-(at)ion nominals' which stands also for nominals with related suffixes (-al, -ment, etc.). Nominals in this group are contrasted with 'simple' or 'referential' nominals, such as 'version' and 'violence', which instead tend to denote a referent.…”