Search

In this paper we bring evidence from English and Italian deverbal zero nominals (to climb > the climb-Ø N) that zero is a possible spell-out of a nominalizer otherwise overtly instantiated in suffixed nominals (examin-ation ). We argue in favor of a Distributed Morphology approach, a separationist theory that recognizes and easily implements zero morphology with underlying syntax-semantics. Abstracting away from other theoretical trends and their foundational reasons to refrain from using zero suffixes, we address three properties that have been argued to fundamentally distinguish zero nominals from overtly suffixed nominals, with the implication that they instantiate a different word formation process: i) realization of verbal argument structure, ii) possibility of embedding verbalizing suffixes, and iii) semantic transparency in relation to the verb. By means of corpus data and two manually collected datasets of 561 English and 174 Italian zero nominals based on lexicographic information, we bring solid evidence against these claims, by arguing that: i) a great proportion of zero nominals do realize verbal argument structure, ii) the zero affix may embed verbalizing affixes within the limits of the selectional restrictions it independently imposes on its base, and iii) zero nominals present the same patterns of polysemy that suffixed nominals display. However, we show that zero nominals also present some idiosyncrasies to the extent that not all of them express compositional event readings with argument structure, a matter that deserves further research within the frame of their competition with suffixed nominals.

In this paper we offer an overview of the linguistic phenomena that have traditionally been handled by means of zero affixes and of the theoretical debate around the advantages and disadvantages of employing such null morphemes in theoretical modeling. While the advantage of positing zero affixes is straightforward from an empirical perspective (see inflectional syncretism and affixless category change, among others), their theoretical legitimacy has been controversially debated for several decades. In this overview, we present the main problems that have been brought forward against zero affixation and some of the mechanisms that have been proposed as an alternative. Finally, we show how the different articles in our edited collection contribute to this debate by addressing the following three research questions: (1) How do current theories of derivational morphology deal with linguistic phenomena that seem to involve zero affixes? (2) How do zero derivational affixes compare with their “overt analogues” (i. e. overt derivational affixes)? (3) What insights can we obtain from language-specific properties, typological generalizations and/or larger data-oriented studies on the question whether zero affixation is necessary in derivational morphology?

This paper offers an overview of the variety of phenomena that can be subsumed by conversion and introduces diagnostic criteria to distinguish between theoreticallymotivated subtypes and corresponding analytical means to model them. By taking a syntactic approach to these formations and focusing on deverbal and deadjectival conversion nouns and nominalizations, I will argue for three types of conversions, determined by the complexity of their internal structure: root-based conversions, which deserve an underspecified analysis; morphological conversions, which resemble suffixed nominalizations; and syntactic conversions, which show no internal nominal properties, just the external distribution of nouns. The broader point I aim to make is that a purely word-based morphological approach to conversion will miss important distinctions between morphological and syntactic conversion, which can only be detected if the syntactic behavior of these formations is also closely examined. [ABSTRACT FROM AUTHOR]

This paper is concerned with the morphosyntax of zero-derived nominals based on verbs with postposed and prefixed particles (e.g. to bail out > the bailout-;; to override > the override-;). In general, zeroderived nominals have received less attention than suffix-based nominals (cf. the climb-ing, the examination, the assign-ment). In the generative literature, in particular, after Grimshaw's (1990) insights on the ability of suffix-based nominals to inherit verbal event and argument structure, zero-derived nouns have been claimed to lack such properties and have been syntactically analyzed as derived not from a verb but from an uncategorized root, as modeled in Borer (2013). Drawing on data from natural text corpora and following Rappaport-Hovav and Levin's (1998) theory of event structure and argument realization, I show that zero-derived nouns derived from verbs with postposed and prefixed particles may realize argument structure on their event readings, which can only come about from the event structure of their base verbs. [ABSTRACT FROM AUTHOR]

This paper is concerned with nominalizations derived from psychological verbs in English. Based on particular properties in their realization of argument structure, which have long been noticed in the literature, I will argue that in a syntax-based approach to word formation such as Distributed Morphology these nominals must be derived from the psychological root alone and cannot include any event structure. This contrasts with non-psych nominals, which more readily include verb event structure. I will show that this difference lies in the different ontological status of the two kinds of roots. Furthermore, psychological verbs and their special roots allow us to conclude that there is no structural difference between derived nominals (based on Latinate suffixes such as -al, -ance, -ation, -ion, and -ment) and zero-derived nominals, whose suffix is covert. A clear difference, however, is posited between these nominals and those based on -ing. [ABSTRACT FROM AUTHOR]

The focus of our study is the syntactic structure of Greek and Romanian stative nominalizations that are derived from alternating object-experiencer (OE) verbs (i.e. OE verbs that have subject-experiencer counterparts) and realize the experiencer argument. These verbs are aspectually ambiguous between eventive and stative readings, so the stative nominalization might be derived either from the result state of the event or from the stative eventuality itself. In contrast to what one would expect, we show that the stative value of these nominalizations is not inherited from the stative reading of the corresponding verb, but they are nominalizations of the root phrase whose experiencer argument and stative properties they preserve. [ABSTRACT FROM AUTHOR]