窪薗 晴夫 (国立国語研究所 理論・対照研究領域 教授)
文法研究班 「名詞修飾表現」
プラシャント・パルデシ (国立国語研究所 理論・対照研究領域 教授)
2019年12月24日 (火) 10:00~12:30
The Department of Linguistics, University of Mumbai, India



What is nominalization? —Towards the theoretical foundations of nominalization— SHIBATANI Masayoshi (Rice University)

Despite the philosopher-linguist Zeno Vendler’s point made more than four decades ago that “the grammar of nominalizations is a centrally important part of linguistic theory” (Vendler 1976:125), linguistic theory has made little progress toward a satisfactory understanding of the phenomenon of nominalization, and how nominalizations, as the products of this process, function in grammar. This seminar attempts to lay theoretical foundations of nominalization, characterize major types of nominalizations, and to explicate how they function in grammar. Proper understandings of these undermine many theoretical assumptions held in the current analysis of such prominent phenomena collectively dabbed “subordinate clauses”, including relativization and complementation. The problems are not unique to a particular theoretical framework, obtaining equally in typological and functional, as well as formal linguistics.

While nominalization phenomena are receiving increasing attention in recent years, most studies are narrow in scope, are shallow in the analysis, and overall lack a proper understanding of the relationship between grammatical structures and their uses. None of the studies delve into the cognitive underpinning of the nominalization process, preventing them from gaining an insight and understanding of the semantic properties of nominalizations. We demonstrate that nominalization is a metonymic process by which a range of denotations is potentially evoked based on the nominalization structures. Just like ordinary metonymic expressions (e.g., “Drink a glass a day”, “India defeated Ireland 5-2”), verbal-based nominalizations, for example, denote a wide range of substantives metonymically related to verbal events, such as Event protagonists/participants (Agent, Patient, etc.), Resultant product, Instrument, and Location, and the precise denotation is contextually determined per the Gricean Cooperative Principle.

We distinguish between lexical and grammatical nominalizations, and between event nominalizations and argument nominalizations. We then demonstrate that grammatical nominalizations, like ordinary nouns, have two major uses; namely, an NP-use, where a nominalization heads an NP and a modification-use, where a nominalization modifies a noun head within an NP. We argue that so-called “complements” and “relative clauses” are simply an NP-use of event nominalizations and a modification-use of argument nominalizations, respectively. One of the major claims made is that there are no distinct structures such as “(clausal) complements” and “relative clauses” apart from grammatical nominalizations. Briefly discussed here is the distinction among nominalizations, clauses, and sentences—one of the most neglected topics in modern linguistic theory.

We then take up the topic of nominal-based nominalizations, where we argue that so-called genitive/possessive forms, such as mine and John’s in English and other languages are in fact nominalizations of nominals, i.e., pronouns, nouns, and noun phrases. Just like verbal-based nominalizations, nominal-based nominalizations have two major uses of NP-use and modification-use. Our analysis obviates such parts of speech distinctions as “possessive pronouns” (e.g., mine, his) and “possessive adjectives” (e.g., my book, his book)—the former are an NP-use of pronoun-based nominalizations and the latter a modification-use of the same. Included here is a discussion on how so-called “genitive case” differs from true case forms both semantically and syntactically, justifying our treatment of it as a nominal-based nominalization. We show that (numeral) classifiers are also nominalizers for numerals and other parts of speech in many languages of the world. The importance of nominal-based nominalizations is demonstrated in terms of the spread of nominalization markers, which often starts out in nominal-based nominalizations and then spreads to verbal-based nominalizations.

Among the far-reaching theoretical conclusions reached by this study is that the various types of relative clauses (e.g., free/headless RCs, internally-headed RCs, restrictive RCs) recognized in the typological literature are actually no more than epiphenomena arising from the different uses of grammatical nominalizations. This and other pertinent issues discussed are illustrated and supported by a wide-spectrum of language data from around the globe, including Indic and Dravidian languages.