Projecting the Adjective: The Syntax and Semantics of Gradability and Comparison

Outstanding Dissertations in Linguistics

by Christopher Kennedy

This book investigates the core meaning and syntactic distribution of gradable adjectives: adjectives such as long, short, bright and dim, with respect to which objects can be ordered and compared.

259 pages, ebook

First published March 1, 1999

About the author

Christopher Kennedy is the author of Ennui Prophet, Encouragement for a Man Falling to His Death, Trouble with the Machine , and Nietzsche's Horse. He is an associate professor of English at Syracuse University where he directs the MFA Program in Creative Writing.

Reviews

[Nat · Rating 5 out of 5]

This is Chris's impressive dissertation on gradable adjectives. It's packed with examples that Chris handles with what, after working through his arguments, seems a more intuitive analysis than the competitors. Chris argues that gradable adjectives, like "tall" or "long", or anything that can be combined with degree morphology or figure in comparative constructions (Nat is taller than Wyeth; Nat is surprisingly tall, etc.) are functions from objects to degrees on a scale--which scale is determined by the comparative adjective. So "tall" is a function from objects to degrees on the height scale. "Long" is ambiguous; the lexical entry for "long" includes at least three different scales onto which objects can be "projected": the linear extent scale, the temporal duration scale, and the "page number" scale (for printed matter). (I would quibble with this last denotation for "long"--a book is long just in case it contains a lot of words, not if it has a lot of pages. A book with enormous print and a correspondingly large number of pages isn't necessarily long, whereas a book with comparatively few pages that is packed with small print can be long.)

If you assume (as Chris, and most semanticists do) that the denotation of a sentence is a truth-value, you need something else in addition to an object and a function from objects to degrees to get to a truth-value. You need something that takes you from degrees to truth-values. In a comparative sentence like "Nat is taller than Wyeth", it's clear what does the additional work: the comparative morpheme "-er" and "than Wyeth". Chris treats the comparative morpheme as a relation between two different values: a standard value, which is a degree on a scale set by the thing against which my height is being measured (in this case, Wyeth), and what Chris calls a reference value, which in this case is set by me. The sentence "Nat is taller than Wyeth" therefore has a meaning composed of the denotation of the comparative morpheme, the degree set by Wyeth, the scale determined by the comparative adjective (the height scale), and the degree onto which I am projected on that scale. The specific meaning of "-er" is the relation >, so that the sentence as a whole is true just in case the degree on the height scale that I'm projected onto is greater than the degree to which Wyeth is projected on the height scale.

Absolute constructions, like "Nat is tall" are handled in the same way--but there has to be a null degree morpheme that does the work of taking the degree on the height scale I'm projected onto and turning it into a truth-value. Basically (this gets complicated later on), the null degree morpheme is the relation ≥ between the degree I'm projected onto and some contextually determined degree d. So "Nat is tall" is true just in case the degree on the height scale I'm projected onto is greater than or equal to some contextually determined degree on the height scale.

This analysis of gradable adjectives has all kinds of advantages and explanatory payoffs that I won't spoil by describing in detail.