Non-subsective adjectives
\[ \newcommand{\expr}[3]{\begin{array}{c} #1 \\ \bbox[lightblue,5px]{#2} \end{array} ⊢ #3} \newcommand{\ct}[1]{\bbox[font-size: 0.8em]{\mathsf{#1}}} \newcommand{\abbr}[1]{\bbox[transform: scale(0.95)]{\mathtt{#1}}} \def\true{\ct{T}} \def\false{\ct{F}} \]
Non-subsective adjectives can be fit into the same mold. Thus let’s assume that they also have syntactic category \((n/n)\) and semantic type \(((e → t) → (e → t))\).
Vanilla non-subsective adjectives
Non-subsective adjectives which aren’t privative—e.g., alleged—can be given lexical entries like (1).
- \(⟨\textit{alleged}, (λf.(λx.\ct{alleged}(f)(x)))⟩ ⊢ (n/n)\)
This lexical entry helps get at the fact that we don’t observe an entailment concerning Bo’s anthood from the sentence Bo is an alleged ant; in particular, (1) helps explain this fact by not including ‘\(∧ f(x)\)’ after ‘\(\ct{alleged}(f)(x)\)’.
Privative adjectives
Privative adjectives contribute the entailment that the modified noun does not hold of an entity of which the whole noun phrase is predicated. To account for this fact, we can give a privative adjective like fake the lexical entry in (2).
- \(⟨\textit{fake}, (λf.(λx.\ct{fake}(f)(x) ∧ ¬f(x)))⟩ ⊢ (n/n)\)
Now if we combine fake with basketball player, we get:
\[⟨\textit{fake basketball player}, (λx.\ct{fake}((λx.\ct{bbp}(x)))(x) ∧ ¬\ct{bbp}(x))⟩ ⊢ n\]