An Explanation of the Veridical Uniformity Universal

Abstract

A semantic universal, which we here dub the Veridical Uniformity Universal, has recently been argued to hold of responsive verbs (those that take both declarative and interrogative complements). This paper offers a preliminary explanation of this universal: verbs satisfying it are easier to learn than those that do not. This claim is supported by a computational experiment using artificial neural networks, mirroring a recent proposal for explaining semantic universals of quantifiers. This preliminary study opens up many avenues for future work on explaining semantic universals more generally, which are discussed in the conclusion.

Prosodically-conditioned factive inferences in Korean: An experimental study

“…The third involves the landscape of EC (projection) variation in languages like Korean, which is characterized by systematic UC variation. The last issue involves the veridicality behaviors of Korean verbs in the domain of interrogative complements (Spector & Egré 2015, Theiler, Roelofsen & Aloni 2018, Steinert-Threlkeld 2020, Roelofsen & Uegaki 2020. On a first approximation, al-type verbs appear to display typologically unexpected patterns (they are selectively veridical with p, but obligatorily veridical with Q complements) worth investigating further.…”

Monotone Quantifiers Emerge via Iterated Learning

“…Beyond the universals of quantification, neural networks have been shown to more easily learn meanings satisfying other semantic universals, such as veridical uniformity (Steinert‐Threlkeld, 2019 ). 17 An iterated learning extension of such models could also offer insights in the way individual biases are reflected on a population of neural network agents.…”

“…Neural networks offer one such model of learning. The previous work has addressed the learnability challenge by showing that quantifiers, responsive predicates, and color terms that satisfy certain semantic universals are easier to learn than ones that do not for neural networks (Steinert‐Threlkeld, 2019 ; Steinert‐Threlkeld & Szymanik, 2019 , 2020 ).…”

Searching for a universal constraint on the possible denotations of clause-embedding predicates

“…Intuitively, this predicate means know when taking an interrogative complement and be opinionated when taking a declarative complement. Steinert-Threlkeld (2020) gives the following lexical entry:…”

“…Recent work has explored semantic universals in the domain of clause-embedding predicates like know, agree, and wonder (Spector & Egré 2015;Theiler, Roelofsen & Aloni 2018;Uegaki 2019;Steinert-Threlkeld 2020). Within this line of work, two basic questions can be distinguished.…”