Dispersion in the Lexicon
We explore the hypothesis that the global organization of the lexicon maximizes the perceptual distinctness of words, by preferentially relying on highly perceptible contrasts, even when the phonotactics of the language permit less perceptible ones. We find that distinctions among words predominantly rely either on few highly perceptible contrasts (e.g. /tɪn/ vs. /kɪn/ with a place contrast in prevocalic position) or on many globally distributed contrasts, with multiple differences keeping the words apart (e.g. /sɪts/ vs. /ɹɑks/ with a postvocalic place contrast further disambiguated by other differences). We present evidence for this hypothesis from a study investigating the typology of minimal pairs in 58 languages from 23 major language families and a study relating the frequency of English minimal pairs to patterns of perceptual confusion.
The languages of the world, despite many structural differences, exhibit properties that make them suitable for communication. Here we explore the hypothesis that the phonological lexicon of languages is partially shaped by a bias against word confusability. Perception experiments have provided evidence that the recognition of consonant-vowel-consonant (CVC) sequences is impaired if the two consonants share certain phonological features. We present cross-linguistic evidence from the lexica of eleven languages from seven major language families that i) CVC sequences where the two consonants share phonological features are under-attested and ii) the degree of their under-attestation is predicted by their relative confusability in comprehension studies.
Constraints on Predication
Are natural languages subject to universal constraints on lexical meaning? Such constraints, if they exist, would restrict the relation between conceptual representation and linguistic expression, and it is not obvious why this relation should be restricted at the lexical level. After all, lexical items encode arbitrary associations of form and meaning; at least in principle, any concept might be associated with a linguistic form. We propose such a constraint on possible lexical meanings, applying to all natural language predicates of conjoinable type. The basic idea is that the meanings of lexical predicates are constrained by restrictions on the set of individuals relevant to their truth. We advance the following generalization: The only individuals relevant to the truth-conditions of an expression involving a natural language predicate are the individuals identified by its arguments. We say that natural language predicates are “myopic” in the sense that they do not see beyond the entities denoted by their arguments.
What does it mean to compare sets of objects along a scale, for example by saying “the men are taller than the women”? We explore comparison of pluralities in two experiments, eliciting comparison judgments while varying the properties of the members of each set. We find that a plurality is judged as “bigger” when the mean size of its members is larger than the mean size of the competing plurality. These results are incompatible with previous accounts, in which plural comparison is inferred from many instances of singular comparison between the members of the sets (Matushansky and Ruys, 2006). Our results suggest the need for a type of predication that ascribes properties to plural entities, not just individuals, based on aggregate statistics of their members. More generally, these results support the idea that sets and their properties are actively represented as single units.
Reality TV and Language Change
We are building a large-scale corpus of the trajectories of different phonetic and phonological parameters over the course of three months in a diverse group of speakers situated in a linguistically and socially closed system: Season 9 of the reality-television show Big Brother (Channel 4, United Kingdom). This show offers a unique opportunity to study medium term phonetic change in individuals, allowing us to assess both cause and medium-term effect of time-dependent patterns in speech. VOT and t-deletion measurements in this data-set show several linguistic effects consistent with previous work. Additionally, change in the means of different speakers is non-linear over time, with speakers’ trajectories neither fluctuating around a single mean, nor drifting in a single direction throughout the observed period; we also present preliminary evidence that social factors can drive phonetic change in individuals over extended periods of time.
