Semantic minimalists adhere to a simple theory of meaning: ‘Jenny’ means Jenny, ‘cut’ means cut and ‘grass’ means grass. One might assume that such a simple theory has difficulties accounting for the phenomenon that most lexical expressions seem to be associated with an often large number of related senses or meanings. Especially contextualists like Recanati have argued that polysemy is evidence against minimalism. In my paper, I first list a number of desiderata for a successful theory of polysemy. I then propose a minimalist theory of polysemy that meets these desiderata.

This talk argues for a novel theory of concepts. Standard approaches in cognitive science take concepts to be structured bodies of information that underlie categorization and inference, such as prototypes or theories. However, the heterogeneity of such representational structures undermines this approach. Recent approaches allow for heterogeneous concepts via "hybridist" (Vicente & Martínez Manrique 2016) or "pluralist" (Weiskopf 2009) views. Another option is Fodorian atomism, on which a concept is an unstructured vehicle with a denotation. Atomism avoids problems faced by hybridist and pluralist views (e.g., compositionality). One problem for atomism is its lack of interaction with the informational structures that are known empirically to underlie categorization and inference. Another problem is the phenomenon of polysemy, wherein single word meanings (and therefore single concepts) can shift their denotations. This talk argues that concepts function as "generative pointers"--i.e., unstructured atoms that point to locations in memory where bodies of information are stored, and the selection of particular subsets of this information can modulate the denotation of the token deployment of the concept. This theory interacts with recent developments in lexical semantics, and provides a form of atomism that moves away from a rigid view of conceptual and semantic reference and takes a less polemical view toward the cognitive science of concepts than classic Fodorian atomism.

Most of our concepts seem to be either purely descriptive, e.g., cake, to swim, or purely normative, e.g., great, ought. The class of thick concepts, such as generous and rude, has long been thought to provide the only exception. Within the past few years, researchers have both discovered and begun to investigate a new class of partly normative concepts, the so-called dual character concepts (Del Pinal & Reuter 2017, Knobe et al. 2013, Leslie 2015). Dual character concepts are unique in that they have two related but independent dimensions for categorization: one descriptive, one normative. Although ample theoretical work has been published on thick concepts, no empirical studies have so far been performed to test the extent to which thick concepts are considered to have a normative component. I therefore designed an empirical study the purpose of which was two-fold: First, the study suggests a way of operationalizing the notion of thickness to determine the normative element of thick concepts. Second, the study was designed to demonstrate the relative independence of the normative thickness from the normative dimension of a dual character concept. Studying thick concepts empirically allows us to examine possible ways in which the normative component of thick concepts and the normative component of dual character concepts interact.

A long-standing datum in cognitive science is that people make semantic inferences, which draw on the meaning of concepts, as well as purely syntactic inferences, which don’t. That contrast is puzzling since the representational theory of mind (RTM) assumes that all inferences are a matter of causal transitions between representational vehicles in virtue of non-semantic properties. Semantic inferences used to be picked out as those that draw on the internal structure of a concept. However, experimental work on concepts has produced a near-consensus that, for a typical lexical concept, there is no single representational structure which is always involved in thinking with that concept. At the same time, the recent conspicuous success of deep convolutional neural networks in modelling various categorisation tasks suggests that much of the information we draw on when using a concept is not conceptually or explicitly represented at all, but is instead implicit in dispositions to apply the concept on the basis of non-conceptual representations. This new landscape has many attractions, however the old contrast between syntactic and semantic inferences seems to have been squeezed out. Can we still explain the contrast, within the strictures of RTM? This paper argues that we can, not by appealing to conceptual structure, but by making a novel distinction between two types of representational processing in which concepts are involved.

Mereological theories of inherent polysemous words (see Arapinis 2015; Gotham 2016) propose that copredicative noums like book or school denote mereological compounds when they appear in copredicative sentences -for example: The book is heavy and interesting; the school was in the middle of the party when it caught fire-.  According to mereological theories, the word school denotes  something that is a building and a group of people and the word book denotes something that is physical and abstract. In this talk, I will give some objections to these theories: (i) the persistence conditions of the whole when one of its parts disappears or is destroyed; (ii) the quantification or counting criteria of complex objects in sentences like: There are three heavy books on the shelf; there are three interesting books; there are three interesting and heavy books.  Finally, I will argue that the proposal of activation packages about polysemy (see Ortega-Andrés & Vicente 2019) solves these puzzles.