Jens Kipper, an assistant professor of philosophy, will present "Why Vagueness Persists but Doesn't Proliferate in Natural Language."

Abstract

Barton Lipman famously argued that vague languages cannot be game-theoretically optimal. This creates a puzzle: if vagueness is of no use to language users, then why is it so prevalent?

In this paper, we use signaling games to address this puzzle. In extant work involving signaling games, vagueness is modeled as a mixed (i.e., non-deterministic) signaling strategy. We argue that this modeling criterion is inadequate and instead propose to model vagueness as the presence of a region in the state space in which signaling is (almost) absent and which is located between two regions of signaling. We show that on this understanding, vague languages can be optimal. Since standard signaling games don't capture the conditions in which vague expressions are learned in the real world, we go on to introduce a new type of signaling game in which agents have to learn a signaling strategy from a limited sample of usage data of a previous generation. Given various common learning algorithms, we observe that vagueness is created and remains stable for many generations. We also demonstrate analytically that vagueness evolves towards a steady state in the conditions we identify. These findings suggest that vagueness arises and stabilizes naturally when speakers try to coordinate signaling strategies based on sparse data in large state spaces.

About the Speaker

Jens Kipper is an assistant professor of philosophy. Before joining the Rochester faculty, he was an assistant professor at the University of Bielefeld in Germany, where he taught undergraduate courses on topics such as the philosophy of language, perception, and scientific explanation. He also led graduate seminars on consciousness, mental content, and philosophy and science fiction, among other subjects.

A specialist in epistemology—the theory of knowledge—as well as philosophy of mind and philosophy of language, Kipper earned a PhD in philosophy at the University of Cologne in 2012. His work also involves the philosophy of artificial intelligence, the philosophy of science, and applied ethics.

He is the author of two books: Research Ethics: An Introduction, with Thomas Fuchs and others (Metzler), and A Two-Dimensionalist Guide to Conceptual Analysis (Ontos). In 2015–16, Kipper carried out an 18-month project that explored how we can know what is necessary or possible compared with the thesis that our mental states (such as beliefs, desires, or sense experiences) represent only things internal to us and not our environment. He conducted the research at MIT, with the support of a research fellowship from the German Research Foundation.