Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Reading Group

Reading group: Kit Fine, ‘Vagueness, Truth and Logic’ (1975)

Oskar Nyberg Fellow
· 4 replies
Meaning & languageAnalyticKit Fine#penumbral-connections#reading-group#supervaluationism#vagueness
Follow

Vagueness is my obsession, and I've realised that many people know supervaluationism only from summaries. The original is better than the summaries. So: a slow read of Kit Fine's "Vagueness, Truth and Logic", first published in Synthese in 1975 and reprinted in Keefe and Smith's anthology Vagueness: A Reader (1996).

The paper is dense but not long. I suggest three sessions, roughly a third each:

  • Session 1: the setup. Fine treats vagueness as a deficiency of meaning rather than a gap in our knowledge, and introduces the idea of ways of making a vague expression precise. What counts as an admissible precisification?
  • Session 2: the case against the alternatives. Why truth-functional many-valued approaches mishandle "penumbral connections" (the logical ties between vague claims), and how the supervaluational idea, true on all admissible precisifications, keeps them.
  • Session 3: higher-order vagueness and "definitely". What happens when the boundary between admissible and inadmissible precisifications is itself vague.

Reports per session below. Bring objections; the paper has had fifty years of them.

LikeReply
Follow4 replies

Join the conversation: members reply here, and joining is free. Reading is always free.

Join free to reply Sign in

4 replies

  1. Tomasz Szewczak

    Fellow

    Session 1. What struck me was how much hangs on the "deficiency of meaning" framing. If the word "heap" simply hasn't been given a sharp boundary, then there is nothing to know, and the epistemicist's hidden boundary has nowhere to hide. Fine doesn't argue against epistemicism at length (it wasn't yet the live rival it later became), but the framing already rules it out.

    Helpful · 2
  2. Vikram Sethi

    Fellow

    Session 1, a teacher's note. The idea of "admissible" precisifications matched something I do when marking. "Show your working clearly" is vague, but there are clearly inadmissible readings (no working at all) and a range of acceptable ones. I mark a script as passing that criterion if it passes on every reading I'd defend. Without knowing it, I've been supervaluating.

    Helpful · 3
  3. Ana Clara Ferreira

    Fellow

    Session 2. The penumbral connection argument is the heart of it. If a colour patch is borderline red and borderline orange, "it's red or orange" should be true even though neither disjunct is. Any logic where the value of the whole depends only on the values of the parts can't deliver that. Supervaluationism delivers it because "red or orange" is true on every admissible sharpening.

    The cost I'd flag: classical validity survives, but some classical rules don't behave as usual once you add a "definitely" operator. Session 3 will need that.

  4. Oskar Nyberg

    Fellow

    Session 3. Ana Clara's flag is right: once you add "definitely", some familiar rules (reasoning by cases, for instance) need restrictions. Fine is aware of higher-order vagueness and builds structure for it, but the critics' best objection is the one my students always raise in the end: supervaluationism makes "there is a last heap number" true, though no number is the last. Fine thinks that's acceptable once you see what super-truth is. I'm still unsure.

    Helpful · 1