Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Reading Group

Reading group: Quine’s ‘On What There Is’, slowly, in three sittings

Tomasz Wójcik Fellow
· 6 replies · 186 views
Existence & beingAnalyticBertrand RussellPlatoW. V. O. Quine#nominalism#ontological-commitment#quine#reading-group
Follow

"On What There Is" first appeared in the Review of Metaphysics in 1948 and opens From a Logical Point of View (1953). It is short, funny in places, and much quoted by people who have not read it. I propose we read it slowly in three sittings, a week or so apart.

Sitting 1: the puzzle of non-being. Plato's beard, Pegasus, and the two imaginary philosophers McX and Wyman, who think Pegasus must be something in order to be denied. Ask yourself whether Quine is fair to them.

Sitting 2: the solution and the slogan. Russell's theory of descriptions, the move from the name "Pegasus" to the predicate "pegasizes", and the criterion: to be is to be the value of a variable. Ask what exactly the criterion measures. It tells us what a theory says there is, not what there is.

Sitting 3: universals and the choice of a scheme. McX returns, now arguing for redness. Then Quine maps the medieval positions on universals onto logicism, intuitionism and formalism in the philosophy of mathematics, and ends with ontology as part of choosing a conceptual scheme, judged partly by simplicity.

Any edition is fine; the text is the same. Please post when you have read each section, with one thing that struck you and one thing you doubt. I will start with my own doubts once a few people have posted.

LikeReply
Follow6 replies

Members reply here. Reading is always free.

Sign in to reply

6 replies

  1. Giulia Ferrante

    Fellow

    Sitting 1. What I enjoyed is that Quine treats the puzzle as Plato did in the Sophist, as a real difficulty about speech, not a confusion to be laughed off. Plato's Eleatic Stranger solved it by treating "what is not" as "what is different from", so that saying something is not is saying it is other than something else. Quine's solution is quite different, but he starts from the same spot. I teach the Sophist to seventeen-year-olds and might steal "Plato's beard" for them.

  2. Pieter de Jong

    Fellow

    Sitting 2. As someone who refactors code for a living, the move from "Pegasus" to "the thing that pegasizes" looks like turning a hard-coded constant into a query. A query can return nothing without crashing; a constant that points nowhere is a bug. I doubt, though, that the trick generalises as cleanly as he suggests. "Pegasizes" is a predicate nobody understands except as "is Pegasus", so it feels like we have moved the problem rather than dissolved it.

    Helpful · 1
  3. Oskar Nyberg

    Fellow

    Sitting 3, on McX and redness. Quine's reply is that predicates need not name anything: "there are red houses" commits you to houses, not to redness. Fair. But I find it worth recording that Quine himself later accepted classes (sets) because mathematics as used in science quantifies over them. He called this, in effect, a reluctant concession. So the essay's nominalist mood is not the final Quine.

    Pieter, I share your doubt about "pegasizes". Quine admits the predicate is artificial; his point is only that it can be introduced without presupposing anything that does the pegasizing.

    Helpful · 2
  4. Henrike Vogel

    Fellow

    The ending surprised me. Choosing a conceptual scheme for its simplicity and usefulness sounds almost Kantian, except that Kant thought the categories were fixed for any finite understanding, whereas Quine treats the scheme as revisable like any other theory. I would like to know whether Quine had read the first Critique closely or was arguing past it.

    Helpful · 3
  5. Derek Hollis

    Fellow

    Wrap-up from me. What I most value is that the criterion makes commitments checkable: you can look at a regimented theory and say what it needs. In my day job that is exactly what a specification does. What I doubt is the last step, that ontology is in the end a matter of which scheme is simplest. Simplicity for whom, and measured how? We argue about that in software too, and it is rarely settled by logic.

  6. Tomasz Wójcik

    Fellow

    Thank you all. My own doubt, for the record, is the one Henrike raises from the other direction: if the scheme is chosen, in what sense does it tell us what there is rather than how we prefer to talk? That is Carnap's challenge, and I suggest we read his 1950 essay next, as a reply.

    Helpful · 1