The conference runs from 10:00-18:30 on Friday 30th of June and Saturday 1st of July and consists of four thematic panels and six talks (details below).
SCHEDULE
Day 1:
9:30-10 arrival and gathering
10:00-10:10 opening remarks
10:10-11:00 Richard Holton: Knowing and Telling
11:00-11:30 coffee break
11:30-13:00 Philosophy of Action and Epistemology Panel:
Jeremy Goodman: The Myth of Full Belief
Natalie Waights Hickman: The Right and The Good; Knowing That and Knowing How
Harvey Lederman: Publicity and Propositional Attitudes
Carlotta Pavese: Knowing How to Solve Kraemer’s Puzzle
13:00-14:30 Sandwich lunch (invited participants only)
14:30-15:20 Mona Simion: Knowledge and Disinformation
15:20-15:30 short break
15:30-16:20 Sarah Moss: A Synthesis View of Counterfactuals
16:20 – 16:50 coffee break
16:50-18:20 Philosophy of Language and Metaphysics Panel:
Daniel Deasy: Permanentist Temporalism
John Hawthorne: Knowledge, Evidence, Reasons
Ofra Magidor: Polysemy and the Ambiguity Tests
Peter Pagin: Can Norms of Assertion be Constitutive?
19:00-19:30 drinks (invited participants only)
19:30 dinner (invited participants only)
Day 2:
9:30-10 arrival and gathering
10:00-10:10: announcements
10:10-11:00 Cian Dorr and Andrew Bacon: Classicism, Weak Classicism, and Necessitism
11:00-11:30 coffee break
11:30-13:00 Logic and Metaphysics Panel:
Peter Fritz: Prospects for Higher-Order Contingentism
Øystein Linnebo: Truly Absolute Generality is Indefinite Generality
Beau Madison Mount: Externalism, Mechanism, and Absolute Provability
Juhani Yli-Vakkuri: Some Results in Semantics
13:00-14:30 Sandwich lunch (invited participants only)
14:30-15:20 Annina Loets and Alexander Roberts: Possibility Plenitude
15:20-15:30 short break
15:30-16:20 Maria Lasonen-Aarnio: Knowledge and Reasonable Belief: The Prescriptive and Hypological
16:20 – 16:50 coffee break
16:50-18:20 Novel Extensions Panel:
Amber Riaz: Moral Learning and Experience (online)
Alexander Bird: The Duhem–Quine Thesis Refuted
Mariona Miyata-Sturm: Elegance and Overfitting
Nico Silins: Indiscriminability and Aesthetic Value
18:20-18:30 Remarks by Tim Williamson
19:30 dinner (invited participants only)
Lunch and dinner is provided for invited participants only: if you haven’t received an email about it, please make your own arrangements
Further participants are
Corine Besson, Michael Blome-Tilmann, Dorothy Edgington, Jane Friedman, Thomas Kroedel, Jennifer Nagel, Thomas Sattig, Levi Spectre, Amia Srinivasan, Bruno Whittle
PANELS
Philosophy of Logic and Mathematics
Øystein Linnebo
Truly Absolute Generality is Indefinite Generality
Williamson’s “Everything” provides a powerful defense of absolute generality. A central argument is that generality relativism is plagued by a serious expressibility deficit. To develop a semantics for absolutely general quantification, Williamson appeals to higher-order logic—in a way that forces him to ascend to higher and higher levels of the type-theoretic hierarchy. This leads to new forms of expressibility deficit, I argue, which are strikingly analogous to the ones that the generality absolutist sought to avoid. I therefore explore the prospects for a form of generality, even more absolute than Williamson’s, which can avoid these residual expressibility deficits. A central idea is that this truly absolute generality is generality across an indefinite universe. Thus, by pursuing an initially Williamsonian line of reasoning, I end up with a strikingly un-Williamsonian conclusion.
Beau Madison Mount
Externalism, Mechanism, and Absolute Provability
In his 1951 Gibbs lecture, Kurt Gödel defended a thesis which has become known as “Gödel’s disjunction”. On Gödel’s view, at least one of the following must be the case: (1) mechanism about the human mind is false or (2) there exists some simple sentence of arithmetic that is true but absolutely unprovable (not merely unprovable within a particular formal system, but unprovable by any means whatsoever).
In a recent paper, Timothy Williamson has argued against Gödel’s disjunction on the basis of an externalist account of mathematical knowledge. Williamson contends that Gödel equivocates on quantifier scope: the appropriate interpretation of the relevant absolute provability claim is not that there can be some human-like agent who knows every simple truth of arithmetic, but that for every simple truth of arithmetic there can be some human-like agent who knows it, and the latter is compatible, given an externalist account of mathematical knowledge, with mechanism about the mind.
I argue that Williamson is correct about the interpretation of the absolute provability claim, but his view faces a different problem. It is a reasonable constraint on absolute provability that the set of mathematical truths that a human-like agent can prove in the limit must be recursively enumerable, and his account of knowledge of mathematical axioms allows for violations of this constraint. The best option for salvaging Williamson-style externalism about mathematical knowledge is to maintain that not every a priori knowable mathematical truth is absolutely provable, but it is far from clear whether this view can be developed in detail.
Peter Fritz
Prospects for Higher-Order Contingentism
In the terminology introduced in Timothy Williamson’s “Modal Logic as Metaphysics” (OUP, 2013), higher-order contingentism is the view that it is not only contingent what individuals there are, but also contingent what propositions, properties and relations there are. In recent work (“Being somehow without (possibly) being something”, forthcoming in Mind), I have argued that higher-order contingentists should hold that even impossibilia can have properties and stand in relations; this is independently motivated and allows the contingentist to express higher-order quantification over mere possibilia. In this talk, I will note that while this quantificational power allows the contingentists to address so-called expressive power challenges, it brings with it a new problem concerning the metasemantics of quantifiers.
Juhani Yli-Vakkuri
Some Results in Semantics
Tarski showed that concepts of the semantics of a language (truth, interpretation, logical consequence, etc.) can be defined in a metalanguage of a higher order. I will discuss some new results on the definability of semantic concepts that do not, unlike Tarski’s, assume that the higher-order logic on which the metatheory is based is an extensional one.
Philosophy of Language and Metaphysics
Daniel Deasy
Permanentist Temporalism
In Modal Logic as Metaphysics (2013), Timothy Williamson argues in favour of combining necessitism with the view that there are contingent truths. As Williamson points out, it is natural to think that analogous arguments speak in favour of combining permanentism with temporalism, the view that there are temporary truths. But this combination of views leaves certain options open. I describe some of the options.
Peter Pagin
Can Norms of Assertion be Constitutive?
Several philosophers, including Tim himself, have claimed that assertion is governed by a constitutive norm. That an utterance of some kind counts as an assertion then consists in its being subject to this norm (e.g. the Knowledge Norm). But what makes the utterance subject to the norm? One plausible answer is that the utterance satisfies certain descriptive conditions. In virtue of satisfying these conditions, the utterance can then be judged as complying or not with the norm. And the problem is that, intuitively, in satisfying these conditions, the utterance already is an assertion. The norm isn’t needed to make it one. This is unlike, say, the moving of a chess piece.
Ofra Magidor
Polysemy and the Ambiguity Tests
A family of familiar linguistic tests purport to help identify when a term is ambiguous. For example, the fact that ‘The ball is spherical and took place at noon’ is zeugmatic suggests that ‘ball’ is used with a different sense in ‘The ball is spherical’ and in ‘The ball took place at noon’. These ambiguity tests are not only of linguistic interest, but also important to other areas of philosophy: a familiar philosophical strategy is to claim that some phenomenon is disunified and its accompanying term is ambiguous. The tests have been used, for example, to evaluate Stanley and Williamson’s claim for the unification of knowledge-that and knowledge-how. Further applications appear in discussions of disunification theses for causation, pain, and wrongness.
These ambiguity tests, however, have come under fire. It has been alleged that the tests fail for polysemy, a common type of ambiguity, and precisely the one that is at issue in philosophically interesting cases. Furthermore, the objection that the tests fail for polysemy is often taken to be an undeniable bit of linguistic data. For example, it is claimed that even though the noun ‘lunch’ has two different but related senses (one denoting some food, and one denoting an event in which the food is consumed), the sentence ‘Lunch was delicious but took hours’ is felicitous, thus demonstrating that the ambiguity test fails for the polysemous noun ‘lunch’.
We argue that this assessment is mistaken. The objection implicitly relies on controversial assumptions about how to account for copredication sentences, in which a single argument is ascribed prima facie incompatible properties. Furthermore, on several viable theories of copredication (including the ones we ourselves defend) the objection fails: polysemy does not fail the ambiguity tests.
(Joint work with David Liebesman)
John Hawthorne
Knowledge, Evidence, Reasons
I discuss some recent challenges to Williamsonian ideas about how knowledge relates to evidence and motivating reasons.
Philosophy of Action and Epistemology
Harvey Lederman
Publicity and Propositional Attitudes
A great deal of social interaction requires coordination. To pass the soup without spilling, people have to coordinate on a moment when the handoff is complete. To avoid collisions on the side-walk, people have to coordinate on which way they’ll go. To have a conversation without interruptions, people have to coordinate on whose turn it is. A standard view holds that coordination in these and other cases depends on, or is at least facilitated by, certain facts becoming public for the relevant people. On one view, these facts become public in the target sense if and only if they are commonly known to the group, where something is commonly known if and only if everyone knows it, everyone knows that they know it, everyone knows that they know that they know it, and so on. Recently, Timothy Williamson has suggested instead that coordination is facilitated not by common knowledge itself, but by a default presumption of common knowledge. In this talk, I’ll critically examine these and other proposals for the target notion of “publicity”, and offer a new argument against them.
Carlotta Pavese
Knowing How to Solve Kraemer’s Puzzle
In certain cases, people judge that agents bring about ends intentionally but that they do not bring about the means that brought about those ends intentionally—even though bringing about the ends and means is just as likely. We call this difference in judgments the “Kraemer effect”. We offer a novel explanation of Kraemer effect: a perceived difference in the extent to which agents know how to bring about the means and the ends explains the Kraemer effect. In four experiments, we replicate the Kraemer effect in a variety of non-moral and moral scenarios, and we find support for our new account. In Experiment 1, we conceptually replicated the Kraemer effect in a variety of non-moral scenarios (Nadelhoffer 2004, 2005), and we found a new effect predicted by our new hypothesis: the know-how effect. We also found that in non-moral scenarios, people’s intentionality judgments that give rise to the Kraemer effect are fully mediated by people’s know-how judgments. In Experiment 2, we found that reducing the know-how effect also reduces the Kraemer effect. In Experiment 3, we rule out a common cause explanation, according to which both Kraemer effect and the know-how effect are independently explained simply by the lack of reliable ways to perform the means. Previous experimental work used morally-loaded scenarios to investigate Kraemer effect (Nadelhoffer 2004, 2005). Experiment 4 looks at morally loaded scenarios: here too, we find both a Know-How effect and a Kraemer effect. We also find that this Know-How effect partially mediates the Kraemer effect. We conclude that know-how explains a lot of the variance in morally-loaded cases similar to those that previously found the Kraemer effect. We end by suggesting that the Know-How Hypothesis provides an overall more satisfying solution to Kraemer’s puzzle than some notable alternatives, and it comports well with a large, burgeoning section of action theory that identifies an important connection between know-how and intentionality (e.g., Malle & Knobe 1997, Stanley and Williamson 2021, Pavese 2020).
Natalia Waights Hickman
The Right and The Good; Knowing That and Knowing How
This recasts the debate over know-how as animated by opposing (teleological vs deontological) conceptions of the foundations of normativity in craft and practical professions, and, through this lens, explores the limits of any attempted rapprochement between Intellectualist and Rylean views. I then present a picture which marries the Intellectualist (deontologist) epistemology of know-how with the Rylean (teleological) metaphysics of practical normativity. This is possible if one recognises values and goods as normatively more fundamental—as setting the ultimate standards of normative evaluation—but knowledge of facts and rules as epistemically prior in the order of learning or hierarchy of competences.
Jeremy Goodman
The Myth of Full Belief
I begin by articulating a broadly knowledge-first and widely (if tacitly) held conception of full(/outright) belief, as a mental state necessary for knowledge, subject to a knowledge norm, and internal to the extent possible (e.g., given content externalism). I then argue that there is no such mental state. On a more positive note, I suggest that a parallel “remembering-first” account of memory belief is much more promising, since in the case of memory belief the relevant kind of internalism is more naturally denied.
Novel Extensions
Amber Riaz (online)
Moral Learning and Experience
Tim has questioned the significance of the a priori/a posteriori distinction in general epistemology. Having applied those ideas to moral epistemology in a separate paper, I am now working on developing my alternative account of moral epistemology. In particular, I explain and defend the view that experience hones and calibrates one’s moral “intuition”.
Nico Silins
Indiscriminability and Aesthetic Value
I argue that even if we can’t tell two objects apart, they can still differ in aesthetic value for us, and I lay out the consequences of this claim for views about aesthetic value.
Alexander Bird
The Duhem–Quine Thesis Refuted
I briefly articulate the Duhem Thesis and its Quinean extension, and then argue that they are false. Hypotheses can be tested (and refuted) singly when all auxiliary hypotheses are known to be true.
Mariona Miyata-Sturm
Elegance and Overfitting
Building on Forster and Sober (1994), Tim has argued (e.g., ‘Abductive Philosophy’, 2016) that simplicity, elegance, and other broadly aesthetic properties are epistemically significant theoretical virtues because they reduce the risks of overfitting. We are less likely to modify theories to fit noise in the data or reject a theory based on faulty data if we stick with aesthetically pleasing theories. This is promising, but important details must be filled out before we can extend the overfitting justification to theory evaluation in general and to broadly aesthetic properties beyond simplicity (like elegance).
I discuss an example from the development of the Vine-Matthews-Morley hypothesis – a cornerstone of the theory of plate tectonics – which shows that feelings of elegance can play a role in finding a good trade-off between simplicity and fit. Relying on such feelings is certainly a looser and less precise method for guarding against overfitting than the one Forster and Sober proposed, but it is also applicable in a wider range of circumstances (in particular, when the algorithmic approach based on the number of independent parameters isn’t applicable). I end by outlining a somewhat speculative framework for explaining how aesthetic feelings can play such an epistemic a significant role which draws on recent work on metacognition.
TALKS
Sarah Moss
A Synthesis View of Counterfactuals
There are two prominent views of counterfactuals in the literature—variably strict conditional accounts developed by Stalnaker and Lewis, and strict conditional accounts defended by von Fintel, Williamson, and others. Unfortunately, both views have problems accounting for ordinary language judgments about counterfactuals. This talk will begin with a brief tour of these problems. In particular, I argue that many strict conditional accounts fail to accommodate our probabilistic judgments about counterfactuals, and I raise questions for recent attempts to explain away these judgments.
Having set out these problems, I introduce a view of counterfactuals that can solve them. This view incorporates a key insight of variably strict accounts—namely, that the antecedent of a counterfactual often influences what worlds are relevant to its truth conditions at a context. However, unlike variably strict accounts, my view preserves a strict conditional semantics according to which Antecedent Strengthening is valid, thereby capturing data that have motivated Williamson and other strict conditional theorists.
Cian Dorr and Andrew Bacon
Classicism, Weak Classicism, and Necessitism
‘Booleanism’ is a collection of natural identities involving the truth-functional connectives, according to which propositions and properties form Boolean algebras under these connectives. There is a natural way of extending these identities to the quantifiers, ‘Classicism’, which entails the thesis of Necessitism defended by Williamson in *Modal Logic as Metaphysics*. In this talk we will argue that Booleans should be Classicists, by showing that under assumptions much weaker than Classicism one can derive the existence of operations conforming to the Classicist identities for the quantifiers, and arguing that if there are such operations, they are the best deservers of the label ‘unrestricted quantifier’.
Annina Loets and Alexander Roberts
Possibility Plenitude
The puzzles of modal variation seem to demonstrate that if an individual could have been slightly different from how it is, it could have been greatly different from how it is (Chisholm 1967, Salmon 1989, Williamson 1990). Building on the discussion in Williamson (1990), we will compare two treatments of the puzzles, one based on the thesis of modal plenitude (Dorr et al. 2021), another based on a novel thesis we call possibility plenitude. And we will explore which role, if any, vagueness plays in driving the puzzles on each of these views.
Maria Lasonen-Aarnio
Knowledge and Reasonable Belief: The Prescriptive and Hypological
I have defended a central epistemic status, that of reasonableness, which I understand in terms of knowledge-conducive dispositions. But how do we square a knowledge norm with a norm of reasonableness? I sketch a picture on which reasonableness is intricately connected with the hypological, not the prescriptive. On my picture the prescriptive and hypological come apart in a radical manner. I discuss whether this marks a difference between the moral and epistemic domains.
Mona Simion
Knowledge and Disinformation
This paper develops a knowledge-first account of the nature of disinformation: X is disinformation in a context C iff X is a content unit communicated at C that has a disposition to generate ignorance at C in normal conditions. It also offers a taxonomy of disinformation, and an account of what it is for a signal to constitute disinformation for a particular agent in a particular context.
Richard Holton
Knowing and Telling
Some have understood telling as the transmission of knowledge; and this fits well into a knowledge-first approach. But when it takes a sentential complement, ‘tells’, unlike ‘knows’ is not factive, and this might seem problematic for such an account. However, with a wh-complement, ’tells’ exhibits something akin to factivity. I explore the possibility of focussing an account of telling on the wh-constructions, and then, working back, on what this might say about ‘knows’ when this takes a wh-complement.
TimFest
Conference in Honour of Timothy Williamson
TimFest 