The leading idea of this article is that one cannot acquire knowledge of any non-epistemic fact by virtue of knowing that one that knows something. The lines of reasoning involved in the surprise exam paradox and in Williamson’s _reductio_ of the KK-principle, which demand that one can, are thereby undermined, and new type of counter-example to epistemic closure emerges.
Hempel‘s paradox of the ravens, and his take on it, are meant to be understood as being restricted to situations where we have no additional background information. According to him, in the absence of any such information, observations of FGs confirm the hypothesis that all Fs are G. In this paper I argue against this principle by way of considering two other paradoxes of confirmation, Goodman‘s 'grue‘ paradox and the 'tacking‘ (or 'irrelevant conjunct‘) paradox. What these paradoxes reveal, I argue, (...) is that a presumption of causal realism is required to ground any confirmation; but once we grant causal realism, we have no reason to accept the central principles giving rise to the paradoxes. (shrink)
On David Lewis's original analysis of causation, c causes e only if c is linked to e by a chain of distinct events such that each event in the chain (counter-factually) depends on the former one. But this requirement precludes the possibility of late pre-emptive causation, of causation by fragile events, and of indeterministic causation. Lewis proposes three different strategies for accommodating these three kinds of cases, but none of these turn out to be satisfactory. I offer a single analysis (...) of causation that resolves these problems in one go but which respects Lewis's initial insights. One distinctive feature of my account is that it accommodates indeterministic causation without resorting to probabilities. (shrink)
Tracking accounts of knowledge were originally motivated by putative counter-examples to epistemic closure. But, as is now well known, these early accounts have many highly counterintuitive consequences. In this note, I motivate a tracking-based account which respects closure but which resolves many of the familiar problems for earlier tracking account along the way.
André Gallois (1998) attempts to defend the occasional identity thesis (OIT), the thesis that objects which are distinct at one time may nonetheless be identical at another time, in the face of two influential lines of argument against it. One argument involves Kripke’s (1971) notion of rigid designation and the other, Leibniz’s law (affirming the indiscernibility of identicals). It is reasonable for advocates of (OIT) to question the picture of rigid designation and the version of Leibniz’s law that these arguments (...) employ, but, the problem is, some form of rigidity is required for one to affirm the occasional identity of objects, and some (restricted) version of Leibniz’s law must be conceded if identity really is involved. Gallois accordingly recommends an account of rigidity and a version of Leibniz’s law to this end.1 We find Gallois’ proposals entirely inadequate to their task. We aim in this paper is to explicate and defend an alternative approach for occasional identity theorists. We do not seek to defend (OIT) per se; our aim, rather, is simply to show that the arguments from rigid designation and Leibniz’s law are inconclusive. Let’s begin with an outline of these arguments. (shrink)
A slight modification to the translation scheme for David Lewis's counterpart theory I put forward in 'An Alternative Translation Scheme for Counterpart Theory' (Analysis 49.3 (1989)) is proposed. The motivation for this change is that it makes for a more plausible account of contingent identity. In particular, contingent identity is accommodated without admitting the contingency of self-identity.
This paper defends a simple, externalist account of knowledge, incorporating familiar conditions mentioned in the literature, and responds to Timothy Williamson’s charge that any such analysis is futile because knowledge is semantically un-analyzable. The response, in short, is that even though such an account may not offer a reductive analysis of knowledge-by way of more basic, non-circular concepts-it still has an explanatory advantage over Williamson’s own position: it explains how belief can fail to be knowledge.
Kripke (1977) presents an argument designed to show that the considerations in Donnellan (1966) concerning attributive and referential uses of (definite) descriptions do not, by themselves, refute Russell’s (1905) unitary theory of description sentences (RTD), which takes (utterances of) them to express purely general, quantificational, propositions. Against Kripke, Marga Reimer (1998) argues that the two uses do indeed reflect a semantic ambiguity (an ambiguity at the level of literal truth conditions). She maintains a Russellian (quantificational) analysis of utterances involving attributively (...) used descriptions but attempts to defend the following two claims about utterances involving referentially used descriptions (referential utterances) (1998, p. 89). (shrink)
A Russellian theory of (definite) descriptions takes an utterance of the form ‘The F is G' to express a purely general proposition that affirms the existence of a (contextually) unique F: there is exactly one F [which is C] and it is G. Strawson, by contrast, takes the utterer to presuppose in some sense that there is exactly one salient F, but this is not part of what is asserted; rather, when the presupposition is not met the utterance simply fails (...) to express a (true or false) proposition. A defender of Strawson's approach, however, must square up to what appear to be straightforward counterexamples to the presupposition thesis, and must also provide an account of certain linguistic phenomena that supposedly demand treating descriptions as quantifiers, as the Russellian theory does. In this paper I propose fresh considerations in favour of Strawson's approach. I shift attention from what the utterer presupposes to preconditions for the use of descriptions, and distinguish between referring and predicative uses of descriptions (not to be confused with referential and attributive uses); importantly, the referring and predicative uses have different preconditions, I argue, and these provide some satisfactory responses to the aforementioned challenges facing the Strawsonian. South African Journal of Philosophy Vol. 27 (3) 2008: pp. 242-257. (shrink)
Timothy Williamson (2000 ch. 5) presents a reductio against the luminosity of knowing, against, that is, the so-called KK-principle: if one knows p, then one knows (or is at least in a position to know) that one knows p.1 I do not endorse the principle, but I do not think Williamson’s argument succeeds in refuting it. My aim here is to show that the KK-principle is not the most obvious culprit behind the contradiction Williamson derives.
It is well known that Russell's theory of descriptions has difficulties with descriptions occurring within desire reports. I consider a flawed argument from such a case to the conclusion that descriptions have a referring use, some responses to this argument on behalf of the Russellian, and finally rejoinders to these responses which press the point home.
Advocates of occasional identity have two ways of interpreting putative cases of fission and fusion. One way?we call it the Creative view?takes fission to involve an object really dividing (or being replicated), thereby creating objects which would not otherwise have existed. The more ontologically parsimonious way takes fission to involve merely the ?separation? of objects that were identical before: strictly speaking, no object actually divides or is replicated, no new objects are created. In this paper we recommend the Creative approach (...) as the best way of dealing with certain problem cases involving teletransportation. Our considerations yield novel takes on psychological-continuity theories of personal identity and survival, and on the puzzle of Theseus' ship. (shrink)
David Lewis’s counterpart-theoretic semantics for quantified modal logic is motivated originally by worries about identifying objects across possible worlds; the counterpart relation is grounded more cautiously on comparative similarity. The possibility of contingent identity is an unsought -- and in some eyes, unwelcome -- consequence of this approach. In this paper I motivate a Kripkean counterpart theory by way of defending the prior, pre-theoretical, coherence of contingent directness. Contingent identity follows for free. The theory is Kripkean in that the counterpart (...) relation is in a sense stipulated rather than grounded on similarity, and is such that no object has more than one counterpart at a world. This avoids a number of objections Fara and Williamson have recently levelled against counterpart theory generally; their other objections are addressed by enriching the theory with special quantifiers and actuality operators. (shrink)
Aims. Saul Kripke’s (1977) argument defending Russell’s theory of (definite) descriptions (RTD) against the possible objection that Donnellan’s (1966) distinction between attributive and referential uses of descriptions marks a semantic ambiguity has been highly influential.1 Yet, as I hope you’ll be persuaded, Kripke’s line of reasoning— in particular, the ‘thought-experiment’ it involves—has not been duly explored. In section II, I argue that while Kripke’s argument does ward off a fairly ill-motivated ambiguity theory, it is far from clear whether it would (...) succeed against more realistic candidates. If the central point I make in this regard is correct, it tells not only against Kripke’s argument but also against what has become a fairly orthodox line against the ambiguity thesis (as I shall call it). In section III, I compare Kripke’s defence of Russell with his ‘schmidentity’ argument (1980, p. 108), which involves essentially the same kind of thought-experiment. But, as I shall show, the latter argument contains an added twist which converts what otherwise would be merely a defence of one semantic theory into an attack against its rival. In section IV, I argue that the offensive strategy is unsound and attempt to locate its error. I conclude by drawing a (not unfamiliar) moral concerning the semantics—pragmatics distinction. (shrink)
In a recent article, Paul Noordhof (1999) has put forward an intriguing account of causation intended to work under the assumption of indeterminism. I am going to present four problems for the account, three which challenge the necessity of the conditions he specifies, and one which challenges their joint-sufficiency.
This paper considers, and rejects, three strategies aimed at showing that the KK-principle fails even in most favourable circumstances (all emerging from Williamson’s Knowledge and its Limits ). The case against the final strategy provides positive grounds for thinking that the principle should hold good in such situations.
In his influential paper 'Speaker's Reference and Semantic Reference', Kripke defends Russell's theory of descriptions against the charge that the existence of referential and attributive uses of descriptions reflects a semantic ambiguity. He presents a purely defensive argument to show that Russell's theory is not refuted by the referential usage and a number of methodological considerations which apparently tell in favour of Russell's unitary theory over an ambiguity theory. In this paper, I put forward a case for the ambiguity theory (...) that thwarts Kripke's defensive strategy and argue that it is not undermined by any of his methodological points. (shrink)
André Gallois’ (1993) modified account of restrictedly rigid designators (RRDs) does indeed block the objection I made to his original account (Gallois 1986; Ramachandran 1992). But, as I shall now show, there is a deeper problem with his approach which his modification does not shake off. The problem stems from the truth of the following compatibility claim: (CC) A term’s restrictedly rigidly designating (RR-designating) an object x is compatible with it designating an object y in a world W where x (...) exists but is distinct from y.1 It follows from (CC) that the necessary (contingent) truth of a sentence of the form “α is identical with β”, where “α” and “β” are RRDs of objects x and y respectively, does not require the necessary (contingent) identity of x and y. This is borne out by Gallois’ original example (see 1986, p. 58-63). Taking W to be the actual world, we have: (1) “Mary is identical with Mary” is necessarily true; yet “Mary” RR- designates Mary and Alice, which are only contingently identical. 1 I leave it to the reader to check that in Gallois’ own example (1986, p. 58), on the view he defends (pp. 62-63), and despite his modified characterisation of restricted rigidity, RDC# (1993, p. xx), “Mary” RR-designates Alice (as well as Mary) in W, but designates Mary and not A/ice in W1. Gallois has accepted this in correspondence. In light of (CC), while I agree with Gallois (1993, p. 153) that: (7) (a=b & ◊(a≠b)) → (∃x)(∃y)(x=y & ◊(x≠y)) is a theorem given RDC#, I dispute his defence of it (on p. 153). For, if (CC) is correct, the antecedent of (7) could be true even though a and b are identical in every world (where either exists)! (shrink)
In this note I revive a lingering (albeit dormant) account of rigid designation from the pages of Mind with the aim of laying it to rest. Why let a sleeping dog lie when you can put it down? André Gallois (1986) has proposed an account of rigid designators that allegedly squares with Saul Kripke’s (1980) characterisation of them as terms which designate the same object in all possible worlds, but on which, contra Kripke, identity sentences involving rigid designators may be (...) merely contingently true. This suits Gallois, as he finds the notion of contingent identity coherent. Thus, the thrust of Gallois’ thesis is that his account of rigidity is preferable to Kripke’s because his accommodates a coherent metaphysical viewpoint, whereas Kripke’s doesn’t. Gallois has thwarted one unconvincing challenge (see Carter 1987; Gallois 1988) and his account, as yet, remains untainted. But not for long, I hope.1 Let us assume, for the sake of argument, that the notion of contingent identity is coherent, that, in other words, it makes—or can make—sense to say that certain (possible) objects are identical in one world but distinct in another. What I shall argue here is that Gallois’ account of rigidity would prevent us from expressing the contingent nonidentity of objects; if so, this is a significant failing of the account, for, as it will emerge, clearly Gallois is committed to the contingency of non-identity. (shrink)
An utterance of a sentence involving an incomplete (definite) description, ‘the F’, where the context—even taking the speaker’s intentions into account—does not determine a unique F, would be unintelligible. But an utterance (in the same context) of the corresponding Russellian paraphrase would not be unintelligible. So I urged in ‘A Strawsonian objection to Russell’s theory of descriptions’ (ANALYSIS 53, 1993, pp. 209-12). I compared an utterance of (1) The table is covered with books. with an utterance of (1)’s Russellian paraphrase..
This note begins by retracing my roundabout route to a well-known, but widely dismissed, view of Even-if conditionals (Even-Ifs), namely, that they affirm their consequents, i.e. that [Even if A, C] entails [C]. The route also explains my reluctance to give up on that view in the face of prima facie overwhelming, countervailing linguistic evidence — evidence which has led most philosophers of conditionals to the rival view that Even-Ifs entail the corresponding Ifs: i.e. that [Even if A, C] entails (...) [If A, C].1 The goal of the paper is to provide fresh motivation for the dismissed view, and to offer responses to prima facie persuasive countervailing considerations. What emerges is an account where prevalent uses of Even-Ifs are accounted for at the level of pragmatics rather than by the semantics — hence the ‘topsy-turvy’ in the title. (shrink)
This paper examines a neglected puzzle about conditionals: namely, the fact that each of a pair of conditionals with incompatible consequents, [A > C] and [B > C*], may be properly affirmable in circumstances when one does not believe, and is not entitled to infer, the denial of the conjunction of the antecedents, i.e. ~(A & B). The puzzle is why this should be so, since the conditionals entail the conjunction on the popular accounts of conditionals. I present a pragmatic (...) solution which distinguishes between two levels of affirmability. (shrink)
One of Strawson's objections to Russell's theory of descriptions is that what are intuitively natural and correct utterances of sentences involving incomplete descriptions come out false by RTD. Russellians have responded, not by challenging Strawson's view that these uses are natural and correct, but by embellishing RTD to accommodate these uses. I pursue an alternative line of attack: I argue that there are circumstances in which "we" would find utterances of such sentences unnatural and improper but "RTD" would sanction. So, (...) RTD clashes with ordinary language, as Strawson suggests. (shrink)
An âinvertedâ reasoner is someone who finds the inferences we find easy, inversely difficult, and those that we find difficult, inversely easy. The notion was initially introduced by Christopher Cherniak in his book, Minimal Rationality, and appealed to by Stephen Stich in The Fragmentation of Reason. While a number of difficulties have been noted about what reasoning would amount to for such a reasoner, what has not been brought out in the literature is that such a reasoner is in fact (...) logically impossible. This is what I hope to demonstrate in this paper. (shrink)
This paper puts forward necessary and suffficient conditions for knowing, and addresses Timothy Williamson's claims that knowledge is an un-analyzable mental state, and that any such analysis is futile. It is argued that such an account has explanatory motivation: it explains when belief fails to be knowledge.
Timothy Williamson (2000) reckons that hardly any mental state is luminous, i.e. is such that if one were in it, then one would invariably be in a position to know that one was. This paper examines an argument he presents against the luminosity of feeling cold, which he claims generalizes to other phenomenal states, such as e.g. being in pain. As we shall see, the argument fails. However, our deliberations do yield two anti-luminosity results: a simple refutation of the claim (...) that one invariably knows whether one feels cold or not,1 and a counterexample to the luminosity of knowing—in effect, a counterexample to the (KK)- principle. (shrink)