22 ideas
| 8820 | Rules of reasoning precede the concept of truth, and they are what characterize it [Pollock] |
| Full Idea: Rather than truth being fundamental and rules for reasoning being derived from it, the rules for reasoning come first and truth is characterized by the rules for reasoning about truth. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Cog.Mach') | |
| A reaction: This nicely disturbs our complacency about such things. There is plenty of reasoning in Homer, but I bet there is no talk of 'truth'. Pontius Pilate seems to have been a pioneer (Idea 8821). Do the truth tables define or describe logical terms? |
| 8819 | We need the concept of truth for defeasible reasoning [Pollock] |
| Full Idea: It might be wondered why we even have a concept of truth. The answer is that this concept is required for defeasible reasoning. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Cog.Mach') | |
| A reaction: His point is that we must be able to think critically about our beliefs ('is p true?') if we are to have any knowledge at all. An excellent point. Give that man a teddy bear. |
| 10928 | Maybe we can quantify modally if the objects are intensional, but it seems unlikely [Quine] |
| Full Idea: Perhaps there is no objection to quantifying into modal contexts as long as the values of any variables thus quantified are limited to intensional objects, but they also lead to disturbing examples. | |
| From: Willard Quine (Reference and Modality [1953], §3) | |
| A reaction: [Quine goes on to give his examples] I take it that possibilities are features of actual reality, not merely objects of thought. The problem is that they are harder to know than actual objects. |
| 10925 | Failure of substitutivity shows that a personal name is not purely referential [Quine] |
| Full Idea: Failure of substitutivity shows that the occurrence of a personal name is not purely referential. | |
| From: Willard Quine (Reference and Modality [1953], §1) | |
| A reaction: I don't think I understand the notion of a name being 'purely' referential, as if it somehow ceased to be a word, and was completely transparent to the named object. |
| 10926 | Quantifying into referentially opaque contexts often produces nonsense [Quine] |
| Full Idea: If to a referentially opaque context of a variable we apply a quantifier, with the intention that it govern that variable from outside the referentially opaque context, then what we commonly end up with is unintended sense or nonsense. | |
| From: Willard Quine (Reference and Modality [1953], §2) |
| 13128 | 'Ultimate sortals' cannot explain ontological categories [Westerhoff on Wiggins] |
| Full Idea: 'Ultimate sortals' are said to be non-subordinated, disjoint from one another, and uniquely paired with each object. Because of this, the ultimate sortal cannot be a satisfactory explication of the notion of an ontological category. | |
| From: comment on David Wiggins (Identity and Spatio-Temporal Continuity [1971], p.75) by Jan Westerhoff - Ontological Categories §26 | |
| A reaction: My strong intuitions are that Wiggins is plain wrong, and Westerhoff gives the most promising reasons for my intuition. The simplest point is that objects can obviously belong to more than one category. |
| 10930 | Quantification into modal contexts requires objects to have an essence [Quine] |
| Full Idea: A reversion to Aristotelian essentialism is required if quantification into modal contexts is to be insisted on. An object must be seen as having some of its traits necessarily. | |
| From: Willard Quine (Reference and Modality [1953], §3) | |
| A reaction: This thought leads directly to Kripke's proposal of rigid designation of objects (and Lewis response of counterparts), which really gets modal logic off the ground. Quine's challenge remains - the modal logic entails a huge metaphysical commitment. |
| 8822 | Statements about necessities need not be necessarily true [Pollock] |
| Full Idea: True statements about the necessary properties of things need not be necessarily true. The well-known example is that the number of planets (9) is necessarily an odd number. The necessity is de re, but not de dicto. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Nat.Internal') | |
| A reaction: This would be a matter of the scope (the placing of the brackets) of the 'necessarily' operator in a formula. The quick course in modal logic should eradicate errors of this kind in your budding philosopher. |
| 14645 | To be necessarily greater than 7 is not a trait of 7, but depends on how 7 is referred to [Quine] |
| Full Idea: To be necessarily greater than 7 is not a trait of a number, but depends on the manner of referring to the number. | |
| From: Willard Quine (Reference and Modality [1953], §2) | |
| A reaction: The most concise quotation of Quine's objection to 'de re' modality. The point is whether the number might have been referred to as 'the number of planets'. So many of these problems are solved by fixing unambiguous propositions first. |
| 9201 | Whether 9 is necessarily greater than 7 depends on how '9' is described [Quine, by Fine,K] |
| Full Idea: Quine's metaphysical argument is that if 9 is 7+2 the number 9 will be necessarily greater than 7, but when 9 is described as the number of planets, the number will not be necessarily greater than 7. The necessity depends on how it is described. | |
| From: report of Willard Quine (Reference and Modality [1953]) by Kit Fine - Intro to 'Modality and Tense' p. 3 | |
| A reaction: Thus necessity would be entirely 'de dicto' and not 'de re'. It sounds like a feeble argument. If I describe the law of identity (a=a) as 'my least favourite logical principle', that won't make it contingent. Describe 9, or refer to it? See Idea 9203. |
| 10927 | Necessity only applies to objects if they are distinctively specified [Quine] |
| Full Idea: Necessity does not properly apply to the fulfilment of conditions by objects (such as the number which numbers the planets), apart from special ways of specifying them. | |
| From: Willard Quine (Reference and Modality [1953], §3) | |
| A reaction: This appears to say that the only necessity is 'de dicto', and that there is no such thing as 'de re' necessity (of the thing in itself). How can Quine deny that there might be de re necessities? His point is epistemological - how can we know them? |
| 9203 | We can't quantify in modal contexts, because the modality depends on descriptions, not objects [Quine, by Fine,K] |
| Full Idea: 'Necessarily 9>7' may be true while the sentence 'necessarily the number of planets < 7' is false, even though it is obtained by substituting a coreferential term. So quantification in these contexts is unintelligible, without a clear object. | |
| From: report of Willard Quine (Reference and Modality [1953]) by Kit Fine - Intro to 'Modality and Tense' p. 4 | |
| A reaction: This is Quine's second argument against modality. See Idea 9201 for his first. Fine attempts to refute it. The standard reply seems to be to insist that 9 must therefore be an object, which pushes materialist philosophers into reluctant platonism. |
| 8818 | Defeasible reasoning requires us to be able to think about our thoughts [Pollock] |
| Full Idea: Defeasible reasoning requires us to be able to think about our thoughts. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Cog.Mach') | |
| A reaction: This is why I do not think animals 'know' anything, though they seem to have lots of true beliefs about their immediate situation. |
| 8811 | What we want to know is - when is it all right to believe something? [Pollock] |
| Full Idea: When we ask whether a belief is justified, we want to know whether it is all right to believe it. The question we must ask is 'when is it permissible (epistemically) to believe P?'. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Ep.Norms') | |
| A reaction: Nice to see someone trying to get the question clear. The question clearly points to the fact that there must at least be some sort of social aspect to criteria of justification. I can't cheerfully follow my intuitions if everyone else laughs at them. |
| 8817 | Logical entailments are not always reasons for beliefs, because they may be irrelevant [Pollock] |
| Full Idea: Epistemologists have noted that logical entailments do not always constitute reasons. P may entail Q without the connection between P and Q being at all obvious. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Ref.of Extern') | |
| A reaction: Graham Priest and others try to develop 'relevance logic' to deal with this. This would deny the peculiar classical claim that everything is entailed by a falsehood. A belief looks promising if it entails lots of truths about the world. |
| 8814 | Epistemic norms are internalised procedural rules for reasoning [Pollock] |
| Full Idea: Epistemic norms are to be understood in terms of procedural knowledge involving internalized rules for reasoning. | |
| From: John L. Pollock (Epistemic Norms [1986], 'How regulate?') | |
| A reaction: He offers analogies with bicycly riding, but the simple fact that something is internalized doesn't make it a norm. Some mention of truth is needed, equivalent to 'don't crash the bike'. |
| 8823 | Reasons are always for beliefs, but a perceptual state is a reason without itself being a belief [Pollock] |
| Full Idea: When one makes a perceptual judgement on the basis of a perceptual state, I want to say that the perceptual state itself is one's reason. ..Reason are always reasons for beliefs, but the reasons themselves need not be beliefs. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Dir.Realism') | |
| A reaction: A crucial issue. I think I prefer the view of Davidson, in Ideas 8801 and 8804. Three options: a pure perception counts as a reason, or perceptions involve some conceptual content, or you only acquire a reason when a proposition is formulated. |
| 8813 | If we have to appeal explicitly to epistemic norms, that will produce an infinite regress [Pollock] |
| Full Idea: If we had to make explicit appeal to epistemic norms for justification (the 'intellectualist model') we would find ourselves in an infinite regress. The norms, their existence and their application would themselves have to be justified. | |
| From: John L. Pollock (Epistemic Norms [1986], 'How regulate?') | |
| A reaction: This is counter to the 'space of reasons' picture, where everything is rationally assessed. There are regresses for both reasons and for experiences, when they are offered as justifications. |
| 8812 | Norm Externalism says norms must be internal, but their selection is partly external [Pollock] |
| Full Idea: Norm Externalism acknowledges that the content of our epistemic norms must be internalist, but employs external considerations in the selection of the norms themselves. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Ep.Norms') | |
| A reaction: It can't be right that you just set your own norms, so this must contain some truth. Equally, even the most hardened externalist can't deny that what goes on in the head of the person concerned must have some relevance. |
| 8816 | Externalists tend to take a third-person point of view of epistemology [Pollock] |
| Full Idea: Externalists tend to take a third-person point of view in discussing epistemology. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Ref.of Extern') | |
| A reaction: Pollock's point, quite reasonably, is that the first-person aspect must precede any objective assessment of whether someone knows. External facts, such as unpublicised information, can undermine high quality internal justification. |
| 8815 | Belief externalism is false, because external considerations cannot be internalized for actual use [Pollock] |
| Full Idea: External considerations of reliability could not be internalized. Consequently, it is in principle impossible for us to actually employ externalist norms. I take this to be a conclusive refutation of belief externalism. | |
| From: John L. Pollock (Epistemic Norms [1986], 'Ref.of Extern') | |
| A reaction: Not so fast. He earlier rejected the 'intellectualist model' (Idea 8813), so he doesn't think norms have to be fully conscious and open to criticism. So they could be innate, or the result of indoctrination (sorry, teaching), or just forgotten. |
| 10931 | We can't say 'necessarily if x is in water then x dissolves' if we can't quantify modally [Quine] |
| Full Idea: To say an object is soluble in water is to say that it would dissolve if it were in water,..which implies that 'necessarily if x is in water then x dissolves'. Yet we do not know if there is a suitable sense of 'necessarily' into which we can so quantify. | |
| From: Willard Quine (Reference and Modality [1953], §4) | |
| A reaction: This is why there has been a huge revival of scientific essentialism - because Krike seems to offer exacty the account which Quine said was missing. So can you have modal logic without rigid designation? |