48 ideas
| 9821 | A definition need not capture the sense of an expression - just get the reference right [Frege, by Dummett] |
| Full Idea: Frege expressly denies that a correct definition need capture the sense of the expression it defines: it need only get the reference right. | |
| From: report of Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894]) by Michael Dummett - Frege philosophy of mathematics Ch.3 | |
| A reaction: This might hit up against the renate/cordate problem, of two co-extensive concepts, where the definition gets the extension right, but the intension wrong. |
| 3593 | The only way to specify the corresponding fact is asserting the sentence [Williams,M] |
| Full Idea: The trouble with appeal to facts in the correspondence theory is that, in general, we have no way of indicating what fact a sentence, when true, corresponds to other than asserting the sentence. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.12) |
| 3585 | Coherence needs positive links, not just absence of conflict [Williams,M] |
| Full Idea: It is often claimed that coherence is more than 'absence of conflict' between beliefs; it also involves 'positive connections'. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.10) |
| 3584 | Justification needs coherence, while truth might be ideal coherence [Williams,M] |
| Full Idea: Contemporary coherence theorists are advancing a theory of justification, not of truth, …with those who argue that truth is also coherence explaining it in terms of ideal coherence, or coherence at the limit of enquiry. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.10) |
| 9585 | Since every definition is an equation, one cannot define equality itself [Frege] |
| Full Idea: Since every definition is an equation, one cannot define equality itself. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.327) | |
| A reaction: This seems a particularly nice instance of the general rule that 'you have to start somewhere'. It is a nice test case for the nature of meaning to ask 'what do you understand when you understand equality?', given that you can't define it. |
| 11211 | If a sound conclusion comes from two errors that cancel out, the path of the argument must matter [Rumfitt] |
| Full Idea: If a designated conclusion follows from the premisses, but the argument involves two howlers which cancel each other out, then the moral is that the path an argument takes from premisses to conclusion does matter to its logical evaluation. | |
| From: Ian Rumfitt ("Yes" and "No" [2000], II) | |
| A reaction: The drift of this is that our view of logic should be a little closer to the reasoning of ordinary language, and we should rely a little less on purely formal accounts. |
| 3599 | Deduction shows entailments, not what to believe [Williams,M] |
| Full Idea: The rules of deduction are rules of entailment, not rules of inference. They tell us what follows from what, not what to believe on the basis of what. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.18) |
| 11210 | Standardly 'and' and 'but' are held to have the same sense by having the same truth table [Rumfitt] |
| Full Idea: If 'and' and 'but' really are alike in sense, in what might that likeness consist? Some philosophers of classical logic will reply that they share a sense by virtue of sharing a truth table. | |
| From: Ian Rumfitt ("Yes" and "No" [2000]) | |
| A reaction: This is the standard view which Rumfitt sets out to challenge. |
| 11212 | The sense of a connective comes from primitively obvious rules of inference [Rumfitt] |
| Full Idea: A connective will possess the sense that it has by virtue of its competent users' finding certain rules of inference involving it to be primitively obvious. | |
| From: Ian Rumfitt ("Yes" and "No" [2000], III) | |
| A reaction: Rumfitt cites Peacocke as endorsing this view, which characterises the logical connectives by their rules of usage rather than by their pure semantic value. |
| 17446 | Counting rests on one-one correspondence, of numerals to objects [Frege] |
| Full Idea: Counting rests itself on a one-one correlation, namely of numerals 1 to n and the objects. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894]), quoted by Richard G. Heck - Cardinality, Counting and Equinumerosity 3 | |
| A reaction: Parsons observes that counting will establish a one-one correspondence, but that doesn't make it the aim of counting, and so Frege hasn't answered Husserl properly. Which of the two is conceptually prior? How do you decide. |
| 9582 | Husserl rests sameness of number on one-one correlation, forgetting the correlation with numbers themselves [Frege] |
| Full Idea: When Husserl says that sameness of number can be shown by one-one correlation, he forgets that this counting itself rests on a univocal one-one correlation, namely that between the numerals 1 to n and the objects of the set. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.326) | |
| A reaction: This is the platonist talking. Neo-logicism is attempting to build numbers just from the one-one correlation of objects. |
| 9586 | In a number-statement, something is predicated of a concept [Frege] |
| Full Idea: In a number-statement, something is predicated of a concept. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.328) | |
| A reaction: A succinct statement of Frege's theory of numbers. By my lights that would make numbers at least second-order abstractions. |
| 9580 | Our concepts recognise existing relations, they don't change them [Frege] |
| Full Idea: The bringing of an object under a concept is merely the recognition of a relation which previously already obtained, [but in the abstractionist view] objects are essentially changed by the process, so that objects brought under a concept become similar. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.324) | |
| A reaction: Frege's view would have to account for occasional misapplications of concepts, like taking a dolphin to be a fish, or falsely thinking there is someone in the cellar. |
| 9589 | Numbers are not real like the sea, but (crucially) they are still objective [Frege] |
| Full Idea: The sea is something real and a number is not; but this does not prevent it from being something objective; and that is the important thing. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.337) | |
| A reaction: This seems a qualification of Frege's platonism. It is why people start talking about abstract items which 'subsist', instead of 'exist'. It shows Frege's motivation in all this, which is to secure logic and maths from the vagaries of psychology. |
| 9577 | The naïve view of number is that it is like a heap of things, or maybe a property of a heap [Frege] |
| Full Idea: The most naïve opinion of number is that it is something like a heap in which things are contained. The next most naïve view is the conception of number as the property of a heap, cleansing the objects of their particulars. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.323) | |
| A reaction: A hundred toothbrushes and a hundred sponges can be seen to contain the same number (by one-to-one mapping), without actually knowing what that number is. There is something numerical in the heap, even if the number is absent. |
| 9578 | If objects are just presentation, we get increasing abstraction by ignoring their properties [Frege] |
| Full Idea: If an object is just presentation, we can pay less attention to a property and it disappears. By letting one characteristic after another disappear, we obtain concepts that are increasingly more abstract. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.324) | |
| A reaction: Frege despises this view. Note there is scope in the despised view for degrees or levels of abstraction, defined in terms of number of properties ignored. Part of Frege's criticism is realist. He retains the object, while Husserl imagines it different. |
| 3591 | We could never pin down how many beliefs we have [Williams,M] |
| Full Idea: Asking how many beliefs I have is like asking how many drops of water there are in a bucket. If I believe my dog is in the garden, do I also believe he is not in the house, or in Siberia? | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.11) |
| 3582 | Propositions make error possible, so basic experiential knowledge is impossible [Williams,M] |
| Full Idea: Propositional content is inseparable from possible error. Therefore no judgement, however modest, is indubitable. So if basic experiential knowledge has to be indubitable, there is no such knowledge. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 8) |
| 3592 | Phenomenalism is a form of idealism [Williams,M] |
| Full Idea: Phenomenalism is a form of idealism. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.12) |
| 3579 | Sense data avoid the danger of misrepresenting the world [Williams,M] |
| Full Idea: The point of insisting on the absolute immediacy of sense data is that representation always seems to involve the possibility of misrepresentation. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 8) |
| 3581 | Sense data can't give us knowledge if they are non-propositional [Williams,M] |
| Full Idea: Acquaintance with sense data is supposed to be a form of non-propositional knowledge, but how can something be non-propositional and yet knowledge? | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 8) |
| 3564 | Is it people who are justified, or propositions? [Williams,M] |
| Full Idea: What exactly is supposed to be 'justified': a person's believing some particular proposition, or the proposition that he believes? | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 1) | |
| A reaction: A key distinction. See my comment on Idea 3752. What would justify a sign saying 'treasure buried here'? People can be justified in believing falsehoods. How could a false proposition be justified? |
| 3595 | What works always takes precedence over theories [Williams,M] |
| Full Idea: A theory that represents working practices as unworkable is a bad theory. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.13) | |
| A reaction: Good point. There's a lot of this about in epistemology, especially accusations of circularity or infinite regress, which (if true) don't somehow seem to worry the cove on the Clapham omnibus. |
| 3580 | Experience must be meaningful to act as foundations [Williams,M] |
| Full Idea: If we are to treat experience as the foundation of knowledge, then experience must itself be understood to involve propositional content. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 8) | |
| A reaction: This sounds right, but since pure 'experience' obviously doesn't have propositional content, because it needs interpretation and evaluation, then this strategy won't work. |
| 3578 | Are empirical foundations judgements or experiences? [Williams,M] |
| Full Idea: Empirical foundationists must decide whether knowledge ultimately rests on either beliefs or judgements about experience, or on the experiences or sensations themselves. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 8) | |
| A reaction: This clarifies the key issue very nicely, and I firmly vote for the former option. The simplest point is that error is possible about what sensations are taken to be of, so they won't do on their own. |
| 3576 | Foundationalists are torn between adequacy and security [Williams,M] |
| Full Idea: The foundationalists dilemma is to define a basis for knowledge modest enough to be secure but rich enough to be adequate. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 7) | |
| A reaction: ..And that is just what they are unable to do, precisely because adequate support would have to have enough content to be defeasibe or fallible. |
| 3577 | Strong justification eliminates error, but also reduces our true beliefs [Williams,M] |
| Full Idea: A strongly justificationist view of rationality may not be so rational; we want the truth, but avoiding all errors and maximising our number of true beliefs are not the same thing. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 7) | |
| A reaction: An interesting dilemma - to avoid all errors, believing nothing; to maximise true belief, believe everything. It is rational to follow intuition, guesses, and a wing and a prayer - once you are experienced and educated. |
| 3589 | Why should diverse parts of our knowledge be connected? [Williams,M] |
| Full Idea: Why should political theory ever have much to do with quantum physics, or pet care with parliamentary history? | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.11) | |
| A reaction: This hardly demolishes the coherence account of justification, since your views on pet care had better be coherent, for your pet's sake. It's a pity people can make their politics cohere with their ethics. |
| 3590 | Coherence theory must give a foundational status to coherence itself [Williams,M] |
| Full Idea: Coherence theory implicitly assigns the criteria of coherence a special status. …In so far as this status is assigned a priori, the coherence theory represents a rationalistic variant of foundationalism. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.11) | |
| A reaction: Nice move, to accuse coherence theorists of foundationalism! Wrong, though, because the a priori principles of coherence are not basic beliefs, but evolved pragmatic procedures (or something...). |
| 3571 | Externalism does not require knowing that you know [Williams,M] |
| Full Idea: From an externalist point of view, knowing about one's reliability is not required for 'first-order' knowledge. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 2) | |
| A reaction: Ah. 'First-order knowledge' - what's that? What we used to call 'true belief', I would say. Adequate for animals, and a good guide to daily life, but uncritical and unjustifiable. |
| 3574 | Externalism ignores the social aspect of knowledge [Williams,M] |
| Full Idea: A problem with pure externalism is that it ignores the social dimension of knowledge. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 2) | |
| A reaction: This seems to be contradicted by Idea 3573, which allows a social dimension to agreement over what is reliable. I am inclined to take knowledge as an entirely social concept. |
| 3569 | In the causal theory of knowledge the facts must cause the belief [Williams,M] |
| Full Idea: According to Goldman's early causal theory of knowledge, my belief that p counts as knowledge if and only if it is caused by the fact that p. This is sufficient as well as necessary, and so does not involve justification. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 2) | |
| A reaction: I take his theory simply to be false because what causes a belief is not what justifies it. I expect my mother to ring; the phone rings; I 'know' it is my mother (and it is), because I strongly expect it. |
| 3567 | How could there be causal relations to mathematical facts? [Williams,M] |
| Full Idea: It is not clear what would even be meant by supposing that there are causal relations to mathematical facts. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 2) | |
| A reaction: I agree, though platonists seem to be willing to entertain the possibility that there are causal relations, for which no further explanation can be given. Better is knowledge without a causal relation. |
| 3586 | Only a belief can justify a belief [Williams,M] |
| Full Idea: Justification requires logical rather than causal connections. That is the point of the slogan that only a belief can justify a belief. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.10) | |
| A reaction: It seems better to talk of 'rational' connections, rather than 'logical' connections. It isn't 'logical' to believe that someone despises me because their lip is faintly curled. |
| 3573 | Externalist reliability refers to a range of conventional conditions [Williams,M] |
| Full Idea: The radical externalists' key notion is 'reliability', which is a normative condition governing adequate performance, involving reference to a range of conditions which we decide rather than discover. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 2) | |
| A reaction: If we can decide whether a source is reliable, we can also decide whether a reliable source has performed well on this occasion, and that will always take precedence. |
| 3565 | Sometimes I ought to distrust sources which are actually reliable [Williams,M] |
| Full Idea: I may reach a belief using a procedure that is in fact reliable, but which I ought to distrust. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 1) | |
| A reaction: The tramp on the park bench who gives good share tips. The clock that is finally working, but has been going haywire for weeks. Reliabilism is a bad theory. |
| 3566 | We control our beliefs by virtue of how we enquire [Williams,M] |
| Full Idea: We control our beliefs by virtue of how we enquire. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 1) |
| 3594 | Scepticism just reveals our limited ability to explain things [Williams,M] |
| Full Idea: All the sceptic's arguments show is that there are limits to our capacity to give reasons or cite evidence. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.13) |
| 3575 | Scepticism can involve discrepancy, relativity, infinity, assumption and circularity [Williams,M] |
| Full Idea: The classical Five Modes of Scepticism are Discrepancy (people always disagree), Relativity ('according to you'), Infinity (infinite regress of questions), Assumption (ending in dogma) and Circularity (end up where you started). | |
| From: Michael Williams (Problems of Knowledge [2001], Ch. 5) | |
| A reaction: I take Relativity to be different from scepticism (because, roughly, it says there is nothing to know), and the others go with Agrippa's Trilemma of justification, which may have solutions. |
| 3587 | Seeing electrons in a cloud chamber requires theory [Williams,M] |
| Full Idea: Armed with enough theory, we can see electrons in a cloud chamber. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.10) |
| 9581 | Many people have the same thought, which is the component, not the private presentation [Frege] |
| Full Idea: The same thought can be grasped by many people. The components of a thought, and even more so the things themselves, must be distinguished from the presentations which in the soul accompany the grasping of a thought. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.325) | |
| A reaction: This is the basic realisation, also found in Russell, of how so much confusion has crept into philosophy, in Berkeley, for example. Frege starts down the road which leads to the externalist view of content. |
| 9579 | Disregarding properties of two cats still leaves different objects, but what is now the difference? [Frege] |
| Full Idea: If from a black cat and a white cat we disregard colour, then posture, then location, ..we finally derive something which is completely without restrictions on content; but what is derived from the objects does differ, although it is not easy to say how. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.324) | |
| A reaction: This is a key objection to abstractionism for Frege - we are counting two cats, not two substrata of essential catness, or whatever. But what makes a cat countable? (Key question!) It isn't its colour, or posture or location. |
| 9587 | How do you find the right level of inattention; you eliminate too many or too few characteristics [Frege] |
| Full Idea: Inattention is a very strong lye which must not be too concentrated, or it dissolves everything (such as the connection between the objects), but must not be too weak, to produce sufficient change. Personally I cannot find the proper dilution. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.330) | |
| A reaction: We may sympathise with the lack of precision here (frustrating for a logician), but it is not difficult to say of a baseball defence 'just concentrate on the relations, and ignore the individuals who implement it'. You retain basic baseball skills. |
| 9588 | Number-abstraction somehow makes things identical without changing them! [Frege] |
| Full Idea: Number-abstraction simply has the wonderful and very fruitful property of making things absolutely the same as one another without altering them. Something like this is possible only in the psychological wash-tub. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.332) | |
| A reaction: Frege can be awfully sarcastic. I don't really see his difficulty. For mathematics we only need to know what is countable about an object - we don't need to know how many hairs there are on the cat, only that it has identity. |
| 9583 | Psychological logicians are concerned with sense of words, but mathematicians study the reference [Frege] |
| Full Idea: The psychological logicians are concerned with the sense of the words and with the presentations, which they do not distinguish from the sense; but the mathematicians are concerned with the matter itself, with the reference of the words. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.326) | |
| A reaction: This is helpful for showing the point of his sense/reference distinction; it is part of his campaign against psychologism, by showing that there is a non-psychological component to language - the reference, where it meets the public world. |
| 9584 | Identity baffles psychologists, since A and B must be presented differently to identify them [Frege] |
| Full Idea: The relation of sameness remains puzzling to a psychological logician. They cannot say 'A is the same as B', because that requires distinguishing A from B, so that these would have to be different presentations. | |
| From: Gottlob Frege (Review of Husserl's 'Phil of Arithmetic' [1894], p.327) | |
| A reaction: This is why Frege needed the concept of reference, so that identity could be outside the mind (as in Hesperus = Phosophorus). Think about an electron; now think about a different electron. |
| 3588 | Foundationalists base meaning in words, coherentists base it in sentences [Williams,M] |
| Full Idea: In the foundationalist picture the meaning of individual words (defined ostensively) is primary, and that of sentences is derivative. For coherentists sentences come first, with meaning understood functionally or inferentially. | |
| From: Michael Williams (Problems of Knowledge [2001], Ch.10) | |
| A reaction: Coherentism about language doesn't imply coherentism about justification. On language I vote for foundationalism, because I am impressed by the phenomenon of compositionality. |
| 11214 | We learn 'not' along with affirmation, by learning to either affirm or deny a sentence [Rumfitt] |
| Full Idea: The standard view is that affirming not-A is more complex than affirming the atomic sentence A itself, with the latter determining its sense. But we could learn 'not' directly, by learning at once how to either affirm A or reject A. | |
| From: Ian Rumfitt ("Yes" and "No" [2000], IV) | |
| A reaction: [compressed] This seems fairly anti-Fregean in spirit, because it looks at the psychology of how we learn 'not' as a way of clarifying what we mean by it, rather than just looking at its logical behaviour (and thus giving it a secondary role). |