48 ideas
| 13939 | No possible evidence could decide the reality of numbers, so it is a pseudo-question [Carnap] |
| Full Idea: I cannot think of any possible evidence that would be regarded as relevant by both nominalists and realists about numbers, and would decide the controversy, or make one side more probable. Hence I regard the external questions as pseudo-questions. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 4) |
| 16252 | Metaphysics uses empty words, or just produces pseudo-statements [Carnap] |
| Full Idea: Since metaphysics doesn't want to assert analytic propositions, nor fall within the domain of physical science, it is compelled to employ words for which no criteria of application are specified, ..or else combine meaningful words..into pseudo-statements. | |
| From: Rudolph Carnap (Elimination of Metaphysics by Analysis of Language [1959]), quoted by Tim Maudlin - The Metaphysics within Physics 2.4 | |
| A reaction: A classic summary of the logical positivist rejection of metaphysics. I incline to treat metaphysics as within the domain of science, but at a level of generality so high that practising scientists become bewildered and give up. |
| 13342 | Carnap defined consequence by contradiction, but this is unintuitive and changes with substitution [Tarski on Carnap] |
| Full Idea: Carnap proposed to define consequence as 'sentence X follows from the sentences K iff the sentences K and the negation of X are contradictory', but 1) this is intuitively impossible, and 2) consequence would be changed by substituting objects. | |
| From: comment on Rudolph Carnap (The Logical Syntax of Language [1934], p.88-) by Alfred Tarski - The Concept of Logical Consequence p.414 | |
| A reaction: This seems to be the first step in the ongoing explicit discussion of the nature of logical consequence, which is now seen by many as the central concept of logic. Tarski brings his new tool of 'satisfaction' to bear. |
| 13251 | Each person is free to build their own logic, just by specifying a syntax [Carnap] |
| Full Idea: In logic, there are no morals. Everyone is at liberty to build his own logic, i.e. his own form of language. All that is required is that he must state his methods clearly, and give syntactical rules instead of philosophical arguments. | |
| From: Rudolph Carnap (The Logical Syntax of Language [1934], §17), quoted by JC Beall / G Restall - Logical Pluralism 7.3 | |
| A reaction: This is understandable, but strikes me as close to daft relativism. If I specify a silly logic, I presume its silliness will be obvious. By what criteria? I say the world dictates the true logic, but this is a minority view. |
| 13936 | Questions about numbers are answered by analysis, and are analytic, and hence logically true [Carnap] |
| Full Idea: For the internal question like 'is there a prime number greater than a hundred?' the answers are found by logical analysis based on the rules for the new expressions. The answers here are analytic, i.e., logically true. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 2) |
| 8748 | Logical positivists incorporated geometry into logicism, saying axioms are just definitions [Carnap, by Shapiro] |
| Full Idea: The logical positivists brought geometry into the fold of logicism. The axioms of, say, Euclidean geometry are simply definitions of primitive terms like 'point' and 'line'. | |
| From: report of Rudolph Carnap (Empiricism, Semantics and Ontology [1950]) by Stewart Shapiro - Thinking About Mathematics 5.3 | |
| A reaction: If the concept of 'line' is actually created by its definition, then we need to know exactly what (say) 'shortest' means. If we are merely describing a line, then our definition can be 'impredicative', using other accepted concepts. |
| 8960 | Internal questions about abstractions are trivial, and external ones deeply problematic [Carnap, by Szabó] |
| Full Idea: Carnap's verdict is that questions regarding the existence of abstracta tend to be trivial when taken as internal and deeply problematic when taken as external. | |
| From: report of Rudolph Carnap (Empiricism, Semantics and Ontology [1950]) by Zoltán Gendler Szabó - Nominalism 6 | |
| A reaction: If the internal aspect of the problem is 'trivial', this would put Carnap in league with fictionalists, who are only committed to entities while playing the current game. What is the status of the theory? Carnap wanted flowers to bloom. |
| 13933 | Existence questions are 'internal' (within a framework) or 'external' (concerning the whole framework) [Carnap] |
| Full Idea: We distinguish two kinds of existence questions: first, entities of a new kind within the framework; we call them 'internal questions'. Second, 'external questions', concerning the existence or reality of the system of entities as a whole. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 2) | |
| A reaction: This nicely disposes of many ontological difficulties, but at the price of labelling most external questions as meaningless, so that the internal answers have very little commitment, and the external (big) questions are now banned. Not for me. |
| 13934 | To be 'real' is to be an element of a system, so we cannot ask reality questions about the system itself [Carnap] |
| Full Idea: To be real in the scientific sense means to be an element of the system; hence this concept cannot be meaningfully applied to the system itself. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 2) |
| 13938 | A linguistic framework involves commitment to entities, so only commitment to the framework is in question [Carnap] |
| Full Idea: If someone accepts a framework for a kind of entities, then he must admit the entities as possible designata. Thus the question of the admissibility of entities is reduced to the question of the acceptability of the linguistic framework for the entities. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 4) | |
| A reaction: Despite the many differences of opinion between Quine and Carnap, this appears to be a straight endorsement by Carnap of the Quinean conception of ontological commitment. |
| 13935 | We only accept 'things' within a language with formation, testing and acceptance rules [Carnap] |
| Full Idea: To accept the thing world means nothing more than to accept a certain form of language, in other words, to accept rules for forming statements and for testing, accepting, or rejecting them. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 2) | |
| A reaction: If you derive your metaphysics from your language, then objects are linguistic conventions. But why do we accept conventions about objects? |
| 2730 | Because 'gold is malleable' is necessary does not mean that it is analytic [Audi,R] |
| Full Idea: Taking the proposition that gold is malleable to be necessary does not commit one to considering it analytic. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], IV p.116) |
| 14305 | In the truth-functional account a burnt-up match was soluble because it never entered water [Carnap] |
| Full Idea: If a wooden match was completely burned up yesterday, and never placed in water at any time, is it not the case, therefore, that the match is soluble (in the truth-functional view). This follows just from the antecedent being false. | |
| From: Rudolph Carnap (Testability and Meaning [1937], I.440), quoted by Stephen Mumford - Dispositions | |
| A reaction: This, along with Edgington's nice example of the conditional command (Idea ) seems conclusive against the truth-functional account. The only defence possible is some sort of pragmatic account about implicature. |
| 2715 | Beliefs are based on perception, memory, introspection or reason [Audi,R] |
| Full Idea: The four basic kinds of belief are rooted in perception, memory, introspective consciousness, and reason. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], Intr.p.7) |
| 2735 | Could you have a single belief on its own? [Audi,R] |
| Full Idea: Could one have just a single belief? | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VII p.198) |
| 2736 | We can make certain of what we know, so knowing does not entail certainty [Audi,R] |
| Full Idea: The possibility of making certain of what we already know suggests that knowing a proposition does not entail its being certain. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VIII p.220) |
| 2722 | Sense-data theory is indirect realism, but phenomenalism is direct irrealism [Audi,R] |
| Full Idea: Where the sense-datum theory is an indirect realism, phenomenalism is a direct irrealism. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], I p.43) |
| 2721 | If you gradually remove a book's sensory properties, what is left at the end? [Audi,R] |
| Full Idea: If you imagine subtracting a book's sensory properties one by one, what is left of it? | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], I p.42) |
| 2727 | Red and green being exclusive colours seems to be rationally graspable but not analytic [Audi,R] |
| Full Idea: The proposition that nothing is red and green all over at once is not analytic, but it is rationally graspable, so it seems to be an a priori synthetic proposition. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], IV p.100) |
| 2728 | The concepts needed for a priori thought may come from experience [Audi,R] |
| Full Idea: I may well need experience to acquire the concepts needed for knowledge of the a priori, such as the concept of a colour. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], IV p.103) |
| 2716 | To see something as a field, I obviously need the concept of a field [Audi,R] |
| Full Idea: The propositional belief which portrays what I see in front of me AS a field requires my having a concept of one. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], I p.17) | |
| A reaction: To me this immediately invites the question of what a cow or horse experiences when they look at a familiar field. They know how to leave and enter it, and register its boundaries and qualities. Concepts? |
| 2717 | How could I see a field and believe nothing regarding it? [Audi,R] |
| Full Idea: How could I see a field and believe nothing regarding it? | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], I p.20) |
| 2720 | Sense-data (and the rival 'adverbial' theory) are to explain illusions and hallucinations [Audi,R] |
| Full Idea: The sense-datum theory is mainly to explain hallucinations and illusions, though there might be other theories, such as the 'adverbial' theory. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], I p.36) |
| 2719 | Sense data imply representative realism, possibly only representing primary qualities [Audi,R] |
| Full Idea: A sense-datum theory might be called a representative realism because it conceives perception as a relation in which sense-data represent perceived external (hence real) objects to us. For Locke they were resemblances only of primary qualities. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], I p.33) |
| 2718 | Perception is first simple, then objectual (with concepts) and then propositional [Audi,R] |
| Full Idea: Simple perceiving gives rise to objectual perceiving (attaching concepts to the object), which gives rise to propositional perceiving. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], I p.23) |
| 2741 | The principles of justification have to be a priori [Audi,R] |
| Full Idea: The crucial principles of justification are a priori. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], X p.311) |
| 2729 | Virtually all rationalists assert that we can have knowledge of synthetic a priori truths [Audi,R] |
| Full Idea: Rationalists virtually always assert or imply that, in addition to knowledge of analytic truths, there is knowledge of synthetic a priori truths. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], IV p.105) |
| 13932 | Empiricists tend to reject abstract entities, and to feel sympathy with nominalism [Carnap] |
| Full Idea: Empiricists are in general rather suspicious with respect to any kind of abstract entities like properties, classes, relations, numbers, propositions etc. They usually feel more sympathy with nominalists than with realists (in the medieval sense). | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 1) | |
| A reaction: The obvious reason is that you can't have sense experiences of abstract entities. I like the question 'what are they made of?' rather than the question 'how can I experience them?'. |
| 13937 | New linguistic claims about entities are not true or false, but just expedient, fruitful or successful [Carnap] |
| Full Idea: The acceptance of new linguistic forms about entities cannot be judged as being either true or false because it is not an assertion. It can only be judged as being more or less expedient, fruitful, conducive to the aim for which the language is intended. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 3) | |
| A reaction: The obvious problem seems to be that a complete pack of lies might be successful for a very long time, if it plugged a critical hole in a major theory. Is success judged financially? How do we judge success without mentioning truth? |
| 2725 | To remember something is to know it [Audi,R] |
| Full Idea: Remembering something is so entails knowing that it is so. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], II p.68) | |
| A reaction: Clearly I can say I "remember" x, but be wrong. Presumably we then say that I didn't really remember, which requires success, like "I know". It is true (as with "know") that as soon as I say that the something is false, I can't claim to remember it. |
| 2724 | I might remember someone I can't recall or image, by recognising them on meeting [Audi,R] |
| Full Idea: If I can neither recall nor image Jane I can still remember her, for on seeing her I might recognise her, and might remember, and even recall, our last meeting. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], II p.66) | |
| A reaction: Hm. I can hardly claim to remember her if I have no concept of her, and don't recall our last meeting. If seeing her triggers recognition, I would say that I NOW remember her, but I didn't before. Memory is more conscious than Audi claims. |
| 2731 | Justification is either unanchored (infinite or circular), or anchored (in knowledge or non-knowledge) [Audi,R] |
| Full Idea: There are four possible kinds of epistemic chain: infinite and unanchored, circular and unanchored, anchored in a belief which is not knowledge, and anchored in a belief which is bedrock knowledge. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VII p.183) | |
| A reaction: About right, though I don't think 'chain' is the right word for what is proposed if justification is to be coherent. The justifications float like lilies in the pond of reason, and a Self (Monet?) seems needed to assess the picture |
| 2739 | Internalism about justification implies that there is a right to believe something [Audi,R] |
| Full Idea: Internalists about justification tend to conceive of it as a matter of having a right to believe something. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VIII p.234) | |
| A reaction: I'm an internalist, but I don't understand this, unless it refers to the social aspect of justification. Can I grant myself internal rights? I can justify my belief to other people. |
| 2732 | Maths may be consistent with observations, but not coherent [Audi,R] |
| Full Idea: That 7+5=12 and that carrots are nourishing are mutually consistent, but do not exhibit coherence. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VII p.192) | |
| A reaction: This shows how difficult it would be to define 'coherent'. Is 'carrots are nourishing' coherent with 'fish are nourishing'? Is the battle of Hastings coherent with the battle of Waterloo? |
| 2733 | It is very hard to show how much coherence is needed for justification [Audi,R] |
| Full Idea: It is very difficult to specify when an explanatory relation generates enough coherence to create justification. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VII p.193) | |
| A reaction: I take coherence to be the key concept in epistemology, and quite impossible to define. This is why the 'space of reasons' is a useful concept. It is a courtroom, in which each case is different. |
| 2734 | A consistent madman could have a very coherent belief system [Audi,R] |
| Full Idea: A schizophrenic who thinks he is Napoleon, if he has a completely consistent story with enough interlocking details, may have a belief system that is superbly coherent. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VII p.194) | |
| A reaction: This is an exaggeration, but the fact is that one isolated lie is totally coherent, so coherence can only emerge when a system is large. Sense experience must be central to coherence. |
| 2738 | Consistent accurate prediction looks like knowledge without justified belief [Audi,R] |
| Full Idea: If someone consistently prophesied the winners of horse races, it appears that this man knows who will win the races, but surely he does not have justified beliefs as to who will win? | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VIII p.229) | |
| A reaction: This is where internalists and externalists (notably reliabilists) sharply part company. IF a reliable clairvoyant appeared, we would eventually accept them as a knower. But they DON'T appear, because knowledge needs justification! |
| 2740 | A reliability theory of knowledge seems to involve truth as correspondence [Audi,R] |
| Full Idea: If one favours a reliability theory of knowledge (which is externalist) the correspondence theory of truth seems the most appropriate. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VIII p.243) | |
| A reaction: Sounds right. Coherence implies some sort of internal assessment, whereas correspondence just needs to plugged into the facts. I like coherence justification and correspondence truth. |
| 2737 | 'Reliable' is a very imprecise term, and may even mean 'justified' [Audi,R] |
| Full Idea: Reliabilism cannot specify how reliable a process must be before it grounds knowledge, and it cannot specify what is reliable in the first place. 'Reliable' may become circular, and may mean 'justified'. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], VIII p.225) | |
| A reaction: The first time you ever read an instrument, or talk to a stranger, you have no indication of reliability. Circularity looks like a big problem. Knowledge must precede reliability? |
| 3061 | Anaxarchus said that he was not even sure that he knew nothing [Anaxarchus, by Diog. Laertius] |
| Full Idea: Anaxarchus said that he was not even sure that he knew nothing. | |
| From: report of Anaxarchus (fragments/reports [c.340 BCE]) by Diogenes Laertius - Lives of Eminent Philosophers 09.10.1 |
| 18699 | Carnap tried to define all scientific predicates in terms of primitive relations, using type theory [Carnap, by Button] |
| Full Idea: Carnap's ultimate ambition in the Aufbau is to provide a constitution-system within which any predicate of any scientific vocabulary can be explicitly defined in terms of primitive relations holding among basic elements, using type theory. | |
| From: report of Rudolph Carnap (The Logical Structure of the World (Aufbau) [1928]) by Tim Button - The Limits of Reason 05.2 | |
| A reaction: David Chalmers has a modern shot at the same project in 'Constructing the World'. Ramsey sentences seem to be part of the same game. |
| 13940 | All linguistic forms in science are merely judged by their efficiency as instruments [Carnap] |
| Full Idea: The acceptance or rejection of abstract (or any other) linguistic forms in any branch of science will finally be decided by their efficiency as instruments. | |
| From: Rudolph Carnap (Empiricism, Semantics and Ontology [1950], 5) |
| 13048 | Good explications are exact, fruitful, simple and similar to the explicandum [Carnap, by Salmon] |
| Full Idea: Carnap's four criteria for giving a good explication are similarity to the explicandum, exactness, fruitfulness and simplicity. | |
| From: report of Rudolph Carnap (Logical Foundations of Probability [1950], Ch.1) by Wesley Salmon - Four Decades of Scientific Explanation 0.1 | |
| A reaction: [compressed] Salmon's view is that this represents the old attitude, that the contribution of philosophy to explanation is the clarification of the key concepts. Carnap is, of course, a logical empiricist. |
| 2726 | We can be ignorant about ourselves, for example, our desires and motives [Audi,R] |
| Full Idea: We can have false beliefs, or some degree of ignorance, about our own mental lives. For example, about our own dispositions, such as not believing that we have a certain ignoble desire. | |
| From: Robert Audi (Epistemology: contemporary introduction [1998], III p.83) | |
| A reaction: This idea, that we don't know ourselves, has become a commonplace of recent philosophy, but I am unconvinced. Mostly we know only too well that we harbour a base desire, and we feel a creeping sense of shame. Total ignorance is very rare. |
| 12131 | All concepts can be derived from a few basics, making possible one science of everything [Carnap, by Brody] |
| Full Idea: In the 'Aufbau', Carnap tried to show how all of our concepts can be derived from a few basic concepts. ..Consequently there can be one science which studied all that existed, the science of the objects corresponding to the basic concepts. | |
| From: report of Rudolph Carnap (The Logical Structure of the World (Aufbau) [1928]) by Baruch Brody - Identity and Essence 2.2 | |
| A reaction: This is Carnap's Constructionist programme. |
| 11968 | The intension of a sentence is the set of all possible worlds in which it is true [Carnap, by Kaplan] |
| Full Idea: Carnap's proposal is to understand the category of intensions appropriate to sentences (his 'propositions') as sets of possible worlds. The intension of the sentence is taken as the set of all possible worlds in which the sentence is true. | |
| From: report of Rudolph Carnap (Meaning and Necessity [1947]) by David Kaplan - Transworld Heir Lines p.90 | |
| A reaction: [reference?] This extension of the truth-conditions view of meaning strikes me as being very attractive. Except that whole worlds hardly seem to be relevant to my remark about how lunch might have been improved. |
| 18285 | All translation loses some content (but language does not create reality) [Carnap] |
| Full Idea: I do not believe in translatability without loss of content, and therefore I think that the content of a world description is influenced to a certain degree by choice of a language form. But that does not mean that reality is created through language. | |
| From: Rudolph Carnap (Letters to Schlick [1935], 1935.12.04), quoted by J. Alberto Coffa - The Semantic Tradition from Kant to Carnap 19 'Truth' | |
| A reaction: It is a mistake to think Quine was the first to spot the interest of translation in philosophy of language. 'Does translation always lose content?' is a very nice question for focusing the problem. |
| 20064 | Actions are not mere effects of reasons, but are under their control [Audi,R] |
| Full Idea: An action for a reason is one that is, in a special way, under the control of reason. It is a response to, not a mere effect of, a reason. | |
| From: Robert Audi (Action, Intention and Reason [1992], p.177), quoted by Rowland Stout - Action 6 'Alien' | |
| A reaction: This modifies Davidson's 'reasons are causes'. Audi has a deviant causal chain which causes trouble for his idea, but Stout says he is right to focus on causal 'processes' (an Aristotelian idea) rather than causal 'chains'. |