44 ideas
| 11147 | Naturalistic philosophers oppose analysis, preferring explanation to a priori intuition [Margolis/Laurence] |
| Full Idea: Philosophers who oppose conceptual analysis identify their approach as being 'naturalistic'. Philosophy is supposed to be continuous with science, and philosophical theories are to be defended on explanatory grounds, not by a priori intuitions. | |
| From: E Margolis/S Laurence (Concepts [2009], 5.2) | |
| A reaction: [They cite Papineau 1993, Devitt 1996 aand Kornblith 2002] I think there is a happy compromise here. I agree that any philosophical knowledge should be continuous with science, but we shouldn't prejudge how the analytic branch of science is done. |
| 13520 | A 'tautology' must include connectives [Wolf,RS] |
| Full Idea: 'For every number x, x = x' is not a tautology, because it includes no connectives. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 1.2) |
| 13524 | Deduction Theorem: T∪{P}|-Q, then T|-(P→Q), which justifies Conditional Proof [Wolf,RS] |
| Full Idea: Deduction Theorem: If T ∪ {P} |- Q, then T |- (P → Q). This is the formal justification of the method of conditional proof (CPP). Its converse holds, and is essentially modus ponens. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 1.3) |
| 13522 | Universal Generalization: If we prove P(x) with no special assumptions, we can conclude ∀xP(x) [Wolf,RS] |
| Full Idea: Universal Generalization: If we can prove P(x), only assuming what sort of object x is, we may conclude ∀xP(x) for the same x. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 1.3) | |
| A reaction: This principle needs watching closely. If you pick one person in London, with no presuppositions, and it happens to be a woman, can you conclude that all the people in London are women? Fine in logic and mathematics, suspect in life. |
| 13521 | Universal Specification: ∀xP(x) implies P(t). True for all? Then true for an instance [Wolf,RS] |
| Full Idea: Universal Specification: from ∀xP(x) we may conclude P(t), where t is an appropriate term. If something is true for all members of a domain, then it is true for some particular one that we specify. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 1.3) |
| 13523 | Existential Generalization (or 'proof by example'): if we can say P(t), then we can say something is P [Wolf,RS] |
| Full Idea: Existential Generalization (or 'proof by example'): From P(t), where t is an appropriate term, we may conclude ∃xP(x). | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 1.3) | |
| A reaction: It is amazing how often this vacuous-sounding principles finds itself being employed in discussions of ontology, but I don't quite understand why. |
| 13529 | Empty Set: ∃x∀y ¬(y∈x). The unique empty set exists [Wolf,RS] |
| Full Idea: Empty Set Axiom: ∃x ∀y ¬ (y ∈ x). There is a set x which has no members (no y's). The empty set exists. There is a set with no members, and by extensionality this set is unique. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 2.3) | |
| A reaction: A bit bewildering for novices. It says there is a box with nothing in it, or a pair of curly brackets with nothing between them. It seems to be the key idea in set theory, because it asserts the idea of a set over and above any possible members. |
| 13526 | Comprehension Axiom: if a collection is clearly specified, it is a set [Wolf,RS] |
| Full Idea: The comprehension axiom says that any collection of objects that can be clearly specified can be considered to be a set. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 2.2) | |
| A reaction: This is virtually tautological, since I presume that 'clearly specified' means pinning down exact which items are the members, which is what a set is (by extensionality). The naïve version is, of course, not so hot. |
| 13534 | In first-order logic syntactic and semantic consequence (|- and |=) nicely coincide [Wolf,RS] |
| Full Idea: One of the most appealing features of first-order logic is that the two 'turnstiles' (the syntactic single |-, and the semantic double |=), which are the two reasonable notions of logical consequence, actually coincide. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.3) | |
| A reaction: In the excitement about the possibility of second-order logic, plural quantification etc., it seems easy to forget the virtues of the basic system that is the target of the rebellion. The issue is how much can be 'expressed' in first-order logic. |
| 13535 | First-order logic is weakly complete (valid sentences are provable); we can't prove every sentence or its negation [Wolf,RS] |
| Full Idea: The 'completeness' of first order-logic does not mean that every sentence or its negation is provable in first-order logic. We have instead the weaker result that every valid sentence is provable. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.3) | |
| A reaction: Peter Smith calls the stronger version 'negation completeness'. |
| 13531 | Model theory reveals the structures of mathematics [Wolf,RS] |
| Full Idea: Model theory helps one to understand what it takes to specify a mathematical structure uniquely. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.1) | |
| A reaction: Thus it is the development of model theory which has led to the 'structuralist' view of mathematics. |
| 13532 | Model theory 'structures' have a 'universe', some 'relations', some 'functions', and some 'constants' [Wolf,RS] |
| Full Idea: A 'structure' in model theory has a non-empty set, the 'universe', as domain of variables, a subset for each 'relation', some 'functions', and 'constants'. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.2) |
| 13519 | Model theory uses sets to show that mathematical deduction fits mathematical truth [Wolf,RS] |
| Full Idea: Model theory uses set theory to show that the theorem-proving power of the usual methods of deduction in mathematics corresponds perfectly to what must be true in actual mathematical structures. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], Pref) | |
| A reaction: That more or less says that model theory demonstrates the 'soundness' of mathematics (though normal arithmetic is famously not 'complete'). Of course, he says they 'correspond' to the truths, rather than entailing them. |
| 13533 | First-order model theory rests on completeness, compactness, and the Löwenheim-Skolem-Tarski theorem [Wolf,RS] |
| Full Idea: The three foundations of first-order model theory are the Completeness theorem, the Compactness theorem, and the Löwenheim-Skolem-Tarski theorem. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.3) | |
| A reaction: On p.180 he notes that Compactness and LST make no mention of |- and are purely semantic, where Completeness shows the equivalence of |- and |=. All three fail for second-order logic (p.223). |
| 13537 | An 'isomorphism' is a bijection that preserves all structural components [Wolf,RS] |
| Full Idea: An 'isomorphism' is a bijection between two sets that preserves all structural components. The interpretations of each constant symbol are mapped across, and functions map the relation and function symbols. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.4) |
| 13539 | The LST Theorem is a serious limitation of first-order logic [Wolf,RS] |
| Full Idea: The Löwenheim-Skolem-Tarski theorem demonstrates a serious limitation of first-order logic, and is one of primary reasons for considering stronger logics. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.7) |
| 13538 | If a theory is complete, only a more powerful language can strengthen it [Wolf,RS] |
| Full Idea: It is valuable to know that a theory is complete, because then we know it cannot be strengthened without passing to a more powerful language. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 5.5) |
| 13525 | Most deductive logic (unlike ordinary reasoning) is 'monotonic' - we don't retract after new givens [Wolf,RS] |
| Full Idea: Deductive logic, including first-order logic and other types of logic used in mathematics, is 'monotonic'. This means that we never retract a theorem on the basis of new givens. If T|-φ and T⊆SW, then S|-φ. Ordinary reasoning is nonmonotonic. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 1.7) | |
| A reaction: The classic example of nonmonotonic reasoning is the induction that 'all birds can fly', which is retracted when the bird turns out to be a penguin. He says nonmonotonic logic is a rich field in computer science. |
| 13530 | An ordinal is an equivalence class of well-orderings, or a transitive set whose members are transitive [Wolf,RS] |
| Full Idea: Less theoretically, an ordinal is an equivalence class of well-orderings. Formally, we say a set is 'transitive' if every member of it is a subset of it, and an ordinal is a transitive set, all of whose members are transitive. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], 2.4) | |
| A reaction: He glosses 'transitive' as 'every member of a member of it is a member of it'. So it's membership all the way down. This is the von Neumann rather than the Zermelo approach (which is based on singletons). |
| 13518 | Modern mathematics has unified all of its objects within set theory [Wolf,RS] |
| Full Idea: One of the great achievements of modern mathematics has been the unification of its many types of objects. It began with showing geometric objects numerically or algebraically, and culminated with set theory representing all the normal objects. | |
| From: Robert S. Wolf (A Tour through Mathematical Logic [2005], Pref) | |
| A reaction: His use of the word 'object' begs all sorts of questions, if you are arriving from the street, where an object is something which can cause a bruise - but get used to it, because the word 'object' has been borrowed for new uses. |
| 11141 | Modern empiricism tends to emphasise psychological connections, not semantic relations [Margolis/Laurence] |
| Full Idea: A growing number of philosophers are attracted to modified forms of empiricism, emphasizing psychological relations between the conceptual system and perceptual and motor states, not semantic relations. | |
| From: E Margolis/S Laurence (Concepts [2009], 3.2) | |
| A reaction: I suddenly spot that this is what I have been drifting towards for some time! The focus is concept formation, where the philosophers need to join forces with the cognitive scientists. |
| 11142 | Body-type seems to affect a mind's cognition and conceptual scheme [Margolis/Laurence] |
| Full Idea: It is claimed, on the basis of empirical research, that the type of body that an organism has profoundly affects it cognitive operations and the way it conceptualises the world. We can't assume that human minds could inhere in wildly different body types. | |
| From: E Margolis/S Laurence (Concepts [2009], 3.2) | |
| A reaction: Sounds interesting. They cite Lawrence Shapiro 2004. It needs a large effort of imagination to think how a snake or whale or albatross might conceptualise the world, in relation to their bodies. |
| 11121 | Language of thought has subject/predicate form and includes logical devices [Margolis/Laurence] |
| Full Idea: The language of thought is taken to have subject/predicate form and include logical devices, such as quantifiers and variables. | |
| From: E Margolis/S Laurence (Concepts [2009], 1.1) |
| 11120 | Concepts are either representations, or abilities, or Fregean senses [Margolis/Laurence] |
| Full Idea: The three main options for the ontological status of concepts are to identify them with mental representations, or with abilities, or with Fregean senses. | |
| From: E Margolis/S Laurence (Concepts [2009], 1) |
| 11122 | A computer may have propositional attitudes without representations [Margolis/Laurence] |
| Full Idea: It may be possible to have propositional attitudes without having the mental representations tokened in one's head. ...We may say a chess-playing computer thinks it should develop its queen early, though we know it has no representation with that content. | |
| From: E Margolis/S Laurence (Concepts [2009], 1.1) | |
| A reaction: [Thye cite Dennett - who talks of the 'intentional stance'] It is, of course, a moot point whether we would attribute a propositional attitude (such as belief) to a machine once we knew that it wasn't representing the relevant concepts. |
| 11124 | Do mental representations just lead to a vicious regress of explanations [Margolis/Laurence] |
| Full Idea: A standard criticism is that the mental representation view of concepts creates just another item whose significance bears explaining. Either we have a vicious regress, or we might as well explain external language directly. | |
| From: E Margolis/S Laurence (Concepts [2009], 1.2) | |
| A reaction: [They cite Dummett, with Wittgenstein in the background] I don't agree, because I think that explanation of concepts only stops when it dovetails into biology. |
| 11123 | Maybe the concept CAT is just the ability to discriminate and infer about cats [Margolis/Laurence] |
| Full Idea: The view that concepts are abilities (e.g. found in Brandom, Dummett and Millikan) would say that the concept CAT amounts to the ability to discriminate cats from non-cats and to draw certain inferences about cats. | |
| From: E Margolis/S Laurence (Concepts [2009], 1.2) | |
| A reaction: Feels wrong. The concept is what makes these abilities possible, but it seems rather behaviourist to identify the concept with what is enabled by the concept. You might understand 'cat', but fail to recognise your first cat (though you might suspect it). |
| 11125 | The abilities view cannot explain the productivity of thought, or mental processes [Margolis/Laurence] |
| Full Idea: The abilities view of concepts, by its rejection of mental representation, is ill-equipped to explain the productivity of thought; and it can say little about mental processes. | |
| From: E Margolis/S Laurence (Concepts [2009], 1.2) | |
| A reaction: The latter point arises from its behaviouristic character, which just gives us a black box with some output of abilities. In avoiding a possible regress, it offers no explanation at all. |
| 11140 | Concept-structure explains typicality, categories, development, reference and composition [Margolis/Laurence] |
| Full Idea: The structures of concepts are invoked to explain typicality effects, reflective categorization, cognitive development, reference determination, and compositionality. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.5) |
| 11128 | Classically, concepts give necessary and sufficient conditions for falling under them [Margolis/Laurence] |
| Full Idea: The classical theory is that a concept has a definitional structure in that it is composed of simpler concepts that express necessary and sufficient conditions for falling under the concept, the stock example being unmarried and a man for 'bachelor'. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.1) | |
| A reaction: This is the background idea to philosophy as analysis, and it makes concepts essentially referential, in that they are defined by their ability to pick things out. There must be some degree of truth in the theory. |
| 11130 | Typicality challenges the classical view; we see better fruit-prototypes in apples than in plums [Margolis/Laurence] |
| Full Idea: The classical view is challenged by the discovery that certain categories are taken to be more typical, with typicality widely correlating with other data. Apples are judged to be more typical of (and have more common features with) fruit than plums are. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.1) | |
| A reaction: This discovery that people use prototypes in thinking has been the biggest idea to ever hit the philosophy of concepts, and simply cannot be ignored (as long as the research keeps reinforcing it, which I believe it does). The classical view might adapt. |
| 11129 | The classical theory explains acquisition, categorization and reference [Margolis/Laurence] |
| Full Idea: The appeal of the classical theory of concepts is that it offers unified treatments of concept acquisition (assembling constituents), categorization (check constituents against target), and reference determination (whether they apply). | |
| From: E Margolis/S Laurence (Concepts [2009], 2.1) | |
| A reaction: [See Idea 11128 for the theory] As so often, I find myself in sympathy with the traditional view which has been relegated to ignominy by our wonderful modern philosophers. |
| 11131 | It may be that our concepts (such as 'knowledge') have no definitional structure [Margolis/Laurence] |
| Full Idea: In the light of problems such as the definition of knowledge, many philosophers now take seriously the possibility that our concepts lack definitional structure. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.1) | |
| A reaction: This challenges the classical view, that there are precise conditions for each concept. That view would obviously be in difficulties with atomic concepts, so our account of those might be applied all the way up. |
| 11132 | The prototype theory is probabilistic, picking something out if it has sufficient of the properties [Margolis/Laurence] |
| Full Idea: In the prototype theory of concepts, a lexical concept has probabilistic structure in that something falls under it if it satisfies a sufficient number of properties encoded by the constituents. It originates in Wittgenstein's 'family resemblance'. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.2) | |
| A reaction: It would seem unlikely to be a matter of the 'number' of properties, and would have to involve some notion of what was essential to the prototype. |
| 11133 | Prototype theory categorises by computing the number of shared constituents [Margolis/Laurence] |
| Full Idea: On the prototype theory, categorization is to be understood as a similarity comparison process, where similarity is computed as a function of the number of constituents that two concepts hold in common. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.2) | |
| A reaction: Again it strikes me that 'computing' similarity by mere 'number' of shared constituents won't do, as there is a prior judgement about which constituents really matter, or are essential. That may even be hard-wired. |
| 11134 | People don't just categorise by apparent similarities [Margolis/Laurence] |
| Full Idea: When it comes to more reflexive judgements, people go beyond the outcome of a similarity comparison. Even children say that a dog surgically altered to look like a raccoon is still a dog. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.2) | |
| A reaction: We can defend the theory by not underestimating people so much. Most categorisation is done on superficial grounds, but even children know there may be hidden similarities (behind the mask, under the bonnet) which are more important. |
| 11135 | Complex concepts have emergent properties not in the ingredient prototypes [Margolis/Laurence] |
| Full Idea: An objection to the prototype view concerns compositionality. A complex concept often has emergent properties, as when it seems that 'pet fish' encodes for brightly coloured, which has no basis in the prototypes for 'pet' or 'fish'. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.2) | |
| A reaction: I would take 'pet fish' to work like a database query. 'Fish' has a very vague prototype, and then 'pet fish' narrows the search to fish which are appropriate to be pets. We might say that the prototype is refined, or the Mk 2 prototype appears. |
| 11136 | Many complex concepts obviously have no prototype [Margolis/Laurence] |
| Full Idea: Many patently complex concepts don't even have a prototype structure, such as 'Chairs that were purchased on a Wednesday'. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.2) | |
| A reaction: [The example seems to be from Fodor] I disagree. If we accept the notion of 'refining' the prototype (see Idea 11135), then the compositionality of the expression will produce a genuine but very unusual prototype. |
| 11137 | The theory theory of concepts says they are parts of theories, defined by their roles [Margolis/Laurence] |
| Full Idea: The theory theory of concepts says that terms are related as in a scientific theory, and that categorization resembles theorising. It is generally assumed that scientific terms are interdefined so that content is determined by its role in the theory. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.3) | |
| A reaction: I never like this sort of account. What are the characteristics of the thing which enable it to fulfil its role? You haven't defined a car when you've said it gets you from A to B. |
| 11138 | The theory theory is holistic, so how can people have identical concepts? [Margolis/Laurence] |
| Full Idea: A problem with the theory theory of concepts is that it is holistic, saying a concept is determined by its role, not by its constituents. It then seems difficult for different people to possess the same concepts (or even the same person, over time). | |
| From: E Margolis/S Laurence (Concepts [2009], 2.3) | |
| A reaction: This seems a good objection to any holistic account of concepts or meaning - spotted by Plato in motivating his theory of Forms, to give the necessary stability to communication. |
| 11139 | Maybe concepts have no structure, and determined by relations to the world, not to other concepts [Margolis/Laurence] |
| Full Idea: According to conceptual atomism, lexical concepts have no semantic structure, and the content of a concept isn't determined by its relation to other concepts but by its relations to the world. | |
| From: E Margolis/S Laurence (Concepts [2009], 2.4) | |
| A reaction: [They cite Fodor 1998 and Millikan 2000] I like the sound of that, because I take the creation of concepts to be (in the first instance) a response to the world, not a response to other concepts. |
| 11146 | People can formulate new concepts which are only named later [Margolis/Laurence] |
| Full Idea: People seem to be able to formulate novel concepts which are left to be named later; the concept comes first, the name second. | |
| From: E Margolis/S Laurence (Concepts [2009], 4.2) | |
| A reaction: [This seems to have empirical support, and he cites Pinker 1994] I do not find this remotely surprising, since I presume that human concepts are a continuous kind with animal concepts, including non-conscious concepts (why not?). |
| 16713 | Philosophers are the forefathers of heretics [Tertullian] |
| Full Idea: Philosophers are the forefathers of heretics. | |
| From: Tertullian (works [c.200]), quoted by Robert Pasnau - Metaphysical Themes 1274-1671 20.2 |
| 6610 | I believe because it is absurd [Tertullian] |
| Full Idea: I believe because it is absurd ('Credo quia absurdum est'). | |
| From: Tertullian (works [c.200]), quoted by Robert Fogelin - Walking the Tightrope of Reason n4.2 | |
| A reaction: This seems to be a rather desperate remark, in response to what must have been rather good hostile arguments. No one would abandon the support of reason if it was easy to acquire. You can't deny its engaging romantic defiance, though. |