54 ideas
| 10455 | Free logic at least allows empty names, but struggles to express non-existence [Bach] |
| Full Idea: Unlike standard first-order logic, free logic can allow empty names, but still has to deny existence by either representing it as a predicate, or invoke some dubious distinction such as between existence and being. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L1) |
| 10676 | The Axiom of Choice is a non-logical principle of set-theory [Hossack] |
| Full Idea: The Axiom of Choice seems better treated as a non-logical principle of set-theory. | |
| From: Keith Hossack (Plurals and Complexes [2000], 4 n8) | |
| A reaction: This reinforces the idea that set theory is not part of logic (and so pure logicism had better not depend on set theory). |
| 10686 | The Axiom of Choice guarantees a one-one correspondence from sets to ordinals [Hossack] |
| Full Idea: We cannot explicitly define one-one correspondence from the sets to the ordinals (because there is no explicit well-ordering of R). Nevertheless, the Axiom of Choice guarantees that a one-one correspondence does exist, even if we cannot define it. | |
| From: Keith Hossack (Plurals and Complexes [2000], 10) |
| 23623 | Predicativism says only predicated sets exist [Hossack] |
| Full Idea: Predicativists doubt the existence of sets with no predicative definition. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], 02.3) | |
| A reaction: This would imply that sets which encounter paradoxes when they try to be predicative do not therefore exist. Surely you can have a set of random objects which don't fall under a single predicate? |
| 23624 | The iterative conception has to appropriate Replacement, to justify the ordinals [Hossack] |
| Full Idea: The iterative conception justifies Power Set, but cannot justify a satisfactory theory of von Neumann ordinals, so ZFC appropriates Replacement from NBG set theory. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], 09.9) | |
| A reaction: The modern approach to axioms, where we want to prove something so we just add an axiom that does the job. |
| 23625 | Limitation of Size justifies Replacement, but then has to appropriate Power Set [Hossack] |
| Full Idea: The limitation of size conception of sets justifies the axiom of Replacement, but cannot justify Power Set, so NBG set theory appropriates the Power Set axiom from ZFC. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], 09.9) | |
| A reaction: Which suggests that the Power Set axiom is not as indispensable as it at first appears to be. |
| 10687 | Maybe we reduce sets to ordinals, rather than the other way round [Hossack] |
| Full Idea: We might reduce sets to ordinal numbers, thereby reversing the standard set-theoretical reduction of ordinals to sets. | |
| From: Keith Hossack (Plurals and Complexes [2000], 10) | |
| A reaction: He has demonstrated that there are as many ordinals as there are sets. |
| 10677 | Extensional mereology needs two definitions and two axioms [Hossack] |
| Full Idea: Extensional mereology defs: 'distinct' things have no parts in common; a 'fusion' has some things all of which are parts, with no further parts. Axioms: (transitivity) a part of a part is part of the whole; (sums) any things have a unique fusion. | |
| From: Keith Hossack (Plurals and Complexes [2000], 5) |
| 10454 | In first-order we can't just assert existence, and it is very hard to deny something's existence [Bach] |
| Full Idea: In standard logic we can't straightforwardly say that n exists. We have to resort to using a formula like '∃x(x=n)', but we can't deny n's existence by negating that formula, because standard first-order logic disallows empty names. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L1) |
| 23628 | The connective 'and' can have an order-sensitive meaning, as 'and then' [Hossack] |
| Full Idea: The sentence connective 'and' also has an order-sensitive meaning, when it means something like 'and then'. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], 10.4) | |
| A reaction: This is support the idea that orders are a feature of reality, just as much as possible concatenation. Relational predicates, he says, refer to series rather than to individuals. Nice point. |
| 10453 | In logic constants play the role of proper names [Bach] |
| Full Idea: In standard first-order logic the role of proper names is played by individual constants. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L1) |
| 23627 | 'Before' and 'after' are not two relations, but one relation with two orders [Hossack] |
| Full Idea: The reason the two predicates 'before' and 'after' are needed is not to express different relations, but to indicate its order. Since there can be difference of order without difference of relation, the nature of relations is not the source of order. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], 10.3) | |
| A reaction: This point is to refute Russell's 1903 claim that order arises from the nature of relations. Hossack claims that it is ordered series which are basic. I'm inclined to agree with him. |
| 10452 | Proper names can be non-referential - even predicate as well as attributive uses [Bach] |
| Full Idea: Like it or not, proper names have non-referential uses, including not only attributive but even predicate uses. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L1) | |
| A reaction: 'He's a right little Hitler'. 'You're doing a George Bush again'. 'Try to live up to the name of Churchill'. |
| 10456 | Millian names struggle with existence, empty names, identities and attitude ascription [Bach] |
| Full Idea: The familiar problems with the Millian view of names are the problem of positive and negative existential statements, empty names, identity sentences, and propositional attitude ascription. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L1) | |
| A reaction: I take this combination of problems to make an overwhelming case against the daft idea that the semantics of a name amounts to the actual object it picks out. It is a category mistake to attempt to insert a person into a sentence. |
| 10440 | An object can be described without being referred to [Bach] |
| Full Idea: An object can be described without being referred to. | |
| From: Kent Bach (What Does It Take to Refer? [2006], Intro) | |
| A reaction: I'm not clear how this is possible for a well-known object, though it is clearly possible for a speculative object, such as a gadget I would like to buy. In the former case reference seems to occur even if the speaker is trying to avoid it. |
| 10444 | Definite descriptions can be used to refer, but are not semantically referential [Bach] |
| Full Idea: If Russell is, as I believe, basically right, then definite descriptions are the paradigm of singular terms that can be used to refer but are not linguistically (semantically) referential. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.1 s5) | |
| A reaction: I'm not sure that we can decide what is 'semantically referential'. Most of the things we refer to don't have names. We don't then 'use' definite descriptions (I'm thinking) - they actually DO the job. If we use them, we can 'use' names too? |
| 10671 | Plural definite descriptions pick out the largest class of things that fit the description [Hossack] |
| Full Idea: If we extend the power of language with plural definite descriptions, these would pick out the largest class of things that fit the description. | |
| From: Keith Hossack (Plurals and Complexes [2000], 3) |
| 10666 | Plural reference will refer to complex facts without postulating complex things [Hossack] |
| Full Idea: It may be that plural reference gives atomism the resources to state complex facts without needing to refer to complex things. | |
| From: Keith Hossack (Plurals and Complexes [2000], 1) | |
| A reaction: This seems the most interesting metaphysical implication of the possibility of plural quantification. |
| 10675 | A plural comprehension principle says there are some things one of which meets some condition [Hossack] |
| Full Idea: Singular comprehension principles have a bad reputation, but the plural comprehension principle says that given a condition on individuals, there are some things such that something is one of them iff it meets the condition. | |
| From: Keith Hossack (Plurals and Complexes [2000], 4) |
| 10669 | Plural reference is just an abbreviation when properties are distributive, but not otherwise [Hossack] |
| Full Idea: If all properties are distributive, plural reference is just a handy abbreviation to avoid repetition (as in 'A and B are hungry', to avoid 'A is hungry and B is hungry'), but not all properties are distributive (as in 'some people surround a table'). | |
| From: Keith Hossack (Plurals and Complexes [2000], 2) | |
| A reaction: The characteristic examples to support plural quantification involve collective activity and relations, which might be weeded out of our basic ontology, thus leaving singular quantification as sufficient. |
| 10673 | Plural language can discuss without inconsistency things that are not members of themselves [Hossack] |
| Full Idea: In a plural language we can discuss without fear of inconsistency the things that are not members of themselves. | |
| From: Keith Hossack (Plurals and Complexes [2000], 4) | |
| A reaction: [see Hossack for details] |
| 10680 | The theory of the transfinite needs the ordinal numbers [Hossack] |
| Full Idea: The theory of the transfinite needs the ordinal numbers. | |
| From: Keith Hossack (Plurals and Complexes [2000], 8) |
| 10684 | I take the real numbers to be just lengths [Hossack] |
| Full Idea: I take the real numbers to be just lengths. | |
| From: Keith Hossack (Plurals and Complexes [2000], 9) | |
| A reaction: I love it. Real numbers are beginning to get on my nerves. They turn up to the party with no invitation and improperly dressed, and then refuse to give their names when challenged. |
| 23626 | Transfinite ordinals are needed in proof theory, and for recursive functions and computability [Hossack] |
| Full Idea: The transfinite ordinal numbers are important in the theory of proofs, and essential in the theory of recursive functions and computability. Mathematics would be incomplete without them. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], 10.1) | |
| A reaction: Hossack offers this as proof that the numbers are not human conceptual creations, but must exist beyond the range of our intellects. Hm. |
| 10674 | A plural language gives a single comprehensive induction axiom for arithmetic [Hossack] |
| Full Idea: A language with plurals is better for arithmetic. Instead of a first-order fragment expressible by an induction schema, we have the complete truth with a plural induction axiom, beginning 'If there are some numbers...'. | |
| From: Keith Hossack (Plurals and Complexes [2000], 4) |
| 10681 | In arithmetic singularists need sets as the instantiator of numeric properties [Hossack] |
| Full Idea: In arithmetic singularists need sets as the instantiator of numeric properties. | |
| From: Keith Hossack (Plurals and Complexes [2000], 8) |
| 10685 | Set theory is the science of infinity [Hossack] |
| Full Idea: Set theory is the science of infinity. | |
| From: Keith Hossack (Plurals and Complexes [2000], 10) |
| 23621 | Numbers are properties, not sets (because numbers are magnitudes) [Hossack] |
| Full Idea: I propose that numbers are properties, not sets. Magnitudes are a kind of property, and numbers are magnitudes. …Natural numbers are properties of pluralities, positive reals of continua, and ordinals of series. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], Intro) | |
| A reaction: Interesting! Since time can have a magnitude (three weeks) just as liquids can (three litres), it is not clear that there is a single natural property we can label 'magnitude'. Anything we can manage to measure has a magnitude. |
| 23622 | We can only mentally construct potential infinities, but maths needs actual infinities [Hossack] |
| Full Idea: Numbers cannot be mental objects constructed by our own minds: there exists at most a potential infinity of mental constructions, whereas the axioms of mathematics require an actual infinity of numbers. | |
| From: Keith Hossack (Knowledge and the Philosophy of Number [2020], Intro 2) | |
| A reaction: Doubt this, but don't know enough to refute it. Actual infinities were a fairly late addition to maths, I think. I would think treating fictional complete infinities as real would be sufficient for the job. Like journeys which include imagined roads. |
| 10668 | We are committed to a 'group' of children, if they are sitting in a circle [Hossack] |
| Full Idea: By Quine's test of ontological commitment, if some children are sitting in a circle, no individual child can sit in a circle, so a singular paraphrase will have us committed to a 'group' of children. | |
| From: Keith Hossack (Plurals and Complexes [2000], 2) | |
| A reaction: Nice of why Quine is committed to the existence of sets. Hossack offers plural quantification as a way of avoiding commitment to sets. But is 'sitting in a circle' a real property (in the Shoemaker sense)? I can sit in a circle without realising it. |
| 10664 | Complex particulars are either masses, or composites, or sets [Hossack] |
| Full Idea: Complex particulars are of at least three types: masses (which sum, of which we do not ask 'how many?' but 'how much?'); composite individuals (how many?, and summing usually fails); and sets (only divisible one way, unlike composites). | |
| From: Keith Hossack (Plurals and Complexes [2000], 1) | |
| A reaction: A composite pile of grains of sand gradually becomes a mass, and drops of water become 'water everywhere'. A set of people divides into individual humans, but redescribe the elements as the union of males and females? |
| 10678 | The relation of composition is indispensable to the part-whole relation for individuals [Hossack] |
| Full Idea: The relation of composition seems to be indispensable in a correct account of the part-whole relation for individuals. | |
| From: Keith Hossack (Plurals and Complexes [2000], 7) | |
| A reaction: This is the culmination of a critical discussion of mereology and ontological atomism. At first blush it doesn't look as if 'composition' has much chance of being a precise notion, and it will be plagued with vagueness. |
| 10665 | Leibniz's Law argues against atomism - water is wet, unlike water molecules [Hossack] |
| Full Idea: We can employ Leibniz's Law against mereological atomism. Water is wet, but no water molecule is wet. The set of infinite numbers is infinite, but no finite number is infinite. ..But with plural reference the atomist can resist this argument. | |
| From: Keith Hossack (Plurals and Complexes [2000], 1) | |
| A reaction: The idea of plural reference is to state plural facts without referring to complex things, which is interesting. The general idea is that we have atomism, and then all the relations, unities, identities etc. are in the facts, not in the things. I like it. |
| 10682 | The fusion of five rectangles can decompose into more than five parts that are rectangles [Hossack] |
| Full Idea: The fusion of five rectangles may have a decomposition into more than five parts that are rectangles. | |
| From: Keith Hossack (Plurals and Complexes [2000], 8) |
| 12900 | How could 'S knows he has hands' not have a fixed content? [Bach] |
| Full Idea: How can it be that a sentence like 'George knows that he has hands', even with time and references fixed, does not have a fixed propositional content? | |
| From: Kent Bach (The Emperor's New 'Knows' [2005], I) | |
| A reaction: The appeal is to G.E. Moore's common sense view of immediate knowledge (Idea 6349). The reply is simply that the word 'knows' shifts its meaning, having high standards in sceptical philosophy classes, and low standards on the street. |
| 12901 | If contextualism is right, knowledge sentences are baffling out of their context [Bach] |
| Full Idea: Contextualism seems to predict that if you encounter a knowledge attribution out of context you won't be in a position to grasp which proposition the sentence expresses. | |
| From: Kent Bach (The Emperor's New 'Knows' [2005], I) | |
| A reaction: It is only the word 'knows' which is at issue in the sentence. If someone is said to 'know' about the world of the fairies, we might well be puzzled as to what proposition was being expressed. Is the word 'flat' baffling out of context? |
| 12902 | Sceptics aren't changing the meaning of 'know', but claiming knowing is tougher than we think [Bach] |
| Full Idea: When a sceptic brings up far-fetched possibilities and argues that we can't rule them out, he is not raising the standard for the word 'know'. He is showing it is tougher than we realise for a belief to qualify as normal knowledge at all. | |
| From: Kent Bach (The Emperor's New 'Knows' [2005], III) | |
| A reaction: [Bach cites Richard Feldman for this idea] I think that what happens in the contextual account is that 'true', 'belief' and 'know' retain their standard meaning, and it is 'justified' which shifts. 'I am fully justified' can have VERY different meanings! |
| 10663 | A thought can refer to many things, but only predicate a universal and affirm a state of affairs [Hossack] |
| Full Idea: A thought can refer to a particular or a universal or a state of affairs, but it can predicate only a universal and it can affirm only a state of affairs. | |
| From: Keith Hossack (Plurals and Complexes [2000], 1) | |
| A reaction: Hossack is summarising Armstrong's view, which he is accepting. To me, 'thought' must allow for animals, unlike language. I think Hossack's picture is much too clear-cut. Do animals grasp universals? Doubtful. Can they predicate? Yes. |
| 10447 | We can refer to fictional entities if they are abstract objects [Bach] |
| Full Idea: If fictional entities, such as characters in a play, are real, albeit abstract entities, then we can genuinely refer to them. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.1) | |
| A reaction: [He cites Nathan Salmon 1998] Personally I would prefer to say that abstract entities are fictions. Fictional characters have uncertain identity conditions. Do they all have a pancreas, if this is never mentioned? |
| 10443 | You 'allude to', not 'refer to', an individual if you keep their identity vague [Bach] |
| Full Idea: If you say 'a special person is coming to visit', you are not referring to but merely 'alluding to' that individual. This does not count as referring because you are not expressing a singular proposition about it. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.1 s2) | |
| A reaction: If you add 'I hope he doesn't wear his red suit, but I hope he plays his tuba', you seem to be expressing singular propositions about the person. Bach seems to want a very strict notion of reference, as really attaching listeners to individuals. |
| 10446 | Fictional reference is different inside and outside the fiction [Bach] |
| Full Idea: We must distinguish 'reference' in a fiction from reference outside the fiction to fictional entities. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.1) | |
| A reaction: This may be more semantically than ontologically significant. It is perhaps best explicated by Coleridge's distinction over whether or not I am 'suspending my disbelief' when I am discussing a character. |
| 10439 | What refers: indefinite or definite or demonstrative descriptions, names, indexicals, demonstratives? [Bach] |
| Full Idea: Philosophers agree that some expressions refer, but disagree over which ones. Few include indefinite descriptions, but some include definite descriptions, or only demonstrative descriptions. Some like proper names, some only indexicals and demonstratives. | |
| From: Kent Bach (What Does It Take to Refer? [2006], Intro) | |
| A reaction: My initial prejudice is rather Strawsonian - that people refer, not language, and it can be done in all sorts of ways. But Bach argues well that only language intrinsically does it. Even pointing fails without linguistic support. |
| 10441 | If we can refer to things which change, we can't be obliged to single out their properties [Bach] |
| Full Idea: We can refer to things which change over time, which suggests that in thinking of and in referring to an individual we are not constrained to represent it as that which has certain properties. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.1) | |
| A reaction: This seems a good argument against the descriptive theory of reference which is not (I think) in Kripke. Problems like vagueness and the Ship of Theseus rear their heads. |
| 10442 | We can think of an individual without have a uniquely characterizing description [Bach] |
| Full Idea: Being able to think of an individual does not require being able to identify that individual by means of a uniquely characterizing description. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.1 s1) | |
| A reaction: There is a bit of an equivocation over 'recognise' here. His example is 'the first child born in the 4th century'. We can't visually recognise such people, but the description does fix them, and a records office might give us 'recognition'. |
| 10445 | It can't be real reference if it could refer to some other thing that satisfies the description [Bach] |
| Full Idea: If one is referring to whatever happens to satisfy a description, and one would be referring to something else were it to have satisfied the description instead, this is known as 'weak' reference,...but surely this is not reference at all. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.1 s7) | |
| A reaction: Bach wants a precise notion of reference, as success in getting the audience to focus on the correct object. He talks of this case as 'singling out' some unfixed thing, and he also has 'alluding to' an unstated thing. Plausible view. |
| 10457 | Since most expressions can be used non-referentially, none of them are inherently referential [Bach] |
| Full Idea: An embarrassingly simple argument is that most expressions can be used literally but not referentially, no variation in meaning explains this fact, so its meaning is compatible with being non-referential, so no expression is inherently referential. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L2) | |
| A reaction: I think I have decided that no expression is 'inherently referential', and that it is all pragmatics. |
| 10463 | Just alluding to or describing an object is not the same as referring to it [Bach] |
| Full Idea: Much of what speakers do that passes for referring is merely alluding or describing. ...It is one thing for a speaker to express a thought about a certain object using an expression, and quite another for the expression to stand for that object. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.3) | |
| A reaction: Bach builds up a persuasive case for this view. If the question, though, is 'what are you talking about?', then saying what is being alluded to or singled out or described seems fine. Bach is being rather stipulative. |
| 10459 | Context does not create reference; it is just something speakers can exploit [Bach] |
| Full Idea: Context does not determine or constitute reference; rather, it is something for the speaker to exploit to enable the listener to determine the intended reference. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L3) | |
| A reaction: Bach thinks linguistic reference is a matter of speaker's intentions, and I think he is right. And this idea is right too. The domain of quantification constantly shifts in a conversation, and good speakers and listeners are sensitive to this. |
| 10460 | 'That duck' may not refer to the most obvious one in the group [Bach] |
| Full Idea: If one ducks starts quacking furiously, and you say 'that duck is excited', it isn't context that makes me take it that you are referring to the quacking duck. You could be referring to a quiet duck you recognise by its distinctive colour. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L3) | |
| A reaction: A persuasive example to make his point against the significance of context in conversational reference. Speaker's intended reference must always trump any apparent reference suggested by context. |
| 10461 | What a pronoun like 'he' refers back to is usually a matter of speaker's intentions [Bach] |
| Full Idea: To illustrate speakers' intentions, consider the anaphoric reference using pronouns in these: "A cop arrested a robber; he was wearing a badge", and "A cop arrested a robber; he was wearing a mask". The natural supposition is not the inevitable one. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L4) | |
| A reaction: I am a convert to speakers' intentions as the source of all reference, and this example seems to illustrate it very well. 'He said..' 'Who said?' |
| 10462 | Information comes from knowing who is speaking, not just from interpretation of the utterance [Bach] |
| Full Idea: It is a fallacy that all the information in an utterance must come from its interpretation, which ignores the essentially pragmatic fact that the speaker is making the utterance. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L4) | |
| A reaction: [He cites Barwise and Perry 1983:34] This is blatantly obvious in indexical remarks like 'I am tired', where the words don't tell you who is tired. But also 'the car has broken down, dear'. |
| 10458 | People slide from contextual variability all the way to contextual determination [Bach] |
| Full Idea: People slide from contextual variability to context relativity to context sensitivity to context dependence to contextual determination. | |
| From: Kent Bach (What Does It Take to Refer? [2006], 22.2 L3) | |
| A reaction: This is reminiscent of the epistemological slide from cultural or individual relativity of some observed things, to a huge metaphysical denial of truth. Bach's warning applies to me, as I have been drifting down his slope lately. Nice. |
| 6005 | Animals are dangerous and nourishing, and can't form contracts of justice [Hermarchus, by Sedley] |
| Full Idea: Hermarchus said that animal killing is justified by considerations of human safety and nourishment and by animals' inability to form contractual relations of justice with us. | |
| From: report of Hermarchus (fragments/reports [c.270 BCE]) by David A. Sedley - Hermarchus | |
| A reaction: Could the last argument be used to justify torturing animals? Or could we eat a human who was too brain-damaged to form contracts? |
| 10683 | We could ignore space, and just talk of the shape of matter [Hossack] |
| Full Idea: We might dispense with substantival space, and say that if the distribution of matter in space could have been different, that just means the matter of the Universe could have been shaped differently (with geometry as the science of shapes). | |
| From: Keith Hossack (Plurals and Complexes [2000], 9) |