Ideas of Bob Hale, by Theme

[British, fl. 2001, Professor at Glasgow Univerity.]

green numbers give full details    |    back to list of philosophers    |     unexpand these ideas    |    
1. Philosophy / E. Nature of Metaphysics / 1. Nature of Metaphysics
You cannot understand what exists without understanding possibility and necessity
     Full Idea: I defend the thesis that questions about what kinds of things there are cannot be properly understood or adequately answered without recourse to considerations about possibility and necessity.
     From: Bob Hale (Necessary Beings [2013], Intro)
     A reaction: Good. I would say that this is a growing realisation in contemporary philosophy. The issue is focused when we ask what are the limitations of Quine's approach to metaphysics. If you don't see possibilities around you, you are a fool.
1. Philosophy / F. Analytic Philosophy / 5. Linguistic Analysis
Questions about objects are questions about certain non-vacuous singular terms
     Full Idea: I understand questions about the Fregean notion of an object to be inseparable from questions in the philosophy of language - questions of the existence of objects are tantamount to questions about non-vacuous singular terms of a certain kind.
     From: Bob Hale (Abstract Objects [1987], Ch.1)
     A reaction: This view hovers somewhere between Quine and J.L. Austin, and Dummett is its originator. I am instinctively deeply opposed to the identification of metaphysics with semantics.
2. Reason / D. Definition / 6. Definition by Essence
A canonical defintion specifies the type of thing, and what distinguish this specimen
     Full Idea: One might think of a full dress, or canonical, definition as specifying what type of thing it is, and what distinguishes it from everything else within its type.
     From: Bob Hale (Necessary Beings [2013], 06.4)
     A reaction: Good! At last someone embraces the Aristotelian ideas that definitions are a) quite extensive and detailed (unlike lexicography), and b) they aim to get right down to the individual. In that sense, an essence is captured by a definition.
2. Reason / D. Definition / 12. Paraphrase
An expression is a genuine singular term if it resists elimination by paraphrase
     Full Idea: An expression ... should be reckoned a genuine singular term only if it resists elimination by paraphrase.
     From: Bob Hale (Abstract Objects [1987], Ch.2.II)
     A reaction: This strikes me as extraordinarily optimistic. It will be relative to a language, and the resources of a given speaker, and seems open to the invention of new expressions to do the job (e.g. an equivalent adjective for every noun in the dictionary).
4. Formal Logic / D. Modal Logic ML / 7. Barcan Formula
The two Barcan principles are easily proved in fairly basic modal logic
     Full Idea: If the Brouwersche principle, p ⊃ □◊p is adjoined to a standard quantified vesion of the weakest modal logic K, then one can prove both the Barcan principle, and its converse.
     From: Bob Hale (Necessary Beings [2013], 09.2)
     A reaction: The Brouwersche principle (that p implies that p must be possible) sounds reasonable, but the Barcan principles strike me as false, so something has to give. They are theorems of S5. Hale proposes giving up classical logic.
With a negative free logic, we can dispense with the Barcan formulae
     Full Idea: I reject both Barcan and Converse Barcan by adopting a negative free logic.
     From: Bob Hale (Necessary Beings [2013], 11.3)
     A reaction: See section 9.2 of Hale's book, where he makes his case. I can't evaluate this bold move, though I don't like the Barcan Formulae. We can anticipate objections to Hale: are you prepared to embrace the unexpected consequences of your new logic?
5. Theory of Logic / A. Overview of Logic / 7. Second-Order Logic
If second-order variables range over sets, those are just objects; properties and relations aren't sets
     Full Idea: Contrary to what Quine supposes, it is neither necessary nor desirable to interpret bound higher-order variables as ranging over sets. Sets are a species of object. They should range over entities of a completely different type: properties and relations.
     From: Bob Hale (Necessary Beings [2013], 08.2)
     A reaction: This helpfully clarifies something which was confusing me. If sets are objects, then 'second-order' logic just seems to be the same as first-order logic (rather than being 'set theory in disguise'). I quantify over properties, but deny their existence!
5. Theory of Logic / C. Ontology of Logic / 4. Logic by Convention
Maybe conventionalism applies to meaning, but not to the truth of propositions expressed
     Full Idea: An old objection to conventionalism claims that it confuses sentences with propositions, confusing what makes sentences mean what they do with what makes them (as propositions) true.
     From: Bob Hale (Necessary Beings [2013], 05.2)
     A reaction: The conventions would presumably apply to the sentences, but not to the propositions. Since I think that focusing on propositions solves a lot of misunderstandings in modern philosophy, I like the sound of this.
5. Theory of Logic / F. Referring in Logic / 1. Naming / d. Singular terms
We should decide whether singular terms are genuine by their usage
     Full Idea: The criteria for a genuine singular term should pick out not the singular terms themselves but their uses, since they may be genuine in one context and not another.
     From: Bob Hale (Abstract Objects [1987], Ch.2.II)
     A reaction: [rephrased] This will certainly meet problems with vagueness (e.g. as the reference of a singular term is gradually clarified).
Often the same singular term does not ensure reliable inference
     Full Idea: In 'the whale is increasingly scarce' and 'the whale is much improved today' (our pet whale), we cannot infer that there is something that is much improved and increasingly scarce, so this singular term fails Dummett's criterion based on inference.
     From: Bob Hale (Abstract Objects [1987], Ch.2)
     A reaction: [much rephrased] This is not just a problem for a few cunningly selected examples. With contortions almost any singular term can be undermined in this way. Singular terms are simply not a useful guide to the existence of abstracta.
Plenty of clear examples have singular terms with no ontological commitment
     Full Idea: Some examples where a definite singular noun phrase is not 'genuine' (giving ontological commitment): 'left us in the lurch'; 'for my mother's sake'; 'given the sack'; 'in the nick of time', 'the whereabouts of the PM', 'the identity of the murderer'.
     From: Bob Hale (Abstract Objects [1987], Ch.2.II)
     A reaction: These are not just freakish examples. If I 'go on a journey', that doesn't involve extra entities called 'journeys', just because the meaning is clearer and a more commonplace part of the language.
If singular terms can't be language-neutral, then we face a relativity about their objects
     Full Idea: If we lack any general, language-neutral characterization of singular terms, must not a parallel linguistic relativity infect the objects which are to be thought of as their non-linguistic correlates?
     From: Bob Hale (Abstract Objects [1987], Ch.2.III)
     A reaction: Hale thinks he can answer this, but I would have thought that this problem dooms the linguistic approach from the start. There needs to be more imagination about how very different a language could be, while still qualifying as a language.
5. Theory of Logic / H. Proof Systems / 4. Natural Deduction
Unlike axiom proofs, natural deduction proofs needn't focus on logical truths and theorems
     Full Idea: In contrast with axiomatic systems, in natural deductions systems of logic neither the premises nor the conclusions of steps in a derivation need themselves be logical truths or theorems of logic.
     From: Bob Hale (Necessary Beings [2013], 09.2 n7)
     A reaction: Not sure I get that. It can't be that everything in an axiomatic proof has to be a logical truth. How would you prove anything about the world that way? I'm obviously missing something.
6. Mathematics / A. Nature of Mathematics / 3. Nature of Numbers / g. Real numbers
The real numbers may be introduced by abstraction as ratios of quantities
     Full Idea: The real numbers may be introduced by abstraction as ratios of quantities. ..They are not defined by Dedekind cuts; rather, the cuts constitute a domain with the properties that are a necessary precondition.
     From: report of Bob Hale (Reals by Abstraction [1998]) by B Hale / C Wright - Intro to 'The Reason's Proper Study' 3.3
     A reaction: This is Hale's neo-logicist attempt to derive the real numbers from Hume's Principle.
6. Mathematics / C. Sources of Mathematics / 6. Logicism / c. Neo-logicism
Add Hume's principle to logic, to get numbers; arithmetic truths rest on the nature of the numbers
     Full Idea: The existence of the natural numbers is not a matter of pure logic - it cannot be proved in pure logic. It can be proved in second-order logic plus Hume's principle. Truths of arithmetic are not logic - they depend on the nature of natural numbers.
     From: Bob Hale (Necessary Beings [2013], 07.4)
     A reaction: Hume's principles needs entities which can be matched to one another, so a certain ontology is needed to get neo-logicism off the ground.
7. Existence / C. Structure of Existence / 5. Supervenience / a. Nature of supervenience
Interesting supervenience must characterise the base quite differently from what supervenes on it
     Full Idea: Any intereresting supervenience thesis requires that the class of facts on which the allegedly supervening facts supervene be characterizable independently, without use or presupposition of the notions involved in stating the supervening facts.
     From: Bob Hale (Necessary Beings [2013], 03.4.1)
     A reaction: There might be intermediate cases here, since having descriptions which are utterly unconnected (at any level) might be rather challenging.
7. Existence / C. Structure of Existence / 7. Abstract/Concrete / a. Abstract/concrete
The abstract/concrete distinction is based on what is perceivable, causal and located
     Full Idea: The 'concrete/abstract' distinction has a strong intuitive feel, and can seem to be drawable by familiar contrasts, between what can/cannot be perceived, what can/cannot be involved in causal interactions, and is/is not located in space and time.
     From: Bob Hale (Abstract Objects [1987], Ch.3.I)
     A reaction: Problems arise, needless to say. The idea of an abstraction can be causal, and abstractions seem to change. If universals are abstract, we seem to perceive some of them. They can hardly be non-spatial if they have a temporal beginning and end.
Colours and points seem to be both concrete and abstract
     Full Idea: It might seem that colours would qualify both as concrete and as abstract objects. ...and geometrical points also seem to be borderline.
     From: Bob Hale (Abstract Objects [1987], Ch.3.II)
     A reaction: The theory of tropes exploits this uncertainty. Dummett (1973:ch.14) notes that we can point to colours, but also slip from an adjectival to a noun usage of colour-terms. He concludes that colours are concrete. I think I agree.
The abstract/concrete distinction is in the relations in the identity-criteria of object-names
     Full Idea: Noonan suggests that the distinction between abstract and concrete objects should be seen as derivative from a difference between the relations centrally involved in criteria of identity associated with names of objects.
     From: Bob Hale (Abstract Objects [1987], Ch.3.III)
     A reaction: [He cites Noonan 1976, but I've lost it] I don't understand this, but collect it as a lead to something that might be interesting. A careful reading of Hale might reveal what Noonan meant.
Token-letters and token-words are concrete objects, type-letters and type-words abstract
     Full Idea: In familiar, though doubtless not wholly problematic jargon, token-letters and token-words are concrete objects, type-letters and type-words abstract.
     From: Bob Hale (Abstract Objects [1987], Ch.3.III)
     A reaction: This is indeed problematic. The marks may be tokens, but the preliminary to identifying the type is to see that the marks are in fact words. To grasp the concrete, grasp the abstraction. An excellent example of the blurring of the distinction.
7. Existence / C. Structure of Existence / 7. Abstract/Concrete / b. Levels of abstraction
There is a hierarchy of abstraction, based on steps taken by equivalence relations
     Full Idea: The domain of the abstract can be seen as exemplifying a hierarchical structure, with differences of level reflecting the number of steps of abstraction, via appropriate equivalence relations, required for recognition at different levels.
     From: Bob Hale (Abstract Objects [1987], Ch.3.III)
     A reaction: I think this is right, and so does almost everyone else, since people cheerfully talk of 'somewhat' abstract and 'highly' abstract. Don't dream of a neat picture though. You might reach a level by two steps from one direction, and four from another.
7. Existence / D. Theories of Reality / 8. Facts / c. Facts and truths
There is no gap between a fact that p, and it is true that p; so we only have the truth-condtions for p
     Full Idea: There is no clear gap between its being a fact that p and its being true that p, no obvious way to individuate the fact a true statement records other than via that statement's truth-conditions.
     From: Bob Hale (Necessary Beings [2013], 03.2)
     A reaction: Typical of philosophers of language. The concept of a fact is of something mind-independent; the concept of a truth is of something mind-dependent. They can't therefore be the same thing (by the contrapositive of the indiscernability of identicals!).
8. Modes of Existence / D. Universals / 1. Universals
If F can't have location, there is no problem of things having F in different locations
     Full Idea: If Fs are incapable of spatial location, it is impossible for a and b to be at the same time in different places and yet be the same F.
     From: Bob Hale (Abstract Objects [1987], Ch.3.III)
     A reaction: A passing remark from Hale which strikes me as incredibly significant. The very idea of a 'one-over-many' is that there are many locations for the thing, so to conclude that the thing is therefore non-located seems to negate the original problem.
It is doubtful if one entity, a universal, can be picked out by both predicates and abstract nouns
     Full Idea: The traditional conception of universals, resting as it does upon the idea that some single type of entity is picked out by expressions of such radically different logical types as predicates and abstract nouns, is of doubtful coherence.
     From: Bob Hale (Abstract Objects [1987], Ch.3 Intro)
     A reaction: A striking case of linguistic metaphysics in action. I don't believe in universals, but I don't find this persuasive, as our capacity to express the same proposition by means of extremely varied syntax is obvious. Is 'horse' an abstract noun?
Realists take universals to be the referrents of both adjectives and of nouns
     Full Idea: On the traditional realist's view abstract qualities (universals) are the common referents of two quite different sorts of expression - of ordinary adjectives (predicates), and of abstract nouns referring to them.
     From: Bob Hale (Abstract Objects [1987], Ch.2.II)
     A reaction: This fact alone should make us suspicious, especially as there isn't an isomorphism between the nouns and the adjectives, and the match-up will vary between languages.
8. Modes of Existence / E. Nominalism / 1. Nominalism / c. Nominalism about abstracta
Objections to Frege: abstracta are unknowable, non-independent, unstatable, unindividuated
     Full Idea: Objections to Frege's argument for abstract objects: that the objects would not have the right sort of independence; that we could have no knowledge of them; that the singular term statements can't be had; that thoughts of abstracta can't be identified.
     From: Bob Hale (Abstract Objects [1987], Ch.1)
     A reaction: [compressed] [See Idea 10309 for the original argument] It is helpful to have this list, even if Hale rejects them all. They are also created but then indestructible, and exist in unlimited profusion, and seem relative to a language. Etc!
9. Objects / A. Existence of Objects / 2. Abstract Objects / a. Nature of abstracta
Shapes and directions are of something, but games and musical compositions are not
     Full Idea: While a shape or a direction is necessarily of something, games, musical compositions or dance routines are not of anything at all.
     From: Bob Hale (Abstract Objects [1987], Ch.3.II)
     A reaction: This seems important, because Frege's abstraction principle works nicely for abstractions 'of' some objects, but is not so clear for abstracta that are sui generis.
Many abstract objects, such as chess, seem non-spatial, but are not atemporal
     Full Idea: There are many plausible example of abstract objects which, though non-spatial, do not appear to satisfy the suggested requirement of atemporality, such as chess, or the English language.
     From: Bob Hale (Abstract Objects [1987], Ch.3.1)
     A reaction: Given the point that modern physics is committed to 'space-time', with no conceivable separation of them, this looks dubious. Though I think the physics could be challenged. Try Idea 7621, for example.
If the mental is non-spatial but temporal, then it must be classified as abstract
     Full Idea: If mental events are genuinely non-spatial, but not atemporal, its effect is to classify them as abstract; the distinction between the abstract and the mental simply collapses.
     From: Bob Hale (Abstract Objects [1987], Ch.3.1)
     A reaction: This is important. You can't discuss this sort of metaphysics in isolation from debates about the ontology of mind. Functionalists do treat mental events as abstractions.
Being abstract is based on a relation between things which are spatially separated
     Full Idea: The abstract/concrete distinction is, roughly, between those sortals whose grounding relations can hold between abstract things which are spatially but not temporally separated, those concrete things whose grounding relations cannot so hold.
     From: Bob Hale (Abstract Objects [1987], Ch.3.III)
     A reaction: Thus being a father is based on 'begat', which does not involve spatial separation, and so is concrete. The relation is one of equivalence.
9. Objects / A. Existence of Objects / 2. Abstract Objects / c. Modern abstracta
The modern Fregean use of the term 'object' is much broader than the ordinary usage
     Full Idea: The notion of an 'object' first introduced by Frege is much broader than that of most comparable ordinary uses of 'object', and is now fairly standard and familiar.
     From: Bob Hale (Abstract Objects [1987], Ch.1)
     A reaction: This makes it very difficult to get to grips with the metaphysical issues involved, since the ontological claims disappear into a mist of semantic vagueness.
9. Objects / A. Existence of Objects / 2. Abstract Objects / d. Problems with abstracta
We can't believe in a 'whereabouts' because we ask 'what kind of object is it?'
     Full Idea: Onotological outrage at such objects as the 'whereabouts of the Prime Minister' derives from the fact that we seem beggared for any convincing answer to the question 'What kind of objects are they?'
     From: Bob Hale (Abstract Objects [1987], Ch.2.II)
     A reaction: I go further and ask of any object 'what is it made of?' When I receive the answer that I am being silly, and that abstract objects are not 'made' of anything, I am tempted to become sarcastic, and say 'thank you - that makes it much clearer'.
9. Objects / C. Structure of Objects / 5. Composition of an Object
If a chair could be made of slightly different material, that could lead to big changes
     Full Idea: How shall we prevent a sorites taking us to the conclusion that a chair might have originated in a completely disjoint lot of wood, or even in some other material altogether?
     From: Bob Hale (Necessary Beings [2013], 11.3.7)
     A reaction: This seems a good criticism of Kripke's implausible claim that his lectern is necessarily (or essentially) made of the piece of wood it is made of. Could his lectern have had a small piece of plastic inserted in it?
9. Objects / F. Identity among Objects / 1. Concept of Identity
The relations featured in criteria of identity are always equivalence relations
     Full Idea: The relations which are featured in criteria of identity are always equivalence relations.
     From: Bob Hale (Abstract Objects [1987], Ch.3.III)
     A reaction: This will only apply to strict identity. If I say 'a is almost identical to b', this will obviously not be endlessly transitive (as when we get to k we may have lost the near-identity to a). Are 'two threes' identical to 'three twos'?
9. Objects / F. Identity among Objects / 3. Relative Identity
We sometimes apply identity without having a real criterion
     Full Idea: Not every (apparent) judgement of identity involves application of anything properly describable as a criterion of identity, ...such as being able to pronounce that mercy is the quality of being merciful.
     From: Bob Hale (Abstract Objects [1987], Ch.2.II)
     A reaction: This suggests some distinction between internal criteria (e.g. grammatical, conceptual) and external criteria (existent, sensed).
10. Modality / A. Necessity / 2. Nature of Necessity
Absolute necessity might be achievable either logically or metaphysically
     Full Idea: Maybe peaceful co-existence between absolute logical necessity and absolute metaphysical necessity can be secured, ..and absolute necessity is their union. ...However, a truth would then qualify as absolutely necessary in two quite different ways.
     From: Bob Hale (Absolute Necessities [1996], 4)
     A reaction: Hale is addressing a really big question for metaphysic (absolute necessity) which others avoid. In the end he votes for rejecting 'metaphysical' necessity. I am tempted to vote for rejecting logical necessity (as being relative). 'Absolute' is an ideal.
10. Modality / A. Necessity / 3. Types of Necessity
Maybe not-p is logically possible, but p is metaphysically necessary, so the latter is not absolute
     Full Idea: It might be metaphysically necessary that p but logically possible that not-p, so that metaphysical necessity is not, after all, absolute.
     From: Bob Hale (Absolute Necessities [1996]), quoted by E.J. Lowe - The Possibility of Metaphysics 1.5
     A reaction: Lowe presents this as dilemma, but it sounds fine to me. Flying pigs etc. have no apparent logical problems, but I can't conceive of a possible world where pigs like ours fly in a world like ours. Earthbound pigs may be metaphysically necessary.
A strong necessity entails a weaker one, but not conversely; possibilities go the other way
     Full Idea: One type of necessity may be said to be 'stronger' than another when the first always entails the second, but not conversely. This will obtain only if the possibility of the first is weaker than the possibility of the second.
     From: Bob Hale (Absolute Necessities [1996], 1)
     A reaction: Thus we would normally say that if something is logically necessary (a very strong claim) then it will have to be naturally necessary. If something is naturally possible, then clearly it will have to be logically possible. Sounds OK.
'Relative' necessity is just a logical consequence of some statements ('strong' if they are all true)
     Full Idea: Necessity is 'relative' if a claim of φ-necessary that p just claims that it is a logical consequence of some statements Φ that p. We have a 'strong' version if we add that the statements in Φ are all true, and a 'weak' version if not.
     From: Bob Hale (Absolute Necessities [1996], 1)
     A reaction: I'm not sure about 'logical' consequence here. It may be necessary that a thing be a certain way in order to qualify for some category (which would be 'relative'), but that seems like 'sortal' necessity rather than logical.
'Absolute necessity' is when there is no restriction on the things which necessitate p
     Full Idea: The strength of the claim that p is 'absolutely necessary' derives from the fact that in its expression as a universally quantified counterfactual ('everything will necessitate p'), the quantifier ranges over all propositions whatever.
     From: Bob Hale (Necessary Beings [2013], 04.1)
     A reaction: Other philosophers don't seem to use the term 'absolute necessity', but it seems a useful concept, in contrast to conditional or local necessities. You can't buy chocolate on the sun.
Logical and metaphysical necessities differ in their vocabulary, and their underlying entities
     Full Idea: The difference between logical and metaphysical necessities lies, not in the range of possibilities for which they hold, but - at the linguistic level - in the kind of vocabulary essential to their expression, and the kinds of entities that explain them.
     From: Bob Hale (Necessary Beings [2013], 04.5)
     A reaction: I don't think much of the idea that the difference is just linguistic, and I don't like the idea of 'entities' as grounding them. I see logical necessities as arising from natural deduction rules, and metaphysical ones coming from the nature of reality.
Absolute necessities are necessarily necessary
     Full Idea: I argue that any absolute necessity is necessarily necessary.
     From: Bob Hale (Necessary Beings [2013], 05.5.2)
     A reaction: This requires the principle of S4 modal logic, that necessity implies necessary necessity. He argues that S5 is the logical of absolute necessity.
10. Modality / A. Necessity / 5. Metaphysical Necessity
Metaphysical necessity says there is no possibility of falsehood
     Full Idea: Friends of metaphysical necessity would want to hold that when it is metaphysically necessary that p, there is no good sense of 'possible' (except, perhaps, an epistemic one) in which it is possible that not-p.
     From: Bob Hale (Absolute Necessities [1996], 2)
     A reaction: We might want to say which possible worlds this refers to (and presumably it won't just be in the actual world). The normal claim would refer to all possible worlds. Adding a '...provided that' clause moves it from absolute to relative necessity.
10. Modality / A. Necessity / 6. Logical Necessity
'Broadly' logical necessities are derived (in a structure) entirely from the concepts
     Full Idea: 'Broadly' logical necessities are propositions whose truth derives entirely from the concepts involved in them (together, of course, with relevant structure).
     From: Bob Hale (Absolute Necessities [1996], 3)
     A reaction: Is the 'logical' part of this necessity bestowed by the concepts, or by the 'structure' (which I take to be a logical structure)?
Logical necessities are true in virtue of the nature of all logical concepts
     Full Idea: The logical necessities can be taken to be the propositions which are true in virtue of the nature of all logical concepts.
     From: Bob Hale (Absolute Necessities [1996], p.10)
     A reaction: This is part of his story of essences giving rise to necessities. His proposal sounds narrow, but logical concepts may have the highest degree of generality which it is possible to have. It must be how the concepts connect that causes the necessities.
Logical necessity is something which is true, no matter what else is the case
     Full Idea: We can identify the belief that the proposition that p is logically necessary, where p may be of any logical form, with the belief that, no matter what else was the case, it would be true that p.
     From: Bob Hale (Necessary Beings [2013], 04.1)
     A reaction: I find this surprising. I take it that logical necessity must be the consequence of logic. That all squares have corners doesn't seem to be a matter of logic. But then he seems to expand logical necessity to include conceptual necessity. Why?
Maybe each type of logic has its own necessity, gradually becoming broader
     Full Idea: We can distinguish between narrower and broader kinds of logical necessity. There are, for example, the logical necessities of propostional logic, those of first-order logic, and so on. Maybe they are necessities expressed using logical vocabulary.
     From: Bob Hale (Necessary Beings [2013], 04.5)
     A reaction: Hale goes on to prefer a view that embraces conceptual necessities. I think in philosophy we should designate the necessities according to their sources. This might clarify a currently rather confused situation. First-order includes propositional logic.
10. Modality / C. Sources of Modality / 1. Sources of Necessity
Explanation of necessity must rest on something necessary or something contingent
     Full Idea: The dilemma is that to give the ultimate source of any necessity, we must either appeal to something which could not have been otherwise (i.e. is itself necessary), or advert to something which could have been otherwise (i.e. is itself merely contingent).
     From: Bob Hale (The Source of Necessity [2002], p.301)
     A reaction: [Hale is summarising Blackburn's view, and going on to disagree with it] Hale looks for a third way, but Blackburn seems to face us with quite a plausible dilemma.
Why is this necessary, and what is necessity in general; why is this necessary truth true, and why necessary?
     Full Idea: We must distinguish between explaining particular necessities and explaining necessity in general; and we ought to distinguish between explaining, in regard to any necessary truth, why it is true, and explaining why it is necessary.
     From: Bob Hale (The Source of Necessity [2002], p.308)
     A reaction: Useful. The pluralist view I associate with Fine says we can explain types of necessity, but not necessity in general. If we seek truthmakers, there is a special case of what adds the necessity to the truth.
The explanation of a necessity can be by a truth (which may only happen to be a necessary truth)
     Full Idea: My claim is that there are non-transitive explanations of necessities, where what explains is indeed necessary, but what explains the necessity of the explanandum is not the explanation's necessity, but its truth simpliciter.
     From: Bob Hale (The Source of Necessity [2002], p.311)
     A reaction: The big idea is to avoid a regress of necessities. The actual truths he proposes are essentialist. An interesting proposal. It might depend on how one views essences (as giving identity, or causal power)
It seems that we cannot show that modal facts depend on non-modal facts
     Full Idea: I think we may conclude that there is no significant version of modal supervenience which both commands acceptance and implies that all modal facts depend asymmetrically on non-modal ones.
     From: Bob Hale (Necessary Beings [2013], 03.4.3)
     A reaction: This is the conclusion of a sustained and careful discussion, recorded here for interest. I'm inclined to think that there are very few, if any, non-modal facts in the world, if those facts are accurately characterised.
10. Modality / C. Sources of Modality / 3. Necessity by Convention
If necessity rests on linguistic conventions, those are contingent, so there is no necessity
     Full Idea: If the alleged necessity, e,g, 2+2=4, really does depend upon a convention governing the use of the words in which we state it, and the existence of that convention is merely a contingent matter, then it can't after all be necessary.
     From: Bob Hale (The Source of Necessity [2002], p.302)
     A reaction: [Hale is citing Blackburn for this claim] Hale suggests replies, by keeping truth and meaning separate, and involving laws of logic. Blackburn clearly has a good point.
10. Modality / C. Sources of Modality / 4. Necessity from Concepts
Conceptual necessities are made true by all concepts
     Full Idea: Conceptual necessities can be taken to be propositions which are true in virtue of the nature of all concepts.
     From: Bob Hale (Absolute Necessities [1996], p.9)
     A reaction: Fine endorse essences for these concepts. Could we then come up with a new concept which contradicted all the others, and destroyed the necessity? Yes, presumably. Presumably witchcraft and astrology are full of 'conceptual necessities'.
Concept-identities explain how we know necessities, not why they are necessary
     Full Idea: It seems to me that identity-relations among concepts have more to do with explaining how we know that vixens are female foxes etc., than with explaining why it is necessary, and, more generally, with explaining why some necessities are knowable a priori.
     From: Bob Hale (The Source of Necessity [2002], P.313)
     A reaction: Hale rejects the conceptual and conventional accounts of necessity, in favour of the essentialist view. This strikes me as a good suggestion of Hale's, since I agree with him about the essentialism.
10. Modality / C. Sources of Modality / 6. Necessity from Essence
The big challenge for essentialist views of modality is things having necessary existence
     Full Idea: Whether the essentialist theory can account for all absolute necessities depends in part on whether the theory can explain the necessities of existence (of certain objects, properties and entities).
     From: Bob Hale (Necessary Beings [2013], Intro)
     A reaction: Hale has a Fregean commitment to all sorts of abstract objects, and then finds difficulty in explaining them from his essentialist viewpoint. His book didn't convince me. I'm more of a nominalist, me, so I sleep better at nights.
Essentialism doesn't explain necessity reductively; it explains all necessities in terms of a few basic natures
     Full Idea: The point of the essentialist theory is not to provide a reductive explanation of necessities. It is, rather, to locate a base class of necessities - those which directly reflect the natures of things - in terms of which the remainder may be explained.
     From: Bob Hale (Necessary Beings [2013], 06.6)
     A reaction: My picture is of most of the necessities being directly explained by the natures of things, rather than a small core of natures generating all the derived ones. All the necessities of squares derive from the nature of the square.
If necessity derives from essences, how do we explain the necessary existence of essences?
     Full Idea: If the essentialist theory of necessity is to be adequate, it must be able to explain how the existence of certain objects - such as the natural numbers - can itself be absolutely necessary.
     From: Bob Hale (Necessary Beings [2013], 07.1)
     A reaction: Hale and his neo-logicist pals think that numbers are 'objects', and they necessarily exist, so he obviously has a problem. I don't see any alternative for essentialists to treating the existing (and possible) natures as brute facts.
10. Modality / E. Possible worlds / 1. Possible Worlds / a. Possible worlds
What are these worlds, that being true in all of them makes something necessary?
     Full Idea: We need an explanation of what worlds are that makes clear why being true at all of them should be necessary and sufficient for being necessary (and true at one of them suffices for being possible).
     From: Bob Hale (Necessary Beings [2013], 03.3.2)
     A reaction: Hale is introducing combinatorial accounts of worlds, as one possible answer to this. Hale observes that all the worlds might be identical to our world. It is always assumed that the worlds are hugely varied. But maybe worlds are constrained.
10. Modality / E. Possible worlds / 1. Possible Worlds / e. Against possible worlds
Possible worlds make every proposition true or false, which endorses classical logic
     Full Idea: The standard conception of worlds incorporates the assumption of bivalence - every proposition is either true or false. But it is infelicitous to build into one's basic semantic machinery a principle endorsing classical logic against its rivals.
     From: Bob Hale (Necessary Beings [2013], 10.3)
     A reaction: No wonder Dummett (with his intuitionist logic) immediately spurned possible worlds. This objection must be central to many recent thinkers who have begun to doubt possible worlds. I heard Kit Fine say 'always kick possible worlds where you can'.
18. Thought / C. Content / 6. Broad Content
The molecules may explain the water, but they are not what 'water' means
     Full Idea: What it is to be (pure) water is to be explained in terms of being composed of H2O molecules, but this is not what the word 'water' means.
     From: Bob Hale (Necessary Beings [2013], 11.2)
     A reaction: Hale says when the real and verbal definitions match, we can know the essence a priori. If they come apart, presumably we need a posteriori research. Interesting. It is certainly dubious to say a stuff-word means its chemical composition.