structure for 'Natural Theory'    |     alphabetical list of themes    |     unexpand these ideas

26. Natural Theory / D. Laws of Nature / 5. Laws from Universals

[laws seen as necessary relations between universals]

13 ideas
In causal laws, 'events' must recur, so they have to be universals, not particulars [Russell]
     Full Idea: An 'event' (in a statement of the 'law of causation') is intended to be something that is likely to recur, since otherwise the law becomes trivial. It follows that an 'event' is not some particular, but a universal of which there may be many instances.
     From: Bertrand Russell (On the Notion of Cause [1912], p.179)
     A reaction: I am very struck by this. It may be a key insight into understanding what a law of nature actually is. It doesn't follow that we must be realists about universals, but the process of abstraction from particulars is at the heart of generalisation.
Rather than take necessitation between universals as primitive, just make laws primitive [Maudlin on Armstrong]
     Full Idea: My own view is simple: the laws of nature ought to be accepted as ontologically primitive. …They are preferable in point of familiarity to such necessitation relations between universals.
     From: comment on David M. Armstrong (What is a Law of Nature? [1983]) by Tim Maudlin - The Metaphysics within Physics 1.4
     A reaction: I think you make natures of things primitive, and reduce laws to regularities and universals to resemblances. Job done. Natures are even more 'familiar' as primitives than laws are.
Armstrong has an unclear notion of contingent necessitation, which can't necessitate anything [Bird on Armstrong]
     Full Idea: The two criticisms levelled against Armstrong are that it is unclear what his relation of contingent necessitation is, and that it is unclear how it is able to necessitate anything.
     From: comment on David M. Armstrong (What is a Law of Nature? [1983]) by Alexander Bird - Nature's Metaphysics 3.1.2
     A reaction: I suppose someone has to explore the middle ground between the mere contingencies of Humean regularities and the strong necessities of scientific essentialism. The area doesn't, however, look promising.
The laws of nature link properties with properties [Armstrong]
     Full Idea: There is an utterly natural idea that the laws of nature link properties with properties.
     From: David M. Armstrong (What is a Law of Nature? [1983], 06.3)
     A reaction: Put it this way: given that properties are expressions of invariant powers, the interaction of two properties will (ceteris paribus) be invariant, and laws are just invariances in natural behaviour.
Maybe laws of nature are just relations between properties? [Harré]
     Full Idea: The idea of the Dretske-Armstrong-Tooley view is very simple: the laws of nature relate properties to properties.
     From: Rom Harré (Laws of Nature [1993], 2)
     A reaction: Presumably the relations are necessary ones. I don't see why we need to mention these wretched 'universals' in order to expound this theory. It sounds much more plausible if you just say a property is defined by the way it relates to other properties.
We might say laws are necessary by combining causal properties with Armstrong-Dretske-Tooley laws [Shoemaker]
     Full Idea: One way to get the conclusion that laws are necessary is to combine my view of properties with the view of Armstrong, Dretske and Tooley, that laws are, or assert, relations between properties.
     From: Sydney Shoemaker (Causal and Metaphysical Necessity [1998], I)
     A reaction: This is interesting, because Armstrong in particular wants the necessity to arise from relations between properties as universals, but if we define properties causally, and make them necessary, we might get the same result without universals.
Originally Humeans based lawlike statements on pure qualities, without particulars [Harré/Madden]
     Full Idea: The original Humean suggestion was that lawlike statements must contain only purely qualitative predicates - that is, predicates which do not require in a statement of their meaning a reference to any particular object or spatio-temporal location.
     From: Harré,R./Madden,E.H. (Causal Powers [1975], 2.II)
     A reaction: Harré and Madden are keen to promote particulars (with powers) as the foundation of scientific theory, and I agree with them. It strikes me as quite elementary that generalisations arise from particulars, so can't fundamentally explain them.
Individuals enter into laws only through their general qualities and relations [McMichael]
     Full Idea: Individuals appear to enter into laws only through their general qualities and relations.
     From: Alan McMichael (The Epistemology of Essentialist Claims [1986], VIII)
     A reaction: This is a very significant chicken-or-egg issue. The remark seems to offer the vision of pre-existing general laws, which individuals then join (like joining a club). But surely the laws are derived from the individuals? Where else could they come from?
Laws of nature are necessary relations between universal properties, rather than about particulars [Mumford]
     Full Idea: The core of the Dretske-Tooley-Armstrong view of the late 70s is that we have a law of nature when we have a relation of natural necessitation between universals. ..The innovation was that laws are about properties, and only indirectly about particulars.
     From: Stephen Mumford (Laws in Nature [2004], 06.2)
     A reaction: It sounds as if we should then be able to know the laws of nature a priori, since that was Russell's 1912 definition of a priori knowledge.
If laws can be uninstantiated, this favours the view of them as connecting universals [Mumford]
     Full Idea: If there are laws that are instantiated in no particulars, then this would seem to favour the theory that laws connect universals rather than particulars.
     From: Stephen Mumford (Laws in Nature [2004], 06.4)
     A reaction: There is a dispute here between the Platonic view of uninstantiated universals (Tooley) and the Aristotelian instantiated view (Armstrong). Mumford and I prefer the dispositional account.
Laws cannot offer unified explanations if they don't involve universals [Bird]
     Full Idea: Laws, or what flow from them, are supposed to provide a unified explanation of the behaviours of particulars. Without universals the explanation of the behaviours of things lacks the required unity.
     From: Alexander Bird (Nature's Metaphysics [2007], 2.1.2)
     A reaction: Sounds a bit question-begging? Gravity seems fairly unified, whereas the frequency of London buses doesn't. Maybe I could unify bus-behaviour by positing a few new universals? The unity should first be in the phenomena, not in the explanation.
If the universals for laws must be instantiated, a vanishing particular could destroy a law [Bird]
     Full Idea: If universals exist only where and when they are instantiated, this make serious trouble for the universals view of laws. It would be most odd if a particular, merely by changing its properties, could cause a law to go out of existence.
     From: Alexander Bird (Nature's Metaphysics [2007], 3.2.2)
     A reaction: This sounds conclusive. He notes that this is probably why Armstrong does not adopt this view (though Lowe seems to favour it). Could there be a possible property (and concomitant law) which was never ever instantiated?
The view that laws are grounded in substance plus external necessity doesn't suit dispositionalism [Vetter]
     Full Idea: The Armstrong/Tooley/Dretske view, which takes laws to be substantial but grounded in a relation of nomic necessitation external to the properties themselves, is not an attractive option for the dispositionalist.
     From: Barbara Vetter (Potentiality [2015], 7.8)
     A reaction: The point is that the dispositionalist sees laws as grounded in the properties. I prefer her other option, of dispositionalism plus a 'shallow' view of laws (which she attributes to Mumford). The laws are as Lewis says, but powers explain them.