Do laws reveal necessities in nature?
In this essay, I will argue that how one answers the question of whether laws reveal necessities in nature is contingent upon the importance one places on 'intuition', to the debate. I will also argue that if one does not consider intuitions to have any bearing on the debate, then the Ramsey-Lewis account seems to be the best Humean account available. However, I will conclude with a modern non-Humean account which seems to better fit our intuitive response to the topic. The problem of whether there are laws of nature is a metaphysical, rather than epistemological, hypothesis; not to do with knowledge of truths but rather the nature of reality. We must distinguish laws of nature from accidental generalizations. When Hume proposed his account of causation; that we never see any necessitating link between cause and effect, he argued that there is, therefore, no a priori knowledge of the type that these kinds of links would provide. Hume simply argues that 'cause' is only associated with 'effect' through the constant conjunction of the two. The problem of induction, then, is that our experiences, of a finite number of cause-effect sequences, is insufficient to provide us with knowledge that the two will be similarly conjoined in the future. So this account of causation seems to make it impossible to distinguish between genuine laws of nature and accidental generalizations. Hume says that all there is to a causal law is something like 'whenever x, then y'. But not every truth of this form expresses a law. For instance, suppose whenever I do some gardening it rains. This statement is true, and suppose it will continue to be so, because I'm not going to do any more gardening ever again. But it clearly isn't a causal law. Clearly my doing something other than gardening doesn't preclude the skies' possibility for downpour. This is the problem of distinguishing laws from accidents, therefore. We have two approaches available to us, which Papineau calls 'Humean' and 'Non-Humean'. By the former, we would argue that causal laws never claim to state connections, but rather constant conjunctions, and attempts to demarcate some conjunctions from others. The latter, constrastingly, argues against Hume, claiming that the difference between laws and accidents is that the former do state necessary connections. Before evaluating these two possibilities, it will be useful to investigate what we are going to require of a successful demarcation between natural laws and accidental generalisations. I will attempt to outline this now. It is often said that laws support counterfactual conditionals, whereas accidents don't. For instance, we intuitively accept as true the counterfactual 'if the temperature had been above 100 degrees Centigrade, the water would have boiled', in virtue of the law that water boils at above 100 degrees. However, intuitively we do not accept it as true, returning to our earlier example, for me to say 'if I had done the gardening, it would have rained'. This is how laws support counterfactuals. However, this does not explain our difference. There are difficulties with the meaning of counterfactuals. The purpose of a counterfactual is to state what occurs in nonactual situations. How could we ever ascertain the truth of one, therefore? Surely,
Buy the full version of these notes or essay plans and more in our Metaphysics Notes.
