On the laws of nature

Abstract

All so-called 'laws of nature' are scientific hypotheses. There exist, however, two drastically distinct major classes of scientific hypo theses some of whose members have been referred to as 'laws of nature'. It is the differences and relations between these two classes of hypotheses that I shall try to examine in this paper. Possibly because the word 'law' may convey some kind of necessity, contem porary scientists, who have extensively realized the fallibility or tenous nature or severe limitations of all of their hypotheses, have shrunk back from calling more recently propounded scientific hypo theses 'laws of nature'. Nevertheless, the title 'law of nature' has been retained for certain hypotheses which have been referred to traditionally as 'laws', such as Ohm's law, or Newton's laws of motion. Thus, while retaining the word 'law' for certain hypotheses of science, as a matter of conventional talk, it would generally be regarded as quite old-fashioned to refer to any of the recently established hypotheses of science as 'laws of nature'. In order to classify 'laws of nature' it is indispensable to distinguish two major classes of hypotheses to one or the other of which most or possibly all (non-ad hoc) hypotheses of science belong. CLASS 1 hypotheses comprise well-formed as well as idealized empirical generalizations where the meaning of 'well-formed' will be explained in the sequel. Those idealized hypotheses of CLASS 1 which are referred to as 'laws of nature' will be called 'CLASS 1 laws of nature.' The other class of scientific hypotheses, which will be refer red to as CLASS 2 hypotheses, consists of the starting hypotheses of hypothetico-deductive theories, such as Newtonian mechanics, elec tromagnetic theory, and quantum mechanics. I shall call those hypo theses of CLASS 2 which are referred to as 'laws of nature' 'CLASS 2 laws of nature.' By discussing the ontological status of hypotheses of CLASS 1 and of CLASS 2, and the interrelations of the two classes of hypotheses, I shall automatically include as special cases the ontological status of CLASS 1 and CLASS 2 laws of nature.