For the class in Philosophy of Science I took last semester, we had to write at least 7 revised essays on various topics pertaining to the philosophy of science. These essays incorporate class discussion and points from Chalmers' "What is This Thing Called Science?"
“Which of the two definitions of a scientific law that Chalmers discusses is more convincing for you? Argue for you position.”
Chalmers provides two basic possible definitions of a scientific law as follows:
Laws as regularities: If A, then B.
Laws as characterizations of powers or dispositions: “the identification and characterization of an object’s dispositions, tendencies, powers or capacities, both actual and potential.”
More specifically, laws as regularities refers to the idea that “events of type A are regularly followed, or accompanied, by events of type B provided disturbing factors are not present;” hence, “if A, then B” does not necessarily imply that A causes B, but rather that a correlation exists between the actualization of events A and B.
Moreover, in class we generalized the “objects” referred to in the definition of laws as characterizations as “entities or material systems.” Chalmers outlines several arguments against “laws as regularities.” The best objection, I think, is that the qualification of the definition – “…providing disturbing factors are not present” – reduces the applicability of a law to experimental situations, situations in which appropriate conditions are met. Essentially, the problem is that the sword cuts both ways: if one wishes to define a law in such a way that it’s dependent upon experimental conditions, then science becomes a series of tautologies. Instead of describing the nature of reality, we would actually be imposing artificial situations onto reality and passing them off as accurate descriptions of reality, a problem similar to the previously discussed “problem of models” (i.e. an ideal representation of reality rather than an actual one). On the other hand, if one wishes to assume laws are applicable outside of experimental situations as well as inside experimental situations, then laws cannot be identified with the regularities that are achievable in experimental situations – this definition of a scientific law would not stand.
Chalmers is optimistic with regards to the second definition of a scientific law. In class, we noted that if we extend Chalmers’ definition as above (i.e. material systems, as well as entities, are objects which can possess dispositions et. al.), there is a possibility that problems Chalmers poses – such as the problem of the applicability of laws as characterization to the laws of thermodynamics – would be solved. The potentiality of objects is also an aspect of the definition Chalmers fails to utilize in favor of laws as characteristics. However, a seemingly insoluble problem was that it was not explained how one comes to characterize an object’s characteristics without presupposing laws as regularities.
Hence, given that both definitions of a scientific law are dependent upon the assumption that the axiom of empiricism is sound (sensation is reliable), that an assumed solution to the problem of unknown variables exists &c., that laws as regularities seems to be the more historically accurate description of scientific laws, and that the latter law presupposes the former leads me to believe that the former (laws as regularities) is more convincing.