Changes

Just as is the case for [[The First Law for Theories (Barseghyan-2015)]], this law does not impose limitations on the sort of methods that can replace employed methods.[[CiteRef::Barseghyan (2015)|p.125]] However, importantly, "the community never remains with no expectations whatsoever. When facing a new theory, the community always has some implicit expectations concerning such theories. These expectations may be very specific or they may be very abstract and vague, but some expectations are always present, for otherwise no theory assessment would be possible."[[CiteRef::Barseghyan(2015)|p.126]]

|Prehistory=This idea is not new to philosophers of science. For example, it is expressed in [[Kuhn (1970a)]]: "scientists behave in the following ways; those modes of behaviour have… the following essential functions; in the absence of an alternative mode that would serve similar functions, scientists should behave essentially as they do if their concern is to improve scientific knowledge."[[CiteRef::Kuhn (1970a)|p.237]] This is similar to our formulation of [[The First Law for Methods (Barseghyan-2015)]], disregarding the normative piece ("scientists should behave").

<blockquote>In the most basic case, a community can reject some of the more specific requirements of its currently employed method and revert to a more abstract method. Alternatively, it can replace those rejected requirements with some new specific requirements. Suppose the employed method stipulates that a new theory must be tested in repeatable experiments and observations. In principle, the community may one day remove some of the ingredients of this method, say, the requirement of repeatability. As a result, the community can either revert to a more abstract method or it can introduce a new requirement to replace the repeatability clause. For instance, the community may revert to the more abstract method which stipulates a new theory must be tested in experiments and observations (no repeatability requirement). Alternatively, it can introduce a new requirement that in addition to empirical testing a new theory must also explain all the facts explained by the accepted theory. Which of these two scenarios materialize at each particular instance is decided by a number of contingent factors.[[CiteRef::Barseghyan (2015)|p.125]]</blockquote>