Figure 1.5 Scientific progress through abduction.
abductive hypotheses to generate (predict) new observations. Each new abductive hypothesis in the relational triangle (Figure 1.4) becomes a part of background (background ideas-theory) and thus creates a new enlarged background (see Figure 1.5). The new background generates novel observations, but these too—because they constitute a background-observation relation—yield opposite identities that require further abductive inferences.
Empirical support for an abductive explanation is the outcome of any assessment of scope. Here, another central feature of a relational metamethod needs to be differentiated from the traditional modern split metamethod. Under the rule of split-off induction and deduction, it was assumed that scientific progress moved forward through the deductive falsification of theories (Popper, 1959). The criterion of falsification, however, fell into disrepute through demonstrations by several historians and philosophers of science (e.g., Hanson, 1958; T. S. Kuhn, 1962; Lakatos, 1978; Laudan, 1977; Putnam, 1983; Quine, 1953) that although deductive logic, and hence falsification, is applicable to a specific experimental hypothesis, falsification does not reach to the level of rich theories (i.e., background is abductive in character, not inductive nor deductive). Within a relational metatheory, these demonstrations lead to the principle that falsified experimental hypotheses are important in that they constitute failures of empirical support for the broader abductive explanation, but they are not important in the sense of constituting a refutation of the explanation. T. S. Kuhn, Lakatos, and Luadan describe these failures as anomalous instances for the background, and as such they require evaluation; but they do not in and of themselves require abandonment of the abductive explanation (see Overton, 1984, 1994a).
To this point a relational metatheory and an integrative metamethod have been described, and the manner in which these ground, constrain, and sustain various developmentally relevant issues, theories, and methods has been illustrated. The next section of this paper presents a broad illustration of the application of relational metatheory to developmental inquiry.
Was this article helpful?