Explaining the Explanations of 100 Million Missing Women - Mechanism and Causality in Biology and Economics - page 223

we notice methodological similarities and dissimilarities among economists and
biomedical scientists. First, we find that Sen's and Blumberg's theorization pro-
cesses may best be understood in terms of Inference to the Best Explanation
formulated by, inter alia, Peter Lipton. Lipton's Inference to the Best Explanation
has a root of Charles Sanders Peirce's abduction , which consists the form of
inference: “The surprising fact, C, is observed; but if A were true, C would be a
matter of course, hence, there is reason to suspect that A is true.” Inference to the
Best Explanation stresses also on inductive inference. Lipton maintains that
scientists infer from available evidence to the hypothesis that would provide the
best explanation for that evidence. For instance, as Lipton states, Darwin inferred
from his biological evidence the theory of natural selection because natural selec-
tion would best explain that evidence (Lipton 2000 , p. 184). Conversely, inference
is guided by explanatory considerations, particularly by those explanations that
provide most understanding if true—the loveliest explanations. 4 This seemingly
circular argument in fact describes well how a hypothesis is formulated in science,
how it is justified as the best hypothesis among many competing hypotheses, and
how it can heuristically guide the inference process. It is also stressed that Inference
to the Best Explanation is fallible. Because all available evidence does not neces-
sarily lead to truth, the best explanation may not be the actual explanation. Lipton's
reconstruction of Carl Hempel's ( 1966 ) discussion of Ignaz Semmelweis's attempt
to explain different rates of childbed fever in two hospital wards well demonstrates
how the best explanation to such observation can be derived from the account of
Inference to the Best Explanation (Lipton 2005 ).
It seems both of Sen's and Blumberg's studies fit well the account of Inference to
the Best Explanation. In our case study, Sen and Blumberg seek to infer from
evidence the best explanations for Asia's missing women. Sen derives missing-
women phenomenon from the sex ratios in Asia from those in the Western
countries, and then he infers from the evidence that the son preference induced
by women's gainful employment is the best explanation for Asia's missing women.
Similarly, Blumberg infers from his data that the hepatitis B hypothesis is the best
biological explanation for the missing women, in which he uses his knowledge of
the hepatitis virus, especially the feature of gender sensitivity. Despite the fact that
both Sen's cultural hypothesis and Blumberg's hepatitis B hypothesis provide the
best explanation for their evidence, we observe that the consideration of explaining
the missing-women phenomenon has been the guiding force to direct them to adopt
a particular inference strategy for carefully developing their accounts by explaining
and explaining away the adequacy of explanatory components. So doing makes
their explanation better than others. Yet it would still be understandable should
Sen's and Blumberg's hypotheses have turned out to be false explanations in the
face of newly acquired data, and have been replaced by a theory providing better
account for the data.
4 By contrast, a likeliest explanation is the explanation that is best warranted by the evidence.