Experimental Discovery, Data Models, and Mechanisms in Biology: An Example from Mendel’s Work - Mechanism and Causality in Biology and Economics - page 115

visible or transparent. Appropriating Darden's metaphor, to discover is to cause a
black or gray box to become a glass (transparent) box. If a scientist's work cannot
help herself or others look inside a black or gray box, she has not discovered
anything. Driesch turned a black box about the development of embryos into a
gray box, but he re-covered it with another facade: the vital force. This prevented
people from seeing inside the gray box. So Driesch did not really discover the
phenomenon of conditional specification. Admittedly, Driesch did discover some-
thing novel, but what he discovered was irrelevant to conditional specification, a
phenomenon that requires a mechanistic explanation. Driesch identified it as a
phenomenon of embryonic development directed by a vital force, so Driesch is
the discoverer of that anomalous phenomenon rather than of conditional specifica-
tion. Of course, Driesch's experiment still made a contribution to the discovery of
conditional specification, by providing an anomaly to be solved.
Kuhnians hold that a scientific discovery needs paradigmatic theories to offer
conceptual recognition. Thus, neither Mendel nor Griffith nor Driesch is a discov-
erer. None of them identified or recognized his findings in accordance with a new
paradigm, because none proposed such a thing. The Kuhnian holistic view is
problematic because it depends on grand paradigms, complete theories, or theoreti-
cal principles in the hierarchical framework in Fig. 6.4 .
The orthodox view presupposes that a discoverer is the first person who sees a
novel phenomenon by observation or experimentation, without conceptual cogni-
tion. This explanation produces inconsistent results in certain cases—for instance,
Driesch's and Mendel's. Driesch should be the discoverer of conditional specifica-
tion, for he first produced and observed “that phenomenon.” Mendel should not be
the discoverer of the Mendelian pattern of trait transmission, because other
hybridists had observed it before Mendel did. The orthodox view is problematic
because it identifies a discovery only on the basis of pure experience, neglecting the
role of conceptual recognition.
My proposal takes a middle way. My judgment of whether Mendel, Driesch, and
Griffith, respectively, made scientific discoveries is in agreement with some historians'
general view (e.g., Mayr's and Magner's), but I made my judgment for quite different
reasons. 14 I agree that a discovery requires a conceptual recognition, but the recognition
may not be theoretical; a data model can play the role. A scientist can use the model to
recognize that a set of data represents a significant phenomenon—a repeatable pattern
or regularity rather than an illusion, trivial appearances, or “mystery experience.”
However, data models by themselves are not sufficient to complete a discovery.
A discoverer should point out a direction that can lead to an advanced discovery. His
data model should indicate possibilities for further research, that is, the envisaging of
mechanisms. In the sense that ED3 can provide the most coherent explanation for
14 By contrast, Mayr and Magner did not explain why they regarded Griffith as a discoverer but
Driesch as not, given the similar structure of Griffith's and Driesch's experiments.