International Society for History, Philosophy, and Social Studies of Biology

twitter 2015
     facebook 2015


FRIDAY, JULY 10  /  09:00 - 10:30  /  DS-R520
Individual papers
Explanation, experiment, and mechanism

Is Brownian motion a mechanism?

Archie Fields (Indiana University, Bloomington, United States)

The mechanistic account of explanation has received a great deal of attention over the past decade, especially in the domain of biology. By many accounts, mechanistic explanation is the dominant mode of explanation in the biological sciences. I argue that there are a large number of biological phenomena that cannot be explained fully in terms of mechanisms. I begin by briefly describing a general mechanistic account of explanation and highlighting some of the required features of mechanisms discussed in the literature, particularly regularity of operation and having fine-tuned organization. I then show that the motion of motor proteins, which are enzymes that metabolize molecular energy sources like ATP to do mechanical work, cannot be completely explained mechanistically. These proteins often rely upon Brownian motion to move them down microtubule tracks within cells in a way that fails to adhere fully to both the regularity and fine-tuned organization that is required for mechanistic explanation. Generalizing from the motor protein case, I argue that Brownian motion cannot be explained mechanistically and that because Brownian motion is causally efficacious in many biological phenomena, those phenomena also cannot be fully explained mechanistically. This demonstrates that Craver’s principal normative requirement of mechanistic explanations, which is that the mechanistic explanation must account fully for the explanandum phenomenon, cannot be fulfilled for explanations of many biological phenomena. Thus, the scope and explanatory power of the mechanistic strategy is drastically reduced since many biological phenomena involving Brownian motion cannot be given complete mechanistic explanations.

"How possibly" explanation in biology: Lessons from Wilhelm His’ "simple experiments" models

Christopher Pearson (Southern Illinois University Edwardsville, United States)

One common characterization of how possibly explanation in biology is that they are incomplete, non-autonomous explanations. For some, this lack of completeness and autonomy does not, however, undermine fully the value of how possibly explanation to the biological sciences. Despite their presenting as incomplete and non-autonomous, how possibly explanations occupy an important heuristic role in biology by, for example, setting out novel research programs. The case favoring an important role for how possibly explanation in biology I think is exceptionally strong however, limiting that role to heuristic, I believe misdiagnoses to some degree the ways in which how possibly explanation functions within biology. Indeed, characterizing the role of how possibly explanation as one of mere heuristic is crucially contingent on the view that how possibly explanations are incomplete and non-autonomous. I maintain, to contrary, that there are how-possibly explanations in biology that are, in a substantive way, both complete and autonomous. My defense of this claim stems from an analysis of Wilhelm His simple experiments , in which His attempted to construct manipulable physical models of development. I argue that His' simple experiments invite 2 distinct interpretations regarding the success of mechanical explanation of development. These 2 interpretations depend on adopting one or the other of 2 contrastive explanatory contexts. The first of these how possibly can mechanical principles explain an embryo s transition from one stage to the next is set against the context of non-mechanical factors being required to explain animal development. The second how possibly can mechanical principles explain an embryo s transition from one state to the next set against the context of seeking an inventory of factors sufficient to account for actual embryological development. The latter of these contrastive explanatory contexts fully supports the view that how possibly explanations are incomplete and non-autonomous, though eminently useful as heuristic. Alternatively, the former contrastive explanatory context demonstrates His’s model derived explanations as both complete and autonomous.

Omics experimental strategy (OES) as a scientific epistemology for system-driven research

Eve Roberts (Dalhousie University, Canada)

Omics disciplines (genomics, proteomics, metabolomics, exomics, and the like) challenge the standard view that all biological/biomedical research must be hypothesis-driven. Previously I have argued that omics research is ‘system-driven research’, in which the complexity of a biological system is addressed directly. Furthermore, although system-driven biological/biomedical research operates within a hierarchy of hypotheses, what actually shapes system-driven experimental design is the system itself, not a proximate hypothesis. Evidently omics methodologies produce new scientific knowledge: the question is how?—given that such research is not hypothesis-driven. Taking proteomics as exemplar, I identify the strength of system-driven research as its ability to generate detailed, inclusive accounts of natural biological systems. I call its scientific epistemology the Omics Experimental Strategy. OES involves a well-designed experiment, specifically with a detailed description of the system under investigation, empirical data produced competently by reliable technology according to peer-generated standards of best practice, and finally contextualization of data within the system. For system-driven research, validation takes place within the system, which is driving the experiment. (Importantly, a similar validating relationship holds for hypothesis-driven research, where validation takes place in relation to the proximate hypothesis driving the experiment.) These contextualized findings provide new insight into how that system functions. Indeed some of these findings may prove to be entirely novel and unexpected. Arguably, for the OES, novelty serves as an epistemic virtue. Supported by a scientific epistemology, system-driven research carried out according to the OES constitutes scientific research, not a run-up to ‘real’ science.