Historians on the role of mechanistic thinking in inter war physicochemical biology

Historians of science have shown an interest in physico-chemical biology and in the institutional policies that governed its implementation. Some have interpreted biologists’ interest in collaborating with physicists and chemists as an attempt to establish biology as a proper science. For example, Garland Allen (2005, pp. 280-81) has described the mechanistic conception of biological phenomena as a strategy used by young biologists in the first decades of the twentieth century to distance themselves from old-fashioned natural history and its non-testable hypotheses. Others have emphasised the role of philanthropic institutions, particularly the Rockefeller Foundation and the director of its natural sciences division, Warren Weaver.³⁰ According to Robert Kohler, physicists and chemists were driven to address biological problems for financial reasons: researchers from the physical sciences collaborated with biologists to acquire funding for expensive instruments which otherwise they would not have been able to afford. In the resulting research collaborations, Kohler (1991, pp. 304-5) wrote, ‘physical scientists provided technical services, and biochemists or biologists the problems’. More recently, Michel Morange (2007, p. 108) has pointed to the larger political dimension of the philanthropic support of the biological sciences, referring to a widely shared notion in the 1930s that biology should provide solutions for social problems.

Some scholars have also made references to certain research strategies and scientists' methodological norms. Evelyn Fox Keller, for example, has concluded that the introduction of physics to the field of biology in the 1930s led to a transformation of language, focus and methodology. According to Fox Keller (1990, pp. 406-7), this transformation resulted in an entire new set of beliefs - among them the belief in ‘a particular kind of (linear, causal) mechanism’. Mechanisms also play a role in Daniel Kevles and Gerald Gei- son's (1995, p. 101) account of the experimental life sciences. In particular, they have ascribed the privileged status of fields such as biochemistry and neurophysiology in the 1930s to these disciplines’ ability to demonstrate the ‘fertility of the mechanistic approach to biological problems - specifically, by showing the extent to which extremely complex events ... could be explained by or “reduced to” electrical-chemical and other basic physical concepts’. Soraya de Chadarevian and Harrnke Kamminga (1998, p. 5) have emphasised that researchers who ‘reduced bodily functions to the interplay of molecules’ made efforts to develop technologies ‘to measure and monitor molecules in the body’. However, usually historians did not specify in detail what the belief in causal mechanism implied or why exactly it was believed that in fields like biochemistry biological phenomena were, in fact, successfully reduced to physico-chemical concepts. Some philosophers of science, by contrast, have endeavoured to clarify these issues.

Philosophers on mechanisms and scientific practice

More than forty years ago the philosopher William C. Wimsatt (1976, p. 671) claimed that most biologists ‘see their work as explaining types of phenomena by discovering mechanisms', and that this is referred to as reduction. William Bechtel and Andrew Hamilton (2007) have, more recently, specified the way in which mechanistic explanations are reductive. They characterise mechanisms as entities and activities organised in such a way that they are responsible for the phenomenon to be explained.³¹ Now, putting forward mechanistic explanations is reductive in that it involves 'decomposing the system responsible for a phenomenon into component parts and component operations’ (Bechtel and Hamilton 2007, p. 405). The new mechanical philosophy thus focuses on an abstract research goal: the objective of discovering how the phenomenon of interest is produced. By specifying what constitutes an adequate description of a mechanism, philosophers outlined a number of general epistemic norms.³²

Assertions about scientific practice were able to be made from this characterisation of mechanisms - inferences that were justified by the assumption that ‘in the discovery of mechanisms, the product shapes the process of discovery’ (Darden 2017, p. 264). Philosophers inferred, for instance, that, in order to put forward an explanation, researchers have to identify the entities and activities of the mechanism, assign entities to activities and show how these components are organised, so that, under specific contextual conditions, the mechanism produces the phenomenon to be explained (Bechtel 2009. p. 553). Philosophers also identified a series of reasoning and experimental strategies to guide the discovery of mechanisms.³³ These strategies are abstract descriptions of research actions which help scientists find the mechanism responsible for the phenomenon of interest - that is, they help scientists achieve the goal of explaining a biological phenomenon in an acceptable way. Last but not least, mechanistic research is claimed to have an integrative effect. According to philosophers Carl Craver and Lindley Darden (2013, pp. 162-3), the 'integration of biology is forged by building mechanism schemas that ... satisfy evidential constraints from many areas of biology (chemistry and physics too)’. Since different fields are suited to the study of different kinds of evidential constraints, the search for mechanisms promotes interfield integration and collaborations across disciplinary boundaries.

In conclusion, the debate on mechanisms held over the past twenty years in Philosophy of Science covers many of the topics raised in historians’ analyses of interwar physico-chemical biology. However, the significance of the new mechanical philosophy is based on the assumption that the search for mechanisms was (and still is) important in actual scientific practice. This assumption cannot be supported by philosophical considerations alone. Rather, historical evidence - for instance, from biophysical and biochemical research which was carried out in the 1920s and 1930s - is better suited to substantiating such a hypothesis.

Unanswered questions and the promise of integration

Historians and philosophers of science have engaged extensively with physico-chemical biology and mechanistic explanations. Nevertheless, some interesting questions remain unanswered. Firstly, it is not precisely known where, when and how researchers came to pursue this type of research. Craver and Darden (2013, p. 3), for instance, were unable to uncover in their recent research ‘just when and how this mechanistic view of science entered the different fields of biology, specifically, and precisely how the idea of mechanism came to so thoroughly triumph as a way of thinking about explanation in biology’. Secondly, the philosophical literature provides more of a sketch of an account of scientific practice than a comprehensive analysis. There is still much to learn about the conceptual, material and social preconditions and implications of mechanistic research. Historical analyses will allow scholars to assess the impact of the epistemic norms worked out in the new mechanical philosophy, since by studying history we can learn more about the prevalence and importance of mechanistic reasoning in science. Similarly, by analysing concrete instantiations of past cross-disciplinary research, we can assess the conclusiveness of philosophical statements about interfield integration.

Similarly, most historians’ analyses tell us little about the theoretical and methodological assumptions of cross-disciplinary research in physico-chemical biology between the world wars. Many accounts affirm the importance of Claude Bernard’s or Jacques Loeb’s methodological views, and show how these views inspired leading scientists and philanthropists to promote a certain style of research.³⁴ There, however, the accounts end. For example, it is not yet agreed how scientists came to realise that certain methods could help researchers find solutions to specific biological problems. It is also not entirely clear why biologists and practitioners from the physical sciences decided to collaborate in the first place and how this affected their experimental procedures and theoretical models. The need to know the answers to these questions would be less urgent if it was clear that the emergence of cross-disciplinary research always hinged upon special infrastructures and funding opportunities.³’’ However, this is clearly not the case. Went and Kogl's joint research project, for example, was not funded by the Rockefeller Foundation, nor were there institutional incentives for cross-disciplinary research at Utrecht University. Such cases rather suggest that researchers sometimes had other reasons - possibly epistemic ones - for collaborating with practitioners from other fields.

Although epistemic factors seldom lie at the heart of historians’ analyses, there is no methodological reason for this bias. After all, recognising the significance of non-epistemic factors does not render the study of epistemic factors obsolete. Whatever it was that motivated researchers to enter the field of physico-chemical biology, eventually they had to decide which problems to solve and how; it was at this point that epistemic norms came into play. We have good reason to believe that in planning their research scientists took into account the criteria that would need to be met for their peers to accept their new solutions. The idea of binding disciplinary norms is not unknown to historians. Quite the contrary: a consensus of what constitutes an interesting question, an appropriate investigation, or an acceptable answer is an integral part of most definitions of the term ’discipline’.³⁶ Historians of science have found that young scientists are routinely ‘instructed’ during their training to follow a certain investigative protocol: they learn how to achieve and present their findings in a way that makes their results acceptable to their peers. Whether the research in physico-chemical biology in the early decades of the twentieth century was really shaped by norms similar to those discussed in the new mechanical philosophy is, of course, an empirical question.

In summary, historians and philosophers examined similar points: (a) the search for explanations in which biological phenomena are ‘reduced’ to physico-chemical events and the properties of living matter; (b) the use of physico-chemical methods in biology; and (c) cross-disciplinary research collaborations. While historians presented these as separate features of research in physico-chemical biology, philosophers offered a unifying explanation: advocates of the new mechanistic philosophy see features (a)-(c) as the consequences of the goal of scientists to find mechanisms. It might eventually be possible to explain researchers’ experiments and their hypotheses as well as their decisions to collaborate with researchers from other fields as a direct result of their desire to discover mechanisms. Hence, philosophy has indeed the potential to advance our understanding of past research practice. In the following and final section, I discuss how to avoid the methodological pitfalls discussed in section 2. how to identify historical actors’ research problems, goals, capacities, norms and actions. I then conclude this chapter by pointing out the relevance of such studies for History of Science and Philosophy of Science.