A central part of scientific enquiry involves constructing representations of the world, or more accurately of those objects, events and processes in the world that the science in question is concerned with. Representations can take many forms, including diagrams, taxonomies, verbal descriptions, physical models, and abstract mathematical models. Thus a diagram of the solar system, a taxonomy of Alpine flora, a ball-and-stick model of a chemical substance, and a mathematical model of the spread of a disease are all examples of representations. Different though they are, each of these scientific constructs aims to represent some system in the world (the “target system”) and can be assessed for how well they achieve this aim. 

A good scientific representation needs to be accurate, that is, to correspond reasonably well to its target. Typically, though, a representation will not correspond to its target in all respects but only in important ones. For example, a diagram of the solar system may depict the relative size of each planet and its distance from the sun, but not its surface topography. Furthermore, many representations involve idealization, that is, they assume away certain features of their target system for simplicity. For example, an economic model of international trade may treat transport costs as zero, despite this not being true of real-world trade. Thus a scientific representation need not exhibit perfect fidelity to its target, and usually does not aspire to. This raises an important philosophical issue: how to distinguish idealized-but-still-useful representations of a target system from misrepresentations. 

To construct a representation of a target system, of any sort, requires the use of concepts. Indeed concepts can be seen as the building blocks of scientific representations. Chemists could only construct ball-and-stick models of chemical substances once they had the concepts of atom, molecule and chemical bond. Epidemiologists could only construct mathematical models of the spread of Covid-19 once they had the concepts of infectious agent, exponential growth, and basic reproductive number. To understand how scientific representation works, therefore, it is essential to consider the concept-formation that underlies it. Moreover, since concepts are essentially bound up with inferences, or patterns of reasoning, the study of scientific representation cannot be divorced from the study of the inferences associated with the concepts from which the representations are built. 

The aim of the proposed project is to examine how biological evolution has been represented – diagrammatically, verbally and mathematically – in the scientific literature, past and present. A further aim is to examine representations of evolution in the context of pedagogy and science communication. “Biological evolution” is taken to include the process of descent with modification that Darwin described; the mechanisms that drive the evolutionary process such as natural selection; and the products to which the process has given rise, such as organic adaptation and diversity. Scientists have constructed representations of each of these elements in their quest to understand how evolution works. The project will offer a systematic study of these representations, the concepts from which they are built, and the associated inferences, from an overarching philosophical perspective.  

The project has six work strands. The first strand examines diagrammatic representations of evolution, such as trees, landscapes and causal graphs. The second examines linguistic representations, particularly the use of metaphors and analogies to describe the evolutionary process. The third examines mathematical representations, as found in the abstract models that evolutionary theorists develop. The fourth strand examines “ways of thinking” about evolution, that is, fundamental cognitive styles that scientists and laypeople alike use to think and reason about evolutionary phenomena. The fifth strand considers the communication of evolutionary ideas, in particular how evolution is represented in science education and non-specialist fora. The sixth strand examines the project of generalizing evolution to the non-biological realm, a project whose feasibility depends in part on which representations of evolution are treated as canonical. 

The importance of the project lies in its integrative ambition. The project will bring together philosophical ideas about the nature of representation and idealization, linguistic ideas about metaphor and analogy, psychological ideas about reasoning and cognitive biases, and educational ideas about science communication. By drawing on such a diverse range of ideas, the project will deepen our understanding of how evolution is, has been, and should be represented. The results will be of interest to both philosophers of science and scientific practitioners alike. 

For further information on the research project, prospective applicants should contact Samir Okasha directly (or project coordinator Charlotte Withers cw1658@bristol.ac.uk)