The Project

Diagrammatic representations of evolution abound . Three types of diagram will be examined in detail:

(i) the “tree of life” in its various incarnations, up to and including contemporary phylogenetic trees; 

(ii) the “ adaptive landscape ” in its multiple versions , used to depict the (supposed) trajectory of a population evolving under natural selection; 

(iii) causal graphs and flow diagrams used to capture aspects of evolution and inheritance , such as Weismann’s famous depiction of the “continuity of the germ plasm”. 

These diagrams are interesting since all have been criticized, in some cases trenchantly, but nonetheless continue to be used today, and arguably are of heuristic value in at least some contexts . 

The aim will be to scrutinize the scientific role, utility , and representational adequacy of these (families of) diagrams. 

The key questions are whether the diagrams constitute useful idealizations, or unwelcome distortions, of biological reality; and whether they aid or impede our understanding of the evolutionary process. 

These issues will be tackled by making explicit the presuppositions on which the diagrams rest and evaluating the controversies to which they have led.

Verbal descriptions of evolution often employ metaphors or invoke analogies with other sorts of process. 

Purposive metaphors use the language of functions, intentions and goals to describe Darwinian evolution and / or the evolved traits of organisms. 

Cognitive metaphors use the language of folk psychology, such as “tries”, “knows” and “wants”, to describe the process of evolution and its products. 

Dynamic metaphors use the language of “ forces ” to describe the causal processes , such as natural selection, that give rise to evolutionary change. 

Engineering metaphors use the language of design, search and problem - solving to describe adaptive evolution. 

The prevalence of these metaphors raises the issue of what heuristic value (if any) they bring ; what their presuppositions are;  whether they could in principle be dispensed with; and whether they can be grounded in fundamental evolutionary theory . 

These issues will be addressed by studying the scientific role and heuristic value of the metaphors and analogies in question , and by drawing on the resources of cognitive metaphor theory (Lakoff and Johnson 1980).

Mathematical models have been the mainstay of evolutionary theory since the 1920s. The story of how Darwin’s ideas came to be mathematicised by the neo - Darwinists has been told many times. 

Nonetheless key issues remain unresolved, such as whether the basic Darwinian argument – that natural selection on heritable variation will lead to adaptive change – is actually vindicated by contemporary models of evolution or not. Another issue is whether models can help assess the validity of non - mathematical representations of evolution , for example, adaptive landscape diagrams and dynamical metaphors . 

A third issue concerns idealization. Since all evolutionary models rely on idealizations, the goal of expressing evolutionary principles in mathematical form must rely, at least implicitly, on pragmatic decisions . 

This prompts the question of whether some mathematical models of evolution are more fundamental than others, in the sense of resting on fewer idealizations, and if so which. 

These issues will be tackled by attending carefully to the practice of evolutionary modelers, and by drawing on the philosophical literature on models and idealization in science.

Certain psychological tendencies, or “styles of thinking”, influence how humans think about the world, including the biological world. 

Teleological thinking imputes aims and goals to processes in the world; Essentialist thinking categorizes worldly objects into fixed categories; Strategic thinking interprets actions and events in the world as resulting from rational agents pursuing their self - interest.

These styles of thinking , at least some of which are cross - cultural and arise early in development, might reasonably be regarded as cognitive biases that need to be overcome by scientific training . ( For example, a predisposition for teleology can make it harder for students to understand that natural selection lacks foresight.) 

One the other hand , it is possible that certain representations of evolution , including heuristically powerful ones, may themselves be influenced by these styles of thinking. For example, the appeal of game - theoretic models and the “ selfish gene ” metaphor arguably derives in part from our predisposition towards strategic thinking; similarly for purposive metaphors and teleological thinking. The relation between representations of evolution and our psychological tendencies thus requires close study . 

These issues will be tackled by integrating the results of strands 1 - 3 of the project with relevant empirical findings from the cognitive psychology literature

The communication of evolutionary ideas is an important issue , given the large number of people who do not accept evolution, and the frequent misunderstandings among those who do. 

There is a substantial discussion of this issue in the science education literature. However, the role that representations play in the communication of evolutionary ideas has not been systematically examined. 

One important question is whether certain representation s of evolution , e.g. particular diagrams or metaphors, can be shown to aid or to hinder accurate understanding of evolution. 

A second question is whether certain evolutionary concepts and constructs, e.g. phylogeny as the basis for classification, natural selection as a causal process, are especially hard to teach and communicate, given ingrained styles of thinking . 

These questions will be tackled by drawing on the results of strands 1 - 4 of the project, drawing on relevant findings from the science education literature, and conducting a systematic study of representations of evolution in textbooks and other non - specialist fora. 

The idea that biological evolution is just one instantiation of a more general evolutionary schema has been propounded by scholars in various fields. Thus evolutionary principles have been applied to cultural practices, languages, scientific theories, and firms in the marketplace, among others. Though much progress has been made with the goal of generalizing evolution beyond the biological case, important philosophical and scientific questions remain unresolved. A number of these questions relate to the representations of evolution – diagrammatic, verbal and mathematical – explored in strands 1-3 above. In particular, can representations of evolution that were originally developed for the biological realm, and the concepts from which they are built, be straightforwardly generalized to the non-biological realm or not? Are some representations more amenable to such generalization than others, and if so why? Are some representations of evolution inherently biological in character and thus not suitable for use in a fully general evolutionary theory?

Two examples can make this issue more concrete. The first concerns mathematical models of evolution. Most traditional models of evolution were developed in the field of population genetics and rely on the assumption of Mendelian inheritance; such models do not naturally apply to non-genetical evolution and their morals do not necessarily generalize. However, alternative models of evolution, such as ones based on the replicator dynamics  and the Price equation , rely on fewer biological assumptions so can be more easily extrapolated beyond the biological realm. However, this generality comes with a cost, as such models often lack the predictive power of population-genetic models and have sometimes been criticized as empirically empty. Therefore, the issue of whether biological evolution should be viewed as one instantiation of a more general evolutionary schema depends in part on which (if any) mathematical representations of biological evolution we treat as fundamental. 


The second example concerns diagrammatic representations. Phylogenetic trees are widely used in contemporary biology to depict patterns of ancestry and descent between taxa, but such trees rely on the assumption of strict branching, i.e. lineages that split do not join up again. This assumption is a good approximation for most (non-microbial) biological taxa but not necessarily in the non-biological realm. Thus where cultural and linguistic evolution are concerned, for example, the assumption of strict branching is inappropriate, so tree-like representations of evolutionary change are not necessarily possible. This suggests that the answer to the much-debated question of whether biological and non-biological evolution are fundamentally alike or not may depend in part on which representations of evolution we employ.