Phil Madgwick
Aug 27, 2017 · 3 min read

I am going to answer your two questions in two parts. First, I am not quite sure what you mean by ‘principles of evolution’, but I think I may be able to answer your questions anyway (!). You will have to correct me if I have misunderstood the question.

I think as my other response made very explicit, I view the problem that natural selection is trying to explain is the appearance of design in nature (which I also clarify here). And, given my mathematical focus, I am inclined to think about ‘principles’ as ‘axioms’ for a mathematical model, i.e. what would I need to tell someone for them to replicate the process of evolution by natural selection in a model. I think the first place to start is thinking about natural selection as a metaphor for something that ‘just happens’ (e.g.). In this line of thinking, it is clear that the onus of axioms comes down to a description of the fundamental ‘unit of selection’ (ignoring the controversy about the use of this term because I think it is obvious what I mean). Dawkins (1976) did exactly this and arrived at the concept of the replicator as the best expression of Neo-Darwinian principles: the idea that if you have replicators then you have Darwinian evolution.

Although I would make a minor amendment to the popular expression of this conception, the replicator is by-and-large the best summation of Neo-Darwinism. In Dawkins discussion, he uses three traits that underlie a successful replicator: fecundity, fidelity and stability. Quite obviously, the title ‘replicator’ implicitly overplays the fecundity and fidelity aspects of a replicator’s strategy, but this three-pronged interpretation of the replicator does somewhat alleviate this confusion. For, of course, one could imagine the evolution of a strategy that does not involve replication if a gene could be immortal (which is not realistic, but is theoretically possible).

Second, I think a good place to start is appreciating the success of current evolutionary theory in making predictions (e.g. sex ratios), but also the constraints on the form of these predictions. Mostly, evolutionary biologists reconstruct the reasons for existing traits to explain their ‘design value’, which is often cryptic (as I mentioned before). Explanation is often understood as the first step toward prediction by philosophers of science, but biology has a problem because explanations often demonstrate some kind of optimisation that is specific to a particular time and place.

What this really comes down to is what it is you think you should be predicting. Maybe you could shed some light on what kinds of predictions you had in mind. From adaptationism, I could predict the form of adaptations given an ecology, but this is quite specific. To avoid specifics and talk about the predictions of the core of evolutionary theory, I think there are two selection as inclusive fitness optimisation is what I consider the fundamental prediction of Neo-Darwinism. In a sense, I take this to rest upon the selfish gene, which I described above as an axiom. What my original article tried to highlight was that there is still some resounding discrepancy in the framework that is used to understand how selfish genes give rise to selfish organisms.

Almost by definition, if we could resolve this problem then we will have a deeper understanding of where the focus of adaptation rests in nature, whether it be the gene, the cell, the organism, the group and so on across ‘hierarchical’ scales of organisation. This deeper understanding should enable us to use inclusive fitness theory without the starting assumption that a particular individuality is the focus of adaptation. Although many examples of this assumption seem valid now, who can tell what oddities there will be lurking in the resolution: for example, is a tree a colony of competing phytomers that are each selected to maximise their direct fitness or a unified organism that maximises the inclusive fitness of the whole?

I am not sure whether this answer will satisfy you, but perhaps you had your own ideas that I could critique for you? This may help me clarify your meaning. Please do reply!

)

Phil Madgwick

Written by

Evolutionary biologist PhD (Milner Centre/University of Bath) on the general theory of Social Evolution, and have a personal interest in applications to AI.