Inheritance and cooperation: Clarke introducing the themes

Apparently some short girl called Clarke dreamt up the idea of linking inheritance and cooperation for the focus of a conference. An audio file of my introductory talk can be found here.  Here is (roughly) what I had to say in my defence:

Cooperation is interesting, there is a lot of it about, but we struggle to explain why it doesn't fall apart in the face of selfish cheaters, free riders, who take the benefits of cooperation without paying their fair share of its costs. Kin selection theory is great because it allows us to explain how cooperation can be maintained despite the free rider threat. But the problem is this theory only helps when cooperation is between relatives, interactors who carry the same genes. But what is it about sharing a copy of the same gene – what is it about being the same – that helps organisms to cooperate?

Our standard story says that a gene is a type, which can manifest in many different physical copies. What a gene ‘wants’ is to have as many copies as possible, but it doesn’t care which ones. So if one of the organisms that carries a copy sacrifices itself in such a way that organisms that carry other copies benefit, then as long as the total number of copies goes up, rather than down, the gene is ‘happy’. It will spread. So in fact two copies of a single gene simply cannot cheat one another. Both copies have their fitness boosted by all and only behaviours that increase the total number of copies – whether they are one of those copies or not.

Something to notice about this explanation is that it doesn't actually rely on genes. All that's necessary is that interaction takes place between multiple tokens of a common type. The question is, are there other, non-genetic, types whose tokens could help us to explain instances of cooperation between non-relatives?

We are becoming  aware that genes are just one among many systems of inheritance, systems for transmitting content between individuals.  Can these extended inheritance systems (epigenetics, ecological inheritance, cultural inheritance, lateral gene transfer and symbiote inheritance) help us to explain any of the interactions that kin selection theory struggles with? 

Our ideas about what processes inheritance systems can support  depend upon the details of how we have chosen to conceptualise natural selection, and vice versa.  For example, If natural selection necessarily involves replicators, this implies that cultural selection can only occur if there are populations of discrete cultural units that differentially replicate themselves: memes. If, on the other hand, cultural evolution appears to take place among cultural artefacts that seem to show lineage-like patterns of change over time, but where variation looks continuous, and its hard to identify particulate units that are being copied, such as with human languages, we might find reason to favour the classical view of evolution.  So thinking about cultural selection might help us to reach new insights about biological selection processes. 

I pose the following questions for future work on the influence of inheritance on cooperation:

Are our explanations of cooperation properly divided into proximate and ultimate kinds, or is this distinction misleading? How should we explain cooperation in humans?

Should we define inheritance in terms of the transmission of information, and if so, what kind of information?

What consequences does extended inheritance have for key theoretical notions in evolutionary biology, such as relatedness, reproduction, fitness? Can we use Hamilton's rule with respect to units of extended inheritance? Do processes of extended inheritance necessitate revisions to such concepts? 

Can extended inheritance support free-standing evolutionary processes, or does selection on genes always act as a brake?

Ought we replace the focus on genetic inheritance with a more inclusive notion?  And if so, how inclusive should it be?

Listen to the talks here.

Next post: Helantera and Uller: Superorganisms as model systems