Darwinizing Gaia

Many neoDarwinians regard Jim Lovelock’s Gaia Hypothesis as outrageous.
That is, to propose that non-reproducing collectives evolve by natural
selection, and thus sport “functions” as selected effects is impossible
in theory. This is because mainstream neoDarwinians adopt formulations
of evolution by natural selection compatible with “Lewontin’s Recipe”,
where fitness cashes out as differential reproduction. If, on the other
hand, we replaced (or supplemented) such formulations with models based
on David Hull’s replicator/interactor framework, we might see
interactions between species making up “holobionts”, ecosystems and even
the biosphere as a whole as under selection. I will attempt to integrate
earlier attempts (“It’s the song not the singer” and clade selection)
within such a framework.