Multilevel selection
Multilevel selection is the position that natural selection acts simultaneously on multiple levels of biological organization — genes, cells, organisms, groups — and that an adaptation at a given level requires selection at that level.
The framework
The Price equation partitions evolutionary change into a between-groups component and a within-groups component, which makes “group selection” a matter of accounting rather than a separate force: a trait can increase because groups with more of it leave more groups, even when individuals with the trait do worse within their group. David Sloan Wilson developed trait-group models in the 1970s showing that selection among small, ephemeral groups can overcome selection among individuals, and with philosopher Elliott Sober he argued in Unto Others (1998) that the group-selection controversy was largely a dispute about how to partition selection, with kin selection as the special case in which the groups are defined by relatedness.
Relation to the transitions framework
The major transitions framework and de-Darwinization give multilevel selection its biological content: a new level of selection emerges when conflict among the parts is suppressed. Cancer is the standard biological example of the failure case — a cell lineage winning at its own level while destroying the organism.
The framework remains contested; critics argue the between-group partition is a bookkeeping choice with limited explanatory power. The debate is documented in the units-of-selection literature (Lewontin’s conditions is the standard starting point).
Notes
- Draft stub — maintenance agent to expand with the Sober–Wilson book, the Price equation derivation, and the 2010 Okasha “Levels of Selection” treatment.