Few debates in evolutionary biology have been as persistent — or as unproductive — as the question of the unit of selection.
Is selection acting on genes, organisms, groups, species, or lineages?
Decades of argument have not resolved the issue. This is usually treated as a sign of conceptual difficulty or empirical complexity.
It is neither.
The Expectation of a Unit
The question presupposes that selection must act on something — a stable entity that persists long enough to be selected, compared, and explained.
This expectation is inherited from a deeper habit: explanation requires a bearer.
If selection explains differential persistence, then something must persist.
And so the search begins.
The Usual Candidates
Each proposed unit offers clarity — and demands a cost.
-
Genes promise discreteness and inheritance, but abstract away development, environment, and phenotype.
-
Organisms preserve intuitive identity, but fracture under symbiosis, microbiomes, and life cycles.
-
Groups capture collective effects, but struggle to define boundaries and reproduction.
-
Species offer historical continuity, but lack clear mechanisms of selection.
-
Lineages capture persistence, but dissolve into trajectories rather than things.
Each candidate works — until it doesn’t.
Why the Debate Never Ends
The persistence of the debate is often attributed to insufficient data or conceptual confusion.
But the pattern is telling.
Each proposed unit stabilises selection at a particular scale — and thereby suppresses relations at others.
The debate cannot resolve because the demand for a unit is itself misplaced.
Selection Without a Bearer
Darwin did not require a unit of selection. He required only differential persistence within a population of varying entities.
Later evolutionary theory demanded more.
It sought a stable bearer to anchor explanation — something that could carry fitness, be acted upon, and persist across time.
This demand reflects an intolerance.
The Intolerance of Distributed Persistence
Selection operates across relations.
It does not act on isolated units, but on patterns of interaction that cut across levels: genetic, developmental, ecological, and historical.
Persistence is distributed.
There is no single locus where selection “happens”.
This is deeply unsettling.
Relational Units
From a relational stance, a “unit” is not a thing but a cut.
What counts as a unit depends on:
-
the timescale of analysis
-
the relations foregrounded
-
the constraints of interest
-
the explanatory purpose
This does not make them arbitrary. It makes them accountable.
Why Gene-Centrism Persists
Gene-centred views persist not because they are more true, but because they offer maximal stability with minimal relational cost.
Genes are discrete, replicable, and abstractable.
They function as ontological anchors in a field that resists anchoring.
This is explanatory convenience, not metaphysical privilege.
Units and Identity Revisited
The unit problem is the evolutionary form of the intolerance of non-identity.
If entities do not preserve strict identity across time, then no single unit can be maintained without distortion.
What Is Being Protected
The insistence on a unit protects the idea that evolution can be explained by reference to stable entities rather than evolving relations.
It preserves the fantasy that explanation can be localised.
But evolution is not local.
It is distributed across histories, environments, and constraints.
Units as Tools, Not Foundations
This does not mean that talk of genes, organisms, or groups should be abandoned.
It means they should be treated as tools rather than foundations.
Each unit cuts the evolutionary field differently. Each reveals something and obscures something else.
No unit deserves ontological primacy.
The Consequences of Refusal
When the intolerance of the unit dominates, evolutionary explanation becomes distorted:
-
debates are framed as zero-sum
-
explanatory levels are treated as competitors
-
relational effects are reduced or ignored
-
ontological commitments masquerade as empirical claims
The question “which unit is correct?” becomes a substitute for the harder question of perspective.
Selection as Cut
Selection is not something that acts on units.
It is the differential actualisation of possibilities across a constrained field.
Units are how we render that field tractable.
They are not what the field consists of.
Looking Ahead
If there is no privileged unit, and fitness is relational, then the next intolerance comes into view.
How does evolutionary theory account for novelty — not recombination, not optimisation, but the emergence of genuinely new possibilities?
The next post will take up The Intolerance of Novelty — where explanation most clearly strains.
No comments:
Post a Comment