When scientists attempt to explain animal behaviour, representation almost always arrives uninvited.
The language may vary — information, signals, internal models, decisions, strategies — but the assumption is the same:
for behaviour to be flexible and appropriate, the organism must somehow represent the situation it is in.
This assumption feels so natural that it often goes unnoticed.
It is also almost certainly wrong.
The representational reflex
Representation enters explanation not because it has been demonstrated, but because it reassures.
If an animal can be said to:
perceive information,
form an internal picture of the world,
select among options,
and execute a plan,
then the mystery dissolves. Behaviour becomes legible by analogy with ourselves.
But this legibility is purchased at a cost: the phenomenon is redescribed rather than explained.
Why representation feels inevitable
There are three reasons representation seems unavoidable.
1. We are semiotic creatures
Humans live in meaning.
We navigate the world through symbols, categories, reasons, and interpretations. When we act flexibly, it is usually because we understand what we are doing.
This makes it extremely difficult not to project the same structure onto other organisms.
But analogy is not evidence.
2. Representation flatters explanation
Representational accounts promise depth:
hidden structures,
internal states,
explanatory mechanisms beneath the surface.
In practice, they often function as labels for ignorance.
To say that an animal “represents the environment” is rarely to specify how behaviour is coordinated. It is to name the coordination and move on.
3. We mistake description for necessity
Because representational language can describe behaviour coherently, we assume it is required to produce it.
This is a mistake.
A vocabulary that makes sense to us need not correspond to the causal or organisational structure of the system itself.
What representation would have to do
To justify its place, representation would need to explain at least four things:
Speed — how appropriate action is produced faster than deliberation would allow.
Robustness — how behaviour remains adequate despite noise and perturbation.
Novelty — how genuinely new situations are handled without prior encoding.
Coordination — how multiple degrees of freedom are brought into alignment in real time.
Representational accounts typically presuppose these achievements.
They do not derive them.
The hidden cost of representation
Once representation is assumed, several consequences follow — often unnoticed.
Teleology creeps in
Representations are always about something.
This introduces goals, correctness conditions, and success criteria that belong to human meaning-making, not to biological coordination.
The environment is externalised
The world becomes input to be processed, rather than part of the system within which behaviour is organised.
This makes genuine coupling harder to see.
Competence is intellectualised
What is in fact a matter of viability, timing, and constraint is recast as a matter of knowledge and decision.
The problem appears solved, but only because it has been transformed into a different problem altogether.
A different question
If we resist the representational reflex, a different line of inquiry opens.
Instead of asking:
What does the animal know or represent?
we can ask:
What structure of possibilities makes this behaviour available to the system at this moment?
This question does not require:
internal pictures,
symbolic content,
or meanings.
It requires only that behaviour be understood as the actualisation of viable possibilities under constraint.
Clearing the ground
Rejecting representation does not mean denying:
neural complexity,
learning,
sensitivity to history,
or sophistication of behaviour.
It means refusing to explain coordination by importing human categories where they do not belong.
In the next post, we will make the most important distinction this series depends on:
the difference between value and meaning.
Without that distinction, representation will always sneak back in.
With it, we can finally begin to explain how animals do what they do — without knowing what they are doing.
No comments:
Post a Comment