PLog3 Not even wrong

Why science today isn't even wrong

A short position paper on recoverable distinction, bound standing and scientific entitlement. Part of the Dot theory PLog series 28/09/’26

≡*  Framework

⊙  Operator

V  Residual

∟  Model

Reality does not arrive already labelled.

It does not announce itself as electron, field, organism, cause, information, state, evidence or law. Science makes such distinctions extraordinarily precise through measurement, mathematics, instrumentation, experiment, comparison and criticism. But the distinction itself still has to acquire scientific standing.

That is where an odd gap appears.

Science possesses many highly developed mechanisms for licensing claims: proof, calibration, replication, statistical inference, experimental controls, citation, model comparison, standards, peer review and domain-specific rules of admissibility.

What it does not possess is an equally systematic cross-domain procedure for stating exactly what epistemological standing the resulting scientific object has acquired.

The licensing exists.
The positional record of the licence generally does not.

This matters because science repeatedly uses the same apparently ordinary words for technically different objects.

“State” in quantum mechanics, thermodynamics, computation and psychology is not one thing.

“Information” in Shannon theory, biology and ordinary language is not one thing.

Neither are “observation”, “model”, “representation”, “measurement”, “cause”, “proof”, “constraint”, “reality” or even “data”.

Their scientific meanings are bound by different relations.

And that suggests a distinction between a physical bound state and epistemological bound standing.

A quantum bound state may be completely legitimate physics. Under a specified Hamiltonian, interaction and boundary structure, a particular state is admissible. Its predictions may be experimentally confirmed to extraordinary precision.

That gives it strong mathematical, empirical and physical standing.

But it does not, by itself, license the further statement:

this mathematical state is what Reality fundamentally is.

That is another transition.

And another transition requires another warrant.

formal admissibility → empirical correspondence → physical standing
does not automatically entail:
→ fundamental ontology

The point is not that the bound state is unreal.

The point is that its physical reality is itself scientifically represented under constituted conditions.

Physics cannot escape epistemology merely because its subject matter is physical.

This gives us a useful second notion:

A scientific object has bound standing when its meaning is sufficiently constrained by recoverable scientific relations that another operator can determine what it refers to, what produced it, what it licenses, and where that licence stops.

Bound standing does not require an exhaustive description of Reality.

Quite the opposite.

Perhaps 99.999% of the page can remain blank.

Science does not need to represent everything. It needs to preserve the distinctions material to the operation being performed.

Hence:

partial representation ≠ arbitrary representation
and:
successful representation ≠ exhaustive ontology

This is already how science succeeds.

A thermometer does not contain temperature itself. A detector does not display a tiny particle announcing its identity. A graph is not the phenomenon. A model is not the world.

What matters is whether the relation between signal, operation, interpretation and claim remains recoverable.

Science may therefore be understood as one of humanity's most sophisticated technologies of publicly recoverable distinction.

That formulation also explains why representation alone is insufficient.

A number may persist after its calibration record is lost.

A sentence may persist after its evidential source disappears.

A graph may persist after its modelling assumptions are forgotten.

A model output may persist after the conditions under which it was valid cease to be recoverable.

The symbol survives.

Its scientific standing may not.

Symbolic persistence ≠ constitutional persistence.

That distinction has become especially obvious with artificial intelligence. An AI can produce an excellent scientific explanation without conducting the experiment, a legal argument without possessing legal authority, or a medical-sounding conclusion without examining a patient.

The representation and its entitlement are different objects.

But AI has not created that problem.

It has exposed it.

Scientific culture frequently allows established concepts to inherit standing because generations of theory, evidence, instrumentation and practice lie behind them. That inheritance is indispensable. Nobody can reconstruct all of physics every time they perform an experiment.

Yet inherited standing and locally demonstrated standing are not the same thing.

Nor is cultural familiarity equivalent to epistemological warrant.

What appears to be missing is surprisingly modest: a formal widget recording the standing itself.

Something like:

referent → framework → constituting conditions → operation → evidence
→ standing → scope → failure condition → successor rights

Not another ontology.

A record of why the scientific object is entitled to mean what we are asking it to mean.

This is where the provocation in the title sits.

Science is very good at being wrong.

That is one of its greatest achievements.

But before something can properly be wrong, there must be a sufficiently constituted object whose failure conditions are intelligible.

If two scientists use the same word under different unrecorded jurisdictions, or move from a successful representation to an ontological conclusion without recording the bridge, they may not yet have constituted the same proposition to disagree about.

At that level science is not wrong.

It has not yet earned the privilege.

The right to be right is conditioned by the right to be wrong.
And the right to be wrong is first an epistemological achievement.

Perhaps science already possesses almost everything required.

What is missing is simply a systematic way of recording:

what distinction was made,
under which operation,
what relation binds its meaning,
what standing it earned,
and how far that standing is allowed to travel.

Reality may remain mostly blank.

Science does not need to fill the page.

It needs to know what licences the marks it makes upon it.

⊙