The Inference from Empirical Success to Ontological Independence: A Formal Critique

Step 1: Define the Target Inference

Let:

• T = a scientific theory

• S(T) = T is empirically successful

• R(T) = T is (approximately) true of a mind-independent reality

• I(T) = The ontology of T describes entities that exist independently of observation

Scientific realism typically moves from:

1. S(T)
   to

2. I(T)

The claim is that empirical success justifies ontological independence.

We will examine whether this inference is valid.

---

Step 2: Empirical Success Has an Internal Definition

A theory is empirically successful if:

• it produces accurate predictions,

• its predictions are confirmed by experiment,

• it systematises observed data.

Crucially:

Empirical success is defined entirely in terms of observational outcomes.

Thus:

S(T) is a property of the relationship between theory and observation.

It does not directly concern unobserved reality.

This is already important.

---

Step 3: Independence Is a Stronger Claim

Ontological independence asserts:

• The entities posited by T exist,

• and they exist independently of observation, measurement, and theoretical framing.

This claim goes beyond empirical adequacy.

It asserts a claim about reality outside observational conditions.

Thus:

S(T) concerns observable alignment.

I(T) concerns observer-independent existence.

These are logically distinct claims.

---

Step 4: No Deductive Link Exists

From:

• A theory matches observations,

it does not follow that:

• The theoretical entities exist independently of observation.

The inference is not deductively valid.

Empirical adequacy does not entail ontological independence.

Many alternative explanations of success are logically possible, including:

• Instrumental adequacy (the theory organizes observations without describing independent entities),

• Structural correspondence without intrinsic independence,

• Contextual realism,

• Relational interpretations,

• Pragmatic models.

Therefore:

Success underdetermines ontology.

---

Step 5: The Under-Determination Principle

For any successful theory T:

There may exist multiple ontological interpretations compatible with S(T).

This is especially evident in quantum mechanics, where:

• Different interpretations yield identical empirical predictions.

Thus:

Empirical success does not uniquely determine ontology.

If multiple incompatible ontologies fit the same data, then empirical success cannot justify one specific metaphysical commitment.

Therefore:

S(T) does not entail I(T).

---

Step 6: The Hidden Additional Premise

For the realism inference to work, one must add an extra premise:

The best explanation of empirical success is that the theory describes mind-independent entities.

Call this premise E.

Then realism becomes:

S(T) + E → I(T)

But E is not derived from physics.

It is a philosophical claim about explanation.

Thus scientific realism depends on a metaphysical assumption in addition to empirical data.

The independence ontology is not forced by science.

It is introduced as an interpretative principle.

---

Step 7: Quantum Theory Weakens E

In classical physics, E seemed plausible.

But quantum mechanics complicates the explanatory story:

• The formalism does not assign global intrinsic properties.

• Contextuality prevents non-contextual hidden variable models.

• Measurement plays a structural role.

Therefore, the simplest explanation of empirical success in quantum theory does not require:

• fully definite, observer-independent intrinsic properties.

The explanatory link between success and independence weakens.

Thus premise E loses support.

---

Step 8: Conclusion of the Formal Argument

We can now state the result clearly:

1. Empirical success concerns alignment between theory and observation.

2. Ontological independence concerns existence beyond observation.

3. The former does not logically entail the latter.

4. Multiple ontologies can explain the same empirical success.

5. Therefore, empirical success does not justify ontological independence.

Hence the inference:

Empirical success → Ontological independence

is invalid.

It requires additional metaphysical commitments not licensed by physics alone.

---

Final Structural Conclusion

Scientific realism often presents itself as the rational inference from success to reality.

But the step from empirical adequacy to observer-independent ontology is not compelled by logic or by the formal structure of physics.

It is a philosophical choice.

Quantum theory, by exposing the contextual and relational structure of physical formalism, makes that choice increasingly difficult to defend as a default position.

The independence ontology is therefore not a scientific necessity.

It is an optional metaphysical overlay — one that modern physics no longer requires.