Does Science Move Toward Truth?

Science progresses.

Few people seriously doubt that.

Modern physics predicts phenomena with extraordinary precision.

Medicine treats diseases once fatal.

Astronomy reconstructs cosmic history.

Genetics traces ancestry through molecular evidence.

But progress is not the same as truth.

A theory can become more useful.

More predictive.

More unified.

Does that mean it becomes closer to what reality is actually like?

This is one of the deepest questions in philosophy of science.

What Would “Closer to Truth” Mean?

Truth is usually treated as all-or-nothing.

A statement is true or false.

But scientific theories are large structures.

They contain equations, models, idealizations, ontology, and approximations.

A theory can be wrong in some respects and extremely accurate in others.

So scientific progress may require a notion of approximate truth.

Newton Was Wrong—and Right

Newtonian gravity is not the deepest known theory of gravity.

Relativity replaced it at fundamental level.

Yet Newtonian mechanics still predicts falling objects, planetary motion, and engineering systems with extraordinary accuracy in appropriate regimes.

Calling Newton simply “false” loses something important.

The theory captured genuine structure within a domain.

Correspondence Limits

New theories often reproduce old theories as approximations.

Relativity approaches Newtonian results when speeds are small relative to light and gravity is weak.

Quantum mechanics can recover classical behavior in suitable limits.

This continuity supports a realist intuition.

Scientific progress does not merely discard the past.

It often explains why the past worked.

Approximate Truth

A realist may say Newtonian theory was approximately true in its tested domain.

The world behaves sufficiently like Newtonian predictions there.

Later theories expand the domain and correct deeper assumptions.

Scientific progress becomes a sequence of increasingly accurate representations.

But measuring “distance from truth” formally is difficult.

Verisimilitude

Philosophers use the term verisimilitude for truth-likeness.

Popper hoped scientific progress could be understood as movement toward theories with greater truth content and less falsity.

Formalizing this idea proved difficult.

Still, the intuition remains attractive:

one theory can be closer to reality than another even if neither is perfectly true.

Predictive Progress

One clear form of progress is improved prediction.

A new theory predicts more phenomena, more accurately, across wider domains.

This is measurable.

If theory B repeatedly predicts better than theory A, science has progressed instrumentally.

The philosophical question is whether predictive improvement implies ontological truth.

Explanatory Progress

Science also progresses by explanation.

Newton unified terrestrial and celestial motion.

Maxwell unified electricity, magnetism, and light.

Evolution unified vast biological patterns.

Explanatory scope can increase even when prediction was already good.

Understanding itself seems to improve.

Unification

A theory that explains many phenomena from fewer assumptions often feels closer to underlying structure.

Why?

Because reality seems less fragmented.

Separate laws become manifestations of one framework.

This is one reason physicists value unification.

But unification is a methodological virtue, not proof of truth.

Nature could contain irreducibly distinct structures.

Manipulative Success

Science also increases control.

We build semiconductors, lasers, vaccines, and satellites.

This practical success suggests theories capture something real.

It is difficult to manipulate nature reliably using completely wrong models.

This is another version of the realist no-miracles intuition.

No-Miracles Argument

Scientific realists argue:

the best explanation for the extraordinary predictive and technological success of mature science is that its theories are at least approximately true.

Otherwise, success would be miraculous.

The argument is abductive.

It does not deductively prove realism.

But it is powerful.

Pessimistic Meta-Induction

Anti-realists respond with history.

Past science also had successful theories later abandoned.

Phlogiston.

Caloric fluid.

Ether models.

If previous generations mistook useful theories for truth, perhaps we are doing the same.

This is the pessimistic meta-induction.

Scientific success may not guarantee correct ontology.

Structural Realism

Structural realism offers a compromise.

Perhaps science becomes increasingly accurate about relations, equations, and causal structure even when its picture of entities changes.

The structure linking observables may survive theory change.

Science then approaches truth primarily at the level of relations.

Entity Realism

Another route focuses on entities we can manipulate.

Electrons.

Genes.

Viruses.

If scientists can reliably intervene using them, their reality seems well supported even if theories surrounding them evolve.

Progress may occur unevenly.

We can become confident about entities before knowing their ultimate nature.

Kuhn’s Challenge

Kuhn complicates the idea of linear progress.

Paradigms can change concepts, standards, and problem definitions.

If standards themselves shift, comparing old and new theories becomes difficult.

Kuhn still saw progress in problem-solving ability.

But he resisted a simplistic picture of science marching directly toward one fixed final representation.

Incommensurability Is Not Incomparability

Even if paradigms are partly incommensurable, they are not totally incomparable.

Relativity predicts everything Newtonian mechanics predicts in its domain and more.

Quantum theory explains phenomena classical mechanics cannot.

Scientific communities can compare accuracy, scope, coherence, and fertility.

Conceptual change complicates progress.

It does not erase it.

Laudan and Problem-Solving Progress

Philosopher Larry Laudan emphasized problem-solving effectiveness.

A research tradition progresses when it solves more empirical and conceptual problems with fewer serious anomalies.

This avoids committing to a final metaphysical truth.

Science can progress by solving problems even if realism remains unsettled.

Instrumentalist Progress

An instrumentalist can define progress without truth.

Better science means more accurate predictions, broader empirical adequacy, and more effective control.

No claim about hidden reality is required.

This view captures practical success but may understate science’s explanatory ambition.

Scientists usually want to know not only what works but what exists.

Convergence

Do independent scientific methods converge?

Often, yes.

Atomic theory is supported by chemistry, spectroscopy, scattering, microscopy, and quantum theory.

The same underlying structures appear across independent methods.

Convergence strengthens realism because one systematic illusion would need to explain many different successful pathways.

Approximation Across Scales

Scientific truth may be layered.

Fluid dynamics is true enough at macroscopic scales.

Molecular physics explains its limits.

Quantum field theory operates deeper still.

One level need not eliminate another.

If reality is hierarchical, then progress may involve discovering increasingly connected levels rather than one final vocabulary.

Is There a Final Theory?

Perhaps science eventually finds fundamental laws.

Perhaps every deeper explanation opens another layer.

We do not know.

Even a final physical theory might not eliminate complex emergence, computational irreducibility, historical contingency, or semantic explanation.

Complete fundamental laws would not necessarily mean complete practical knowledge.

Error Reduction

One modest way to define progress is:

science reduces error.

Measurements become more precise.

Predictions improve.

Systematic biases are discovered.

Domains of failure become known.

Even if final truth remains unreachable, moving away from demonstrable error counts as progress.

Science Does Not Move Uniformly

Progress is uneven.

One field may advance rapidly.

Another may stagnate.

Some theories persist for social reasons.

Some discoveries are lost and rediscovered.

Science is a human institution.

There is no guarantee of monotonic improvement every year.

Long-term progress can coexist with local regression.

Does Science Converge?

Many areas show convergence.

Independent laboratories measure the same constants.

Different instruments detect the same phenomena.

Competing theories narrow toward shared constraints.

This suggests that reality exerts a stabilizing pressure.

Scientists can disagree.

The world does not change to accommodate the disagreement.

Truth as a Regulative Ideal

Perhaps truth is best treated as a regulative ideal.

Science acts as though there is a reality independent of our wishes and attempts to make representations increasingly answerable to it.

We may never know that a final description has arrived.

But the ideal guides testing, correction, and comparison.

Truth functions as a direction even without a guaranteed endpoint.

A Careful Answer

Does science move toward truth?

A strong but cautious answer is:

Science clearly moves toward greater empirical accuracy, wider explanatory scope, stronger unification, and better control. These forms of convergence give substantial reason to think mature science often approaches real structure, even though history gives us no guarantee that current theories are final or literally true in every detail.

Progress is real.

Finality is not assured.

The End of Part IV

We began this part by asking what science is.

We examined observation, measurement, evidence, models, experiment, prediction, computation, induction, explanation, causality, simplicity, demarcation, falsifiability, verification, realism, underdetermination, paradigms, and scientific change.

We now understand science not as a machine that produces certainty, but as a disciplined method for learning under uncertainty.

The next part widens the frame.

Science is not humanity’s only attempt to understand existence.

Religion, myth, and mysticism also offer narratives, practices, and forms of meaning.

The next question is therefore:

How should science and religion relate to one another?