Scientific Realism vs. Antirealism: What Science Actually Tells Us About Reality

The debate between scientific realism vs antirealism is one of the most profound and enduring questions in the philosophy of science, probing whether our best scientific theories actually describe a mind-independent reality or merely provide useful tools for prediction and manipulation. At its core, this dispute asks: Do electrons, black holes, and quantum fields really exist, or are they just convenient fictions that help us make sense of observations? By examining competing arguments through the lens of modern physics, from the unfathomable smallness of quantum particles to the cosmic scale of General Relativity, we can see how this philosophical tension shapes our understanding of what science actually tells us about the nature of the universe.

The Core Positions in the Scientific Realism vs Antirealism Debate

Scientific Realism: The “Truth-Seeking” View

Scientific realism is the view that our best scientific theories are approximately true descriptions of a world that exists independently of our minds. According to this position, the entities posited by successful theories, such as electrons, genes, and spacetime curvature, are genuine features of reality. A key variant is entity realism, championed by philosopher Ian Hacking, which argues that if we can manipulate an entity in a laboratory (like using electrons in a cathode ray tube), we have good reason to believe it exists. Hacking famously wrote, “If you can spray them, then they are real.”

Realists also rely on the “no miracles argument,” first formulated by Hilary Putnam. The argument goes: if scientific theories were not at least approximately true, their stunning predictive success, from predicting the existence of Neptune to the bending of light during a solar eclipse, would be a miracle. The simplest explanation for this success is that our theories latch onto real structures in the world.

Antirealism: The “Empirically Adequate” View

Antirealism is a family of positions that deny some or all of the realist claims. The most influential antirealist stance is empiricism (or constructive empiricism), championed by Bas van Fraassen. The empiricist holds that the goal of science is not truth about unobservable entities, but empirical adequacy – that is, theories should be able to save the observable phenomena. Van Fraassen argues that we can have belief in the truth of what we observe, but we should remain agnostic about unobservables like quarks or the Higgs boson.

Another antirealist tradition is instrumentalism, which treats theories as mere instruments or tools for prediction, without any commitment to the reality of their theoretical terms. For example, a pure instrumentalist would treat the concept of “space-time curvature” not as a literal description of geometry, but as a convenient way to calculate the orbits of planets.

A more radical form is scientific antirealism about laws, sometimes associated with Nancy Cartwright’s image of the “dappled world.” In works like How the Laws of Physics Lie, Cartwright argues that the fundamental laws of physics are only true in highly idealized, artificially simple circumstances – they do not accurately describe the messy, complex reality we actually inhabit.

How Physics Drives the Debate: Three Case Studies

Modern physics provides the most challenging testing ground for the scientific realism vs antirealism debate, because its objects are often so remote from everyday experience. Let’s examine three key areas.

Quantum Mechanics: The Challenge of Indeterminacy

Quantum mechanics is perhaps the greatest battleground. Realists often point to phenomena like the quantum Zeno effect (where frequent observation can “freeze” a quantum state) as evidence that measurement genuinely interacts with a real system. Yet, antirealists highlight the measurement problem: if quantum states (described by the wavefunction) are real, then why does a single observation collapse the wavefunction into a definite value? This leads to bizarre interpretations like the Many-Worlds Interpretation, where all possible outcomes are real in parallel universes. In contrast, the Copenhagen interpretation (championed by Niels Bohr) is explicitly antirealist: it treats the wavefunction as a tool for calculating probabilities, not as a literal description of reality. Bohr famously said, “There is no quantum world. There is only an abstract quantum mechanical description.”

General Relativity and the Reality of Spacetime

Einstein’s General Relativity (GR) describes gravity as the curvature of spacetime caused by mass and energy. For a scientific realist, spacetime curvature is a real property of the universe, confirmed by experiments like the Gravity Probe B satellite NASA’s Gravity Probe B page, which measured frame-dragging effects. Yet, antirealists can point out that GR is only a classical, non-quantum theory. We know it breaks down at singularities (like inside black holes), and attempts to unify it with quantum mechanics (e.g., string theory or loop quantum gravity) propose radically different views of spacetime. If our best current theory is inevitably incomplete, the antirealist argues, we should be cautious about taking its entities as literally real.

The Unobservable in Particle Physics: The Case of Quarks

Quarks are never observed directly – they are confined inside protons and neutrons. A strong antirealist argument comes from the underdetermination of theory by data: there are multiple mathematically equivalent ways to describe the same phenomena. For example, in the mid-20th century, the S-matrix theory (developed by Geoffrey Chew) described particle interactions without positing quarks at all, treating particles as “bootstrapped” from each other. Yet today, the quark model is the standard, and we even have indirect evidence from deep inelastic scattering experiments at the SLAC National Accelerator Laboratory, which showed a point-like substructure inside protons. A realist would say this is strong evidence for quarks; an antirealist would note that we still cannot isolate a single quark, and the theoretical framework (quantum chromodynamics) is so abstract that it often relies on perturbative approximations that fail for low-energy processes.

The Pessimistic Induction and the Meta-Induction

A powerful antirealist argument is the pessimistic induction, also called the “argument from the history of science.” If past successful theories, like Newtonian mechanics, the caloric theory of heat, or the Ptolemaic system, turned out to be false, then we should expect our current best theories to eventually be superseded. This is often paired with a meta-induction over the historical record: science progresses by discarding theoretical entities (phlogiston, ether, Martian canals) even while retaining their empirical successes.

Realists have a counter-argument: the “selective optimistic induction” or “divide and conquer” strategy. They argue that while earlier theories may have been false in their specifics, they often preserved structural truths. For example, Newtonian gravity correctly predicts planetary orbits to high precision, even if its ontological claim (about action-at-a-distance forces) is replaced by the more accurate ontology of curved spacetime. Realists like Philip Kitcher and Richard Boyd argue that the success of a theory comes from the parts that actually refer to real entities and processes – these “working posits” survive theory change, even as “idle posits” (like phlogiston) are discarded.

The Pragmatic Turn: Is the Debate Solvable?

Many philosophers, including Larry Laudan and Helen Longino, have suggested that the debate may be irresolvable in a definitive sense. Laudan argued that the no-miracles argument fails because the success of science is often due to the predictive power of mathematical structures, not necessarily their ontological truth. Meanwhile, pragmatists (following C.S. Peirce and William James) sidestep the debate entirely: they argue that we should accept as “real” whatever entities are indispensable to our best explanations and predictions, without worrying about a final, transcendent truth.

A contemporary synthesis is found in structural realism, defended by John Worrall and James Ladyman. Structural realism holds that we should be realists about the mathematical structure of theories, not the entities they posit. For example, the transition from Fresnel’s ether-based theory to Maxwell’s electromagnetic theory preserved the mathematical relations between waves and fields, even though the entity “ether” was abandoned. This view attempts to thread the needle between the historical evidence of theory change and the impressive mathematical success of physics.

Why It Matters Beyond Philosophy

The scientific realism vs antirealism debate is not merely an academic exercise. How we answer it has concrete implications:

• For funding and research: If antirealism is correct, then funding highly speculative theories like string theory (which makes few testable predictions) might be seen as a waste of resources. If realism is correct, then searching for the ultimate constituents of reality (like the Higgs boson) is a worthy goal.
• For science education: If we teach students that atoms are “real,” we instill a commitment to realism. If we emphasize that a theory is only “a model,” we might better prepare them for scientific revolutions.
• For public trust: When scientists claim that climate models or medical theories are “true,” the public is more likely to trust them. If antirealism were widely accepted, it could undermine the perceived authority of science – though defenders of antirealism argue that empirical adequacy is sufficient for trust.

Sources & References

• van Fraassen, Bas C. The Scientific Image. Oxford University Press, 1980. (Foundational text for constructive empiricism.)
• Putnam, Hilary. Mind, Language and Reality: Philosophical Papers, Volume 2. Cambridge University Press, 1975. (Contains the “no miracles argument.”)
• Cartwright, Nancy. How the Laws of Physics Lie. Oxford University Press, 1983. (Argues for antirealism about laws.)
• Worrall, John. “Structural Realism: The Best of Both Worlds?” Dialectica, vol. 43, no. 1–2, 1989, pp. 99–124. (Key paper on structural realism.)
• NASA. “Gravity Probe B: Testing Einstein’s Universe.” NASA’s GPB page (Accessed 2025). (Source for the empirical confirmation of frame-dragging.)
• Stanford Encyclopedia of Philosophy. “Constructive Empiricism.” Plato.stanford.edu (Accessed 2025). (Comprehensive overview of Van Fraassen’s views.)

Further reading: Scientific realism on Wikipedia