Scientific Realism vs Instrumentalism

Direct Answer

Scientific realism and instrumentalism are two answers to a single question: what are scientific theories really telling us? The realist answers: the truth. According to scientific realism, mature scientific theories are approximately true descriptions of a mind-independent world; the entities they posit — electrons, genes, black holes — genuinely exist; and science progresses by getting closer to the truth about reality. The instrumentalist answers: usefulness. Theories are instruments, tools for organizing observations and making predictions. Their unobservable entities are convenient fictions or placeholders, and what matters is not whether they "exist" but whether they work. On the instrumentalist view, the question "do electrons really exist?" is like asking whether the lines of longitude and latitude really exist: useful devices, not discoveries. The debate is not an idle philosophical squabble — it determines what we think science has achieved when it succeeds.

Historical Context

The debate has ancient roots and a distinctively modern form. Aristotle treated the celestial spheres of astronomy as mathematical devices, a position his successors used to reconcile astronomy with physics — an early instrumentalism. Copernicus, Kepler, and Galileo, by contrast, insisted that the heliocentric system described reality, inaugurating realism's modern career. The great systematizer of instrumentalism was the eighteenth-century bishop George Berkeley, who argued that Newton's forces and absolute space were mathematical conveniences rather than realities; David Hume agreed, holding that we never observe causal powers, only constant conjunctions. In the nineteenth century, the positivist tradition — continued by logical empiricists such as A. J. Ayer — treated theories as calculational devices whose content reduces to observation statements. In the twentieth century the debate sharpened. Karl Popper was a realist about falsifiable theories, while Thomas Kuhn and Paul Feyerabend raised doubts about whether paradigm-relative theories could claim literal truth, and W. V. O. Quine argued that the choice between theories is underdetermined by all possible evidence.

Key Distinctions

The debate organizes around three arguments. The no-miracles argument, associated with Hilary Putnam: scientific theories are stunningly successful — they predict eclipses, build microchips, and cure diseases. If theories were not at least approximately true, this success would be a miracle; realism is the only explanation that does not invoke magic. The pessimistic induction, associated with Larry Laudan: virtually every past scientific theory — phlogiston, caloric, the ether — turned out to be false. By induction, our current theories will likely suffer the same fate, so realism's claim that mature theories are true is hubris. The semantic distinction: realists treat theory claims as truth-evaluable descriptions; instrumentalists treat them as calculational devices. The middle position is structural realism: we can be realists about the mathematical structure of theories while remaining agnostic about the entities, because structure survives theory change even when the entities do not.

Contemporary Debates

The contemporary literature offers nuanced positions rather than two camps. Entity realism (with Nancy Cartwright and Ian Hacking) holds that we should be realists about well-manipulated entities — electrons that we can spray and count — while remaining skeptical about theories as wholes. Constructive empiricism, defended by Bas van Fraassen, holds that the aim of science is not truth but empirical adequacy: theories must save the phenomena, and belief should be restricted to what observation shows, not extended to the unobservable. This is instrumentalism's sophisticated heir. The debates now touch the quantum domain, where interpretations differ over whether the wave function is real or a predictive device, and the debate connects to underdetermination: if multiple theories fit all possible data, evidence alone cannot make us realists about any one of them. Realists respond that convergence — successive theories preserving the successes of their predecessors — is best explained by approaching truth.

Practical Implications

The philosophical debate has practical echoes. In medicine, the question of whether a "disease entity" really exists versus whether a diagnostic category merely organizes symptoms affects treatment decisions. In climate science, realism about models — treating them as genuine descriptions of the Earth system — justifies action, while a purely instrumental reading would treat them as mere prediction machines. For the individual inquirer, the mature lesson is a balance: treat well-established scientific theories as the best descriptions we have, while remaining open to revision — the pragmatist attitude of Peirce, who held that belief is warranted by successful practice yet always provisional. You do not need to settle the metaphysics to use science well, but understanding why the debate persists deepens your appreciation of what knowledge and truth can mean.

Related Concepts

• What Is a Theory? — the object of the debate
• What Is Underdetermination? — the evidence problem behind anti-realism
• What Is Knowledge? — what theories are supposed to produce
• What Is Truth? — the realist's target and the instrumentalist's question
• What Is a Paradigm Shift? — the historical engine of the pessimistic induction

Further Learning

• Read the SEP entry on Scientific Realism
• Consult the IEP entry on Scientific Realism and Antirealism
• Trace the epistemology in Critique of Pure Reason
• Visit the Understanding Reality collection for the full learning path