Entity realism is a position in the philosophy of science holding that theoretical entities posited by mature scientific theories - electrons, genes, quarks, and the like - genuinely exist in the world, even when the theories describing their behavior may be incomplete, idealized, or ultimately replaced. It occupies a middle ground between full-blown scientific realism, which extends commitment to both entities and theories, and anti-realist positions such as instrumentalism and constructive empiricism, which deny or suspend judgment about the existence of unobservable things. The position is most closely associated with philosophers Ian Hacking and Nancy Cartwright, who developed it independently in the early 1980s.
The core claim of entity realism is epistemological: scientists are entitled to believe in the existence of entities they can manipulate causally, even if the best theoretical account of those entities is likely wrong or partial. Hacking's formulation is often summarized as “if you can spray them, they're real” - referring to his analysis of experiments in which electrons are used as tools to investigate other phenomena. The ability to exploit an entity's causal properties reliably and systematically is taken as sufficient grounds for ontological commitment to that entity, independent of whether any particular theory about its nature is correct.
This separates entity realism from scientific realism more broadly. A scientific realist typically holds that successful theories are approximately true descriptions of the world, including their theoretical posits. An entity realist makes no such commitment to theories, accepting that the history of science is filled with successful but ultimately abandoned theoretical frameworks, while maintaining that the entities those frameworks were built around may nevertheless be real.
Hacking's Representing and Intervening (1983) introduced the entity realist position through detailed examination of experimental practice. His argument centers on the distinction between merely detecting an entity and genuinely using it as an instrument of investigation. When physicists use positrons to probe the structure of atomic nuclei, they do not merely theorize about positrons - they treat them as tools with reliable causal properties. Hacking argued this practical mastery constitutes a form of knowledge that is more robust than theoretical knowledge.
Nancy Cartwright, in How the Laws of Physics Lie (1983), approached adjacent territory through a critique of fundamental laws. She argued that the abstract laws of physics do not accurately describe the behavior of real objects, while the phenomenological models scientists actually use for prediction and intervention do track real causal factors. Her position is sometimes called causal realism rather than entity realism proper, but both converge on the view that causes and entities are better known than covering theories or laws.
Critics of entity realism have challenged several of its assumptions. The boundary between entities and theories is contested: it is not clear that a commitment to an entity can be cleanly separated from theoretical commitments about what that entity is, since the same experimental setup that licenses belief in electrons also encodes theoretical assumptions about charge, mass, and electromagnetic interaction. This objection is associated with structural realists and anti-realists who argue that Hacking's distinction does not hold under close analysis. Debates about entity realism address these objections in detail.
A further challenge is the problem of unconceived alternatives: even if manipulation licenses ontological commitment, the history of science suggests that our conception of the entity being manipulated may be radically revised or abandoned. The discovery that what was called “the electron” is better described in quantum mechanical terms raises questions about whether pre-quantum physicists were committed to the same entity as their successors.
Entity realism is partly motivated by the pessimistic meta-induction, the argument that the historical track record of abandoned successful theories gives us reason to doubt the truth of current theories. Standard scientific realism struggles to respond to this argument while entity realism attempts to absorb it: theoretical turnover is conceded, but the continuity of experimental manipulation across theoretical change is offered as evidence that at least the entities persist across revisions. Whether this response succeeds is a matter of ongoing discussion in the philosophy of science.
There is no scientific or philosophical consensus on entity realism as a position. It is a live topic within the philosophy of science, where realist and anti-realist positions remain actively contested. See Scientific Realism - Philosophy of Science Consensus for discussion of where the broader debate stands.