Imagine a map of a city you have never visited. Even if the map mislabels a building—calling a post office a "cathedral"—the map remains immensely useful if it correctly identifies that the building is three blocks north of the river and two blocks east of the park. In the philosophy of science, **Structural Realism** suggests that our best scientific theories are like this map: while we might be wrong about what the "buildings" (entities) actually are, we are remarkably right about the "streets" (the mathematical relationships) connecting them.
## The Best of Both Worlds
Structural realism was famously reinvigorated by John Worrall in his 1989 paper, [Structural Realism: The Best of Both Worlds](https://doi.org/10.1093/bjps/40.2.139). It seeks to resolve the two most powerful, yet conflicting, arguments in the history of science:
1. **The No Miracles Argument:** Science is so successful at making predictions that it would be a "miracle" if our theories were not tracking some objective truth.
2. **The Pessimistic Meta-Induction:** History is a graveyard of discarded theories (like phlogiston or the luminiferous ether). If past scientists were certain and wrong, we are likely wrong too.
Worrall argued that when a theory is replaced, the "nature" of the entities often vanishes, but the mathematical structure is preserved.
## The Case of the Vanishing Ether
The classic illustration involves Augustin-Fresnel’s 19th-century theory of light. Fresnel believed light was a pressure wave traveling through an elastic solid medium called the "ether." When James Clerk Maxwell later developed electromagnetism, the "ether" was discarded as a non-existent entity.
However, Fresnel’s mathematical equations describing how light reflects and refracts remained perfectly intact within Maxwell’s framework. The *entities* changed, but the *structure* of the relationships endured. As Worrall explains:
> "There was continuity or accumulation in the shift, but the continuity is one of form or structure, not of content."
## Epistemic vs. Ontic Structural Realism
While Worrall’s original thesis is known as **Epistemic Structural Realism (ESR)**—the view that we can only *know* the structure of the world—later thinkers pushed the concept further.
**Ontic Structural Realism (OSR)**, championed by philosophers like [James Ladyman](https://plato.stanford.edu/entries/structural-realism/), suggests a more radical metaphysical claim: there are no "things" or "entities" at all. In this view, the world is not made of particles that have relationships; the world is *made* of the relationships themselves. This aligns with modern quantum mechanics, where subatomic "particles" often lack individual identity and are better described as nodes in a mathematical field.
## Why It Matters
Structural realism challenges our deep-seated intuition that the world must be composed of "stuff." If the history of science teaches us that entities are transitory but structures are permanent, we must ask: Is "structure" merely a description of reality, or is reality, at its most fundamental level, nothing but mathematical structure? This remains one of the most provocative questions in contemporary [philosophy of science](https://en.wikipedia.org/wiki/Philosophy_of_science).