Have measurements done with modern machinery (Atlas detector, Icecube neutrino observatory, etc.) made the human intellect cross the Kantian

When Bas van Fraassen introduced constructive empiricism, he landed a radical blow against both traditional scientific realism and classical instrumentalism. A common misconception is that van Fraassen’s views match classical instrumentalism—which reduces unobservable terms like *singularity* or *weak force* to mere mathematical shorthands, convenient fictions, or non-referential calculating devices. Instead, van Fraassen asserts that theoretical terms ought to be interpreted **literally**. According to constructive empiricism, when a theory talks about an unobservable vector field or a weak force, it makes a genuine, truth-evaluable claim about the universe. The departure from realism lies not in the *meaning* of the terms, but in our *epistemic attitude* toward them. > "Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate." > — Bas van Fraassen, [*The Scientific Image*](https://doi.org/10.1093/0198244274.001.0001) ## Literal Semantics Without Ontological Commitment To understand what theoretical terms stand for in this framework, it helps to distinguish between semantic content and epistemic belief. Classical instrumentalists, such as the logical positivists in their early phases, attempted semantic reductionism: they claimed theoretical sentences could be translated without loss of meaning into statements about observable phenomena. Van Fraassen rejects this reductionist strategy entirely. Under constructive empiricism: - **Literal Meaning**: Theoretical terms like *quantum field* or *gravitational singularity* refer directly to hypothetical entities, properties, or structures. The sentence "a Higgs field permeates space" has the exact same truth conditions for an empiricist as it does for a realist. - **Empirical Adequacy over Truth**: While the realist asserts that the theory is *true* (and thus that the Higgs field exists), the constructive empiricist holds that we only need to accept the theory as *empirically adequate*. A theory is empirically adequate if what it says about the *observable* phenomena is true. - **Agnosticism, Not Denial**: Theoretical terms do not stand for "fictions" or "useful lies." Rather, we remain strictly agnostic about whether their theoretical referents exist in the world. For a concrete illustration, consider how a semantic realist and a constructive empiricist view a theoretical construct like dark matter. The realist believes the term *dark matter* refers to an actual, unobservable physical substance. The classical instrumentalist claims *dark matter* is just an equation component used to fix galactic rotation curves. Van Fraassen's constructive empiricist takes a middle path: the term *dark matter* literally purports to refer to an unseen mass, but scientists only need to commit to the claim that the gravitational equations successfully predict our macroscopic, observable astronomical observations. ## The Semantic Representation: Theoretical Models If theoretical terms are not reduced to observations, how do they function within a scientific theory? Van Fraassen shifts the focus of philosophy of science from a syntactic view (theories as axiomatic sets of sentences) to a **semantic view** (theories as collections of models). In this structural model-based framework: 1. Theoretical terms derive their meaning from their place within abstract mathematical structures (models). 2. A theory provides a family of candidate models that attempt to represent physical reality. 3. The crucial linkage is the **empirical substructure**. The theory specifies a structural part of its model that is isomorphic to the observable phenomena (the "appearances"). As van Fraassen expands in [*Quantum Mechanics: An Empiricist View*](https://doi.org/10.1093/019824861X.001.0001), theoretical terms like *wavefunction* or *spin* do not need to be tethered to unobservable micro-entities in the physical world. Instead, they operate as necessary components within an abstract model that allows the empirical substructure of that model to align perfectly with observable measurement outcomes. Thus, theoretical terms stand for elements within an abstract, mathematical representation. Whether those elements correspond to real, unobservable objects in nature is a question van Fraassen deems scientifically unnecessary to answer. Science succeeds by producing empirically adequate models, making true belief about the unobservable realm a superfluous metaphysical add-on. For deeper exploration of these semantic dynamics, see the [Stanford Encyclopedia of Philosophy entry on Constructive Empiricism](https://plato.stanford.edu/entries/constructive-empiricism/).