Can semantic facts (facts about meaning and truth) reduce to natural facts investigated by science? Naturalism says yes; non-naturalism says semantic facts are special. This debate connects to questions about whether intentionality is ultimately physical and how to explain the normativity of semantics—that some uses are correct and others incorrect.
Study attempts to naturalize reference: causal theories, informational theories, teleological approaches. For each, identify what natural facts ground semantic facts. Then examine normativity: why is "dog" correctly applied to dogs and not cats? Is this a natural property or a normative one? Consider the metaphysical status of semantic facts and whether they supervene on physical facts.
You come to this topic with an understanding of physicalism about mind — the view that mental facts are ultimately physical facts — and some familiarity with the philosophy of language. Semantic naturalism asks: can the facts that make expressions mean what they mean be explained in terms that a natural science could in principle investigate? The stakes are high because meaning seems stubbornly unlike ordinary physical properties. A rock doesn't refer to anything; a thermostat doesn't *misrepresent* temperature — it just mechanically responds. Yet words refer, and uses of words can be correct or incorrect. Where does this semantic dimension come from?
The most direct naturalist strategy is the causal theory of reference: a word like "water" refers to H₂O because our uses of the word are causally linked, via perception and testimony chains, to actual samples of H₂O in the world. This is a natural fact — a causal-historical relation between linguistic practice and a physical kind. Hilary Putnam's work on natural kinds pushed this idea: "gold" refers to gold in all possible worlds because our term is anchored to the actual substance, whatever its underlying nature turns out to be. The causal theory naturalizes *reference* by identifying it with a type of physical-causal relation.
The harder problem is normativity: meaning isn't just about which things a word is causally connected to — it's about which things the word is *correctly* applied to. "Dog" should be used for dogs, not cats, even if on some occasion I mistakenly call a cat a dog. This correctness dimension — that there is a right and wrong answer — seems to resist purely causal explanation. A causal chain also connects my word "dog" to cats I see and call "dog," yet those uses are wrong. The teleological theory (Ruth Millikan) offers a naturalist response: the correct function of a mental or linguistic state is determined by its evolutionary history — what it was selected to do. Just as the function of the heart is to pump blood (because ancestors with heart-like organs that pumped blood survived), the function of "dog" representations is to track dogs, because these representations were selected for in the evolutionary history of the cognitive systems that use them.
Whether this fully captures normativity remains contested. Kripke's presentation of Wittgenstein's rule-following considerations raises the worry that no natural fact — no causal history, no pattern of use — can determine what a rule *requires* in a new case, because any such fact is compatible with infinitely many different rules. If so, semantic facts have a normative dimension that outruns any natural property. The naturalist has several responses: deflationary accounts that deny the gap is as wide as it seems, communitarian accounts that locate normativity in social practice, or pluralist accounts that treat semantic naturalism as compatible with irreducibly normative semantic facts supervening on natural ones. Your task is to understand which strategy best preserves the naturalist commitment — that semantic facts are not a ghostly addition to the physical world — while accounting for why meaning is normatively structured.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.