Modal realism, developed by David Lewis, holds that possible worlds are concrete, spatiotemporally isolated universes just as real as the actual world — 'actual' is an indexical term like 'here', picking out whichever world the speaker inhabits. This view analyzes modality without invoking primitive modal notions: to say something is possible just is to say it exists in some world. Lewis argued modal realism earns its ontological cost through explanatory power across modality, causation, mental content, and properties. Critics, including the ersatzist tradition, argue we can get the same semantic benefits with abstracta rather than concrete worlds.
Read Lewis's On the Plurality of Worlds Chapter 1 for the positive case, then Stalnaker's Our Knowledge of the Internal World for the ersatzist response. Evaluate whether Lewis's parsimony arguments are compelling.
From your work on possible-worlds semantics, you know that philosophers use possible worlds as a framework for analyzing modal claims — "it is possible that P" means P is true in some possible world; "it is necessary that P" means P is true in all possible worlds. This framework is extremely useful. But you may have bracketed a pressing question: what *are* possible worlds? David Lewis's modal realism gives the most radical and arguably most coherent answer.
Lewis's central claim is that possible worlds are concrete entities — spatiotemporally isolated universes, fully as real and material as the one we inhabit. Every way a universe could be is a way some universe actually is. There is a world where you took a different job, where the dinosaurs survived, where the laws of physics differ. These are not abstract descriptions, fictional entities, or sets of sentences — they are real places, fully populated with concrete objects, events, and processes. The crucial move is that "actual" is an indexical — like "here" or "now," it picks out different things depending on who uses it. We call our world "the actual world" for the same reason we call our location "here" — it's the world *we* happen to be in, not a metaphysically privileged world.
Lewis's case for this radical ontology is cost-benefit: modal realism pays for its profligate ontology through theoretical unification. Without modal realism, you need to invoke primitive modal notions — possibility, necessity, counterfactual dependence — as unexplained basics. With modal realism, these all reduce to non-modal quantification over worlds and their inhabitants. "Possibly P" = there exists a world where P. "Necessarily P" = every world is one where P. Counterfactuals, properties (as sets of possible individuals), propositions (as sets of possible worlds), and even mental content can all be analyzed in terms of concrete worlds and their denizens. Lewis argued this is *ideologically* parsimonious even if *ontologically* extravagant — the primitive ideology needed is smaller than any alternative.
The main opposition comes from ersatzists, who accept possible-worlds talk but deny that worlds are concrete. Instead, they propose worlds are abstract objects: maximal consistent sets of propositions (linguistic ersatzism), sets of states of affairs (structural ersatzism), or maximal possible properties (Adams). The ersatzist preserves the semantic machinery at far lower ontological cost — no need to posit infinitely many concrete universes. But Lewis's counter-argument is that ersatz worlds must use modal or intentional concepts to specify which abstract structures are *possible*, reintroducing the very primitives the possible-worlds framework was meant to analyze. Modal realism, Lewis claimed, achieves genuine reduction; ersatzism achieves only repackaging.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.