Abstract objects — numbers, sets, propositions, properties, types — are entities that (if they exist) are non-spatial, non-temporal, and causally inert. Platonism holds that abstract objects exist independently of minds and languages: the number 7 exists whether or not anyone thinks about it, and mathematical truths are discovered, not invented. Nominalism denies the existence of abstract objects and attempts to paraphrase apparently abstract-referring language in terms of concrete particulars or linguistic conventions. The debate is driven by the indispensability argument: if our best scientific theories quantify over numbers and functions, and we accept those theories, we seem committed to the existence of those abstract entities. The epistemological challenge cuts the other way: if abstract objects are causally inert, how can we have knowledge of them? Benacerraf's dilemma crystallizes this tension between ontological robustness and epistemic access.
Read Benacerraf's 'Mathematical Truth' for the core dilemma, then compare Quine's indispensability argument (for Platonism) with Field's Science Without Numbers (for nominalism). Ask: can we have knowledge of things that have no causal contact with us?
From your study of universals and particulars you know that some things in the world — properties, relations, kinds — seem to be shared across many individual objects. Abstract objects raise a harder version of that question: what about entities that don't seem to be located in or around concrete objects at all? The number 7 is not located under your chair or above the moon. The proposition that snow is white is not somewhere in space. The set of all prime numbers is not a thing you can bump into. Abstract objects — if they exist — are non-spatial, non-temporal, and causally inert. This puts them in a radically different ontological category from anything in your perceptual experience.
Platonism (or Platonic realism) holds that abstract objects exist mind-independently and timelessly. The number 7 existed before humans existed and will continue to exist after we are gone. Mathematical truths are not invented; they are discovered. When a mathematician proves a theorem, they are learning something about a domain of objects as real and determinate as the physical world — just inaccessible to the senses. This view has powerful support: it explains why mathematics is unreasonably effective at describing physical reality, why mathematicians across cultures converge on the same truths, and why mathematical truths feel necessary in a way that physical truths don't. The indispensability argument, associated with Quine and Putnam, makes this support formal: our best scientific theories quantify over numbers and functions as indispensably as they quantify over electrons; if we accept the theories, intellectual honesty requires accepting the objects they posit.
Nominalism denies the existence of abstract objects and faces a two-part challenge: explain away apparent reference to abstract objects, and explain how mathematics works without them. Nominalists argue that sentences like "7 is a prime number" need not be read as asserting the existence of an object called 7; they can be paraphrased as claims about structures, or as useful fictions, or as claims about what certain formal systems license us to say. Hartry Field's *Science Without Numbers* attempts to show that physics can be reformulated without quantifying over mathematical entities at all — making mathematics a useful but ontologically innocent tool rather than a window into abstract reality.
Benacerraf's dilemma crystallizes why the debate is genuinely hard. On one horn: if abstract objects exist (the Platonist position), they are causally inert. Our mathematical beliefs cannot be caused by causal contact with numbers, since numbers don't cause anything. So how do we have mathematical knowledge? On the other horn: if mathematical knowledge requires causal contact with its objects (as knowledge of the physical world does), then either abstract objects are causally potent (and we need a new account of them) or we don't have genuine mathematical knowledge. Any adequate theory of abstract objects must either explain how knowledge without causal contact is possible, or explain why abstract objects don't require such contact, or reconceive what mathematical "knowledge" really is. The dilemma doesn't refute Platonism — but it establishes that the ontological cost of admitting abstract objects is a serious epistemological problem that cannot be waved away.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.