If Numbers Aren’t Real, Why Do We Trust Math?
When Numbers Play Hide-and-Seek
You sit down with your math homework. The problem says 4 + 5 = ?. You instantly write 9. But pause for a moment: what is the number 9? You cannot point to “nine” the way you can point to nine apples. If every apple disappeared, would the number 9 still exist? This puzzle—what are numbers, and where are they?—has kept philosophers arguing for thousands of years.
One bold answer says that numbers are completely real, but they are not made of matter. They exist outside space and time, in a special kind of reality. This view is called platonism (after the ancient Greek philosopher Plato, c. 428–348 BC). Platonists think mathematical objects are abstract objects: things that are real but not physical or mental. A chair is physical; a thought is mental; but a number is neither. It lives in a kind of ghostly, timeless world.
Platonism is attractive because it takes everyday math talk seriously. When we say “7 is prime” it certainly sounds like we are claiming something about an object—the number 7—just as “the apple is red” claims something about an apple. Platonists say that is exactly what is happening. Mathematical sentences really are about objects, and those objects are abstract. This gives math a clear foundation. But it also leaves a huge mystery: if numbers exist in a world beyond space and time, how do we humans, stuck inside spacetime, ever get to know anything about them?
Plato’s Hidden Kingdom: Abstract Objects
For a platonist, every equation you learn uncovers a tiny piece of an unseen kingdom. The equation 2 + 2 = 4 would be true even if no humans had ever existed. The number 4, and the fact that it is the sum of two 2s, simply is, eternally.
The philosopher Gottlob Frege (1848–1925) championed this idea. He argued that numbers are objective, real objects that we discover, not invent. Platonism seemed to fit how mathematicians actually work: they talk as if they are exploring territory, not making it up. Yet the worry about knowledge would not go away. In the twentieth century, Paul Benacerraf (1931–2025) sharpened the challenge: human knowers are physical creatures who learn through sight, touch, and causal contact. Abstract objects, being outside spacetime, cannot bump into us or send signals. So how could we ever get reliable information about them? This epistemological argument—the problem of how we know abstract objects—made many philosophers uncomfortable with platonism. They began hunting for alternatives.
If Numbers Aren’t Abstract, What Are They?
Some thinkers tried to put numbers back into the ordinary world. John Stuart Mill (1806–1873) proposed physicalism: math is really about physical piles of objects. “2 + 3 = 5” just tells you that when you push a pile of two rocks together with a pile of three rocks, you get a pile of five. But this runs into trouble with infinity. Mathematics talks about infinite sets that are vastly bigger than any collection of physical things. The universe simply doesn’t hold enough rocks.
Another group endorsed psychologism: numbers are ideas inside our heads. On this view, “3 is prime” describes a mental object. This was popular in the late 1800s (the early Edmund Husserl, 1859–1938, believed it for a while). But Frege demolished the view with a simple observation: mathematical theories claim there are infinitely many different numbers; yet our minds contain only finitely many ideas. Psychologism cannot make sense of a genuinely infinite realm of objects.
A third strategy, paraphrase nominalism, avoids objects entirely by changing what mathematical sentences mean. For example, “3 is prime” might really say “If there were numbers, then 3 would be prime.” But when you tell your friend “3 is prime,” you do not mean anything about an imaginary “if-then.” You mean it is prime. Treating ordinary math talk as a hidden conditional feels forced. So by the late twentieth century, many philosophers were stuck: platonism seemed right about what words mean, yet wrong about what really exists.
Fictionalism: The Story of Arithmetic
Into this stalled debate stepped a group of thinkers with a playful-sounding idea. A fictionalist about mathematics agrees with the platonist on one crucial point: sentences like “4 is even” really do claim to be about objects. However, the fictionalist denies that those objects exist. Therefore, strictly speaking, “4 is even” is false—just as “Sherlock Holmes lives on Baker Street” is false. But Holmes is true in the story written by Arthur Conan Doyle. Similarly, the fictionalist says, “4 is even” is true in the story of mathematics. That story is not a novel written by a person; it is the network of accepted axioms and definitions that mathematicians build together.
The American philosopher Hartry Field (born 1946) developed one of the most influential versions of fictionalism. Mark Balaguer (a contemporary philosopher) and others have refined the view. They argue that fictionalism gives you the best of both worlds. You keep the platonist’s natural reading of math sentences—they are about objects—without having to believe in a ghostly realm of abstract things. You avoid the implausible rewrites of physicalism, psychologism, and paraphrase nominalism. Math becomes a grand, disciplined make‑believe that works.
But a good story still has rules. Fictionalists must explain why “2 + 2 = 4” is correct while “2 + 2 = 5” is wrong. Their answer: correctness in mathematics is correctness‑according‑to‑the‑story. A sentence is right if it follows logically from the accepted mathematical rules (axioms), just as a claim about Wizarding Britain is right if it fits the Harry Potter books. Some fictionalists even say the story is what would be true if abstract objects existed. That preserves the object‑oriented feel of math without the heavy ontology.
The Big Pushback: Science Needs Real Numbers
Fictionalism sounds clever, but it faces a powerful objection—one pressed by the philosophers W.V.O. Quine (1908–2000) and Hilary Putnam (1926–2016). The indispensability argument says: mathematics is not a separate game you can take or leave; it is woven deeply into our best scientific theories of the physical world. Physics, chemistry, and astronomy use numbers and equations to describe how things move, heat up, or orbit. If we think those scientific theories are true (and they do their job stunningly well), then we seem forced to accept that the math they rely on is true too. You cannot say “gravitation is real” while also calling the numbers in Newton’s equations a fiction. The numbers are part of the truth‑story.
Fictionalists have two main replies. The hard‑road response, pursued by Field, tries to show that science does not actually need numbers. Field spent years rewriting a chunk of Newtonian physics in a language that never mentions numbers or any other abstract object—a task of staggering difficulty, and one that many philosophers doubt can be carried out for all modern science (especially quantum mechanics). The easy‑road reply, developed by Balaguer and others, concedes—for the sake of argument—that math is embedded in our scientific language. But it argues that math serves only as a representational aid, a kind of descriptive map. The map can be useful without being literally true. A city map with a fictional grid can still guide you home. The physical world, on this view, behaves exactly as it needs to for science to succeed; the mathematics is just the colored overlay that helps us talk about it.
The debate rages on. A recent twist asks whether mathematics actually explains physical facts, not just describes them. If biological traits, like the life cycles of periodic cicadas, turn out to depend on prime numbers (which minimize overlap with predators), perhaps math is doing explanatory work that no mere fiction could do. If so, the easy‑road may be in trouble. But fictionalists have responses here too. The fight is deep and unresolved.
Why the Fight Isn’t Over—and Why It Matters to You
So where do we stand? Fictionalists have not delivered a knockout blow to platonism. The epistemological argument against abstract objects is powerful but widely agreed to be inconclusive. The multiple‑reductions argument (that there are too many candidate number‑sequences) also fails to settle the matter. Many fictionalists appeal to Ockham’s razor: the principle that, all else being equal, we should prefer the view that posits fewer kinds of things. Since both platonism and fictionalism account for the same mathematical data, but fictionalism avoids adding abstract objects, the razor seems to favor fictionalism. Yet not everyone accepts that the razor applies here—after all, if abstract objects really existed, they wouldn’t be an “extra” luxury; they’d be the indispensable furniture of reality. The debate, for now, is a stalemate.
You might wonder whether any of this actually touches your life. It does, quietly. The question “What makes a mathematical claim correct?” is a miniature version of bigger questions: “What makes any claim true?” and “When are we justified in saying something exists?” Your smartphone, your class schedule, the science that lands rovers on Mars—all rely on math. If math turned out to be just a story, you would still be able to rely on it, the way you rely on a map. But you might start to see that the map and the territory are not the same. The next time you solve an equation, you might pause and wonder: am I reading a real universe of timeless truths, or am I playing the most powerful make‑believe game ever invented? Both possibilities are astonishing.
Think about it
- If you think numbers don’t exist, does that mean “2+2=4” is just an opinion, like your taste in music? Why or why not?
- Imagine you meet an alien whose math is completely different from ours—maybe 2+2=5 in their system. Would that prove that math is invented by minds, or would our math still be correct in some sense? How could we decide?
- Engineers use math to design safe bridges. If math were merely a very useful fiction, could you still trust the bridge? What does that tell you about the difference between truth and usefulness?
