How Many Real Numbers Are There?

Imagine you’re counting. You start: one, two, three… and you keep going. You know there’s no end—there are infinitely many whole numbers. But here’s a stranger question: how many real numbers are there? Not just the whole numbers, but every decimal, every fraction, every number like π or √2 or 0.3333… — all the points on a number line.

The mathematician Georg Cantor discovered something weird in the late 1800s: there are more real numbers than there are whole numbers. Not just “more” in the sense that one infinity is bigger than another—he actually proved it. The set of real numbers is bigger than the set of whole numbers. The whole numbers are “countably infinite,” and the real numbers are “uncountably infinite.”

This leads to a natural question: just how many real numbers are there? Is it the next infinity after the whole numbers? Or something much larger?

This is the Continuum Hypothesis (CH). It says: there is no infinity between the size of the whole numbers and the size of the real numbers. In other words, the real numbers are exactly the next infinity up.

For over 140 years, mathematicians have been trying to figure out whether this is true or false. And here’s the strangest part: we now know that using the standard rules of mathematics, you cannot prove it true, and you cannot prove it false. The question is what philosophers call an “independent” statement—like asking whether a rule exists in a game where the rulebook doesn’t mention it.

This article is about what that means, and what philosophers and mathematicians have done about it.

The Discovery That Changed Everything

When Cantor first proposed the Continuum Hypothesis, people tried to prove it or disprove it. They failed for decades. Then, in the 1930s and 1960s, two mathematicians showed why they had failed—and their discoveries changed how we think about mathematics itself.

First, Kurt Gödel showed that you could add CH to the normal rules of mathematics (a system called ZFC) and everything would still be consistent. No contradictions would arise. Then Paul Cohen showed the opposite: you could also add the opposite of CH to ZFC, and that would be consistent too.

Think about what this means. Imagine you’re playing chess, and someone asks: “Can a pawn capture a piece by moving backward?” You check the rules. The rules don’t say anything about this. So some people could decide that yes, you can, and others could say no, and both groups could play perfectly consistent games of chess, as long as they added their own rule. The question isn’t settled by the original rules—it’s “independent.”

CH is like that. The standard rules of mathematics (ZFC) don’t settle it. You can have a mathematical universe where CH is true, and another where it’s false, and both are perfectly logical.

Three Ways of Asking the Question

Philosophers noticed something interesting about CH. When mathematicians talk about it, they actually mean three different things—and these three versions are equivalent under the normal rules, but they come apart when you start thinking about “definable” sets.

The first version is the interpolant version: there is no set of real numbers whose size is strictly between the whole numbers and all real numbers. The second is the well-ordering version: any way of arranging the real numbers in order has to stop before a certain point. The third is the surjection version: you can’t map the real numbers onto a certain larger infinity.

Why does this matter? Because when you look at “definable” sets—sets you can actually describe with a finite definition—these three versions behave differently. For the first two versions, we know that CH holds for all definable sets. For the third version, it gets more complicated, and philosophers still argue about whether this tells us anything about CH itself.

There’s an asymmetry worth noticing. If you find a definable counterexample to CH—a definable set of real numbers whose size is between the whole numbers and all reals—that’s a real counterexample. But no matter how many definable versions of CH you prove true, you never prove CH itself. As one philosopher put it, this approach could refute CH but it could never prove it.

A Strange Model Where CH Fails

Around the year 2000, Hugh Woodin constructed a very special mathematical model—call it a “possible universe” of mathematics—where CH is false. What’s special about this model is that it’s “maximal” in a certain sense: it contains all the things that could possibly exist, based on the rules, and nothing more.

In this model, the size of the real numbers turns out to be the second infinity after the whole numbers—not the first, but the second. This is written as 2^(ℵ₀) = ℵ₂. (Don’t worry about the symbols; the important thing is that it’s larger than CH would predict.)

Woodin argued that this model is special—that there are good reasons to think it captures something true about mathematics. But not everyone agrees.

The Multiverse Response

Some philosophers look at the independence results and draw a different conclusion. They say: there isn’t a single “true” universe of mathematics. Instead, there’s a multiverse—many different mathematical universes, each equally valid. CH is true in some of these universes and false in others, and neither answer is “the” answer.

One version of this view is the generic multiverse, which includes all universes you can reach by something called “forcing” (the technique Cohen invented to show CH was independent). According to this view, CH is indeterminate—it doesn’t have a truth value, any more than “the king is on the left side of the board” has a truth value in chess without knowing which side you mean.

But this view faces problems. It turns out that if certain mathematical conjectures are true (the Ω Conjecture and others), then the generic multiverse view collapses in a strange way: truth in the multiverse becomes reducible to truth in a small fragment of mathematics, which violates principles that seem fundamental to how we think about the universe of sets. This is technical, but the upshot is that the multiverse view has a hard time surviving its own logic.

The Case for CH

There’s also a parallel argument that CH might be true. If you work at a different level of mathematical complexity (a level called Σ²₂ instead of the level Woodin worked with), you can construct a similar argument: there are “good” theories, and all of them imply CH. There’s even a maximal theory, just like in Woodin’s case.

So we have two parallel arguments, one pointing toward CH and one pointing away from it. Which one wins? It depends on which level of mathematical complexity you think is more fundamental. Philosophers haven’t settled this.

The Ultimate Inner Model

There’s a third approach that might settle everything. Mathematicians have been building “inner models”—small, well-behaved universes inside the larger universe of sets. The simplest one is called L, constructed by Gödel. In L, CH is true. But L can’t accommodate the largest kinds of infinities (called large cardinals) that many mathematicians accept.

Recently, work by Woodin and others suggests there might be an “ultimate” inner model—call it L^Ω—that can accommodate all large cardinals, that would be “effectively complete” (meaning it would answer almost every question), and that would settle CH one way or another.

The catch: there are multiple candidates for this ultimate model, and they disagree about CH. Some would make CH true; others would make it false. So the question becomes: which of these candidates has the right structure? This is what researchers are working on now.

Why This Matters

You might wonder why anyone cares. The Continuum Hypothesis seems abstract—it’s about infinities that no one will ever count. But it’s become a test case for a deeper question: can every mathematical question have an answer?

Some philosophers think yes—that given enough work, we’ll find new axioms that settle CH. Others think no—that some questions are genuinely undecidable, and we have to live with that.

This debate matters beyond mathematics. It touches on what it means for something to be “true” at all. If there can be statements that are neither true nor false—statements that are perfectly meaningful but have no answer—then truth is stranger than we thought.

For now, nobody really knows whether CH is true or false. The smartest people in the world disagree. And that’s part of what makes it fascinating: a simple question about counting has opened up the deepest questions about what mathematics is.

Key Terms

Term	What it does in this debate
Continuum Hypothesis (CH)	The claim that there is no infinity between the size of the whole numbers and the size of the real numbers
ZFC	The standard set of rules for modern mathematics; CH is independent of ZFC
Independent	A statement that can neither be proved nor disproved from a given set of rules
Generic multiverse	The collection of all mathematical universes reachable by forcing; a way of being pluralist about CH
Ω-logic	A very strong logic that is “well-behaved” under forcing; used to build arguments about CH
Inner model	A small, well-understood universe inside the full universe of sets (like L, Gödel’s model)
Large cardinal axioms	Strong assumptions about very large infinities; accepted by many set theorists
ℙ_max	A special construction that produces a model where CH fails and the real numbers have size ℵ₂

Key People

Georg Cantor: The mathematician who first discovered that there are different sizes of infinity and proposed the Continuum Hypothesis.
Kurt Gödel: Showed that CH could be added to standard mathematics without causing contradictions; built the inner model L where CH is true.
Paul Cohen: Showed that the opposite of CH could also be added without contradictions; invented the technique of “forcing.”
Hugh Woodin: A contemporary set theorist who constructed the ℙ_max model where CH fails and argued that CH is probably false; also working on the “ultimate inner model.”
W. Hugh Woodin’s critics: Philosophers and mathematicians who argue that the multiverse view is more plausible, or who defend the case for CH.

Things to Think About

If CH is independent of the normal rules of mathematics, does that mean it “doesn’t have an answer”? Or does it just mean we need better rules?
The multiverse view says there are many equally valid mathematical universes. How is that different from saying mathematics is just whatever we decide it is? Is there a way to tell the difference?
Suppose someone builds a machine that could check every mathematical proof. Would that machine ever find the answer to CH? If not, does that mean the question is meaningless?
Parallel arguments exist for both CH and its opposite, depending on which level of mathematical complexity you look at. How should we decide which level is the “right” one to look at?

Where This Shows Up

Computer science: Independence results like CH show the limits of what can be proved by any computer program—there will always be true statements it can’t verify.
Physics: Some physicists have wondered whether the size of the continuum might connect to the structure of space-time (is space continuous or discrete?).
Everyday reasoning: The debate about CH is a real example of something that happens in ordinary arguments: two people can agree on all the facts, disagree on a conclusion, and neither be “wrong” by the rules they share. The question is what to do when that happens.