Are Some Infinities Bigger Than Others? Cantor's Discovery
December 1873: Two Infinities That Are Not the Same
In a small German town in late 1873, Georg Cantor (1845–1918) sat late into the night, trying to pair up two kinds of numbers. He did not yet know he was about to tear open mathematics.
Cantor was comparing the set of natural numbers (1, 2, 3, …) with the set
of real numbers — every point on an unbroken number line, including
decimals that go on forever.
Both sets are obviously infinite. But Cantor asked a sharper question: can you
put them into a perfect one‑to‑one correspondence? That means pairing
each natural number with exactly one real number, with no leftover real numbers
on either side. If such a perfect match exists, the two sets have the same
cardinality — the same size.
You can think of a party where every guest finds exactly one chair and no chair stays empty. If that works, you know there are exactly as many guests as chairs, without counting anyone. Cantor tried to do the same for infinite sets.
He found something shocking. Even though both sets are infinite, the real numbers cannot be matched one‑to‑one with the naturals. No matter how you try, there will always be real numbers left unmatched. The reals are, in a profound sense, more numerous. The natural numbers are countable — you could, in principle, list them one after another forever. The real numbers are uncountable — any attempt to list them always misses almost all of them.
Cantor’s proof, later called the diagonal argument, was elegantly simple. Imagine writing down every real number between 0 and 1 as an infinite decimal. Then build a new decimal by walking down the diagonal of this list and changing each digit. The new number cannot be anywhere in the list, because it differs from the first number in the first digit, from the second in the second, and so on. The list was supposed to contain them all, yet you just produced one that was missing. So the reals cannot be captured in a countable list.
Infinity had just split in two.
Cantor’s Dangerous Question: Is There an Infinity Between?
Cantor did not stop there. He quickly realized there might be a whole ladder of infinities, each larger than the last. He named the smallest infinite cardinality ℵ₀ (“aleph‑null”), the size of the natural numbers. The very next size he called ℵ₁.
He then asked the question that would obsess mathematicians for over a century: exactly where does the size of the real numbers sit on this ladder? He suspected it was exactly ℵ₁ — that there is no size of infinity squeezed between the countable and the continuum. This guess became the famous Continuum Hypothesis (CH). In plain words: every infinite set of real numbers is either countably infinite (like the naturals) or as big as the whole continuum.
Proving the CH, however, turned out to be maddeningly difficult. Cantor himself tried for years and never succeeded. David Hilbert (1862–1943), the most influential mathematician of his generation, called it the first problem on his famous list of 23 unsolved challenges for the twentieth century. Thousands of pages were written. And yet no one could decide whether Cantor’s intuition was right.
Behind the scenes, a deeper problem was threatening to make the whole discussion meaningless.
When “The Set of All Sets” Exploded
Before Cantor, many mathematicians assumed that any property you can describe collects things into a set. The property “is a golden retriever” picks out the set of all golden retrievers. The property “is a number” picks out the set of numbers. So why not “is a set”? That would give you the set of all sets.
Bertrand Russell (1872–1970) spotted a fatal flaw in this rosy picture — now called Russell’s Paradox. Consider the property “is a set that does not contain itself.” Most ordinary sets have this property (the set of all chairs is not itself a chair, so it does not belong to itself). But now ask: does the set of all sets that do not contain themselves belong to itself? If it does, then by its own definition it should not; if it does not, then it should. The contradiction snaps the concept.
The naive idea that any description makes a set had to be abandoned. Pioneers like Ernst Zermelo (1871–1953) and Abraham Fraenkel (1891–1965) built a careful system of rules — axioms — that tell you precisely which sets are allowed. The standard system, called ZFC (Zermelo‑Fraenkel set theory with the Axiom of Choice), carefully restricts how sets can be formed. Some collections, such as the collection of all sets or all ordinal numbers, are now called proper classes — they are too big to be treated as sets without inviting paradox. ZFC became the safe foundation for nearly all of modern mathematics.
The universe of all sets that ZFC permits can be pictured as a tower, denoted V, built layer by layer. The bottom layer is the empty set. The next layer is the set containing the empty set. Keep taking power sets and unions, climbing through the ordinals, and you never stop — you get a rich, endless hierarchy.
But as mathematicians would soon discover, this safe tower is also deeply incomplete.
Gödel’s Invisible Cage: A Universe Where the CH Holds
Kurt Gödel (1906–1978) delivered the first seismic shock in 1938. He showed that if the ZF axioms (ZFC without the Axiom of Choice) are consistent — that is, they do not lead to a contradiction — then you can safely add the CH and the Axiom of Choice without creating any new inconsistency. In other words, you cannot disprove the CH from the usual axioms of set theory.
He did this by building a special mini‑universe inside the tower V. He called it L, the constructible universe. In L, at each stage, instead of taking all subsets of the previous layer, you only take the subsets that can be defined precisely by a formula. This trims the universe down so severely that it forces the real numbers to line up in a neat, predictable way. In L, there is no uncountable set of reals with a size smaller than the continuum — the CH is true.
Gödel’s construction was like showing that a particular kind of tidy room could exist inside a messy house without breaking the house’s walls. The CH might not be forced on us, but it is definitely not ruled out.
For a quarter‑century, many hoped the CH could still be proved true. The next discovery shattered that hope.
Cohen’s Explosion: Forcing the CH to Fail
In 1963, Paul Cohen (1934–2007) invented a technique called forcing that let mathematicians expand a universe. Starting with a small, countable model of ZFC (which Gödel had shown exists if ZFC is consistent), Cohen showed how to carefully add new subsets of the natural numbers that live outside the original model, without wrecking the axioms.
The trick is to add the new set generically — bit by bit, using only partial information that never fully reveals the final set. Because the construction is so controlled, the enlarged model still satisfies all of ZFC. But in the new model, there are suddenly many more real numbers — so many that the continuum becomes, say, ℵ₂ (the second uncountable cardinal) or even larger. The CH fails dramatically.
Cohen’s result, together with Gödel’s, proved that the Continuum Hypothesis is undecidable from ZFC. You cannot prove it true, and you cannot prove it false. It is an independent sentence, floating free of the usual rules.
Cohen’s forcing method opened a floodgate. Over the following decades, mathematicians used it to show that hundreds of other natural‑sounding questions — about infinite games, the structure of the real line, even problems in algebra — are also undecidable from ZFC alone.
Why It Still Matters: The Never‑Ending Story of Infinity
What do we do when a question as basic as “how many points are on a line?” has no answer inside standard mathematics? The undecidability of the CH tells us that ZFC, the gold‑standard foundation, is not enough to settle every reasonable mathematical question. Some philosophers and mathematicians think this means the CH simply has no definite truth value — it is like asking whether a colour is heavier than a sound. Others, following Gödel, believe the CH has a true answer; we just need to discover new axioms that go beyond ZFC.
The most promising candidates for new axioms involve large cardinals — infinities so staggeringly huge that their existence cannot be proved from ZFC alone. Think of weakly inaccessible cardinals, measurable cardinals, or Woodin cardinals — each level far bigger than anything ZFC can guarantee. These gigantic infinities act like road signs on the interpretability highway: if a statement can only be proved by assuming a very large cardinal, you know it was deeply hidden. While large cardinals do not yet settle the CH, they give us a framework for comparing the strength of theories and for deciding many other open questions (such as whether every projective set of reals is determined — yes, if Woodin cardinals exist).
So why should a twelve‑year‑old care? Every time you use a real number — to measure a length, to graph a curve, to send a rocket to Mars — you are leaning on a continuum whose exact size remains a mystery. Mathematics is not a finished museum of dusty facts; it is a living argument. The rules that govern infinity are still being written. And maybe one day, someone reading this article will add a new rule that finally tells us how many points there really are on a line.
Think about it
- If you could magically add one new axiom to mathematics that would decide the Continuum Hypothesis, would you want the answer to be “yes” (no missing size) or “no” (there is a size between)? Why would you lean that way?
- Some mathematicians think the CH is neither true nor false — it’s just a question that has no answer in any possible universe. Can you think of a question in your own life that might also have no right answer, no matter how hard you look?
- Imagine discovering a brand‑new size of infinity, bigger than all the alephs we know. Would that create a dangerous paradox, or could we just add it to the family and keep going?
