Yes: some infinities are strictly bigger than others. The set of counting numbers 1, 2, 3, … is infinite, and so is the set of real numbers between 0 and 1, but Georg Cantor proved in 1891 that the second can never be listed against the first. His test for “same size” needs no counting at all: two collections match if every member of one can be paired with a different member of the other and nothing is left over. By that test the counting numbers match the even numbers, the integers and every fraction, but not the real numbers.
Here are two infinite collections. One is strictly bigger than the other, and proving it helped turn a mathematician into a target. For two thousand years thinkers argued over infinity: was it something completed, or only a process that never ends? Then, in the 1870s, Georg Cantor asked how you compare sizes without counting.
A shepherd without number words could still match one pebble to each sheep. If every sheep gets one pebble and no pebbles remain, the collections match. No totals needed, just partners, with nobody left out. Cantor took that simple test somewhere deeply strange.
Counting without numbers
Take the counting numbers, 1, 2, 3, continuing forever, and write only the even numbers underneath. Surely there are half as many; we threw away every other number. But pair 1 with 2, 2 with 4, 3 with 6, and in general pair n with 2n. Every counting number gets a partner, every even number gets reached, and nobody is left over. By the shepherd’s rule the two collections have exactly the same size. An infinite collection can match a proper part of itself; finite collections cannot do that. Cantor treated that as a feature, not a bug.
Now include the negative integers. They stretch forever in both directions, but we can still make a list: 0, 1, −1, 2, −2, zigzagging outward until every integer gets a position. Cantor named this size aleph null, the size of an endless list with nothing missing. Collections of that size are called countably infinite.
Every fraction in a line
Fractions look like a harder case. Between any two distinct fractions sit infinitely many more, and even between 0 and 1 they never run out. Every interval contains them, though they do not fill the line. Surely this collection must be bigger.
The trick is to stop picturing points and start picturing pairs. Put numerators across and denominators down: every positive fraction has a place in this infinite grid. Reading row by row fails immediately, because you never finish the first row. Instead, zigzag along the anti-diagonals. Each sweep is finite, and every cell is reached after finitely many steps. Skip repeated values (2/2 is already 1), start with zero, and alternate each new positive fraction with its negative. Every rational number appears exactly once. Aleph null again.
Something dense throughout the number line fits into one queue. Perhaps all infinities really are the same size?
The diagonal
In 1891 Cantor published his diagonal argument. Here is its familiar decimal version. Take the real numbers between 0 and 1: a half, a third, π − 3, the square root of 2 minus 1. Write their decimals, using trailing zeros when necessary. Could every real number in this interval go into a list?
Suppose someone says yes and hands us that list: first number, second number, third number, continuing forever. Their claim is simple: every real number between 0 and 1 appears somewhere. Now build another decimal. Inspect the first digit of the first entry, then the second digit of the second entry, and continue down the diagonal. Whenever the inspected digit is 1, write 2; otherwise, write 1. Using only ones and twos avoids the trap of decimals with alternative endings, like trailing nines.
Where can our number appear in the list? Not first, because its first digit differs. Not second, because its second digit differs. At any position n it differs from that entry in digit n, so every possible position is ruled out. The list missed a number, and the same works against any proposed list.
There are strictly more reals than counting numbers: two infinities of different sizes. Almost every real is irrational. More strongly, the algebraic numbers are also countable, so almost every real solves no nonzero polynomial with integer coefficients.
The hotel and the gap
David Hilbert, a leading mathematician of his age, made countable infinity feel like somewhere you could check in. Picture infinitely many rooms numbered 1, 2, 3, onward, and every room is occupied. A new guest arrives. The manager asks every current guest to move one room along: room 1 to room 2, 2 to 3, room n to room n + 1. Nobody loses a room, but room 1 becomes empty, and the full hotel welcomes another guest.
Now infinitely many newcomers arrive, one for each counting number. Move every existing guest from room n to room 2n. All the odd rooms are free, and the newcomers fit with nobody left outside. But one guest for every real number? No rearrangement can accommodate that crowd. There simply is no matching.
Cantor proved something stronger: the collection of all subsets of any set is strictly bigger than the original set. Repeat that operation and you get an endless tower of larger infinities, never a largest one.
But what happens at the first gap? Aleph null is the smallest infinite size, and the real numbers have a larger size. Is any size strictly between them, too big to list but too small to match the line? Cantor believed not. This is the continuum hypothesis, and he struggled for years without proving it. In 1900 Hilbert placed it first on his famous list of unsolved problems for the new century.
Gödel’s result, published in 1940, showed that the hypothesis cannot be disproved from the usual axioms of set theory, assuming those axioms are consistent. In 1963 Paul Cohen showed it cannot be proved from them either, under the same assumption. The hypothesis is independent: if those axioms are consistent, adding either answer preserves consistency. The axioms leave the gap undecided, and choosing an answer means adopting something more.
Paradise
Not everyone welcomed Cantor’s ideas. Leopold Kronecker, his former teacher and a powerful Berlin mathematician, fiercely opposed them. He is remembered for saying that God made the integers and everything else was humanity’s work. Their conflict was bitter, but claims that he systematically blocked Cantor’s papers and a Berlin appointment go beyond secure evidence.
Cantor remained at the smaller University of Halle. From 1884 he experienced recurring mental health crises. Neither criticism nor one mathematical problem can adequately explain his illness, and he was not simply driven out of mathematics. He died in a sanatorium in 1918, amid wartime deprivation. Hilbert’s defense, published in 1926, declared that no one shall expel us from the paradise that Cantor has created.
Today pairing is foundational, and students learn diagonalization early. Related arguments reveal uncomputable problems and limits to formal proof. Infinity was not merely big: finite, checkable arguments could reveal its structure, and questions that our chosen axioms cannot settle.
