ultrafinitism.org

Some numbers run out.

You can write 10↑↑10 in six characters. It is a tower of ten 10s, each the exponent of the one below. Nobody has ever counted to it, and nothing in the universe could hold its digits.

Ultrafinitism asks whether a number that can be named but never reached is a number at all.

The ladder

Pick a number. Each is put through three tests: can you count to it, can you write out all its digits, and can you name it.

    Three dots

    The same question shows up in small things. We write 1/3 = 0.333… and the dots do a lot of work. Is there a finished, infinite object behind them? Or only a rule that produces another digit whenever someone asks, and has never been asked for all of them?

    Roughly 40 digits of π fix the circumference of the observable universe to within the width of a hydrogen atom. The digits past that are, as far as measurement goes, never going to differ from anything.

    The people

    This is a serious position in the foundations of mathematics. Finitists reject completed infinities. Ultrafinitists go one step further and doubt the very large finite numbers too.

    • Alexander Yessenin-Volpin first argued for it, asking what it means for a number to exist if nobody could ever write it.
    • Rohit Parikh gave it formal shape with the idea of feasible numbers.
    • Edward Nelson, a working mathematician with deep results, doubted that exponentiation is total.

    The questions run into complexity theory, bounded arithmetic, and the physical limits of computation.

    Say something

    You don't have to agree. A good objection beats a good nod.

    Find us on Bluesky →