“It from Bit”: Is Information the Stuff of Reality?

In 1989 the physicist John Archibald Wheeler — the man who coined "black hole" and supervised Feynman — stood up and proposed that physics had the hierarchy backwards. Matter and energy, he said, are not fundamental. Information is. His slogan: "it from bit." "Every it — every particle, every field of force, even the spacetime continuum itself — derives its function, its meaning, its very existence entirely from the apparatus-elicited answers to yes-or-no questions, binary choices, bits." Reality, in his participatory universe, sits in "indeterminate limbo" until a question is asked — like his parable of the game of twenty questions, where the players realize the word "wasn't in the room when I came in": the answers create the thing being asked about.

Now split Wheeler's provocation into two halves — one settled, one wide open. The settled half: "information is physical." In 1961 Rolf Landauer at IBM proved that erasing one bit of information has a minimum thermodynamic cost — about 3 zeptojoules at room temperature — because erasure squeezes two possibilities into one, and the lost entropy must go somewhere as heat. It was verified experimentally in 2012. Information is not an abstraction floating above physics; it is made of physics. Your quantum computer obeys this too: every quantum gate is reversible (unitary), so the energy cost of computation enters only where irreversibility enters — initialization and measurement.

The contested half: "physics is information." Critics reply — Landauer's own dictum cuts both ways — that information "makes no sense in the absence of something to be informed," and ask the awkward question: "Matter can clearly exist without mind, but where do we see mind existing without matter?" Is "it from bit" profound physics or poetry? The honest answer: it's an open research program, not a result. And note the irony — QBism (last lesson) agrees that information is central but locates it in the agent, not in the furniture of the world.

But here is where it gets serious: instead of arguing about the slogan, physicists tried to rebuild quantum theory from information principles — the way Einstein rebuilt physics from two crisp postulates of relativity. Lucien Hardy (2001) showed that quantum theory follows from five "reasonable axioms" about how systems combine and transform; the axiom that makes it quantum rather than classical is continuity — delete that one word and you get ordinary probability back. Chiribella, D'Ariano, and Perinotti (2011) derived quantum theory from six informational principles, crowned by the purification principle: "ignorance about a part is always compatible with maximal knowledge of a whole" — every mixed state is part of some pure entangled state. Their punchline: "Quantum theory is the only standard theory of information compatible with the purity and reversibility of physical processes."

The most radical descendant is constructor theory (Deutsch and Marletto, 2015): stop writing physics as "initial conditions plus laws of motion" and write it as statements about which tasks are possible and which are impossible. On this view, quantum information is a special case of "superinformation" — information with the properties that you can't copy all of it at once and every transformation is reversible — and from superinformation fall out the no-cloning theorem, complementarity, and objective unpredictability as theorems. Status flag, stated honestly: as of 2026 it has produced no confirmed prediction that only it makes. It is Deutsch's current great project — philosophy as a research program, in real time.

So what does a quantum computer mean on this view? It is physics doing information processing at the deepest level the theory allows: the machine manipulates superinformation — unclonable, complementary, reversibly transformable — rather than classical bits. Wheeler's slogan becomes a design principle. And it closes the circle of this module: Deutsch could read computation as a physical notion in 1985 because, at bottom, computation and physics are made of the same stuff. Whether that stuff is "it" or "bit" is the question Wheeler left us.

Common myth: "'It from bit' is established physics — scientists proved reality is made of information." No: the settled part is Landauer's "information is physical." The "physics is information" direction is contested philosophy plus an active (unfinished) research program.