Two statements that must not be confused
Modern science ties information tightly to the physical world. Every real record needs a medium, and information processing obeys thermodynamics and quantum mechanics.
But no experiment has shown that matter, energy, space and time are nothing but information. “Information is physical” and “physics is only information” sound similar, yet the second is a much stronger claim.
1948Shannon mathematises information
kBT ln 2minimum cost of erasing one bit
it from bitWheeler’s hypothesis, not proof
0 1
medium → distinction → informationOne bit, many physical forms
A bit is the smallest classical unit for distinguishing between two possibilities: 0 or 1, yes or no. It is not a particle like an electron. It is an abstract way to describe a choice.
The same bit can be stored as two voltage levels, two magnetic directions, the presence or absence of charge, or two states of light. The medium can change while the logical information remains the same.
In 1948 Claude Shannon showed how to measure a source’s uncertainty and how much information can be transmitted reliably through a noisy channel. In his theory, “information” does not mean knowledge, truth or meaning. The mathematics does not decide whether a message is a discovery, a lie, a poem or random symbols.
The first limit: Shannon information, quantum information, biological information and everyday “meaning” are not one concept with several names.
Theory confirmed experimentallyErasing information really produces heat
Rolf Landauer showed that logical operations have physical consequences. If a memory can initially contain either 0 or 1 and an operation always resets it to zero, two possible histories are compressed into one result. Looking only at the final state, the initial state cannot be recovered.
Qmin = kBT ln 2minimum heat released when one bit is ideally erased
At room temperature this is about 3 × 10−21 joules — tiny, but not zero. In 2012 Antoine Bérut’s team experimentally verified Landauer’s bound using a microscopic particle trapped in a double-well potential representing 0 and 1.
What does this prove? Processing information is a physical process. It does not prove the converse claim that everything physical is only information.
Established physicsA qubit is not a smaller bit
A qubit can be prepared in a superposition, yet measurement cannot extract an arbitrary amount of classical data from it. Nor can an unknown quantum state be copied perfectly and deterministically. Wootters and Zurek established this in their 1982 no-cloning theorem.
This is not a flaw in today’s devices. It follows from the structure of quantum mechanics. Quantum information can be teleported using entanglement and classical communication, but no perfect copy of the unknown state can be kept along the way.
Strong theoretical resultCan information explain quantum mechanics?
Chiribella, D’Ariano and Perinotti showed that quantum theory’s formalism can be reconstructed from principles governing the preparation, transformation and reading of information — including a purification principle.
This matters: quantum mechanics can be understood as a theory of limits on information processing. It still does not establish that information is the basic “substance” of the world. The fact that geometry describes planetary motion does not mean a planet is made of geometry.
Serious but restricted modelWhen a surface says more than a volume
Black-hole physics supplied an extraordinary clue: a black hole’s entropy is proportional to its horizon area, not the volume inside. This helped inspire the holographic principle — the possibility that a gravitational region can be encoded on its boundary.
The sharpest laboratory is the AdS/CFT correspondence. Within it, the Ryu–Takayanagi formula links entanglement entropy in the boundary theory to the area of a corresponding surface in the geometric “bulk”. Later work showed that reconstructing bulk information has a structure resembling quantum error correction.
This is where the claim that space may “emerge from entanglement” comes from. It has strong mathematical support in particular models. We do not know whether the same mechanism describes our Universe: AdS geometry differs from observed cosmology, and no experimentally confirmed theory of quantum gravity yet settles the question.
Philosophical-physical hypothesis“It from bit”: from a question to reality
John Archibald Wheeler condensed the radical intuition into the phrase it from bit. In his proposal, physical properties acquire definiteness through elementary yes-or-no distinctions. Quantum mechanics was central to the idea because the kind of measurement performed helps determine which result can be registered.
Wheeler did not provide a complete theory with new predictions that distinguish it experimentally from standard physics. It was a research programme and a philosophical guide, not a finished proof.
Two stronger claims must be kept separate: that the Universe is a computer, and that we live in a simulation. Neither follows automatically from Shannon’s theory, Landauer’s principle or quantum mechanics.
Three claims — three different answers
| Claim | State of knowledge |
|---|
| Physical systems can be described using information. | Yes |
| Storing and processing information requires physical processes. | Yes |
| Matter, energy and spacetime are nothing but information. | Unknown |
The difference between the second and third statements is the heart of the debate. A map must be stored somewhere and the cost of processing it can be measured. That does not mean the territory is made of the map.
What could settle the argument?
For “reality is information” to become a testable physical theory, it must offer more than a different language. It would need predictions that differ from existing theories: a characteristic limit to spacetime continuity, a new quantum effect, or a measurable trace of a specific informational structure.
If “the world is matter” and “the world is information” lead to exactly the same observations, no experiment can choose between them. The dispute remains ontological — about how equations are interpreted, rather than what a detector records.
Research on quantum gravity, holography, information thermodynamics and reconstructions of quantum theory could change that. For now, these fields provide important clues, not a final verdict.
Conclusion: the bit has not replaced the “it”
Information is not a disembodied add-on to physics. Every record has a medium, operations on it obey thermodynamics, and the quantum world imposes fundamental rules on what may be done with information. In some theories, informational relations are even tightly linked to geometry.
But the reverse relation has not been established — that matter, energy, space and time are merely manifestations of bits. “It from bit” remains a fertile idea at the boundary of physics and philosophy: serious enough to investigate, and still too weakly confirmed to present as fact.
In one sentence: physics has shown that information is tied to matter; it has not shown that matter is only information.
Scientific sources
- Claude E. Shannon, “A Mathematical Theory of Communication”, Bell System Technical Journal (1948)
- Rolf Landauer, “Irreversibility and Heat Generation in the Computing Process”, IBM Journal of Research and Development (1961)
- Antoine Bérut et al., “Experimental verification of Landauer’s principle linking information and thermodynamics”, Nature (2012)
- William K. Wootters, Wojciech H. Zurek, “A single quantum cannot be cloned”, Nature (1982)
- Giulio Chiribella, Giacomo M. D’Ariano, Paolo Perinotti, “Informational derivation of quantum theory”, Physical Review A (2011)
- Jacob D. Bekenstein, “Black Holes and Entropy”, Physical Review D (1973)
- Juan Maldacena, “The Large N Limit of Superconformal Field Theories and Supergravity” (1997/1998)
- Shinsei Ryu, Tadashi Takayanagi, “Holographic Derivation of Entanglement Entropy from AdS/CFT”, Physical Review Letters (2006)
- Ahmed Almheiri, Xi Dong, Daniel Harlow, “Bulk Locality and Quantum Error Correction in AdS/CFT”, JHEP (2015)
- John Archibald Wheeler, “Information, Physics, Quantum: The Search for Links” (1990)
The evidence labels refer to specific scientific results. The ontological claim that “everything is information” currently has no decisive experimental test.
← Back to the Science room