Author’s Note: Hawking’s Paradox: Answered by Two Theories consists of 18 chapters divided into four parts. This preprint presents Preface and Chapter One of the first theory. Interested publishers are welcome to contact the author through this website.

Hawking's Paradox
Answered by Two Theories

Juliet Zhong

Independent Researcher | London, United Kingdom | July 2026 |

ORCID: 0009-0006-5099-3671 | AI research Tool: Claude, ChatGPT

[Preface]

Cosmology and A Fish

In his masterpiece A Brief History of Time, Stephen Hawking opens with a story about a man lecturing on cosmology. When he finished, an old woman in the back stood up. ‘What you have described is nonsense,’ she said. ‘The world is a flat plate resting on the back of a giant tortoise.’ The wise man smiled indulgently. ‘And what is the tortoise standing on?’ ‘You are very clever,’ the old woman said, ‘but it is tortoises all the way down.’

Hawking wrote it as a gentle joke about the limits of folk cosmology. But he did not notice that he was also describing himself.

Hawking spent fifty years asking where the information goes when a black hole evaporates. He assumed it had to be somewhere in the radiation, or somewhere in the remnant, or somehow encoded in the geometry of the evaporating spacetime. He looked in every corner of the room he knew. He never considered that the information had left the room entirely—that the room he was searching in was not the only room, and that what looked like a paradox from inside it was not a paradox at all, viewed from the outside.

Consider there is a fish in the ocean. The fish has never seen a fishing net. One day, the water above it darkens in a strange grid pattern, something descends, and the fish is caught. The fish is astonished: ‘Wow, what a black hole! It swallowed everything!’ From the fish's perspective, something inexplicable has happened: an invisible force materialized from the surface of the sea, captured it without warning, and defied every principle of underwater physics it had ever known. It does not make sense. The fish calls it a paradox. The fisherman calls it Tuesday.

Hawking's information paradox is the fishing net. From inside the three-dimensional material domain that physics has always called the universe, the disappearance of information into a black hole is a genuine impossibility, a violation of quantum mechanics with no known resolution. Hawking was right about that. What he did not have, and what this book provides, is the frame outside the water: the cross-dimensional ledger that records what modern physics can only report as missing.

The argument proceeds in three parts. The first establishes the accounting tool—the structural node, the only physical quantity in 3D that satisfies every constraint for a conserved variable that crosses a dimensional boundary. The second applies it to black holes: the horizon redefined, the radiation reidentified, the information paradox answered. The third follows the migrating content to its destination—the consciousness field, where arrival is not storage, where the three great unsolved problems of modern physics (entropy, black holes, and consciousness itself) turn out to be three descriptions of the same process.

The fish eventually learns what a net is. Hawking did not. He spent a lifetime studying that net yet never knew—the answer was above the water.

Chapter One: Hawking's Paradox

In modern physics, there is a famous unsolved puzzle known as Hawking's paradox.

In the autumn of 1973, Stephen Hawking set out to prove that black holes do not glow. He was thirty-one years old, already unable to write without assistance, already composing the elaborate geometric arguments that would make him famous in his head rather than on paper, already occupying the kind of position in British theoretical physics that made other physicists pay attention when he spoke. The motor neurone disease had arrived in 1963, during his first year as a graduate student at Cambridge, when he had been a somewhat directionless and, by his own later account, underachieving young man who had coasted through Oxford on a natural facility for physics without ever working especially hard at it. The diagnosis concentrated everything. He had been given two years to live. The knowledge that time was finite—more finite than the finite time available to everyone—did not break him. It did, by his own account, the opposite: it gave every hour of work a weight it had not previously carried, a consequence it had not previously possessed. He worked. He survived the two years. He kept working. He lived for fifty-five more years.

The physics he produced in them was shaped in ways both obvious and subtle by the constraint the disease imposed. He worked at Cambridge, in the Department of Applied Mathematics and Theoretical Physics. For the specific kind of physics Hawking was doing—the global topology of spacetime, the large-scale causal structure of cosmological solutions to Einstein's field equations—geometric thinking was not a substitute for the real thing. It was the real thing. The ability to see the shape of a spacetime as a whole, to understand its causal structure the way one understands the layout of a city from above, was exactly the ability the questions required, and Hawking possessed it in a degree that had no obvious relationship to the method of composition he had been forced to develop by necessity. Since he could not fill blackboards with equations, he learned to perform calculations geometrically, in pictures, rotating and manipulating mental representations of spacetime that he carried entirely in his head. It was with this geometric intuition that he turned, in the autumn of 1973, to the question of black hole thermodynamics.

He expected to confirm that black holes were cold. Silent. Perfectly dark.

In the formal language of physics, this expectation was expressed as the statement that black holes had no temperature—that the concept of thermal emission, which applies to any object in thermal equilibrium with a surrounding radiation field, simply did not apply to something from which nothing could escape. Jacob Bekenstein, a young Israeli-American graduate student at Princeton, had been arguing for the previous two years that black holes must have entropy—that the second law of thermodynamics required it—and Bekenstein's argument was logically compelling in a way that made most physicists uncomfortable rather than convinced. If a black hole has entropy, it must have temperature. If it has temperature, it must radiate. And if it radiates, things come out of it, which general relativity said was impossible. The argument was too clever, the conclusion too paradoxical. Most physicists, including Hawking, believed Bekenstein had made a category error—had mistaken a formal mathematical analogy for a physical reality. The calculation Hawking undertook in the autumn of 1973 was, in his own understanding of it, the calculation that would settle the matter. It would show that the temperature was zero. Bekenstein would be wrong, gracefully wrong, wrong in the way that produces insight rather than embarrassment, but wrong.

The calculation refused to cooperate.

What Hawking was doing was technically demanding in a way that required him to operate at the intersection of two frameworks that had never been seriously joined: quantum field theory, the language in which particle physics was written, and the curved spacetime geometry of general relativity, the language in which gravity was written. The marriage of these two frameworks was known to be necessary—any complete theory of nature would have to treat gravity quantum mechanically—and known to be ferociously difficult. Hawking did not attempt the full marriage. He did something more modest: he asked what quantum fields would do in the presence of a classical black hole, treating the spacetime geometry as fixed and using the full machinery of quantum field theory to calculate how the fields propagated through it. This semiclassical approach had obvious limitations, but it was valid in the regime that mattered: for black holes large compared to the Planck scale, where quantum gravity corrections were negligible.

However, when applying this approach, he found something he had not expected and, initially, did not believe.

The vacuum of quantum field theory—the state containing no particles—is not the same for all observers. In flat spacetime, the vacuum is unambiguous: every inertial observer agrees on what it is and how many particles it contains, which is none. In curved spacetime, this unanimity breaks down. An observer freely falling through a black hole's event horizon, whose local experience of spacetime is smooth and unremarkable—exactly as the equivalence principle requires—defines the vacuum in terms of modes that make sense in their own locally flat reference frame. A stationary observer far from the black hole defines the vacuum differently, in terms of modes appropriate to the asymptotic flat spacetime far from the hole. When the distant observer calculates what the freely-falling observer's vacuum state looks like in terms of their own modes, the answer is not zero particles. It is a thermal distribution of particles, at a specific temperature inversely proportional to the black hole's mass, propagating outward from the horizon to infinity.

Hawking found this at every stage of his calculation, in every form of the analysis. He checked it against the results other people had obtained using different methods. He looked for the error, because the conclusion was impossible. The error was not there. He wrote up the result and sent it to Nature, which published it in March 1974. The black hole radiates. The temperature is real. Bekenstein was right.

The paper was received by the physics community with something between astonishment and relief. The astonishment was obvious: this was the kind of result that reorganized the field, the kind of calculation whose implications would take years to work out. The relief was subtler, and to understand it requires a brief account of what Bekenstein had done in the two years before Hawking's calculation, and why it had made everyone so uncomfortable.

Jacob Bekenstein was a graduate student at Princeton working under John Wheeler, who had been one of the central figures in bringing general relativity into the mainstream of American physics and who had coined the phrase black hole to replace the more cumbersome gravitationally completely collapsed star. In 1972, Bekenstein published an argument that black holes must have entropy—specifically, that the entropy of a black hole is proportional to the area of its event horizon, measured in natural Planck units. The argument was based on a thermodynamic consistency requirement: if a high-entropy system could be thrown into a black hole, and if the black hole carried no entropy, then the total entropy of the universe would decrease when the system crossed the horizon, in violation of the second law of thermodynamics. The only way to preserve the second law was to assign the black hole itself an entropy that increased appropriately when it swallowed something.

The argument was logically compelling. It was also, for most physicists, deeply uncomfortable, and the discomfort had a specific source: if a black hole has entropy, the thermodynamic analogy between black hole mechanics and ordinary thermodynamics—a formal analogy that Carter, Bardeen, and Hawking himself had worked out in 1973—becomes physical rather than formal. And if the analogy is physical, the temperature in the analogy is a real temperature. And if the temperature is real, the black hole radiates. But general relativity said nothing could escape from a black hole. The conclusion was impossible.

Hawking had been one of the most forceful critics of Bekenstein's argument. He found the thermodynamic identification between the surface gravity of a black hole and a temperature physically nonsensical: a black hole at a real temperature would have to emit real radiation, and nothing came out of black holes. Bekenstein was making a category error, confusing a mathematical analogy for a physical reality. The calculation Hawking undertook in 1973 was intended to demonstrate this definitively. He would show that the temperature was exactly zero, that the radiation rate was exactly nothing, and that Bekenstein's pretty analogy was exactly that—an analogy, not a physical identity.

The calculation showed the opposite. And then, almost immediately, the physicists who had absorbed the result began to see what it implied.

If a black hole radiates, it loses energy. Mass and energy are equivalent, by Einstein's most famous equation, and the radiation carries energy away from the black hole, so the black hole must shrink as it radiates. A shrinking black hole has a higher temperature—the temperature is inversely proportional to the mass—and a higher temperature means more intense radiation, which means faster mass loss, which means faster temperature rise. The process accelerates. For a black hole of stellar mass, the timescale for complete evaporation is longer than the current age of the universe by many orders of magnitude, a number so large that astrophysical black holes are not evaporating in any practical sense. But the physics is the same regardless of the timescale. If a black hole is left in isolation long enough, it evaporates. It shrinks. Eventually, when it has become small enough that the semiclassical approximation breaks down and quantum gravity can no longer be ignored, something happens—and the most natural assumption, the one that gives the tidiest physics, is that the black hole disappears entirely, leaving nothing behind.

Everything that ever fell in goes with it. Every star that collapsed to form it. Every asteroid, every gas cloud, every quantum particle that subsequently crossed the horizon. All of it is gone. The radiation that carried the black hole's energy away is thermal—it is featureless heat, characterized entirely by the temperature at the moment of emission, carrying no information about what produced it. A black hole that formed from the collapse of a star of pure hydrogen and a black hole that formed from the collapse of an identical-mass star made of exotic matter would radiate in exactly the same way and leave exactly the same nothing behind. Whatever was different about them—whatever information distinguished them—has been erased.

Quantum mechanics says this cannot happen. The unitarity of quantum-mechanical time evolution is not a principle that admits exceptions. Every quantum system evolves according to the Schrödinger equation, and the Schrödinger equation is reversible: distinct initial states always evolve into distinct final states, and the complete specification of the present state of a system is always sufficient to reconstruct any past state. This is what information conservation means in quantum mechanics: not that you happen to know the state, but that the state exists uniquely, that there is a fact of the matter about the history, that the past can in principle be recovered from the present. A physical process that converts distinct initial states into identical final states—into the same thermal radiation—violates this. It is not quantum mechanics with a correction. It is not quantum mechanics at all.

Hawking saw the contradiction and accepted it. His position, stated with characteristic directness in papers and lectures throughout the late 1970s and 1980s, was that the black hole was an exception to quantum unitarity. Information was genuinely destroyed in the formation and evaporation of a black hole. The laws of physics were not deterministic at the fundamental level once black holes were taken into account.

This was a dramatic claim, and Hawking made it dramatically. He argued that quantum mechanics would need to be modified—that the framework that had succeeded in describing every other physical phenomenon would have to be extended to permit the kind of non-unitary evolution that black holes apparently produced. This was also the period in which Hawking became famous in a sense that theoretical physicists rarely achieve—famous to people who had no particular interest in physics, famous in the way that Einstein had been famous, as a cultural symbol of the scientific enterprise. A Brief History of Time was published in 1988 and remained on the Sunday Times bestseller list for four years, a feat that no popular science book had previously managed and that most publishers had thought impossible for a book about cosmology and black holes. Hawking had made a bet with his editor that every equation he included would halve the sales; the editor prevailed on him to include only Einstein's E equals mc squared, which Hawking judged irremovable. The book sold ten million copies. His face—the thin, angular face above the wheelchair, the slightly tilted head, the expression that was often described as a smile but was more accurately the expression of a man for whom smiling required effort—appeared on magazine covers in a dozen countries. He was a phenomenon.

The fame coexisted with the scientific position, and the scientific position was becoming increasingly isolated. Most of his colleagues in quantum gravity and string theory believed he was wrong. The argument about information loss was not settled, but the weight of expert opinion had been shifting since the early 1990s, and by the middle of that decade it had shifted substantially. Leonard Susskind, at Stanford, had made information preservation something close to a personal crusade—had written papers, given lectures, and eventually written a popular book specifically in response to Hawking's position. Susskind believed, and argued with considerable force, that Hawking's calculation was correct within its semiclassical approximation but that the approximation was not valid for the question of information preservation, that the correlations that would carry information out of the black hole in the outgoing radiation were precisely the kind of subtle, long-range quantum correlations that the semiclassical approach was structurally unable to detect. Hawking was calculating a quantity that appeared to be zero using a method that had zero resolution for the very quantity he was calculating. The method was right for what it could see. What it could see was not what mattered.

This argument Hawking found unpersuasive. He had done the calculation. The calculation gave a thermal spectrum. A thermal spectrum carries no information. He did not see why the argument from quantum gravity corrections, which no one could actually calculate, should override the calculation that had actually been done. The disagreement had the quality of a collision between two kinds of confidence: Hawking's confidence in the explicit calculation, Susskind's confidence in the principle. Both were real. Neither could resolve the other.

In 1997, the disagreement crystallized into a bet. Hawking, together with the Caltech physicist Kip Thorne, wagered against John Preskill, also of Caltech, on the question of whether information was destroyed or preserved in black hole evaporation. The stakes were an encyclopedia of the winner's choice—information, of a rather more ordinary kind, that could be retrieved at will. Hawking and Thorne bet that information was lost. Preskill bet that it was preserved. The bet was signed and witnessed, and copies circulated in the physics community the way the best physics bets always do, accumulating a significance that exceeded their nominal stakes.

The argument that eventually shifted the consensus came not from any calculation about black holes directly but from a mathematical structure that emerged from string theory: the Anti-de Sitter/Conformal Field Theory correspondence, proposed by Juan Maldacena in 1997. The correspondence established a precise duality between a theory of gravity in anti-de Sitter spacetime—a spacetime with specific negative curvature—and a quantum field theory living on the boundary of that spacetime, with no gravity. The boundary theory was a standard quantum field theory, explicitly unitary by construction, and the duality related every observable in the bulk gravitational theory—black holes included—to an observable in the boundary. If the duality was exact, and the evidence accumulated over the following years suggested that it was, then the bulk theory had to be unitary too. Black holes in the bulk could not destroy information, because their boundary duals could not. The argument was indirect—it established that information was preserved without saying how—but it was mathematically rigorous and, for most of the field, convincing.

Hawking took seven years to accept it.

In the summer of 2004, at the Seventeenth International Conference on General Relativity and Gravitation, held in Dublin, he announced that he had changed his mind. The announcement was not widely anticipated. Hawking had booked a slot at the conference, giving the organizers no indication of what he intended to say beyond the title of his talk: The Information Paradox for Black Holes. When word spread through the conference in the days before the talk that Hawking was planning something significant, the room filled in the way conference rooms only fill when something unusual is expected. Journalists arrived. Photographers crowded the back. The physicists who had been listening to Hawking argue for information loss for thirty years settled in with the slightly suspended quality of attention that attends any occasion when a major figure in any field prepares to say something that cannot be taken back.

He arrived, as he always arrived, in his wheelchair, the voice synthesizer that had replaced his own voice projecting the measured, slightly robotic American-accented English that had become, by 2004, as recognizable as any speaking voice in science. He had prepared a new argument—not a capitulation to Susskind's position or to the AdS/CFT argument, but his own route to the same conclusion, based on the Euclidean path-integral approach to quantum gravity, in which the sum over all possible spacetime geometries that could contribute to the quantum-gravitational amplitude included contributions from spacetimes with trivial topology, spacetimes in which the black hole had not formed, that restored unitarity to the total sum. The argument was technical, contested, and not universally accepted on the day or afterward. But it was not the argument that mattered. What mattered was the statement that accompanied it.

Hawking said, in the direct way that left no room for qualification, that he had been wrong. Information was not lost.

He conceded the bet, at sixty-two years old.

The moment carried an echo that another physicist’s moment of 1929 recognized. Back in 1917, Einstein had introduced the cosmological constant to produce a static universe, believed in it for over a decade, and called it his greatest blunder after Hubble's observations of galactic recession in 1929 showed that the universe was expanding. The public reversal—the willingness to say, plainly, that a position held with great conviction and great public authority had been wrong—was a rare thing in science, where the incentive structure typically rewards holding ground rather than yielding it. It requires a specific kind of intellectual integrity: the ability to distinguish between defending a position and defending the evidence. Einstein had it. Hawking, in Dublin in 2004, had it too. Both men had the stature to absorb the cost of the concession. Both had the honesty to pay it without equivocation. Neither attempted to reframe the retreat as a strategic repositioning, or to suggest that his original position had been subtly correct all along. They were wrong, and they said so.

What neither Einstein nor Hawking managed, having conceded the error, was the mechanism. Einstein removed the cosmological constant and offered nothing in its place; the constant's eventual return, in the context of dark energy observations from 1998 onward, came from a direction Einstein had not anticipated and would not have predicted. He removed the constant because the universe was expanding. He did not explain dark energy. Hawking conceded that information was preserved and offered, as a replacement mechanism, a succession of sketches—the Euclidean path integral argument in Dublin, the soft hair proposal in 2016—each more technically developed than the last, none of them complete.

He spent the remaining fourteen years of his life working on the mechanism, and the work produced nothing definitive. The most sustained attempt was the soft hair proposal, developed with Malcolm Perry at Cambridge and Andrew Strominger at Harvard, published in a series of papers from 2015 to 2016. Soft hair referred to low-energy quantum excitations on the horizon of the black hole, associated with a symmetry group of the gravitational theory known as the BMS group, after Bondi, van der Burg, Metzner, and Sachs, who had identified it decades earlier in the context of gravitational radiation. The BMS group is an infinite-dimensional symmetry of the asymptotic structure of flat spacetime, and the associated conserved charges—the soft gravitons and soft photons—are zero-energy excitations that leave a permanent imprint on the geometry of spacetime even as ordinary particles pass through it. The proposal was that infalling matter, as it crosses the horizon, deposits information on these soft excitations—imprints its quantum state on the gravitational equivalent of a permanent but extremely faint record kept at the horizon surface. The soft excitations would then participate in the outgoing radiation, providing a channel by which the information made it back out, encoded in correlations between the radiation and the horizon's soft modes.

The proposal was published in Physical Review Letters and attracted immediate and serious engagement. It was elegant, rooted in real mathematical structure, and it connected black hole physics to a body of work on symmetries and conservation laws that had been developing independently in several other contexts. Strominger, in particular, had been building a unified picture in which soft theorems, memory effects, and BMS symmetries formed a triangle of related phenomena, and the soft hair proposal placed the information paradox inside this larger structure. It was not dismissed. It was examined carefully and found insufficient. The soft excitations carry too little entropy—by a factor that grows without bound as the black hole mass increases—to account for the full quantum state of everything that has ever fallen into a macroscopic black hole. The calculation that would derive the Page curve from the soft hair mechanism was not performed, because the mechanism did not obviously produce it. The proposal pointed toward something, without arriving there.

Hawking published revisions and extensions in 2018, working at a speed that his physical condition made remarkable and that the people closest to him understood as something other than urgency about publication. He died in March 2018, leaving the final paper in a form that Perry and others completed and submitted after his death. The paradox that he had created in 1974 was, at the moment of his death, exactly as open as it had been when the Nature paper appeared.

Since then, the state of the problem is a proliferation of sophisticated frameworks and an absence of resolution. The island formula, derived from replica wormhole calculations by Penington, Almheiri, Mahajan, Maldacena, and Zhao in 2019, reproduces the Page curve that unitarity requires—the curve of radiation entropy that must rise and then fall, as the black hole loses mass, rather than rising monotonically as Hawking's original calculation predicted. This was taken by many in the field as a significant advance, and it is: for the first time, a calculation in a gravitational theory produces the right entropy curve, the curve that information conservation demands. But the physical interpretation of the replica wormholes—Euclidean geometric objects that dominate the gravitational path integral at late times in the calculation—remains contested. No one knows what they correspond to in real, Lorentzian spacetime, the spacetime in which actual black holes form and evaporate. The calculation gives the right answer, by whatever measure of rightness is appropriate, without explaining what is physically happening. The Page curve has been reproduced; the mechanism remains unspecified.

The AdS/CFT correspondence, which had shifted the consensus in 1997, continues to guarantee information preservation in principle and continues to provide no account of how it is achieved in practice. The duality relates the bulk gravitational physics, where black holes live, to the boundary quantum field theory, where unitarity is manifest—but it relates them without providing the translation dictionary that would allow a statement about information in the bulk to be converted into a statement about what the boundary degrees of freedom are doing. Information is preserved, the correspondence says. Where it is, and in what form, and by what physical process it escapes the horizon—these the correspondence does not say.

The firewall argument, which AMPS had sharpened in 2012 to show that the paradox was not merely a problem at the end of evaporation but a local physics problem at the horizon, remains without a universally accepted resolution. If the outgoing radiation is entangled with early radiation, as unitarity requires, it cannot also be entangled with the interior of the black hole, as the smooth horizon of general relativity requires—because quantum entanglement cannot be shared. One of the three frameworks must be wrong, or incomplete, or valid only within a domain whose boundary is the black hole horizon. Which one, and in what way, and whether the wrongness is a modification or a replacement—this remains open.

What unites every proposed resolution is what none of them specifies: the physical quantity that crosses the boundary, the thing that carries information from inside to outside in a universe where the geometric structure of spacetime insists that no causal influence does so. Every proposal that accepts information conservation must accept that something makes this crossing. The proposals that have been explored identify properties that the crossing thing must have, without identifying what it is. Hawking spent forty years looking for it. The field has spent fifty years looking for it. The question is not more tractable now than it was in 1974, except in the sense that the tools for failing to answer it have become substantially more sophisticated.

This is not a puzzle about a number or an approximation that will yield to a better calculation within the existing framework. It is a structural incompatibility between three independently valid and verified frameworks, each of which describes some aspect of the situation correctly, and none of which can be extended to cover what the other two describe. Thermodynamics accounts for the entropy, balancing perfectly. Quantum mechanics requires unitarity, absolutely. General relativity specifies the causal structure, exactly. The three specifications, placed together, produce a contradiction that half a century of the best theoretical physics in the world has not dissolved. The three frameworks do not merely disagree. Each one, taken alone, is internally consistent and confirmed by experiment. The contradiction appears only when all three are required to hold simultaneously—which they must, because the black hole is simultaneously a gravitational, thermodynamic, and quantum object. A system with three independently valid but mutually exclusive constraints has no solution within the space those constraints define. It requires a variable outside that space. To resolve it requires a quantity that none of these three frameworks contains—a quantity that crosses the boundary the geometric ledger says cannot be crossed, that conserves what the quantum ledger says cannot be lost, and that does so without disturbing the energy account that the thermodynamic ledger says is balanced.

Identifying this quantity requires going further back than Hawking, further back than the frameworks within which the paradox was formulated, to the point where the confusion about what is actually being tracked in these three ledgers began. The instinct is to place that point in 1974, when Hawking wrote down the radiation formula, or perhaps in 1973, when Bekenstein first assigned a quantity called entropy to a geometric surface, but the confusion was not introduced there. It was already present in the assumptions that both calculations took for granted—assumptions about what entropy is, what it measures, and what it means for entropy to increase. Unbeknownst to themselves, Hawking and Bekenstein inherited a problem they did not create. To work out why Hawking’s paradox became a question that even one of the greatest scientific minds of the time could not resolve, we need to begin where Clausius left off in 1865.

—— End of Chapter One ——

Search This Blog

Juliet Zhong