Assumption Map of the Paradox

An information paradox is not a contradiction between slogans. It is a derivation from a specified set of assumptions about the state space, horizon, dynamics, and observable algebra. An assumption map makes that derivation falsifiable: every premise is written mathematically, the inconsistent subset is isolated, and removing any one member is tested by an explicit alternative model.

Required background. Quantum Information supplies purification, entropy inequalities, channels, and finite-dimensional countermodels used in the deletion tests.

Helpful background. Semiclassical Gravity supplies the evaporation regime; Quantum Fields in Curved Spacetime supplies the near-horizon pair state; Causality and Global Structure supplies complete slices, horizons, and endpoint alternatives.

Minimal inconsistent assumption sets

Let $\mathcal A=\{A_1,\ldots,A_n\}$ be assumptions translated into propositions. A subset $M\subseteq\mathcal A$ is inconsistent if its members imply both a statement $P$ and its negation. It is minimal if every proper subset loses the contradiction. Minimality is logical, not sociological: it does not identify which premise nature abandons or which premise is least plausible.

For a quantitative argument, each $A_i$ should include its error. “The horizon is smooth” becomes a bound on the distance between a near-horizon state and the local vacuum. “Information escapes” becomes a statement about a specified radiation entropy or recovery fidelity. “Locality” becomes factorization or approximate commutation for named algebras on a named slice. Without these translations, two authors can appear to disagree while testing different propositions.

A practical deletion test has four steps:

1. name every system and the common state in which the premises are evaluated;
2. derive the contradiction while carrying approximation errors;
3. delete one premise and construct a witness model satisfying the rest;
4. repeat until no further premise can be removed without restoring consistency.

Failure to find a witness is not proof that the premise is necessary, but a witness is decisive evidence that the original contradiction used it.

A Hawking-curve assumption set

Consider radiation $R_k$ after $k$ emission steps and a new exterior–interior pair $b_kc_k$. One useful assumption set is:

• C — complete exterior endpoint: after complete evaporation, the final physical algebra is the radiation algebra $R_N$, with no remnant or disconnected purifier;
• U — closed unitarity: a pure initial state gives $S(R_N)=0$ when $R_N$ is complete;
• P — semiclassical pair structure: each step creates $b_kc_k$ close to a pure entangled pair and $b_k$ is nearly independent of $R_{k-1}$;
• F — persistent factorization: $R_{k-1}$, $b_k$, and $c_k$ are independent tensor factors to the accuracy used by the entropy inequality.

Under exact pair structure and factorization,

$$\rho_{R_{k-1}b_k}=\rho_{R_{k-1}}\otimes\rho_{b_k}, \qquad S(R_k)=S(R_{k-1})+S(b_k).$$

For an entangled Hawking pair, $S(b_k)\gt0$, so the radiation entropy grows at each step. Small corrections leave a positive lower bound under the hypotheses quantified by Mathur Mathur 2009, §§ 4–6. Repeated growth conflicts with U and C at the endpoint. Hawking’s pure-to-mixed proposal retains C, P, and F but rejects U Hawking 1976, §§ II–IV.

Each deletion points to a different response. Delete C and a remnant $H$ can purify $R_N$. Delete U and treat the mixed radiation map as fundamental. Delete P and emit a late qubit correlated with early radiation rather than a fresh independent partner. Dropping F instead requires reformulating P: its displayed tensor-product statement cannot be retained literally once the factors are removed. An encoded interior may reproduce selected vacuum correlators, but an explicit common operator model must show which statements survive. These possibilities explain why “unitarity versus smoothness” is not the only formulation; they do not, by themselves, prove this informal four-item set logically minimal.

From assumptions to response classes

The same method prevents proposals from talking past one another. A remnant changes endpoint completeness. A firewall changes the near-horizon state. Complementarity restricts simultaneous operator assignments or comparisons. State dependence changes the domain on which interior operators are represented. Islands change the entropy prescription and effective subsystem assignment in controlled gravitational models. Fundamental loss changes closed-system dynamics.

These moves are not automatically consistent theories. They merely identify which premise is being modified. A proposal must also reproduce ordinary exterior effective field theory where it is tested, preserve linearity on its declared state space, and state whether its conclusion concerns a formal entropy, an asymptotic amplitude, or an executable recovery protocol.

The AMPS argument supplies a sharper late-time example: early radiation $E$, a late exterior mode $B$, and an interior partner $A$ cannot satisfy all of late-radiation purification, near-vacuum $AB$ entanglement, and ordinary factorization to the asserted accuracy Almheiri et al. 2013, § 2. That proof is developed later; here it illustrates why labels and approximation norms belong in the assumption map.

Results and open questions as of 2026-08-19

The logical method is stable, but the minimal set depends on formulation and parameter regime. It remains unresolved which assumption a complete four-dimensional theory changes. Current evidence supports several competing positions rather than a universal verdict about which assumption fails.

Argument or proposal	Result for the assumption set	Scope and limitation
Hawking endpoint Hawking 1976, §§ II–IV	Retains the semiclassical endpoint and rejects exact unitary purity	Leading semiclassical spacetime does not fix the microscopic completion
Small-corrections theorem Mathur 2009, §§ 4–6	Pairwise perturbative corrections are insufficient to reverse entropy growth	Depends on factorization and a stated norm bounding corrections
AMPS late-mode argument Almheiri et al. 2013, § 2	Old-hole purification, vacuum entanglement, and semiclassical factorization are jointly inconsistent	Applies to the named $EBA$ algebras and accuracy assumptions; it does not select a unique response
Complementarity and restricted descriptions Harlow 2016, §§ 5.5–5.7	Attempts to prevent one observer from operationally comparing duplicated descriptions	Operational restrictions alone need not remove a common-state entropy contradiction

The conservative position treats the map as a diagnostic of where semiclassical reasoning loses control. More permissive positions reject a particular factorization, locality assumption, or operator domain while retaining approximate predictions. A third position accepts nonunitarity or an enlarged endpoint. Their conclusions differ because their Hilbert space and common-state assumptions differ.

The conclusions could change after a theorem controls approximate factorization errors, or after a microscopic single-theory model realizes one deletion witness through complete evaporation, or after an operational experiment distinguishes formal inconsistency from inaccessible comparison without unacceptable backreaction.

Independent deletion test

Use three qubits $R,b,c$. The state $\lvert0\rangle_R\otimes\lvert\Phi^+\rangle_{bc}$ realizes exact P and F and raises $S(Rb)$ by $\ln2$. Replacing it by $\lvert\Phi^+\rangle_{Rb}\otimes\lvert0\rangle_c$ makes the new radiation purify $R$, but the horizon pair $bc$ is no longer vacuum-entangled. Identifying $c$ with a logical operator already encoded in $R$ avoids an independent factor but changes F. These alternatives expose which premise moves; they do not prove that gravity realizes any of them.

Generic entropy methods remain with Quantum Information, fixed-background pair production with Quantum Fields in Curved Spacetime, and stable thermodynamics with Semiclassical Gravity. Explicit string or holographic realizations are outside this page.

Exercise

An argument uses U, P, and F but never asserts C; a hidden system $H$ remains after radiation emission. Does monotonic $S(R_k)$ contradict unitarity of the complete evolution?

Solution

No. Complete unitarity constrains the entropy of $RH$, not of $R$ alone. The hidden system can purify the radiation, so $S(R)\gt0$ is compatible with a pure state on $RH$. To recover an information-loss contradiction one must add C, asserting that $R$ is the complete final algebra, or require a later process that transfers the correlations from $H$ to $R$. The missing premise is therefore substantive, not cosmetic.

References

• Almheiri, Ahmed, Donald Marolf, Joseph Polchinski, and James Sully. 2013. “Black Holes: Complementarity or Firewalls?” Journal of High Energy Physics 2013 (02): 062. Open text.
• Harlow, Daniel. 2016. “Jerusalem Lectures on Black Holes and Quantum Information.” Reviews of Modern Physics 88: 015002. Open text.
• Hawking, Stephen W. 1976. “Breakdown of Predictability in Gravitational Collapse.” Physical Review D 14: 2460–2473. DOI.
• Mathur, Samir D. 2009. “The Information Paradox: A Pedagogical Introduction.” Classical and Quantum Gravity 26: 224001. Open text.