What physical structure makes an acting, remembering system a person, and what makes its later experience belong to the same subject? Recognition Science connects this question to retained comparisons and the relationships an action changes. We derive exact constraints on that connection. A fixed reversible update stores and rereads a finite history; its main-state trajectory through one tape cycle determines the entire encoded initial state. Linearity excludes exact readout of every encoded pure state when the apparatus must also leave it unchanged and uncorrelated with the record. Exact universal duplication of a native signal is incompatible with linear evolution, while reversible transfer is possible. Participant consent survives grouping of subjects precisely when authorship is constant within each group. A unique fixed-total ledger extension retains native account updates and the external bond’s participant effects. At a common native operand product, endpoint pulse-energy changes equal the corresponding scaled cost and signed-harm changes. These results constrain physical history, transfer and authority. Identifying that history with an experiencing subject still requires an occurrence-level experiential identification and a sound and complete law of subject composition. We state those obligations on the same physical objects.
What physical structure makes an acting, remembering system a person, and what makes its later experience belong to the same subject? Recognition Science connects this question to retained comparisons and the relationships an action changes. We derive exact constraints on that connection. A fixed reversible update stores and rereads a finite history; its main-state trajectory through one tape cycle determines the entire encoded initial state. Linearity excludes exact readout of every encoded pure state when the apparatus must also leave it unchanged and uncorrelated with the record. Exact universal duplication of a native signal is incompatible with linear evolution, while reversible transfer is possible. Participant consent survives grouping of subjects precisely when authorship is constant within each group. A unique fixed-total ledger extension retains native account updates and the external bond’s participant effects. At a common native operand product, endpoint pulse-energy changes equal the corresponding scaled cost and signed-harm changes. These results constrain physical history, transfer and authority. Identifying that history with an experiencing subject still requires an occurrence-level experiential identification and a sound and complete law of subject composition. We state those obligations on the same physical objects.
Can distinct subjects belong to one experiencing whole? Within Recognition Science’s shared physical architecture, we derive an exact criterion: every finite collection of felt occurrences belongs jointly to a subject if and only if the all-occurrences profile lies in the closure of existing subjects’ profiles. For a nonempty felt domain, closedness yields an existing whole; nested membership preserves distinct participants. The physical derivations establish matter–photon correlation under the complete atomic source, a fixed observable with an absolute correlation margin, and all-time record retention under a specified holding interaction. Signed operators give exact coherent code capture. A complete finite receiving Hamiltonian captures coherent input with error at most C_v/√N for N bank cells and any normalized initial stock, while retaining the native stock clock and every finite passive reference. A normalized product partner evolving under its independent native clock preserves the same bound. With injectively matched contacts covering the bank, the full receiving and contact Hamiltonian also gives a bank-size-independent constant, including coherent inputs and every finite passive reference. These results assume supplied stock and contacts. The remaining connections are physical realization and completeness, recognition’s experiential identity, actual finite subject composition, and realization of the whole-subject limit. These conditions distinguish the positive account from its unfinished universal derivation. AI-use disclosure: OpenAI Codex assisted with source analysis, Lean formalization, manuscript drafting, and independent critique within the author’s Recognition Science research.
Two physical states can have the same cost and still respond differently to the same action, so a description that keeps only the cost cannot say what a participant will undergo. We ask what a description must keep so that no accessible consequence is lost. Squaring a comparison of two counts as signed postings and reading the residual against the comparison's own cross postings gives the reciprocal cost J(x) = (x + 1/x)/2 - 1; a fixed composition law for chained comparisons, with a unit calibration, singles it out among all costs. Measured responses to that cost recover a directed ledger of bonds between participants. Economic transitions act on the same ledger, and a record of the path keeps what endpoint sums lose. On a finite quantum system, a finite list of preserved binary decisions (two, for a qubit) forces a physical channel that recovers the whole state, including its correlations with a retained reference. Each result holds under stated physical premises. The Born weights are fixed by stated record rules, chief among them register noncontextuality. What remains to be derived is which outcome a read gives and when, and which carriers and interactions the ledger's laws produce.
Two physical states can have the same cost and still respond differently to the same action, so a description that keeps only the cost cannot say what a participant will undergo. We ask what a description must keep so that no accessible consequence is lost. Squaring a comparison of two counts as signed postings and reading the residual against the comparison's own cross postings gives the reciprocal cost J(x) = (x + x−1)/2 − 1; a fixed composition law for chained comparisons, with a unit calibration, singles it out among all costs. Measured responses to that cost recover a directed ledger of bonds between participants. Economic transitions act on the same ledger, and a record of the path keeps what endpoint sums lose. On a finite quantum system, a finite list of preserved binary decisions (two, for a qubit) forces a physical channel that recovers the whole state, including its correlations with a retained reference. Each result holds under stated physical premises. The Born weights are fixed by stated record rules, chief among them register noncontextuality. What remains to be derived is which outcome a read gives and when, and which carriers and interactions the ledger's laws produce. Version 1.1 adds: the confusion of a comparison with its reverse is the only one any cost obeying the three laws of identity can make, so a cost-only description loses exactly the direction of each bond; the shape of the composition law is forced for every rational cost with no ceiling, and the record reading is its member k = 1 with the unit fixed by calibration; a test that tells the cost apart from a quadratic logarithmic rival; harm counted once, with a borne share that sums to the change of the ledger's total cost; the sphere of a qubit read as the sphere of ratios, on which the cost is sec(latitude) − 1; the characterization of local subjects composing into a whole; and the Born weights from stated record rules. It corrects version 1.0's abstract, which said that two preserved binary decisions force recovery on any finite quantum system: two suffice for a qubit, and d levels need a longer finite list (Theorem 6.1).
A physical description must preserve distinguishable occurrences, the information needed to compare them, and records available to later interactions. We examine what these requirements force within Recognition Science, keeping the premises of each implication explicit. Finite record constructions give arithmetic with a ledger of logical commitments. A specified comparison apparatus, continuous reporting and reference calibration fix the reciprocal cost, while a positive additive reading and a first costly join force golden scaling. A linking requirement selects three dimensions within the stated spherical spatial class; a complete one-bit cyclic pass through three binary coordinates then has minimum length eight. Factorization, order and calibration determine the corresponding recognition weight. In a finite quantum carrier, preservation of calibrated sharp-decision optima forces exact state recovery, including correlations with a retained reference. We also identify what these mathematical results require of a physical realization: acquisition of the supplier state, complete resource accounting, attachment of controller and record dynamics, turnoff, and record retention. Ethical and economic effects act on the same physical bond ledger. The experiential interpretation requires a further admissibility law and a distinct subject-composition condition. The result is a connected account of forced structure and its physical premises, with explicit boundaries between a uniqueness theorem, a constructed representation and an enacted process.
Two physical states can have the same cost and still respond differently to the same action, so a description that keeps only the cost cannot say what a participant will undergo. We ask what a description must keep so that no accessible consequence is lost. Squaring a comparison of two counts as signed postings and reading the residual against the comparison's own cross postings gives the reciprocal cost J(x) = (x + 1/x)/2 - 1; a fixed composition law for chained comparisons, with a unit calibration, singles it out among all costs and fixes that choice of reference. Measured responses to that cost recover a directed ledger of bonds between participants. Economic transitions act on the same ledger, and a record of the path keeps what endpoint sums lose. On a finite quantum system, two preserved binary decisions force a physical channel that recovers the whole state, including its correlations with a retained reference. Each result holds under stated physical premises. Which carriers, interactions and occurrence probabilities nature admits remains to be derived.
A physical identification requires a domain of experiments, a realization, and an observation that distinguishes competing mechanisms. We construct these for the δ framework. Within finite causal quantum experiments, computable Choi data are equivalent to finite arithmetic programs producing valid instrument approximations at every requested accuracy. A separate pointer reader preserves the quantum posterior and supports adaptive composition. For the native two-atom census, the actual join update and common-mode retention select golden scaling and a signed contraction. Coherent return selects its minimal conserving completion; an independently calibrated spectral band then selects its constant interaction. A bright-mode construction and a stable short-time calibration procedure supply an explicit realization and uncertainty budget; measured duration and frequency supply the dimensional units. We derive a general endpoint identity between stored record energy and comparison imbalance, and distinguish the monotone fresh-record trajectory from coherent recurrence. The logical-to-physical bridge also has a sharp boundary: an explicit instance of the proposed completion generator has no effective value-reflecting physical reader or numerical comparison receipt. Sequence completion requires precisely the corresponding halting-oracle resource, not an unexplained scalar cost. This obstructs operational completion while leaving purely proof-language transport distinct. Native implementations of identical logical strength nevertheless carry different physical receipts. Joint covariance and the exact endpoint law state the valid relationship between those quantities. Finally, calibrated return and three-outcome tests distinguish the relevant phase and recording alternatives, with quantitative error bounds. First public preprint release, version 1.0. The files include the PDF and the main TeX file. AI-use disclosure: AI assistance was used for source analysis, mathematical exploration, manuscript drafting and revision, diagram preparation, and computational verification support. GPT-6 Astra reviews assessed arguments, scope, readability, and schematics. These were internal model reviews, not external peer review. Jonathan Washburn is the sole listed author and is responsible for the released manuscript.