We revisit whether a semiclassical closed baby universe in AdS/CFT necessarily possess a trivial one-dimensional Hilbert space or may instead carry a large entropy. Recent results on Haar random encoding suggest a breakdown of complementary recovery, in which no logical operators can be reconstructed from individual bipartite subsystems. Motivated by this, we propose an interpretation where a baby universe emerges as logical degrees of freedom that cannot be accessed from either boundary alone, assuming pseudorandom dynamics in holographic CFT correlators. We then analyze two conceptual puzzles: an apparent cloning of baby-universe microstates and its eventual fate at the singularity. Both puzzles are avoided because no single boundary observer can access the baby-universe degrees of freedom, be it classical or quantum, reflecting an emergent form of complementarity due to the structure of random encoding. In this interpretation, observers arise naturally: the same heavy operator that prepares the baby-universe geometry also serves as observer-like degrees of freedom that define an observer-dependent baby-universe microstate.
Conventional relativity places four-dimensional spacetime connected by Lorentz transformations, and fourdimensional spacetime curved by gravity, at the foundation of physical spacetime, whereas quantum mechanics places Hilbert space at the foundation of state description. As long as these two structures are simultaneously treated as fundamental realities, relativistic quantum theory and quantum gravity must begin with the problem of how a dynamically curved spacetime is to be quantized and made consistent with Hilbert-space states. Nakaza New Relativity Theory (NNRT) changes this very point of departure. NNRT takes universal time and universal three-dimensional space as physical spacetime and connects the real motions of all inertial systems by the Galilean Transformation (GT), while relocating the Lorentz transformation, time dilation, length contraction, and the gravitational metric to the observational and measurement layer obtained through electromagnetic phase. The successful mathematics and experimental values of conventional relativity are therefore retained, but the ontology in which curved and deformable four-dimensional spacetime is the unique physical spacetime is not retained. This paper uses NNRT as a new entry into quantum mechanics and repositions Hilbert space not as physical space, but as a mathematical representation space that preserves states, phases, probabilities, and correlations. Using the atomic clock as a concrete example, the paper further presents a continuous observational chain in which quantum levels generate a frequency scale and the resulting phase signal is compared between systems through relative motion and gravitational fields. Accordingly, the conventional entry point—how to quantize a physically real curved four-dimensional spacetime and incorporate it into Hilbert space—loses its premise within NNRT, and quantum theory and relativity are reconstructed as distinct representational and observational structures arising from a common universal physical spacetime. Schrödinger’s cat, EPR/Bell correlations, time dilation, length contraction, the twin problem, and the reification of the gravitational metric can then be organized as manifestations of a common ontological operation: identifying a mathematical representation with physical reality itself. The significance of this paper is therefore not to add another interpretation alongside conventional relativity, but to absorb its successful mathematics into the observational layer of NNRT and, by turning the ontological point of departure, to provide a new formulation of foundational problems that includes quantum mechanics.
We combine the ideas of Dirac's orthonormal representation, Everett's relative state, and 't Hooft's ontological basis to define the notion of a world for quantum mechanics. Mathematically, for a quantum system
Q with an associated Hilbert space
H, a world of
Q is defined to be an orthonormal basis of
H. The evolution of the system is governed by Schrödinger's equation for the worlds of it. An observable in a certain world is a self-adjoint operator diagonal under the corresponding basis. Moreover, a state is defined in an associated world but can be uniquely extended to the whole system as proved recently by Marcus, Spielman, and Srivastava. Although the states described by unit vectors in
H may be determined in different worlds, there are the so-called topology-compact states which must be determined by the totality of a world. We can apply the Copenhagen interpretation to a world for regarding a quantum state as an external observation, and obtain the Born rule of random outcomes. Therefore, we present a mathematical formalism of quantum mechanics based on the notion of a world instead of a quantum state.
We reformulate the problem of the "interpretation of quantum mechanics" as the problem of DERIVING the quantum mechanical formalism from a set of simple physical postulates. We suggest that the common unease with taking quantum mechanics as a fundamental description of nature could derive from the use of an incorrect notion, as the unease with the Lorentz transformations before Einstein derived from the notion of observer independent time. Following an an analysis of the measurement process as seen by different observers, we propose a reformulation of quantum mechanics in terms of INFORMATION THEORY. We propose three different postulates out of which the formalism of the theory can be reconstructed; these are based on the notion of information about each other that systems contain. All systems are assumed to be equivalent: no observer-observed distinction, and information is interpreted as correlation. We then suggest that the incorrect notion that generates the unease with quantum mechanichs is the notion of OBSERVER INDEPENDENT state of a system.
The definition of entropy in both classical and quantum contexts relies on a coarse-graining of the underlying microscopic description. While this dependence is widely acknowledged, the physical constraints that determine which coarse-grainings are admissible are rarely made explicit. This issue becomes particularly acute when the universe is treated as a closed system, where no external observer or privileged level of description can be assumed. Any characterization of entropy — including claims about the low-entropy initial conditions of the universe — must then be grounded in descriptions that can be physically instantiated from within the system itself. In this work, we introduce a rigorous operational framework that establishes explicit physical selection principles over the space of admissible descriptions. We organize these criteria into necessary conditions (local accessibility and physical implementation via internal interactions) and practical sufficiency conditions (finite resource requirements anchored in Landauer’s principle, dynamical stability, and effective consistency). To demonstrate the validity of our framework, we introduce a concrete one-dimensional spin-chain toy model, illustrating how mathematically valid global partitions fail to be physically realizable due to dynamic propagation limits and informational constraints. We analyze the far-reaching implications of this restriction across three domains. In classical statistical mechanics, it limits the arbitrariness of macrostate definitions. In quantum theory, it identifies environment-induced decoherence (einselection) not merely as a formal tool, but as the objective dynamical mechanism that forces the emergence of admissible internal partitions. In cosmology, it resolves foundational circularities in the quasi-classical histories framework and reframes the thermodynamic arrow of time as a "doubly relational" property—dependent both on cosmic boundary conditions and on the material limits of internal subsystems. Our framework clarifies the operational status of entropy as an emergent property of physically instantiated descriptions within a universe that observes itself.
We associate a gravitational Hilbert space
Hσ to any closed compact
(d−1)-manifold
σ with real metric. A quantum state
J(σ) is a
d-manifold bounded by
σ and equipped with elliptic data. An inner product is defined by gluing states pairwise across
σ and evaluated by viewing the resulting closed
d-manifold as a boundary condition on the gravitational path integral (GPI) over
(d+1)-manifolds.
If
σ is nonempty and the GPI is dominated by a single
(d+1)-manifold
M in the
GN→0 limit, then
M contains a Lorentzian CRT fixed-point set, providing
J(σ) with a classical spacetime interpretation. Conversely, given a finite Lorentzian domain with edge
σ, a state
J(σ) may be associated to it by deforming its initial data off the real Lorentzian section and retaining only elliptic data. This establishes a broad correspondence between non-asymptotic spacetime regions and quantum states.
Assuming that
Hσ factorizes over connected components of
σ, our framework admits operators and partial traces. This allows us to explore the information-theoretic structure of the states we define. As an example, we construct a family of states by deforming partial Cauchy slices
Σ that straddle a two-sided black hole;
σ consists of two spheres. We construct the reduced state on one sphere and find that its Rényi entropies are positive, monotonic, and sensitive to all aspects of
Σ and its complex deformation. The von Neumann entropy, however, is controlled only by the maximin surface in the causal domain of
Σ, independently of other parameters, so long as the complex deformation does not vanish. Our proposal may thus explain the efficacy of tensor network toy models of holography while transcending their limitations.