A Framework for Universality in Physics, Computer Science, and Beyond

TL;DR

The work develops a unified, category-theoretic framework for universality across disparate domains, ranging from universal Turing machines to universal spin models. By formalizing an ambient category $\mathcal{A}$ with targets $T$, contexts $C$, and a simulator $s: P\otimes C\to T\otimes C$, it defines universality via reductions to a trivial simulator, and introduces a shadow-based observable behavior through a behavior structure. It achieves two main pillars: (i) a No-Go result showing universal spin models must involve infinitely many Hamiltonians, and (ii) a Lawvere-inspired fixed-point analysis linking universality to unreachability/undecidability, with concrete instantiations for TM, NP-completeness, and spin systems. The framework further provides a simulator-category for comparing universality (parsimony) and proves functoriality results to relate different instances, laying groundwork for cross-disciplinary transfer of universality results. Overall, the paper offers a principled, modular approach to classify, compare, and extend universality phenomena across computation, physics, and mathematics, while highlighting fundamental limits via fixed points and unreachability.

Abstract

Turing machines and spin models share a notion of universality according to which some simulate all others. Is there a theory of universality that captures this notion? We set up a categorical framework for universality which includes as instances universal Turing machines, universal spin models, NP completeness, top of a preorder, denseness of a subset, and more. By identifying necessary conditions for universality, we show that universal spin models cannot be finite. We also characterize when universality can be distinguished from a trivial one and use it to show that universal Turing machines are non-trivial in this sense. Our framework allows not only to compare universalities within each instance, but also instances themselves. We leverage a Fixed Point Theorem inspired by a result of Lawvere to establish that universality and negation give rise to unreachability (such as uncomputability). As such, this work sets the basis for a unified approach to universality and invites the study of further examples within the framework.

Figures (12)

• Figure 1: (a) A target--context category consists of an ambient category $\mathcal{A}$ (which is quasi-total gs-monoidal) with distinguished objects $T$ (targets) and $C$ (contexts), as well as an ambient preorder relation $\mathord{ \gtrdot }$ on every set $\mathcal{A}(A,T \otimes C)$. (b) This preorder is preserved by precomposition.
• Figure 2: While the ambient category $\mathcal{A}$ describes abstract processes, their "observable consequences" are manifested as relations between sets provided by the functor $\mathrm{Sha}$. To provide a behavior structure is to specify the latter (\ref{['def:beh_structure']})
• Figure 3: An example of an enhancement map $\textrm{enh} \colon U\to V$ where $U$ and $V$ are subsets of $X= \{\top, y, z,\bot\}$, and where $\succeq$ is generated by the Hasse diagram depicted with solid lines. The map takes any element in $U$ to an element in $V$ which is above in the $\succeq$ relation. Note that in this example $\mathrm{enh}$ is not a degradation map (see \ref{['fig:degradation']}).
• Figure 4: An example of a degradation map $\textrm{deg} \colon V\to U$ where $U$ and $V$ are subsets of $X= \{\top, y,z, \bot\}$, and where $\succeq$ is generated by the Hasse diagram depicted with solid lines. The degradation map takes any element in $V$ to an element in $U$ which is below in the $\succeq$ relation. Note that in this example $\mathrm{deg}$ is not an enhancement map (see \ref{['fig:enhancement']}).
• Figure 5: (a) The ambient imitation preorder $\mathord{ \gtrdot }$ (\ref{['def:mrel']}) between morphisms of type $A \to T \otimes C$ in the ambient category (left-hand side) is defined via the imitation preorder of their shadows ($\overline{f}$ and $\overline{g}$) in the category of relations, when composed with the evaluation function. Evaluation is used to generate a "behavior-valued relation" of type $\overline{A} \to \overline{B}$. (b) Behaviors of shadows ($\overline{\mathrm{eval}}\circ\overline{f}$ and $\overline{\mathrm{eval}}\circ\overline{g}$) are related by the imitation relation $\mathord{ \gtrdot^{\mathrm{im}}_{\hbox{\tiny$B$}}}$ (\ref{['def:brel']}) precisely if there are $a$-dependent enhancement and degradation maps as specified in the figure.

Theorems & Definitions (124)

• proof
• remark 1: Levels of abstraction
• example 1: Category of relations $\mathsf{Rel}$
• Definition 2.1: fritz2022lax
• Definition 2.2
• Definition 2.3
• Definition 2.4
• remark 2: Functional $\neq$ single-valued
• Definition 2.5
• Definition 2.6