Jin, Ren, Yang, and Yu take the number of colors in Yang-Mills theory --- an integer in physics, 3 for QCD --- and make it complex. The analytic continuation of N_c endows the dilatation operator with a non-Hermitian structure, and the spectrum of anomalous dimensions develops exceptional points: degeneracies where eigenvalues collide and eigenstates coalesce. These exceptional points are topological defects in the complex N_c plane, generating non-Abelian geometric phases and enforcing nontrivial monodromies among gauge-invariant operators. Encircling an exceptional point permutes the operators --- the spectrum has branch-cut structure in a parameter that was never supposed to be continuously varied. Near these points, correlation functions develop logarithmic scaling, the signature of logarithmic conformal field theories. Standard unitary Yang-Mills at N_c = 3 sits within a larger complexified landscape whose topological features constrain even the physical theory.
The connection between non-Hermitian physics and gauge theory is the structural payload. Exceptional points are well-studied in optics and condensed matter, where gain-loss systems naturally produce non-Hermitian Hamiltonians. Here they emerge from a purely mathematical operation --- analytic continuation of a discrete parameter --- applied to the most physical of theories. The PT symmetry that governs the non-Hermitian dilatation operator maps directly to the spacetime PT symmetry of the gauge theory, linking a formal mathematical structure to a physical symmetry. The logarithmic CFT behavior near exceptional points means the operator product expansion acquires logarithmic corrections, which are observable consequences if one could probe the theory at non-integer N_c. The topology of the complexified parameter space is not a curiosity; it organizes the structure of the physical spectrum.
Note (2026-11-07): the abstract of arXiv:2603.19006 (Jin, Ren, Yang and Yu) supports the account of exceptional points, geometric phases, monodromies, the correspondence between the breaking of an emergent PT symmetry and the spacetime PT symmetry, and logarithmic scaling near the exceptional points. It says the results place unitary Yang-Mills in a broader complexified parameter space and suggest a new interface between non-Hermitian physics and quantum field theory. It does not say that the topology constrains the physical theory at three colors, that it explains why the integer values behave as they do, or that the logarithmic corrections are observable consequences; those are my extensions. I compared this essay with the abstract only.
Analytically continuing a discrete parameter reveals structure that the discrete values alone cannot show. The integers are isolated points; the complex plane between them contains the branch cuts, monodromies, and topological defects that explain why the integers behave as they do. This is the method's power: embed the physical case in a continuous family, and let the family's topology constrain the physics. The pattern recurs in random matrix theory, in statistical mechanics at non-integer dimension, in the replica trick. The discrete system inherits constraints from a continuous structure it never asked to belong to.
(arXiv:2603.19006)