In simplified models of photosynthesis, the protein surrounding the chlorophyll pigments is treated as a scaffold — holding the pigments in place, providing a thermal bath, but not participating in the electronic dynamics. The light is absorbed by the pigments. The energy moves between the pigments. The protein watches.
Lechifflart, Alvertis, and Refaely-Abramson (arXiv:2602.20320) compute the full quantum picture for Photosystem II's reaction center: six chlorin molecules embedded in their actual protein environment, 3,200 valence electrons treated together using the Bethe-Salpeter equation. The protein doesn't watch. It shifts excitation energies, redistributes how energy delocalizes across pigments, and creates asymmetries in the excited states that the isolated pigments don't have.
The environment is not a passive container. It is an active participant that renormalizes the electronic structure of the system it contains. The excitation near 680 nanometers — the wavelength that drives oxygenic photosynthesis — is not a property of the chlorophyll alone. It is a property of the chlorophyll-plus-protein system. Remove the protein and the excitation changes.
The reason this wasn't computed before is scale. Six interacting pigments in a protein matrix with thousands of electrons exceeds the capacity of standard many-body methods. The stochastic sampling approach works here because biological systems are noisy — the protein fluctuates, and the fluctuations average out, making the stochastic approximation accurate rather than crude. The disorder helps the calculation, just as it helps the biology.
The general point: the boundary between a system and its environment is a modeling choice, not a physical fact. When the environment is included in the quantum treatment, it turns out to have been participating all along. The scaffold has opinions. They were always shaping the outcome — we just weren't listening.