The Hubble tension and the observer's causal diamond

The Hubble constant from the local distance ladder exceeds the cosmic microwave background (CMB) value by more than 5σ, and baryon acoustic oscillations (BAO) return an intermediate value. Proposed changes to early- or late-time physics have not yet resolved the tension. This paper explores whether the ladder, BAO and CMB values can be reconciled within a quantum recording architecture that operationalises Wheeler's participatory universe. The proposed mechanism is an endpoint gate on the boundary of the observer's causal diamond that shortens the distance scale recorded by a measurement. The diamond can change in five directions: its size, the time of its centre and the three spatial coordinates of its centre. Each direction that a probe fixes in setting its scale passes through the gate once and survives with probability σ = exp(−3/(16π²)). A probe that fixes m of these five directions, its metrological rank, records a Hubble length shortened by σ^m and reads σ^(−m) times the Hubble constant of the realised geometry. The CMB takes its scale from last scattering, rank zero. BAO fix separations across and along the line of sight, rank two. The ladder's chain of distances can fix all five, and the ladder data select rank five. A CMB fit with three parameters fixed by the architecture gives H0* = 66.80 km s⁻¹ Mpc⁻¹ for the realised geometry. The predicted readings 66.8, 69.4 and 73.5 match the CMB, BAO and ladder without fitting a continuous parameter to late-time data. With the rank left free, DESI DR2 BAO prefer 1.72 ± 0.26, close to rank two. Reading BAO at rank two lowers the preference for evolving dark energy from 2.5–4.0σ to 0.3–1.6σ across three supernova samples. A recording architecture centred on the observer may reconcile the ladder, BAO and CMB values. The main current limitation is that the physical process by which the gate shortens the scale recorded by each measurement has not yet been derived. Falsifiable predictions are presented for forthcoming surveys.