Gravitational lenses can weigh in on distances twice, and their second testimony is just beginning to be heard. Beyond time delays, lensed quasar systems yield angular-diameter distances directly: combine the image geometry, the time delays, and the lens galaxy's stellar kinematics, and the distance to the lens follows with systematics largely orthogonal to every standard-candle method, no ladder, no calibration chain, a geometric measurement of D_A at redshifts the local methods never reach. The pioneering applications (Paraficz-Hjorth, then Jee, Suyu and collaborators) established the technique on the H0LiCOW lens sample, yielding lens distances whose implied H0 sits in the low 70s with 10-to-20 percent per-lens precision, and double-source-plane systems add distance-ratio constraints that test the expansion history's shape independent of H0's absolute scale.
The method's promise and its predicament arrived together: the same mass-sheet and kinematic degeneracies that complicated the time-delay program propagate here, since the lens model's mass profile and the line-of-sight convergence enter the distance inference directly, and the small sample sizes leave the technique's verdict on the Hubble tension suggestive rather than decisive. The structural interest is what the method measures that others cannot: angular-diameter distances to individual systems at z of 0.3 to 0.9, each along its own line of sight, with environmental convergence as an explicit model ingredient rather than a buried assumption, the configuration any environment-dependent expansion would need to be tested system by system.
The standing is a technique scaling up: Rubin and Euclid will deliver hundreds of suitable lenses, JWST and ELT-class kinematics will break the profile degeneracies, and lens distances will graduate from supporting evidence to an independent map of D_A across redshift and environment.