Bulges were supposed to be mergers' signatures. In hierarchical ΛCDM, disk galaxies grow through repeated merging, and mergers scramble orbits, funneling low-angular-momentum material into classical bulges whose mass, age, and kinematics should be tightly woven into the disk's assembly history (Mayer et al. 2008).
The census disagrees on two fronts. First, pure-disk and weak-bulge galaxies are common, including massive systems that should have suffered many bulge-building mergers; the existence of large bulgeless disks at late times is a long-standing embarrassment for merger-driven assembly (Kormendy et al. 2010). Second, where bulges exist, they are often decoupled: pseudo-bulges with disk-like kinematics, stellar populations whose ages do not track the disk's, and structural parameters that vary independently of the surrounding disk, as if the two components had separate origins rather than one interwoven history (Kormendy and Kennicutt 2004; Brooks and Christensen 2016). The model can suppress bulge formation with feedback in individual cases, but suppressing it often enough, while preserving merger rates that other observables require, has not been demonstrated.
The standing is part of the broader pattern in which galaxy structure looks set early and preserved, rather than churned by ongoing assembly. JWST's morphological census at high redshift, where pure disks already appear, keeps tightening the timeline against the merger-built-bulge picture.