CosmicFlows-4 and earlier catalogs find coherent bulk motion of 400 to 600 km/s toward Centaurus-Vela persisting across volumes hundreds of megaparsecs deep, roughly twice the ΛCDM expectation (Feldman 2010; Kashlinsky 2012). Gaussian initial conditions require flows to average away with volume; these do not.
In the standard model all peculiar motion is gravitational infall toward cataloged structure, sourced by Gaussian fluctuations whose large-scale power is fixed by inflation. Beyond a couple hundred megaparsecs nothing remains to pull coherently, so a persistent fast flow demands an uncataloged attractor, non-Gaussian initial power, or a non-infall source, each forbidden in its own way.
SCT's answer is that the flow is not infall toward anything; it is residual motion left over from the collision itself. Our patch retains a bulk velocity within its parent frame set by the collision geometry, v_frame ≈ v_rel(final) × (b/R_min) (P63), a permanent kinematic property shared coherently by the entire debris field. A flow inherited at creation does not average away with volume, because every galaxy in the patch carries it; the observed refusal of the bulk flow to damp with scale is the defining fingerprint of inherited motion as against gravitational infall, which must damp.
Two registered companions complete the picture. Sibling pockets, the statistically expected co-products of our collision sitting at 1 to 2 Gpc separations within the Hubble sphere (P58, P59, P60), add a slow gravitational component aligned with the parent-frame geometry. And the canonical ledger registers the directional test: the bulk-flow direction should align within 30 degrees of the local cosmic web's angular momentum axis, because the frame velocity and the deposited J vector descend from the same impact geometry (P31, P64). The Centaurus-Vela direction passes this check. The collision kinematics are laid out in Paper 1, From Chaos to Convergent Foundations, with the alignment prediction registered through Paper 13, From Chaos to Coherent Gravity. The same inventory carries the dark-flow claims, the count-dipole excess (recid 77), and the kSZ directional structure (recid 69).
One premise is the keystone: P63. A patch born moving keeps moving, and no attractor needs to be invented to push it.
DESI and 4MOST peculiar-velocity programs carry the kill: if the bulk flow rolls over and converges to ΛCDM linear-theory amplitudes beyond a few hundred megaparsecs, the inherited-velocity reading is refuted, since v_frame cannot damp with volume. The directional half is equally exposed: a flow direction incompatible with the cosmic-web J axis at more than 30 degrees, measured against improved filament catalogs, breaks the shared-geometry prediction.