Structural equation models cannot adequately represent grounding and metaphysical explanation

Against the account, a couple of things can and have been said. First, just as there are important analogies between grounding and causation (of the ordinary, run-of-the mill sort: §1), there are disanalogies as well. This is accepted by both friends and foes of the view. The question then is whether those differences are serious enough to undermine the work the analogy does in the account (proponents of the account, not surprisingly don’t think so; for some who do: see, e.g., Bernstein 2016,