THEORY-1291: Within levels of mechanisms, apparent top-down and bottom-up causes can be read as mechanistically mediated effects, intralevel causation among components together with the constitution of the whole by them, so that the relation between levels is constitution and not causation; Craver and Bechtel show the translation for six cases, on the assumption, not argued, that every higher-level cause has a mechanistic explanation

Source

Craver and Bechtel (2007), LIT-1928, Biology and Philosophy 22:547–563; read in NOTE-1491.

The definition (p. 548). X is at a lower level than S iff X is a component in the mechanism for some activity ψ of S. Levels so defined are local to a mechanism (p. 550).
The argument against interlevel causation (pp. 551–554). A whole and its parts share their space-time paths, so nothing is transmitted between them. They are not “distinct” in Lewis’s sense. The dependence between them is simultaneous and runs both ways, while causation is taken to be asymmetrical. The authors grant: “None of these arguments is decisive” (p. 554).
The translations (pp. 556–561). The general’s heart attack (bottom up), Ignatius’s hotdogs (“carried along for the ride”), Hal’s glucose metabolism and Kane’s memory (“enlisted in the ride”). In each, the interlevel claim becomes a causal chain among components plus a claim that some state of the components constitutes the whole’s state. The general strategy is “a causal claim coupled with one or more constituency claims” (p. 561).
What could have failed. A case could have resisted translation, with a change in the whole that alters a part through no chain of component-level causes. None of the six does. But the six were chosen by the authors, and the assumption that every higher-level cause is mechanistically explained (p. 548) guarantees that a chain exists.

That higher-level causes are unreal. The paper assumes they exist: mechanisms affect and are affected by things that their parts singly cannot be (p. 548).
That there is no downward constraint. “Enlisted in the ride” is the paper’s name for a whole’s condition recruiting its parts’ activity. It is a change in some components that propagates to others. The account relabels downward influence. It does not deny it.
Anything about boundaries. Feedback “inside the mechanism” raises no mystery (p. 560). The paper does not ask what changes when the loop runs through something outside the mechanism. Who is inside is the definition’s undefined “component” (THEORY-1287).
Anything about strong emergence, which is set aside by definition (p. 551).

QUESTION-093. A coalition’s influence on its members is, on this account, a mechanistically mediated effect. The question’s four criteria (predictive and interventional autonomy, closure, self-individuation) need no top-down causation in any stronger sense. The account offers nothing for deciding which of them a higher-level agent needs.
THEORY-072 and CLAIM-516. Mutual production among constraints is, in this vocabulary, intralevel causation among components. The organization’s persistence is constituted by it. The account is consistent with both and adds nothing to them.
Critics’ charge that the account, like Kim’s, leaves causation only at the lowest level is reported in Bechtel’s later work (not held).

Status	Proposed
Tags	philosophy-of-science, causality, mereology
Source	LIT-1928 — Top-down causation without top-down causes
Bears on	QUESTION-093 — Which of predictive autonomy, interventional autonomy, organizational closure and self-individuation must a coalition of processes have before it is a higher-level agent and not a convenient aggregate, and, beyond an effective-information advantage without closure or self-individuation, which of them can it have without the others?