Why it is a question
CLAIM-047 says possessing a regime is not the same as its being effective, and CLAIM-027 says which regime is effective depends on maintained relations: accessibility, precedence, persistence, jurisdiction, suspension and transformation. Each of the six relates regimes to one another or to the agent’s situation. None is the agent’s representation of what it can do.
Metzinger’s mind-wandering is a case where that is the variable. During an episode the capacity to suspend or end one’s own train of thought is, on his “moderate” reading, kept: “what we have lost is a specific form of knowledge, and not the ability itself” (LIT-1941, p. 282). It is not exercised, and he says why: “one can only deliberately and autonomously exert an ability if that ability is explicitly represented in one’s phenomenal self-model” (p. 280). If so, the representation is a condition of efficacy, and the agent’s will-organization includes a relation between a regime and the agent’s model of having it. Two years later he withholds the causal reading. The epistemic agent model may be “a mere epiphenomenon”, or “exactly the missing computational element that constitutes” control, or may “not constitute, but represent” and “causally enable relevant second-order functional properties like veto control”; the data “do not help to decide” (LIT-1945, pp. 13–14).
The question matters to the record beyond the will line. CLAIM-355 keeps real organization apart from the owner a self-model represents, and A26’s reply to plain fictionalism is that “self-representations may themselves be parts of the real causal organization they represent”. This question asks whether, for one representation, the agent’s model of its own control capacity, that reply is true.
Metzinger’s own material shows the two can come apart both ways:
- Control without the representation. “Successful inhibition can be an entirely subpersonal process” (LIT-1942, p. 13). If so, the representation is not necessary for control as such, and the principle holds only for “deliberate” control, which may be defined by the representation.
- The representation without control. “the phenomenology of M-autonomy does not justify the claim that the functional property of M-autonomy was actually present” (LIT-1941, p. 290), and the return of control is represented as self-caused whatever caused it (THEORY-1298).
What would count as an answer
An intervention on the representation alone, with the capacity measured by a route that does not pass through it. Metzinger’s H1 and H3 (LIT-1945, p. 14) are of this kind: block the epistemic agent model’s introspective availability and see whether mind-wandering begins; restore predictive self-control and see whether an episode ends. Held fixed must be the competing mental processes and the task (as CLAIM-122 asks), and measured separately must be the ability, for instance by externally cued stopping.
- Condition of efficacy: blocking the representation ends deliberate control while cued stopping is intact, and restoring it restores control. Then the representation is a relation the record should name, closest to accessibility.
- Accompaniment: blocking the representation leaves control as it was, or control lapses only when cued stopping lapses too. Then the representation tracks efficacy without governing it, and the record’s six relations stand as they are.
What it is not
- Not CLAIM-523’s legibility, which is about whether a discrepancy is recognized, not whether the agent represents its own capacity to act on it. Metzinger’s meta-awareness (that one is off task) is legibility; his epistemic agent model (that one can stop) is this question’s variable.
- Not CLAIM-022’s question of whose the will is. A representation could be a condition of efficacy and still be authored or installed by others; Metzinger says the adult’s sense of origination is installed by caregivers before autobiographical memory (LIT-1945, pp. 20–21), which bears on QUESTION-364.