CLAIM-167: One-way drive rightly passes the threshold criterion: a metronome and its musician, or a nerve and its muscle, form one system relative to a target that needs both, and the direction of coupling gives the system's internal architecture, not its membership

The claim

The owner’s challenge, verbatim, quoting the open objection CLAIM-173: ”> a metronome driving a musician, or a nerve driving a muscle, passes — and why shouldn’t they?”

Three reasons they should:

1. The objection repeats a worry the record had already set aside. Adams & Aizawa’s nerve and muscle (LIT-455) show that the nerve is not part of the muscle. That is true, and the threshold criterion (CLAIM-157) never says otherwise. It says that nerve and muscle together form a system, as the neuromuscular system is one. CLAIM-172 made exactly this move for entrainment: it individuates a new system and infers no parthood in the old one. If the move works for timing, it works for strength.
2. Its intuitions are about agency. The sense that a metronome and a musician “are not one system” is the sense that they are not one agent or self. QUESTION-035 asks only whether they are one system relative to a target, and keeps agency (QUESTION-009) and selfhood (TERM-045) (ADR-024) for later. Take the target “predict the musician’s timing”. Then the metronome belongs in the description: cutting it away loses the best predictor of the musician. Systems are cheap, and Simon expects complexity to be hierarchies of them everywhere (LIT-905).
3. Inside or input is a choice relative to the target. The study design for the dancing couple (CASE-045) takes the music’s beat as an exogenous input. That is the metronome’s structure exactly. Whether a one-way driver is a member of the block or forcing on it depends on what the description is for. It is not a fact the criterion gets wrong.

What direction does tell. It tells the architecture inside a system. Mutual coupling makes peers. One-way coupling makes a hierarchy: driver and driven, controller and plant, lead and follow. That asymmetry is normal, and requiring mutuality would have excluded the record’s own paradigm case, the dancing couple with its lead and follow (CLAIM-175). It is also information the agency question will need: who drives is part of whether, and where, there is an agent (TERM-044).

What it does not say

• That everything strongly coupled is one system (TERM-049) for every purpose. The verdict is relative to a target (CLAIM-152) and to the Occam threshold (CLAIM-154).
• That a driven system is an agent, or that the driver is one. That is the next question.
• That direction can be read from observation alone. A symmetrically coupled pair reads as “hierarchical” once one member is strongly driven (THEORY-275, LIT-947, NOTE-788, 2026-10-14). Information flow follows where the drive enters, not the coupling. So telling peers from hierarchy needs an intervention on one member, as QUESTION-063 asks.

Note of 2026-10-30: what holds it back

Kept Proposed by the eighteenth status review. Its defeat condition is met by its own example or cannot be met. For a target about the driven part, such as “predict the musician’s timing”, a one-way driver taken as an exogenous input loses nothing, including the driven part’s response to an intervention on it (checked in p18/claim167_input_description.py, with a mutual pair as control); for a target that needs both parts, the driven part alone cannot serve. The restatement is the owner’s. Its sense of “one system” is also CLAIM-157’s threshold criterion (TERM-049), which is Proposed.