1.1The question
A decision about force, or the emergency around it, is made at one time and carried out at another. Between the two lies what the decision could not know: whether supplies will arrive, whether a report was true, whether those who must act will be able to. The classical vocabulary for this gap is Clausewitz's: friction, and uncertainty.
This article asks whether the emergency questions published on Meridian treat uncertainty as a friction that civilians must take on. It asks first what follows, for civilian authority and those who execute its decisions, if Clausewitz's claim that policy governs war is read with his claim that war resists every plan; and second, of two records: when a civilian body writes down an emergency decision before its outcome is known, does it keep the uncertainty for itself, hand it to those who execute, or pass it to those who wait?
The two records come from Meridian, the decision-record platform of the RISCON family, which publishes decision questions set in fictional polities, each with thirteen judgments generated by AI. In one, a budget council must decide how to extend an emergency budget that feeds and houses displaced residents after an audit found that some payment claims lack delivery confirmation. In the other, a continuity committee must decide who may approve civilian protection spending while the elected parliament cannot reach its quorum.
1.2The contribution
The article makes three contributions, each narrower than the work it builds on.
The first is a reading of On War. The parent study's reading note gives three claims from Book I: that war is a continuation of policy by other means (Chapter 1), that friction makes the simplest thing difficult (Chapter 7), and that much of what action in war rests on is hidden in uncertainty (Chapter 3). This article reads those chapters at length, adds Chapters 6 and 8 of Book I and a chapter of Book VIII, and argues that the claims form a single problem: policy must govern a medium whose resistance is known only through experience at the point of execution. Clausewitz's own answers to that problem are part of the reading.
The second contribution brings friction into the civil-military debate. The reading notes on Samuel Huntington (1957), Peter Feaver (2003) and Eliot Cohen (2002) describe who holds authority and how it is checked; none mentions friction or uncertainty. Feaver's principals monitor their agents; a principal who monitors by results, which is this article's case, faces a diagnostic problem when results fall short: the cause may be shirking, or it may be friction that no one could avoid. The argument draws on Feaver's note for the terms principal, monitor, work and shirk; the diagnostic problem is this article's.
The third contribution reads the two records with that problem in hand. The records are modern summaries written by AI for the platform, set in fictional polities; they report no real state, budget, parliament or person. The budget record shows a civilian body that refuses to treat a gap in paperwork as proof of wrongdoing and writes a calendar for telling the two apart. The quorum record shows friction striking the civilian principal itself, and shows the judgments parting over whether an argument from speed should move the power of approval to the executive.
1.3The debate joined and the limits of the claim
Section 2 lays out the positions of Huntington, Feaver and Cohen side by side and adds three works that bear on decisions made under uncertainty: Lee Clarke (1999) on plans that could not work as written, David Collingridge (1980) on keeping decisions corrigible, and Niklas Luhmann (1993) on the difference between risk and danger. Placing these works in one debate, setting them beside Clausewitz and applying them to the two records is this article's own doing. Each work is described only as far as the parent study's reading note goes; where the argument goes further, the step is the article's.
The article does not decide among the positions, and it does not say which option either fictional body should choose.