Advanced Wizardry is unusually willing to say what its own magic will not do. Across the spellbook and the trades, a working is repeatedly gated — and the gates are not all the same kind of thing, which is what this page is about.
Some gates are mechanical. The recognition charm only works if this person is truly hard working and actually doing a great job, and the reason falls out of how the charm is described as working: it casts a remembering zone around the person, and a zone that holds a memory has nothing to hold when nothing happened.1 The confession charm will have no effect if the person hasn't committed any crimes, and its effect is a guilt that will only be alleviated by confessing — no crime, no guilt to relieve.2 In both cases the interlock is not a rule bolted on. It is the mechanism, seen from the side.
Other gates are motive. The blessed-speech charm cannot be used for selfish means, for it will naturally not work.3 The countenance charm — which changes a person's aura to that of an extremely trustworthy one — carries the harshest warning in the volume: used with selfish ends or with the intention to take advantage of other people, it will fail with the direst of consequences.4 And the hero conjuration will only help if the fight is fair, just and proper.5
Put the two kinds side by side and one of them has an evaluator and the other does not.
The mechanical gates need nobody to judge anything. A remembering zone with no memory in it does nothing because there is nothing in it. The guilt charm produces a discomfort with one exit and a person who did nothing has nothing to walk out of.
The motive gates need somebody to look at the caster and decide. And in exactly one case, this volume says who: the hero judges the fight before helping. Somebody turns up, assesses, and declines.
In the other two, nobody is named. It will naturally not work. It will fail with the direst of consequences. Naturally, according to what — and consequences delivered by whom?
⚠ And the two unattended gates sit on the two charms in the whole spellbook that most obviously invite abuse. One makes every word out of your mouth almost compelling and quite difficult to resist, so that listeners follow the speaker's suggestions. The other makes strangers want to trust you, with no requirement that you be trustworthy.
Falsifiable claim: in this corpus, an interlock binds if and only if the mechanism does the checking. Every gate that depends on the caster's motive is decorative, and the corpus reveals this by never naming an evaluator for one.
The test is mechanical and nobody has run it. Take every stated limit across the 165 pages and sort it into three bins — mechanism-checked, evaluator-named, motive-only. The claim predicts that the motive-only bin contains the charms with the highest abuse potential, and that no charm in it has a described failure mode beyond "it does not work."
Every system that tries to gate a capability on intention has this problem, and most of them hide it better. A rule that says do not use this selfishly is not a constraint on the tool; it is a description of who the author hopes will hold it. A constraint that lives in the mechanism — the charm needs a real memory, the guilt needs a real crime — cannot be talked around, because there is nothing to talk to.
The interesting move here is that one corpus contains both kinds, side by side, in the same schools, without noticing the difference.
Weakest point: the two bins may not be as clean as the claim needs. The hero conjuration is evaluator-named and motive-gated, which suggests a third category rather than a binary — and if there are three kinds, "decorative" is too strong for the middle one. The honest reading until the sort is run is that motive-only gates are unverifiable, which is weaker and more defensible than decorative.