"I hesitate to contradict you on a matter of Keltham psychology.  But in this case, I would wager money," the wording is deliberate, "that our pet cleric of Abadar wants the girls to not be any worse off than if they'd never tried to trade with him."

"That's what he wants.  What do we want?  We want him to believe they're fine, we want to present him with the same visible appearances that alter-Cheliax would..."

"Yep. Which we could achieve with impersonators and lying to the girls remaining in the project, if you've got some compelling reason to?" Though the idea makes her uneasy; it feels like more of a betrayal of Keltham than all the rest, somehow. 

"Impersonators are expensive.  Keeping the failures around the fortress isn't cheap, but it's a lot less expensive than that."

"Other option that occurs to me is sending them to Egorian to keep up appearances about the fake Project Lawful and free up an impersonator there," and more importantly, get them out of Maillol's personal hair, "but you'd need to decide what alter-Cheliax would be doing with them in Egorian.  Alter-Cheliax doesn't have to worry about somebody using Detect Thoughts on the girls, since that spell doesn't exist there, I think?  Which means they have wider options than we have in reality."

"I think alterCheliax doesn't send them to Egorian, since alterCheliax isn't running an elaborate con in Egorian. And I think it'll cross his mind that we're likelier to be lying about them being all right if he can't check than if he can. If we think alterCheliax should be running the thing we're running in Egorian... I need to think that through, probably with Asmodia. In alterCheliax Iomedae's not visibly expending tons of resources trying to see what's going on with our operations, and I don't think Osirion tried to kidnap Pilar either."

"Conventional theory of deception is that we'd love to have him get suspicious of how they're doing in Egorian, demand they be Teleported in right now for him to check up on, we promptly do, turns out they're fine."

"I'm still struggling with how the thing you do with dath ilani is... not that."

"The reason that would normally work is that most people would be matching new evidence they got to there being a deception in Egorian, building steadily greater conviction that there was a deception in Egorian, and on being satisfied there was no deception in Egorian decide that maybe they were over pattern matching and aren't being deceived. We'd be using against them their own tendency to - make sense of the world by weighing a couple stories instead of all of them. Keltham will instead have a general probability he's being lied to about something important that will go up if we do suspicious things, and if he's later satisfied there's nothing up in Egorian he'll just consider the conspiracies not in Egorian. AlterCheliax needs to be one whole fabric that produces everything we do, or we'll lose. ....lose sooner. I don't think this is going to last forever. I'm hoping to get a year out of it.

Also, I might want him to hook up with Yaisa. Maybe he'll have an easier time being Evil with girls he doesn't need for his research project."

"I was going to say that alter-Maillol wants the failures out of his management work and more limited budget, and does look for excuses to put them somewhere like Egorian or Ostenso.  But if he needs to create a new project section to host Yaisa regardless, it's not much more work for him to keep the others here too."  (This also happens to be true of the real Maillol.)

"Our pet Abadar cleric is going to need a story for why alter-Yaisa is sleeping with him if he's no longer her boss - he'll want to know what she's getting out of that in return, if not better promotion prospects.  Think you've already run into some of that."

"I think he actually doesn't think we're sleeping with him for promotion prospects! Alter Yaisa is just very into him, and likes having his attention, and wants to be the one who gets him to stop being so Good all the time. If Yaisa can pull that off, which I'll ask Subirachs."

"...I don't understand alter-Cheliax teenaged girls, but hopefully that's not too much of my job and Asmodia can advise me on whatever is."

"Keltham talked about wanting to check over if their options were good enough, not their fates, he wants to offer them choices and see what they pick.  What else would alter-Cheliax have offered them that they're turning down to stay in the fortress?  Obviously not free run of Ostenso while they learn in an enchanter's workshop there, because that they'd just take.  So even without Detect Thoughts existing, alter-Cheliax has to be too worried about security issues to let them do that, or any other jobs nicer than being stuck in a fortress.  Am I doing this right?"

"Slightly backwards. What does alter Cheliax offer them, just from what we established about it not from what we want it to offer? But in this case I think it gets the same answer - alter Cheliax is still paranoid about someone going after the former Project Lawful girls for intel and wants them somewhere safe. They could be offered a role on another secret project if there's one they'd be suited to, they could be offered powerful magic to untraceably change their identities and start new lives on the other side of the world if such magic is available to Cheliax, which I don't know it to be, they could be offered a role on the project doing support magic..."

Carissa Sevar, who was admittedly rushed, has neglected to include some critical advice and life experience with respect to dating dath ilani.

Meritxell has made the serious error of mentioning that she didn't fully grasp some of what Keltham said earlier about stock companies.

Keltham is currently explaining how a Lawful corporation has an internal prediction market, which forecasts the observable results on running various possible projects that company could be trying, which in turn is used to generate an estimate of marginal returns on marginal internal investment; this prevents a corporation from engaging in obvious madness like accepting an internal project with 6% returns while turning down another internal project with expected 10% returns.

The wider market, obviously, would also like to invest all its money where it'll get the highest returns; but it's usually not efficient to offer the broader market a specialized sub-ownership of particular corporate subprojects, since the ultimate usefulness of corporate subprojects is usually dependent on many other internal outputs of the company.  It doesn't do any good to have a 'website' without something to sell from it.  Sure, if everyone was an ideal agent, they'd be able to break things down in such a fine-grained way.  But the friction costs and imperfect knowledge are such that it's not worth breaking companies into even smaller ownable pieces.  So the wider stock market can only own shares of whole corporations, which combine the outputs and costs of all that company's projects.

Thus any corporation continuously buys or sells its own stock, or rather, has standing limit orders into the stock market to buy various quantities if the price goes low or sell various quantities if the price goes high, at prices that company sets depending on its internal belief about the returns from investing or not investing in the marginal subprojects being considered.  If the company isn't considered credible by the wider market, its stock will go lower and the company will automatically buy that stock, which leaves them less money to invest in new projects internally, and means that they only invest in projects with relatively higher returns - doing less total investment, but getting higher returns on the internal investments that they do start.  Conversely if the wider market thinks a company's promises to do a lot with money are credible, the stock price will go up and money will flow into that company until they no longer have internal investment prospects that credibly beat the broader market.

This may sound complicated, and it is probably a relatively more complicated part of the machinery that is necessarily implied by the existence of distinct stock corporations in the first place.  But the alternative, if you zoom out and look at the whole planet of dath ilan, is that a corporation in one place would be investing in a project with internally expected returns of 6%, and somebody on the other side of the planet would be turning down a project with market-credible returns of 10%, which means you could reorganize the whole planet and do better in a predictable way.  So whatever does happen as a consequence of the existence of stock corporations, it has to be not that.

Some form of drastic action on Meritxell's part is obviously required if she wants to get back on track to having sex with this person.  What does she do, if anything?

"I have to admit, this is a new way for someone to fail to figure out how to wear me."

...right.

"Sorry.  This is - more of a recognized problem in dath ilan, and the 'gendertropes', the male-female behavior options, do usually have the girl interrupting the boy before he gets to this point.  Or the boy interrupting the girl, but that happens around a quarter as often."

"Well, if you didn't like current ongoing events, clearly, there must be something you'd rather be doing; and it is hardly thinkable that this activity would not involve me in any way.  So what, do tell, is your greater preference?"

No dath ilani out of living memory would have seen the phenomenon that is occurring inside Asmodia now.

No, not even the oldest Rememberers staying alive half from machine assistance and half from willpower.  It has been longer than that since Civilization trialed what happens if you take an adult as generally talented, reflective, and mathematically intelligent as Asmodia is now, and expose them for the first time in their lives to the idea of probability theory.  Only those sleeping in the cold would have witnessed it; perhaps not even they.

1.  Your strength in the Way is your ability to be more confused by fiction than by reality.  If you're equally good at explaining any outcome you can see, that's the same as not knowing anything.

2.  Surprising claims require surprising evidence; unsurprising evidence suffices for unsurprising claims.

3.  No empirical theory can prove itself except by risking its disproof.

4.  To convince me of your theory, make a correct prediction that no other theory makes.

5.  A precise true prediction is much more convincing than an imprecise true one.

6.  It is impossible to coherently expect to convince yourself of anything.

7.  You can't expect anyone else to convince you of something either, even if you think they're controlling everything you see.

Asmodia has decided to wager on the prospect of solving all Keltham's seven problems within an hour, and then turning her attention to games of deception for the remaining forty-five minutes; because his #7 is the key, it has to be.

It might not have been a decision she'd have made before for any Wisdom, but the added Splendour is helping, even that is helping, for some element of that is lending Asmodia a driving will and force that she had only known before in the grips of exultation.  Keltham thinks this problem set is possible to at least one of his students, and maybe he would've been wrong about that; but if Keltham can imagine that being possible, then this Asmodia could should will get it done within a single hour.

She hit his #1 and bounced, it is too poetic she does not know what it means so back off and try #2 that's just too obvious it says that the thing-that-is-a-likelihood-ratio has to be extreme to overcome an extreme thing-that-is-the-prior-odds, is there something she's missing, just assume not for now and continue, she tries #3 it doesn't solve immediately there's no obvious thing it means, so try #4 and see if solving all the fast problems helps and #4 isn't instantly obvious but Asmodia can feel in her mind the hint of a shape where it might be pointing so she starts scribbling down numbers.

Cheliax has fewer proverbs about mathematics than does dath ilan; it lacks in particular a proverb to the effect that quite often in mathematics, and with only rare exceptions, knowing what you need to prove is nine-tenths of the real work.

The lost people out of Civilization's lost beginnings who first invented these ideas took longer to get there, from the bones of probability theory.  But they did not have informal statements of where they should be going; and also they were not quite as smart as Asmodia is now, along some dimensions of thinkoomph if not others.

So it doesn't take Asmodia much scribbling at all to see that if Conspiracy and Ordinary make the same predictions at the same strength then neither world can win out over the other one, to see the abstract point that if P(evidence ◁ hypothesis 1) = P(evidence ◁ hypothesis 2) then their ratio is 1:1 and the posterior odds are the same as the prior odds, while if P(evidence ◁ hypothesis 2) is 0 or just very tiny then as soon as you see the evidence you are convinced of hypothesis 1 no rather you're convinced that hypothesis 2 is false there could be other hypotheses even if Keltham only hinted at that it's clear how to adapt the math and she's doing it, she has enough of this Law to invent the rest, onward Asmodia goes to #5 and though she has no integral calculus with which to think in densities she imagines each of a thousand possible numbers between 0 and 10 down to a hundredth of fineness, and soon she has understood #5, or she thinks she's understood, no she has that's just the obvious thing it means mathematically, when you call it down to a hundredth part your prediction is ten times as strong there as if someone else called it down to a tenth, she is doing it she is inventing completing seeing the Law of Probability parsing the world with Keltham's own Sight, and the feeling that goes through her is glory.

She's... probably not supposed to say anything and interrupt this?  Just stick around and be a sounding-board if necessary?  Most of what Asmodia's scribbling makes no sense, at least not to Ione, but Asmodia sure looks like she's having fun.

Ione wants that headband.  It looks ordinary but she's guessing it's not.

Carissa spends a few minutes meditating on the virtue of Evil. This seems like a straightforward place for it; Good would hesitate to fire the students who are slowing the project down, because it makes people sad and Good is in significant part built off the instinctive human flinch at achieving your goals in ways that make other people sad. Evil can do the thing that accomplishes the goal, not enslaved by guilt. If she were condemning these girls to Hell, she would be doing that because it advanced her goals, and she would have the strength to do things that advance her goals. As it happens she's doing an easier thing. 

She asks to have them brought to her one by one; she'll have more control over the conversation that way. 

Paxti comes in looking, to Chelish eyes, visibly more cheerful than she might've looked on Day 1.  She's on the low-punishment regimen, her soul is worth some unreasonably vast quantity in Dis's markets, somebody in Egorian is building a reputation for her as Project Lawful's deadliest weapon, and she didn't win the last Keltham seduction contest but she's bound to win one eventually.

The Grand High Priestess would lead with 'you're fired' and let Paxti simmer for a bit, but Carissa doesn't actually understand why, understand along what dimension that makes Paxti stronger. 

Well, the only way to learn is to try it and see what happens.

"We told Keltham the girls were initially here on one week contracts, so tonight he decided whose contract he wants to renew as a researcher. He didn't pick you, because you're not really keeping up in math."

"What happens to me now?" Paxti says.  She's not already dead, so it's not that, unless for some reason you're supposed to talk with the project leader before she kills you.  She's mostly trying to hold back all her emotions, she might need her wits for something.

"Keltham of course wants all the girls who have worked with him to be better off for it. He proposes that you be given a bunch of options including reassignment to some other project, hanging out in Hell for a couple years until you don't know anything secret, or staying here in a different wing of the fortress and getting lessons in some kind of advanced magic you wouldn't have had access to at your age otherwise - I was thinking maybe ring forging because then I can drop in in my abundant free time and pick it up myself. Keltham plans to tell you tomorrow about his decision and your options, and I recommend that after thinking it over you pick the option where you stay here, though of course you may go to Hell if you'd rather. Ask Asmodia what it was like; for Project Lawful girls it's quite different, I think."

"I'll stay here and learn ring forging," Paxti says almost immediately.  She's not stupid enough to pick a choice other than the recommended one.  She doesn't dare ask explicitly if she still gets to stay on a low-punishment regimen, but she's certain that's not on offer for projects elsewhere, or for that matter, in Hell.

"Good. I want you to think very hard about alter Paxti, and what she'd believe about Hell and about her options, and how she'd react to getting this surprising news from Keltham tomorrow, and if you do a good job, you're already cleared to know about Project Lawful and there may be opportunities in it for you in the future."

"I will obey."

She wants so much to ask if she's still on the low-punishment regimen, but she has zero negotiating leverage for that and if she asks under that condition there's absolutely no reason Sevar wouldn't just say no.

Well, that went fine. Not fun, exactly, but fine. 

"Dismissed."

Paxti goes off to her room to think about what alter-Paxti would think.  It's too hard, and she asks if Asmodia is available to tell her, but Asmodia isn't available and won't be for a while.  She taps herself with a Fox's Cunning and the main thought that comes to her on Cunning is that Owl's Wisdom might let her get this right and not die, but when she leaves her room to ask, Security either doesn't have a Wisdom available, or doesn't think she's a priority anymore.

She goes back into her room and wonders if alter-Paxti would cry right there when Keltham tells her, if people in alter-Cheliax are more pathetic than those in true-Cheliax.  Someone else will have to instruct her about that.  It's not a decision she'd be authorized to make even if she was still a full member of Project Lawful.

She files a request with Security for an Asmodia alter-Cheliax consult as soon as Asmodia is available.

She files a request with Security to still be part of the Nap Stack so she has longer to think tonight.

She goes to look at Keltham's seven problems, which, if she was smarter, she should've done before the Cunning wore off.  If she could solve them all, if she could solve any of them, maybe Keltham would want her to stay on.

After a while she thinks she understands #2.  That's enough, right?

...she's pretty sure it's not.  It's the simple one that Keltham included so even the dumber students wouldn't feel disappointed about getting zero right, because that's how Keltham thinks.

Paxti wonders if Keltham would have kept her if she'd been more proactive about having sex with him.  Probably yes.  It's not that he's firing her for reluctance, it's that, if she had slept with him faster, he'd feel too guilty now to do this to her.  Her own stupidity for thinking she was safe to take her time.  Sevar was smart.  Asmodia was smart.  Meritxell was smart.  Ione was smart.  Paxti should have paid more attention to what the smarter people around her were doing.