"What's the plan if it looks like in a few years they'll have a massive, spellsilver driven military advantage -"

"Pray."

"Less than a ten percent chance, in my view, that they're going to hold on to him for that long. Though - they've surprised me so far."

Years?  Keltham totally intends to be producing massive amounts of spellsilver in months.  He's graphed out the supply networks, Cheliax has located a potential mine for the key ore, stockpiling preliminary quantities of the ore is not that expensive, he's got a process that doubles the amount of sulfuric acid on every cycle, his tier-1s are picking up Prestidigitation at the requisite level, and people currently think that ten pounds of spellsilver is a lot.

PL-timestamp:  Day 51

PL-timestamp:  Day 52

Contessa Lrilatha reports to Keltham on the presentation to the Worldwound-treaty countries, including the offer to bet. Mostly the complaint of the other countries was that this would be incredibly expensive and they aren't as rich as Cheliax so the fact it's worth it to very-rich Cheliax doesn't mean it's worth it to anyone else. Also no one wanted to bet, on the grounds that betting is a bizarre thing to do. She thinks they'd be more receptive if there were good industrial processes for metals such that alternatives to lead for pipes were affordable. 

Cheliax has also tried Neutralize Poison on lots of people. An effect on Intelligence is not observable immediately, but even if the Neutralize Poisoning is working it might take time to have effects; Neutralize Poison doesn't restore you instantly to an unpoisoned state, just gets the poison out. They went over everyone with Detect Intelligence beforehand and will do so again in two weeks.

Lead is famous for having bad developmental effects.  People will have grown up with damaged neurons not making the right connections.  Maybe Restoration works against that; it's worth trying - but Keltham wouldn't bet on it.

The obvious thing to try here to get faster info would be raising mice, rats, or some other mammal with shorter developmental times.  Keltham will sketch out some basic experimental techniques for measuring the thinkoomph of mice so that developmental damage can be quickly observed within an experimental versus control group.  That can verify the problems with lead and prove them to other countries; and they can also see whether Neutralize Poison or Restoration helps at all there, once a mouse has already grown up damaged.

It's - so bizarre that an element as high on the periodic table as 82 would be the most affordable metal for pipes.  Civilization uses complicated synthetic liners for pipes that carry drinking water, but Keltham knows that underneath the liners are Element-29 - what they currently think is 'copper' - and Civilization would use something failover safe underneath the linings, so copper pipes would probably be safe.

Keltham will look into copper mining, see if anything there stands out as obviously improvable.  Keltham was thinking about electrorefining / electroplating anyways.  Maybe they can line the lead pipes with a thin sheen of copper via electroplating, it wouldn't - wouldn't be a great idea - but it's probably better than exposing the lead outright - though Keltham will need to figure out, some way to check, that the metal-to-metal contacts aren't generating voltages and putting even more lead ions into the water -

It's - probably not what he should actually be prioritizing.

He should be prioritizing scaling up spellsilver.  It's famously bad for startup projects to split their attention.

He gave them warning.  He doesn't - understand, Golarion, but - he tried -

Having lots and lots of +4 intelligence headbands to sell is step one on a lot of things, not the least of which is earning enough money and credibility to scale up the Project further, get more people besides Keltham who could work out how to line pipes that carry drinking water...

Are they at least not - using lead cooking pots, to make food for children.

Well, Cheliax is requiring everyone to turn in their lead cooking pots and get replacements. Other places - might be doing that. A bit slowly, but they'll probably get around to doing it.

But - what are the Lawful countries, who aren't Cheliax - is there a stated reason they're not moving faster on this?  It's been three days!

- right, and things don't really happen in three days, most places. 

He doesn't understand.  Three days might not be a lot of time for him, but governments - have rapid-response teams, people working in shifts, precalculated plans for contingencies -

...none of that is true here, is it.

Okay.

The Project can solve all of those problems itself, if it has to, it'll just take months and Keltham should make sure spellsilver production is scaling first.

He'll ask a lot of questions.  Just to make sure he's not missing something.

The answers can all be almost entirely true, and swiftly supplied, since Cheliax actually did do the thing.

PL-timestamp:  Day 53

Keltham holds forth today on the answer to a question Pilar asked him a while ago, about dath ilan's view on what people should want, and how they should want it.

It's not really the way dath ilan sees things, that 'should'.  It's not advertised as a decision society gets to make for you - how would that even work, why would any agent agree to coordinate their own part, in that.  But if you want to know how an average dath ilani sees things -

So Keltham holds forth then upon the Light, and its origins.

Natural selection singlemindedly optimized humans for only one single simple thing, in the 'outer optimization target':  Inclusive genetic fitness.

But its optimization method was via accidental mutation producing many random tweaks, and selection keeping only the few tweaks that improved inclusive genetic fitness.

And when that kind of optimization produces entities that do their own optimizing, like humans, using that kind of hill-climbing method that moves slowly and in many directions through the search space - rather than analytically designing a solution - it doesn't reproduce the single simple constant 'outer optimization target' inside the 'inner optimizer' of mortals.

Like, at all.

Some humans explicitly want kids, and not just sex, is about as close as natural selection got.  Pretty few humans explicitly want to maximize the number of copies of their DNA in the next generation.  Keltham himself was a lot closer to having a goal like that than most dath ilani, and it really wasn't about the DNA, for him, it was about making a point.  Keltham definitely doesn't want relative inclusive genetic fitness; if he had two brothers, he wouldn't have been just as happy with each of those brothers having 144 kids, even though that's the same amount of total genetic-relatedness passed down.

So what do the 'inner optimizers' end up wanting, in a hill-climbing 'outer optimization loop' like that?

They end up wanting 42 major things and 314 minor things (on the current count of what's known and thought to be distinct in the way of adaptation) that correlated with having more kids - or with your relatives having more kids - across the range of varying situations and environments where the 'outer optimization loop' was optimizing.

Keltham doesn't have them all memorized.

But, for example, people generally want to eat food, especially if they haven't eaten food in a while.  Potential ancestors who didn't want to eat any food starved to death and failed to become actual ancestors.

People like green fields, blossoming flowers, the sound of running water in the distance, because people who gravitated to fertile places like that had more kids, in the ancestral environment.  Keltham isn't one of the people who have a strong need for that sort of thing - he's fine in this Fortress for a while yet - but if it starts to bother him at some point, he will have to employ Interior Decorating Principles that try to create interior environments that satisfy the same instincts.

Mothers love their children - as do fathers, if they get the kinds of cues that the ancestral environment would have delivered, about the kid being theirs, which dath ilan is correspondingly very careful to deliver.  You'd expect that to correlate with having more surviving kids who made grandkids, in the ancestral environment.  Mothers who casually tossed their babies aside did not become ancestors.

People who have brothers and sisters usually care about those; they share half your genes on average.  There's a joke about how you shouldn't sacrifice your life to save your brother or sister's life, but you should sacrifice your life to save 2 siblings or 8 cousins.  It's funny because, of course, human beings don't care at all about their inclusive genetic fitness once they learn that such a thing exists.  They're not aligned to that which created them.  Like, you can imagine some creator smarter than natural selection devising an analytic solution for things that actually pursued a particular goal, but if you look around at yourself and you're not that, you're not.  As humans evaluate shouldness, there's no reason a human should be aligned to their creator, natural selection.  There isn't any principle of Law saying that things created will end up aligned to their creators.

Keltham's lessons sure did emphasize that point a lot!  He's guessing it's because there's some silly thing kids might otherwise do, where they imagine that inclusive genetic fitness is the 'correct' or 'intended' goal behind all of the things that they actually want.  Like, they start incorrectly thinking that they're supposed to calculate how much reproductive fitness they'll get out of food, instead of correctly understanding that they want the food because it's tasty and that's all there is to their own utilityfunction.  If a mistake like that carried over to adulthood, people might start thinking that the objective point of loving their own children, was to spread their DNA, rather than to love their children, which is in fact the point if you're a human.  So it's important to understand that there's just no reason to think like that!

Thaaaaat does sound like somebody might've been warding dath ilani specifically against modes of thought that are common in Golarion, as Asmodia will highlight for Korva later.

It's fine to want lots and lots of things.  The ideal of being coherent isn't the same idea as being simple, not at all.  Lawfulness just says, don't make weird patterns of decisions where you could rearrange your resources, and get more of everything you wanted simultaneously.  If you're going to dance a complicated dance, that's fine, just don't step on your own feet.

The 356 known fragments of desire, the shattered correlated pieces of inclusive genetic fitness in the ancestral environment, are not like components of a utility function; they are not born into human beings as consistently valued long-term goals.  They are things people want, dislike, under particular circumstances, patterns of thought that they flinch towards, flinch away from.

Human beings are born, bluntly, as giant messes.

And when they weave themselves into anything more sensible than that - they should take care not to become less complicated, and leave anything out that they really wanted.

Life, consciousness, and activity; health and strength; the myriad pleasures and satisfactions that may be natural to a person, from good food to good sex; happiness, contentment; beauty, harmony, aesthetic experience; love, friendship, working together; justice and the receiving of earned deserts; power and achievement; self-actualization of what has been designated as virtue; self-expression, freedom; adventure and novelty; ultimate safety from the worst harms; the esteem of others; familial bonds; children and intellectual legacy; honor in one's conduct and trading...

All these are some things close to base human desires in the moment, that seem reasonably reifiable into components of the utility function - things to be pursued as ends in themselves, and not for the sake of anything else.  These are some of the things that dath ilan tries to measure when it measures its Planetary Utility Function Achievement Index, as is bet upon in important conditional prediction markets.

There's a lot of things to value!  And dath ilan endeavors not to leave any of it around free for the taking, if prediction markets say that it would be just as easy to do things a little differently, and get more of things designated as nice.

Ordinary life and politics in dath ilan, then, takes place close to the tradeoff-optimal frontier - where you can't get more of one thing you want, without sacrificing something else - which is the sort of situation you'd mostly expect to find yourself inside, if you'd otherwise planned and acted like a sane person during your previous life.

The ordinary politics of dath ilan are all about the tradeoffs.  Is it more important to have more and smarter geniuses in the next generation, or to avoid assortative mating between the Very Smart People as might lead them to begin to diverge from the rest?  There are benefits and costs on both sides...

Yet there is a word in Baseline that means the harmony and song between the many varieties of what dath ilan considers (on average) to be betterness in thought and action: kindness and altruism and honor and beauty.  It is a word for those rare occasions where, instead of tradeoffs, there is a nontrivial policy question on which utilitarianism and deontology and virtue ethics and aesthetics all agree.

A high-functioning society should present its individuals with only rare true moments like that, because usually things should not be going wrong in that many ways at once; there is a dath ilani proverb that, like acts of individual heroism, such moments should be found mainly in fiction.

Keltham didn't like that fiction, in fact.  He found it annoying.

Keltham didn't really understand, on a deep level, why anybody in real life needed the concept of what, in Baseline, would be called 'the Light'.

Until he got to Golarion and found that it - was apparently not that easy - to get the lead out of cooking pots being used to feed children.

That's it.  That's all Keltham had to say for today's lecture.

Does he just not realize how many problems are bigger than that one even if you're Good.

Well, lots of problems cause children to die horribly, which is worse if you're one of the dead children. You'd rather be an alive child than a slightly smarter dead one.

But this one makes them more worthless. It doesn't seem strange, to Carissa, that Keltham is fixated on it. If there were a reverse-lead, that made you more Lawful and smarter, she'd drink it all the way until she was a god.

It's not particularly the answer Pilar was hoping to hear, though, in retrospect, it's obvious enough that dath ilan wasn't going to have an Asmodean-helpful answer there.  They would think that Pilar ought to be crystallizing all of her human wants into her utilityfunction, instead of simplifying it down like Sevar spoke of, to 'wanting to be a good slave to Asmodeus' and then to simply serving Asmodeus as the Most High does.

She also needs to warn Sevar, later, against exposing Keltham to facts that activate his Good impulses.  She can see, now, some of what the Chosen was talking about when she spoke of mortals being quivering slimes that flinch about in disorganized directions in response to their own fleeting thoughts.  If they want to corrupt Keltham to Evil, they need to have him not thinking at all about children, or anything else for which he feels sympathy, for a long-enough period that his Evil impulses have a chance to crystallize unopposed into a full-fledged Evil utilityfunction.

"Are there any impulses on that list of 356 desires that dath ilan says to just cast out of yourself?" Pilar says.

She realizes only after speaking that she forgot to ask if alterPilar would have the same question.  AlterPilar could, obviously, but Asmodia is always sternly admonishing them all to ask what their alterselves would do, not could do.  You can rationalize anything as a possibility to yourself, it doesn't mean it'll fool Keltham about probabilities.

"Dath ilani tend to be pretty conservative about passing that judgment.  They wouldn't even have told you that I had too much self-interest, just that I had too little altruism - said that I wanted too little, not too much.  And they wouldn't have said that about me, either, wouldn't have told me I was wrong to be what I was.  Just, I wasn't what they hoped their own kids would turn out like."

"Human beings, I was told as a kid, have a 'vengeful' impulse," he uses the Baseline word instead of Taldane, to name a feeling rather than an act, "to hit somebody else back as hard as you think they've hit you, or harder, if you hold yourself injured by them.  That way, they won't dare injure you in the first place, right?  The obvious atomic problem with that baseline emotion, from the standpoint of the Law of Decision, is that it's a 'punishment', something you'd do only because you expected the other agent to respond to threats.  The obvious problem with it molecularly is that if people disagree about who owes the net debt of revenge, they get locked into an infinite resonating retaliation.  The obvious problem with it globally is that if you've got lots of agents running around trying to invert each other's utility functions, for acts of vengeance or any other reason, it's pretty easy to imagine Reality as a whole getting into a state where everyone's utility functions are being inverted all the time and you'd all collectively prefer unleashing Rovagug on the whole thing."

"So to first order, Civilization tells children, yeah, don't just flow along with your first instinct about revenge.  Channel it into something more organized, with better effects on the larger gameboard."

"When somebody stops coordinating with you, stop coordinating with them.  It's not 'punishment' to Defect against somebody who's Defecting against you, in a cooperation-defection dilemma; you're just withdrawing the Cooperation you were trying to coordinate with them.  It's not 'punishment' to catch a thief or send them to the Last Resort; it's simply not in your own interest to have them wandering around taking your stuff in the future."

"But you can still draw on something like a remaining trace of the emotion of revenge, for that.  And that reflects a meta-level policy of rescuing as many emotions as possible, rather than throwing them away and trying to run on pure Law."