Sister blog of Physicists of the Caribbean in which I babble about non-astronomy stuff, because everyone needs a hobby

Wednesday, 20 April 2022

Slaying the utility monsters

I've previously gone on something of a rant against utilitarianism, but since reading On Liberty (I will eventually finish that planned post series, I promise) I've become far more sympathetic

As I see it, neither liberalism nor utilitarianism provide a complete moral system, but both provide important parts of the whole. In particular, you can't treat moral actions as a linear sum : allowing the happiness of one to balance out the misery of the many is pretty much guaranteed to lead straight to hell. Rather, you have to actively seek first and foremost to minimise the suffering so that as few people as possible experience the smallest amount of suffering possible. Minimising suffering is the priority, over and above increasing happiness.

Furthermore, whatever action you take you have to justify everything by some independent criteria : you don't ever cause suffering for the explicit purpose of improving happiness, otherwise you're in cliché territory about the ends justifying the means. Your goals are to minimise suffering and maximise happiness (that seems a pretty safe set of goals for any would-be social engineering*), but you can't let your goals dictate what is just and fair. The two must be kept strictly independent. Furthermore, in a liberal system you should always seek to employ the smallest possible action that gives the greatest possible gain.

*Yeah, this is a simplification, but without simplifications we'll get nowhere.

On then to this post. I believe the above caveats should address most of the objections, e.g. :

Say you’re a surgeon. You have 5 patients who need organ transplants, plus 1 healthy patient who is compatible with the other 5. Should you murder the healthy patient so you can distribute his organs, thus saving 5 lives ?

Duuhhh, no ! This would be letting the goals dictate justice. And going around murdering people for their organs is so obviously wrong that I will have Words with anyone who thinks I need to explain why that's the case. Of course it's wrong. You know it, I know it, so let's not pretend otherwise. Never mind that like the Trolley Problem the details of the situation (which are here lacking) could profoundly change the moral problem being addressed, you just can't justify murder.

Most of the other examples are very similar : 

You have a tasty cookie that will produce harmless pleasure with no other effects. You can give it to either serial killer Ted Bundy, or the saintly Mother Teresa. Bundy enjoys cookies slightly more than Teresa. Should you therefore give it to Bundy?

Egads no. Why should I ever reward the evil ? First I must minimise suffering, which means punishing the evil, not giving them cookies. Good grief. Enjoyment alone cannot possibly directly equate with moral good.

There is a large number of Nazis who would enjoy seeing an innocent Jewish person tortured – so many that their total pleasure would be greater than the victim’s suffering. Should you torture an innocent Jewish person so you can give pleasure to all these Nazis?

Pfft. Again, happiness does not equate with morality, that's plainly daft. And as I've said before, because they can't be mathematically quantified, you can't you linearly sum happiness and misery anyway. That's like trying to take the smell of rancid cheese and divide it by a sausage : it just doesn't make any sense at all. You've just lost Godwin's Law, methinks.

...because failing to save lives is just as wrong as actively killing...

This is bollocks, though, isn't it ? I mean yeah, if you see someone drowning and can throw them a lifebelt but actively choose not you, sure, you've basically murdered them. Choosing your life career based around what makes you happy instead of what saves the most lives isn't the same at all. In the former, you have a straightforward binary choice. In the latter, you have a whole slew of hypotheticals and a maddeningly complex optimisation problem : what exactly does save the most lives ? Being a doctor ? Being a politician out for healthcare reform ? 

I think again these things cannot be so easily quantified. And you cannot spend your entire life constantly worrying about other people because that's just not human and will inevitably lead to your own ruin. Sorry, but it's just not nearly as simple as choosing to become a medical doctor versus an athlete, because that's not how people work. And any moral framework must be constructed with respect to how real human beings actually function, rather than treating them as linear machines capable of any arbitrary action.

So yes, I agree with piece as far as that particular conception of utilitarianism goes : if we treat pleasure itself as moral good, and we seek to maximise this with no other concerns... absolutely, that is a catastrophically appalling idea. But I don't see any reason why we have to take this useless definition of "utilitarianism" to be the only possibly interpretation of it. It seems to me that there are far better, more worthy implementations which deserve consideration instead. 

Tuesday, 19 April 2022

Make it worth defending

One of my favourite quotes comes from someone about whom I otherwise know exactly nothing. When asked if Fermilab contributed to the defence of the United States, the founding director R. R. Wilson allegedly said :

No sir, I don’t believe so... Nothing at all. It has nothing to do directly with defending our country, except to make it worth defending.

Given that a lot of the following rant is about lies, you might want to read both of those links, though the accuracy of that particular quote is not relevant. It's the principle that's important.

First we might briefly consider this rather nice piece from the Atlantic, which in brief summary says that Trump supporters believe the election was stolen because of vague feelings and intuitions that it should have been. They've been primed over many years to believe Trump and disbelieve anyone else. And like I’ve been saying for a while, the point of misinformation isn’t (usually) to convince but to confuse. Once you stop trusting facts, you fall back on gut instinct and emotion. While you can often and all too easily manipulate statistics toward whatever end you see fit, it is altogether harder to do this when it comes to the final binary choice : who got more votes ? At this level data cannot be bent or broken. Facts are facts. 

Feelings, however, are altogether more malleable. Sowing distrust ahead of time means that when you reach that final choice, the data no longer matters. Feelings are all that’s left : the election must have been stolen - so the unconscious thoughts go - because the alternative does not fit my intuition.

It's a nice article, but I do have one major critique :

These voters aren’t bad or unintelligent people.

Yes they are. They are a bunch of bigoted racist, filter-bubble-brained, idiotic, useless twats. I could go on, but I now turn my attention to my main target : the Tories.

Boris Johnson is a pathological liar. We already know this, but he insists on reminding us so often that one sometimes feels he's deliberately trying to one's nose in it. Much like Trump proclaiming he could shoot someone, Johnson's verbal equivalent of writing an apology in Comic Sans is damn close to saying it doesn't matter if he broke the law or not, he can do what he wants.

That Johnson attended parties is itself shameful. There's no way around this. The vast majority of people stuck rigidly to the rules despite having to make enormous sacrifices to do so. For some, staying inside and watching Netflix is hardly a burden, but for plenty of others, missing essential contact with loved ones is asking the Earth. If you wanted to go to the pub then it sucks to be you, but if you had to attend a funeral, or wanted to see your ill or elderly relatives, then this is altogether different. 

Tory claims that nurses and teachers were all indulging in similar parties is wrong every which way you look at it. It's wrong because it's factually untrue. It's wrong also because one illegal act doesn't justify someone's else's (I mean, good grief, do these people not have mothers ?!? This is parenting 101 !). It's just a plainly horrible thing to say : it doesn't matter that the Prime Minister broke the law because everybody else did too ? They bloody well didn't, because those laws were in place for a very good reason, you muppet. Not everyone's thought processes are such a bunch of bullshit, least of all nurses and teachers.

Then we have Johnson's claim that it never occurred to him he might be breaking the law. This is a farcically stupid thing to say, because his instructions were, for once, as clear as day. Everybody understood them. So for him to say that he and his cronies alone did not means that either he is at best monumentally incompetent or, worse, uncaring of the rules and their necessity. Either he's incredibly stupid or he's a liar. And these are not mutually exclusive conditions by any means - he could well be both. Stupid, malevolent or both, he clearly shouldn't be in charge anything at all.

In the end, though, the parties themselves are a secondary concern. Boris may have taken the definition of "political party" rather too literally, but there's no real chance that the events actually did contribute much to the spread of the coronavirus. What's actually dangerous is the lying. I need not summarise this because Boris lied at every turn. He lied when he said no rules were broken, he lied when he said he didn't know about the events, he lied when he said he didn't know he was at a party, he lies practically every time he turns up the Commons to fail to answer questions. He repeatedly refuses to correct the record on the most minor of issues. And in that classic behaviour of a liar, he continuously changes his story. It's just not credible to suggest that even Boris Johnson is so hapless that he's doing anything other than lying through his teeth. Being a buffoon doesn't automatically mean someone is nice.

Then we have the absurd and tiresomely predictable claims that we can't change leader during a war. Again, this is wrong every way you look at it. For starters, we're not at war, idiot ! One country we don't like is fighting another country which we do like. The only sense in which we're fighting is vicariously, by proxy. If you let this criteria prevent us from changing leader then we'll probably never be able to change leader again. Secondly, for God's sake, we changed Prime Minister during World War II. It's absurd to suggest that we can change leader during an actual, direct war but not during a remote, proxy conflict !

But this too is by-the-by. It's the lies that matter. Lies are how it all starts. Frank Herbert said that it's truth that comes closest to being the basis for all morality, and I think he was right. True, philosophically, establishing "truth" is hard, but we don't need to excogitate on the nature of reality in day-to-day politics. We all know the domain of sensory perception and evidence we're considering. We implicitly understand the applicant conditions of what's being asked, that by "Boris Johnson" we mean a collection of what we call atoms arranged in the repulsive shape of a blonde twerp, and by "party" we understand that a multitude of similar such badly-arranged atoms gathered for an explicitly social event. It doesn't matter that rigorously pinning down the precise definitions of even such things as "chairs" is remarkably difficult, because we know, at a deep, inexpressible, gut level, what it is we have to consider. We know what's relevant, even if a definition of "know" sometimes eludes us.

No, philosophy won't save Boris here. The only credible interpretation of the facts is that Boris Johnson wilfully and repeatedly misled Parliament because he wants to stay in power. Any other viewpoint is ridiculous.

It's been suggested that this doesn't matter because in other countries politicians are held to much lower standards. Well, fuck that. I say the fact that all these lies are over trivialities like cake are not excusable, but are exactly why Boris needs to be removed from office. How could you ever trust a man who goes to enormous lengths to evade and deceive over such petty details as to whether he had a slice of bloody cake or not ? How could you ever trust such a man to make the really big decisions ? No, war in Ukraine is not a justification for keeping such a twat in charge - it's the very opposite, a clear indication of an urgent need to kick him out. For all that Britain has done well with military support, it's been pretty pathetic on the humanitarian front, something a new leader could remedy but Boris and his racist foreign secretary surely won't.

Lies matter for two reasons. First, when you stop respecting the truth, you end up with Trumps. You end up with vile creatures like Priti Patel... here I must digress. There's no alternative to sending asylum seeks to Rwanda ? Are you fucking mad ?* You're saying that Rwanda has better facilities than us and there's nothing we can do about it ? What kind of idiot do you think I am ? I'll tell you an alternative : anything. Literally anything**. For God's sake, they're people crossing in small boats who are drowning out there. What the hell kind of threat do you think they pose ? What the hell are you so afraid of ? If it was up to me, I would simply let them all in. They're just people.

* I'm glad to see this idea being shot down in flames from many quarters, but I feel stronger language needs to be used. "Racist cunt" would be about right, but I'd be prepared to settle for "heartless bitch".
**Except possibly sending them to Russia. But sssh, don't give the evil witch ideas.

And to return to my point : ultimately, lies lead you to monsters like Putin.

Now I'm not saying Boris is comparable, though Patel might be. But once you start setting precedents like this you risk a very slippery slope. If breaking the law isn't a good reason to depose a Prime Minister, then just what is, exactly ? If the Prime Minister openly lying is acceptable, in what sense do we have a meaningful democracy ? None at all, it seems to me. None. He can just make up whatever excuse he likes about whatever he wants to do and the whole process of holding officials to account goes out the window. 

So yeah, today's Boris might hardly be someone who deliberately slaughters innocents (needless pandemic causalities aside, which are not the same), but this idea of accepting a proven liar in charge opens the door to things which are so very much worse tomorrow. Presuming that this won't happen is folly of the highest order. Without red lines for our leaders, we are setting them above the law. It is very hard to see how any democracy can survive such an eventuality. I'm not saying the country will instantly burst into flames - that's daft - but that road is an awful lot shorter than we might like to think. Our political system is too precious to take it for granted.

No, the cake doesn't matter. The parties aren't that important. The lies are what matters. To have a political system in which honesty is valued do much that even trivial lies are not excused is very much like Fermilab : it doesn't directly help us with defence, but by God it makes us worth defending. And to add something that R. R. Wilson probably wouldn't, but needs to be said : fuck you, Boris, you fascistic lying clown-faced imbecile. Go throw yourself down a well and rid us of your stupidity. Enough.

Monday, 18 April 2022

The Western square and the Russian tower

I'd just like to draw your attention to this old post on societal networks, which is relevant once again. Here I add some comments in light of the geopolitical situation.


In his 2018 book The Square and the Tower, Niall Ferguson described the different strengths and weaknesses of hierarchical, feudal societies and more free-form, network-structure market communities. What I keep thinking of is his description of how the strictly hierarchical Soviet Union appeared strong and stable to the outside world, but in the end it disintegrated extremely rapidly*. In part this may be because disagreement within a hierarchy is extremely problematic. Since information and commands can only flow linearly up and down, and there is not much sideways transfer of information, a single determined link that doesn’t obey the diktats (or finds out something that hierarchy just isn’t equipped to deal with) has to be dealt with incredibly harshly. In order for this particular type of network to function, disagreement has to be supressed, all too often with brute force. HBO’s Chernobyl does an excellent job of getting across a strong vibe of how the hierarchical structure created problems when dealing with the eponymous nuclear crisis.

* I never got around to blogging it, but in Huxely's Brave New World Revisited, the author was quite certain that the whole world would be Communist by about now.

For example, if your own advancement (or even just your ability to maintain your position) depends on the actions of your underlings, you have to make sure those underlings do what you tell them. This is true regardless of the organisational structure. But in a very strict hierarchy, being so linear in organisation your underlings have little or no access to anyone but yourself, so there’s a tendency to shout at them until they do whatever you yourself have been told to do by those above you in the hierarchy. The lack of sideways transfer (i.e. checks and balances between peers) means that one way or another you either get on with it or get out. Nobody’s giving you any reason to doubt your superiors. Nobody's able to question whether the orders are a good idea or not, because every relationship is one of power and control, not discussion and debate - there's no equality whatever. So it’s up to you and you alone to keep your underlings in their place. In essence, the system ensures that everybody’s continuously punching downwards.

At best, when disagreement occurs, the most efficient hierarchy can only change its goal very slowly - it takes a long time to propagate information throughout the whole structure, or even just to the top and back down again, especially if that information is unwanted. More likely, those in such a situation will simply remove the disagreement and replace any stubborn personnel with those who are more pliable, thus tending towards making more and more mistakes. The more strict the hierarchy, and the more complex the problems it has to overcome, the worse this will be. This may be why medieval feudalism was more successful, being not so strict as the Soviet version nor having problems as complex that needed to be solved (certainly nothing as complex as a nuclear reactor, at any rate). A comparison of medieval feudalism versus the hierarchies of modern dictatorships would be a fascinating study...

A flatter, more generic network structure, by contrast, is far more robust to disturbances*. Sideways transfers of information means that there are, in effect, continuous checks and balances - peers who are equal and have no authority over each other can much more freely discuss and argue without immediate consequence. This means that opinions can be much more easily re-evaluated in the face of contrary evidence. So in this organisation, just as you might now experience contrary information to what your superiors direct (for pure, flat networks are rare, most still have a degree of hierarchy), so too do your superiors themselves. Thus changing the goals becomes very much easier. Punching upwards is a lot less risky : not only because collectively the lower orders can more readily challenge the authority of the elites, but also because everyone is used to re-evaluation as a matter of course.

*I'm going to follow Ferguson and use "network" in this very loose sense. Obviously this is all strongly dependent on the network structure, as a hierarchy is itself a type of network.

Furthermore, a single stubborn rogue who disagrees with the status quo can be circumvented - there are alternative pathways by which any broad consensus can enact its goals, without needing to remove the problem. Disagreement can be comfortably tolerated, incorporated, and in the very best cases it is actively encouraged.

Indeed, a strong, robust consensus only arises because those who can disagree are able to voice dissent. Contrarians are valued because they’re not always wrong (except for the extreme cases, who are few enough that they can be neglected), nor inevitably fatal to any goals. Thus the network is able to formulate better goals in the first place, because it’s able to formulate more accurate conclusions as the basis for its goals. It draws on a much more diverse range of sources for its raw facts and can analyse them from a diverse set of perspectives and techniques. A consensus formed in such a way is incredibly powerful, and far less likely to make stupid errors than a hierarchical system.

Ferguson says that it takes a network to beat a network, and this seems extremely topical. The neo-Soviet Empire that is now Russia appears to be very hierarchical indeed, but not even a particularly efficient one. Its goals are stupid, its methods are stupid, its implementation is stupid, and it seems utterly unable to learn anything. It seems less like a pure feudal hierarchy and more of a series of disconnected, broken little sub-hierarchies with little or no overall coherency.

The opposing western powers, by contrast, form a highly complex network. They too have flaws. But while dissent within Russia simply cannot be allowed, within the west it’s something of an asset. True, it makes implementation of goals less than optimally efficient, and sometimes that even leaves it at a complete impasse. But the network learns. It considers alternatives. It doesn’t chuck anyone out just because they have one terrible suggestion - it keeps them around it case they have good ideas in the future. Even really extreme viewpoints can be considered because generally the numerically larger more moderate forces mitigate the worse excesses, and sometimes those extremists do turn out to be correct.

Of course, no network is ever a pure hierarchy nor a perfect ideal analytic system (we don’t even know what the latter really is). Most groups require a degree of hierarchy; a tendency towards groupthink through having too dense internal connections cannot ever be eliminated. But overall, the network vs. hierarchy analogy feels very apt. Not that this means Russia is necessarily prone to an imminent regime change, sadly, but it would go a long way towards explaining why the Russian government appears to be such a catastrophically absurd omnishambles. Essentially, it's filter bubbles on a grand scale... and Russia has fatally misunderstood liberalism. What appears as division and incompetence is only sometimes so. A lot of the time disagreements, however unpleasant they may be to experience, are actually a manifestation of the systems' greatest strength : considering alternatives, changing its mind, doubting, and above all, learning.

Whose cloud is it anyway ?

I really don't understand the most militant climate activists who are also opposed to geoengineering . Or rather, I think I understand t...