Flat Earth Economics

    Say you're trying to navigate around your neighbourhood and you want to keep the total distance travelled to minimum. Let's pretend you're not limited to roads, and can always travel in a straight line to any particular destination. You go 4 km to your first destination, and then turn 90 degrees to your left to go 3 km to the next destination, and then you can go home. By the Pythagorean Theorem, the straight-line distance to get home is the square root of (3 squared + 4 squared), or 5 km.

    That's the correct answer, or close enough for any practical purpose. In reality the world is not a flat surface, but the curvature is so big that for distances of only a few kilometres we can just treat it as flat without any significant error in calculation. Or, to put it another way, the 40,000 km circumference of the Earth is so big compared to 5 km that we might as well treat it as infinite, which is what it would be if the Earth were in fact a truly flat surface.

    Of course, the math breaks down when you start dealing with bigger portions of the whole planet. If you go 10,000 km on a flat surface, turn right and go another 10,000 km, you'll be about 14,000 km from your starting point, but on the Earth you'll still only be 10,000 km away. And if you travel 20,000 km from your starting point, the next step you take, in any direction, will take you closer to your starting point. The farthest you can possibly be from any other point on the planet is 20,000 km. (I am assuming all distances are measured along the surface, rather than taking a shortcut through the mantle...)

    Scale makes a difference. It's fine to approximate small portions of the planet's surface as flat, but it's a mistake to apply those assumptions to the whole. And the same principle applies to economies.

    When I sit down to balance my household budget, the total number of dollars I have anything to do with is a negligible fraction of all the Canadian dollars circulating around out there. For my purposes, I might as well treat the dollars that aren't mine as infinite in quantity. If I spend $100, it'll take me $100 of income to get back where I started. Simple, straight, flat-plane geometry. 

    And that's more or less true for most businesses and even local municipal governments or individual government agencies. If you spend X dollars, you generally need at least X dollars in revenue to cover it. 

    But that's when the dollars you have anything to do with are a negligible fraction of the sea of dollars sloshing around out there. At the level of a national government with a sovereign currency, that's no longer the case. The federal government of Canada has something to do with literally every Canadian dollar in circulation anywhere, because every Canadian dollar is a creation of the Bank of Canada. To continue with our analogy, the Canadian government operates at a scale that encompasses the entire globe of the Canadian economy. 

    There's a lot more to this than I've described here, and I don't mean to suggest that national governments are completely immune to the sorts of considerations that apply to private individuals and corporations. Nor am I making specific claims here about the mathematics of government debt and deficits and how much spending and taxation is appropriate. All I'm saying at this point, in this post, is that the common rhetoric about government spending is based on an inappropriate comparison. Yes, it's completely reasonable to worry about revenue and expenditures at the scale of the individual household or business, just as it's completely reasonable to use Pythagorean calculations while navigating around your neighbourhood. It's just that the math is different when you get to global scales; you have to take into account curvature if you don't want to get lost.

Rhetorical Heavy Lifting

      I get into a fair number of arguments, and I daresay I'm reasonably good at it. I tend to "win" more often than I lose, if you define it in terms of persuading people that you're more likely to be right. (I actually think it's counterproductive to think of argument that way, as I elaborate on here.) I prefer to think of it as a process leading to greater understanding of the issue by all parties, a discussion rather than a debate, but there is usually some adversarial element, and I can usually hold my own pretty well when that's the case.

    So one thing that sometimes happens when I appear to be "winning" a debate is this: my opponent expresses some frustration that I'm only winning because I happen to be skilled at rhetoric, and if only they were better able to express their ideas more clearly, they'd be able to convince me. And that's certainly a possibility; there are many subtle concepts that are very difficult to express clearly but which turn out to be true (or at least, to have great explanatory/predictive power). But often difficulty in communicating an idea isn't so much due to a lack of rhetorical ability as it is due to the idea itself just not being as well-formed and coherent as it feels. That is, the feeling of being right or of knowing something to be so isn't identical with actually being right. (Like when I dreamed I came up with a mathematical proof of the immortality of the soul.) Goshdarnit, I know I'm right, but I just can't put it into words!

    A metaphor I've found useful for the way debates like that go is that it's like a rock-lifting contest. I choose a rock and you choose a rock, and the winner is the one who can lift their rock the highest. Now, you might think that what you need to win such a contest is to be very strong, but in fact, most often the winner is the person who is able to choose the lightest rock. 

    Now, skill at rhetoric is like being very strong in that it allows you to lift bigger argumentative rocks higher than you would otherwise be able to lift them. But the bigger impact is that it makes you a better judge of the weight of rocks. So much of the time, when someone tells me they'd be winning this argument if only they were better at rhetoric, I want to say hey, don't feel bad. I couldn't lift that rock, either.

    And to do that, I have to make a good faith effort to try to lift their rock.

    

Not All Humans

     When I was very young, my feelings were hurt when the cute little squirrel ran away from me, when all I wanted to do was pat it. Why was it scared of me? Why did it think I was mean? 

    In time, I understood that in the eyes of a squirrel, I was just a big non-squirrel animal, in a world where squirrels are often eaten by big non-squirrel animals. And so I wished I could talk to animals and they could understand me, so I could tell them that I was a nice human and I just wanted to be their friend. Not all humans are mean!

    But then I learned about lying, and understood that even if I could tell squirrels not to be afraid of me, that's exactly the sort of thing a human would tell a squirrel to make it easier to catch and eat. And so a smart squirrel shouldn't believe me when I said I wasn't going to eat them. And that, too, hurt my feelings, to understand not only that there were bad people who told lies, but that because there were bad people who told lies, I should not expect people (or squirrels) to trust me by default, however earnestly I might believe myself to be worthy of trust.

    I also came to understand that, if I genuinely cared about the well-being of squirrels, I should not want them to trust humans (including me) too easily. It is in their best interest to be wary of potential predators, even if the majority of us larger animals have no interest whatsoever in eating a squirrel. And if I still really want to befriend a squirrel, at the very least I should expect to have to earn their trust.

    This applies to my fellow human beings as well, though obviously with somewhat different parameters, since humans are a social species and live in communities where a certain amount of default trust is essential, but where there is always potential for betrayal. And so there are appropriate boundaries. It's no big deal if a stranger at a bus stop asks you what time it is, but unsettling if even a fairly close friends asks to look through your smartphone. The very attempt to take shortcuts across these boundaries is itself deeply creepy; there aren't many bigger redder flags than "What's the matter, don't you trust me?" 

    And protesting #notallmen! whenever someone talks about misogyny or #metoo or anything of the sort is exactly that kind of red flag. Just as you shouldn't take it personally that a squirrel doesn't trust humans, you shouldn't take it as a personal affront that you're not immediately assumed to be one of the good ones, and you're not entitled to the benefit of the doubt. No one is, and it's childish to expect otherwise.

What if it wasn't Trumpists?

      Some defenders of Donald Trump have been trying to claim that real troublemakers at the Capitol riot on January 6 of this year might have been Antifa infiltrators, perhaps trying to make Trump look bad. This is a pretty preposterous idea, but let’s take it at face value for the sake of argument. In fact, let’s go all the way and presume that all of the rioters, every single one of them, were there specifically for the purpose of discrediting Trump and his movement, and that all the real Trumpists were all perfectly nonviolent and understood that they should under no circumstances give even the appearance of a threat of violence, and so they scrupulously avoided any kind of unlawfulness. Fine. Let’s say that’s the case, and that 100% of the mayhem was deliberately engineered by a coalition of Antifa, BLM, Democrats and never-Trumpers, just to embarrass Trump.

     That doesn’t exonerate him, and indeed maybe makes it even worse. Why? Because Trump's own actions made it way, way too easy for the rioters to fool us all into believing they were his supporters. He goaded on a crowd with his speech, and whether he actually meant for them to force their way into the Capitol and break stuff, anyone in that crowd could be forgiven for believing that’s what he did mean. It was a completely plausible interpretation, and if it was a misinterpretation, it was a completely predictable one. In failing to predict it, Trump essentially gave his enemies a blank cheque, which they predictably would have cashed. 

     Now, I don’t actually believe that any of the rioters were secretly Antifa. The simplest, most parsimonious explanation is that they were exactly what they appeared to be: a mob of Trump dupes and deplorables. There might also have been a few foreign intelligence agents seizing the opportunity to gain access to the Capitol and plant listening devices or steal laptops, but again, Trump himself bears the blame for creating that opportunity. It never could have happened but for his ill-conceived encouragement. 

     So it really isn’t a defence of Trump to allege that the riot was carried out by his enemies. It’s true, though, because he is and always has been his own worst enemy.

"Do your own research"

     "Do your own research," they say when you ask for evidence or support for whatever conspiracy theory they're advocating. "I'm not going to do your homework for you," is another one they'll sometimes throw in. 

     This is an evasion tactic, of course. If they had good evidence and understood it well enough to be so confident in their conclusions, they'd be able to explain it to you. But it's not just about evasion; it's also about posturing, trying to imply that they're smarter and more informed than you and their time is too important to be wasted on imparting their vast knowledge to you. 

     As a general rule, if you make a claim then the onus is on you to provide support for the claim, so telling someone to "do their own research" when you tell them the latest conspiracy theory is just lazy bad form. But the reason this ploy resonates is that "do your own research" is generally good advice. The trouble is that the "research" they've done usually amounts to following whatever rabbit hole the YouTube algorithm generates for them, and those rabbit holes often lead to echo chambers.

     So what should it actually mean to do your own research? Broadly speaking, it means to gather evidence, evaluate its credibility and weighting, and attempt to synthesize it all into a coherent theory that allows you to make reliable inferences. The precise methods you use to gather primary evidence may vary from field to field, but in practice most of the evidence you gather will be the reports or testimony or work of other people, and so a good part of your task will be deciding how reliable these sources are.

     And here's where the "do your own research" crowd go horribly wrong, because the source whose credibility it's most important to evaluate is yourself. When they exhort you to do your own research, there's often the implication that it's lazy or stupid to trust the research of other people, as if your own research is inherently better or more reliable. But your opinions and beliefs are just as likely to be wrong as anyone else's, and you should not treat them as the answer key that everyone else's answer should match to be deemed reliable. 

     Here's an example I've used before. A doctor recommends surgery, and describes the procedure to you. Upon realizing that she's proposing cutting into you with a knife, you reject her expertise, because even you as a non-expert know that cutting people is bad. This is the wrong way to judge the doctor's expertise, because while it's true that any expert should know that cutting people is bad, the true expert may also know other things (such as how to stitch people back up, and how to cut them so as to make stitching the back up easier) that make it not so bad. 

     Instead, the better approach is to say, "Wait, you're going to cut me with a knife? Won't that hurt? Won't that put me at risk of bleeding to death?" and listen -- listen -- to understand the answer. Obviously if the doctor is surprised to learn that cutting people is generally bad, they're probably not a real doctor, but a real doctor is prepared to answer this sort of question honestly and reliably. You don't necessarily need to understand every detail of what they're telling you, but if you have a good basic lay-person's grasp of the vocabulary and subject matter, you should be able to tell when they're just making up stuff and bluffing. 

     The trouble is, not everyone has a good basic lay-person's grasp of the vocabulary and subject matter, and even if you do, interrogating an expert to decide if they really know what they're talking about takes a lot of effort and time you don't always have. This is why we have shortcuts: credentials and certifications. If someone has a PhD from an accredited university in the subject matter, or a license to practice the profession in question, it's reasonable to assume they are in fact qualified experts. It's not an absolute guarantee, of course, but in general it's reasonable to defer to their judgment in their particular field, and checking out their credentials usually enough to qualify as having "done your own research" before you adopt the results of their research.

     It's important to be clear here. You do have to trust your own research in one sense: whatever conclusion you adopt, whether it originates with you or some other source, it is inescapably the result of your decision. You're stuck with it; there's no "I was just following orders" absolution. You have a choice what conclusion to adopt, but you cannot choose not to choose; even suspending judgment is a choice (and often the right one). And more often than not, the wisest judgment is to adopt the conclusion of the experts.

About that vaccine video...

      A friend just sent me a video (which I will not share here) asking me for my synopsis of its content, and it occurs to me that many others may receive this video and be seeking answers to the same question, so I thought I'd do a post about it. The video is from someone examining the package of the new Astrazeneca vaccine, and calling our attention to certain words printed on the packaging they think we should be very alarmed about. 

     Now, the usual caveat should apply here: I AM NOT A TRAINED SCIENTIST. However, I've learned a lot of the very very basics from a lifetime of being a nerd very interested in science, and I have a strong background in general critical thinking and making sense of stuff, which I suspect is why I have friends who come to me to ask about this kind of thing. As well, my son us currently studying cell biology and it's been fascinating talking with him about how much more staggeringly complex and beautiful and interesting these things are than I ever imagined (and I've always imagined them to be pretty darned cool). What I'm offering here is not an expert explanation, because I'm not an expert. It is simply an educated lay-person's reading of the matter, intended as an alternative to the panicky less-educated lay-person's account given in the video. 

     I'm not sharing the video here for a couple of reasons, but mainly because I don't think it needs more bandwidth, and also because I suspect there are multiple similar videos and forwarded emails making the same claims. I will attempt to charitably present the concerns it raises, and then explain why they aren't nearly as problematic as it seems from the video.

     The first word the video attends to in the vaccine packaging is "ChAdOx1-S (recombinant)". The narrator seems to think "ChAdOx1-S" is just a meaningless serial number name of the vaccine itself, and focuses on the word "recombinant" as the frightening bit, which is in a way kind of amusing for reasons I'll get to in a moment. "Recombinant" means more or less what it sounds like: re-combining DNA from two or more different organisms. This is actually not a new process: it happens every time a baby is conceived or a flower is pollinated. Even asexually reproducing bacteria often absorb bits of genetic material from various sources, sometimes incorporating it into their own genome. And naturally-occuring retroviruses splice their own code into the genome of the cells they infect. 

     What is new is our ability to do this artificially in a test tube, which we've only been able to do for a few decades. And it's tremendously useful, first for research and later as we get better at it for practical and therapeutic applications. If you don't know exactly what a strip of DNA does, you can sometimes figure it out by snipping it out of a cell to see what happens when it's gone, and then splice it into a cell that normally doesn't have it to compare the results. And then when you understand things better, you can do stuff like take the gene that produces insulin and splice it into some E. coli to produce this important life-saving hormone in industrial quantities without having to harvest it from animals. 

     So the ChAdOx1-S (recombinant) vaccine is, presumably, a vaccine made by recombining a sample of genetic material from the pandemic virus with some other genetic material that made up the precursor to the ChAdOx1-S vaccine. In other words, the vaccine is the result of a recombinant process, not something that will cause a recombination in your own cells. But if you thought that it was the latter, then the original vaccine is even scarier, because it turns out that "ChAdOx1-S" actually refers to Chimpanzee Adenovirus Vector 1, evoking a terrifying Island of Dr. Moreau scenario. So I find it hilarious that the person in the video missed this detail. But of course, it's not actual chimpanzee DNA; it refers to an adenovirus that infects chimpanzees. And humans and chimps being very very similar, viruses that infect chimps can often infect humans and vice versa.

     A vaccine is often just a de-activated version of a virus, something that resembles the actual virus enough that the immune system learns to recognize it as something to be destroyed, but isn't actually itself infectious. Think of a wanted poster: it has an image of the face of the bad guy, so you know what he looks like, but the wanted poster can't rob your stagecoach. But a wanted poster is more than just a photo: it also contains information that alerts you to why you should beware of the guy in the picture, whom you should call if you see him, maybe a reward or other motivation for doing so, and so on. 

     So it's not enough to just present some molecule to the immune system. You have to present it in a way that the immune system will recognize it as a pathogen and start producing antibodies against it. It's like you have to include all the "WANTED" text from the poster, except that with molecules we don't know how to generate all the relevant text from scratch. So a recombinant vaccine is sort of like taking a successful wanted poster you already have for some other virus, and cutting and pasting a photo of the new virus into it.

     That's what I think they've done with ChAdOx1-S (recombinant). They've taken a vaccine that seems to work for the chimp adenovirus, and spliced in some part to make it work for the new pandemic virus.

     The next bit the video gets very alarmed about is something called "MRC-5", which turns out to be a human cell line derived from fetal lung tissue. It sounds like the person in the video is deeply disturbed that aborted human fetal tissue is an ingredient of the vaccine you'd be injected with. It's not. Human cell cultures are a kind of lab rat: they test the vaccine on those cultures, to see how it affects human cells. They do not use it as an ingredient in actually making the vaccine itself.

     Now, you might have moral reservations about using a product that was tested on aborted fetal cells, but there's an important detail I need to point out here. Nobody is getting pregnant to produce fetuses for the purpose of harvesting tissue. These are fetuses who were going to be aborted for other reasons, possibly even naturally as a miscarriage. Using fetal tissue for medical research is no different, morally, from using the tissues of an accident victim. You can regard the abortion itself as an appalling tragedy, and so it might well be, but so is the death of any other organ donor; that doesn't make deriving some good out of their tragic sacrifice inherently immoral, especially if we are appropriately respectful and appreciative.

     (Also, it's interesting to note that the particular fetus MRC-5 is descended from died in 1966. Fetal cells are particularly useful for culturing because they're so early in the development process, and have so much more growth ahead of them. It's usually easier to make them immortal than it is to do so with cells of a mature adult. That said, the first human cell line to be immortalized came from Henrietta Lacks, a cancer patient who died in 1951. Cancer's weird that way.)

     Finally, the video emphasizes a passage in some of the research documents about the vaccine calling for AI resources to go through the high volume of expected ADR ("Adverse Drug Reaction") reports and make sure every detail is recorded and analyzed. Yeah, at first glance, this sounds scary, like they expect the vaccine to be horribly dangerous and hurt a whole lot of people. But here I want to repeat a theme I brought up here: if they know the vaccine is going to have a lot of ADRs, and they still intend to go ahead with it, what are we missing? Either we should assume they're diabolically evil or stupid, or maybe, just maybe, having a lot of ADR reports isn't quite the terror it seems.

     We already know that developing a vaccine is going to take (has taken) a long time, and a very large part of that is safety testing. They know that there are always risks with developing any new therapy. And they know they're going to get a lot of reports of adverse reactions. A report of an adverse reaction, however, is just that: a report. Many, probably most, of those reports will turn out to be something else. Someone gets a shot, and happens to get totally unrelated food poisoning the next day. Someone else has a heart attack. Someone else doesn't realize yet she's pregnant and reports some of her symptoms as a potential adverse reaction. When you're testing a drug, you want all of this data, whether or not it's actually related to the drug, so you can look through it all for patterns to figure out what, if anything, really is due to the drug and what isn't. And finding those patterns is an absolutely monumental task, which is why an AI system would be so incredibly useful in sifting through all the data.

     I am not saying that there is nothing to be wary of with the Astrazeneca vaccine, or indeed any of the new vaccines they're bringing out. There's a lot of pressure to get these vaccines in use very quickly, and it's not unreasonable to fear that corners might have been cut, or there just hasn't been enough time for unknown side effects to become apparent. Of course there are risks; the real question is, as always, are the risks of not being vaccinated greater or less than the risks of being vaccinated?

     What I am saying is that many of the fears people have of this new vaccine are unfounded and based on an extremely incomplete (even more incomplete than mine) understanding of what these words mean. When terrified people urge you to "do your own research", understand that research involves more than just googling; you need to know how to interpret the words you're looking up, and how they are actually used by the experts doing the work. 

     A little knowledge is a dangerous thing, especially when you think it's a lot. 

Some thoughts on lying, and why I (somewhat) trust the mainstream media

     I've written elsewhere that lying is the strategy of the stupid. That's not to say that everyone who ever lies is stupid, or that there can never be a time when lying can be done intelligently. Just that, as a general rule, it's better to tell the truth than to lie, and not just for moral reasons. I have two arguments here.

    First, I want to suggest that intelligence generally is in some sense about truth. That is, if we define intelligence as an overall problem-solving ability based in the acquisition and application of information, the solutions that intelligence comes up with for problems will tend to be better when the information it uses is true. That's not to say that intelligent people are those who know the truth, or that they don't deal in  hypotheticals. Quite the contrary, because being able to entertain and explore counterfactuals is itself an important and useful way to discover more truths. Rather, I'm saying that the enterprise of intelligence is largely about evaluating what things are or are likely to be true, and making decisions or choices that take considerations of truth or falsehood into account. An intelligent solution to the problem "Should I bring an umbrella?" will be one that considers (among other things) the likelihood that "it will rain" is a true statement.

     Now, there is a difference between wanting to know the truth yourself and lying, which is wanting someone else to believe a falsehood, so it doesn't necessarily follow from any of this that smart people would not lie. I'm making a softer claim here, namely, that as intelligence consists in large part of habits of truth-seeking, there is necessarily going to be a certain amount of conflict between one's inner dialog of truth-speaking and an outward practice of lying. The habits interfere with each other, and while that doesn't mean an intelligent person cannot lie (after all, intelligence is about solving problems, including the problem of lying), it does mean it's more work.

     And that leads into my second argument: it really is more work. Any meaningful lie you might tell has some way it might be revealed as a lie. If I say that it is raining, and you look outside and see it's not, you know I'm an unreliable witness. But most lies are about things that are a little harder to disprove, and the best lies are the ones that go completely undetected as lies, which is more likely if they are nearly impossible to disprove.

     Recall that we're talking about intelligence as an overall problem-solving ability, and note that confirming a statement as true or disproving it as a lie is itself a problem calling for an exercise of that intelligence. When you craft a lie and decide whether or not to tell it, you will want to consider how easy it is to disprove it, but here you are limited by your own intelligence. Just because you think disproving your lie would be prohibitively difficult doesn't mean someone else might not find it trivially easy. 

     The problem is compounded for people who, in addition to not being particularly bright to begin with, wrongly think they are significantly smarter than average, because they will tend to believe that a difficult problem for them (unmasking a lie) would be downright impossible for lesser minds. But it's important to recognize that intelligence isn't a simple linear quantity, and smart people recognize that there's a lot even they don't know; some absolute moron might just happen to know the one crucial fact that shatters an otherwise impenetrable lie. An intelligent person knows that for every way they can imagine their lie being discovered, there are a thousand ways they haven't thought of.

     That's why I say that lying is the strategy of the stupid. The stupid tend to think that it's easy to lie, and of course superficially it is: all you have to do is say something that's not true. But that's the shortcut. Telling a robust, consistent lie that will withstand concerted and intelligent scrutiny? That is ferociously hard.

     And here it's important to point out that the same superficial shortcut applies to the business of rejecting a falsehood. It is also trivially easy to dismiss some claim as a lie or "fake news"; you don't need to consider a shred of evidence. Boom. "Liar!" and you're done. Of course, since you can do this equally well with any claim regardless of its truth or falsehood, it has zero probative value.

     So why do I tend to (somewhat) trust the mainstream media when they report that, for example, Covid-19 is a pandemic that's killed over a million people worldwide in the past year (1.49 million as of this writing)?

     It's not because I think media companies have our best interests at heart or that they somehow find the idea of lying morally repugnant and would never ever dream of doing such an evil thing. True, I tend to think that the majority of people employed in reporting and publishing the news, or in almost any industry, are probably decent human beings who aren't completely diabolical and might balk at the more obvious sins asked of them, but I'm well aware of how decent human beings can be gradually and subtly corrupted by an unjust system, so I have little doubt that mainstream news sources would lie like crazy if they thought it were in their interests to do so and they could get away with it. 

     But that's just it. It isn't in their interests to be caught flagrantly lying. There are, of course, powerful economic interests behind most every news outlet, and there's definitely a bias in what gets covered and what doesn't, and pretty strong spin in how any particular issue is reported, but when it comes to straight up lies? Those can and usually will be revealed somehow, especially given that there are multiple competing news outlets who would just love to discredit each other. If they could. 

     And some "news" outlets do take shortcuts discrediting their competitors, of course. They boast to their viewers that only they can be trusted, that their competitors are just full of lies. Such a lazy shortcut really just discredits them, because it is so lazy and so independent of the truth or falsehood of the claim. 

     I don't buy the claim that mainstream media is just full of lies, because the overall coherence of the stories is just too damned hard to fake. Yeah, there's going to be lots of stuff in the papers that's wrong, misreported or spun, and sometimes just plain lies. That's so with all sources of information (including and perhaps especially the ones decrying everyone else as "fake news"), and there's no getting around the hard work of evaluating and assessing and integrating all the data into a coherent world view. Just tossing the bulk of the data into the box marked "lies" is a lazy shortcut.