If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:
Super Thinking: The Big Book of Mental Models By: Gabriel Weinberg and Lauren McCann
Human Compatible: Artificial Intelligence and the Problem of Control By: Stuart J. Russell
Joseph Smith’s First Vision: Confirming Evidences and Contemporary Accounts By: Milton Vaughn Backman
The Cultural Evolution Inside of Mormonism By: Greg Trimble
The Worth of War By: Benjamin Ginsberg
The Pioneers: The Heroic Story of the Settlers Who Brought the American Ideal West By: David McCullough
Sex and Culture By: J. D. Unwin
It’s been another month where most of my thoughts have revolved around COVID-19. In particular, like most people, I’ve been thinking about the end game. It would seem to me that there are four ways out:
(Edit: In between writing this and publishing this I came across a spreadsheet that did a much better job of outlining the various options. You should probably just check it out and skip the rest of the intro.)
The one that everyone’s hoping for is the development of an effective vaccine. I’ve heard that Oxford is hoping to have something by September, which is faster than I would have expected, but I’m still not sure that gives us the “vaccine solution” much before the beginning of the year, and that assumes that there are no logistical difficulties in trying to get the vaccine to the billions who would need it. And regardless of all of that, even under this most optimistic of all scenarios, no one thinks we can maintain the current measures until then.
The second possibility is that we get so much better at treating it that it becomes no worse than similar illnesses. I’m not sure how close we are to this, mostly what I hear is news about how treatments we thought would work aren’t. That 88% of people still die even on ventilators, and that even young people are suffering strokes. Despite this, I would assume that we can’t help but get better, and it is true that the longer it takes someone to get COVID the more likely they are to get treatment informed by all the knowledge accumulated up to that point. But I don’t think this does or should play a major role in deciding when to open things up in the same way hospital capacity does.
The third possibility is we control things so well that we completely stop the spread of the disease. China claims to have done this, but that claim comes with a lot of caveats, and even if it’s true, it seems clear that we won’t be able to duplicate their methodology in the US.
The final possibility is herd immunity, which seems the most likely outcome, particularly given the limitations mentioned above. To get there a significant percentage of everyone will have to get COVID-19, and the only knob we can turn is how fast or slow that happens. Initially it appeared that, since we were going to need to get there eventually, the primary reason for going slower was to make sure the hospitals didn’t get overwhelmed, not to keep people from getting sick. Especially since slowing down happens to be really hard on the economy. Having done that It appears that in most places the hospitals aren’t overwhelmed which is awesome, but would also suggest that maybe the dial needs moved to a higher speed of transmission. Which is kind of what states are doing by reopening (Utah re-opened on Friday.) So my point is less that we’re doing anything wrong and more that people seem to have lost sight of the fact that herd immunity is still the most probable ending, and that such immunity is going to require that a lot more people get infected…
I- Eschatological Reviews
This book came to my attention after I read a review of it on Slate Star Codex, and if you’re just looking for a general review I would direct you there. When it comes to the actual contents of the book, I don’t have much to add, and given that I have another 8 books to cover I don’t think it’s worth repeating anything Alexander already said. No, what I’m interested in are the books eschatological implications, so let’s move straight to there.
What This Book Says About Eschatology
As has been discussed extensively here and elsewhere many smart people have significant worries about the AI control problem. That is, how do you ensure that if and when we get around to creating an artificial intelligence that it doesn’t end up doing things we would rather it didn’t. Things that might conceivably include eliminating humanity entirely.
Previous attempts to address this problem have notable weaknesses. The first challenge is getting the AI to obey our instructions in the first place, but even once you have mastered that issue, the AI might take your instructions too literally. The famous example being the so-called paperclip maximizer which takes a simple instruction to make paperclips and turns it into a drive to turn everything into paperclips, including us. This led to people imagining that the instructions needed to include a clause for making us happy, which led to other people imagining an AI which stuck an electrode directly into the pleasure center of our brain, which they labeled wireheading.
As one of the key features of the book, Russell offers up a new system which is designed to solve these previous problems. It revolves around the idea of telling the AI it needs to keep us happy, but giving it very little information on what that means. This forces the AI to come up with guesses on how to make that happen with each guess getting a certain probability of being correct. Then it uses our behavior as a way to update that probability and narrow things down to the best guess. And, If our behavior is information, it’s not going to stop us from doing anything, because it wants the information encoded in our actions. Meaning it won’t stop us from shutting it off, because that’s potentially the most valuable information of all.
To use the example of an order to make paper clips, the AI might make two guesses it might assign odds of 30% that we want a big bar of metal to be made into paperclips and odds of 70% that we want the dog to be made into paperclips. This is obviously incorrect, and exactly the kind of thing we’re worried about, but under Russell’s proposal when we race across the room and snatch the dog out of it’s robot pincers it will use that information to change the distribution to 99% bar of metal, 1% Fido.
This methodology is Indisputably superior to what came before, but I still think it has some problems. In particular I think there’s a danger that the AIs evaluations will end up converging around the same supernormal stimuli that we ourselves, and the market in general have converged on. One of the best arguments for capitalism is that it acts as a distributed intelligence for fulfilling people’s revealed desires, and I’m a fan of capitalism, particularly given the alternatives, but I’m not sure the best choice is to turn the dial on it to 11.
All of which is to say, if you’re worried about the eschatology of AI Risk, the main effect of Russell’s proposal may be avoiding an artificial doom in favor of hastening the natural doom we were already headed for.
By: Yuval Levin
As I mentioned in my last post, if you’re one of those people who feels like something is wrong with the modern world, then the next step is identifying what that something is. This book is Levin’s stab at that and from his perspective the problem is that all of our institutions have been gutted in the service of narcissist self promotion.
To elaborate, in the past attending a venerable institution, say Harvard, was supposed to be about absorbing the lessons, traditions and values of that institution. And with that a certain responsibility to protect and maintain the dignity of the institution. This responsibility continued even after you departed. You were always a Harvard man, and that carried certain expectations. But these days attending Harvard is less about absorbing its history and ideals, and more about making sure Harvard reflects your ideals, and conforms to current social norms, with very little attention paid to institutional values. From this foundation Levin goes on to make arguments about collective action being healthier and more effective than individual action, and how institutions are repositories of virtue, and stuff like that.
I thought it was a pretty good book, and if my review is insufficient there are plenty more out there, but in the end it was another example of discussing symptoms rather than identifying the underlying disease. Which I hope to take a stab at.
What This Book Says About Eschatology
Back in 2013 Scott Alexander of Slate Star Codex put forward a theory for the divide between left and right. He theorized that from an evolutionary perspective humans have two modes. Most of the time they’re in survival mode, but occasionally they get lucky and conditions are such that they can move into a thrive mode. To quote from the post:
It seems broadly plausible that there could be one of these switches for something like “social stability”. If the brain finds itself in a stable environment where everything is abundant, it sort of lowers the mental threat level and concludes that everything will always be okay and its job is to enjoy itself and win signaling games. If it finds itself in an environment of scarcity, it will raise the mental threat level and set its job to “survive at any cost”.
There’s much more to it than that, and if you want to dig deeper read his post, but as this is just a stepping stone, let’s grant that this might be happening and move on. My question, which I explored in a post I wrote back 2016, was if we assume that this is true, and further that the number of people in “thrive mode” is increasing, what consequences follow? There were a lot of them, but one I didn’t explore was institutional decline, but I think it slots in nicely.
If you’re in survival mode then institutions end up being very important. If you protect them they protect you. So much so that historically getting kicked out of an institution was one of the worst punishments that could be inflicted. This most commonly happened with the institution of a city and was called banishment, but being excommunicated from the Catholic Church during the Middle Ages worked very similarly. But now that more and more people are moving to thrive mode the protections an institution can offer mean next to nothing. Instead it’s all about how the institutions can be used as a platform for increasing the visibility of an individual.
As long as this is the case, it seems unlikely that we’re going to ever rebuild institutions in the manner Levin hopes for, because the very nature of the people who make up those institutions has changed. The world is slowly and unalterably becoming a very different place, and I don’t think there’s a simple path back.
By: J. D. Unwin
I covered this in my last post.
II- Capsule Reviews
By: Gabriel Weinberg and Lauren McCann
In certain respects this is just one more self-help book, to sit on the shelf alongside all of the others which have been published over the years. But, having read quite a few of those books, I would say that this one is not only different, but better. To begin with, nearly all self-help books claim to introduce some new way of thinking or some clever system that will radically improve your productivity or at least change your life for the better. Most of these books do not in fact do this, frequently because the idea(s) they introduce aren’t truly new. (For an example see my review of You Are a Badass: How to Stop Doubting Your Greatness and Start Living an Awesome Life which was just a repackaging of The Secret.)
I understand that there are very few truly new things out there, and some of the better books take one principle and really dig into it, for example the value of habits (eg The Power of Habits by Charles Duhigg) or the importance of focusing just on what’s essential (eg Essentialism by Greg McKeown), but this book doesn’t do that either, the approach this book takes is to assemble every single helpful mental model there is and pack it into a single book.
It would be easy for such a book to feel rushed, or choppy, but somehow it was neither. Does this mean that the book never makes a mistake? No, when you’re including everything some of it is going to turn out to not work as well as initially advertised or end up a victim of the replication crisis (for example the growth mindset). That said I didn’t come across anything harmful, and while I was familiar with most of the models they included, I gained that familiarity after reading dozens of books. It probably would have been preferable to just read this one.
In the final analysis all self-help books can be divided into two categories, those where the knowledge gained was of more value than the time required to read them, and those that were a waste of time. And while this book isn’t the best ever, I would definitely put it in the first category.
By: Candice Millard
This is the same author who wrote River of Doubt which I reviewed back in February. This time she tackled the assassination of James A. Garfield. It’s a fascinating story. To begin with Garfield is a lot more awesome than I imagined. I always had the feeling that he was a mediocre president, and perhaps he was, though if so, that was probably just because he wasn’t in office long enough to accomplish anything. But his life before the presidency was pretty incredible. He was born in a log cabin, fatherless before he turned two, horribly poor, but he managed to get a good education by working like a maniac. Eventually he was elected to the House of Representatives (after serving as a general in the Civil War) and then over his strenuous objections, he was nominated to be the Republican Presidential candidate in 1880 on the 36th ballot, after it was clear that no other candidate could secure a majority.
This sounds pretty exciting all on its own, but then on top of all you have the awful story of how Garfield wasn’t killed by the bullet, but by the horrible treatment he received from doctors who didn’t believe in sterilization. And then, if that weren’t exciting enough, there’s the additional story of how Alexander Graham Bell worked 16 hour days for months creating a metal detector in an attempt to find the bullet. The two stories collide when Bell succeeds in creating the detector, but fails to find the bullet because the doctors would only allow him to use it on one half of Garfield’s body and that wasn’t the half the bullet was in. I’ve read better history books, but this was up there, and it has the advantage of being about an event that I knew almost nothing about beforehand.
Similar to War! What Is It Good For?: Conflict and the Progress of Civilization from Primates to Robots by Ian Morris which I reviewed back in November this is another book that makes the case that war has been fundamental to the development of civilizations and nations, and that it’s absence might bring unforeseen harms. Overall I was less impressed with this book. It didn’t seem quite as tight, for example the chapter on “beating swords into malign plowshares” was a particular slog.
That said I’m a fan of contrarians, and this is certainly a very contrarian book. And it’s possible that just by explaining how war is an instrument of rationality, that the book is worth the cover price. As an example of what that means, recall the optimism which preceded the second Iraq War. It’s safe to say that many people including those at the highest level of government, genuinely believed that we would quickly overthrow Saddam, easily establish a functioning and peaceful democracy, and do both with minimal cost in terms of time and money. As we know, the first part kind of happened. On everything else the expectations were tragically mistaken.
The question then becomes how much damage would maintaining those mistake expectations have caused? Is it better that we learned our lesson through the crucible of war, or would it be better if we had never learned that lesson? Or is it possible we could have learned it in some other way? It is indisputable that war is an instrument of rationality, it’s just not clear that this is sufficient to make it necessary.
By: David McCullough
I like McCullough, though I frequently get him confused with Ron Chernow, leading me to believe that I had read more of his books than I actually had, but this is actually just the second of his I’ve read, the first being John Adams of course.
I’m not sure how best to review this book. Though I suppose I can at least keep you from making the same mistake I made. For some reason I expected the book to cover the entire westward expansion, and in reality most of the action is confined to a single town in Ohio, Marietta. But it is impressive how much mileage McCullough is able to get out of this limited geographic focus. He manages to wrap in the Revolutionary War, Washington and his veterans, slavery, the frankly amazing Northwest Ordinance, and the conspiracy by Aaron Burr to form a new nation in the middle of the continent.
I expect you already know what kind of book this is, and if you like that sort of book you’ll like this.
As I continue to read these ancient Greek tragedies, I become more aware of how frequently the playwright manages to point out, that, in addition to everything else that’s going on, isn’t Athens awesome! And when I remember that, comparatively at least, Athens really was awesome, I wonder how much of it was due to art and attitudes like this.
Beyond that I don’t have much to add to the enormous amount of commentary and scholarship which has been devoted to these plays, except to say that from my perspective, if you only had time to read one play, and you wanted that play to be representative of the entire genre, Medea would be my current recommendation.
(She’s best known for murdering her children, but there’s a lot going on in addition to that.)
III- Religious Reviews
Since I have some readers that are uninterested or less interested in my religious stuff I decided to create a separate section for my reviews of religious books. Though really, as long as you’re here you might as well read them.
At the October General Conference of The Church of Jesus Christ of Latter-day Saints (LDS), President Nelson announced that the next conference, in April, would be dedicated to a celebration of the 200th Anniversary of the First Vision, Joseph Smith’s Theophany. My next door neighbor lent this book to me and suggested I read it in anticipation of the event. I ended up finishing it just before Conference, and I’m glad I did. For people steeped in LDS apologetics, There probably won’t be many surprises, but it is interesting how long people have been having the same debates over the same subjects.
Also, despite the fact that standards of proof and citation have tightened up in the intervening decades, I think the book, written 40 years ago, and its research have aged well.
By: Greg Trimble
Once again I’m not sure who recommended this book to me. I should start writing it down. If I enjoyed a book (which I generally do) it doesn’t matter. In the future I can just continue to do what comes naturally, but if I didn’t like a book then I need to exercise caution before accepting another recommendation from the same source. Which is a roundabout way of saying that this was kind of a mediocre book. Perhaps it’s biggest problem was that it wasn’t a book, it was a collection of essays, but not billed as such. The chapters/essays had just enough of a connection that it made me wonder if there was a deeper connection that I was just missing, which tied the essays together into a book. But I don’t think there was.
Also even if you considered the chapters as essays rather than parts of a cohesive whole, some were pretty good, but a lot weren’t. As an example many of the essays had an apologetic theme, but were so superficial that they actually had the opposite effect on me, and I’m a committed member! (It’s possible that’s the point, that his presentation works best on people who aren’t already in the deep end, but I kind of doubt it.)
The title essay (though not labeled as such, just the first chapter) was directed at members within the Church, arguing that as a whole we need to be less dogmatic and more accepting. Trimble is not the first to suggest this, in fact I would argue that it’s almost a cliche. And it’s precisely for that reason that I think it needs to be examined more closely. I’m sure that improvements could be made in this area, but I worry that it obscures the true root problem. Allow me to provide an example of what I mean.
I was out to lunch with an old co-worker the other day (take-out which we ate while walking), and he told me about an incident that had happened in his congregation. He’s in the young men’s and they had a boy who wanted to stop attending church. In an effort to reach out to him they decided to let his father teach a lesson, hoping either the setting or the instructor would make a difference. But as soon as the lesson started the boy got up to leave. And the father and everyone else did exactly what Trimble and others like him would recommend, they asked him nicely (meekly) to stay. He blew them off and left.
Now I don’t know about anyone else who might be reading this blog, but I cannot imagine in a million years doing something like that to my father. Nor can I imagine what he or the other adults would have done. So what’s the difference? Is this a problem with the boy? Is he so hardened that he would have walked out even if it had been 30 years ago? I really doubt that. Was it the fault of the Dad? Based on the story I don’t think there’s any way he could have been nicer or more understanding, which people claim is the answer. Could he have been meaner? Sure, but is there any doubt that he would have been viewed as the bad guy?
So what’s the difference between when I was a boy and now? Who screwed up? Was it the Boy? The father? I would contend that it was society. That in our drive to be accepting that we have abandoned the principle that, if you’re part of a community, there are certain expectations. (This is closely related to what Levin was saying.) That essentially the center of gravity has shifted from the majority of people thinking that such behavior is totally unacceptable to the majority of people thinking that we have to treat our kids with infinite tolerance regardless of what they do. This is a cultural evolution, just as the title of Trimble’s book would suggest, but I would contend that this evolution is just as likely to be the problem as it is to be the solution.
This review is already long, and no one’s saying that this is not a tough subject, but the key question is, in the end, if your goal is to keep this boy in the church, what method works better. The method I and my contemporaries experienced 30 years ago, or the method we’re using now of being super tolerant? Trimble strenuously argues for the latter, and I don’t think the evidence is as clear cut as he thinks. Kids are dumb, and having a community agreement that they are going to do certain things until a certain age, i.e. how it worked in all ages and societies up until about 10 years ago, might not be as awful as people claim. At a bare minimum is it possible the pendulum has swung too far?
Summer is just around the corner, which is unfortunate because it’s my least favorite season (The order is fall, winter, spring, summer.) If you have any desire to help me through this difficult time, or if you’re also a curmudgeon who hates summer as well consider donating.