Chris Hallquist's Blog

Hey everybody! I want you all to know that I have a new blog at topherhallquist.wordpress.com. I know a lot of people have been unhappy about the awful ads and awful commenting system on my Patheos blog, which has in turn left me not-very-motivated to blog more, which has led to the blog just sort of dying. This is my solution to that.

Also, it’s been at least a year since I’ve done much atheism blogging here. I may still post here when I have things to say about religion (which I still do, in spite of being busy with other stuff lately). But non-atheism stuff will likely be at the WordPress blog.

I’m currently just using a default WordPress theme, and may try to prettify the blog in the future, but when I set it up I wanted to make sure I got an actual blog post out right away. So here it is: The case for donating to animal rights orgs:

This year, I am seriously considering making the bulk of my charitable donations to organizations working on animal rights, specifically Animal Charity Evaluators and two of the charities they recommend, Animal Equality and Mercy for Animals. I wanted to talk about my reasons for this, both to get feedback from other people in the effective altruism movement, and to convince some of them to do the same if I’m right.

Ozy has an excellent post up titled “On Exclusion”, that makes a distinction between anti-evangelistic, non-evangelistic, and evangelistic groups. Anti-evangelistic groups  are the ones that active throw up barriers to entry–religions that prefer members be born into them, like Judaism. Evangelistic groups are the obvious ones–political movements, Christianity.

The interesting case is non-evangelistic groups. They just don’t care one way or the other. If someone randomly encounters the group and likes what they see, it will be easy to join, but non-evangelistic groups also don’t mind turning many potential members off:

So let’s imagine Joe comes in to his local anime convention. Joe likes watching anime! He would be very interested in the panels! But Joe thinks that all those people running around in costumes are very weird. It’s not even Halloween. They should get a life. Why are people hugging each other when they don’t even know each other? What’s this ‘pocky’ stuff? Why are fourteen-year-olds squealing about ‘yaoi’? That’s porn! You shouldn’t talk about porn in public! Why are there suddenly a bunch of people dancing? Why is that man selling body pillows? If you can’t get laid you should at least have the dignity not to mention it. Joe feels that anime conventions have a long way to go before they are inclusive of him.

I think that the otaku community would be perfectly justified in saying “Joe, cosplay, hugging, pocky, dancing, public conversation about porn, and body pillows are all part of what it means to be a member of our community. It’s fine if you watch anime and don’t want to be around people who like cosplay and body pillows, but we’re not going to change our community so it caters to Joe-like preferences.”

Ozy applies this to the LessWrong community (which I’m going to be using here to refer to the actual website LessWrong and various descendant communities):

For instance, I think Less Wrong should be nonevangelistic. I think my friends are awesome and I love hanging out with them, but I’m not convinced that we offer any unique insights that can’t be found just as well elsewhere, and I don’t think anyone has enough information to make correct predictions about the Singularity. On the other hand, someone who believes very strongly in raising the sanity waterline might think Less Wrong should be evangelistic: they believe the world would be a much better place if everyone had just read the Sequences.

Similarly, I think Less Wrong should be aimed at people who have what I called the not geek not autism thing. This is purely selfish: I like not-geek-not-autism people, I am one myself, and I would be annoyed at having to find a new community. On the other hand, someone else might have joined Less Wrong because they want to become more epistemically or instrumentally rational. They are probably going to spend a lot of time muttering to themselves about how half of these people aren’t interested in self-improvement at all, and there is all this low-hanging fruit just sitting there, and WHY ARE YOU PEOPLE CUDDLING ALL THE TIME WHAT DOES THAT HAVE TO DO WITH ANYTHING

My observation here is that even more typical LessWrongers–people who buy into the ideology more than Ozy does–rarely seem to want LessWrong to be evangelistic. I think Eliezer Yudkowsky does, maybe a few CFAR folks do, but I don’t think I’ve ever seen anybody else express real enthusiasm for evangelism. Sure, people may talk about “raising the sanity waterline,” but when they do I tend to get a “toying with something that’s probably hopeless” vibe.

The Effective Altruism movement, on the other hand, is very evangelistic. Everyone’s very interested in how we convince more people to support proven interventions to fight global poverty (or in some cases other causes, like ending factory farming). And I think this explains why the two communities, in spite of their overlap, end up looking so very different in practice. A major function of LessWrong is to act as a safe space for obnoxious contrarians. EA can be contrarian but people are much more likely to worry about whether a particular contrarian idea is worth pushing, and if so how to most effectively push it.

I was originally going to call this post, “Effective Altruism and the LessWrong-o-sphere: a proposal,” but I don’t really have much to propose. The separating out of EA and LessWrong is already happening (in the Bay Area–in Oxford they were much less entangled to begin with). But sometimes it may be helpful to be consciously aware of the divide.

This isn’t to say people people can’t participate in both communities, any more than anime fans can’t be part of EA. But there’s a difference between being an anime fan in EA and expecting anime fandom and EA to be deeply entangled. Likewise, it probably doesn’t make sense to expect EA and LessWrong to remain deeply entangled.

Disclaimer stolen from Ben’s blog: “this should probably not be the first thing you read about effective altruism. It’ll give you a pretty biased impression! If that’s you, try something from the reading list I compiled instead.”

So there’s been a bit of arguing on the internet in the past week over whether the Effective Altruism (EA) movement needs to be more inclusive. I was hesitant to comment on this at first, because one of the big things I like about online EA spaces is their relative freedom from the kind of stupid arguments that tend to consume much of the internet. But enough people are already arguing about it that I don’t think adding my 2 cents at this point could hurt. (And, I think sensible people’s aversion to these arguments is sometimes part of the problem.)

Initial disclaimer, the recent discussion has focused on feminism. Ozy’s reaction to that is to say that the EA movement really needs to be more welcoming to religious people. In principle, I agree that we shouldn’t be gratuitously pissing off religious people and it would be wonderful if someone had any great ideas for how to redirect some of the bajillions of dollars that Catholics donate to charity every year to the Against Malaria Foundation. But I don’t have any such great ideas, and I do have things to say on the feminism angle, so I’m going to talk about that.

TLDR; I don’t think it’s a good idea to have a blanket ban on anti-feminists in EA. Apparently some people have (in private) seriously discussed doing that. When people say things like that, they’re thinking of the guy who literally thinks women should be subservient to men, and not thinking of the sex worker who thinks the feminist movement is unfixably messed-up on sex worker’s rights. The latter might be wrong, but that doesn’t mean she should be excluded.

That said, I think there are some cases where people need to be more supportive when someone says, “Hey, here’s this thing that makes me uncomfortable.”

My take on this is partly influenced by the problems by the problems I see in the LessWrong community. LessWrong is distinct from EA, but it helped birth the Bay Area EA community and there’s still a lot of overlap. These are issues I mostly don’t see in more strongly EA-branded spaces like various EA Facebook groups and the EA Forum, but because of the overlap they still end up mattering.

I have a lot of issues with LessWrong, but relevant here is (1) tribalism, in the form of many people having an inflated opinion of the LessWrong in-group and being dismissive of people not in the in-group (2) thinking that worrying about PR is beneath them, and indeed a threat to good rationalist discourse.

So for example, take the original that got people arguing about EA and feminism: Ben Kuhn wrote a post on making the Effective Altruism movement more welcoming, and one of the comments on the post was:

I think this is a great post, but find it ironic that you start with a quote from one of the least welcoming people I’ve encountered in the movement. Being a woman I find it pretty alienating to read posts that talk about ‘the already thin line between feminism and literally being Voldemort’, and the more often I see his vitriolic posts about feminism being endorsed by people in EA the more I question whether I can stay in the movement.

She’s talking about Scott Alexander, who used to be one of the most popular posters on LessWrong before he started his own extremely popular blog, Slate Star Codex. He actually said the Voldemort thing, and Ben had linked to and quoted from another one of Scott’s anti-feminist rants in his post. Mind you, the quote wasn’t about feminism, but Ben apologized for giving the impression he endorsed the post and removed the link (while leaving the quote).

I think this (annonymous) woman’s expression of discomfort was reasonable. Yet people yelled at her for it. And the justifications given for yelling at her are absurd. I’ve seen people–both in Ben’s comments and discussion elsewhere–accusing her of wanting to enforce uniform pro-feminist politics, or dismissing her complaint as being about “insufficiently progressive poltiics”, or claiming talk of being “welcoming” is just ideological code.

But she never said any of that. She said she’s uncomfortable seeing feminism compared to Voldemort. It isn’t hard to imagine an EA movement that isn’t conspicuously feminist or progressive or whatever, but is also free of that kind of vitriol being directed at feminists. This isn’t hard.

Then there’s the people who say, “How could anyone be made to feel uncomfortable by Scott? He’s so charitable to his opponents!” Not when talking about feminists he isn’t–as is obvious when you read his posts on the subject. I think this actually makes the problem worse–Scott likes to talk about how much he believes in being charitable to your opponents, but when he’s charitable to neoreactionaries but not feminists, I suspect for a lot of people this speaks volumes about him.

(Note that I think this is partly LessWrong in-groupyness: there were some neoreactionaries in the LessWrong in-group, but feminists get seen as outsiders.)

Oh, and for the people trying to defend Scott’s feminist credentials, I think it’s worth listening to Scott’s own words on the subject:

I notice that, no matter how many long rants against feminism I write, everyone continues to assume I am a feminist. It’s like, “He doesn’t make too many spelling errors, his writing isn’t peppered with racial slurs – he’s got to be a feminist. He probably just forgot the word ‘not’ in each of his last 228 sentences.”

His writing is not terribly ambiguous on this point. Also Ozy would like people to stop using the fact that zie used to date Scott as evidence of Scott’s feminist credentials, and I suggest people listen to zir.

I say all this, but I also repeat what I said at the beginning of this post that “ban anti-feminists” is a bad idea. If that confuses you, it shouldn’t. When a woman says, “hey, this thing makes me uncomfortable,” there are a lot of ways you can deal with the issue that are in between “ban all anti-feminists” and “yell at her for trying to force a single ideology on your group.”

In this case, you don’t need to ban Scott. If people were more careful about endorsing, or seeming to endorse, his anti-feminist vitriol, I suspect Ben’s commenter would feel a lot more comfortable. Or, heck, if more of Scott’s fans would recognize that “being criticized” kind of comes with the territory of “being a popular blogger who writes about controversial subjects,” and be less knee-jerk defensive of him.

Again, as I indicated at the beginning of this post, there are anti-feminists I sympathize with. That said, I think announcing “I hate feminism!” is still a generally bad idea, because a lot of people are going to take it the wrong way. The fact that I can sympathize with why someone would say p and don’t think they should be kicked out of the movement for saying p doesn’t mean I think saying p is a good idea, or that we shouldn’t be discouraging people from saying things like it, or that it’s a good idea to signal-boost blog posts saying p.

A final point: sensible people rightly dislike these kinds of nasty internet arguments, and tend to wish they’d just go away. And one way to try to make them to away is to offer up a formulaic, “well, both sides did bad things,” or position yourself as the middle ground between the extremes. Which is sometimes the right stance to take, sometimes both sides really are being about equally stupid.

But I think it can be tempting to take that stance even when oh, say, someone voices a reasonable complaint and a bunch of people yell at them for it. Even when you think the initial complaint is reasonable, it can be tempting to think, “well, maybe if I throw the other side a bone, they’ll stop yelling.” But this is a bad dynamic: it means that when something is making someone uncomfortable, they’ll find that even people who think their complaint is reasonable won’t stick up for them, which isn’t going to help them feel comfortable in the movement.

So please: when people are talking about things that make them uncomfortable within the movement, don’t play the “well, both sides…” game, if that’s not what you actually believe.

In my previous post, I compiled citations from historical thinkers on how awful life was back in their day. But such citations aren’t conclusive–maybe they were exaggerating for effect. Or maybe people with severe depression are grossly overrepresented in the ranks of intellectuals. Or who knows. So in this post, I want to examine the question more directly.

Specifically, I want to ask how bad life really is at subsistence. That is, how subjectively bad is it to live a life where you have just barely enough resources to survive?

I think the naive answer is “pretty bad.” In principle you could imagine a centrally planned society where there’s no extra food but people never have more children than society will be able to feed, in practice subsistence means population is kept in check by early death. “Misery” in Malthus’ words, though that begs the question, as you could imagine a society where most people die young but are happy until their death.

Beyond that observation, I think the question is very hard to think about for people living in 21st century America. It’s too far removed from our experience. That was part of my motivation for my previous post–just to convey a sense of how alien reading about the past can be at times.

I have another reason, though, for suspecting that life at subsistence is mostly pretty miserable. Evolution has designed animals to be motivated to survive and reproduce by a combination of pain and pleasure, but it seems to have gone heavy on the pain. Any time an animal faces a serious threat to its continued existence, the main things moving it to act will be pain and fear, not the pleasantness of the actions that bring survival nor hope of future pleasure. Eating is pleasant, but hunger is what keeps us from starving.

And life at subsistence means facing continual threats to your existence. The possibility that you could go from having barely enough food to not enough at any time. Indeed, it likely means never having quite enough food–it used to be almost everyone grew up with their growth stunted by malnutrition, and that chronic low-level malnutrition meant they were more likely to get sick, and more likely to die if they did.

This is not a very conclusive argument. I have no first-hand experience with chronic malnutrition, and don’t know anybody who has. Maybe people just get used to it? And others I respect have thought about this issue and come to different conclusions. Brian Tomasik, for example, is known for arguing that wild animals mostly don’t have lives worth living, but focuses on pain of death:

I personally believe that most animals (except maybe those that live a long time, like >3 years) probably have lives not worth living, because I would trade away several years of life to avoid the pain of the average death, and this is assuming that even their lives are net positive (which is dubious in view of cold, hunger, disease, fear of predators, and all the rest).

However, this belief of mine is somewhat controversial. I think the claim of net expected suffering in nature needs only a weaker assertion: namely, that almost all of the expected happiness and suffering in nature come from small animals (e.g., minnows and insects). The adults of these species live at most a few years, often just a few months or weeks, so it’s even harder in these cases for the happiness of life to outweigh the pain of death. Moreover, almost all the babies of these species die (possibly painfully) after just a few days or weeks of being born, because most of these species are “r-selected” — see Type III in this chart.

He seems to think there are evolutionary reasons why the pain of death is likely to be especially bad:

One can advance some argument that evolution should avoid making animal lives excessively horrifying for extended periods prior to death because doing so might, at least in more complex species, induce PTSD, depression, or other debilitating side-effects. Of course, we see empirically that evolution does induce such disorders when traumatic incidents happen, like exposure to a predator. But there’s probably some kind of reasonable bound on how bad these can be most of the time if animals are to remain functional. Death itself is a different matter because, once it reaches the point of inevitability, evolutionary pressures don’t constrain the emotional experience. Death can be as good as painless (for a few lucky animals) or as bad as torture (for many others). Evolution has no reason to prevent death from feeling unbearably awful.

In short, “it seems unlikely that species would gain an adaptive advantage by feeling constant hardship, since stress does entail a metabolic cost.”

This seems backwards to me. I suspect the negative side-effects of pain and stress are actually adaptive in the contexts they evolved to handle. For example, I’ve heard speculation that what we call PTSD may actually be adaptive in situations with a constant risk of repeated trauma. The metabolic costs of stress likely come from redirecting resources away from other metabolic functions towards avoiding danger, which may be adaptive if an animal faces constant risk of predation.

While Brian makes a good point that some forms of death, like being burned alive, may be more painful than anything an animal is likely to experience during its life, I expect death is on average not any more painful than many other traumatic life events. Specifically, for evolutionary reasons I expect a typical death is about as painful as a typical life-threatening event that ultimately ends up being non-fatal.

Experience seems to back this up, for example, childbirth used to be a major killer of women, and its often reckoned to be one of the most painful things a human can possibly experience. On the other hand, humans may be unusual in this regard. For heavily “r-selected” species, it may be that they rarely almost die, and that instead, life-threatening problems are almost always actually fatal. I’m not quite sure what to think here, and the fact that Brian’s reasoning is so different than mine gives me pause.

Robin Hanson has argued for a much more optimistic view, mainly in the context of his forecast of a future world of trillions of sentient digital mind emulations, most of which will just barely be able to pay to rent and power their bodies. With cheap virtual entertainment and the ability to engineer-out pain, perhaps the emulations will be happy. Robin is less convincing, though, on actual people who’ve lived a subsistence existence.

In fact, as far as I can tell he just assumes they experience far more pleasure than pain, and then goes on to psychologize rich people who would doubt this. And those psychologizing explanations aren’t terribly convincing either, for example, he says, “Rich folks would personal be horrified to have to live so poor. They are very used to their wealth, and for them poverty would be a huge horrifying shameful fall in social status.”

This seems to miss what we are actually talking about here–not poverty as it exists in 21st century America, but what we 21st century Americans call “extreme poverty” or “absolute poverty” (really just the state of almost everyone for most of human history). If Robin lost his job, couldn’t get another one, and his family had to go on food stamps, no doubt he’d find that humiliating, but his family wouldn’t starve. By contrast, the problem with living at subsistence isn’t the loss of status; it’s being hungry all the time because you don’t have enough food.

As I’ve previously discussed, Robin defends raising animals for food, and one of his arguments is that farm animals don’t typically commit suicide. Some people, like Brian, may respond that modern factory farming is so awful that this just proves failure to commit suicide is a bad indicator of whether an animal’s life is worth living.

But set that aside. The argument seems hard to swallow in any case, unless you’re a pure preference utilitarian and very committed to the idea of revealed preferences. There seems to be no contradiction in imagining a mad scientist engineering a species to feel constant pain but never attempt suicide. It doesn’t seem too far-fetched to imagine evolution doing a lesser form of this, given that evolution should select strongly against suicide yet most animals’ experiences seem heavily biased towards pain over pleasure.

But I’m not very confident about any of this. I repeat how alien subsistence existence is to modern experience. The kind of poverty that was the norm through most of human history is declining rapidly even in sub-saharan Africa. So I’m not even sure how you would answer this question definitively. But I wish it were discussed more.

The possibility that some–perhaps many–lives may not be worth living is depressing to think about. But it’s one that should worry effective altruists and people interested in utilitarian ethics. Among EAs interested in animal welfare, it’s widely thought that the lives of animals in factory farms are so bad as to not be worth living. Some have expressed similar thoughts about wild animal suffering.

Similar concerns apply when we look ahead to the far future: Robin Hanson predicts a future of trillions of digital minds eking out a subsistence existence, earning just enough money to pay for rent and electricity for their bodies. He says they’ll be happy anyway, but many of his readers are skeptical. Even ignoring such an artificial population explosion, there’s a question of how many flesh-and-blood humans you could have before additional ones would be a net negative on the world.

21st-century humans live in an age of unprecedented material prosperity, so when I think about this question the first place I’m inclined to go is to historical sources on living conditions in eras past. I’ve stumbled across a lot of striking material in my readings of philosophy. Basically, go back more than a mere hundred years (or even slightly less), and the world you’re looking at is deeply alien.

Right now, I’m reading Montaigne’s essays, and I’m in the middle of a section on how ineffective the medicine of his day was (of course, for Montaigne, the topic is just “medicine”). As a consequence, it was normal for people to suffer for a long time with extremely painful diseases and no effective treatments.

Montaigne himself suffered for years with kidney stones, and writes vividly about how excruciating the experience was, and his futile search for any kind of effective treatment. (The main treatment he tries is taking in the waters of a hot spring. He’s not confident it helps, but recommends it because he’s sure it won’t hurt, unlike many of the other things doctors of his day did to their patients.)

Hume’s Dialogues Concerning Natural Religion contain an especially vivid description of human suffering. After the arguments for the existence of God have been dispensed with, two of the participants (Demea and Philo) begin talking about how (in Philo’s words) “the only method of bringing every one to a due sense of religion, is by just representations of the misery and wickedness of men.” What’s even more striking is that the universal misery of man is treated as universally accepted:

The people, indeed, replied Demea, are sufficiently convinced of this great and melancholy truth. The miseries of life; the unhappiness of man; the general corruptions of our nature; the unsatisfactory enjoyment of pleasures, riches, honours; these phrases have become almost proverbial in all languages. And who can doubt of what all men declare from their own immediate feeling and experience?

In this point, said Philo, the learned are perfectly agreed with the vulgar; and in all letters, sacred and profane, the topic of human misery has been insisted on with the most pathetic eloquence that sorrow and melancholy could inspire. The poets, who speak from sentiment, without a system, and whose testimony has therefore the more authority, abound in images of this nature. From Homer down to Dr. Young, the whole inspired tribe have ever been sensible, that no other representation of things would suit the feeling and observation of each individual.

As to authorities, replied Demea, you need not seek them. Look round this library of Cleanthes. I shall venture to affirm, that, except authors of particular sciences, such as chemistry or botany, who have no occasion to treat of human life, there is scarce one of those innumerable writers, from whom the sense of human misery has not, in some passage or other, extorted a complaint and confession of it. At least, the chance is entirely on that side; and no one author has ever, so far as I can recollect, been so extravagant as to deny it.

There you must excuse me, said Philo: Leibnitz has denied it; and is perhaps the first who ventured upon so bold and paradoxical an opinion; at least, the first who made it essential to his philosophical system.

Cleanthes, the third participant in the dialog, briefly pipes up against the claim of universal human misery, but this is dismissed as denying what is obvious to everyone And this being Hume, this is all a setup to launch into the problem of evil. This leads to an observation I want to highlight:

Formerly it was a most popular theological topic to maintain, that human life was vanity and misery, and to exaggerate all the ills and pains which are incident to men. But of late years, divines, we find, begin to retract this position; and maintain, though still with some hesitation, that there are more goods than evils, more pleasures than pains, even in this life. When religion stood entirely upon temper and education, it was thought proper to encourage melancholy; as indeed mankind never have recourse to superior powers so readily as in that disposition. But as men have now learned to form principles, and to draw consequences, it is necessary to change the batteries, and to make use of such arguments as will endure at least some scrutiny and examination.

I have a theory about this change that Hume observers, an alternative to his theory that the shift in theological style was not simply about people becoming more logical. Rather, in previous times, human misery was truly nigh-universal, so it was easy for people to convince themselves that things really did have to be that way.

But in Hume’s day, the mid-18th century, the world was starting to see the first glimmerings of true material prosperity. At that time, many people still looked back to the days ancient Rome as a golden age which would never be equaled. By Hume’s time, he was able to convincingly argue against that view, but as far as I know he was one of the first to do so.

Yet even if people didn’t consciously realize how much better things had gotten, they could still look around them and see that there was an alternative to universal misery, one that did not involve waiting for an afterlife or Christ’s Millennial Kingdom. But misery was still the norm.

One last quote from Hume, that may be relevant to future posts: “Ask yourself, ask any of your acquaintance, whether they would live over again the last ten or twenty years of their lives. No! but the next twenty, they say, will be better.”

No discussion of this sort would be complete without mentioning Malthus. Like Hume, Malthus takes it as universally accepted that misery was universal, at least among the poor, and seeks to explain this fact. There are also some good descriptions of what this meant in practice, for example, the fact that it was the norm for poor children in the country side to grow up stunted due to malnutrition.

As an aside, another point about Malthus that often goes overlooked is that he did not quite say that there was no hope to improve the lot of the poor. He said that the population must inevitably be kept in check by misery or vice, the latter category for Malthus including birth control. When it comes to that claim, anything you may have heard about Malthus being refuted by modern prosperity is wrong.

Malthus estimates, based on data from Britain’s North American colonies, where farmland was abundant, that unchecked the population can double every 25 years. The Earth’s population was about a billion people in Malthus’ day, if it had doubled every 25 years since then, we would have over a quarter trillion people today. The Green Revolution was never going to feed that many people.

The problem with Malthus isn’t his facts, but his ethics. He never betrays the slightest preference for birth control over population control via disease and starvation. I suppose for him the question of birth control vs. early death is a bit like the question “would you rather starve to death or die of Ebola?” Rather morbid, and beside the point: both are very bad, right? Reading Malthus is like listening to an alien dreamed up by Eliezer Yudkowsky to argue a point about metaethics.

Jumping forward over half a century, we have John Stuart Mill’s Utiltarianism. In that book, one o the criticisms of utilitarian philosophy that Mill addresses is that “happiness, in any form, cannot be the rational purpose of human life and action; because, in the first place, it is unattainable.”

In his response, Mill does not say (surprisingly, or not, if you’ve read this far) something like, “Nonsense! Plenty of people are happy!” Instead, he says, “unquestionably it is possible to do without happiness; it is done involuntarily by nineteen-twentieths of mankind.” But Mill, the social reformer, thought this could be changed: “The present wretched education, and wretched social arrangements, are the only real hindrance to its being attainable by almost all.”

Things seem to have definitely turned the corner by the early 20th century. In 1930, Bertrand Russell wrote:

The injustice, the cruelty, and the misery that exist in the modern world are an inheritance from the past, and their ultimate source is economic, since life-and-death competition for the means of subsistence was in former days inevitable. It is not inevitable in our age. With our present industrial technique we can, if we choose, provide a tolerable subsistence for everybody. We could also secure that the world’s population should be stationary if we were not prevented by the political influence of churches which prefer war, pestilence, and famine to contraception. The knowledge exists by which universal happiness can be secured.

Yet to say it was possible then to provide a “tolerable subsistence” for everyone was to say a great many people didn’t have it. As I’ve previously noted, Russell took for granted that in his day, losing one’s job meant starving for most people.

On Tumblr, Kelsey (a.k.a. theunitofcaring) had a really good post on the politics of coming out (I’ve bolded a paragraph that I think is especially important):

Hmm. Harvey Milk’s time is sufficiently different from ours that I’m not sure I can disagree with him, but I think I do disagree about today.

My sister and I (both cis) teamed up on our parents to explain trans stuff. My parents started at ‘if they’ve gotten the surgeries so you can’t tell them from people who were born that gender then I suppose I’d call them it, but they should stop forcing those things on children’ and after a few weeks of patient reasoned disagreement they’d -

- well, last year my mom was really angry at my (nonbinary) roommate and in private she went off on a long and angry rant about them. And used the correct pronouns every single time. That’s sort of the litmus test I used for whether people really get it; do they stop being respectful when it’s toward someone they rightly are angry at? She didn’t.

But I haven’t done that with my parents about gay stuff. I couldn’t. Those weeks would, instead of being empowering, be emotionally exhausting and hellish and awful. And I don’t even think it would have been morally good, even if I had the emotional strength to deal with it. Having my (both straight, afaik) brother and sister team up on my parents to deal with that one honestly seems like it would have been better. My parents wouldn’t have been afraid of asking them honest hard questions, which they were afraid of asking me and thus tiptoed around. And so those questions wouldn’t have burst out at three in the morning while my mom was yelling at me over something else.

I don’t think coming out to people is more conducive to earning their tolerance than, as an ally, repeatedly and carefully explaining things. I think the difference in statistics is because allies really really rarely do that work. Like, let’s say that both LGB people and straight allies could both convince 50% of the people close to them, the LGB people via a few late-night screaming matches and the straight people via a few weeks of talking about it incessantly but cheerfully. But pretty much all LGB people do that at some point, because no choice, and less than 1% of straight allies are going to bother, and so you get the statistic that most people who are tolerant of gay people know one.

This doesn’t make it any different for an LGBT+ person to come out than for a straight person to just make a nuisance of themselves next holiday. But it seems like framing ‘coming out’ as a political act, and a morally important political act, is making a claim much stronger than ‘it’s actually pretty good activism to annoy your ignorant relatives at holidays.’

Telling people who don’t know much about LGBT+ issues that those issues are very important to you, firmly and unapologetically but respectfully, answering their questions and emphasizing the significance to you of their support, is morally good. The only context in which people seem to realize it’s appropriate to do this is while coming out, but the coming out isn’t what makes it morally good.

Also if coming out won’t even possibly have that consequence, and it’ll just make them hurt you, then it is absolutely not morally good to do it.

This is sort of reasoning from one case, but it fits with something I’ve suspected for other reasons, namely: activism is most effective when people have a personal stake in the cause, for the very non-mysterious reason that most people won’t care enough to put in the work of activism unless they have a personal stake in the cause.

This seems to be fit with my observations of the US anti-war movement. It also explains why democracies often seem dominated by self-interested interest groups, even though self-interest doesn’t affect voting much. And unfortunately, it reinforces the fairly pessimistic view I have of animal-rights activism.

Last month, I wrote a post where I talked a bit about Bryan Caplan’s book The Myth of the Rational Voter. Specifically, I talked about the idea that because any one voter is unlikely to swing the outcome of an election, voting provides a way for people to feel altruistic on the cheap. Therefore, in some situations, it may be easier to get for people to vote for a policy than to make analogous changes to their individual behavior.

But how altruistic are voters, really? It’s easy to think of examples of democracies at least acting as if their policies are driven by the self interest of their voters. One example I’ve just been discussing: slavery. In the United States, it was the northern states, the ones that had few slaves and no industries to which slavery was especially well-suited to, that abolished slavery through ordinary democratic means. In the south, where slavery was enormously profitable, abolition needed to be imposed from the outside through a bloody civil war.

Examples are easy to multiply. Extending the slavery example a bit, after the civil war, blacks were disenfranchised, and treatment of blacks by the state governments was correspondingly awful. Today, when Bryan himself goes looking for examples of things future generations might view as evil in plain sight, all of his examples involve groups that can’t vote, such as foreigners. I agree with him there–the US government may occasionally mistreat its own citizens, but what it does to its own citizens is nothing compared to how it treats foreigners.

Bryan might blame these examples on racism and nationalism, and indeed, I agree that these are great problems. But we also seem to see similar patterns when we look at the behavior of politicians representing local constituencies. In the Bay Area, where I live, our housing crisis can largely be blamed on the fact that city governments are generally controlled by local homeowners, who seek to protect their property values without regard to what effect it has on the rest of the area–including people who take long commutes to work in their city, but would prefer to live there.

Yet Bryan has a trump card here: studies of actual voter behavior seem to show little if any correlation with self interest. Wealth has little effect on party identification, age has no effect on support for Social Security and Medicare, and so on. On one level, this data seems very convincing. But how to reconcile it with the actual behavior of democracies?

I’m genuinely unsure of the answer here, but I can think of a number of guesses. I don’t think all of the following theories are equally plausible, but I’m going to throw out a range of possibilities because I don’t want to dismiss any prematurely.

The miracle of aggregation: In The Myth of the Rational Voter, Bryan discusses “the miracle of aggregation,” the idea that a small number of informed voter and a much larger number of ignorant voters can produce informed policy, if the ignorant voters vote at random and are sufficiently numerous to cancel each other out. A central point of the book is that this is wrong–ignorant voters vote not at random but in accordance with predictable biases.

Still, you could imagine a version of this holding for voter selfishness. If only a small number of voters are selfish, but the altruistic majority votes more or less at random, the selfish voters could have a disproportionate impact. This seems like a promising hypothesis for explaining cases where the correlation between self-interest and voting behavior is small but present. Obviously, though, it doesn’t work for cases where there’s no correlation whatsoever.

Self interest doesn’t affect people’s views, but does effect whether they vote on them: For example, Bryan claims there’s little evidence that self-interest affected people’s views of the Vietnam war, yet people’s concern for drafted friends and relatives seems to have played an important role in ending the draft.

Perhaps, then, having a relative drafted didn’t change people’s opinions much–but did mean that if they opposed the war, it made their opposition more likely to be a determining factor in how they voted. This explanation may be promising for explaining lack of correlation between views on individual issues and self-interest, but doesn’t help in the face of data showing self-interest and actual voting are uncorrelated.

Self interest doesn’t affect votes, but does affect other forms of democratic participation: Related to the previous hypothesis, self interest might have less of an effect on voting than on whether people will go to protests, write letters, put a campaign sign on their lawn, or even simply try to persuade their friends to vote the same way as themselves. This hypothesis seems to be at least partly true, though I’m not sure how much it actually explains in practice.

Blame special interests: What looks like democracies caring mainly about the interests of their voters may actually be democracies mainly caring about certain subsets of their voters, i.e. special interests. This could be seen as a variation on the previous hypothesis. It would explain why the governments of southern states in the lead-up to the Civil War acted as they did, even though many of their voters owned no slaves: wealthy slave owners were a special interest group with influence out of proportion to their number of votes.

Enfranchisement and disenfranchisement affects who people think matters morally: In other words, when people vote, they try to vote the interest of their fellow voters. This hypothesis would accept that Bryan is right that a lot of bad US policy towards voters can be explained by nationalism–while also making nationalism an effect of the fact that foreigners have no say in our government. It would also suggest that disenfranchising uneducated voters, an idea Bryan toys with, could push the remaining voters towards policies more hostile to those with less education.

Voter irrationality is the big problem–but voters manage to avoid the worst effects of this: One could argue that the US government’s appalling treatment of foreigners isn’t really selfish, since those policies hurt natives too. Immigration restrictions prevent natives from benefitting from foreign labor, while military adventures waste trillions of dollars in ways that often backfire.

Yet these costs are small compared to the harm these policies inflict on foreigners. Better to lose the opportunity to hire a Haitian, than be forbidden from taking a job outside of Haiti. Better to pay higher taxes to finance military adventures, than become “collateral damage” in a drone strike.

This suggests a two-part model: the root cause of these problems is irrationality, but if the cost to natives was as high as the cost to foreigners, voters would notice and reject them. But so long as the costs to natives are non-obvious, few voters know or care what happens to foreigners.

None of the above–instead, blame the belief that voters are selfish: In his book, Bryan portrays the self-interested voter hypothesis as false but nevertheless widely believed by people who aren’t political scientists. But if politicians believe it, that would be enough for them–and the governments they run–to act much as they would as if it were actually true. Even if politicians have read the political science, if they find themselves surrounded journalists and political allies who haven’t, they might fear being attacked as political naifs if they take stances that demand too much altruism from their constituents.

I suspect Bryan will like this hypothesis a lot. It’s also an optimistic one: perhaps we could fix Bay Area housing policy and US immigration policy simply by spreading the truth about the falsity of the self-interested voter hypothesis. But that may be too optimistic. In the bubble that Bryan and I live in, it may be tempting to believe that Americans care a lot about the lives of foreigners, but every time I poke my nose into a media outlet catering to mainstream conservatives (not libertarians), I find that nationalism and xenophobia are rampant.

In response to my previous post on animal rights and slavery, Wayne of Direct Action Everywhere suggested that the argument made in that post was at odds with the findings of economist Robert Fogel as to how slavery was abolished. So I picked up Fogel’s book Without Consent or Contract: The Rise and Fall of American Slavery. Having read it, I’m honestly confused as to why Wayne would think it supports his views of social movements. 

Let me start by clearing up a possible misunderstanding of my previous post. My point was not that slavery disappeared because it was everywhere economically inefficient. My point was that Northerners were willing to vote to abolish slavery, because they lost relatively little economically from doing so. I agree that slavery was very profitable in the South–that’s why emancipation had to be imposed on the South from outside.

Fogel’s work seems broadly in line with this view. Fogel describes the “gang system” that was used to organize slave labor on plantations. He reports that this system was very efficient–but only for certain crops. “Slavery was marginal in the northern U.S. colonies because no major crops lent themselves to the gang system”–p. 35. (Specifically, indigo, hemp, sugar, tobacco, cotton, and rice were the only crops that benefitted from the gang system–pp. 79-80.)

Regarding the British abolition of the slave trade, Fogel writes (p. 215):

The British abolition of the slave trade was hailed by supporters of the measure and by subsequent commentators as “the most humane and merciful Act which was ever passed by any Legislature in the world” and as one of “three or four perfectly virtuous pages comprised in the history of nations.” Even if one accepts the purity of the motives of those who engineered the campaign, such statements exaggerate the degree of British self-denial. The economic cost of the act to the British, or even to the West Indian planters, was small. The only clear economic losers were the merchants directly engaged in the African slave trade (located mainly in Bristol and Liverpool) who were forced to switch their capital to other enterprises.

Later, he writes, “What made the American struggle against slavery so different from the British one, and politically so much more difficult, was not so much that American soil was trodden by slaves but that it was also trodden by their owners” (p. 238).

More striking than this, though, is Fogel’s description of how the antislavery movement had to morally compromise itself in order to be successful. This is in stark contrast to claims made by DxE to the effect that credit for abolishing slavery should go entirely to the radicals. Here’s Fogel (p. 60):

In the United States the antislavery struggle was also conducted primarily on moral and humanitarian grounds from pre-Revolutionary times until the mid-1850s. Indeed, such abolitionist leaders as William Lloyd Garrison adamantly resisted all proposals to shift the basis of their assault from religious to economic grounds, proposals that emanated from the more worldly leaders of the movement. To turn their appeal from the conscience “to the pocketbook,” from “the duty of Christian reformation” to “the love of political preferment,” he warned, would inevitably corrupt and subvert the moral principles on which their movement was based. Nevertheless, the principal basis of the antislavery appeal did suddenly shift from “Christian duty” to “the pocketbook.” The shift took place between 1854 and 1856 and the political success was immediate and spectacular. The new approach transformed the antislavery movement from a minor political factor into a powerful political force that could control the national agenda.

Fogel mentions several such arguments, but says that, “ultimately perhaps the most politically effective of all the abolitionist arguments, was the contention that the growth of the slave population and its spread into non-slave states and territories constituted an imminent threat to the living standards of free workers and farmers” (p. 121).

The economic argument was important to the success of the Republican party, because it allowed them to tell a story about how slavery would be peacefully, gradually abolished: “the economic arguments enabled Republicans to present a peaceful, long-run solution to the problem of slavery: Since it was economic forces that impelled the South into its expansionist policies, if slaveholders were bottled up in their own region, slavery would gradually die” (p. 354).

It’s important to emphasize that Fogel did not think these economic arguments were correct. His work on the economics of slavery leads him to the conclusion that slavery was not as economically backwards as the antislavery movement of the 1850s claimed. The arguments were nevertheless effective, in spite of the fact that even many people at the time could see they were wrong–which means that the antislavery movement arguably won through “deceptive tactics” (a phrase Fogel uses).

It’s worth contemplating the implications of this for the animal rights movement. I believe that with a little work, a vegan diet can be as healthy as an omnivorous diet. But I’ve long been skeptical of arguments that veganism is significantly healthier. That sounds too much like convenient rationalization. On the other hand, I also suspect health-based arguments for veganism have advantages over moral arguments (consider the decline of smoking, driven more or less entirely by health concerns).

Anyway, back to Fogel. Not only did the antislavery coalition that made Lincoln president rely on dubious economic arguments, it even relied on conspiracy theories. In the North, it was widely believed that there was a “Slave Power” conspiracy to force slavery on the North, including free whites (p. 338ff). Lincoln himself was an advocate of this conspiracy theory (p. 343), and attempts to expand slavery into territories where it had been previously been banned through the 1854 Kansas-Nebraska act were widely seen as evidence of this conspiracy.

Fogel also goes into considerable detail about the fact that the successful Republican coalition of 1860 relied a lot of outright racists for support. This and other points are covered in the quotes below–I’ve bolded one paragraph that strikes me as especially important:

Prior to the 1860 election and well into the war years neither Lincoln nor the Republican party was committed to freeing the slaves or to granting political and social equality to the free blacks of the North. Quite the contrary, when the Democrats sought to portray them as the advocates of black rights, many Republican leaders, including Lincoln, emphasized their commitment to maintaining “the superior position assigned to the white race.” (p. 386-387)

It is easy to invent an antislavery policy that would be far more congenial to the predominant moral standards of our own age than the policies actually put forward by any of the major historical players. To censure the abolitionists for not fully anticipating our values, however, would be foolish. A policy congenial to us could not have built the political coalition needed to destroy slavery. Even policies that now seem conservative were too radical to succeed in 1860. Here then is a dilemma posed not only by the antislavery struggle but by most other moral movements: It is difficult in a democracy, if not impossible, to transform a moral movement into a winning political coalition without deeply compromising its moral integrity; it is also difficult to pursue one moral goal without compromising or even sacrificing other important moral goals. (pp. 388-389)

I believe that we should no longer interpret the moral problem of slavery within the framework of the indictment fashioned by the winners of the antislavery struggle. The contention that we need a new indictment does not mean that all continuity with past views need be broken. Quite the contrary, a new indictment could restore continuity with the earlier and more frankly ethical aspects of the attack, which leaders of the popular movement abandoned, or at least attenuated, because it was politically expedient to do so. Such an indictment, then, would be more closely bound to the radical position on slavery than to the later and more politically successful one. (p. 393)

Many Republican candidates, including some of the most radical leaders of the party, detested blacks. They were quite sincere when they assured voters that as “true Republicans” they “cared nothing for the nigger” and that the aim of the Republican party was to “make ‘white labor respectable and honorable’ by keeping Negroes, free and slave, out of the West. (p. 401)

As late as March 1862 the majority of Republicans in the Senate voted to subsidize the colonization of freedmen. Five months later Lincoln called a group of black leaders to the White House to enlist their support for colonization. Arguing that racial differences created insuperable barriers to citizenship for blacks, and noting the terrible injustices that blacks had suffered at white hands and the white blood being shed in the Civil War on their behalf, Lincoln called on the black leaders “to make clear to their people that removal was the only solution.” Those who refused to leave the country, he continued, were taking “an extremely selfish view of the case.” (p. 402)

To some radical abolitionists, then, the Republican call to resistance of the Slave Power in the territories smacked of sheer opportunism. It was a cry to oppose slavery “where it is not” instead of “where it is.” They suspected that those who called only for resistance in the territories were more interested in preventing free blacks from competing with whites for western lands than in striking a death blow at slavery. The nonextension rhetoric seemed far more likely to feed the popular prejudices against blacks than to promote their civil rights… Yet without pandering to northern negrophobia, the antislavery coalition would surely have lost the 1860 election. As it was, Lincoln just barely squeezed through. The antislavery forces were caught on the horns of a genuine dilemma. Although the negrophobic strategy undoubtedly contributed to patterns of discrimination that still plague the nation, the alternative—an indefinite continuation of slavery—was far more evil. The radical program was unattainable. Under the conditions of 1860, rigid adherence to the purest abolitionist principles, unwillingness to join hands with opportunists, and squeamishness over deceptive tactics would have led the antislavery forces to certain defeat. (pp. 408-409)

When I talk to people involved in animal rights, I frequently hear people make analogies to the abolition of slavery, made with an optimistic spin: “if we managed to abolish slavery, we can abolish animal agriculture,” “if these tactics worked for abolishing slavery, they should work for abolishing animal agriculture,” and so on.

I think there’s an obvious problem with these analogies, so obvious that it didn’t occur to me to point it out in any of my previous posts in this vicinity. The problem is this: slavery was primarily abolished by people who never owned slaves, and would’ve been unlikely to become slaveowners in any case. In other words, it was abolished by people for whom abolitionism was a relatively easy position to take.

I’ve always had the vague impression that in the United States, the slave population was never as high in northern states as in the southern states. But I decided to look up numbers, just to make sure this wasn’t just some myth I’d picked up. Looking at Wikipedia’s pages on slave and free states and the 1790 census, slavery had already been abolished in Main, Massachusetts, and Vermont in 1790. Slaves as a percentage of population in other states were as follows:

New Hampshire: 0.1%Rhode Island: 1.4%Connecticut: 1.2%New York: 6.3%New Jersey: 6.2%Pennsylvania: 0.9%Delaware: 15.0%Maryland: 32.2%Virginia: 39.1%Kentucky: 16.9%North Carolina: 25.5%South Carolina: 43.0%Georgia: 35.5%

Wikipedia indicates that the states with less than 2% slaves were already in the process of gradually abolishing slavery. But none of them had very large non-white free populations (Rhode Island was the highest, around 5%, and even that number was a bit of an outlier). This suggests they never had very large slave populations. New York and New Jersey, similarly, began abolishing slavery in 1799 and 1804, respectively.

On the other hand, in every state that had a 15% or more slave population in 1790, slavery remained 100% legal into the Civil War. This includes Delaware, Maryland, and Kentucky, which never joined the Confederacy.

I don’t doubt that moral arguments played a role in the abolition of slavery. But those moral arguments were primarily successful among people who had little to lose, economically, by accepting them.

This is why I suspect that one of the most likely routes to abolishing animal agriculture is if lab-grown meat were to become as cheap, or better yet cheaper, than farmed meat. If lab-grown meat became cheaper than farmed meat, the latter might continue for awhile as a luxury item. But it would then be easy to attack as a taste of morally callous rich people–much as veal or fox hunting (in Britain) are today.

So, Eliezer Yudkowsky’s long-running Harry Potter fanfic, Harry Potter and the Methods of Rationality, is now complete. I’ve been holding off until now to post some thoughts on it I’ve had since the climax was posted ~2 weeks ago.

Readers of TVTropes will be familiar with the concept of deconstruction–not in the pretentious post-modern sense, but in the sense of literary works which try to poke holes in the conventions of other literary works, often (as TVTropes puts it) by asking, “‘How would this trope play out with Real Life consequences applied to it?’ or ‘What would cause this trope to appear in Real Life?’”

Alan Moore’s Watchmen is the go-to example of deconstruction in superhero comics. George R. R. Martin’s A Song of Ice and Fire novels play the same role for epic fantasy. And these days, lots of works dabble in it: think of Christopher Nolan’s attempt to give us a more “realistic” take on the Batman mythos, or the Daniel Craig Bond movies, which tried to do the same thing with James Bond.

HPMOR works in a similar vein, by giving us a version of Harry who was raised by an Oxford professor, one who’s smart enough to notice all the things that don’t quite make sense in J. K. Rowling’s fictional universe. It’s not everybody’s cup of tea, but I found much of it wildly entertaining.

Unfortunately, I also found the ending to be a huge let-down–which is a real problem for a story like this, because at least half the story is set up as a giant puzzle, tempting the reader to believe that any apparent plot holes would be plugged in some incredibly clever way in the end. When the ending doesn’t deliver, it makes those parts of the story look a lot less impressive in retrospect.

At this point, I have say SPOILER ALERT!, not just for HPMOR but Watchmen, Joss Whedon’s Avengers film, and the first three books in A Song of Ice and Fire (/slash the first three seasons of the latter’s TV adaptation A Game of Thrones.)

Still with me?

Okay, so at the end of HPMOR it’s revealed that in this version of the story, the reason Voldemort was such a great threat in the First Wizarding War is that he was a tactical and strategic genius unmatched in the rest of the the magical world. Among other things, it’s hinted early on that this version of Voldemort wrote the Evil Overlord List, a famous internet posting on how to avoid all the mistakes of James Bond-esque villains.

When he heard the prophecy about Harry defeating him, instead of stupidly walking straight into the trap like in canon, he tried to devise a plan to fulfill the prophecy on his own terms, which involved overwriting the baby Harry’s personality with his own, strategic genius included, to set Harry up has his “equal.” Except this plan blew up in his face anyway (because something something magic resonance), so Voldemort had to wait to return until Harry’s first year at Hogwarts like in the books.

Finding this out was my first major disappointment with HPMOR–it had been hinted that this version of Voldemort was a lot smarter than the canon version, but I figured that would mean he faked his own death rather than being tricked into destroying himself, for some inscrutable reason that would later be revealed. What actually ended up being revealed was just a more elaborate justification for the same result we got in Rowling’s books.

Okay, but that was sort-of understandable. What’s really inexcusable is the climax. You see, originally, Voldemort was planning on using the mini-self he’d created in Harry to rule Magical Britain for him, but then he heard another prophecy saying Harry was going to destroy the world, which Voldemort wasn’t going to stand for, because you can’t rule the world if it’s been destroyed. And Voldemort decided based on past experience that the whole “fulfill prophecy on your own terms” thing was a mistake, so he decides to just kill Harry.

Which leads to Harry finding himself surrounded by Voldemort and 37 Death Eaters, all with their wands pointed at them, and the Voldemort says… that before he kills Harry, he’ll give Harry a chance to trade any remaining secrets Harry knows for the lives of his friends and family.

Harry, of course, uses the extra time to kill all the Death Eaters and incapacitate Voldemort indefinitely, through incredibly clever means the details of which aren’t relevant to this review, because of course he does, you didn’t think this story was going to have an unhappy ending, did you?

What makes this ending such a disaster is that having a hero’s victory depend on the villain not just shooting him at the first opportunity is probably the single most infamous brand of fictional villain stupidity in existence. Seeing that trope used in 2015 is unavoidably painful. But using that trope after a hundred chapters of not-so-subtle hints that your villain so much smarter than all those other fictional villains? That’s just unforgivable.

A bunch of people on the HPMOR subreddit noticed this and complained, and somewhere Eliezer justified it by saying that Voldemort “underestimated Harry’s threat level.” Which seems to misunderstand why Bond villain stupidity is so grating.  In any given instance, you can always say, “well, Blofeld underestimated Bond’s threat level.” But once you notice the pattern, it becomes obvious that the real reason for the mistake isn’t whatever in-universe justification is given, but because it was the only way the writers could get their happy ending.

The first time a writer did this, it was fine. After all, in the real world it’s extremely uncommon for captured secret agents to not only escape but kill the leader of the organization that captured them in the process. When, in the 2012 Avengers movie, after decades of such scenarios, Loki fails to just stab Tony any instead throws him out the window (giving him a chance to suit up as Iron Man mid-fall), it’s significantly more annoying.

Even then, though, it’s forgivable, because if you manage to overlook it, it makes the movie more fun. Even if Robert Downey Jr. hadn’t been under contract for another Iron Man movie, the Avengers is the kind of movie that demands you don’t kill of such a major character without giving him a better send-off. And while you could have avoided both the cliche and the character death by skipping the scene entirely, it would have deprived us of Tony quipping at Loki, and everyone knows Tony’s quipping is one of the best things about the Marvel Cinematic Universe.

Therein lies the dilemma for the writer of something like HPMOR: these story telling conventions don’t exist because writers are stupid, but because they make for fun stories. If you’re going to build your story around attacking them, eventually you’re going to have to do something your readers won’t like, like letting the villain successfully complete his plot before the heroes even arrive (Watchmen). Or you have to kill your hero, have you ex-protagonist’s son try to avenge him, and then kill the son for good measure (A Song of Ice and Fire).

The problem with doing this devoted connoisseurs of your genre will appreciate having their expectations subverted, most of your potential audience won’t. So sure, Watchmen is widely regarded as the greatest comic book of all time, but will its characters ever enjoy the broad popularity of Batman and Superman? I suppose George R. R. Martin may avoid that fate by never having had a single main character, letting him kill off a few of his protagonists while keeping others around to deliver a satisfying ending–we’ll see.

The alternative is to dabble in deconstruction, then chicken out. This is what Christopher Nolan’s Batman movies and Daniel Craig’s Bond movies do and I think the result is basically a disaster. Pointing out the implausibilities of previous iterations of your franchise, but still trying to give your audience the genre-fiction experience they want, just means you’ll end up falling back on versions of those implausibilities in the end, and they’ll be all the more painful because you pointed them out yourself.

If Eliezer had had the courage of his convictions, the story would have ended with Harry dying in a way where he could not possibly come back from the dead and then Hermione having to defeat Voldemort after him. Not necessarily that, but something on roughly that level of weirdness, and definitely not an what amounted to setting up an elaborate justification to have Voldemort make basically the same mistakes he made in canon.

Okay, I could say more about the issues I have with HPMOR, but this is the one I most wanted to get off my chest. And now I have to go, because just because I found the ending a terrible disappointment doesn’t mean I’m not going to the HPMOR wrap party in Berkeley tonight.