Monday, July 21, 2014







can someone actually make the case against capitalism to me? the general consensus here seems to be that it is Awful and Should Be Overthrown when I was under the impression that it’s better at incentives and at allocating resources than all the other systems we’ve tried and every one that’s been seriously proposed

#every communist I have read is a complete imbecile and it’s making me unfairly prejudiced against communists#capitalism

Well, capitalism by default results in some people being poor, and that’s really terrible. Ideally, you fix that with basic income, somebody should try that in real life so we can see if it works (preliminary trials suggest it almost certainly will).

The sorta fundamental ideal of capitalism is that you set up a system where doing pro-social things gets you rich, and being rich gets you access to cool toys, and everybody runs around doing pro-social things all the time.

And it’s an awesome fundamental ideal, and it works out precisely as intended way the hell more often than anticaps will acknowledge.

But sometimes you wind up with people running around doing anti-social things and getting rich off that instead. Economists call this “rent-seeking behavior”. The classic story about this is the kids with a lemonade stand who give out free peanuts, soaked in capsaicin, so now everyone desperately wants lemonade. These kids did not do anything to, y’know, increase net-utility. Profit motive didn’t lead them to pro-social behavior. And I think this happens in real life. The most obvious case to me is advertising. A lot of times advertising just makes you feel bad unless you have the thing. Sometimes this causes large and negative changes at the societal level. Advertisers invented body odor, they invented halitosis, they invented the two-months-salary rule and the idea of the diamond engagement ring as mandatory.

This is all related to the fact that humans are irrational. Like, we capitalists sometimes point out that socialists are wrong about human nature, that humans are not perfect communitarians who are going to be super motivated to work for the common good every day. But I think that, like, saying “Adam Smith, therefore capitalism is ever and always awesome” makes a similar mistake, and assumes that humans are perfect rational selfish agents, and we’re not. We discount hyperbolically. We experience hedonic treadmills. Sometimes that means that making more stuff available doesn’t make us as well off as it should. Sometimes making more stuff available makes us feel less good about what we have.

Communists sometimes talk about the alienation of labor. Like, under capitalism, you do one thing super well and you make money for it and the money solves your other problems, so there’s this, like, indirection between the labor you do and your well-being, you’re not directly engaged in self-care the way you would be if you were like making all your stuff for yourself. There’s probably more to this and I’m probably not understanding it that well, but a lot of people seem to think it’s important?

So far, empirically, capitalism sure appears to have done a poor job of dealing with long-term renewability of natural resources. Like, I know you can probably come up with brilliant solutions involving like ownership of futures contracts and special property rights arrangements, but historically this hasn’t happened.

Also, capitalism is not libertarianism, but Scott’s anti-libertarian FAQ (Why I Hate Your Freedom) may be relevant

So there’s a vote in Switzerland to bring a guaranteed basic income in, and proposals elsewhere, but to my knowledge it hasn’t been tried and failed anywhere.

If you’ll all forgive me for being cliché, Marx sort of covered the “what’s wrong with capitalism” bit already. To sum it up, a middle class—which is really what all “nice” proponents of capitalism want a lot of—is not a natural feature of capital accumulation. Wealth pools at the top without government intervention; I think the original response did a good job of explaining why this is. We see the evidence now; the less regulation on capital, the more wealth disparity. The market is actually shit at allocating resources; we have enough food to feed everyone, but we waste tons of food while the developing world starves, etc. It may be, at the present day, slightly less shit than certain planned economies like Stalin’s USSR, but in no way effective in purely utilitarian, let alone ethical terms.

Well, why not more government intervention—market socialism? I am not a good doctrinaire communist in that I actually think this is fine as an intermediate measure and prefer it to, say, a revolution in which I would inevitably be killed horribly, but I think it’s only a stop-gap measure. Why? Two reasons:

1) People consistently do not vote in their best interest. There are a ton of cultural and psychological reasons for this, but essentially people, at least in North America and increasingly elsewhere, will take a 50 cent tax cut over transit or health care infrastructure that will save them substantially more in the long-run. Economics is haaaaaard.

2) Governments can be bought and are bought, frequently. If you look at surveys of people’s actual political beliefs versus what lobbyists make happen, you’ll see that a strong state apparatus is no guarantee of market regulation.

Thank you both! So I agree that

1) people not getting the things they need is a bad outcome

2) unregulated markets result in worse outcomes than regulated markets in lots and lots of cases

(I do believe that nearly all of the problems with capitalism you mention could be solved with halfway-sensible externality contracts, but if I’m going to hold communism accountable for the fact that whenever attempted it kills millions of people, I have to also hold capitalism accountable for the fact it hasn’t solved those problems.)

… but I think I am entirely unconvinced that as a result we should give up on markets. Or that we even could, actually. In basically every environment where you attempt to replace markets with another means of distributing things, markets pop right up anyway, because as long as some people have things that other people want, they’re going to trade them. Any possible strategy for preventing this does far, far more harm than is done by people making suboptimal choices in the markets themselves. 

It might be useful here to distinguish between a few things? I can imagine a country with a universal basic income which nonetheless allows markets to set the price of labor and goods. That country sounds fantastic and I’d love to live there. I’d consider that country capitalist; would you?

I can also imagine a country where some criterion other than supply and demand is used to set the price of labor and goods. I am reasonably certain that such a society would be a disaster resulting in far more suffering than even the most safety-net-less captialist societies.

I also think that empirically wealthier nations have better safety nets, so the economic system that probably results in the most generous benefits fifty years from now is the one that creates the most wealth now. Do we disagree on that?

To sum it up, a middle class—which is really what all “nice” proponents of capitalism want a lot of—is not a natural feature of capital accumulation.

All ‘nice’ proponents of capitalism want this? It’s not that it seems obviously wrong to me to want this, but it doesn’t seem obviously right either. I want peoples’ preferences to be fulfilled. I want middle class people exactly to the extent they are happier than rich people or working-class people or people living on benefits. I can’t find good research on this, but as far as self-reported happiness goes the poor are happier than the middle class.  (The rich are happiest). 

And Michael, thanks for the mention of dignity-of-labor type considerations; that’s one aspect of communist ideology I find particularly impossible to understand. I would prefer no one ever have to do work that’s related to their self-care and survival; self-care and survival are my least favorite parts of existing, and if my work and intellectual productivity were also directly connected to them (for example, if I had to build my house or grow my food or sew my clothes) I would be utterly miserable. Sometimes I’m cripplingly unhappy over the fact that I have to brush my teeth every single damn day. And that’s about as directly-related-to-your-own-health-and-wellbeing as you can possibly get. 

 Is this psychologically rare? Are most people longing to do work that’s more connected to their immediate needs? Or is that a desire that only a very small percentage of the population has, and mostly because it’s been romanticized beyond all recognition?

(I will change my mind here if enough people chime in to say that this is a desire of theirs; I’m a psychological outlier along enough axes that it would not surprise me to learn that I am an outlier here as well.)

You all seem to be using a very different concept of “alienation of labor” than the one I have. My understanding of the concept of “alienation of labor” is that workers are not working towards intrinsic goals, but extrinsic ones set by people other than themselves. The goal of work is not doing a job and feeling proud and satisfied in a job well done, but rather making money so that you can survive. The worker doesn’t necessarily feel like their labor is needed or useful or helpful to society - just like their employer will pay them to do it. They end up feeling like cogs in the machine or like tools of the system rather than like fulfilled people or vital, contributing members of society. Because they require money to survive and are paid money to work, the primary motivation to work becomes the wage working pays. So in a communist system (or other system without alienation of labor), people would not be forced to allocate their labor based on what makes them the most money, and would instead work jobs that they are skilled at and feel satisfaction from performing and feel like are actually making a positive difference in society. Basically, it’s not about whether your work is self-care or base survival, it’s about whether you feel like you’re actually using your skills and talents in a meaningful way.

Yep, I did not understand this part very well, thanks for the clarification =)

(…) A systems programmer has seen the terrors of the world and understood the intrinsic horror of existence. James Mickens - The Night Watch (via hot-gay-rationalist)
Wednesday, July 16, 2014






If bees become extinct we will have exactly 4 YEARS to live on this planet. I don’t understand how “not giving a fuck” is more important than your life…

okay, I have a thing to say about this. I’m no expert on bees, but I am a biologist (and entomologist) so I think there is something I can contribute that’ll be of worth.

I agree entirely with the sentiment that we must protect honeybees. Obviously they are massively important for biodiversity, as well as pollinating food crops for humans. There is no doubt that if all the honeybees in the world were to vanish in a day that the consequences would be dire.

However, I disagree that the main cause for concern regarding honeybee death is the use of Genetically Modified (GM) crops. I’d be very interested to read a research paper that says ‘GM crops have killed millions of honeybees’, if indeed such a paper exists because in all honesty I find it highly unlikely that this is a true statement.

Let’s start with some facts about GM crops:

1. The development of GM crops is a highly regulated process, bound by strict country-specific legislature. A great number of trials are carried out long before commercial planting of a GM crop is even considered. It is these trials, and accompanying laboratory studies, that ensure a GM crop is safe to non-target organisms (such as honeybees) by investigating direct and indirect effects (Nap et al. 2003).

2. Crops that are genetically modified to express insecticidal proteins (for crop pest control) have a high level of specificity. This means that the insecticidal proteins being produced by the GM plant will only affect a narrow range of insect groups because of the chemical properties of the protein. For example, GM crops expressing insecticidal proteins sourced from the bacterium Bacillus thuringiensis (Bt) will only target some Lepidopteran pests (caterpillars; Romeis et al. 2006). Furthermore, a recent meta-analysis of the literature found that GM Bt crops do not negatively affect the survival of adult honeybees or their larvae (Duan et al. 2008).

3. GM crops can be tailored such that the novel gene is expressed only in particular parts of the plant. For example, GM Bt rice plants express the toxin in the stems but not the grains (Datta et al. 1998). This technique means that gene expression can be excluded from the flowers/pollen of the crop plant, so that bees and other pollinators would not be affected. Neat, huh?

So those are a token few reasons why GM crops are safer than perhaps many people believe (as the result of a lot of questionable, non-scientific articles). To come back to our main point about honeybee death, I would like to briefly mention a few alternative explanations for the recent decline in honeybee populations. These are as follows:

1. Many bees have died as the result of broad-spectrum insecticide use. These are pesticides that lack specificity, and can be harmful to non-target organisms. Neonicotinoids are a well-studied example of this (Decourtye & Devillers, 2010). Not to worry, though, because many broad-spectrum pesticides including neonics are well on their way out. Indeed, the EU recently banned a large cohort of neonic pesticides. This is still a topic of controversy, mind (Goulson, 2013).

2. Many bees have died as the result of Varroa mite infestation. Imagine you’ve been bitten by several ticks, except those ticks are the size of dinner plates. That gives you an idea of the severity of a Varroa mite infestation on a single developing bee. The parasitisation of bees by Varroa mites and other parasites is often accompanied by disease transmission. This can result in colonies dying within two years after infestation (Johnson, 2011).

3. Many bees have died as the result of ‘colony collapse disorder’.  This is a phrase that has popped up a lot recently, and is basically an umbrella term for the various causes of bee death including parasite infestation, disease transmission, environmental stresses, and management stresses such as poor nutrition (Johnson, 2011). Colony collapse has been attributed to broad-spectrum pesticide use in some instances. However, it is has still been observed in countries where broad-spectrum pesticides have been withdrawn (in the EU, like I mentioned earlier; Johnson, 2011).

So those are my main points. Please excuse the bullet-point nature of this; I was trying to keep it fairly short. Not sure I managed that haha. But anyway, my take-home message is that GM crops are not the enemy when it comes to honeybee decline. If anything, bees are at much greater danger from the use of broad-spectrum pesticides and from parasites and diseases. Using GM can even help to alleviate some of the problems associated with broad-spectrum pesticides, as they greatly reduce the need to apply such chemicals (Romeis et al. 2006).

A finishing note: Do your homework. Go on google scholar and read some of the literature, making sure it is recent (within the past 10-15 years). Literature reviews are a great way to find out what the consensus is on any given topic. Don’t use popular media as your main source of information where science is concerned; they tend to favour scandal and exaggeration. You want to know what’s really going on? Check out some research articles and see for yourself.

Thanks for sticking it through to the end of this impromptu mini-essay! —Alice


Datta, K., Vasquez, A., Tu, J., Torrizo, L., Alam, M. F., Oliva, N., Abrigo, E., Khush, G. S., & Datta, S. K. (1998). Constitutive and tissue-specific differential expression of the cryIA (b) gene in transgenic rice plants conferring resistance to rice insect pest. Theoretical and Applied Genetics, 97(1-2), 20-30.

Decourtye, A., & Devillers, J. (2010). Ecotoxicity of neonicotinoid insecticides to bees. In Insect nicotinic acetylcholine receptors (pp. 85-95). Springer New York.

Duan, J. J., Marvier, M., Huesing, J., Dively, G., & Huang, Z. Y. (2008). A meta-analysis of effects of Bt crops on honey bees (Hymenoptera: Apidae). PLoS One, 3(1), e1415.

Goulson, D. (2013). Neonicotinoids and bees: What’s all the buzz?. Significance, 10(3), 6-11.

Johnson, R. (2011). Honey bee colony collapse disorder. DIANE Publishing.

Nap, J. P., Metz, P. L., Escaler, M., & Conner, A. J. (2003). The release of genetically modified crops into the environment. The Plant Journal, 33(1), 1-18.

Romeis, J., Meissle, M., & Bigler, F. (2006). Transgenic crops expressing Bacillus thuringiensis toxins and biological control. Nature biotechnology, 24(1), 63-71.

This commentary is SO important. Succinct and with proper sourcing; beautiful.

It infuriates me when people blame GMO for everything without actually examining the evidence.


just everyone save bees from swimming pools like me okay?

(Source: antinwo)



honestly if you dont think like, the tumblr feminist scene, with all the occasionalyl cheesy kawaii-aesthetic misandry art, hasn’t had an impact on anyone at all like

you dont remember what the average teen girl in a fandom was like before this. you don’t remember how we used to make hate-sites about female characters who “got in the way”, games where you could beat them up, how much we hate our gender and bragged about not being like other girls, used to completely reject everything girly. a lot of us just wanted to be one of the guys. there was a lot of internalised misogyny there

now you get these 15 year old girls loving other girls and loving themselves fiercely, even at the total cost of male approval and just. god. if like 14-year-old me could see this shit now. 

and like if you dont think teen girls learning to love themselves and their body and each other isnt important than i do not know what to say to you

I know I’m always focusing on the flaws in tumblr’s brand of feminism—because I want it to improve tbh—but this is really true as well.

(Source: cephalodogs)

Starting to liveblog madoka’s third movie!


On my other blog.

Monday, July 14, 2014

why i’m not into FAI







Posting this for scientiststhesis, who asked.  (Asked under his other account but I can’t seem to at-sign it for reasons probably having to do with hyphens?)

Attention conservation notice: this was originally written in an email to someone who asked what I thought of Friendly AI.  If the phrase “Friendly AI” means nothing to you then this is probably not going to make any sense.

As I said one post ago: paints in broad strokes, is very informal about everything, and I imagine much of it will not meet your [scientiststhesis’] standards for argument, but it does sketch a set of arguments that could be fleshed out more.

Read More

This is… weird? Because almost everything you say “LWers are” or “LWers do” is almost universally not true of LWers I know, or myself. But anyway.

Read More

Warning: long.  (Can you tell I’m procrastinating?)

Read More

Here we go.

Read More

Thanks for the response.  I think we have gone most of the way we can go here?

Most of my objections to FAI research are the result of skepticism about AI.  But that still causes me to consider FAI research less important.  When I say I’m “not into FAI” I’m not saying that current FAI research is poorly done for what it is and should be done differently.  I’m saying that FAI research doesn’t seem to me to be especially important, and I don’t understand the rationale behind, say, donating money to people to do it.

(Is FAI uniquely deserving of donations relative to all other math or philosophy research?  Does it make sense to donate to MIRI while not offering money to, say, any academics who are having trouble getting grants?)

Right. But if you take the premise that self-modifying AI is possible, which a large part of the research community does, then the extra assumptions needed for FAI to be relevant are just UF is the default, F is a hard problem, AIs can become reasonably superintelligent (FOOM is not strictly necessary). If one believes that, then donating to MIRI is one of the best decisions they could make.

There is also some semantics here about what counts as “AI.”  It’s easy for me to imagine a world where the only superintelligences we can make are biological super-brains made of real neurons.  I would call these “AI” but maybe you wouldn’t; in any case some of FAI would not apply to them (e.g. because it’s hard to build a brain to precisely specified order without molecular nanotech, you can’t have “tiling agents” that make nearly identical copies of themselves).

I would certainly call them AI, and I’m an adept of the philosophy “Some things I cannot change, but ‘til I try I’ll never know.” If it turns out that AI won’t be reasonably strong without molecular nanotech (a thing to which I still ascribe such tiny probabilities that I’m feeling kinda guilty for even spending time thinking about it), I’m not going to throw my hands up in the air and give up until and unless I find positive evidence that other kinds of AI are impossible and so is molecular nanotech.

I still feel like I haven’t made the “creativity” point clear even to myself, and probably need to think about it more.  I think what I’m probably trying to say is that FAI people ignore time and space efficiency — for instance one MIRI report cites Marcus Hutter’s AIXI-related definition of intelligence which makes no mention of time or space efficiency.  Again, I’m not saying specifically that FAI research could be better as FAI research if it included these things (I’m not even sure what that would mean), but that much of the justification for thinking FAI is important and encouraging support for MIRI is the terrifying specter of unfriendly superintelligences way above us in terms of how much use they get per bit, and these aren’t very scary if computational complexity means it takes them 10^10 years to get anything done.

Sure, I suppose? But looking at that ideal definition and pointing to how inefficiently computable it is is like looking at Bayes’ Theorem and doing the same. It doesn’t really matter, we just need to find ways that provably approximate them in effectively computable ways such that we have a definitive upper-bound on the error we’re making (one example of a situation where we have that is in the difference between orthodox test statistics and Bayes’ Theorem).

(I have seen one LWer link that Yudkowsky post I linked earlier to illustrate “what is at stake” in FAI/UFAI.)

Bwahahahahah. People taking Yudkowsky fiction as evidence of anything in real life has to be the most ironic thing ever.

When I have more time I want to go through some MIRI reports closely and try to find points, if there are any, where I think they depend on specific assumptions about what AI will be like.  I’m too tired and busy right now to make this a very good post, but I’ll keep thinking about it in the coming weeks.

I’ll be eagerly awaiting! :)

raginrayguns said: I don't think I ever met any old people resigned to death? But I have met a few old people that saw it as unfortuante thattheir deathw ould be relatively soon, or at least thought there were specific major unfortunate things about that. Where... are you getting your impressions of old ppl's attitude towards death, 'cause it's either sampling a different subset of old ppl or just not reliable evidence or i need to question my observations

I mean, most people I know are resigned to death, young or old. Oh it’s just a part of life, or it gives life meaning, or some such bullshit, it’s not uncommon.

But anyway, I’d expect older people to be more resigned to dying that younger people on general principles. Are these old people you’ve met religious at all?

Idk though, lots of atheists I’ve heard of also spout the conformist “death is part of life” bullshit, sooo…