91Joined Nov 2014


"I think this article would have been better if it had noted these issues."

Yes, it would have! Very glad you raised them. This is part of what I had in mind when mentioning "reputational risk" but I'm glad you fleshed it out more fully.

That being said, I think there is a low cost way to reap the benefits I'm talking about with integrity. Perhaps we have different standards/expectations of what's misleading on a resume, and what kind of achievements should be required for certain accolades. Maybe a 20 min presentation that required a short application should be required before doing this. I don't know. But I find it hard to believe that we couldn't be much more generous with bestowing accolades to dedicated members of the community without engaging in deception.

Maybe I can try to restate this in a way that would seem less deceptive...

I genuinely believe that there are tons of deserving candidates for accolades and speaking engagements in our community. I think that we can do more to provide opportunities for these people at a very low cost. I hope to help organize an event like this in NYC. I probably wouldn't leave it open to just anyone to participate, but I would guess (from my experience with the NYC community) that few people would volunteer to speak who didn't have an interesting and informed perspective to share in a 15 minute presentation. Perhaps, I have an overly positive impression of the EA community though.

(ps. I think your response is a model of polite and constructive criticism. thanks for that!)

I recognize that I'm a novice on these issues, and I'm open to being persuaded about this, but that position just seems incredibly counterintuitive to me. Of course, that's doesn't mean it's not true, and if I had to guess right now, I would say it's right (based on the weight of opinions of those who have been more involved in AI than me). But I'd really like to see more discussion about the strategy here.

So do you think other kinds of non-event programming could be useful? Like an in depth blog post, or a podcast episode?

I didn't realize that both parties had to consent. What about the WTO?

Great comment. I'd like to follow up in even more detail later. But my main reaction is (1) I was wrong about some important things and will explain below; and (2) that yes, everyone should be skeptical about international law and doubly so about obligations under the ICESCR (and triply so when I say something about it ☹). As a matter of course, international law is regularly violated and not enforced. I should have been more clear about this, although my question "why does this matter?" and my modest response ("at least we can shame them") implicitly recognized that we can't just go to court and double foreign aid spending.

So on the details of enforcement, I did some more research and found that I was really wrong about at least two important things. First, the European human rights system did not directly incorporate the ICESCR and second, their primary economic rights document, the European Social Charter (ESC), is enforced by European Committee of Social Rights (ECSR), which issues non-binding judgments, unlike the European Court of Human Rights. While their decisions are non-binding, my impression from my classes is that their decisions are taken very seriously and states often adjust their national policy to conform to the rulings of the Committee. The biggest problem for my idea is not that their decisions are non-binding but that there doesn’t appear to be any similar obligation to provide foreign assistance in the ESC. I’ll have to look elsewhere.

So ultimately I’m not sure if there is a venue where this could be litigated with even a small potential of success. I’m less optimistic than I was before but I’ll keep looking to see whether any other regional or national human rights systems have incorporated this extraterritorial obligation into their laws. One potentially viable option is the International Court of Justice, although this would require finding a developing country to sponsor the litigation and effectively sue a wealthy country. That would be pretty cool but I’m not sure if it’s possible.

But as I stated earlier, I think knowledge of international legal obligations is useful regardless of whether they can be enforced through a court. If countries bring themselves into compliance through their legislatures to meet the demands of non-binding UN bodies, then there may still be utility in raising the issue of their noncompliance/violations in our advocacy. Again the UK just two years ago cited to their international obligation as the reason for raising their aid spending to the magic .7% mark. This to me shows that such international obligations are not completely useless. Personally, I think the EA community has much more learn about whether/how effective it is to advocate for the global poor through human rights mechanisms. It’s certainly not an unreasonable prior to be very skeptical, but there is a vibrant academic debate on the subject that seems very empirically grounded. I think we could profit a lot by looking into this further.

I'm going to reply with a long response soon. But unsurprisingly, I was very wrong about the potential for enforcement in the EU. That being said, still think the idea is useful. I'll explain. Thanks for your criticism. I was indeed confused.

Does abnormal self-sacrifice inspire social change?

The conventional wisdom is that many movements throughout history (christianity, buddhism, Indian independence) were inspired and fueled by the extraordinary sacrifice of leaders/early adopters (or at least myths about such extreme altruism). The conventional wisdom may be wrong, but maybe we need more abnormal sacrifice in our movement, not less. In fact, I think it's plausibly a good idea for us to donate our kidneys, precisely as a symbol of our commitment-- not necessarily in the hopes that others will follow suit-- but in the hopes that it inspires people to take altruism more seriously.

In theory, local EA chapters can do all this already without focusing on local problems. This is not an argument for donating locally as much as an argument for focusing how to take on collaborative, visible, and engaging projects in our communities.

Of course work for effective causes might not be as engaging or enticing in our communities but then this just becomes an extension of the argument against weirdness: "AI is weird, but so is global poverty relative to the local soup kitchen."

Anyway, I welcome the broad criticism: maybe we should be thinking more about how to create engaging events in our communities, partnering with non-EAs and non EA organizations, and being more visible in the local sphere.

I'm so excited by all the recent public discussion about movement building. It's really encouraging to see so many brilliant people investing their time and energy into this neglected area.

That being said, I am concerned that we are reinventing the wheel, and ignoring a substantial body of empirical and theoretical work that has already been done on the subject.

Why are we starting from scratch and developing novel theories of social change? Why are we focusing on mathematics and philosophy instead of academic sociology research? I'm not an expert but, I'm familiar enough to know that lots of other smart people have studied the issues addressed in this paper. Lots of people are interested the growth and strategy of social movements.

If we were talking about ending global poverty, we would not be postulating new models of economic development. Why should we demand any less empirical/academic rigor in the context of movement building? Why are we so willing to trust our intuitions here?

I think there are two common reasons for ignoring academic sociology research here (but both of them are pretty weak):

  1. The research on movement building is extremely shallow and of poor quality
  2. The EA commitment to cause neutrality is so unique that analogies to other movements (and to existing academic research) are not very useful

To address the first point, I think that we have to consider "EA expert overconfidence" bias. As Rob Wiblin has pointed out, people who are experts in one area are often radically overconfident in other areas. I think EAs succumb to this pretty severely: we are all so shocked (rightly so) at how much cause prioritization is neglected by smart people that we think we have to basically do everything from scratch. But this isn't quite right. We need to distinguish between "effective means" and "effective ends." EA's might be world leaders when it comes to thinking about effective ends (i.e. worthwhile causes like global poverty, animal suffering, far future suffering etc.) but we have no reason to think we are superior when it comes to effective means. Smart people have been trying to understand the spread of ideas and the grow movements for a long time. We should be shocked if there isn't at least some good work done on the subject. My own shallow research has left me convinced that there is a lot of good stuff out there. Even though sociology has the reputation for being less rigorous than economics, there is a lot of serious, rigorous empirical and theoretical work out there.

I think the second point is also not a huge issue. First, lots of other social movements have faced the problem of maintain a broad base of support in the face of ever-changing goals/priorities (political parties and religions seem like good examples). But more importantly, even if we are unique in this regard, it seems that many of the big questions in movement building apply equally well to either case.

Ultimately, if I was GPP, I would try to convene a working group of non-EA academic experts on social movements before trying to do any more original thinking on the issue.