ishaan

May 2020: I've now been working at Charity Entrepreneurship for a little under a year. My primary responsibility is designing the curriculum for next year's incubation program.

May 2019: I've been lurking / supporting EA from the sidelines since around 2011, reading content from here, LW, and OB. This year, my involvement has increased, as I attended my first EAG in London, stayed at the EA hotel, and volunteered a bit for Rethink Priorities. At time of writing, I'm about to do CharityEntrepreneurship's incubation program. Most of my recent activity on the forum is geared to improving my knowledge of things which I think will help me succeed at creating an impact via that avenue.

-Ishaan Guptasarma

ishaan's Comments

Where is it most effective to found a charity?

I've never done this myself, but here's bits of info I've absorbed through osmosis by working with people who have.
-Budget about 50-100 hours of work for registration. Not sure which countries require more work in this regard.
-If you're working with a lot of international partners, some countries have processes that are more recognized than others. The most internationally well-known registration type is America's 501(c)(3) - which means that even if you were to for example work somewhere like India, people are accustomed to working with 501(c)(3) and know the system. Less important if you aren't working with partners.
-If you are planning to get donations from mostly individuals, consider where those individuals are likely to live and what the laws regarding tax deductibleness are. Large grantmakers are more likely to be location agnostic.
-You don't need to live where you register, but if you want to grant a work visa to fly in an employee to a location, generally you will need to be registered in that location.

If you're interested in starting a charity you should consider auditing Charity Entrepreneurship's incubation program, and apply for the full course next year. Audit course will have information about how to pick locations for the actual intervention (which usually matters more than where you register for your impact). The full course for admitted students additionally provides guidance and support for operations/registration type stuff.

EA Forum feature suggestion thread

I posted some things in this comment, and then realized the feature I wanted already existed and I just hadn't noticed it - which brings to mind another issue: how come one can retract, overwrite, but not delete a comment?

Dignity as alternative EA priority - request for feedback
What evidence would you value to help resolve what weight an EA should place on dignity?

Many EAs tend to think that most interventions fail, so if you can't measure how well something works, chances are high that it doesn't work at all. To convince people who think that way, it helps to have a strong justification to incorporate a metric which is harder to measure over a well established and easier to measure metrics such as mortality and morbidity.

In the post on happiness you linked by Michael, you'll notice that he has a section on comparing subjective well being to traditional health metrics. A case is made that improving health does not necessarily improve happiness. This is important, because death and disability is easier to measure than things like happiness and dignity, so if it's a good proxy it should be used. If it turned out the that the best way to improve dignity is e.g. prevent disability, then in light of how much easier to measure disability prevention is, it would not be productive to switch focus. (Well, maybe. You might also take a close association between metrics as a positive sign that you're measuring something real. )

To get the EA community excited about a new metric, if it seems realistically possible then i'd recommend following Michael's example in this respect. After establishing a metric for dignity, try to determine how well existing top givewell interventions do on it, see what the relationship is with other metrics, and then see if there are any interventions that plausibly do better.

I think this could plausibly be done. I think there's a lot of people who favor donations to GiveDirectly because of the dignity/autonomy angle (cash performs well on quite a few metrics and perspectives, of course) - I wouldn't be surprised if there are donors who would be interested in whether you can do better than cash from that perspective.

EA considerations regarding increasing political polarization
Why effective altruists should care

Opposing view: I don't think these are real concerns. The Future of Animal Consciousness Research citation boils down to "what if research in animal cognition is one day suppressed due to being labeled speciesist" - that's not a realistic worry. The vox thinkpeice emphasizes that we are in fact efficiently saving lives - I see no critiques there that we haven't also internally voiced to ourselves, as a community. I don't think it's realistic to expect coverage of us not to include these critiques, regardless of political climate. According to google search, the only folks even discussing that paper are long-termist EAs. I don't think AI alignment is any more politically polarized except as a special case of "vague resentment towards silicon valley elites" in general.

Sensible people on every part of the political spectrum will agree that animal and human EA interventions are good or at least neutral. The most controversial it gets is that people will disagree with the implication that they are best ways to do good...and why not? We internally often disagree on that too. Most people won't understand ai alignment enough to have an opinion beyond vague ideas about tech and tech-people. Polarization is occurring, but none of this constitutes evidence regarding political polarization's potential effect on EA.

EA and tackling racism

a) Well, I think the "most work is low-quality aspect" is true, but also fully-general to almost everything (even EA). Engagement requires doing that filtering process.

b) I think seeking not to be "divisive" here isn't possible - issues of inequality on global scales and ethnic tension on local scales are in part caused by some groups of humans using violence to lock another group of humans out of access to resources. Even for me to point that out is inherently divisive. Those who feel aligned with the higher-power group will tend to feel defensive and will wish not to discuss the topic, while those who feel aligned with lower-power groups as well as those who have fully internalized that all people matter equally will tend to feel resentful about the state of affairs and will keep bringing up the topic. The process of mind changing is slow, but I think if one tries to let go of in-group biases (especially, recognizing that the biases exist) and internalizes that everyone matters equally, one will tend to shift in attitude.

EA and tackling racism
I've seen a lot of discussion of criminal justice reform

Well, I do think discussion of it is good, but if you're referring to resources directed to the cause area...it's not that I want EAs to re-direct resources away from low-income countries to instead solving disparities in high income countries, and I don't necessarily consider this related to the self-criticism as a community issue. I haven't really looked into this issue, but: on prior intuition I'd be surprised if American criminal justice reform compares very favorably in terms of cost-effectiveness to e.g. GiveWell top charities, reforms in low income countries, or reforms regarding other issues. (Of course, prior intuitions aren't a good way to make these judgements, so right now that's just a "strong opinion, weakly held".)

My stance is basically no on redirecting resources away from basic interventions in low income countries and towards other stuff, but yes on advocating that each individual tries to become more self-reflective and knowledgeable about these issues.

I suppose the average EA might be more supportive of capitalism than the average graduate of a prestigious university, but I struggle to see that as an example of bias

I agree, that's not an example of bias. This is one of those situations where a word gets too big to be useful - "supportive of capitalism" has come to stand for a uselessly large range of concepts. The same person might be critical about private property, or think it has sinister/exploitative roots, and also support sensible growth focused economic policies which improve outcomes via market forces.

I think the fact that EA has common sense appeal to a wide variety of people with various ideas is a great feature. If you are actually focused on doing the most good you will start becoming less abstractly ideological and more practical and I think that is the right way to be. (Although I think a lot of EAs unfortunately stay abstract and end up supporting anything that's labeled "EA", which is also wrong).

My main point is that if someone is serious about doing the most good, and is working on a topic that requires a broad knowledge base, then a reasonable understanding the structural roots of inequality (including how gender and race and class and geopolitics play into it) should be one part of their practical toolkit. In my personal opinion, while a good understanding of this sort of thing generally does lead to a certain political outlook, it's really more about adding things to your conceptual toolbox than it is about which -ism you rally around.

EA and tackling racism
What are some of the biases you're thinking of here? And are there any groups of people that you think are especially good at correcting for these biases?

The longer answer to this question: I am not sure how to give a productive answer to this question. In the classic "cognitive bias" literature, people tend to immediately accept that the biases exist once they learn about them (…as long as you don't point them out right at the moment they are engaged in them). That is not the case for these issues.

I had to think carefully about how to answer because (when speaking to the aforementioned "randomly selected people who went to prestigious universities", as well as when speaking to EAs) such issues can be controversial and trigger defensiveness. These topics are political and cannot be de-politicized, I don't think there is any bias I can simply state that isn't going to be upvoted by those who agree and dismissed as a controversial political opinion by those who don't already agree, which isn't helpful.

It's analogous to if you walked into a random town hall and proclaimed "There's a lot of anthropomorphic bias going on in this community, for example look at all the religiosity" or "There's a lot of species-ism going on in this community, look at all the meat eating". You would not necessarily make any progress on getting people to understand. The only people who would understand are those who know exactly what you mean and already agree with you. In some circles, the level of understanding would be such that people would get it. In others, such statements would produce minor defensiveness and hostility. The level of "understanding" vs "defensiveness and hostility" in the EA community regarding these issues is similar to that of randomly selected prestigious university students (that is, much more understanding than the population average, but less than ideal). As with "anthropomorphic bias" and as with "speciesism", there are some communities where certain concepts are implicitly understood by most people and need no explanation, and some communities where they aren't. It comes down to what someone's point of view is.

Acquiring an accurate point of view, and moving a community towards an accurate point of view, is a long process of truth seeking. It is a process of un-learning a lot of things that you very implicitly hold true. It wouldn't work to just list biases. If I start listing out things like (unfortunately poorly named) "privilege-blindness" and (unfortunately poorly named) "white-fragility" I doubt it's not going to have any positive effect other than to make people who already agree nod to themselves, while other people roll their eyes, and other people google the terms and then roll their eyes. Criticizing things such that something actually goes through is pretty hard.

The productive process involves talking to individual people, hearing their stories, having first-hand exposure to things, reading a variety of writings on the topic and evaluating them. I think a lot of people think of these issues as "identity political topics" or "topics that affect those less fortunate" or "poorly formed arguments to be dismissed". I think progress occurs when we frame-shift towards thinking of them as "practical every day issues that affect our lives", and "how can I better articulate these real issues to myself and others" and "these issues are important factors in generating global inequality and suffering, an issue which affects us all".

EA and tackling racism
What are some of the biases you're thinking of here?

This is a tough question to answer properly, both because it is complicated and because I think not everyone will like the answer. There is a short answer and a long answer.

Here is the short answer. I'll put the long answer in a different comment.

Refer to Sanjay's statement above

There are some who would argue that you can't tackle such a structural issue without looking at yourselves too, and understanding your own perspectives, biases and privileges...But I worried that tackling the topic of racism without even mentioning the risk that this might be a problem risked seeming over-confident.

At time of writing, this is sitting at negative-5 karma. Maybe it won't stay there, but this innocuous comment was sufficiently controversial that it's there now. Why is that? Is anything written there wrong? I think it's a very mild comment pointing out an obviously true fact - that a communities should also be self-reflective and self-critical when discussing structural racism. Normally EAs love self-critical, skeptical behavior. What is different here? Even people who believe that "all people matter equally" and "racism is bad" are still very resistant to having self-critical discussions about it.

I think that understanding the psychology of defensiveness surrounding the response to comments such as this one is the key to understanding the sorts of biases I'm talking about here. (And to be clear - I don't think this push back against this line of criticism is specific to the EA community, I think the EA community is responding as any demographically similar group would...meaning, this is general civilizational inadequacy at work, not something about EA in particular)

EA and tackling racism

I broadly agree, but in my view the important part to emphasize is what you said on the final thoughts (about seeking to ask more questions about this to ourselves and as a community) and less on intervention recommendations.

Is EA really all about taking every question and twisting it back to malaria nets ...?... we want is to tackle systemic racism at a national level (e.g. in the US, or the UK).

I bite this bullet. I think you do ultimately need to circle back to the malaria nets (especially if you are talking more about directing money than about directing labor). I say this as someone who considers myself as much a part of the social justice movement as I do part of the EA movement.Realistically, I don't think it's really plausible that tackling stuff in high income countries is going to be more morally important than malaria net-type activities, at least when it comes to fungible resources such as donations (the picture gets more complex with respect to direct work of course). It's good to think about what the cost-effective ways to improve matters in high income countries might be, but realistically I bet once you start crunching numbers you will probably find that malaria-net-type-activities should still the top priority by a wide margin if you are dealing with fungible resources. I think the logical conclusions of anti-racist/anti-colonialist thought converge upon this as well. In my view, the things that social justice activists are fighting for ultimately do come down to the basics of food, shelter, medical care, and the scale of that fight has always been global even if the more visible portion generally plays out on ones more local circles.

However, I still think putting thought into how one would design such interventions should be encouraged, because:

our doubts about the malign influence of institutional prejudice...should reach ourselves as well.

I agree with this, and would encourage more emphasis on this. The EA community (especially on the rationality/lesswrong part of the community) puts a lot of effort into getting rid of cognitive biases. But when it comes to acknowledging and internally correcting for the types of biases which result from growing up in a society which is built upon exploitation, I don't really think the EA community does better than any other randomly selected group of people who are from a similar demographic (lets say, randomly selected people who went to prestigious universities). And that's kind of weird. We're a group of people who are trying to achieve social impact. We're often people who wield considerable resources and have to work with power structures all the time. It's a bit concerning that the community level of knowledge of the bodies of work that deal with these issues is just average.I don't really mean this as a call to action (realistically, I think given the low current state of awareness it seems probable that attempting action is going to result in misguided or heavy-handed solutions). What I do suggest is - a lot of you spend some of your spare time reading and thinking about cognitive biases, trying to better understand yourself and the world, and consider this a worthwhile activity. I think, it would be worth applying a similar spirit to spending time to really understand these issues as well.

Effective Animal Advocacy Resources

Super helpful, I'm about to cite this in the CE curriculum :)

Load More