Why not to rush to translate effective altruism into other languages

Over the last year, I’ve been doing research into the intersection between effective altruism in China (see our recent article on the topic). I participated in a retreat in Hong Kong on this topic, and oversaw the funding and mentoring of someone to work full-time on research and advising to help Western organisations learn about the country. I also speak basic conversational Chinese, have lived for almost a year in the country, and was heavily involved in our first efforts to promote effective altruism in the West.


All this has made me very cautious about trying to translate effective altruism materials into Chinese, or do broad based outreach in the country.


I think many of the arguments behind this position also apply to translating effective altruism materials into other languages. So, I think most efforts to do this translation work should probably be delayed. The more different the language, culture and politics, the longer we should wait.


Instead, I think that efforts to expand effective altruism into other languages should initially focus on person-to-person outreach to a small number of people with key expertise.


In the rest of this post, I outline four reasons why, and sketch an alternative approach.


1) Doing mass outreach in another language creates irreversible “lock in”


We made many mistakes doing outreach in the UK and USA, which have been difficult to unwind. For instance, even today many people think 80,000 Hours is primarily about earning to give, despite us saying many times we don’t think earning to give is typically the highest-impact option.


In this way, any kind of broad based outreach is risky because it’s hard to reverse. Once your message is out there, it tends to stick around for years, so if you get the message wrong, you’ve harmed years of future efforts. We call this the risk of “lock in”.


Lock in is partly caused because once there are websites and media articles about you, they stick around. But it’s also because first impressions are difficult to shift.

The existence of lock in means it’s worth delaying broad based outreach to increase the chance of getting the messages right first time.


China faces especially high risk of lock in, because you also face the risk of government censorship, but the consideration applies in any language.

2) It’s very difficult to translate effective altruism


There’s a huge amount we still don’t understand about how to explain effective altruism in English, and translating it into a different language creates a whole host of other problems.


Making translations with the right nuance is difficult. For instance, the direct Chinese translation of “effective altruism” that was initially used (有效利他主义) had the following problems:


1. Used a term for “altruism” that implied a great deal of self-sacrifice.

2. Sounds obviously foreign, which we expect would make it less appealing.

3. Sounds like a political ideology, which may not be viewed positively by the government.


Likewise, one of the possible translations of “existential risk” (生存危机) is very close to the the name of a computer game (生化危机), so doesn’t have the credibility one might want.


What’s more, each culture is different, and ideas often need to be presented in different ways depending on the context. For instance, Chinese writing makes a much greater use of historical analogies and ancient quotes than similar English writing. Rather than quote Peter Singer, we might consider quoting Mozi -- arguably the earliest consequentialist philosopher in history who in wrote around 400 BC of the importance of “universal concern” () towards all people.


Likewise, many materials in the West focus on the value of donating to charities in Africa, but internationally focused philanthropy is much rarer in China, and the Chinese government has prohibited foreign nonprofit organisations fundraising in the country, so this doesn’t seem like a promising route.


The ideas can also combine in unexpected ways with the local culture, leading to surprising results. For instance, effective altruism culture is fairly different in Oxford, the Bay Area and the German-speaking world.


It’s unlikely we’ll successfully navigate these kinds of problems unless we have people who are both excellent writers and marketers in the new language and have an in-depth understanding of effective altruism as it currently exists. But these kinds of people are often in short supply, and it would be better to wait until we find some.


This problem is greater the more culturally different the country of the new language is, so the greater the difference the longer we should wait to translate. My guess is that French is relatively safe (though likely still worth delaying), while we should wait longest to translate and promote materials in Chinese, Japanese, Russian and Arabic.

3) Existing English language materials are often out of date


Our views have changed significantly in just the last couple of years, so many existing materials don’t reflect our current views.


For instance, personally, I think Doing Good Better is too focused on donating to charity, contributing to the widely repeated assertion that effective altruism is “about how to donate to effective charities,” when actually it’s about all ways to do good, and where we think decisions about your career or policy change are often more important than decisions about where to donate.


Likewise, Doing Good Better is focused on global health and short-term issues, when today we think global catastrophic risks, emerging technologies and long-term issues are more urgent.


Translating effective altruism into a new language gives us a chance to correct outdated views and avoid lock in with those old views, and it’s a shame to waste that opportunity by directly translating old materials.


4) Mass outreach is not a good way to promote effective altruism in general


The ideas of effective altruism are unusually complex, and mass outreach tends to oversimplify them, leading to misunderstandings getting locked in. We still have a lot of work to do working out how best to frame the ideas in English, and this favours small-scale outreach where you can get rapid feedback on how well you’re being understood.


What’s more, it’s usually higher-impact to have a small number of highly involved people than hundreds of are merely aware of the ideas, or only participating in a superficial way. This is because impact is a product of how much effort someone invests and how effective their efforts are. A highly engaged person might exert 10-times the effort (e.g. donate 20% rather than 2%) and work 10-times more effectively, making for 100-times the impact.


More concretely, if you’re running a local group, then finding one other dedicated volunteer can mean the difference between the group surviving and failing, which is more important than having hundreds of people who attend events but don’t help out. Likewise, finding one person who speaks the language and really gets the ideas, or has knowledge relevant to a top problem area, is more useful than having thousands of people know about the ideas in a superficial way.


For both reasons, we think it’s usually better to focus on in-depth outreach to a small number of people, ideally through person-to-person discussions, rather than widely promoted short-form content or other mass marketing.


This can feel a bit counterintuitive. We find that people new to building the effective altruism community often think they should focus on getting media attention, inviting famous speakers to events, and “partnerships” with prestigious groups. But these approaches don’t normally yield as many results as building strong person-to-person relationships with genuinely enthusiastic people, and they also pose greater risks.


If written materials are used, then it’s better to focus on books, academic articles and podcasts aimed at a niche audience. Read more about the “fidelity model", and “how valuable is movement growth, which argues that we should focus on how attractive the ideas are before spreading them widely. The approach we advocate here also has similarities with Y Combinator'sdo things that don’t scale advice.

What should we do instead?


Rather than “translating” effective altruism into new languages, it seems better to think in terms of creating a local movement from scratch that’s inspired by the ideas of effective altruism, but is highly adapted to the local context. Imagine that Will MacAskill was Chinese: then what would he have written?


To do this well, we’ll need people who are both experts in the local culture and effective altruism in the West. We’ll also need people who are excellent writer and communicators in the new language.


Initial efforts to expand effective altruism into new languages should focus on making strong connections with a small number of people who have relevant expertise, via person-to-person outreach instead of mass media.


Instead of creating irreversible risks, this strategy has two benefits.


First, it has a significant short-term impact, because finding even a small number of really engaged community members is valuable.


Second, it puts us in a better position to do outreach in the future, because these initial connections will better understand the new language and culture, enabling better translations and outreach in the future.


One counterargument is that you face a chicken and egg problem. Without mass outreach, you could argue that it’s hard to find anyone interested in the ideas, or gain credibility. However, I don’t think this argument holds up.


The recent survey of the community, showed that the media has played a comparatively small role in getting people involved. Rather, the biggest single entry route was personal referral. Some of these entrances into the community might have been aided by the existence of media, but I suspect many could have happened without.


For this reason, I expect that it’s possible to start getting people involved in new countries with more person-to-person outreach. This would mean speaking to and arranging smalls events with (i) friends (ii) existing community members who speak the new language, and (iii) connections of connections from the previous two groups.


Beyond this, you could find more people by targeting lectures and discussion groups at especially promising groups. With these, it’s better to have a small event where the ideas are accurately expressed than one with a well-known speaker who doesn’t quite get what effective altruism is about.


Once you have an initial community of around 100 people, you could start creating written materials to support these efforts. But these materials would initially be tested on people you already know rather than promoted widely. If you need more credibility to support these efforts, rather than seek new press coverage, you could mention the achievements of effective altruism in other countries.


Only after significant testing would you release public written materials. Even then, it would be better to focus on in-depth materials like books and podcasts, rather than short-form media articles.


Translating effective altruism into new languages has the potential to have a tremendous impact, but could also spoil future efforts if done badly, slowing efforts by many years. If we start with in-depth, targeted, in-person outreach, we can make significant progress while reducing these risks.

Comments (43)

Comment author: Evan_Gaensbauer 05 March 2018 08:06:08AM 14 points [-]

Thanks for articulating arguments for this. There is a strong bias in favour of growth of various kinds in EA. There is an elementary growth strategy of naively pursuing growth as fast as possible. I also know several community members who are opposed to growing the movement much at all, as opposed to doing so carefully. However, hardly any effective altruists opposed to different kinds of movement growth lay out their arguments against them. This frustrates me as I'm genuinely curious to separate the good and bad arguments against rapid movement growth in EA, and that they're not publicly written out like this makes that difficult.

Giving arguments for how to do movement growth to allow for nuanced discussion, rather than if we should do much movement growth at all, is very helpful.

Comment author: DavidMoss 05 March 2018 10:04:02PM 4 points [-]

There is a strong bias in favour of growth of various kinds in EA.

This seemed more the case a couple of years ago. I think the pendulum has swung pretty hard in the other direction among EA thought leaders.

Comment author: Arepo 07 March 2018 06:03:21PM *  4 points [-]

Somewhat tangentially, am I unusual in finding the idea of 'thought leaders' for a movement about careful and conscientious consideration of ideas profoundly uncomfortable?

Comment author: DavidMoss 07 March 2018 06:27:23PM 2 points [-]

Definitely not. Often when I see the term used in EA it's being used negatively. To be fair though, the alternative terms I was considering using, "EA insiders", "EA elites", aren't too comfortable either.

Comment author: Evan_Gaensbauer 06 March 2018 04:47:43PM 2 points [-]

Yeah, I haven't been checking. What data gave you that impression?

Comment author: ThomasSittler 05 March 2018 09:48:54PM *  3 points [-]

The two articles cited in the OP are good:

Read more about the “fidelity model", and “how valuable is movement growth”, which argues that we should focus on how attractive the ideas are before spreading them widely.

Comment author: DavidMoss 06 March 2018 12:31:41AM 11 points [-]

Most of the arguments here seem to be more about not rushing into conducting mass (media) outreach in foreign cultures, rather than about not rushing to translate materials.

Regarding the arguments which do concern translation:

Argument #3 (existing English materials are out of date) concerns translation, but doesn't seem an argument against translation tout court, but only against translating Doing Good Better (if we suppose this is dangerously out-dated). From #3 we might instead conclude that we need to hurry to produce new English materials, since DGB is still influencing English-speakers, and then ensure these new materials are translated for foreign readers, lest they too be corrupted by the impression given by DGB. Alternatively we might think that we should generally hold off on any kind of popularisation, in order that we can wait to popularise a superior later version of EA (but then this is just another argument against mass outreach in general).

Argument #2 (translation is very difficult) seems strong when applied to China but, as you acknowledge, less compelling when applied to closer languages/cultures like French/German. This is an important qualification, since it seems plausible that most (or almost all) active non-English-speaking EA groups fall a lot closer to France/Germany than they do to China, and EAs may reasonably disagree about whether on balance producing translations in the contexts.

The end of your post proposes an alternative to mass media outreach, the alternative 'small group, close connections' model seems very compatible with using/needing translation work, of at least a few up to date texts.

You do state in that section that you favour EA-inspired new texts being created in those new contexts, rather than translations, it seems like fidelity considerations may push in the opposite direction. If you are concerned with the fidelity of EA ideas (so much so that you think Doing Good Better, translated, might be too off-message), then it might be better to ensure that some translated EA materials are accessible to new groups rather than having people around the world try to create EA-inspired approaches.

Comment author: Ben_Todd 07 March 2018 10:36:12PM 2 points [-]

Hey David,

I'm not able to fully respond, but a few quick comments which might help to clarify:

I agree I'm combining mass media and translation somewhat. In principle, someone could propose spending translating lots of materials but not sharing them widely, but in practice people don't. Rather, the options are more like (i) translate existing materials into the new language and share widely, or (ii) speak to people in small groups.

I'm arguing in favour of (ii) initially. But then like I say near the end of the post, you'd then start to work on translating materials on the side. The aim is that we'll end up with much better translations if we do lots of in-person outreach first, work on the translations iteratively, and first build up a base of people who really understand both EA and the local language and culture.

(Whereas instead people often move directly to ideas like translating DGB and releasing, or creating an EA website in the local country with lots of content on it).

I think "lock in" might also be pretty significant. DGB creates less problems in English since we've already been locked into those misconceptions about EA. But in countries where there are no existing materials, we want to avoid those mistakes and get a fresh start.

(I also think we should prioritise having a new DGB or alternative intro resource in English, but it will take some time - this might be the best we have right now: https://www.effectivealtruism.org/resources/)

Comment author: DavidMoss 08 March 2018 06:34:33PM 4 points [-]

Thanks for the reply Ben.

In principle, someone could propose spending translating lots of materials but not sharing them widely, but in practice people don't. Rather, the options are more like (i) translate existing materials into the new language and share widely, or (ii) speak to people in small groups.

We may be speaking to different people. I know a number of small group leaders who want materials in their native language to share with their (small) groups, rather than to broadcast a translation on a mass scale.

I suspect that if the alternative is groups struggling through English-language materials or not using written EA materials in their groups at all (and just going it alone), then it's probably better that they have some basic translated materials (one can always advise them not to try to broadcast them widely). I agree groups can develop without mass media outreach, but I'm not sure how well they can develop (e.g. up to the 100 members you suggest in the OP) without EA materials in their language.


I think the question of 'lock-in' is trickier, though am not sure how far it applies to small groups rather than the mass media outreach. I agree that in the English speaking world memes like 'EA = effective charity' and 'EA careers = ETG' are prevalent and perhaps impossible to reverse en mass. But how 'locked in' are individual EAs and small EA groups who have, for example, read Doing Good Better? i.e. how intractable is changing their view from the Doing Good Better view to the updated view (and how far would it have been better to delay publication of Doing Good Better a few years until the ideas were more developed)? If we're dealing with translations for small groups, the situation looks more similar to small groups of EAs who have read Doing Good Better, than to mass-media broadcasting of EAish memes to the general populace. Since it's not clear how far small groups of English speaking EAs are locked in or that it would have been better to delay EA messaging in English speaking areas a few years, it's not clear to me that we should be trying to avoid/delay 'lock in' wherever else we can e.g. in the non-English speaking world.

I think this is especially so if we think less in terms of the possibility of avoiding irreversible lock-in and more in terms of a trade-off between incrementally improving EA messaging and delaying EA messaging several years (assuming that we don't think that the next iteration of EA ideas will be final, but will themselves need to be updated a handful of years later).

Comment author: Evan_Gaensbauer 06 March 2018 04:56:55PM 2 points [-]

I've tried to initiate translation projects for EA into non-English languages in the past. I was looking for EAs who were (close to) fluent in a language and local to where outreach would take place. This was a couple years ago. So, the local EA communities outside the English-speaking world were new, small and didn't have enough people to start up their own translation project. Given the arguments in Ben's post, I don't think necessarily much was lost in not having capitalized on the opportunity to translate EA content into other languages as well.

The most successful case of translation of EA content, and moreover, the generation of brand new EA content, outside of English is in Germany. This was started by EAs who were native speakers of German, and the work of their EA Foundation (EAF). Depending on how much one thinks their circumstances could generalize, it might be best for the movement to work with local groups which successfully develop over a few years to generate new content in other languages. This content could be specialized in its messaging to the culture.

Comment author: vollmer 10 March 2018 03:51:02PM *  4 points [-]

Based on EAF's experience in Germany and Switzerland, I strongly agree with Ben's main points in the post. In the early days we made several mistakes that could have been prevented fairly easily. In particular, it seems hard to correct the perception that EA is not just about donating (to GiveWell top charities). It also remains very difficult to counter the impression that EA is mainly the practical implementation of Singer's views; e.g. Singer's views on infanticide get quoted in many media articles about EA.

Some of the challenges that might have led to this:

  • DGB and Singer's EA book were translated to German, but much of the more advanced content is only available in English.
  • Quickly translating English content is easy. However, it takes much more time to ensure high quality both in terms of language and framings/nuance, and it's even more challenging to keep these translations up to date. See the "fidelity model" blog post referenced above for more discussion of this.
  • The media frequently interview members of the community. Community members are more or less up to date with recent EA publications and would explain EA well, but the media very proactively ask about charitable donations and related issues. It takes a lot of active effort and experience with media interviews to counter this pigeonholing, which is hard to do without much practice. I personally find it pretty hard to give good guidance on this.

So as a conclusion, I think the expansion to Germany, Switzerland, and Austria could have gone much better still, and while I agree it could be the deemed most successful case of translation of EA content, I think it was worse than what we should be aiming for.

Comment author: Evan_Gaensbauer 10 March 2018 06:42:03PM 1 point [-]

That's really great feedback. Thanks.

Comment author: weeatquince  (EA Profile) 05 March 2018 07:32:32PM *  8 points [-]

Good article Ben!


I think similar risks arise with translating effective altruism to new domains or new audiances with particular expertise.

I've felt this when interacting with people looking to apply effective altruism ideas in policy. Such exercises should be approached with caution: you cannot just tell policy makers to use evidence (they've already heard about evidence) or to put all their resources to whatever looks most effective (wouldn't work) etc.

Similarly I suspect there is something to the fact that I find EA materials have had limited acceptance among experts in international development.


I would go a step further and say that the aim should not solely be one of translating EA ideas but also of improving EA ideas. Currently EA is fairly un-diverse in terms of cultures, plurality of ethical views, academic background, etc. I think we can learn a lot from those we are trying to reach out to.


(Minor aside I think mass outreach efforts done well have been are still are valuable and this article underplays that)

Comment author: Evan_Gaensbauer 06 March 2018 05:19:59PM 3 points [-]

It's my impression most policy efforts coming out of EA in most countries are from experienced, professional organizations which work with or hire policy experts. The Centre for Effective Altruism (CEA) has worked with university institutes at Cambridge and Oxford to produce policy reports of global catastrophic risks for European governments. The Effective Altruism Foundation (EAF) has in Germany and Switzerland done policy advocacy, initiated by philosophy post-docs and the like. Before involvement in EA, they weren't particularly experienced in philosophy, but their efforts haven't backfired in any sense. I haven't tracked what portion of their campaigns succeeded at the ballot box, but being able to start things like referendums on animal rights/welfare without opposition and backlash from the public could be considered successes in themselves.

There isn't centralization across the EA community worldwide for work in the policy sector, so technically a group some country could start doing policy work in the name of EA without any kind of external assessment. So a culture of pursuing policy work much more cautiously can definitely still be worth promoting within EA. I notice the examples I gave were about causes like animal advocacy and global catastrophic risks, compared to your example of international development. My examples are of sectors which aren't already as common in academia and policy. So the EA community has been able to effectively break a lot of new ground in policy research and advocacy regarding these causes. Fields like international development and others with a history of more extensive institutional support are more complicated. They require more specialization and expertise to do effective work upon.

Comment author: weeatquince  (EA Profile) 09 March 2018 03:24:45PM 2 points [-]

My point was not trying to pick up policy interventions specifically. I think more broadly there is too often an attitude of arrogance among EAs who think that because they can do cause prioritisation better than their peers they can also solve difficult problems better than experts in those fields. (I know I have been guilty of this at points).


In policy, I agree with you that EA policy projects fall across a large spectrum from highly professional to poorly thought-out.

That said I think that even at the better end of the spectrum there is a lack of professional lobbyists being employed by EA organisations and more of a do-it-ourselves attitude. EA orgs often prefer to hire enthusiastic EAs rather than expensive experts (which maybe a totally legitimate approach, I have no strong view on the matter).

Comment author: Evan_Gaensbauer 10 March 2018 12:01:44AM 1 point [-]

What fields come to mind specifically as ones EAs arrogantly think they can solve problems better than experts can? EA is diverse, so we could be talking about a lot of different things. Re: policy, I think a large part of it is EA is still too small a movement to afford expensive experts or otherwise have a high degree of influence. I consider policy efforts in EA successes as a young movement it's been able to get tabled initiatives in a few countries. These don't often result in much happening, so these are relatively modest outcomes. Certainly there is greater opportunity in the future for EA to influence policy, hopefully by working more with experts in the future.

Comment author: ThomasSittler 05 March 2018 11:15:34AM *  7 points [-]

I agree, and this is a mistake I've made in the past in a big way, especially around 2015 when I was enthusiastically promoting mass outreach with low-fidelity translations.

Here's an additional data point, which is more about the mechanics of translation itself rather than the broader strategy of delaying outreach, but does tend to support the views defended in this post: I've found that writing articles in French about EA from scratch results in much better text than translating from English. The translated stuff always feels a bit alien and uncool. (And this is for what I believe were high-quality translations, with less-than-stellar translations, the results could be outright disastrous.)

Comment author: RyanCarey 05 March 2018 12:08:49PM 6 points [-]

Is it still alien and uncool if you look at an article as a whole and just rewrite it from scratch in French, rather than translating each line? (Kind-of like if I lose my copy of an essay and then rewrite the same ideas in new prose.)

Comment author: ThomasSittler 05 March 2018 09:27:40PM *  2 points [-]

It's a spectrum, but the further away from translation and the more towards "presenting standalone idea" the better in my experience. So taking the high-level ideas and broad structure of an article can be fine. However, even without translating, the temptation can be strong to excessively lean on the original text, it has to be actively resisted.

Two examples: completely written from scratch, and loosely inspired by an existing article.

Comment author: Evan_Gaensbauer 06 March 2018 05:23:36PM 1 point [-]

The EA Handbook you put together included essays originally published as blog posts by individual EA community members in English. There's no reason there couldn't be an EA Handbook for original content generated in and for other languages. There's probably enough content in German to produce a "German EA Handbook". I'm not confident there is enough original EA content in other languages such that a corresponding EA handbook could be produced.

Comment author: ThomasSittler 05 March 2018 09:41:04PM *  1 point [-]

Can whoever downvoted this say why so I can improve? Anonymously if you want.

Comment author: Evan_Gaensbauer 06 March 2018 05:29:48PM 2 points [-]

If you don't give feedback, here's a tip for how I figure why I might have been downvoted when it's not at first clear to me why. I read my comment again, and imagine from a perspective of someone in EA very different from me. Other people's perceptions and intuitions of how our words come across on the internet can be very different from our own. I often notice things in my own comments after the fact as a result. The part of your comment which stood out to me with the typical tone on the EA Forum was referring to some earnest translation effort as "alien and uncool". For various reasons, I can imagine someone seeing arguments from coolness are bad or inappropriate for the EA Forum, although I assume ostensibly you weren't that serious about the comment.

Comment author: Jan_Kulveit 13 March 2018 08:15:29AM *  5 points [-]

I just published a short history of creating effective altruism movement in the Czech Republic http://effective-altruism.com/ea/1ls/introducing_czech_association_for_effective/ and I think it is highly relevant to this discussion

Compared to Ben's conclusions I would use it as a data-point showing

  • it can be done

  • it may not be worth delaying

  • there are intermediate forms of communication in between "mass outreach" and "person-to-person outreach"

  • you should consider more complex model of communication than just (personal vs. mass media): specifically, a viable model in new country could be something like "very short message in mass media, few articles translated in national language to lower the bareer and point in the right direction, much larger amount transmitted via conferences & similar

Putting too much weight into "person to person" interaction runs into the problem you are less likely to find the right persons (consider how such connections may be created)

Btw it seems to me the way e.g. 80k hours and CEA works are inadeqate in creating the required personal connections in new countries, so it's questionable if it makes sense to focus on it

(I completely agree China is extremely difficult, but I don't think China should be considered a typical example - considering mentality it's possibly one of the most remote countries from from Eurpoean POV)

Comment author: Ben_Todd 14 March 2018 09:07:09PM 1 point [-]

Hi Jan,

It's a useful case study, however, two quick responses:

1) To some extent you were following the suggested approach, because you only pushed ahead having already built a core of native speakers who had been involved in the past with English language materials (e.g. Ales Flidr was head of EA Harvard; core of LW people helped to start it).

You also mention how doing things like meeting CFAR and attending EAGxOxford were very useful in building the group. This suggests to me that doing even more to build expertise and connections with the core English-speaking EA community before pushing ahead with Czech outreach might have led to even better results.

I also guess that most of the group can read English language materials? If so, that makes the situation much easier. As I say, the less the distance, the weaker the arguments for waiting.

2) You don't directly address my main concern. I'm suggesting that if we try to spread EA in new languages and cultures without laying out groundwork could lead to a suboptimal version of EA being locked into the new audience. However, in your report, you don't directly respond to this concern.

You do give some evidence of short-term impact, which is evidence that benefits outweighed the opportunity costs. But I'd also want to look at questions like: (i) how accurately was EA portrayed in your press coverage? (ii) how well do people getting involved in the group understand EA? (iii) might you have put off important groups in ways that could have been avoided?

Comment author: Jan_Kulveit 16 March 2018 12:39:43AM *  2 points [-]

Hi Ben,

1) I understand your concerns.

On the other hand I'm not sure if you take into account the difficuties

  • e.g. going to EAG could require something like "heroic effort". If my academic job was my only source of income, going to EAGxOxford would have meant spending more than my monthly disposable income on it (even if EAGxOxford organizers were great in offering me big reduction in fees)

  • I'm not sure if you are modelling correctly barriers in building connections to the core of the community(*). Here is my guess at some problems people from different countries ol cultures may run into when trying to make connections, e.g. by applying for internships, help, conferences, etc.

1) People unconsciously take language sophistication as a proxy for inteligence. By not being proficient in English you loose points.

2) People are evaluated on proxies like "attending prestigeous university". Universities outside of US or UK are generally discounted as regional

3) People are often in fact evaluated based on social network distance; this leads to "rich gets richer" dynamics

4) People are evaluated based on previous "EA achievements" which are easier to achieve in places where EA is allready present.

(*) you may object e.g. Ales Flidr is a good counterexample, but Harvard alumni are typically relatively "delocalized", in demand everywhere, and may percieve greater value in working in the core than spreading ideas from core to emerging local groups. (Prestige and other incentives also point that way)

one risk I see in your article is it may influence the people who would be best at mitigating risks of "wrong local EA movements" being founded to not work on it at all.

I dont think "the barriers" should be zero, as such barriers in a way help the selection of motivated people. Just in my impression they may be higher than they appear from inside. Asking people to first build conections, while building such connections is not systematically supported, may make the barriers higher than optimal.

Btw yes the core of the group can read English materials. Also it could do research in machine learning, found a startup, work in quantitive finance, in part get elected to the parliament, move out of the country, and more. What I want to point at, if you imagine members of a group of people working on "translation" of EA into new culture you would like, they are likely paying huge opportunity costs in doing so. It may be hard to keep them at some state of waiting and building connections.

In our case, counterfactually it seems plausible "waiting even more" could have also led to the group not working, or worse organization beeing created, as the core people would loose motivation / pursue other opportunities.

2) In counting long-term impact and the lock-in effect you should consider the chance movements in new languages and cultures develop in some respects better versions of effective altruism, and beneficial mutations can than spread, even back to the core. More countries may mean more experiments running and faster evolution of better versions of EA. It's unclear to me how these terms (lock in, more optimization power) add up but both should be counted. One possible resolution may be to prioritize founding movements in smaller countries where you create experience, but the damage is limited.

To your questions

i] with one exception, reasonably well taken from the viewpoint of strategic communication (communicating easily communicable concepts, eg impact and effectivity). I don't think the damage is great, and part of misconceptions is unavoidable givent the name "effective altruism".

ii] it has a distribution... IMO the understanding at the moment is highly correlated with the enagagement in the group. Which may be more important than criteria like "average understanding" or "public understanding"

iii] yes, it's complicated, depends on some misalignments, and I dont want to disscuss it publicly.

Comment author: Ben_Todd 17 March 2018 11:11:54PM 0 points [-]

Quick addition that I realise the lack of support for local groups is not ideal, but this capacity constraint is another reason to go slow. I'd favour a more "all or nothing" approach, where we select a small number of countries / languages / locations and then make a major attempt to get them going (e.g. ideally supplying enough money so that 1-2 people can go full-time, pay for trips to visit other groups etc., plus provide in-depth mentoring from CEA), while in other locations we minimise outward facing activities. The middle ground of lots of small groups with few resources doesn't seem ideal. I'm optimistic we're moving in this direction with things like the EA Grants and http://effective-altruism.com/ea/1l3/announcing_effective_altruism_community_building/

Comment author: Risto_Uuk 11 March 2018 12:37:50PM 3 points [-]

Thank you for the post!

I agree that from the point of view of translation Doing Good Better might be too focused on donating to charity and on global health, but this doesn't seem to be an issue at all when it comes small in-depth discussion groups. I guess this is another argument in favor of focusing on these types of activities rather than large-scale outreach.

Comment author: DavidMoss 11 March 2018 03:01:11PM 2 points [-]

I'm curious how much mass outreach there actually is in EA and/or what people have in mind when they talk about mass outreach.

Aside from Doing Good Better and Will/CEA's other public intellectual work, which they seem to be retreating from, it's not clear to me what mass outreach has actually been done.

Comment author: Risto_Uuk 11 March 2018 03:45:43PM 2 points [-]

I think this depends on how we define mass outreach. I would consider a lot of activities organized in EA community to be mass outreach. For example, EAG, books, articles in popular media outlets, FB posts in EA group, 80 000 Hours podcast, etc. They are mass outreach because they reach a lot of people and very often don't enable an in-depth work on. Exceptions would be career coaching session at EAG event and discussing books/articles in discussion groups.

Comment author: DavidMoss 11 March 2018 04:45:09PM 2 points [-]

I agree re. books and articles in the mass media- and these are the kinds of things it seems people are stepping back from.

I think of the EA FB group, EAG etc. to be more insider-facing than outreach (EAG used to seem to be more about general outreach, but not anymore).

The 80K podcast is an interesting middle case, since it's clearly broadcast generally, but I imagine the actual audience is pretty niche and it's a lot more in depth than any media article or even Doing Good Better (in my view). I have to wonder how far, in a couple of years, people will be saying of the podcast the same things being said of DGB, i.e. the content is sub-optimal or out of date and so we wouldn't want to spread it around. The same considerations seem, in principle, to apply in both cases, since even if people within the English-speaking world are already locked into some bad ideas, we don't want to continue locking them into new ideas, which we will judge in 2020 to have been premature/mistaken/sub-optimal.

Comment author: letianw 11 March 2018 11:28:20PM 4 points [-]

Hi Ben,

As a Chinese national currently living in the west, I think I broadly agree with your argument that "efforts to expand effective altruism into other languages should initially focus on person-to-person outreach to a small number of people with key expertise." I also appreciate your grasp of the complexity of cultural and linguistic barriers in promoting EA ideas in the Chinese context, which can often be lost on EAs who are less familiar with other cultures.

One potential objection to this is that not rushing into massive translation effort does not equal to not at least attempt some translation at all. A set of core materials can still be useful, if it is carefully curated by professional translators (not merely bilingual volunteers like me). Without written material, it can be difficult to make ideas stick, even among a small group of personal contacts. A counter argument to this, however, is that the initial promising groups are very likely elite college students and urban professionals who would have no problem reading English materials. I don't have a strong opinion on this.

Another potential problem I can foresee regarding 'personal contact' approach is that to my knowledge, the Chinese government keeps close tabs on any recruitment activities by foreign social movements. Anecdotes from missionary friends 10 years ago suggest that their religious activities, especially when involving locals, were closely monitored by the police, kept under 20 people, and sometimes harassed. I cannot speak with any confidence that this is still the case, or if it will be applied to EAs equally. But this is something to keep in mind when evaluating personal outreach versus media effort.

Comment author: Ben_Todd 07 March 2018 10:39:30PM 3 points [-]

Some concerning data in this recent post about local groups: http://effective-altruism.com/ea/1l7/2017_lean_impact_assessment_evaluation_strategic/

One other striking feature of this category is that all of the top groups [in terms of new event attendees] were from non-Anglo-American countries. While this is purely speculative, an explanation for this pattern might be that these groups are aggressively reaching out to people unfamiliar with EA in their areas, getting them to attend events, but largely not seeing success in transferring this into increased group membership.

Comment author: DavidMoss 08 March 2018 07:33:42PM *  3 points [-]

Thanks for the citation!

We agree this doesn't look good for (non-Anglo-American) groups running large events as outreach, in that it looks like it doesn't bear fruit in terms of increasing members or other outputs. But note the rest of the paragraph you quote, where we say:

it seems plausible that EA groups outside of the traditional geographical areas may face distinct challenges and require more tailored support (such as translation of materials).

One possible explanation for the observation above is that these groups' large events in non-Anglo-American countries don't bear so much fruit because they lack the supporting background and infrastructure (like translated materials). So, for example, if someone attends a large event in London they can easily immediately check out lots of EA websites and materials, find places to follow-up and so on; if someone attends a large event in a different country without these touch-points and critical mass, then not so much. Of course, it may also just be that these areas were less fertile ground for receiving EA message in some other way.

It's also important to note that it's not clear from the data we provided above that non-Anglo-American groups distinctively receive low payoffs from large events. If you look at the specific graph you'll see that these groups are pretty clear outliers, reporting events with many more people who are unfamiliar with effective altruism, but it's not that Anglo-American groups are running large events with lots of people unfamiliar with EA and receiving comparatively larger payoffs: rather it's that most other groups are just not running such large events with so many people unfamiliar with effective altruism. So what is distinctive about these groups is that they are running large events with lots of unfamiliar attendees at all, not that they are running these large events and failing to receive payoff.

Comment author: Arepo 08 March 2018 12:10:13AM *  3 points [-]

Seconding Evan - it's great to have this laid out as a clear argument.

Re this:

In this way, any kind of broad based outreach is risky because it’s hard to reverse. Once your message is out there, it tends to stick around for years, so if you get the message wrong, you’ve harmed years of future efforts. We call this the risk of “lock in”.

I think there are some ways that this could still pan out as net positive, in reverse order of importance:

1) It relies on the arguments against E2G as a premium EA cause, which I'm still sceptical of given numerous very large funding gaps in EA causes and orgs. Admittedly In the case of China (and other semideveloped countries) the case against E2G seems stronger though, since the potential earnings are substantially lower, and the potential for direct work might be as strong or higher.

2) Depending on how you discount over time, and (relatedly) how seriously you take the haste consideration, getting a bunch of people involved sooner might be worth slower takeup later.

3) You mentioned somewhere in the discussion that you've rarely known anyone to be more amenable to EA because they'd encountered the ideas, but this seems like underestimating the nudge effects on which 99% of marketing are based. Almost no-one ever consciously thinks 'given that advert, I'm going to buy that product' - but when you see the product on the shelf, it just feels marginally more trustworthy because you already 'know' it. It seems like mass media EA outreach could function similarly. If so, lock-in might be a price worth paying.

This isn't to say that I think your argument is wrong, just that I don't yet think it's clear-cut.

It also seems like the risks/reward ratio might vary substantially from country to country, so it's perhaps worth thinking about at least each major economy separately?

To the degree that the argument does vary from country to country, I wonder whether there's any mileage in running some experiments with outreach in less economically significant countries, esp when they have historically similar cultures? Eg perhaps for China, it would be worth trialling a comparatively short termist strategy in Taiwan.

Comment author: SiebeRozendal 05 March 2018 04:41:10PM 3 points [-]

Regarding Doing Good Better, is there any follow-up in the pipeline that is more up-to-date?

I find the book a great introduction into EA, but I have had multiple instances where I needed to point out to new members who'd just read the book that for some points "that's not actually what's thought anymore".

Comment author: Ben_Todd 05 March 2018 10:53:04PM 5 points [-]

Hopefully, but it may take a few years.

In the meantime, this is good: https://www.effectivealtruism.org/resources/

And we're working on an advanced series of articles at 80k.

Comment author: avacyn 10 March 2018 06:28:07PM 2 points [-]

I found this article useful and convincing. Thanks for writing, Ben!

However, I was surprised to see that this has become one of the most upvoted posts of all time on the EA forum. I would expect an insightful and convincing post like this to get between 20 and 30 upvotes. I'm worried that I'm missing a more important takeaway. Can someone explain why this has been so positively received?

Comment author: adamaero  (EA Profile) 06 March 2018 04:14:23PM *  1 point [-]

Side note - Have you looked at the Wikipedia pages for Effective Altruism in different languages and translated to [English]?

Examples, sv.wikipedia.org/wiki/Effektiv_altruism ~ marginal impact, opartic thinking, contraceptive thinking. Es.wikipedia.org ~ comparative wealth, etc.

Just something someone here may find interesting.

Comment author: Tuukka_Sarvi 06 March 2018 03:08:10PM *  1 point [-]

Great points!

I like the book Doing Good Better a lot: it is the single most important source that introduced me to EA and convinced me that it is a remarkable and very exciting idea and movement.

Does having "doing good better" available at stores do more good than bad? -- In my estimation it does more good. It is one of the vehicles of spreading the word about EA.

How much less viable vehicle of spreading word a translation of the book would be given that the translation is distributed in a similar cultural environment as the original book (eg. parts of Europe - Sweden, Norway, Finland, France etc) ? -- If additionally given that translation is done professionally by or commissioned by EA-minded locals, in my estimation he difference to the effect of the original language book would be small. Thus, this option is worth considering in my opinion.

I think a translation of short career guide (a book) summarizing the key content in 80kh could be even more interesting to people. 80kh does a superb job of examining at careers from the perspective of having as much impact as possible. I am not aware of any other career guide that take this perspective rigorously.

Comment author: Ben_Todd 07 March 2018 10:37:57PM 1 point [-]

I agree it's a great book, and it has net positive value. My concern is that we might be able to get something even better.

Also I don't think professional translators are in a good position to do this work, since they don't understand the nuance and aims of EA well enough to decide what the terms should be in the local language. We already find it really hard to pick the right terms in English. You'd need someone who really understands EA marketing and a professional translator working together.

Comment author: michaelchen 21 March 2018 11:29:20PM 0 points [-]

Are there new terms for EA and x-risk in Chinese besides 有效利他主义 and 生存危机, by the way?

Comment author: Ben_Todd 07 March 2018 10:40:45PM 0 points [-]

Another consideration I'm not sure of is that a mainly English speaking community will be easier to coordinate than one of the same size split across many languages and cultures, so this might be reason to focus initially on one language (to the extent that efforts across different languages funge with each other).

Comment author: Arepo 07 March 2018 11:34:39PM *  2 points [-]

This seems no more (to me, less) of a concern than that having a diversity of languages and cultures would help avoid it becoming tribalised.

Also, re the idea of coordination, cf my comment above about 'thought leaders'. I know it's something Will's been pushing for, but I'm concerned about the overconcentration of influence in eg EA funds (although that's a slightly different issue from an overemphasis on the ideas of certain people)