Introducing CEA's Guiding Principles

By William_MacAskill @ 2017-03-08T01:57 (+45)

Over the last year, I’ve given a lot of thought to the question of how the effective altruism community can stay true to its best elements and avoid problems that often bring movements down. Failure is the default outcome for a social movement, and so we should be proactive in investing time and attention to help the community as a whole flourish.

In a previous post, I noted that there’s very little in the way of self-governing infrastructure for the community. There’s very little to deal with people representing EA in ways that seem to be harmful; this means that the only response is community action, which is slow, unpleasant for all involved, and risks unfairness through lack of good process. In that post, I suggested we create two things: (i) a set of guiding principles agreed upon by all of EA; (ii) a community panel that could make recommendations to the community regarding violations of those principles.

There was healthy discussion of this idea, both on the forum and in feedback that we sought from people in the community. Some particularly important worries, it seemed to me, were: (i) the risk of consolidating too much influence over EA in any one organisation or panel; (ii) the risk of it being impossible to get agreement, leading to an increase in politicisation and squabbling; (iii) the risk of losing flexibility by enforcing what is an “EA view” or not (in a way that other broad movements don’t do*). I think these were important concerns. In response, we toned back the ambitions of the proposed ideas.

Instead of trying to create a document that we claim to represent all of EA, enforced by a community panel as I suggested, we’ve done two things:

 

(i)  Written down CEA’s understanding of EA (based in part on discussion with other community members), and invited other organisations to share and uphold that understanding if they found it matched their views. This will become a community-wide vision only to the extent that it resonates with the community.

(ii) Created a small advisory panel of community members that will provide input on important and potentially controversial community-relevant decisions that CEA might have to make (such as when we changed the Giving What We Can pledge to be cause-neutral). The initial panel members will be Alexander Gordon-Brown, Peter Hurford, Claire Zabel, and Julia Wise.

 

The panel, in particular, is quite different from my original proposal. In the original proposal, it was a way of EA self-regulating as a community. In this new form, it’s a way of ensuring that some of CEA’s decisions get appropriate input from the community. Julia Wise, who serves as community liaison at CEA, has put together the advisory panel and has written about this panel here. The rest of this post is about how CEA understands EA and what guiding principles it finds appropriate.

How CEA understands EA is given in its Guiding Principles document. I’ve also copied and pasted the contents of this document below. 

Even if few organisations or people were to endorse this understanding of EA, it would still have a useful role. It would: 

However, we hope that the definition and values are broad enough that the large majority of the EA community will be on board with them. And indeed, a number of EA organisations (or leaders of EA organisations) have already endorsed this understanding (see the bottom of this post). If this understanding of EA were widely adopted, I think there could be a number of benefits. It could help newcomers, including academics and journalists, to get a sense of what EA is about. It could help avoid dilution of EA (such that donating $5/month to a charity with low overheads becomes ‘effective altruism’) or corruption of the idea EA (such as EA = earning to give to donate to RCT-backed charities, and nothing else). It might help create community cohesion by stating, in broad terms, what brings us all together (even if many of us focus on very different areas). And it might give us a shared language for discussing problematic events happening in the community. In general, I think if we all upheld these values, we’d create a very powerful force for good.

There is still a risk of having a widely-agreed-upon set of values, which is that effective altruism could ossify or become unduly narrow. However, I hope that the openness of the definition and values (and lack of enforcement mechanism beyond community norms) should minimise that risk.

 

Here is the text of the document:

The Centre for Effective Altruism’s understanding of effective altruism and its guiding principles


What is effective altruism?

Effective altruism is about using evidence and reason to figure out how to benefit others as much as possible, and taking action on that basis.

What is the effective altruism community?

The effective altruism community is a global community of people who care deeply about the world, make benefiting others a significant part of their lives, and use evidence and reason to figure out how best to do so. 

Putting effective altruism into practice means acting in accordance with its core principles:

The guiding principles of effective altruism:


Commitment to Others: 

We take the well-being of others very seriously, and are willing to take significant personal action in order to benefit others. What this entails can vary from person to person, and it's ultimately up to individuals to figure out what significant personal action looks like for them. In each case, however, the most essential commitment of effective altruism is to actively try to make the world a better place.


Scientific Mindset:

We strive to base our actions on the best available evidence and reasoning about how the world works. We recognise how difficult it is to know how to do the most good, and therefore try to avoid overconfidence, to seek out informed critiques of our own views, to be open to unusual ideas, and to take alternative points of view seriously.


Openness: 

We are a community united by our commitment to these principles, not to a specific cause. Our goal is to do as much good as we can, and we evaluate ways to do that without committing ourselves at the outset to any particular cause. We are open to focusing our efforts on any group of beneficiaries, and to using any reasonable methods to help them. If good arguments or evidence show that our current plans are not the best way of helping, we will change our beliefs and actions.


Integrity: 

Because we believe that trust, cooperation, and accurate information are essential to doing good, we strive to be honest and trustworthy. More broadly, we strive to follow those rules of good conduct that allow communities (and the people within them) to thrive. We also value the reputation of effective altruism, and recognize that our actions reflect on it.


Collaborative Spirit: 

We affirm a commitment to building a friendly, open, and welcoming environment in which many different approaches can flourish, and in which a wide range of perspectives can be evaluated on their merits. In order to encourage cooperation and collaboration between people with widely varying circumstances and ways of thinking, we resolve to treat people of different worldviews, values, backgrounds, and identities kindly and respectfully.

The following organizations wish to voice their support for these definitions and guiding principles:

Additionally, some individuals voice their support: 

This doesn’t represent an exhaustive list of all organisations or people involved with effective altruism. We want to invite any other organisations to endorse the above guiding principles if they wish by writing us at hello@centreforeffectivealtruism.org.

Julia and I want to thank all the many people who helped develop this document, with particular thanks to Rob Bensinger, Jeff Alstott, and Hilary Mayhew who went above and beyond in providing comments and suggested wording.

 


null @ 2017-03-08T01:59 (+12)

In my previous post I wrote: “The existence of this would bring us into alignment with other societies, which usually have some document that describes the principles that the society stands for, and has some mechanism for ensuring that those who choose to represent themselves as part of that society abides by those principles.” I now think that’s an incorrect statement. EA, currently, is all of the following: an idea/movement, a community, and a small group of organisations. On the ‘movement’ understanding of EA, analogies of EA don’t have a community panel similar to what I suggested, and only some have ‘guiding principles’. (Though communities and organisations, or groups of organisations, often do.)

Julia created a list of potential analogies here:

[https://docs.google.com/document/d/1aXQp_9pGauMK9rKES9W3Uk3soW6c1oSx68bhDmY73p4/edit?usp=sharing].

The closest analogy to what we want to do is given by the open source community: many but not all of the organisations within the open source community created their own codes of conduct, many of them very similar to each other.

RayTaylor @ 2020-03-10T10:10 (+2)

William I wonder if EA is also, whether we accept this or not, a part of a wider/older historical effort?

I don't mean just people like Esther Duflo at the MIT Poverty Lab, health economists, bottom billion / development economists, Joseph Rowntree Foundation and its social wellbeing research, Oxfam, IDS Sussex, public health people, epidemiologists and (obviously) utilitarian philosophers ....

... but also older roots such as Quakers like prison reformer Elisabeth Fry and anti-slave trade groups, or various buddhists and Christians prioritising health care, the relief of poverty etc.

Of course many of these will have been less mathematical than many modern EAs, and we could identify other differences. However all of these were to significant degrees interested in evidenced improvements and policy improvement, and some still are.

By acknowledging and exchanging with their existing knowledge bases and experiences, wouldn't we be better placed to expand and mainstream the best that EA can offer? And be less ignorant about effective and altruistic work and research that has already been done, and lessons already learned, perhaps especially when it comes to creating and maintaining a movement?!

null @ 2017-03-08T16:19 (+5)

This is pretty much exactly what I was hoping for! Thank you!

null @ 2017-03-09T11:16 (+4)

Congrats on making this, it seems like a modest but still powerful way forward.

Have you thought about making it possible for community members to officially endorsement these principles?

null @ 2017-03-09T14:36 (+3)

We did think about it, but didn't come up with anything that seemed particularly good. (Happy to hear ideas.)

I'm thinking of other parallel documents: for example, when the Universal Declaration on Human Rights came out, there was a list of official signatories (nations). Then there were organizations that promoted the declaration on their own (without participation of the UN). Then there were individuals who supported it by telling their friends they approved of it, writing letters to the editor in support of it, etc.

The main way I would love to see individuals endorsing the principles is to refer to them when disagreements arise and it's unclear how people should behave or make decisions. That's where the rubber meets the road, much more than whose name is on what list.

null @ 2017-03-11T12:27 (+3)

Will this be publicly available on the internet? eg on https://www.effectivealtruism.org/?

null @ 2017-03-11T22:40 (+4)

Yes! Working on it.

Greg_Colbourn @ 2019-03-12T14:49 (+1)

https://www.centreforeffectivealtruism.org/ceas-guiding-principles/

null @ 2017-03-08T13:44 (+3)

I'd like to respond to your description of what some people's worries about your previous proposal were, and highlight how some of those worries could be addressed, hopefully without reducing how helpfully ambitious your initial proposal was. Here goes:

the risk of losing flexibility by enforcing what is an “EA view” or not

It seems to me like the primary goal of the panel in the original proposal was to address instances of people lowering the standard of trustworthiness within EA and imposing unreasonable costs (including unreasonable time costs) on individual EAs. I suspect that enumerating what sorts of things "count" as EA endeavors isn't a strictly necessary prerequisite for forming such a panel.

I can see why some people held this concern, partly because "defining what does and doesn't count as an EA endeavor" clusters in thing-space with "keeping an eye out for people acting in untrustworthy and non-cooperative ways towards EAs", but these two things don't have to go hand in hand.

the risk of consolidating too much influence over EA in any one organisation or panel

Fair enough. As with the last point, the panel would likely consolidate less unwanted influence over EA if it focused solely on calling out sufficiently dishonestly harmful behavior by anyone who self-identified as an EA, and made no claims as to whether any individuals or organizations "counted" as EAs.

the risk of it being impossible to get agreement, leading to an increase in politicisation and squabbling

This seems like a concern that's good, in that a bit harder for me to address satisfactorily. Hopefully, though, there would some clear-cut cases the panel could choose to consider, too; the case of Intentional Insights' poor behavior was eventually quite clear, for one. I would guess that the less clear cases would tend to be the ones where a clear resolution would be less impactful.

In response, we toned back the ambitions of the proposed ideas.

I'd have likely done the same. But that's the wrong thing to do.

In this case, the counterfactual to having some sort of panel to call out behavior which causes unreasonable amounts of harm to EAs is relying on the initiative of individuals to call out such behavior. This is not a sustainable solution. Your summary of your previous post puts it well:

There’s very little to deal with people representing EA in ways that seem to be harmful; this means that the only response is community action, which is slow, unpleasant for all involved, and risks unfairness through lack of good process.

Community action is all that we had before the Intentional Insights fiasco, and community action is all that we're back to having now.

I didn't get to watch the formation of the panel you discuss, but it seems like a nontrivial amount of momentum, which was riled up by the harm Intentional Insights caused EA, went into its creation. To the extent that that momentum is no longer available because some of it was channeled into the creation of this panel, we've lost a chance at building a tool to protect ourselves against agents and organizations who would impose costs on, and harm EAs and EA overall. Pending further developments, I have lowered my opinion of everyone directly involved accordingly.

null @ 2017-03-09T13:36 (+5)

FWIW, as someone who contributed to the InIn document, I approve of (and recommended during discussion) the less ambitious project this represents.

null @ 2017-03-08T07:07 (+3)

I like this, but I think collaborative spirit should be augmented by remembering the value of unity and solidarity, which is rather different than mere collaboration and cooperation. Curious why it didn't get included.

null @ 2017-03-09T14:23 (+9)

We recognize that there are major areas of disagreement between people who are committed to the core ideas of EA, and we don't want emphasis on "unity" to sound like "you have to agree with the majority on specific topics, or you should leave."

null @ 2017-03-09T14:43 (+3)

Ah, I see. Thanks for the response. I agree 100%.

I was framing it in more, well, tribalistic terms, almost to the opposite effect. Basically, if you're an EA trying to achieve your goals and have to deal with problems from outsiders, then regardless of whether we agree, I'm "on your team" so to speak.

null @ 2017-03-08T05:44 (+3)

I'm very much in favor of this