At the 2014 Effective Altruism Summit, each of Geoff Anders, Peter Thiel, and Holden Karnofsky identified three heuristic criteria for effective altruists to use in selecting a cause area:
neglected
valuable
tractable
I. Why Does Effective Altruism Neglect Ant-Aging?
Anti-aging doesn’t seem very tractable on its face, but neither does existential risk reduction. Despite both being causes within emphasized by rationalists and transhumanists, anti-aging has been left outside of effective altruism thus far. I believe this is because the rationalist community as a precursor to effective altruism better coordinated their concern over existential risk better than their concern over anti-aging efforts.
Like, through Less Wrong, and the Machine Intelligence Research Institute, (almost) every existential risk reduction organization got in touch with another. This formed a solid voice advocating for this cause when effective altruism started. On the other hand, Aubrey de Grey and his organization, SENS, seem like the only one(s) in contact with effective altruism, while the rest of the major anti-aging advocates run their organizations out of touch with us, and each other.
Another thing about ant-aging is that while it on its own may seem like a worthy intervention, it often gets lumped with cryonics, and other transhumanist technologies, that seem even less tractable than anti-aging research. That is, those aspects are frequently dismissed by rationalist, let alone effective altruists. So, if the most vocal advocates for anti-aging research only communicate that signal with a bunch of noise, effective altruists may be less likely to consider it.
This seems like a historical contingency to me, based on how the rationalist community organized itself with some circles but not others. This makes possible but by no means definite that the rationalist community has not emphasized anti-aging enough within effective altruism, relative to existential risk reduction.
II. Why Does Effective Altruism Neglect (Better) Policy Advocacy?
This also seems to be due in part to historical contingency. First of all, there is the wariness among the rationalist community that delving into the trenches of politics will be much less tractable than aiding the world through other means. I believe I mildly perceive the same strain of thinking as an undercurrent of utilitarians such as Toby Ord, or Peter Singer.
Also, Givewell thought it much more difficult to assess policy advocacy when measuring impact qualifications for it was much more difficult. In other words, Givewell wanted to cut their teeth, and gain experience, in an area more measurable than policy advocacy. Before, like other charity evaluators they were giving recommendations to individual donors. Now, with Good Ventures, they’re giving recommendations for foundations, with much more money.
In conversation with my friend Joey Savoie a few weeks ago, we discussed maybe Givewell is exploring policy advocacy through the Open Philanthropy Project now because noticeable gains in policy change can only be affected with large investments, and it’s only now with Good Ventures that Givewell has an ally with sufficient weight to get that happening.
My opinion was that anti-aging and existential risk seem roughly equally neglected and roughly equally tractable, but existential risk seems a whole lot more valuable, so hence the focus on that instead.
I concur. This explanation works for why precursor movements to effective altruism such as the rationalist community would have emphasized existential risk over anti-ageing research as well.
II. Why Does Effective Altruism Neglect (Better) Policy Advocacy?
I’m not sure that this is necessarily the case among EA orgs with full-time staff. The Centre for Effective Altruism (in particular the Global Priorities Project, which is our collaboration with FHI), The Open Philanthropy Project and the Cambridge Centre on Existential Risk are putting considerable effort into policy work. For example, I and others at CEA put the majority of our time over the past week into policy research, and our trustees were at a meeting at No. 10 Downing Street yesterday. I have written up some of my thoughts on our early policy work at http://effective-altruism.com/ea/7e/good_policy_ideas_that_wont_happen_yet/
I think that there are a few effects going on here which cause policy to appear under-neglected among the community at large...
There is a relatively larger barrier to entry in policy work (compared to e.g. making a donation to a GiveWell recommendation), which means that policy work is often done by people working in this area full-time, or who have past experience in the area. This may be one of the reasons why the community at large isn’t doing more policy analysis. I think it would be useful if the EA community did do more policy analysis, in particular making recommendations of policies that could feasibly happen (i.e. tweak this thing, not ban agriculture subsidies) and doing analyses of the type I outline in my post above (e.g. what are the benefits, what are the costs, who will be in favour, who will be against, how can we change the policy to make it more feasible while retaining most of the benefits, how would we actually make this change, and who do we ultimately need to convince about this to make it happen, etc.). I for one would find this useful in informing the work that I do in this area, and if the ideas are good enough they would likely be taken forwards.
Policy work is often under-publicised unless there are major breakthroughs. In doing this work we are developing ongoing relationships with people, and if we were to publicise these relationships on the internet we could damage them. For this reason we often find it difficult to talk about our policy work extensively in public.
There may also be cultural and path-dependent effects at play here, which people have mentioned above/below and elsewhere, so I won’t go into them in detail.
Effective altruist organizations with full-time staff definitely aren’t neglecting policy advocacy. I meant the broader community at large, in the sense that for the last two years it’s been focusing upon: reducing global poverty and illness; animal advocacy; reducing existential risk.
I think it would be useful if the EA community did do more policy analysis, in particular making recommendations of policies that could feasibly happen[...]I for one would find this useful in informing the work that I do in this area, and if the ideas are good enough they would likely be taken forwards.
At the 2014 Effective Altruism Summit, each of Geoff Anders, Peter Thiel, and Holden Karnofsky identified three heuristic criteria for effective altruists to use in selecting a cause area:
neglected
valuable
tractable
This “three factor model” of cause assessment has been used by 80,000 Hours for a long time (they use the terms ‘crowdedness’, ‘importance’ and ‘tractability’). Do we know where it originated?
I think it originated with GiveWell—they used something like this framework for assessing cause areas, which 80k then based their framework on. It’s possible I’m misremembering this though.
GiveWell has used this “three factor model” as well (they also use the terms ‘crowdedness,’ ‘importance,’ and ‘tractability’). I’m not sure about the dates when either organization started using this model, and I wouldn’t be at all surprised if people started using it independently, since it’s rather intuitive.
It makes sense, but I didn’t know about the model before the Effective Altruism Summit. Having it crystallized is great, and everyone should know about it, so I want to write a post about it for this forum.
This comment is a reminder to myself to write it.
If anyone wants to help me write it, or give feedback, please send me a private message.
At the 2014 Effective Altruism Summit, each of Geoff Anders, Peter Thiel, and Holden Karnofsky identified three heuristic criteria for effective altruists to use in selecting a cause area:
neglected
valuable
tractable
I. Why Does Effective Altruism Neglect Ant-Aging?
Anti-aging doesn’t seem very tractable on its face, but neither does existential risk reduction. Despite both being causes within emphasized by rationalists and transhumanists, anti-aging has been left outside of effective altruism thus far. I believe this is because the rationalist community as a precursor to effective altruism better coordinated their concern over existential risk better than their concern over anti-aging efforts.
Like, through Less Wrong, and the Machine Intelligence Research Institute, (almost) every existential risk reduction organization got in touch with another. This formed a solid voice advocating for this cause when effective altruism started. On the other hand, Aubrey de Grey and his organization, SENS, seem like the only one(s) in contact with effective altruism, while the rest of the major anti-aging advocates run their organizations out of touch with us, and each other.
Another thing about ant-aging is that while it on its own may seem like a worthy intervention, it often gets lumped with cryonics, and other transhumanist technologies, that seem even less tractable than anti-aging research. That is, those aspects are frequently dismissed by rationalist, let alone effective altruists. So, if the most vocal advocates for anti-aging research only communicate that signal with a bunch of noise, effective altruists may be less likely to consider it.
This seems like a historical contingency to me, based on how the rationalist community organized itself with some circles but not others. This makes possible but by no means definite that the rationalist community has not emphasized anti-aging enough within effective altruism, relative to existential risk reduction.
II. Why Does Effective Altruism Neglect (Better) Policy Advocacy?
This also seems to be due in part to historical contingency. First of all, there is the wariness among the rationalist community that delving into the trenches of politics will be much less tractable than aiding the world through other means. I believe I mildly perceive the same strain of thinking as an undercurrent of utilitarians such as Toby Ord, or Peter Singer.
Also, Givewell thought it much more difficult to assess policy advocacy when measuring impact qualifications for it was much more difficult. In other words, Givewell wanted to cut their teeth, and gain experience, in an area more measurable than policy advocacy. Before, like other charity evaluators they were giving recommendations to individual donors. Now, with Good Ventures, they’re giving recommendations for foundations, with much more money.
In conversation with my friend Joey Savoie a few weeks ago, we discussed maybe Givewell is exploring policy advocacy through the Open Philanthropy Project now because noticeable gains in policy change can only be affected with large investments, and it’s only now with Good Ventures that Givewell has an ally with sufficient weight to get that happening.
My opinion was that anti-aging and existential risk seem roughly equally neglected and roughly equally tractable, but existential risk seems a whole lot more valuable, so hence the focus on that instead.
I concur. This explanation works for why precursor movements to effective altruism such as the rationalist community would have emphasized existential risk over anti-ageing research as well.
I’m not sure that this is necessarily the case among EA orgs with full-time staff. The Centre for Effective Altruism (in particular the Global Priorities Project, which is our collaboration with FHI), The Open Philanthropy Project and the Cambridge Centre on Existential Risk are putting considerable effort into policy work. For example, I and others at CEA put the majority of our time over the past week into policy research, and our trustees were at a meeting at No. 10 Downing Street yesterday. I have written up some of my thoughts on our early policy work at http://effective-altruism.com/ea/7e/good_policy_ideas_that_wont_happen_yet/
I think that there are a few effects going on here which cause policy to appear under-neglected among the community at large...
There is a relatively larger barrier to entry in policy work (compared to e.g. making a donation to a GiveWell recommendation), which means that policy work is often done by people working in this area full-time, or who have past experience in the area. This may be one of the reasons why the community at large isn’t doing more policy analysis. I think it would be useful if the EA community did do more policy analysis, in particular making recommendations of policies that could feasibly happen (i.e. tweak this thing, not ban agriculture subsidies) and doing analyses of the type I outline in my post above (e.g. what are the benefits, what are the costs, who will be in favour, who will be against, how can we change the policy to make it more feasible while retaining most of the benefits, how would we actually make this change, and who do we ultimately need to convince about this to make it happen, etc.). I for one would find this useful in informing the work that I do in this area, and if the ideas are good enough they would likely be taken forwards.
Policy work is often under-publicised unless there are major breakthroughs. In doing this work we are developing ongoing relationships with people, and if we were to publicise these relationships on the internet we could damage them. For this reason we often find it difficult to talk about our policy work extensively in public.
There may also be cultural and path-dependent effects at play here, which people have mentioned above/below and elsewhere, so I won’t go into them in detail.
Effective altruist organizations with full-time staff definitely aren’t neglecting policy advocacy. I meant the broader community at large, in the sense that for the last two years it’s been focusing upon: reducing global poverty and illness; animal advocacy; reducing existential risk.
How can the rest of us help?
This “three factor model” of cause assessment has been used by 80,000 Hours for a long time (they use the terms ‘crowdedness’, ‘importance’ and ‘tractability’). Do we know where it originated?
I think it originated with GiveWell—they used something like this framework for assessing cause areas, which 80k then based their framework on. It’s possible I’m misremembering this though.
Yeah I concur that GiveWell started it.
GiveWell has used this “three factor model” as well (they also use the terms ‘crowdedness,’ ‘importance,’ and ‘tractability’). I’m not sure about the dates when either organization started using this model, and I wouldn’t be at all surprised if people started using it independently, since it’s rather intuitive.
It makes sense, but I didn’t know about the model before the Effective Altruism Summit. Having it crystallized is great, and everyone should know about it, so I want to write a post about it for this forum.
This comment is a reminder to myself to write it.
If anyone wants to help me write it, or give feedback, please send me a private message.