"Reducing astronomical waste need not involve preventing human extinction—it can involve other changes in humanity’s long-term trajectory."

Glad to see this gaining more traction in the x-risk community!

I don't "see other things in the vicinity such as targeting value improvements where it's almost certain that at least one of them is better." That's where I was asking for better alternative subgoals to reaching value improvements.

I think we practically understand each other's points now though. Thanks for the discussion and clarification.

"I think the arguments we've discussed show reasonably conclusively that animal welfare isn't the best instrumental goal -- because we can see other things in the vicinity such as targeting value improvements where it's almost certain that at least one of them is better."

It seems the "improving animal welfare" interventions could very well be the best ways to improve those values. I think that's a key point where we disagree. I'd be interested in hearing what you think are better alternatives at some point.

If there are clearly better options for proxies related to "improving animal welfare," but not clearly better options for proxies related to "improving human welfare," then "improving animal welfare" could still be the better option of the two. Analogy: if we have two car races with five separate cars each, the worst car in one race could still be better than all five in the other.

Thanks for the clarification. Claim (iii) with "unlikely" rather than "extremely unlikely" is a tenable view, and the specifics, of course, depend on other ways we can affect the far future. Do you think it's fair to put the modified (iii) claim in the same category as...

(v) That improving human welfare with interventions like antimalaria nets, deworming pills, or cash transfers seems "unlikely" to be the best way of meeting the goal of making the long-term future flourish.

I take it you do put these in the same category as you say you are (vi) not making the claim: That improving human welfare is a better instrumental sub-goal for improving the long-term future than improving animal welfare.

But you also claim (vii) "In contrast, we do not have similarly strong reasons to think “improve human welfare” is definitely not the best approach." There seems to clearly be a tension between (vi) and (vii).

Could you resolve it?

I am glad you are considering far-future effects in terms of proxies like "increase empathy." I think this is a useful method. Also, I commend you for considering far-future impacts in general, as increasing our confidence in effecting them seems to have high potential upside for EA decision-making.

On the specifics of this post, I am unsure exactly what your claim is now - whether it's (i) that a happier non-human farm animal, all else equal, seems "extremely unlikely" to meet the goal of making the long-term future flourish, or (ii) that improving animal welfare with interventions like vegan leafleting, public demonstrations, antispeciesism essays, or lobbying for farm animals seems "extremely unlikely" to meet the goal of making the long-term future flourish.

(i) seems clearly true, but if you mean (ii), I think that's debatable and am unsure how you arrive at "extremely unlikely." Most non-human animal activists see great value in helping farm animals now as well as potential impact on the far future. I think strong evidence is required to reject their claims with such confidence.

For example, even Vegan Outreach, arguably the most short-run impact focused organization in the field, considers its impact in both ways: "In addition to influencing the diets of individuals, VO wants to change the ways that people view farm animals and teach people that farm animals are capable of suffering. VO aims to influence public opinion to affect long-term public policy." (http://files.givewell.org/files/conversations/Jack%20Norris%205-20-14%20(public).pdf)

Some organizations are much more explicit about their goals for promoting general antispeciesism sentiment and helping animals in the long-run, like Direct Action Everywhere (http://directactioneverywhere.com), and concern themselves very little (if at all) with individual dietary changes.

Now, this isn't clearly the best route to improving the long-term future, but I think it would take very strong evidence to say it is "extremely unlikely" and the point merits further consideration. I know my personal decisions could change with new information on these potential far-future effects.

I'm happy to hear you state your views outside the post. They seem reasonable and open-minded, which was not my original impression. I look forward to reading more of your work. Always feel free to send me articles/ideas for critique/discussion.

If that is your assertion, I feel the post is misrepresenting your view as something much stronger (i.e human-focused causes have significant positive impact that non-human animal-focused causes do not, therefore human-focused causes are better). This is disingenuous and caused our negative reactions.

"Cost-effectiveness is one of the classic tools used in prioritisation, and at least in theory a higher level of cost-effectiveness should exactly imply higher priority. Now the issue is that we don't trust our estimates, because they may omit important consequences that we have some awareness of, or track the wrong variables."

I totally agree.

"But when people bring cost-effectiveness estimates up, there is often an implicit claim to priority (or one may be read in even if not intended)."

I would agree with the point in parentheses, but often it's just brought up as one factor in a multitude of decision-making criteria. And I think that's a good place for it, at least until we get better at it.

Owen,

I appreciate that you're thinking about flow-through/long-term effects and definitely agree we need more discussion and understanding in the area.

My "hostility" (although it isn't that extreme =] really) is primarily due to the propagation of the assumption that "human-focused causes have positive significant flow-through effects while non-human animal-focused causes do not." We have a lot more research to do into questions like this before we have that sort of confidence.

So the danger here is that impact-focused people might read the post and say "Wow, I should stop trying to support non-human animal-welfare since it doesn't matter in the long-run!" I realize that your personal view is more nuanced, and I wish that came across more in your post. The possible flow-through effects: (i) promoting antispeciesism, (ii) scope sensitivity, (iii) reducing cognitive dissonance, and many more seem perfectly viable.

Hope that makes sense.

Although I generally encourage dissenting opinions in the EA community, I think the idea expressed by this post is harmful and dangerous for similar reasons as those expressed by Brian and Peter.

1) "Some have argued for estimates showing animal welfare interventions to be much more cost-effective per unit of suffering averted, with an implication that animal welfare should perhaps therefore be prioritised."

This seems to be a misrepresentation of the views held by many EAs. Cost-effectiveness calculations are employed by every EA prioritization organization, and nobody is claiming they imply a higher priority. They are only one of many factors we consider when evaluating causes.

2) "Moreover, if it could achieve a lasting improvement in societal values, it might have a large benefit in improved animal welfare over the long-term."

I am glad this sentence was included, but it is relatively deep in the post and is one of the strongest reasons EAs advocate against factory farming and speciesism. I posted my thoughts on the subject here: ( http://thebestwecan.org/2014/04/29/indirect-impact-of-animal-advocacy/ )

3) "The upshot of this is that it is likely interventions in human welfare, as well as being immediately effective to relieve suffering and improve lives, also tend to have a significant long-term impact. This is often more difficult to measure, but the short-term impact can generally be used as a reasonable proxy. ... "

I could replace "human" with "non-human animal" welfare here and the argument would be just as valid. It's a grand assumption to think this applies to human-focused causes but not others. If you have further justification for this, I think that would be an interesting post.

4) "For many types of human welfare intervention, we can use the short-term benefits to humans as a proxy for ongoing improvements in a way that is not possible – and may be misleading – when it comes to improvements to animal welfare."

I think we'd all be happy for you to defend this assertion, since it is quite controversial within EA and the broader community.