Thanks for sharing your perspective.

• Fish Welfare Initiative: Hasn't worked very well, and seemed like it wouldn't in advance, doesn't do CE's original proposed idea anymore, more feedback ahead of time would have told them not to do the original idea

I posted about the cost-effectiveness of the fish welfare interventions recommended by Ambitious Impact (AIM), and Fish Welfare Initiative’s farm program.

Thanks for sharing, JP and Lizka! Great principles.

Thanks for the update!

Over the past year or so, I’ve become increasingly convinced by arguments that we are clueless about the sign (in terms of expected total suffering reduced) of interventions aimed at reducing s-risk.

I believe one can positively influence futures which have an astronomical positive or negative value, but negligibly so. I think the effects of one's actions are well approximated by considering just the 1st 100 years or so.

I think this first example isn’t comparable, and a bit of a strawman. The Vox article is about how bad factory farming is, and how we don’t need to do that to help humans to flourish. This discussion is about potentially withdrawing life-saving interventions because they might be detrimental to animals. This directly connects the saving lives to harming animals – the Vox article doesn’t.

I understand they are not directly comparable, but I guess the newsletters from Vox also reach a much wider audience who is less acceptable of controversial takes compared to readers of the EA Forum.

I’d say I would have allowed for a much bigger behaviour score range

Do you mean you would account for other behavioural proxies for welfare capacity besides the 90 RP considered? Which ones? RP seemed to be super comprehensive.

perhaps through having more than a binary yes/no system on some of the behaviours

I do not understand. RP did not have a binary system to determine the probability of sentience.

The program runs 10,000 simulations where the presence or absence of each proxy in the “Simple_Scoring” spreadsheet was a random variable. For a given organism, the steps taken in a single simulation to generate the proxies possessed by that organism were:

1. Start with a dictionary containing each proxy and an empty list to add scores to.
2. For each proxy in the Simple Scoring sheet:
   1. First, randomly select the probability that the organism possesses the proxy from a uniform distribution whose range maps onto the judgment determined by our contractors. The probability map is as follows:
      1. No: [0.00, 0.00] (Used for the “motile” proxy)
      2. Likely no: [0, 0.25)
      3. Lean no: [0.25, 0.5)
      4. Lean yes: [0.5, 0.75)
      5. Likely yes: [0.75, 1.0]
      6. Yes: [1.00, 1.00] (Used for the “motile” proxy)

For example, if pigs scored a “Likely yes” on taste aversion behavior, then the probability that pigs exhibit taste aversion behavior is sampled uniformly over the interval [0.75, 1.0]. If a proxy was judged “Unknown”, then we defaulted to giving a zero probability of it being present; however, this default can be changed at the start of running the program.

1. Second, generate a Bernoulli random variable using this probability of the species possessing the trait, where 1 indicates that the trait is present and 0 indicates that it is absent.
2. Add the score (0 or 1) to the list corresponding to its particular proxy in the dictionary.

For a given organism, this process was repeated for 10,000 simulations, where each proxy’s score in a given simulation was appended to its respective list. Then, we repeated this procedure for all eighteen organism types studied and saved the simulated proxy data.

RP did not use a binary system to determine the welfare range conditional on sentience, and actually underestimated this by giving zero weight to proxies for which there was no information (see what I bolded below).

To generate the distributions of welfare ranges across species and models, the user must answer the same three questions about whether to give non-zero probability to “Unknown,” “Lean no,” and “Likely no” judgments and what weight should be given to proxies we’re highly confident matter for welfare ranges as were asked for the probabilities of sentience. As before, users can change the probabilities given to “Unknowns” for one or more species of their choosing. 

In our final simulations: 

1. We chose not to assign any weight to proxies with “unknown” judgments for any species. This likely leads to underestimating the welfare ranges for several animal types.
2. We chose to assign positive probabilities to proxies whose judgments are “likely no” and “lean no.”
3. We weighted the proxies that we are highly confident matter for welfare ranges as being five times as important as all other proxies.

giving less complex pain response behaviors a fraction of more complex ones

I do not understand what you are referring to. Which specific proxies do you think should be weighted more heavily?

I think the 4^th input seems absurd and I won’t rehash this much as many others have made arguments against your reasoning on this thread. You’re translating a figure which is on the upper bound of judging severe human pain (which like Bruce said, by definition can’t last long) directly onto what you think might be happening in mosquitos – a wildly different organism.

For all the analyses relying on pain intensities I am aware of, from AIM and RP, the ratio between the intensity of a pain of a given category and that of another is the same across species. I have now asked Cynthia Schuck-Paim, who is the research director of WFP (the organisation defining the pain intensities).

I agree excruciating pain "can’t last long", but this is consistent with my estimate that ITNs cause 119 s of excruciating pain per mosquito they kill.

On what grounds really would mosquitos dying of poisoning likely cause that severe pain at a best guess? I think its possible but very unlikely so I think it would be reasonable for the sake of conservatism to reduce this number by orders of magnitude.

As I say in the post, my estimates for the time in pain come from aggregating 3 sets of estimates provided by WFP's GPT Pain-Track. My estimate of 119 s of excruciating pain may well not be accurate, but which evidence do you have for it being "possible but very unlikely" to be that long?

4. On the number of mosquitos front for a start I don’t like comments like “my takeways would probably be the same even if….” Multipliers can add up, and we’re trying to move towards accuracy so I think it can be an unhelpful copout to question how much any element of an analysis matters – Rethink Priorities said things like this a number of times during their moral weights project which was a small red flag for me.

I agree. At the same time, I think it is worth having in mind how far one is from reversing the conclusions.

I agree there’s no empirical research on the mosquito number front, but from my perspective having travelled around Africa and living in a grass thatched hut and sleeping under a mosquito net for the last 10 years, 24 mosquitos killed a day on average per net seems extremely unlikely. That would be something like 240 million mosquitos killed by nets alone every day in Uganda – which seems to me perhaps plausible but unlikely. From a distance I think you could have been more conservative with your “best guess”

As I replied to Bruce:

I estimate GW's last grant to AMF will kill 0.0183 % as many mosquitoes as the ones currently alive globally over the lifetime of the bednets it funds. This would correspond to killing 1.19 % (= 1.83*10^-4*195/3) of the mosquitoes in DRC per year assuming mosquitoes were uniformly distributed across the existing 195 countries, and that the nets funded by the grant are distributed over 3 years. In reality, it would be less because DRC should have more mosquitoes than a random country. AMF being responsible for killing less than 1.19 % of the mosquitoes in DRC does not sound implausible.

Hi Cynthia,

Under your framework, if one supposes hurtful, disabling and excruciating pain are a, b and c times as intense as annoying pain in humans, should the same apply to other species? Or could the intensity ratios vary across species? My understanding is that the ratios are supposed to be constant across species.

On the other hand, it is much easier to scale even short periods of excruciating suffering with high numbers of animals, especially when you're happy to consider ~8 million mosquitos killed per human life saved by a bednet - I don't have empirical evidence to the contrary, but this seems wildly high.

Great point! It makes sense that harm is proportional to the number of individuals holding the pain caused to each individual constant. So my numbers would only be off if there was a reason to think mosquitoes cannot experience excruciating pain for as much time as humans.

This means it requires approximately 100 bednets over the course of 1 year to save 1 life/~50 DALYs.

"GW calculates a cost per distributed net and life saved of 6.78 $ and 5.10 k$ [in DRC]. These imply AMF has to distribute 752 nets [= 5.10*10^3/6.78] to save a life in DRC".

At your preferred rate of 1 mosquito death per hour per net^[1] this comes to approximately 880,000 mosquito deaths per life saved,^[2] which is 3 OOMs lower than the ~8 million you would reach if you do the "excruciating pain" calculation, assuming your 763x claim is correct$.$^[3]

There are 2 issues here:

• 880 k is 1 order of magniture lower than 8 M, not 3.
• I got 7.91 M (= 1*24*365.25*1.20*752) with the 1st approach, which practically matches the 8 M you got with the 2nd approach, multiplying:
  • 1 mosquito killed per net-hour.
  • 24 net-hours per net-day.
  • 365.25 net-days per net-year.
  • 1.20 net-years per net in DRC, as estimated by GW.
  • 752 nets per life saved in DRC, as estimated by GW (see above).

So I think it would be better if you edited the claim below you make at the start of your 1st comment in this thread.

I think you are probably at least a few OOMs off with these figures, even granting most of your assumptions, as this implies (iiuc) ~8 million mosquito deaths per human death averted.

I do not think you uncovered any errors in my calculations.

For example, by your preferred working definition of excruciating pain, it definitionally can't exist for more than a few minutes at a time before neurological shutdown. I think this isn't necessarily unreasonable, but there might be failure modes in your approach when basically all of your BOTECs come down to "which organisms have more aggregate seconds of species-adjusted excruciating pain".

I estimate each mosquito affected by ITNs experiences 1.98 min (= 119/60) of excruciating pain, which is in line with the definition.

Thanks, Nick.

I don't want this to be the kind of discussion which is front and center on our public EA forum for external facing reasons and PR risk. This might well be the wrong response and many people might disagree with me on this which is fair enough.

Bryan Walsh published a newsletter for Vox, Human progress has come at the expense of animals. It doesn’t have to., where there is a discussion of the EA Forum post Net global welfare may be negative and declining by Kyle Fish. In addition, "A version of this [Bryan's] newsletter originally appeared in the Future Perfect newsletter".

If AMF and the nets weren't doing their job, I would guess the alternatives might be at least as bad for mosquito welfare.

I do not think this is the relevant counterfactual. I assume ITNs (or malaria chemoprevention, or vaccines) save human lives more cost-effectively than the alternatives you mentioned. Otherwise, I would expect GiveWell to recommend them. So, to the extent people want to save lives as cost-effectively as possible, they will use the available resources to fight malaria to distribute ITNs. As a result, I think less donations to AMF simply result in less ITNs around. 

• As many have mentioned, @Vasco Grilo🔸's blunt tool of estimating "excruciating pain" seems almost absurd. The idea that a mosquito experiences that degree of suffering as it dies of poison is possible but so unlikely that I think that best guesses for that number should be revised down by orders of magnitude. See other comments for in-depth arguments, I appreciated this comment from @bruce 
  
  "I don't know where exactly to draw the line here, but 14.3 mosquito days of excruciating suffering for one happy human life seems clearly beyond it."

I estimate 14.3 mosquito-days (= 51*365.25/0.013*10^-5) of excruciating pain neutralise the additional human welfare from saving a life under GW's moral weights multiplying:

• 51 DALYs averted per life saved under GW's moral weights.
• 365.25 days per year.
• The ratio between the intensity of fully healthy human and mosquito life, which I set to the reciprocal of RP’s median welfare range of black soldier flies of 0.013.
• The ratio between the intensity of fully healthy life and excruciating pain of 10^-5 (= 0.1/(10*10^3)). This implies 1 day of fully healthy life is neutralised by 0.864 s (= 24*60^2*10^-5) of "scalding and severe burning events [in large parts of the body]", or "dismemberment, or extreme torture".

Do the 3rd and 4th inputs seem "almost absurd"?

• There's just no way mosquito nets kill anything like 24 mosquitos a day on average. I would guess this at under 5. Basing this estimate on Vasco getting a lot of bites while sitting outdoors at night (Mosquito nets are indoors) doesn't make much sense.

I was seeing much more than 1 mosquito per hour during dust outdoors in Moshi (Tanzania). People most affected by malaria may sleep indoors, but have houses with very limited ability to prevent mosquitoes from entering. The roofs can be made of turf too, which would attract mosquitoes.

I did not easily find empirical estimates for the number of mosquitoes killed by ITNs. I would be happy to update the number if you find them, although my takeaways would be the same even if ITNs only killed e.g. 10 % as many mosquitoes as I assumed.

• Anchoring on RP's moral weights which I argue might be too animal friendly will often mean that any human helping intervention which hurts animals at all might look "bad". Especially in the case of smaller animals which are extremely different from humans, their use of binary behavioural proxies is almost guaranteed to give high moral weights.

As Bob Fischer said, "I don’t think you’ve said anything that should cause someone to question that headline result", "the ultimate question is whether our [RP's] decisions were wrong, not whether they can be construed as animal-friendly".

this, like some of your other estimates relies rather heavily on an unconventional and extremely skewed pain scale, whereby a certain degree of pain is worth many times more than maximal pleasure

My guess that excruciating pain is 100 k (= 10*10^3/0.1) times as intense as fully healthy life ( 100 k times as intense as the maximally happy instant) implies that 0.864 s (= 24*60^2/(100*10^3)) of excruciating pain in humans neutralises 1 day of fully healthy life in humans. Do you think this is "unconventional and extremely skewed"? Examples of excruciating pain include "scalding and severe burning events [in large parts of the body]", or "dismemberment, or extreme torture". I have not experienced excruciating pain myself, so I would not be surprised if it was 10 % as intense as I assumed, but this would not change my takeaways. I estimate the harms caused to mosquitoes are proportional to the intensity of excruciating pain, so AMF would cause 76.3 (= 0.1*763) times as much harm to mosquitoes as it benefits humans according to the updated mainline numbers.

I am not assuming excruciating pain is 100 k times as intense as the maximally happy instant (as you implied above). This is much more intense than fully healthy life, so I guess excruciating pain is much less than 100 k times as intense as the maximally happy instant.

I 'm not sure this is actually how RP intend their welfare ranges to be used. My understanding (and I welcome clarification/correction from RP on this point) is that when their researchers estimate that $creature's welfare range is 1.3% that of humans, they intend that to be interpreted as "$creature's pain sensations are at most 1.3% as intense as human experience", not "to establish how intensely $creature feels pain, multiply 1.3% by a pain scale which may contain an arbitrarily large number of digits, to reach the conclusion that this creature's pain is potentially thousands of times as intense as human pleasure."

I assumed the welfare per living time of fully healthy life in a given species is proportional to the welfare range of that species. This does not directly follow from RP's welfare ranges, but does not obviously favour helping animals or humans. My understanding it that RP's median welfare ranges refer to the ratio ("welfare per time of the practically maximally happy animal-year (positive)" - "welfare per time of the practically maximally sad animal-year (negative)")/("welfare per time of the practically maximally happy human-year" - "welfare per time of the practically maximally sad human-year").