An approach to doing good is to come up with a metric for what constitutes good or bad, and then trying to do things that will optimally increase or decrease such metric, as the case may be.
If you do this, you have to be careful about what metric you choose.
If you have an ontology where you measure good by “number of people who feel benefited by you”, you might end up doing things like sending everyone you can a doughnut with a signed note. If instead your metric is “number of people classified as poor” you might do best to focus on interventions that get people just over the hump of poverty as defined by your scale. And so on.
Conscientious and systematic altruists tend to see problems with metrics like those above. They realize that “people impressed” and “being poor according to an economic metric” are not metrics that really carve nature at its joints.
Dissatisfied with misleading metrics, one then tends to look closer at the world and arrive at metrics that take into account the length of different lives, their quality, their instrumental effect in the world, how much are they exactly being benefited by the intervention relative to other cost-effective alternatives, and so on. And that’s how you get things like Quality Adjusted Life-Years (QALY), micromorts, and the happiness index.
This is, I think, all moving in the right direction. Metrics that make an effort to carve nature at its joints can provide new lenses to see the world. And looking through those lenses tends to generate novel angles and approaches to do a lot of good.
This is why today I will suggest we consider a new metric: The Hell-Index.
A country’s Hell-Index could be defined as the yearly total of people-seconds in pain and suffering that are at or above 20 in the McGill Pain Index (or equivalent)*. This index captures the intuition that intense suffering can be in some ways qualitatively different and more serious than lesser suffering in a way that isn’t really captured by a linear pain scale.

What does this metric suggest we should do to make the world better? Here is an idea (told as if narrated from the future):
Between 2030 and 2050 it was very common for people to wear Get-Out-Of-Hell-Free Necklaces. People had an incredible variety of custom-fit aesthetic and practical additives to their necklaces. But in every single one of them, you could rest assured, you would find a couple of doses of each of these agents:
- N,N-DMT (in case of Cluster Headaches)
- Quetiapine (in case of severe acute psychosis)
- Novocaine + menthol (for very painful stings)
- Ketamine (for severe suicidal feelings)
- Microdosed Ibogaine + cocktail of partial mu-opioid agonists (for acute severe physical pain and panic attack, e.g.. kidney stones)
Some other people would get additional things like:
- Beta blocker (to take right after a traumatic event)
- Agmatine (to take in case you suspect of having being brainwashed recently), and
- Caffeine (if you absolutely need to operate heavy machinery and you are sleep-deprived)
In all cases, the antidote needed would be administered as soon as requested by the wearer. And the wearer would request the antidote as indicated by a very short test done with an app to determine the need for it.
But why? What’s this all about?
The Get-Out-Of-Hell-Free Necklace contents were chosen based on a cost-benefit analysis for how to reduce the world’s Hell-Index as much as possible. Cluster-headaches, kidney stones, bad stings, severe psychotic episodes, suicidal depression, panic attacks, and many types of acute physical pain turned out to account for a surprisingly large percentage of each country’s Hell-Index. And in many of these cases, a substantial amount of the suffering was experienced before medical help could be able to arrive to the scene and do anything about it. A lot of that intense suffering happened to be tightly concentrated in acute episodes rather than in chronic problems (save for some notable examples). And by incredible luck, it turned out that there were simple antidotes to most of these states of agony, all of them small enough to fit in a single light necklace. So it was determined that subsidizing Get-Out-Of-Hell-Free Necklaces was a no-brainer as a cost-effective altruistic intervention.
By 2050 safe and cheap genetic vaccines against almost all of these unpleasant states of consciousness had been discovered. This, in turn, made the use of the Get-Out-Of-Hell-Free Necklaces unnecessary. But many who benefited from it- who had been unlucky enough to have needed it- kept it on for many years. The piece was thought of as a symbol to commemorate humanity’s progress in the destruction of hell. An achievement certainly worth celebrating.
* Admittedly, a more refined index would also distinguish between the intensity of different types of pain/suffering above 20 in the McGill Pain Index (or equivalent). Such index would try to integrate a fair “total amount of hellish qualia” by adding up the pain of each state weighted by its most likely “true intensity” as determined by a model, and then do so for each model you have and weight the contribution of each model by its likelihood. E.g. do both a quadratic and an exponential conversion of values in the 0 to 10 visual analogue scale into dolors per second, and then do a likelihood-weighted average to combine those results into a final value.
[Note: Cross-posted in Qualia Computing]
For context, this post is motivated by the realization that pain-scales are logarithmic compressions of what is probably an exponentially-increasing capacity for pain and suffering in sentient beings. Here is a simple example, as told by the guy who stung himself with 80+ insects to put the pain on a scale:
4:28 - The harvester ants is what got the sting pain scale going in the first place. I had been stung by honey bees, old jackers, paper wasps, etc. the garden variety stuff, that you get bitten by various beetles and things. I went down to Georgia, which has the Eastern-most extension of the harvester. I got stung and I said “Wooooow! This is DIFFERENT!” You know? I thought I knew everything there was about sting insects, I was just this dumb little kid. And realized “Wait a minute! There is something different going on here”, and that’s what got me to do the comparative analysis. Is this unique to harvester ants? Or are there others that are like that. It turns out while the answer is now we know what’s later - it’s unique! [unique type of pain].
7:09 - I didn’t really wanted to go out and get stung for fun. I was this desperate graduate student trying to get a thesis, so I could get out and get a real job, and stop being a student eventually. And I realized that, oh, we can measure toxicity, you know, the killing power of something, but we can’t measure pain… ouch, that one hurst, and that one hurts, and ouch that one over there also hurts… but I can’t put that on a computer program and mathematically analyze what it means for the pain of the insect. So I said, aha! We need a pains scale. A computer can analyze one, two, three, and four, but it can’t analyze “ouch!”. So I decided that I had to make a pain scale, with the harvester ant (cutting to the chase) was a 3. Honey bees was a 2. And I kind of tell people that each number is like 10 equivalent of the number before. So 10 honey bee stings are equal to 1 harvester ant sting, and 10 harvester ant stings would equal one bullet ant sting.
Some EAs have already pointed this out, but I suspect that this has yet to make it into "EA Canon" and it thus hasn't really given rise to novel interventions. Brian Tomasik, Jonathan Leighton, Manu Herran, David Pearce, and other "suffering focused ethics" people have been saying this for a while. Read more here.
I realize that it is always kind of depressing to look into how bad suffering gets, but it does seem important to know if one is indeed trying to reduce suffering as much as possible.
I've suggested something similar for happiness (https://www.lesswrong.com/posts/7Kv5cik4JWoayHYPD/nonlinear-perception-of-happiness ). If you don't want to introduce the weird asymmetry where negative counts and positive not, what you get out of that could be somewhat surprising - it possibly recovers more "common folk" altruism where helping people who are already quite well off could be good, and if you allow more speculative views on the space on mind-states, you are at risk of recovering something closely resembling some sort of "buddhist utilitarian calculus".