Hey everyone,
I wrote a paper on why I think humane pesticides is the best cause area. I've had a few people read it and decided to post it here. Buck Shlegeris had an interesting criticism, which is that I fail to demonstrate why humane pesticides is superior to values spreading and reducing x-risk, and I'll address these in the comments section. Let me know what you think.
(I've been having trouble with formatting, so it might be better to read the google doc. It includes footnotes)
https://docs.google.com/document/d/1t1pnpOjcqtkOTAphMKRHtOpOZAqPYuvVGhCzZn_l8Eo/edit?usp=sharing
Abstract :
This paper argues for the importance of reducing insect suffering through the use of humane pesticides, advancing the claim that it is the best cause area on which to focus marginal time and money. I will first review the arguments for giving insects moral consideration; my main argument is that if insects have any reasonable chance of being sentient, the sheer number of insects we have an effect on necessitates their moral consideration. I will then focus on a feasible way to reduce potential insect suffering, which is the subsidization of humane pesticides. I will conclude that this approach is the marginally best cause-area because of its cost-efficiency, minimal bureaucratic needs, and dearth of people focusing on it. I will also attempt to dispel some concerns that focus on humane pesticides will lead to reduced pesticide use overall, which would potentially cause more insect suffering in the long-term.
Paper :
Why Insects are Morally Relevant
I will put forward two arguments for considering insect suffering. The first is that insects are sentient, thus making them worthy of moral consideration. By ‘sentience’ I mean phenomenal consciousness, which is the ability to have subjective internal experiences. (Note that I am assuming that sentience and ability to feel pain are the necessary factors for moral consideration. Anyone who thinks this is not the case, or thinks that animals don’t deserve moral consideration will be unconvinced by this argument) While I personally believe that insects are sentient, I don’t think this belief is necessary to advance the claim that we ought to support humane pesticides. My second argument is that if there is even a small chance that insects are sentient, the number of insects in existence is so large that we should take measures to avoid potentially harming them. This assumes William MacAskill’s position that in cases of moral uncertainty, we ought to maximize expected value. There are other schools of thought on how one ought to act when the moral status of a being is uncertain, but settling that dispute is beyond the scope of this paper. This argument places the deciding factor instead on insects’ percent chance of sentience, meaning that even if someone thinks insects are probably not sentient they still ought to support humane pesticides.
Insects engage in many behaviors that suggest they have an internal, subjective experience, and are thus sentient. Insects exhibit clear aversion to negative stimuli, which itself suggests they experience pain. However, this does not rule out the possibility that insects are more like a simple computer program than a sentient being. Very few people would consider a simple reinforcement algorithm as having an internal experience, so we require something else in the way insects respond to pain that suggests an internal experience over a machine-like negative reinforcement loop.
One fact which suggests sentience is insects’ response to bodily harm while under the influence of morphine. Insects produce natural opiates in a way similar to humans, and show much less reaction to injury when under the influence of artificial opiates. “... crustaceans, insects, and mollusks show less reaction to a noxious stimulus when they receive morphine. For example, morphine reduces the reaction of mantis shrimps to electric shock, praying mantises to electric shock, and land snails to a hot surface.” (Speciesism, 128). The administration of morphine seems to be hampering some kind of decision procedure that insects use to respond to harm. Rather than having an instantaneous aversion to harm, insects seem to be evaluating a sensation internally, then deciding on a certain response (in this case, moving away from said stimulus). Of course, morphine could be inhibiting some reflexive response to harmful stimuli rather than dulling an internal experience. For example, in humans spinal reflexes can generate responses to harmful stimuli with no conscious input, such as withdrawing one’s hand away from a flame. Insects have been shown to react to harmful stimuli even when their peripheral nervous system is isolated from their central nervous system (analogous to the spinal cord and brain respectively). Distinguishing between reflexive responses and an actual experience of pain is difficult, and I don’t expect to settle the question here. However, insects exhibit certain behaviors that suggest their pain is more than simple reflexive responses. In A Question of Pain in Invertebrates, bioethicist Jane Smith suggests which behaviors can help us distinguish between reflexes and pain. “ ...animals may also learn in the future to avoid situations similar to the one in which the pain occurred. Such responses, while not proof that the animals have experienced pain, can indicate that something more than a simple nociceptive reflex is involved”. If we accept that learning is suggestive of subjective experience, insects exhibit learning responses that suggest they experience pain. In a 1972 study on fruit fly larvae, larvae were successfully trained to avoid odorants released in conjunction with an electric shock. This not only suggests that insects have an internal experience, but more importantly an experience of pain and suffering.
Other behaviors further the case that insects experience phenomenal consciousness. Bees are capable of remembering task-relevant information in simple working-memory tests. Ants are able to recognize and cooperate with ants from the same colony, even across continents. Cockroaches display negative psychological effects when socially isolated at birth. Studies on insects’ social behavior are particularly compelling, as they suggest a certain degree of self-consciousness inherent in any social structure. Considering this evidence, I would find it odd if beings that respond to psychoactive drugs, are capable of learning, and form reasonably complex social structures weren’t capable of feeling pain. Thus, because it is more likely that insects are sentient we ought to give them moral consideration. I’m very aware of my position as an interested amateur and not an entomologist, and I concede that I could be wrong about insect sentience. Because of this, I’ll also advance the more substantial claim that the sheer number of insects requires us to give them moral consideration with even a small chance of sentience.
There are roughly 10^18 insects in the world. Suppose we give insects a .1% chance of being sentient, with their sentience being .1% of a human’s. (These values are intentionally small to demonstrate the scale to which insect suffering dominates) Assuming we assign moral weight to categories of beings by their number and the intensity of their inner experiences, this assignment gives each insect 1/1,000,000 of the moral weight for a human, meaning that the suffering of 1,000,000 insects equals the suffering of one human. Even when assigning insects this absurdly low moral weight, their suffering still dominates, as 10^18 insects comes out to 1 trillion human equivalents. If the number of insects were smaller, say around 7 billion, the consequences of not considering insect suffering might be acceptable. Unfortunately this isn’t the case, and as we shall see, ignoring insect suffering even if we assign a low probability to insect consciousness presents an unacceptably high risk of ignoring a catastrophic moral harm.
Potential Argumentum Ad Abusrdum
Before we continue, I’d like to address an argumentum ad absurdum my line of reasoning is subject to. Some may argue that my argument requires people to focus on alleviating suffering for increasingly strange categories. There are more bacteria than insects, so even if we give bacteria a small percent chance of sentience, we should focus on alleviating bacteria suffering. This can continue almost indefinitely, eventually generating the conclusion that we should focus all our resources on alleviating electron suffering. There are two options for avoiding such absurd conclusions, which I will explain by using the electron example as a test case.
There are 10^80 electrons in the universe. Because the number of electrons is so large, even if we assign an extremely small chance of sentience to electrons we ought to be taking measures to reduce their suffering. One option to this argument is to bite the bullet and accept the dominance of electron suffering. Some may opt for this option, but I find this conclusion absurd enough to discredit my argument. To salvage my argument, I must demonstrate that electron suffering doesn’t dominate all other considerations. The first argument against this is simply that the percent chance of electrons’ being sentient is low enough that their potential suffering doesn’t dominate. This argument works, but it requires that we assign an extremely low chance of sentience to electrons. Furthermore, if we think of sentience as a gradient, electrons are likely orders of magnitude less sentient than insects. I personally think this is the case, but many would object that assigning such a low probability to these events is inherently absurd. Our second argument doesn’t disprove the dominance of electron suffering, but instead addresses the practical difficulties of preventing electron suffering. In other words, we have no idea of what constitutes suffering to an electron. Arguments trying to extrapolate electron preferences from their physical behavior (like saying they prefer to be closer to protons) are analogous to saying that humans prefer to be closer to the earth rather than far away from it because of gravity. The amount of money and energy it would require to figure out potential sources of electron suffering are immense, and the chances of success low. Likewise, the resources it would take to prevent electron suffering would be exorbitant.
Analysis of a Potential Humane Pesticides Intervention
Of course, finding large sources of suffering is only one part of cause-prioritization; we must also make the case that there are effective interventions that can reduce said suffering. In this section, I will be focusing on the intervention specified in Brian Tomasik’s Humane Insecticides, which is the subsidization of less-painful pesticides. I will make the case that this intervention is the most marginally effective cause, scaling linearly per dollar spent and asymptotically per additional person working for it. Also, it is my understanding that Brian Tomasik thinks we shouldn’t advocate for humane pesticides, lest our arguments be taken as being against pesticides in general. I hope to dispel these concerns, making the case that the harm that may come from distortion of our views is less than the harm we fail to prevent by doing nothing.
As we have just seen, there is a strong argument that insects are sentient. We have also established that valuing one insect at 1/1,000,000 of the value of a single human still results in 1 trillion human equivalents at any given moment. If insects really are sentient, by doing nothing we are ignoring potentially the largest source of suffering that currently exists. Effective Altruism has lived up to its name by choosing cause areas that address the largest sources of suffering. Suffering in farmed animals dwarfs human suffering, and is itself dwarfed by potential suffering in the far future. One trend I’ve noticed is that as the size of the suffering increases, we accept more uncertainty in how much effect our actions will have, making the choice between focusing on animal advocacy vs. existential risk difficult despite ex-risk’s suffering space being much larger. Insect’s suffering space is much larger than farmed animals yet humane pesticides offers much less uncertainty about whether our donations have their intended effect.
Subsidization of humane pesticides is described in detail in Brian Tomasik’s Human Insecticides, and I recommend reading it for an in-depth explanation of this intervention. The intervention described involves simply paying farmers to use more humane pesticides. Brian estimates this intervention to cost one dollar per 250,000 less-painful deaths. For comparison, The Humane League prevents around 8.8 animals from existing on factory farms per dollar spent on advertising, and the Against Malaria Foundation prevents 1 child death per every $3,340 spent on malaria nets. To make comparing these numbers easier, let’s assume insects have a 1% chance of being sentient, with their degree of sentience being 1% of a human’s. With these values, humane pesticides saves 25 human equivalents from a more painful death per dollar. Depending on how we value a less painful death, subsidizing human pesticides is competitive (or even superior) with the most effective animal and human related interventions. This intervention would likely consist of subsidizing a switch from growth-regulators, pathogens, and parasites to neurotoxins, which entomologist Jeff Lockwood tentatively ranks as less painful than the former methods. I also recommend focusing on neurotoxins because of their standard effect across all species of insects. Other interventions, such as habitat destruction and use of predators, only affect targeted insect species. It is possible that most insects have lives that aren’t worth living could potentially make the situation worse. meaning the fewer insects in existence the better. I’m uncertain about whether this is the case, but arguments for this position raised enough doubt that I decided to account for this possibility. It also seems like most people focusing on wild animal suffering think this is the case, so I’ve decided to operate under this assumption. Pesticides that kill all species of insects rather than a targeted few are much more likely to reduce the overall insect population, which should produce a larger gain in utility.
Subsidization of humane pesticides is also the most efficient use of our human resources. First of all, the intervention described would require comparatively little bureaucratic support. A theoretical insect-charity would need to 1) locate farms using more painful pesticides, 2) distribute the subsidies to targeted farms, and 3) perform some kind of follow-up to ensure the humane pesticides are actually being used. Such an intervention would require an extremely low burden of proof, which would consist of the check that the subsidized pesticides are actually being used. Implementing this intervention would only require finding the farms and distributing the subsidies, which could be easily handled by a small team of people working full-time. Compared to other animal-charities, a charity doing the intervention described would need far fewer people to help a far greater amount of animals with much less ambiguity as to whether their efforts had the intended effect. (I think this paragraph can be slotted into the previous one - we’ll talk about this).
Furthermore, I think the flow-through effects of subsidizing humane pesticides could be extremely valuable. As I’ve hopefully driven home, insect suffering probably dominates all other sources of suffering. Despite this, I cannot name a single person doing something concrete to alleviate insect suffering. Entomologists such as Jeff Lockwood seem to be thinking seriously about whether insects suffer, but aren’t concentrating on interventions to prevent said suffering. The only people (or rather, person) who seems to be thinking seriously about this subject is Brian Tomasik. One person thinking about how to alleviate suffering for 10^18 organisms seems like a dangerously inadequate level of attention (though I’m not entirely surprised by this; it’s a weird topic). Promoting humane insecticides not only has the direct effect of giving insects more humane deaths, it also helps normalize concern for insect and wild animal suffering. More importantly, it helps normalize concern for wild animals in accordance with EAA ideals. Aside from Animal Ethics, the only charities that come close to being about wild animals are all focused on habitat preservation, which many EAAs think is the exact opposite of what will decrease wild animal suffering. Another important flow-through effect is that such an intervention would help us gain more knowledge on how switches from one pesticide to another affect insect population levels. From my limited understanding, most research along these lines seems to only measure the populations of targeted bugs. Subsidies would provide more opportunities to measure the effects of different pesticides, helping us develop a better understanding of what the most humane method really is.
As a last note, I’d like to address concerns that publicly supporting such an intervention would be misunderstood as being against all pesticide use. Currently, effective altruist views on wild animal suffering have zero traction among the general public. The only people who are publicly concerning themselves with the issue are focusing on habitat preservation, which might be the worst possible intervention to decrease wild animal suffering. I think we shouldn’t worry about making the problem worse because the opinions exactly opposite ours already dominate all public conversation on the topic. We are already in the worst-case-scenario. Furthermore, I think that if we are careful about how we market ourselves, we can avoid misinterpretation. Explicitly stating ‘We are not against habitat destruction. We are not against pesticide use. We simply want animals to die less painful deaths’ seems hard to misconstrue. A mission statement along these lines seems more agreeable to the general public than the current rhetoric around pesticides. I’m not sure of the exact opinion polls on this, but I would imagine the general public is more in favor of pesticide use than not, and stating that we actually support some pesticide use is amenable message than hard line anti-pesticide advocacy.
In conclusion, even if we assign a low probably of sentience to insects, their potential suffering dominates most other sources. Switching to humane pesticides saves an enormous number of insects from painful deaths, has a low burden of proof to demonstrate effectiveness, and helps normalize concern towards wild animal suffering. More importantly, publicly advocating for humane pesticides provides an essential counterpoint to other organizations focused on wild animal suffering. I’m not fully convinced that most wild animals have lives not worth living, but I am convinced that the standard anti-pesticide, pro habitat preservation stance needs to be seriously questioned. Subsidizing humane pesticides takes us from zero to one on wild animal suffering, making it the best use of marginal resources.
I want to respond to a small but salient part of this post, its treatment of moral uncertainty. I first heard this observation from Carl Shulman, though I believe that both Brian and Buck are familiar with it.
You write:
Compare:
Needless to say, many thoughtful people endorse multiplicative factor considerably larger than 10^9 for the ratio of human and insect moral relevance, and assigning their view an infinitesimal probability seems to require overconfidence. (10^8 is about the ratio between [number of synapses in human brain] and [number of synapses in insect brain]).
As you might guess from their differing conclusions, neither of these arguments work. This is not an appropriate way to reason under any kind of uncertainty, either moral or empirical.
Setting aside the technical issues, the simplest reason that I reach a different conclusion:
Thanks, Paul. :) It's worth noting that if the uncertainty is regarded as wholly factual (which is unrealistic, of course), then the insect wager does work, since then there's only a single unit of moral value in play. For example, if the uncertainty were only a 0.1% chance that insects in fact possess specific neural structures that I know for certain I care about equally with the brains of humans, then the argument works. This sort of reasoning would be most applicable to, say, those property dualists who think there's a single, ontological property of consciousness that's either present or not in a binary way, since then there's little moral uncertainty (consciousness matters and unconsciousness doesn't) but still lots of factual uncertainty (is ontologically primitive phenomenal consciousness present or not?).
I'm a little unclear about what this means.
I think what you are saying is something like: if I'm certain that it's the number of neurons which matter morally, but I'm just uncertain how many neurons this organism has, then a 2 envelope problem doesn't apply.
But why can't I leverage that trick to avoid the 2 envelope problem in general? I'm confident that utilons are what matter morally, but I'm uncertain how many utilons an insect has.
The problem to me doesn't seem to be "factual" versus "moral", but rather if there are some nonarbitrary units.
Yeah, that's a cartoon version of what I was saying.
The problem is that "No, Virginia, there really is no such thing as a utilon in any non-arbitrary sense. Happiness and suffering are not actually cardinal numbers that live in the physics of the universe that we can measure. Rather, we use numbers to express how much we care about an experience."
If moral realism were true and the moral truth were utilitarianism, then I suppose there would be a "right answer" for how many utilons a given system possessed (up to positive affine transformation). But I don't take moral realism seriously. In the moral-realism case, the two-envelopes problem for moral uncertainty would be analogous to the ordinary two-envelopes problem, where it's also the case that there's a right answer. For moral non-realists, the two-envelopes problem is just a way to describe the paradox that calculations over moral uncertainty depend on one's unit of measurement.
Okay, I think I understand now, thanks for the explanation.
For what it's worth, the "factual" versus "moral" contrast you are drawing seems to me to be a distinction without a difference. Both the moral realist and the moral non-realist are looking for "nonarbitrary" units of measurement, and an argument that a certain unit was nonarbitrary seems like it would probably be persuasive to both the realist and the non-realist.
Well, the moral realist just assumes there exist non-arbitrary units by faith, since moral realism implies non-arbitrariness. The non-realist believes no such thing. :)
Under some assumptions, that might be correct, but not in the general case. In the general case, the degree of sentience of a human compared to an insect could plausibly be from 1 to 10^9+.
To push through the reasoning about the wager, you need an extra assumption - that none of your probability mass has the human being much more important (or sentient) than the insect. You need to enforce an upper-bound ratio between importance of humans compared to insects. And that's not specific to the moral or empirical question, unless you argue from other philosophical assumptions like certain types of property dualism, but you ought to also apply some uncertainty over those things.
Yeah. My point was that the "two envelopes problem for moral uncertainty" doesn't apply to factual uncertainty. But certainly what conclusion results from applying factual uncertainty depends on one's priors.
I agree that the two-envelope problem doesn't apply to factual uncertainty. The whole thing gets significantly complicated by situations where it's hard to specify how much of the uncertainty is moral and how much is factual.
I regard getting some good and usable answers to this as an important research topic within philosophy. Given that questions of pure moral uncertainty are still unresolved, you'd probably need to make assumptions about that half of things in order to get going.
Some links to discussions of this.
I agree that the reasoning based on epsilon chance of a high multiplier is suspicious, for the reasons that you say. However, I think the pro-insect version is more likely to work out than the pro-human version. In the pro-insect version, you're taking the moral value of humans as fixed, and taking the expectation of the moral value of insects, whereas in the pro-human version, you're taking the moral value of insects as fixed, and taking the expectation of the moral value of humans. My uncertainty seems a lot closer to the first: I have a pretty good handle on what being a human is like, and how much I care about humans, but I'm not so sure about insects.
I don't think you can avoid the problem that way, since it's logically equivalent to re-cast your statement in terms of insect value units?
I think you at least get an edge on the problem from this (related to the point you make about factual uncertainty -- familiarity with humans means we have more factual uncertainty about insect experience). But I wouldn't want to rest the whole of my argument on this edge.
I agree that "I assign probability p that humans are 10^3 times as important as insects and probability q that humans are 10^10 times as important as insects" is logically equivalent to "I assign probability p that insects are 10^-3 times as important as humans and probability q that insects are 10^-10 times as important as humans". However, when people turn these into the argument that "you should care about insects/humans much more than about humans/insects", they're implicitly doing an expected utility calculation.
In one version of the calculation, you're saying that insect lives have utility 1, and human lives have utility 10^3 with probability p and 10^10 with probability q. When you take the expectation (for certain values of p and q), you end up with the human population being more important than the insect population.
In another version of the calculation, you're saying that human lives have utility 1, and insect lives have utility 10^-3 with probability p and 10^-10 with probability q. Then, when you take the expectation (for values of p and q compatible with the pro-human conclusion above), you end up with the insect population being more important than the human population.
My point is that although both these probability assignments are compatible with the description "humans are 10^3 times as important with probability p and 10^10 times as important with probability q", they are not equivalent distributions over utilities. Furthermore, I think my probability distribution looks more like the second than the first.