Opinions expressed here are solely my own and do not express the views or opinions of my employer.

Summary

Some wild animal welfare (WAW) advocates promote research of WAW in academia, hoping that this will help us to lobby governments (which have the most control over nature) to pursue WAW interventions, and to know what interventions to lobby for. In this article, I first explain the theory of change in more detail, and list changes that governments could do if lobbied successfully. Then I discuss potential obstacles to influencing governments:

It’s likely that the welfare of small but very numerous wild animals dominate WAW considerations. But it’s difficult to imagine governments caring about the well-being of small uncharismatic animals like insects or fish in a scope-sensitive way and taking major actions to improve their welfare.
- Counterarguments: WAW is huge in scale even if it only focuses on vertebrates. Also, it could have been similarly difficult to imagine governments funding species conservation efforts until it happened.
The consequences of interventions on the welfare of all affected wild animals seem nearly impossible to determine, even with a lot of research. Also, research in one ecosystem might not generalize to other ecosystems. Hence, we are unlikely to reach solid conclusions on which WAW interventions are good.
- Counterargument: This is the same as the concern of cluelessness that applies to all causes. Hence, it’s unclear whether this is a legitimate argument to prioritize other causes over WAW.
  - However, in my experience, contrary to other causes, people seem to notice that we are clueless about WAW impacts right away, which might make influencing governments more difficult.
Even if we determine consequences, people with different moral views might disagree on which consequences they prefer. For example, people may disagree on how to weigh various positive and negative experiences, how to weigh the welfare of different animal species, etc. This may eventually divide the WAW movement into many camps and hurt overall efforts.

I know very little about the history of other social movements, so I don’t know how important these obstacles are.

Finally, I provide some thoughts on what should be prioritized within WAW.

Context

I started writing this article as a criticism of some of the current WAW movement-building efforts. Compared to longtermist causes, influencing governments to care about WAW to me seemed similarly speculative but much less important. It also didn’t have the certainty of impact of some farmed animal welfare interventions. My main concern was that influencing governments to care about WAW seems very difficult to achieve. However, comments and counterarguments from reviewers made me less skeptical about it. So in this article I now just talk about this approach as a potential avenue to improve WAW, and leave it to readers to form opinions on how promising it is.

Current theory of change

Here is the current theory of change for WAW in the medium term, according to some WAW proponents I talked to:

Consider the following vision of the future. WAW is an established academic field, with many scientists working on determining the animal welfare consequences of various actions. This research informs us on how to help wild animals and also legitimizes caring for WAW. Governmental organizations, the effective altruism (EA) movement, and possibly environmental organizations use the conclusions of this research to guide their actions to improve WAW, much like how various entities currently use research to reduce their environmental footprint and to protect endangered species. This potential future could help wild animals much more than any direct intervention we could pursue today. Hence, we should prioritize activities that would make it more likely that this vision becomes reality.

Some WAW proponents think that it’s most important to influence governments as they have much more say over what happens to wild animals than charities and companies. Because of this, WAW work by EA groups and conservationists might also often involve lobbying governments, with some exceptions (like paying farmers to use more humane insecticides). Hence, I will largely focus on influencing governments in the rest of this article.

Why not just pursue WAW interventions more directly?

Most WAW researchers I talked to thought that in the immediate future, we are unlikely to find WAW interventions that would be competitive with farmed animal welfare interventions in terms of direct short-term cost-effectiveness. After spending some months trying to find such interventions myself,^[1] I tentatively agree. I think that the lives of farmed animals are easier to improve than the lives of wild animals partly because all aspects of farmed animals’ lives (including their genetics) are in human control. This makes it easier to make changes, and to observe the consequences of these changes.

But if we look at a slightly longer timeframe (25-300 years), then perhaps WAW could be more promising than farmed animal welfare because (1) the ultimate scale of possible WAW improvements is bigger as wild animal populations are much bigger, and (2) we might run out of low-hanging fruit in farmed animal advocacy. Since it seems that EA will have more money in the future, it might make sense to pursue both approaches.

What could influencing governments achieve?

Below is a list of interventions that could perhaps directly have a significant positive impact on WAW.

Here are some more ideas suggested by various people of what governments could do:

Vaccinate wild animals against the most painful diseases.
Control wild animal populations in more humane ways (e.g., use humane insecticides, use sterilants instead of slow-acting poisons, ban glue traps for rodents, rodent birth control, pigeon birth control).
Alleviate resource scarcity with contraceptives.
Give antiparasitic medicine to wild animals.^[2]
Lobby the United States Environmental Protection Agency to incorporate welfare assessments into project approvals.^[3]
Have stricter regulations for contaminants that cause wild animal suffering. For example, we could require reduced levels of:
- Lead (perhaps by banning lead in fishing gear) and other heavy metals.
- Industrial chemicals such as PAHs and PCBs (Hontela et al. (1992) suggest that together with mercury, they might cause constant stress to fish).
- Improperly disposed medications.
- Ocean litter (abandoned fishing gear, plastics).
- Oil spills.
Reduce aquatic noise pollution from ships and other sources.
Reduce light pollution, perhaps by dimming artificial lights at night as was done in Germany.
Ban fireworks as they scare animals and cause pollution.
Help animals affected by weather events.
Incentivize things like more natural forage for bees and other insects, bird-safe glass, and mesh in sticky traps.
Eradicate or reduce populations of invasive species that suffer a lot (e.g., have very many offspring) or otherwise create a lot of suffering (e.g., parasites like screwworms).
Protect and increase populations of large herbivores as they probably decrease the populations of small animals, which seems good for utilitarians who think that those small animals have net-negative lives (i.e., on average experience more suffering than happiness).
Actively create conditions that foster welfare. For example, we often try to rehabilitate areas damaged by humans to support life. If we are already trying to change this modified ecosystem to something better, why not include welfare considerations when determining what to change it to?^[3] It could include things like planting plants that lead to higher-welfare ecosystems, although it might be too difficult to determine what such plants would be.
Perhaps in more distant future, gene drives or similar methods could be used to: increase pain thresholds, make animals have fewer offspring per parent, or otherwise improve animals’ lives. See Tomasik (2016) for arguments about why using gene drives for such purposes might be difficult.

Note that for all the WAW intervention ideas listed above, it’s unclear if even their short-term impact on WAW would be good or bad. For example, decreasing pollution would likely increase the average welfare of wild animals, but it would also likely increase overall wild animal populations, which could be bad for WAW if most of these wild animals have lives that are worse than non-existence. I discuss such uncertainties more in the section on cluelessness.

Some more intervention ideas are discussed in Tomasik (2013). Many of them involve reducing overall wild animal numbers in some way, which could decrease suffering. Some people still associate the WAW movement with the view that the goal should be to reduce wild animal numbers. However, it seems that few (if any) of the people who actually work in WAW organizations hold this view, partly due to the uncertainty of whether there is more happiness than suffering in nature and non-utilitarian values.

This list is not exhaustive; I may have forgotten some interventions that were already proposed. Also, since the WAW movement is in its infancy, we may think of other interventions. Note that the list excludes interventions that might affect fewer animals but might be useful for WAW movement building, as I want to understand the ultimate goals.

In my post about WAW in the far future, I argue that affecting future scenarios where humans spread wildlife beyond Earth is more important than improving WAW on Earth with interventions like the ones listed above (e.g., see this Guesstimate model). But if we make governments pursue such interventions and normalize trying to improve WAW, it seems more likely that WAW concerns will be taken into account when spreading wildlife beyond Earth too. However, since influencing governments might be difficult (due to reasons I discuss below), it’s unclear if trying to do it is among the best ways to affect wildlife-spreading scenarios. I list other ways to affect wildlife-spreading scenarios here.

Obstacles to influencing governments and academia

Governments might not care about small animals

According to Bar-On, Phillips, and Milo (2018)’s estimates, for every vertebrate, there are very roughly 100,000 arthropods. Most of the arthropods are small animals like insects, mites, spiders, ticks, copepods, etc. There is a significant probability that some of these small invertebrates are sentient. If at least some of them are, their welfare might be by far the most important thing within WAW. Without properly taking into account small invertebrates’ welfare, any analysis of WAW interventions might be very incomplete, and hence likely incorrect.

But it’s difficult to imagine people starting to care about small wild invertebrate welfare in a scope-sensitive way. We also don’t seem to currently know what is good for the welfare of most small invertebrates, even on the level of an individual.

These same arguments could also be made (to a lesser degree) for small uncharismatic vertebrates like small fish, rodents, etc.

Counterargument: WAW is very important even if we only help vertebrates

According to Bar-On, Phillips, and Milo (2018)’s estimates, most of the wild vertebrates in the world are fishes. There are very roughly 7,000 wild fishes for every vertebrate farmed for food (most of whom are also fishes).^[4] If we assume that farmed vertebrates on average suffer 7 times more than wild fishes (because of usually worse conditions and possibly higher moral weight for some species), then an intervention that would reduce the suffering of 1% of wild fishes by 10% for one year^[5] would be as good as getting rid of factory farming of all vertebrates for one year (ignoring all the indirect effects). It’s somewhat unclear to me which of these tasks is harder. Hence, it could make sense to pursue both.

Difficulty in determining consequences (cluelessness)

Whenever you change one thing in an ecosystem, many other things also change (see this video as an example). Determining in advance what all these changes will be, and whether they will be good or bad for WAW, seems nearly impossible.

An ecologist once told me that it’s still very difficult to predict how population sizes of animals will change given an intervention, despite there being relatively a lot of research on this. Predicting how overall welfare will change is likely to be many times more difficult, as it introduces many new uncertainties like:

Which wild animal experiences will be more prevalent?
How should we weigh the different experiences against each other?
How should we weigh the interests of animals from different species against each other?

And even if we find out the answers for one ecosystem, the consequences of the same intervention in another place might be different. Academia doesn’t tend to produce conclusive takeaways for problems that are this messy. This in turn might make it much harder to use that academic research to influence governments.

This complexity can even apply for making population control methods more humane. For example, chemical insecticides may have different effects on non-target species compared to insecticides that introduce predators, parasitoids, or pathogens to suppress insect populations.

Counterargument: We are clueless about the effects of all interventions

However, this is the same cluelessness concern that applies to all interventions (e.g., see this talk by Greaves). In fact, WAW is one of the reasons for cluelessness for most interventions because most changes in the world affect WAW in complex ways.^[6] Hence, it might be unfair to deprioritize WAW because of cluelessness when the same concern applies to all interventions. I am unsure whether it would make sense to argue that we are more clueless about the effects of WAW interventions than other interventions.^[7]

If we found WAW interventions that are also cost-effective in the effects we can predict, then perhaps we should pursue them, assuming that all indirect effects cancel out (just like we implicitly do with farmed animal welfare interventions when we don’t consider their impact on WAW).^[6] It doesn’t feel satisfactory but the alternative might be never doing anything.

However, in my experience, people notice that we are clueless about WAW interventions right away, and can easily think of scenarios illustrating how it might go wrong. I haven’t noticed this that much with other interventions. This might make it more difficult to convince governments and others to pursue WAW interventions. But it’s possible that people quickly notice that we are clueless about WAW only when talking about WAW interventions in the abstract, but not so much when talking about concrete WAW interventions.

Overcoming moral disagreements

Even if we somehow figured out all the consequences of WAW interventions, people may still disagree about what to do. It seems to me that when thinking about WAW, even small differences in values may lead to large differences in opinions of what should be done.

Here are potential disagreements I am worried about:

Often we would have to trade off the welfare of different species against each other, and people may disagree about what moral weights to give to animals of various species and ages.
Interventions that would make the biggest difference will likely involve certain individual animals being worse off, which some people might find unacceptable, if they think that helping one animal doesn’t compensate for hurting another.
In some cases, there may be disagreements about how to weigh short but very intense suffering against longer but less intense suffering.
Many interventions that improve the welfare of an average wild animal also increase overall wild animal populations (e.g., reducing pollution). But unless the average welfare improvement is very significant, increasing overall populations might seem bad for negative and classical utilitarians, who think that there is much more suffering than happiness in wild nature.
Many people seem to believe that there is intrinsic value in nature being the way it was before the most recent human changes to the environment. This would eliminate many potential interventions that tackle natural wild animal suffering. Perhaps WAW advocates could change this view for some people, but I don’t think they could change it for everyone.
Interventions may also need to be good or neutral from environmental and conservationist perspectives, as such views are important to many people and we need allies not enemies, especially if we want to influence governments. But such a requirement would further decrease the number of viable interventions.

The above are just moral views I am familiar with. This thread and some of the talks here discuss some more views that also support caring in WAW. But again, the question is whether these moral views converge on which concrete actions should be taken for WAW.^[8] Currently, everyone in the WAW movement agrees that something should be done about WAW and that more research is needed, but I worry that it might get more complicated once we are talking about concrete large-scale interventions. Such disagreements might make it difficult to agree on what interventions to pursue and result in no united front.

One reviewer of this text pointed out that “many movements start without a united front and are wildly unpopular, yet succeed in (relative) short amounts of time” (e.g., the US civil rights movement, the anti-slavery movement, and the women’s suffrage movement).^[9] I know very little about these social movements, but I imagine that in these movements, people at least mostly agreed on what changes they want to achieve. In my WAW intervention search project, it was difficult to find interventions that would be uncontroversially good (or at least neutral) from all perspectives mentioned above. I don’t think that many other movements would have such a problem.

I am not alone in thinking that this is a significant problem. Wild Animal Initiative also claims that “uncertainty like this about fundamental ethics is, in my view, the greatest roadblock to large-scale intervention for wild animal welfare.” Hence, the article proposes that “welfare models should be as philosophically open-ended as possible.” However, this would only postpone the more fundamental question of how to choose which moral view to use in the model.

All that said, it is possible that one way of thinking about WAW will come to dominate which might make worries about no united front less important.

Also, perhaps we don’t need to decide which moral views should be used. We can just try to make it more common to discuss WAW under any moral view and try to make public and academic discourse about WAW more sophisticated and less short-sighted. It’s difficult to evaluate the value of this, as it’s unclear what it might lead to and even what moral view to use to evaluate it. For example, how good this would be from a utilitarian perspective largely depends on how utilitarian we expect the outcome to be.

Appendix: Thoughts on what should be prioritized within WAW

According to Bar-On, Phillips, and Milo (2018), there are 500 marine arthropods for every terrestrial arthropod. It also seems that fish might significantly outnumber all land vertebrates. Hence, it could make sense to direct more of our WAW efforts to the ocean.
Perhaps we should research which interventions we expect to ultimately help animals the most in the longer term. For example, if we think the best long-term intervention is preventing diseases, then it could be useful to start building expertise in this field and start a charity that specializes in this area. But before starting such a charity, ideally we should already have in mind what exactly we are building towards (e.g., what kind of diseases we would want to tackle in the longer term).
The two points above assume that the most important consequences of current WAW work are on wild animals on Earth. But in this article, I argued that affecting wildlife-spreading beyond Earth is much more important. Hence, perhaps we should instead prioritize spreading values that would positively impact WAW in wildlife-spreading scenarios.
The conclusions of WAW research may be too complicated and nuanced for voters and politicians to understand compared to something like “reduce carbon emissions” or “free farmed animals from cages.” Perhaps this could be mitigated by developing rules of thumb within WAW like “increase large herbivore populations.”
Since most wild animal are small invertebrates, perhaps first we should try to promote research into small invertebrate welfare in academia and elsewhere. It might begin with promoting more humane treatment of invertebrates used by humans, perhaps starting with small invertebrates used for experiments and insects farmed for animal feed (as this industry predicted to grow exponentially). But such work would also increase the ceiling of impact of WAW work too.

References

Bar-On, Y. M., Phillips, R., & Milo, R. (2018). The biomass distribution on Earth. Proceedings of the National Academy of Sciences, 115(25), 6506-6511.

MacAskill, M., Bykvist, K., & Ord, T. (2020). Moral uncertainty (p. 240). Oxford University Press.

Tomasik, B. (2013). Applied Welfare Biology and Why Wild-Animal Advocates Should Focus on Not Spreading Nature

Credits

This article was written by Saulius Šimčikas in a personal capacity. Opinions expressed are solely my own and do not express the views or opinions of my employer or reviewers.

Thanks to Holly Elmore, Jacob Peacock, Marcus A. Davis, Michael St. Jules, Neil Dullaghan. Oscar Horta, Willem Sleegers, William McAuliffe for reviewing drafts of this post and making valuable comments. Thanks to Cameron Meyer Shorb, Max Carpendale, and others for conversations that informed the post. All mistakes are my own.

Footnotes

^
The only article I published about this search is Reducing aquatic noise as a wild animal welfare intervention. My investigations into other interventions were much more shallow, mostly because I couldn’t find how to estimate their cost-effectiveness.
^
Antiparasitic medicine has been suggested in passing here, but this article is no longer online, not sure why.
^
The idea is taken from this talk (~29:00).
^
There are roughly 140 billion (1.4*10^11) vertebrates farmed for food (100 billion of which are fishes), and very very roughly 1,000,000 billion (10^15) wild fishes.
^
I don’t know what such an intervention for wild fishes could be, but I can imagine it being done by reducing diseases, switching which fish species humans fish, reducing some pollutants, or perhaps even reducing aquatic noise (although my impression is that reducing noise wouldn’t have an impact of this scale).
^
For example, animal farming uses ~27% of all land (and 38.5% of habitable land), and hence has a huge impact on wild animal suffering. As it was argued here and here, the most important short-term consequences of reducing meat consumption may be to wild rather than farmed animals. And we don’t know if these consequences are positive or negative for WAW. Even farmed animal welfare reforms likely have significant WAW consequences.

Similarly, reducing human poverty or deaths may increase consumption and that could in turn decrease wild animal populations by decreasing their habitats. Or it may not; this is all very speculative. But that is the point: we don’t know all the consequences of our actions.
^
I feel that perhaps the cluelessness concern applies to WAW interventions more than to some other short-termist interventions. For example:
- I imagine that most people who support global poverty interventions think that animals don’t matter much morally, and believe that their preferred interventions lead to economic growth which they believe to be good from a long-term perspective. Under such views, global poverty interventions might seem quite robustly positive.
- If you take an intervention like farmed chicken welfare reforms, we at least know that they have a huge proximate impact. Difficult-to-predict second-order effects (listed here) of these reforms also seem very important but not obviously more important than direct effects, contrary to most WAW interventions I looked at. For most WAW interventions I considered (other than some interventions lobbying for more humane population control), it seemed that difficult-to-predict second-order effects are obviously much more important than the proximate impact. Although I think they are easier to predict — partly because I know more about chicken welfare reforms than about WAW interventions because I studied them more, and because welfare reforms are already happening. Future interventions like gene drives could have an even bigger proximate impact. But I’m not sure this would hold if I went beyond second-order effects and looked at all the ripple effects, like effects on artificial sentience.
^
Perhaps we could use research like MacAskill et al. (2020) to try to make decisions under moral uncertainty, giving some weight to various positions. It would still require everyone to agree to do it and on how to do it. This would also reduce what the WAW movement can achieve from the perspective of any one moral view. Additionally, I imagine that this approach might be difficult to communicate to the general public and lawmakers, so it might only be useful for internal decision making.
^
Also, corporate animal welfare campaigns don’t seem to be majorly slowed down by some animal rights advocates being opposed to them.

Lobbying governments to improve wild animal welfare