Context and summary: Many of the strategies aimed at helping animals, such as advocacy towards considering the suffering of mammals farmed for their meat, seems to be contributing to the growth of the farming of smaller animals, which arguably leads to more animal suffering overall since the smaller the animals, the more need to be farmed and killed to produce similar quantities of food. This is one version of what has been called the small animal replacement problem (SARP). Still, many of the animal advocates aware of negative externalities of this sort endorse the strategies that contribute to them. This can be rationalized with the “temporary setback” view, according to which these strategies also have long-term positive externalities (like expanding people’s moral circle) that offset the negative ones. This post is an attempt at formalizing this view and clarifying its premises. Uncovering these premises makes defending the "temporary setback" view seem much harder than it might at first.
 

The most compelling version of the “temporary setback” view is that even if some strategies aimed at helping animals increase overall suffering in the near future because of the small animal replacement problem, they still decrease it in the long run by improving our society’s values. Consider the following hypothetical scenario.

Sentientia – a post-SARP better world: After a very long period where the farming of fish, shrimps, and insects was a major economic sector (especially following the global ban of the farming of most land animals, which animal advocates and other actors pushed for), it is finally collapsing. Some countries have made the farming of (likely) sentient animals (including insects) illegal. The others have strict and enforced animal-welfare norms that make the lives of these small animals not nearly as bad as they used to be. Regardless, the number of farmed fish, shrimps, and insects in these countries has drastically reduced as humans and their pets now eat almost exclusively plant-based food. While experts in ethics and animal welfare were initially overwhelmingly worried about the suffering of wild animals, many more of which were incidentally brought into existence after the ban of mammal farming (as forests replaced crops grown to feed animals and their grazing space), humanity – with the help of AI – deeply modified ecosystems to make diseases, starvation, and painful juvenile deaths all extremely rare. Wild habitats are now sanctuaries where animals are taken care of when needed. Some people oppose this and believe that humans should not intervene in nature, but they are a minority with negligible influence. There are also treaties aimed at preventing the replication of non-human animals off-Earth. If humanity ever colonizes space, it will do so without spreading animal farming or wild animal suffering with them. Overall, despite initial setbacks like SARP, there now is significantly less total animal suffering that there was at the beginning of the 21st century (or than in the counterfactual scenario where animal advocates avoided strategies contributing to SARP), and things will remain this way for eons.

Many of the strategies incidentally contributing to SARP appear as indispensable steps toward a world comparable to Sentientia. While this may seem like enough to endorse the “temporary setback” view, it isn’t. The problem is that we are not facing a mere Sentientia vs status quo dilemma (in which case there would be no doubt that anything that makes Sentientia more likely to be reached is worth pursuing). There is a third set of possibilities. For the sake of illustration, consider the following example.

Reversomelas – a post-SARP worse world: Now that most countries have at least severely restricted the farming of most land animals, the farming of fish, shrimps, crickets, and other small sentient creatures is blossoming. While there are more vegetarians abstaining from eating any animal that there ever was, this does not go anywhere near making up for the fact that the total number of animals farmed and killed for food went through the roof. Surveys show that most people believe that the small animals they eat are sentient and should be farmed in better conditions or maybe even not farmed at all, but rare of those who care enough to do anything about it. The immense growth of the sector has made these small animals a very cheap source of protein and other essential nutrients. Plant-based alternatives are often more expensive and/or considered less tasty. Moreover, most of the products resulting from the small-animal farming actually end up in pet food.^[1] And there are many more pets in the world than there used to be due to human population growth, and people becoming more likely to adopt companion animals as they get richer and more compassionate toward the non-human animals that resemble them the most. Even if all humans went fully vegan, there would still be vastly more animals farmed and killed than there were at the beginning of the 21st century. The rare people advocating for increased shrimp and insect welfare, while there are still many abandoned dogs and cats, wild animal species going extinct, and humans living in poverty, are given incredulous stares. These advocates still are able to celebrate incremental changes in the conditions in which fish and other animals are raised, but those changes will never go anywhere near compensating for the massive growth of the industry in terms of total animal suffering. The science of wild animal welfare has grown much slower than the science of rewilding and wild habitat conservation. Occasionally, welfare biologists manage to get enough funding to vaccinate some wild animal populations against some painful viruses, but life in nature mostly remains the same from the perspective of wild animals, and there are now many more of them (due to the land freed by the decline in mammal farming). The vast majority of people, including most animal advocates, believe that the world has gotten much better for non-human animals. There are many more vegans. Many horrible farm and slaughterhouse practices have been made illegal. People generally seem much more altruistic toward other animals. However, the world has gotten much worse in terms of overall animal suffering. Furthermore, the situation worsens as humans start colonizing space, building permanent settlements outside of Earth where they start farming small fish, shrimps, and insects. Eventually, after eons have passed, things finally end up improving for non-human animals, but nowhere near the point where total animal suffering is lower than it was at the beginning of the 21st century (or than in the counterfactual scenario where animal advocates avoided strategies contributing to SARP).

While many of the strategies incidentally contributing to SARP arguably constitute a necessary step to reach a world like Sentientia, this step has a cost, namely, the risk of ending in a world comparable to Reversomelas. Hence, holding the “temporary setback” view necessitates explicitly arguing that the benefit of these strategies outweighs this cost. This, in turn, requires defending that the improved societal values (and/or other positive flow-through effects) resulting from strategies contributing to SARP will

1. counterfactually lead humanity to either
   1. ban or substantially decrease the farming of small animals to the point where it eliminates SARP, or
   2. drastically improve the conditions in which they are farmed to the point where this makes up for SARP, or
   3. better act in the interests of other beings (e.g., wild animals and potential digital minds) to the point where this morally compensates for the harm caused to small farmed animals; or
   4. do some combination between almost-A, almost-B, and almost-C that overall makes up for SARP; and
2. also outweigh the potential negative externalities other than SARP^[2]. 
    

I'm curious whether people agree with all this and what their cruxes for or against 1 and 2 are.

1. ^{^}

   In our present world, this already seems to be the case for insects: "Currently, pet food is the largest market for insect proteins" (deJonk & Nikolik 2021)

2. ^{^}

   E.g., the wild animal replacement problem (see Tomasik 2019; Shulman 2013)-- which is also alluded to in my descriptions of Sentientia and Reversomelas -- and backfire risks from moral circle expansion (see, e.g., Vinding 2018).