Animal Welfare and Existential Risk: A Thought Experiment

This post regards a thought experiment about the nexus of animal welfare and existential risk that I've been thinking about on and off for a long time. It's written specifically about climate change because that's been a fixation of mine for more than a decade, but climate change can be rhetorically replaced with your existential risk of choice without altering the overall calculus all that much.

I know that this proposition sounds insane at first blush, but could decisive climate action actually be disastrous for animals?

Let me explain.

For some brief context, starting in 2018, I was getting increasingly freaked out by the progressively dire IPCC reports that confirmed climate breakdown as a pressingly existential risk to civilization. Then, in 2023 (March 29, to be precise), I watched an ostensibly comedic video (primarily) about factory farming (warning for those who would like one: brief on-and-off meat imagery from 4:08 to 5:24). Subsequent reading confirmed for me the awful truth: Humans, through animal agriculture, torture billions upon billions of sentient creatures each year. In a grim way, this realization actually improved my mental health, in that it made me way less panicked about climate change, since the end of human civilization (which I know is now considered an unlikely outcome of climate change; I get to that later) no longer seemed so bad.

In a vacuum, I want human civilization to endure. I want to see what we can accomplish, especially in an era of rapid technological advancement. But what I learned in 2023 led me to the inescapable conclusion that human civilization is worth saving only to the extent that it has the potential to become less evil — and that civilizational collapse would be preferable to the indefinite continuation of human civilization as it currently exists. Any less extreme conclusion would require motivated reasoning.

Climate change, as perhaps the biggest existential risk to civilization other than nuclear weapons (although the two are related, as I touch on a bit later), thus carries an enormous potential upside, in that if it does succeed in destroying civilization, it would spare many billion and potentially trillions of sentient creatures lives of extreme suffering, depending on how factory farming changes and doesn't change in the counterfactual in which climate change is curbed, since a post-collapse human species would be incapable of operating factory farms at anywhere near the scale of what exists today.

At this point, I need to draw a distinction between suffering and extreme suffering (i.e., when inflicted by another, torture). Climate change will undoubtedly make life worse for billions of wild animals, as bodies honed by millions of years of evolution to thrive in specific habitats fail to cope with rapid change. But this level of suffering isn't really any worse than that which has befallen animals pretty much since the dawn of carnivorous life — after all, for a predator to avoid a slow death by starvation, many prey animals have to die rather brutally. The kinds of wild-animal suffering that can even hold a candle to factory farming in terms of intensity are over relatively quickly. Even a slow death by starvation, perhaps the worst fate regularly suffered in the wild, is over within weeks, compared to the months or even years for which animals are tortured on factory farms. From an extreme-suffering perspective, wild-animal suffering is so insignificant compared to factory farming that I don't think it should even factor into ethics analyses.

Still, it's far from certain — and it may not even be likely — that addressing climate change will increase extreme suffering. Most climatologists now believe that climate change is unlikely to outright destroy civilization. It certainly could — we don't know exactly how tipping points will work, and it's entirely plausible that, say, thawing the Siberian permafrost will release enough greenhouse gas to trigger several other tipping points and ultimately cause a level of warming incompatible with human civilization, not to mention that conflicts between superpowers over dwindling resources could easily tip into nuclear war — but I'll defer to the experts who say it's unlikely. 

The more likely result of our current climate trajectory, that human civilization is strained and weakened but ultimately endures, is still likely to mean less-widespread extreme suffering, since decreased agricultural yields would probably make it impossible to sustain the world's current level of meat production, but the level of individual suffering on factory farms is unlikely to decrease — most humans don't care about factory farming now (or at least don't care enough to change their behavior), so it's hard to see how they will when their own lives are far more unstable and more of a struggle to survive.

The biggest wild card here is the scenario in which climate change is adequately addressed.  I imagine that factory farming would become less bad in this timeline as well, due to some combination of increased moral opposition and technological improvement. In the best-case scenario (barring an enormous shift in societal attitudes toward animal cruelty akin to the shifts that saw humans finally turn against slavery), lab-grown meat (along with lab-grown animal by-products) becomes so economical that factory farming is made obsolete.

Ultimately, this thought experiment comes down to three variables (albeit each with many sub-variables): x, the probability that climate change will destroy civilization; y, the extent to which factory farming will decline if climate change is left unabated (but civilization survives); and z, the extent to which factory farming will decline if climate change is decisively addressed (this oversimplifies by treating climate action as a binary rather than a range of possibilities, but more variables can be added to represent in-between climate outcomes without fundamentally changing the overall analysis). To avoid the conclusion that climate action will increase extreme suffering, z must be greater than y. If x is low, z need be only a little greater than y. The greater that x is, the greater the gap between y and z necessary to support the conclusion that climate action will decrease extreme suffering.

This is where my analysis of all of this has kind of come to an end. I don't know how to estimate any of the three variables, and there seems to be very little written about y and z in particular, let alone the overall thought experiment (for some reason, "What if the end of civilization is good, actually?" isn't a popular research question). This is one of the only settings with a significant concentration of people who are likely to avoid the reflexive motivated reasoning that would lead most people to dismiss these questions out of hand and who have experience analyzing similar uncertainty-heavy ethical issues, so I would be interested to hear what you all think about it!