Veganism policy
I am vegan, with the following caveats.
- I eat smoked mussels.
- I will occasionally eat left-over beef at the office, when I'm confident that I'm not thereby contributing to aggregate demand for animal products.
Factory farming is incredibly and transparently evil, and I do not want to subsidize it or in any way perpetuate it. Eating industrially-produced meat seems to me to be about as straightforward an ethical failure as purchasing cotton produced by slaves in the the American Antebellum South, one of the most unethical manifestations of slavery ever practiced on planet earth, thereby enriching slaveholders and (indirectly) slave traders.
Animal consciousness?
I'm deeply ignorant about what consciousness is and the mechanisms that are necessary for it.
There is an anthropic argument (about which I don't know how to think clearly) that conscious experience is rare among non-human animals, and perhaps among humans. But from my state of knowledge, it seems like a better than even chance that many mammals have some kind of "inner listener". And if they have an inner listener at all, pain seems like one of the simplest and most convergent experiences for evolved beings.
Which makes industrial factory farming an unconscionable atrocity, much worse than American slavery. It is not okay to treat conscious beings like that, no matter how dumb they are, or how little they narrativize about themselves.
There are currently about 100 billion individual animals living in factory farms (not counting the farmed fish). Assuming that they are all conscious beings, that's 100 billion experience-years in factory farms every year.
It seems to me that, in my state of uncertainty, it is extremely irresponsible to say "eh, whatever" to the possible moral atrocity. We should shut up and multiply. My uncertainty about animal consciousness only reduces the expected number of experience-years of torture by a factor of 2 or so.
An expected 50 billion moral patients getting tortured as a matter of course is the worst moral catastrophe perpetrated by humans ever (with the possible exception of anthropogenic existential risk).
Even if someone has more philosophical clarity than I do, they have to be confident at at a level of around 100,000 to 1 that livestock animals are not experiencing beings, before the expected value of this moral catastrophe starts being comparable in scale to well-known moral catastrophes like the Holocaust, American slavery, and the Mongol invasions. Anything less than that, and the expected value of industrial meat production is beating every other moral catastrophe by orders of magnitude (again, with the exception of anthroprogenic x-risk).
Admittedly this analysis is making some assumptions about the moral value of pain and fear, relative to other good and bad things that can happen to a person, which might influence how we weight the experiences of non-human animals compared to humans. But "pain and terror are really bad, and it is really bad for someone to persistently experience pain and terror" seems like a very reasonable assumption. That the inner lives of non-human animals are less complex than my own suggests a decrease in moral badness of maybe two orders of magnitude, at best. (I think there are plausible claims that the lessened complexity actually increases the moral weight.) I don't think there are plausible ways for the numbers to work out such that factory farming isn't an atrocity.
What about happy animals?
What about raising animals to have happy lives in humane circumstances and killing them to eat?
I'm uncertain on this point. I think it ultimately hinges on whether persons/agents or consciousness-moments ought to be the (main) relevant units of our moral calculus. It seems plausible to me that the equilibrium of my moral philosophy assigns a special value to personhood, such that creating conscious beings with the explicit intent of later ending their existence is sad, even if their lives are net positive, and even if they add to the aggregate utility of the universe. In contrast, it also seems plausible to me that positive conscious experience is a robust good or even that personal identity is ultimately a confused concept that isn't loadbearing for anything in equilibrium-ethics. It might turn out to depend on complicated features beyond consciousness, such as whether a thing knows itself and knows that it doesn't want to die.
Given that, I'm not sure what the moral sign of humanely raising animals for slaughter is. It seems plausible that it could be a positive or a negative. But regardless, the magnitude of this moral dilemma is radically smaller than that of industrial farming. I'm not very invested in this question, compared to the enormously overdetermined question of whether industrial farming is morally acceptable.
At my current level of uncertainty, eating animal products exclusively from humanely raised animals seems like a reasonable alternative to vegansim, to me, as long as one is actually careful about verifying the humane conditions, rather than (as I did at one point), intending to eat only humanely-raised meat, but not putting much thought into how to identify which was which, and then flexibly bending that standard when meat was available.
But as noted, I could imagine changing my mind about this.
What about offsets?
I'm skeptical of eating industrial meat while also paying an offset to negate the harm. I think that I would be in favor of offsetting in principle, but only of offsetting schemes that are clearly robust and universalizable.
Most offsets for eating industrial meat (or at least the ones I've heard about) seem pretty sketchy to me. They're usually some form of "pay for interventions that statistically cause people to go vegan." I'm suspicious that these actually work as advertised (especially since some large fraction of self proclaimed "vegetarians" also report having eaten meat in the past week). Offsets of this sort depend on a specific, complicated, world model being correct for the money paid to translate into reduced demand for industrial farming.
In order for the offset to work...
- The investigative process that found that those interventions cause people to go vegan was unbaised and reliable, avoiding very common experimental or statistical errors, social desirability bias that might make people exaggerate their veganism, and motivated reasoning on behalf of the researchers, among other possible problems.
- The tested intervention translates to the inevitably different context where it is being deployed.
- The organization to which you are donating money is actually performing that intervention, and not doing something else, or simply scamming you.
That is a lot of junctures where the logical chain could break. I don't trust that kind of modeling enough to rely on offsets as a substitute for the simple policy “don't eat factory farmed meat”.
The demonological injunction has a virtue of simplicity. Excluding bizarre scenarios, not consuming industrial meat reliably removes the negative impacts of consuming industrial meat. It is robust to details of my world model turning out to be wrong, a way that "pay for ads that make people go vegan" very much isn't.
My consequentialist altruistic projects are aiming for maximum expected value, and so in those domains, I'm willing to rely on relatively fragile chains of argument, if that's justified by EV estimates. (Though I still try very hard to verify my impact models! Most of the potential value of ambitious altruistic consequentialism is squandered by altruists acting on wrong impact models.) But the point of having deontology at all is to to be robust to my own too-clever arguments, the black swans that I didn't account for, and ordinary self-serving biases. So in those domains, I rely on simple and conservative reasoning as much as possible.
Furthermore, even if this offset scheme does work exactly as hoped, it doesn't universalize. It is logically impossible for everyone in the world to adopt the policy of eating industrial meat and offsetting the harm by inducing someone else to go vegan. With this sort of offsetting plan, what happens when all the easily convincible proto-vegans have gone vegan already is left indeterminate. Someone offsetting their meat consumption today is, at best, passing the buck to their future self.
In contrast, this offsetting scheme is simple and robust, and nicely universalizes. If this were an available option, I would participate in it, instead of foregoing eggs entirely, because I can trust the one-to-one correspondence between my offset and the harm of consuming eggs and because if everyone adopted this policy, industrial egg farming would then be solved.
That's the standard to which I hold any scheme for offsetting industrial meat consumption.
Appendix: My personal veganism trajectory
~2009 - I believe that I first became vegetarian the summer between 8th and 9th grades. I stopped eating meat for a combination of 3 reasons:
- I reasoned that I didn't care if something ate me after I died, but I didn't want to be killed to be eaten, and by the golden rule, it seemed that I should extend the same courtesy to other animals.
- I read a book about Ancient Atlantis that claimed that the Atlantean priests were vegetarians, and I was (for some reason?) motivated to be like them.
- At the time I was engaging in a practice of focusing fully on my food while I ate it—paying attention to the sensations and textures, (and, if I remember correctly, thinking about where the food came from). I paying attention to the details of my experience like this I found some kinds of meat, such as chicken on the bone, disgusting, for having rubbery tendons, like there are in my own body. I was not grossed out by more processed meat like hamburgers and hotdogs. But it struck me as sus to be willing to eat meat, only when the form factor put me in less direct contact with what I was doing, and for consistency, I should stop all together.
I have an allergy to milk, so I was functionally ovo-vegetarian.
~2017 - I had been using my vegetarianism as an example for crux-finding at CFAR workshops. In particular, I had been saying that if being vegetarian was hurting my health or cognitive performance, that would be a crux for me: a reason for me to eat a little meat. Participants offered me some evidence that vegetarianism did have health costs, which compelled me. (In retrospect, the evidence wasn't very good, and I didn't evaluate it very critically.)
I started adding a small amount of meat back into my diet out of nutritional concerns, but was otherwise vegetarian.
2020 - I read that eggs are among the most suffering-intensive animal products (though I again did not assess this claim particularly critically), and so decided to stop eating any eggs and instead eat small amounts of beef or turkey bacon on a regular basis.
2021 - After thinking more about the ethics of factory farming, I became more hardline about veganism—it seemed less like a "personal choice" and more like an obvious ethical imperative that almost everyone was failing—and stopped eating beef.
2023 - I became more concerned about the cognitive impacts of avoiding all animal products, and provisionally decided to add eggs back into my diet as an occasional supplement, with much more attention to sourcing, aiming to only consume eggs produced by happy hens, specifically from the brand Vital Farms. (At the time, I had flagged my uncertainty about how I should count the suffering of the male chicks that were the siblings of those hens, and how morally responsible I am for their fate, by consuming the eggs of their sisters.)
2024 - A friend of mine argued that even the most humanely raised egg laying hens still have pretty awful lives. I decided to provisionally stop eating vital farms eggs until I did the due diligence to assess for myself.
I haven't done that due diligence yet, and so still don't eat Vital Farms eggs.