I'm Karthik, a scientist by training in biochemical engineering and systems/quantitative biology. I've been keen on the Effective Altruism movement since first hearing about it on Sam Harris' podcast in 2016. I'm particularly motivated by the problem of farmed animal agriculture, and have steered my life and pursuits accordingly. I currently work as a data scientist at Climax Foods to develop next-generation vegan food products. And I've also authored a book about the impending end of animal agriculture--focusing on how poor and inefficient animals are as a production system (the "technological" argument against animal ag). Book is titled After Meat, and more info is here: https://aftermeatbook.com/
I love the story and helpful summarization of Singer's mindset. His emphasis on the truth and listening to your detractors is an inspiration.
Reading an essay from him in college was my first step toward appreciating animal welfare and steering my life accordingly. I can't imagine how different my focus would be if it weren't for Peter Singer.
Thanks Cornelis, I agree about the empirical evidence. And indeed, emphasizing Ami and how long it's been around would have obviated a lot of confusion here.
We seem to disagree about (1) the variance of meat-based retail products versus vegan ones and (2) whether or not the "trial-by-fire" standard is more helpful than just simply the criteria that AAFCO/FDA defines regarding nutritional, toxicity, digestibility, and safety.
(1) Sounds like your priors for the intra-variance of meat-based cat food are lower than the inter-variance between validated meat and vegan cat food. I don't share these same priors, and the best I can offer is Chapter 6 of After Meat where I explain in detail how I think about nutrition including the fungibility of food. Long story short, I really don't think there's anything special nutritionally about meat that can't be recapitulated elsewhere, but I understand that not everyone has that intuition.
(2) AAFCO updates their standards in light of new evidence. The "trial-by-fire" is baked into their standards. And I suspect if we had similar priors regarding (1), this point may be moot.
But I appreciate your good faith, and I'll leave it here.
Thanks Cornelis, I sincerely appreciate the good will shown.
For me: I agreed with you and felt like my mind was being changed to being pro-vegan-cat - until I read Elizabeth's comment pointed out the issues in the study. So for me it is mostly because you haven't engaged with that specific comment and pointed out why the concerns that are highlighted in her screenshots (from the actual study!) are not something that I need to worry about.
I conceded on Domínguez-Oliva et al., and Elizabeth's concerns were entirely valid. However, it's one study and one diet, and I felt that Elizabeth was ignoring the basis of nutrition and biochemistry that I emphasized throughout the post. Thanks for highlighting that the food was lacking necessary, known supplements. That was a key point that would have been helpful to broach earlier.
A RCT study will likely be just a formality for something like the Ami vegan cat food. And yes, it's frustrating that this hasn't been done/published yet! As I understand from Andrew Knight, there's a better study coming out this year.
Convince Elizabeth and you, by proxy, convince me I'm pretty sure.
I haven't found Elizabeth willing to falsify their thinking as much as you and perceive general antagonism and defensiveness.
She meant the diet that we have the most empirical evidence doesn't harm/kill them.
We don't have empirical evidence of the same happening with meat-based diets.
Do we have empirical evidence that a specific meat-based food is consistently safe over many years? My understanding is that many change constantly.
So modern nutritionally complete meat-based diets presently have a 100%-wont-cause-major-adverse-health-outcomes rate. Is this not what the studies seem to show?
Not really. Check out the recall withdrawals over the year: https://www.fda.gov/animal-veterinary/safety-health/recalls-withdrawals
Best we can do is what AAFCO and FDA already do. AAFCO sets nutritional guidelines and ingredient requirements. FDA regulates the safety.
We have some empirical evidence that vegan diets appear to quckly give cats major bad health outcomes without supplementation? The first comment in this thread by Elizabeth pointed this out.
Why is that diet representative of for example nutritionally complete Ami, which has been around for years? Isn't it much better to just defer to AAFCO's and FDA's standards, which Ami meets?
The most parsimonious explanation is that the lack of supplements was the problem, not the "vegan"-ness. If I get sick from my potato chip diet, it doesn't mean I should avoid a plant-based diet with all required nutrients.
If you were to consider all possible meals a cat could reaonably be fed today. On average, it seems reasonable to suspect that they would be healthier if more of those meals were meat-based than plant-based.
Based on what? I don't intuit this at all. Furthermore, we're not saying any plant ingredients. We're saying the ones that meet nutritional, toxicity, and digestibility requirements for cats.
Both you and Elizabeth have offered something fairly distinct from AAFCO's/FDA's standards. And so far, I'm finding neither to be better.
Given how nascent the field is and how we're only just finding out what supplementation we might have to give cats
It's not nascent. AAFCO has been providing guidance on the nutritional requirements for cats for decades.
Something weird and unexpected like the NO pathway could explain why cats on vegan diets still get health issues as the pro-vegan study Elizabeth linked to shows.
But we have the same uncertainty with retail meat-based cat food, which I've highlighted is quite distinct from what cats evolved on.
An escape hatch from this would of course be lab grown meat that is to the molecule identical to meat.
I don't understand the obeisance to molecularly-exact meat. Evolution doesn't select for health and well being. It selects for propagation for a specific niche in a specific environment. Our goals with domestic cats are different than what evolution optimized for.
Consider human evolution. For most of it, life expectancy was far lower than today. The diet of prehistoric times isn't by default aspirational. Instead, current nutrition studies focus on health outcomes, e.g. life expectancy, blood pressure, rate of obesity, etc.
So recommendations focus directly on cause (food eaten) and effect (health outcomes). And that's what we should do with cats. We should not put meat on a pedestal and beeline for that.
Feed animals close to their natural diet while researching how to do better. You dismiss this as "appeal to nature", but I would describe it as "the burden is on the attempt to change the default".
"Meeting nutritional requirements" is a far better default standard than what's "natural". Few problems with the "natural" standard:
I assure you I am at least as obnoxious about human nutrition testing, which is better studied and features a more adaptable subject.
Yep, human nutrition is better studied. There's more funding and more interest in the subject.
As discussed before, we're both in agreement for more studies. Funding is needed.
Thanks Lilly. FYI, as far as tractability goes, I ask for ingredient approvals and a few RCTs. These wouldn't require much in terms of cost (~1 million USD). There's millions of vegans around the world, and I bet many are relishing the opportunity to rear vegan cats, including myself.
I also want to flag another ancillary benefit that developed in the days since discussing this post with others: Many folks understand the ethical and environmental benefits to going vegan, but won't do so themselves because of perceived health concerns. If "carnivore" cats can be vegan safely, that's quite the signal to non-vegan humans.
Thanks for the ideas. I do have a few concerns.
1.) Would owners be that much more receptive to switching to grain-free for ethical reasons? I would imagine similar head wind compare to vegan diets. I agree with you though about the cost point.
2.) I think this too would also be a tough sell. I'm guessing most owners think more protein is better for health.
3.) One issue here is the carcass balancing problem of animal agriculture. Cat food is generally the offal of the industry. I'm dubious of this approach without it exploding the cost.
Okay, it sounds like your argument is "vegan cat food is capable of meeting the same standard as meat-based food".
Indeed.
From skimming the AAFCO document it's not obvious to me AAFCO thinks meeting its standard is sufficient for health (see screenshot below, from page 2).
As you pointed out earlier, we don't know everything that we could know. What do you propose we do? It sounds like your concerns are with food testing for cats period, and they're not specific to the vegan formulations.
My stance is that if it's okay to feed cats meat-based food, then it's fine to feed them vegan food.
You say you've done research satisfying you that this is all biochemistry and we know everything we need.
To be clear, this is not what I'm arguing. Biochemistry research is never complete. I'm arguing that it's safe to feed cats vegan based on what's known.
Can you write that up?
Here's the AAFCO list of required nutrients for cats and what I view as authoritative (pages 13-14):
How are we sure we've identified every useful nutrient? How do we know the bioavailability tests are any good?
I'm not sure. But these questions persist for Fancy Feast and any other meat cat food as much as they apply to vegan cat food. Are we sure they have every useful nutrient and that the measured bioavailability is good?
Jakub, thank you for this piece. I loved it when I first read it, and I find myself coming back to it often.
I also appreciate how much the EA movement has normalized eating vegan. In 2013, only about 25% of professors in ethics reported to not eat meat. So, it's striking and quite the signal that every EA event I've been too (2 EAGs, 1 EAGx, many small ones) has served fully vegan food.