With the holiday season coming up, effective altruists around the globe will gather with friends and family to celebrate their good fortune and wonderful community. Community meals are not only fun, but also an important place for us to think about our values and impact. Even though some effective altruists think personal dietary change is too difficult to be the best use of their efforts, I think we can agree on a less demanding norm that still captures many of the benefits of year-round veganism by making sure that EA events are (at least) vegetarian.
Individual dietary change
Individual consumption of animal products is a common topic in effective altruism. It’s been suggested as one of the small things EAs can do to most help others, and many effective altruists support charities that work to inspire people to reduce their consumption of animal products or helped farmed animals in other ways.
But there’s plenty of opposition from people towards changing their diet from the current social norm of eating animals, and the effective altruism community is no exception. The most common reasons people eat animals have been called:
The Four 'N's: that meat eating is Natural ("humans are natural carnivores"), Necessary ("meat provides essential nutrients"), Normal ("I was raised eating meat") and Nice ("it's delicious"). Of all these reasons "necessary" was the most common.
As effective altruists focused on evidence-based reasoning, we can easily overcome the first two roadblocks because we know (i) the first is a fallacious appeal to nature — naturalness simply isn’t a proper justification for goodness with obvious counterexamples like murder and cannibalism, and (ii) a balanced vegetarian/vegan diet can supply all necessary nutrients and even have substantial health benefits. We could also add that (iii) something that’s normal isn’t necessarily good. Slavery, murder, racism, genocide have all been normal at various times in human history. Tackling the last reason is tough, although (iv) any vegan can tell you that when they serve vegan food without mentioning its lack of animal products, the reaction is often much better than if they say it’s vegan beforehand, so we should be mindful of our biases that make us overestimate the costs of veganism. Most importantly, even if it is nice for you, it’s not nice for the animals!
Social norms and institution-focused messaging
Regardless, it’s clear vegetarianism/veganism is very difficult for some people, even if that difficulty is often based in social norms or biases. Some have even suggested that going vegetarian or vegan is so difficult that it’s actually not a good choice for many effective altruists. While many animal-friendly effective altruists disagree with the reasoning in those arguments, I think they’re right that the individual diets of effective altruists shouldn’t be our primary focus.
In general, focusing on individual consumption doesn’t seem like a good idea for most social movements, largely because it lacks the provocation of moral outrage that facilitates the viral growth of new ideas, attitudes, and behavior. For example, if you tell me the source of the issue is my own actions, I immediately become defensive, try to rationalize my behavior, and am unlikely to adopt your views. But if you tell me that an institution is at fault, like the animal agriculture industry, I’m free to be as outraged as I should be at the horrors of animal farming.
I think something we can all do, given the difficulty of personal dietary change, is to promote the norm that all effective altruism events should be (at least) vegetarian. I think this avoids the issue of personal defensiveness or difficulty, since even the most steadfast omnivores can’t be threatened by the simple prospect of a few vegetarian meals, while preserving the most important effect we’d get from every effective altruist going vegetarian, that we acknowledge the moral atrocities of animal agriculture and support a future where all sentient beings, human or otherwise, are given due consideration.
Potential concerns
Some might worry that vegetarian events will weird people out, but this concern seems to almost exclusively exist in the rationalist community — the anecdotal experience of myself and many others I know indicates most people react to vegetarian fare as a health-conscious, ethical decision. I’d be open to evidence to the contrary here, but of all the weird things effective altruists do, it seems this is a very minor concern if appearing normal is our priority. Similarly, someone might worry that vegetarian fare will be hard to find. Again, I only hear this concern in very specific social circles, and I’ve never seen it happen in practice.
Another concern might be that even if all effective altruist events being vegetarian would be a powerful message, one individual event serving animal bodies surely wouldn’t make the difference. I think this is a reasonable perspective, but EA is large enough now that we really need to consider our decisions as a community rather than just as individuals or isolated groups. If we want to make the huge impact on the world we all desire, we need to work together and look beyond our individual situations because that allows us to collectively achieve things none of us could alone, like making this powerful statement against the torture and objectification of sentient beings.
Others might believe that helping animals abused in animal agriculture isn’t the most important issue for them to work on, so they shouldn’t even spend small amounts of effort on the cause. I think this view makes sense in an isolated context for some individuals, but I think it’s a bad norm for the effective altruism movement. Much of our strength comes from having a strong community of do-gooders working on a diversity of causes who are able to join together for different projects to multiply their impact. I think an important community norm in EA is: we shouldn’t take actions that obviously and severely harm the work many other EAs are doing. Moreover, we should be willing to take small actions that others think are very important, even if they don’t make sense given our particular world-view.
Of course, this is challenging because only one major focus area of effective altruism, as far as I know, is focused on an issue most humans clearly participate in on a day-to-day basis. So there’s not an analogous situation to help other people understand this from an animal advocate’s perspective, but to put it mildly, when other people eat animals at EA events, it feels as if some people at that event gathered in a circle and began writing hate articles against the Centre for Effective Altruism or cutting up malaria nets that the Against Malaria Foundation was planning to distribute. It feels like a slap in the face to our work, and worse, like a dismissal of the plight of the billions of suffering farmed animals.
It’s also worth mentioning that some people have allergies or other medical conditions that could make certain dietary restrictions more difficult. I’ve mostly heard this levered against vegan fare, but it could apply to vegetarian as well, not to mention that milk, eggs, and fish are some of the most common allergies in the United States. For example, something that prevents you from eating beans would exclude several common vegetarian entrees. But it seems that a vegetarian event can still accommodate these people if care is taken in preparation and choice of dishes to serve, especially given the diversity of medical needs I’ve seen at other vegetarian/vegan events. So, as would still be the case with a non-vegetarian event, event organizers should take care to customize their fare so that all serious medical needs can be accommodated.
Looking forward
It’s also worth mentioning that there are several other benefits to EA events being vegetarian, like that vegetarian events could help some effective altruists transition to a vegetarian diet because it shows the viability of vegetarian meals and overcomes the crucial first step in the formation of a new habit. Eating animals also leads people to judge animals as less important than they would otherwise. We want to get as close to clear-headed, debiased moral decision-making as possible, and actively participating in the issue at hand seems like it makes this decision-making much more difficult.
Perhaps more importantly, many effective altruists are bothered by the sight of animal products because it reminds them of the massive suffering in animal agriculture. Although this may seem weird because a piece of meat is so far removed from that tragedy, remember that, “The mark of a civilized human is the ability to look at a column of numbers, and weep.” If we cry at numbers, doesn’t it make sense to feel disturbed by a cut up body that represents billions of abused sentient creatures?
So this holiday season, as you eat and make merry with your fellow do-gooders, I’d ask that you try leaving animals off your plate and encourage others to do the same. The personal costs are minimal (many would even say it’s a benefit!), and the effective altruism community can do huge amounts of good by adopting this simple norm.
An alternative, although generally less preferable, policy, if someone really insists that they need to eat animal bodies at the event, would be "no birds or fish," because these directly cause much more harm due to the small size of each individual.
Edit: As Jeff points out, there are some plausible arguments for why the difference of harm might not be as strong in total as the size consideration suggests. Still, almost all animal-focused EAs think eating birds/fish does more harm than eating cows/pigs, so I think it's still a better policy than not having one at all.
I supported not serving meat at big EA conferences where 1) food is purchased collectively by the conference; 2) symbolic and media effects are greatest; 3) the alternative is simply serving farmed chicken/cow/pig/fish.
For less official events and ones where individual purchases are distinguished (e.g. restaurant orders), the argument seems much weaker to me.
Once you start assessing things in terms of amount of harm while allowing animal products to be served, using Animal Charity Evaluators' figures you could do better by requiring a donation of $0.02 per omnivorous diner. To counter the (very plausible) possibility their figures are much too optimistic, and to get a much larger net benefit than the harm of increased farm demand, make it $1. That can be either an extra fee charged when submitting dietary preferences in advance, or a 'tip jar' passed around at a restaurant.
I suspect this may actually be a better norm to spread: many more people pay extra for (not really) "humane meat" and eggs than go veg*n, with total spending of many billions of dollars, vastly more than all cultured meat research or animal advocacy funding. Redirecting a substantial portion of that to leveraged animal protection charity would have massive effects. And the norm is easier to spread because it has far lower personal costs, addressing this problem in the OP:
"In general, focusing on individual consumption doesn’t seem like a good idea for most social movements, largely because it lacks the provocation of moral outrage that facilitates the viral growth of new ideas, attitudes, and behavior. For example, if you tell me the source of the issue is my own actions, I immediately become defensive, try to rationalize my behavior, and am unlikely to adopt your views. But if you tell me that an institution is at fault, like the animal agriculture industry, I’m free to be as outraged as I should be at the horrors of animal farming."
Also, the 'indulgence/tip jar for meat' scheme would result in a reduction in farmed animal suffering with the meal, which should help with this:
"Although this may seem weird because a piece of meat is so far removed from that tragedy, remember that, “The mark of a civilized human is the ability to look at a column of numbers, and weep.” If we cry at numbers, doesn’t it make sense to feel disturbed by a cut up body that represents billions of abused sentient creatures?"
I don't see how the three distinctions you point out are particularly relevant to:
1) the harm to camaraderie in the EA community 2) the harm to our ability to reason objectively about decisions affecting nonhuman animals
Both of these seem fully in effect even if we can individually order parts of grass-fed cows at a private dinner with other EAs.
Regarding the donation versus meat eating (and I think this would apply to the coffee example too), even if the donation norm is a better one to spread, it may not be the better one to attempt to spread because it is much harder to spread. It's hard to imagine there being a particularly strong norm that "at any EA event, those who eat animals must pay $X" while it's easy to imagine a norm taking hold that at any EA event, people do not eat animals.
Yes, this is the best reason not to do it. But it would also be nice if people could embrace Pareto-improving deals that reduce farm animal demand even more as signs of camaraderie. Would freegan consumption be very different for this?
The study that everyone (including me) cites on that reported a weak effect of thinking about eating animals leading to reduced reported belief in those animals' consciousness possibly to feel better about themselves.
The people I know who make donations to animal charities to offset meat consumption (with extra to spare) and on net reduce demand for factory farming almost all accept animal consciousness, and offsetting makes acknowledging animal consciousness and well-being less personally costly or threatening.
Seems easy to imagine for me.
Lots of events have this setup for carbon emissions, and far more have 'different food options for different prices.' On a dietary preferences form where people can ask for 'no peanuts' or the like, one has an extra box with [+$X] next to it, just like one might have a box for including alcohol, or adding banquet attendance to a conference attendance fee. If someone checks the box and pays the money, then that amount is ordered and accessible with the ticket.
If people can't pay in advance and an organizer is creating food they can make the extra donations and announce the offsets.
For a restaurant get-together the invitations can announce the collection plate will be passed around for $1.
Mechanically these are all pretty manageable for one person or group organizing a dinner to implement.
" even if the donation norm is a better one to spread, it may not be the better one to attempt to spread because it is much harder to spread"
Why think it is so hard to spread when pseudo-'humane meat' has spread so much farther than veg*nism, and carbon offsets have spread so much further than things like air travel boycotts?
Whole Foods asks for charitable contributions every time a customer makes a purchase, and could raise for animal charities or cultured meat research.
Many branded products specify that a certain percentage of sales goes to a certain charity. Ethos at Starbucks, for example:
Imagine Whole Foods having a similar line of animal products funding highly effective interventions.
1) I think freegan consumption actually would be very different. You're not paying for the torture and killing. Knowing that would make it far less salient I would think. Plus, the whole reason it harms camaraderie is that we don't believe it's Pareto-improving since the norm of offsetting seems less sticky.
2) I don't really think belief in animal consciousness shows that people are thinking rationally. It's a pretty low and trivial bar.
Regarding the offsetting norm, do you have evidence that offsets have traveled further than avoiding carbon-intensive technology? I know many people who bike, use public transit, live in an urban area, etc. much of which is to some degree driven by the carbon emissions of cars. I can't think of anyone off the top of my head who uses offsets.
It would be great - and far preferable in my view - for Whole Foods to do this offset system instead of 'humane' meat (someone should propose this to them actually), but I would definitely prefer that they stop selling meat entirely. I can't imagine the offsets option would have as strong a normative effect as that.
I was thinking of 'green power' purchase programs and airline offsets as well as things like carbon-neutral data centers using credits. And contrasting that to boycotts rather than marginal reductions. [There's also a huge involuntary credit market, of course, which is more clearly larger than the voluntary responses but isn't directly comparable].
There is a normative effect of doing offsets in getting others to do offsets. If each player doing offsets has more effect than each player changing its own production/consumption, then that can be a win. And the offset charities presumably have normative effects. Would Whole Foods going back to being vegetarian do more than $20MM or $100MM to the most effective animal charities?
Yes, I can't imagine an effective animal charity doing as much good as Whole Foods going back to being vegetarian.
Regarding the normative effects of offsets charities, I think the cost effectiveness figures are far too optimistic here (the most reasonable ones apply to corporate outreach, which I think has the smallest spillovers). I don't see a case for the effectiveness of a donation outweighing the increased contagiousness of a dietary norm.
How much do you think it costs to get 3 people to adopt the dietary norm (with associated follow-on effects)?
And what do you think of the prospects for things like meat substitute R&D, cultured meat/eggs or this chicken-sexing technology?
As mentioned in the post, I don't think preventing direct harm is the major argument for having EA events be vegetarian, so it seems weird that you've chosen that argument to rebut and used it as evidence that the overall case seems "much weaker."
Regardless, I don't think the 'tip jar' or offset style norms are better than vegetarianism, mainly because they seem much less salient and harder to maintain. It seems difficult to remember to bring out a tip jar, coordinate the donations, etc. at every EA event. Simply not purchasing animal bodies seems much more straightforward.
Moreover, I think the offsets norm does even worse than individual consumption change in terms of provoking moral outrage and nonlinear change, and that the personal cost of serving vegetarian food at EA events is sufficiently low, as argued in the OP. The personal cost of year-round personal dietary change is what seems troubling.
Also, offsets seem weird and confusing to the general population. Vegetarian events is a clear statement from EA that we oppose the suffering of farmed animals that's much easier to understand.
I don't think the offsets would do much to relieve the discomfort people have at seeing their friends consume animal bodies. It doesn't seem to remove the broader norm-issues. It just seems to potentially alleviate the direct harm consideration, which, again, isn't the important point here.
"As mentioned in the post, I don't think preventing direct harm is the major argument for having EA events be vegetarian, so it seems weird that you've chosen that argument to rebut and used it as evidence that the overall case seems "much weaker.""
You seem to have misread my first two paragraphs. First I said I bought the argument for not serving meat at big conferences with merged tabs and elevated symbolic value. I then wrote:
At less official events the media/symbolic impacts are smaller, and when tabs are not pooled people don't have to feel that they are funding evil or violating deontology. So the argument is 'much weaker' for dinner with some EA friends. That is about symbolism rather than direct harm.
I mentioned harm underneath a comment where you suggested 'no fish or chicken' as an inferior alternative, on the basis of lower direct harm (and presumably that it signals trying to have smaller animal impacts).
See my response to Zach.
And I said I am much more supportive for big events than for small events. But as one generalizes to smaller and more frequent events, the two converge. I would like to be able to have a dinner get-together (of mostly effective altruists) where veg*n and offsetting omnivore friends are both happy, and which will reduce rather than increase demand for factory farming. That could be 50+ times a year. Sharing a kitchen and dining room table with (EA) housemates can be far more frequent than that.
There's not full agreement among EAs here. Yes, birds and fish are smaller, but some people think mammals matter more.
EDIT: Sorry, I misremembered this. The EAs who think "eat beef instead of chicken" might result in more animal suffering aren't going just from "mammal" status but also from environmental concerns:
-- http://reducing-suffering.org/how-much-direct-suffering-is-caused-by-various-animal-foods/
and
-- https://www.reddit.com/r/Foodforthought/comments/3fer63//ctnzbc1
Either way, I think this isn't settled.
All vertebrates have similar physiological pain receptors (http://philosophyforprogrammers.blogspot.com/2010/11/who-feels-pain.html), and it seems like there's only a (possible) significant difference in ability to feel pain when you get down to invertebrates, like insects.
"All vertebrates have similar physiological pain receptors ." This might be true, but doesn't really decide the issue. Pain receptors are in the peripheral nervous system, but pain is mediated by the brain. In humans, there are many cases where we have activation in pain receptors without consciously feeling pain (soldiers during battle), and similarly there are many cases where we consciously feel pain without any activation in pain receptors (phantom limb pain, chronic pains). Also, it's worth noting that the link you provide refers to a table in Gary Varner's 1998 book. I'm a former grad student of his, and I don't think even then he would have said the argument for pain is "just as strong," in other groups as in mammals, but he does have an updated table in his 2012 book "Personhood, Ethics, and Animal Cognition" that incorporates some of the points I mentioned above. And, in particular, he writes (p. 123) that "...the argument by analogy for pain in non-human animals is strongest in the case of our fellow mammals, and weaker for all of the other taxa." He does still conclude that the best place to draw the line is between vertebrates and invertebrates, but he's also not saying the arguments are equally strong in al cases.
I don't know of any plausible basis besides arbitrary taxonomy that mammals would be that much more morally valuable (several hundred chickens per cow). Some night say the neocortex, but it seems moon-mammals have functionally analogous structures like the dorsal ventricular ridge.
Though I think that fish and chickens probably feel pain, and that we certainly should base our moral decisions on the assumption that they do feel pain, it also seems pretty clear that there is a stronger argument by analogy in the case of mammals. In particular, the anterior cingulate cortex and the insula cortex play a central role in what is known as the affective dimension of pain in humans and other mammals. The most striking evidence for this is that humans with lesions to these areas will report "feeling pain but no longer finding it unpleasant," but there also is extensive evidence from fMRI on humans and other mammals, single-unit recordings on humans during surgeries (and on other mammals during invasive research), direct stimulation (in the case of the insula) on humans and other mammals, lesion studies on other mammals, and knock-out and knock-in studies on mammalian brain structures.
This is relevant because all and only mammals have these brain structures (the cingulate and insula). There aren't any particular behavioral responses to noxious stimuli that you would see in, say, a mouse, but not in a chicken, which is why I think we should assume that they also consciously feel pain. Nevertheless, it is pretty clear that saying "organism M probably feels pain in a manner similar to humans because M has all of the same brain regions which seem to be doing the same things during noxious stimulation and M also has behavioral similarities" is a stronger argument by analogy than saying "organism M probably feels pain because it has behavioral similarities to humans." So even if the best place to draw the line is between vertebrates and invertebrates, we shouldn't confuse that claim with the claim that the evidence is equally strong in both cases. Whether the evidence is strong enough to outweigh the large difference in numbers between eating beef versus chicken, probably not, though I'm not exactly sure how to weigh the "probability of similarity" from arguments by analogy.
I think a better analogy here is holding events that are more expensive than they need to be because they're more luxurious than they need to be. Some EAs try to keep their consumption down so they can use their remaining resources to help others as much as possible, and expensive events really cut against this.
This is better in the sense that it's more realistic (and happens) but worse in that it doesn't capture the directness and visibility of the harm. Obviously most of us don't think that matters on its own, but it does for the setting of norms and the general camaraderie among EAs. I know I'm only one of many who feel alienated from and lukewarm about the EA community in light of how much resistance there is to what should really be a fairly easy policy.
It seems like you think there should be a norm that people don't eat animals or animal products, and I think there should be a norm that people minimize their consumption and donate the rest to help others. We're both trying to build norms here. When I see lots of being spent on things that are just "nice to have" I think of the people who needed that additional money so much more.
I would argue that increasing the size of the movement is more important than trying to enforce individual morality. If Effective Altruists are truly effective then each additional EA is worth more than having a large number of EAs stopping eating meat at EA events.
I agree with your conclusion, and co-lead an EA group that puts it into practice. But I'm incredibly uncomfortable with framing this as "because it upsets people" rather than "because it's the right thing to do." My group lost at least one member because the concept of QALYs was profoundly upsetting to them- should we stop using QALYs to prevent that? Where is the line?
It's definitely a trade-off. I think many more EAs are bothered by other people eating animals than the use of QALYs, that eating animals is far less useful than QALYs, and (less certain here) they are bothered for more EA reasons. If a large number of EAs opposed the use of QALYs because, for example, they felt QALYs painted a very misleading picture of what makes for the worst health issues, then I do think the EA community should seriously reconsider their use.
I do worry, for example, that people could start acting upset by something in order to make changes in the EA community. Although that abuse is possible, I think accepting some risk of it is worth making people more comfortable in cases of genuine discomfort. If I started seeing more abuse, I could change my views, but right now I think there's basically none. So given the lack of these issues, I'm okay with a norm of, "When something is upsetting a lot of community members and doesn't have a clear, substantial benefit for other community members, we should strongly reconsider including it in the community."
It's extremely difficult to persuade people that being veg*an is the right thing to do. It's relatively easy to persuade people that they shouldn't do things that strongly upset many members of their community.
I find my instincts on this slide back and forth quite readily depending on what the context is, and in particular on how intrusive enforcing vegetarian options would need to be. Some example contexts:
I'm hosting an EA event where I live and providing food. Here I'm reasonably inclined to only serve vegetarian options unless and until someone indicates that it's a problem.
I'm organising a local EA event at a third-party location where both vegetarian and non-vegetarian options are available. Now I'm more inclined to let people order what they want. It's not going to possible to enforce otherwise without speaking to more-or-less every attendee individually, which is quite a high cost and sets off defensive reactions that we were trying to avoid in the first place.
A reasonable middle-ground is to go to vegetarian-only restaurants or pubs, but this cuts down the choice of location significantly and it can be hard to both do this and enforce this point from the OP:
"It’s also worth mentioning that some people have allergies or other medical conditions that could make certain dietary restrictions more difficult...So, as would still be the case with a non-vegetarian event, event organizers should take care to customize their fare so that all serious medical needs can be accommodated."
(EA Global SF). An EA conference which is providing meals onsite, with other options in walking distance. Now that we're back to being able to 'quietly' enforce vegetarianism, I'm ok with it.
(EA Global Oxford). An EA conference providing only snacks, not meals. Randomly formed groups go to third-party locations afterwards. You could try and enforce a norm of 'if you're eating with other Effective Altruists please only eat vegetarian here'. And again I'm mostly inclined to think that it's not worth it.
In practice most EA events where I live (London) fall squarely into category 2.
A mostly-unrelated thought on the 'symbolism' concept. I think symbolism cuts both ways here. I don't associate with a lot of altruistically-focused movements with whom I share goals because I think they focus too much on moral purity/moral grandstanding and too little on actually making a difference. Examples would be the green/environmentalist movement and the social justice movement. This is one of the few things within EA that strikes me the same way; it's a low (direct) impact action whose main point is to signal things to others. Such actions tend not to look so impressive once their purpose is known.
In isolation I don't consider this too concerning, but I wouldn't like any significant proportion of community time to be spent on such things. A corollary is that we should spend that limited symbolism budget as efficiently as possible. Which probably feeds into why I'm much more in favour of this when the enforcement is easy.
"So there’s not an analogous situation to help other people understand this from an animal advocate’s perspective, but to put it mildly, when other people eat animals at EA events, it feels as if some people at that event gathered in a circle and began writing hate articles against the Centre for Effective Altruism or cutting up malaria nets that the Against Malaria Foundation was planning to distribute. It feels like a slap in the face to our work, and worse, like a dismissal of the plight of the billions of suffering farmed animals."
I think that it is important for EA to be a pluralistic movement. That is, there may be some EAs who are completely onboard with animal rights and global poverty reduction, but think existential risk due to AI is silly. There may be some EAs who think that the figures are such that existential risk outweighs every other cause and that the people working on other causes are silly. I don't see this as problematic. On one hand, I can see that having all events being vegetarian makes it easier to recruit animal rights activists into the movement. On the other hand, it can create a situation where non-vegetarians feel out of place, which could be very bad if it reduces the eventual size of the movement.
But ACE's best estimate is that the cost of offsetting one meat meal is around 2 cents. That's not a severe harm, that is a very trivial one. The costs of avoiding such small harms almost definitely outweigh the harms themselves. Considering that many people are willing to pay much more than 2 cents to eat meat, it seems very ineffective to require they not do so.
How does the high effectiveness of the recommended ACE charities make a harm more trivial? AMF can save a human life very cheaply; does that make taking a human life a trivial harm?
Same for the willingness of many people to pay much more than 2 cents to eat meat. Why does their strong preference for eating meat make it very ineffective to avoid eating meat? We should be thinking about what does the most good here, not just what satisfies people's personal preferences.
Because people generally care about animals in an aggregative sense - they care about the total amount of suffering.
No, firstly because it costs AMF over $3,000 per life, which is 600,000x more than the figure I was discussing. Multiplying a trivial number by 600,000 can yield non-trivial numbers!
Secondly because we generally think of human lives as being less interchangeable. Killing one human to save one other is not acceptable.
It's unreasonable to expect people to dedicate 100% of their resources to altruism. But what people are willing to dedicate, we should dedicate in the most efficient manner. It's better for both the individual and animals in aggregate for someone to eat meat for lunch and donate $1 than to abstain from meat.
That doesn't seem to do the work you imply it does. Being able to spare 100 lives is a huge feat of good, even if the total amount of people suffering is much greater.
But it's still very cheap, even if it's much larger than other very cheap figures.
It seems speciesist to apply some moral standards to humans but not nonhumans.
As argued elsewhere on this page, it seems dietary change has many more benefits than a small donation that has roughly the same (or even better) direct impact. And your original argument, that "many people are willing to pay much more than 2 cents to eat meat," doesn't do any work in addressing those additional benefits and simply draws from personal preference.
It is pretty hard to offset a human life, as estimates from givewell suggest a cost per marginal life saved in the thousands of dollars.
Lark's point (I imagine) is something like this. If you think ACE's figures are about right, the direct harm of eating meat can be offset fairly cheaply, so better a carnivore giving a few dollars a year to THL than a vegetarian giving nothing. You might say this is a false dilemma, but in reality people are imperfect, and often try and allocate their limited altruistic resources as effectively as possible. If they find refraining from meat to be much more difficult than giving a few dollars (or earning a few more dollars to give away) it seem better all-things-considered they keep eating meat and give money.
So the harm of EA venues eating meat is primarily symbolic, as I don't think animal advocates would be happy if EA venues kept serving meat but gave $100 or whatever to THL, despite this being enough to offset the direct harm. In public facing events, fair enough (although I'm tempted to suggest that offsets etc. might be a good 'EA message'), yet this seems unclear in non-front-facing events.
That seems basically right, but different from what Lark actually said.
I've considered this, but my issue with it is: people are unlikely to actually go and donate 2 cents per meal to 'offset their meat' after a meal.
If we're talking a pre-paid conference or similar with pre-ordered food, it may be worth considering including a 2 cent surcharge (basically a token amount) to raise awareness. I'm not sure about the logistics of this... would you only charge people who don't select a vegetarian/vegan option? There might be backlash regarding discrimination...