Deontological Constraints on Animal Products

Introduction

There is a memetically powerful argument within animal advocacy circles which goes like the following: “We would never ask child abusers to commit less child abuse, so we can’t ask other people to reduce their animal product consumption. We must ask them to end it.”

In this post I try to construct and evaluate a part of this argument. First, I explain my motivation for evaluating the strength of this argument. Second, I note that it’s morally permissible to ask for reductions in some kinds of wrongdoings and list different ways animal production consumption can be morally wrong. I create the category of “non-negotiably wrong” to refer to actions that can’t be asked to be reduced. Third, I look into whether animal product use might be non-negotiably wrong by listing several deontological constraints that might be non-negotiable.

A Venn Diagram summarising results

I don’t have any strong conclusions. I aim to reduce my own confusion and get more input from professional moral philosophers on this topic through this post. I’m also not sure if I should keep writing such posts, so if this post is helpful to you in any way, please let me know.

Many thanks to Michael St. Jules and Bob Fischer for their helpful feedback. All errors are my own.

Motivation

Some animal advocates argue for the following norms because they believe people have a non-negotiable duty to avoid consuming animal products:

Only vegans can speak at animal advocacy events
Only vegans can be members of animal advocacy organisations
Non-vegans shouldn’t join animal advocacy protests
All animal advocacy organisations have a responsibility to prominently advocate for veganism because it’s the main obligation to animals
It’s morally forbidden to use the following sentences because they condone some animal product use or don’t explicitly reject all animal product use:
-Go vegetarian.
-Meat should be taxed.
-Our school should have Meatless Mondays.
-Costco should go cage-free.
-The default school meals in Grenoble should be vegetarian.
-The public schools in New York City should serve exclusively plant-based food on Fridays.
-Take the vegan-22 challenge, and go vegan for 22 days.
-Maybe you should try going plant-based except for cheese.

According to this line of argument, animal product use is not merely harmful(akin to carbon emissions) but also violation of a very strong moral constraint(akin to direct physical violence or owning slaves). It is non-negotiably wrong. For that reason, including non-vegans in animal advocacy is similar to including slave-owners in anti-slavery advocacy. Asking for a reduction in animal product use is akin to asking for a reduction in physical violence(“don’t beat your wife in January”).

To clarify, as it is the case with many issues, there is a spectrum of opinions here. Some people will endorse some of the positions above while rejecting others.

I have been sympathetic to these arguments when it comes to my own consumption. I’m very sympathetic to the idea that since animals are not well-represented, we’re likely to have a bias against their interests. When animal interests and my own interests get into conflict, it makes sense for me to be extra cautious to compensate for my own bias. So I’m happy with being strict in avoiding animal products in clothing and food.

On the other hand, I also suspect being too restrictive in animal advocacy might result in more animals being killed and tortured compared to alternatives. Some reasons offered are the following:

-There might be a Laffer curve to the behaviour change created by your demands. Being too demanding might result in less change than being moderately demanding. (Example: The New York City officials won’t take you seriously if you ask for entirely vegan schools, but will be willing to work with you to have plant-based Fridays.)

-Compliance with small demands might be a stepping stone for compliance with larger demands.

-There is some evidence that some realistic policy changes have benefits for the animals. To form a winning coalition to pass and protect these policies, you need to cooperate with non-vegans and use some of the incrementalist sentences above.

-Some believe that asking people to go vegan is a distraction from pursuing institutional change. According to this argument, vegan advocacy is similar to trying to get 100% of the population to boycott all of the slave-made products. A faster approach would be to just change people’s attitudes/beliefs/political behaviours and get them to vote/protest/fight for abolishing slavery.

It’s often the case that people in favour of restrictive deontological constraints in activism also believe violating these constraints will lead to worse outcomes for the animals. That’s a big empirical question outside of the scope of this post.

Nonetheless, some restrictive advocates also believe we should avoid violating the constraints on advocacy even if avoidance leads to worse outcomes for the animals. That is a very significant cost. For that reason, I want to assess whether these constraints are really sound.

I think there are two parts to be examined in the restrictive argument:

Animal product use is morally forbidden to a very strong degree.
When someone is engaging in a behaviour that is strongly forbidden, it’s also forbidden to use sentences that countenance that behaviour to some extent.

I hope to write more on 2 in another post, but in this post I’m interested in 1. In the next section I intend to clarify what I mean by “morally forbidden to a very strong degree”.

I feel anxious writing this post, because when you try to argue that animal product consumption is not exactly like murder, you can be interpreted to legitimise animal product consumption altogether. I want to clarify that in the overwhelming majority of cases, the harms of buying animal products are much larger than the benefits, so I don’t want to incentivise buying animal products.

Different types of morally wrong

According to some people, some actions are morally forbidden in a stronger way than others. Sometimes I have conversations similar to the following:

Me: What do you think about human rights organisations trying to improve the prison conditions of unfairly imprisoned people? What they do seems to be commendable. So it looks like in some cases it’s commendable to get other people to do less bad things.

Restrictive advocate: I don’t think that’s analogous because the right to life is fundamental and non-negotiable. Negotiating when it comes to prison conditions might be okay but we can’t negotiate on someone else’s life.

So it seems to me that this person claims unfairly imprisoning someone is morally forbidden and negotiable whereas consuming animal products is morally forbidden and non-negotiable.

Here is a list of different positions one might have on this topic.

Animal product use is morally equivalent to refraining from preventing harm to animals. Refraining from donating to an animal charity that will spare one animal from being born into animal farming is morally equivalent to eating one animal and harming one additional animal as a result.(ignoring elasticity)
Animal product use and refraining from preventing harm are qualitatively equivalent but practically different. To avoid naive consequentialism, we should have a tendency towards avoiding harm and therefore avoiding animal products. This rule of thumb might even be so strict that it might be better to be strictly plant-based. Nonetheless, there is no intrinsic moral difference between eating an animal and refraining from preventing harm to an animal.
Threshold deontology: In some cases, it’s wrong to use animal products even when doing so results in making the world the best place possible. But consequences might in some cases override the constraint against consuming animal products.
Absolutism against animal products: Except for medical needs where using animal products is necessary for our own survival, animal product use is to be avoided in all circumstances. No consequence (except medical needs) can justify using animal products.
Non-negotiability of the constraint against animal products: When using imperative sentences directed at people using animal products, it’s morally forbidden to use sentences that countenance some (non-medical) animal product use.

My personal position is that it’s commendable to negotiate to reduce/reform the animal product use of other actors when the expected results of the negotiation are i. ex-ante Pareto superior for the animals ii. sufficiently close to the best possible results given the limitations faced by the negotiator.

I’m leaving it as an open question whether absolutism and non-negotiability imply each other. It’s not obvious to me that they do.

I have seen people that seem to argue in ways that suggest absolutism doesn’t imply non-negotiability. According to these people the following two sentences are true at the same time:

i. It is absolutely forbidden for me to unfairly imprison someone.

ii. It would be permissible for me to negotiate to improve prison conditions of a person who doesn’t deserve to be imprisoned.

I have also seen some other people arguing in ways that suggest non-negotiability doesn’t imply absolutism:

i. The right to bodily integrity of the animals is not absolute. It can be violated for the best interest of animals. For example it’s commendable to neuter cats and dogs to prevent the birth of new animals that will have terrible lives.

ii. The right to bodily integrity of the animals is non-negotiable, we should never countenance some bodily integrity violation to prevent another agent from committing more bodily integrity violations.

For example, assume that Buzz intends to wound an animal for no reason. He asks you to choose an individual animal for him to wound. If you don’t choose an animal, he credibly threatens to wound x number of animals instead. According to this position, no matter how large x is, it’s forbidden to choose an animal to be wounded. Nonetheless, it’s permissible to violate the bodily integrity of animals for the interest of the animals in some circumstances.

I’m mostly interested in assessing the plausibility of the non-negotiability of the constraint against animal products. In the next section I will list several candidate constraints that might be violated by animal product consumption. Then I will assess whether these constraints are non-negotiable.

As a starting point, I will assume that many animals including mammals, birds and fishes have a right to bodily integrity and life. At least in some cases we are forbidden from violating these rights even when it’s optimific to do so. I’ll also assume that the production of animal products violates these rights fairly regularly and it is (typically) morally wrong.

The usual argument for non-negotiability of animal product consumption is to focus on the moral wrongs committed during the production and conclude that consuming animal products is morally equivalent to committing the wrongs during production. Since dairy production involves artificial insemination without consent, dairy is rape. Since rape is non-negotiably wrong, dairy consumption is non-negotiably wrong. It’s wrong to ask for fewer rapes, so we can never ask for a reduction in animal product use.

I think this inference is too quick. When it comes to the consumption of immorally produced products there are a few complications involved:

The primary moral wrong(killing and bodily integrity violation) takes place before the act of purchasing and consumption.
The act of purchasing/consuming products is more continuous than acts like murder or rape. Materials can be split into many small pieces and mixed into everything. We live in a complex economy with large supply chains where it’s very usual for immorally produced materials to be mixed into products we buy. Raw materials are often sold by states ruled by autocrats that seize these resources through violent means. Similarly, certain animal products like gelatin are mixed into so many random things including toilet paper, packaging, cars, planes, electronic devices, shoes, paints and furniture.

I don’t think these complications undermine the wrongness of animal product purchase. But I think these are serious problems for absolutism and non-negotiability of the constraint against animal product use. It’s much easier to sustain absolutism against murder. But people don’t tend to commit murder in their daily lives because they fail to read ingredient lists. In consumption there are way more contexts in which absolutism is visibly tested.

Furthermore, the literature on non-negotiability is not sufficiently developed. There is a lot of philosophical literature on doing harm to bring about good in the world. But there isn’t that much literature on the conditions for non-negotiability. This essay is an attempt to make progress on this concept.

Clarification on terms

In the rest of this post, I will use the word “unnecessary” for some products. I want to clarify what I mean by that.

It’s often claimed that morality can’t demand us to die, so it’s morally permissible to use animal products when it’s necessary for our survival. For example, if some animal-derived drug is necessary to cure our illness, it’s permissible to use it.

Similarly, we have to eat food to survive, so even though the production of plants harms some animals, it’s permissible to eat bread.

I don’t think there is a sharp boundary between necessary and unnecessary products, but I won’t press this point for now. For my purposes it’s enough for me to clarify the concept through a few more examples:

Tobacco → Unnecessary

Black tea → Unnecessary. It has no calories and very low nutrients, so unlike bread, it’s not replacing anything when we consume it. It’s possible to live a perfectly healthy life without any black tea.

Any item of consumption beyond what is needed for a healthy life → Unnecessary

Having at least one electronic device → Necessary

Candidate constraint 1: Purchasing “unnecessary” products that are produced in ways that foreseeably harm others is non-negotiably wrong.

Here are a few suggested examples of such products(and services):

-Any products with carbon emissions

-Buying products of a factory that deposes toxic waste in a nearby lake, harming nearby villagers and animals

-(Maybe?) Purchasing products of AI companies as they impose significant risks on others

It seems clear to me that violating this constraint is not non-negotiably wrong. It’s okay to ask people to reduce their carbon emissions. You are not obliged to gatekeep your activists on whether they buy harmful products. This doesn’t seem controversial to me so I won’t delve deeper.

Candidate constraint 2: Purchasing “unnecessary” products that are produced in ways that intentionally violate the rights of others is non-negotiably wrong.

Here are a few examples of such products:

-Products made by slaves

-Coconuts collected by monkeys, truffles discovered by dogs

-Products tested on animals

-”Unnecessary” crops like coffee, tea and tobacco where pesticide is used during production

This constraint is very frequently invoked when asking people to go vegan. A popular statement of it is “don’t finance murder”.

The free-produce movement used similar arguments against purchasing products made by slaves:

“The slave dealer, the slave holder, and the slave driver, are virtually agents of the consumer, and may be considered as employed and hired by him to procure the commodity … In every pound of sugar used we may be considered as consuming two ounces of human flesh.” (William Fox, 1791, in Address to the People of Great Britain)

This line of argument holds the consumer directly responsible for the intentional wrongdoings committed during the production.

It seems to me we don’t currently view boycotting slave-made products as a non-negotiable requirement.

Free produce movement was abandoned by many abolitionists in the United States. In Great Britain, it focused mainly on sugar, ignoring other products like cotton.

When I think about abolitionists who tried to reduce their country’s imports of slave-made products in partial ways, I don’t think they betrayed enslaved people by asking for reduction. I don’t have any negative attitudes towards abolitionist organisers who didn’t require all the members of their organisations to completely boycott slave-made products.

If there was really something non-negotiable going on we would have much more negative attitudes towards many of the historical abolitionists. I think negotiability was “vindicated by history” in this specific case.

I also think this principle forbids most agricultural products that are unnecessary for a healthy life. Pesticide use is an act of intentional killing since its name is literally pest killer and it aims to kill the animals. Any consumption that isn’t part of the minimum consumption set for a healthy life(coffee, black tea, tobacco, additional clothing, additional furniture made by plants) would be forbidden by this constraint.

I’m very open to the possibility that morality might actually be demanding and maybe these products are actually forbidden. But I think it’s important to notice that consistent application of this principle can’t really distinguish between pesticide treated black tea and animal tested cosmetics.

Candidate constraint 3: Consuming things that cannot legitimately be commodities is non-negotiably wrong.

According to this argument, eggs/meat/dairy etc. are not commodities. They are not food, they cannot be used as objects. Sometimes people who use this argument make analogies with human organ trade: “I acknowledge that there are many harmful products, but at the very least we can be categorically against organ trade and commodifying animal products in general”. When I draw analogies with other immorally produced products, sometimes I get the response “It’s not about how it’s produced. There are moral ways to eat sugar. But eggs are not food. There is no moral way to eat eggs”.

The distinguishing quality of this constraint is forbidding the consumption(rather than purchase) of animal products not because of the way they are obtained, but because of being eggs/meat/milk per se. It’s a categorical rejection.

I will consider other constraints that focus on the way animal products are obtained, such as them being stolen from animals. But in this section, I want to focus on the specific claim that some physical substances can’t legitimately be commodified no matter how they are obtained.

I think the intuition behind this constraint is stronger when it comes to meat. There are people that are categorically against trading human organs even if they are consensually obtained. Using meat might be viewed in a similar way.

I agree that there are certain “objects” that cannot be commodified and traded no matter how they are obtained. Assume there was an assembly machine that can create any object by just assembling atoms in the right way. You ask it to produce a copy of an individual, it just assembles the atoms from its ethically obtained storage and creates an exact copy. I think it would be morally wrong to commodify the copy of an individual, as it would be a conscious and sentient being. Similarly, we can claim that sentient animals themselves can’t be commodified no matter how they are “produced”.

But when it comes to eggs/dairy, it seems obvious to me that they are not individuals. They are mere things. There could be moral ways to eat eggs and dairy. If someone created exact copies of eggs, dairy or wool through ethical means, then I don’t see any problem with consuming these products. If it was possible through precision fermentation to produce a substance chemically identical to dairy without using any animals, it’s clear to me that it would be permissible to consume that.

My intuitions are similar when it comes to body parts. If it were possible to artificially produce human organs it would be morally permissible to trade them. There was even some guy who served his body to his friends after his foot got amputated. Even though his action feels weird and repulsive, it doesn’t look like something morally forbidden to me. I think the real reasons behind the controversy over organ trade are pragmatic or distributive ones over whether consensual trade of organs could leave disadvantaged people worse-off/exploited rather than concerns over whether organs can be commodities.

The only thing that cannot be commodified is individuals. For that reason I don’t think violating this constraint has any weight when it comes to animal products.

Candidate constraint 4: Violating animals’ bodily integrity by using their body parts is non-negotiably wrong.

Examples:

Body parts of humans and non-humans

This argument makes a somewhat controversial assumption that dead individuals have rights, but I think it has some weight. We tend to have strong taboos against using the body parts of humans. Very recently, Harvard University removed human skin from the binding of a very old book in its library. I’m not sure if the real reasons behind this taboo is a pragmatic one or an intrinsically important moral constraint, but uncertainty is in favour of the restrictive advocate here.

Bob Fischer argues that we respect the bodies of dead individuals primarily because we respect their preferences from their lifetime about how to treat their bodies. He further argues that animals don’t have preferences about how their corpses should be treated. I’m not sure about either of these.

I don’t have arguments against non-negotiability of this constraint.

Other things to note are that:

This constraint forbids the consumption of animals’ body parts, not merely purchase.

This constraint covers merely the bodies of animals but not animal products in general.

Candidate constraint 5: Respecting the self-ownership of animals. Consumption of animal products steals something that fundamentally belongs to animals and therefore it is non-negotiably wrong.

According to this line of argument, animals own themselves and things that come out of their bodies. Since milk, eggs and body parts always belong to the animals due to self-ownership, they can never be ours. Using animal products is using stolen property, therefore it is theft or exploitation.

This principle can also explain why not merely purchasing but also the consumption of animal products is wrong. It’s wrong because the products do not and cannot belong to you. When you use animal products you are perpetuating theft.

A slogan for this constraint includes “veganism is getting your hands off from things that don’t belong to you”. This was what Syrian vegan poet Al-Ma’arri was also getting at in 11th century:

“Do not unjustly eat fish the water has given up, and do not

desire as food the flesh of slaughtered animals,

Or the white milk of mothers who intended its pure draught for

their young, not for noble ladies.

And do not grieve the unsuspecting birds by taking their eggs;

for injustice is the worst of crimes.”

This principle comes closest to being extensionally equivalent with veganism(though it doesn’t cover products tested on animals). My opinion is that veganism as it is commonly practised is primarily an anti-theft movement(rather than a movement to minimise killings or right-violations).

I think this constraint has some weight. I don’t have any knockdown arguments against the non-negotiability of this constraint. But it seems to me that at the very least this constraint is weaker than some of the previous constraints. If I had to choose between violating constraint 1 versus violating constraint 5, I think from the perspective of animals it seems much more important to prevent further harm and avoid violating constraint 1. The victims would care much more about not being harmed compared to the usage of their stolen property with zero impact on their life experiences.

For example, in a case where I have to choose between using my old duvet that is filled with wool versus purchasing a new plant-based duvet and harming additional animals as a result of cotton farming; it seems to me that I should keep my old duvet and avoid harming additional animals. I’m assuming that a duvet is part of the minimum consumption set for a healthy life, so we’re entitled to have at least one duvet. In general, it seems much more important to avoid causing additional animal killings and animal suffering than to avoid using products stolen from animals. Given that constraints 1 & 2 seem to be negotiable and that constraint 5 seems weaker than constraints 1 & 2, it seems to me that constraint 5 is also negotiable.

I’m also not sure if there is a deep difference between stolen body products versus stolen products in general. When it comes to raw materials stolen from human communities, my intuitions in favour of non-negotiability are much weaker.

I personally find my intuitions in favour of property rights much weaker than my intuitions in favour of bodily integrity, but I don’t know if that’s a common belief. For example, I have negative attitudes towards violating the bodily integrity of people with extreme mental disabilities. But I’m much less emotionally invested in the notion of property rights when it comes to people with extreme mental disabilities.

Candidate constraint 6: Harming too many animals for very little benefit is non-negotiably wrong.

This constraint is more exploratory. I am not satisfied by how I constructed it but I’m adding it in case someone might be interested in developing it further.

Assume we rank behaviours by their harm to animals and by their benefit. Avoiding using medically necessary drugs is very costly. Avoiding caviar isn’t costly at all. Use of gelatin in toilet papers harms very few animals, and avoiding toilet papers is also moderately costly and so on.

Maybe there is a conversion rate between benefits to us and harm to animals. Some amount of benefit justifies some amount of harm to animals during consumption. That doesn’t mean that initial production of harmful products is morally permitted. For example, we should still strive for a world where gelatin isn’t used in the production of toilet papers. The benefit we derive from toilet papers justifies using them only as a consumer.

But there is a limit to how much we can harm animals for our own benefit. Beyond that limit, it’s wrong to harm animals. There is also a further limit, after which the harms to animals are so serious that behaviours beyond that limit are non-negotiably wrong.

I think a merit of this kind of constraint is that it addresses “grey area fallacy” complaints of restrictive advocates. Pragmatic advocates tend to list examples of common animal product uses by vegans to show that we’re all harming animals to some extent. Restrictive advocates believe it’s fallacious to argue from “We’re all harming animals to some extent” to “We can negotiate for anything as long as it’s the best deal we can get for the animals”. Framing the constraint this way addresses their concern.

Nonetheless, I’m unsatisfied by this constraint for several reasons. First, it’s not clear which benefits are to be included. Assume that we merely include benefits to the agent. In that case this constraint seems repugnantly selfish. Animal product use for the benefit of the agent can save the agent at most. One single person. It reeks of selfishness to me to claim that use of animal products would be forbidden to save dozens of animals while it would be permissible for me to use animal products to save my own life.

I don’t think such selfishness is a devastating flaw. There is a tendency in “common-sense morality” to privilege the agent in such ways. There usually is a higher bar for meddling into other people’s business compared to protecting ourselves. For example, some believe that i. We can kill an innocent threat to save our own lives. ii. It would be forbidden for us to kill an innocent threat that threatens the lives of 2 strangers. Similarly, it might be consistent to claim that benefits to others should be excluded for this constraint.

Nonetheless, ascribing zero moral weight to benefits to the animals would be very unattractive. For that reason I expect many restrictive advocates will be willing to include benefits to animals when they compare benefits to harms. In that case, the distance between the positions of pragmatic and restrictive advocates would be reduced to some extent.

But I think restrictive advocates would be insistent in excluding some benefits to animals. Consider the following examples:

Some goats are stranded on a cliffside. You need a rope to save them from starvation. The only equipment available nearby is made from wool. You buy the wool rope and save the goats.
Bribe: Someone credibly promises to treat 10 animals with painful illnesses for free if you wear a wool sweater.
Threat: Someone credibly threatens to torture 10 animals if you don’t wear a wool sweater.

I think the restrictive advocate would be willing to harm animals in cases like 1, but would be unwilling to harm animals in cases like 2 and 3. I can understand the motivation behind having a moral theory that is impervious to blackmailing and bribing by other agents.

We don’t want to be willing to commit wrongdoing to change the behaviour of other agents. But I’m not sure what exact principle could capture this intuition. Here is a problematic case:

4. Precious painting: You use a woollen rope to save a precious painting from a fire. You sell the painting and use the money to treat illnesses of dozens of animals.

Saving the painting is an action to change the behaviour of other agents. The buyer of the painting could have offered us the same money even if we didn’t save the painting. We can protest that “I shouldn’t have committed wrongdoing for you to do the right thing”. But I believe animal product use in this case should be permissible just like it is in the stranded goat scenario.

To be honest I haven’t seen restrictive advocates provide adequate criteria for the relevant benefits to be included. A more common argumentative strategy is to deny that we’re making a choice when we use animal-based electronics, flights or drugs. These are categorised as “necessities” and people insist avoiding these products is literally impossible. This strategy seems disingenuous to me. People should just admit that refusing to use these products is just too costly (rather than impossible), and have a proper theory for what kinds of benefits could justify animal product use.

Restrictive advocates might protest that “there are no common realistic scenarios in which we have to use animal products or harm animals to help other animals. No one’s donation is conditioned upon me eating animal products. Why should we bother with all of this”. The problem is that according to pragmatic advocates, the benefits of pragmatism are much larger than the benefits we get from “unavoidable” animal product uses.

Assuming that transitioning a large university to Meatless Mondays reduces the animal product consumption of the students by 1%; working on a Meatless Monday campaign might prevent the suffering of 10000 animals per year at the very least. Restrictive advocates should have a non-selfish explanation for how both of these statements are true:

It’s morally permissible for me to buy tickets for a transatlantic flight even if such flights cause around at least 0.2 foreseeable vertebrate killings per ticket^[1].
It’s morally forbidden to utter the sentence “Let’s introduce Meatless Mondays to our campus” to prevent the killing and suffering of 10000 animals per year.

Similarly, there is something that reeks of selfishness to me in claiming that it’s permissible to swallow drugs made from animal bones to relieve my joint pain, but it’s forbidden to utter the sentence “Costco should go cage-free” to relieve the joint pain of millions of hens. Harms to animals we tolerate in our daily lives are for trivial benefits compared to the stakes involved in pragmatic advocacy.

A second problem I have with this constraint is that the “benefits” are unclear for institutional agents. Should it refer to the total profits for the company? Profit per shareholder? Impact on the probability to get re-elected? Or should it refer to the cost for the person we negotiate with?

When you negotiate with company representatives, they have limited power over what the company can do. Committing to transition their entire company to plant-based would get them fired in most cases. That is a very major cost. Would that licence negotiating for less than 100% plant-based?

A third problem for this constraint is to find a theoretical justification for the location of the boundary. It’s difficult to find an argument to justify at what point harm to animals becomes non-negotiable. I’m personally less concerned by this problem.

A final remark about this kind of constraint is that “benefit” is very context-dependent. Some might find it very difficult to avoid dairy but might find it easy to avoid transatlantic flights. Drawing the line by “benefit” allows for different people to have different responsibilities with respect to animal products. I don’t think this is a major problem for the constraint itself.

However, the existing framework used by restrictive advocates is almost exclusively product/service based. People start by thinking whether a certain product/service is vegan, rather than focusing on the benefits and the costs of the specific behaviour. Claiming that “in some circumstances, dairy consumption is negotiably wrong whereas air travel is non-negotiably wrong” might be an unpalatable position for the restrictive advocates.

^
“The mortality cost of carbon” by Daniel Bressler estimates 2.26 ×10^-4 additional human deaths per tonne of CO2 emitted. I assume one tonne of CO2 emissions by a transatlantic flight, and 1000x more wild vertebrates being killed by climate change compared to human beings. For context, there are at least 10¹³ wild vertebrates compared to 10¹⁰ humans. 0.2 additional vertebrate killings might seem small, but it’s a much higher number compared to the number of killings caused by purchasing 10 litres of milk. 10 litres of milk correspond to half a day in a dairy cow’s life.