Introduction: Thinking About Vegetarianism

It is difficult to find, or come up with, arguments for vegetarianism. To anyone familiar with the academic literature on the ethics of eating meat, or anyone who has ever offered sincere thought to the issue, this might sound like a silly claim. Is the philosophical literature not replete with arguments for vegetarianism? Are there not arguments for vegetarianism in the work of some of the most prominent ethicists of the twentieth and twenty-first centuries? Are the reasons in favor of vegetarianism not obvious to anyone who has read about the suffering of animals in agriculture and the environmental impact of the meat industry?

In order to answer these questions, vegetarianism needs to be distinguished from a range of other diets. Vegetarians are people who do not eat meat. For our purposes, “meat” includes the flesh or body of any animal, including fish and invertebrates. Less restrictive diets include what is often called omnivorism , which is a diet including a more or less “normal” amount of meat. Using omnivorism in this sense is problematic: strictly speaking, all humans are omnivores. Omnivory is a matter of biology, not of practice, culture, attitude, or ethics. Nonetheless, this common convention will be followed. Precisely how much meat an omnivorous diet contains will depend on the social, economic, and cultural context. Vegetarianism must also be distinguished from various kinds of demi-vegetarian diets, including pescetarianism (like vegetarianism, but including the meat of fish) and reducetarianism (which involves a conscious attempt to reduce, but not necessarily eliminate, meat – and perhaps other animal products – in one’s diet). Crucially, however, it must also be distinguished from more restrictive diets, including veganism. Vegans avoid all animal products in their diets (and likely more broadly) – not just meat. And what we find when we consider prominent and mainstream arguments for vegetarianism is that they are really arguments for veganism.

Can anything be said for vegetarianism, rather than veganism? Specifically, are there any arguments that should convince individuals to be vegetarian, but not vegan? Exploring that question is the purpose of this chapter. Thus, the chapter will not review standard arguments for and against the consumption of animal products. For recent and in-depth reviews of these questions, readers are invited to consult Abbate 2020, Doggett 2018, Fischer 2018, Katz and McPherson 2019 [present volume], and McPherson 2018. Instead, this chapter will explore arguments for vegetarianism in contrast to omnivorous diets, but also – perhaps more importantly – in contrast to vegan diets.

In the second section, the chapter will indicate how classic cases for vegetarianism are actually cases for veganism. It will then move on to explore, in the third section, some relatively straightforward reasons that people may favor vegetarianism over veganism that do not really draw upon moral reasons at all. In the fourth section, the chapter will ask whether one could defend vegetarianism as being enough for one to do one’s duty and critique veganism as being overly demanding. We will see, though, that any argument for vegetarianism on these grounds faces some tricky problems. While these problems may not be insurmountable, they do mean that the argument holds only very contingently. In the fifth section, the chapter will turn to the arguments of Tzachi Zamir, the only writer in the philosophical literature to argue at length for vegetarianism over veganism. His case relies on a vision of a vegetarian utopia contrasted with a vision of a vegan utopia and a claim about effective campaigning on behalf of animals. Nonetheless, we will see that there are serious problems with his arguments. In the final substantial section, the chapter will explore whether there may be something about the eating of animals’ bodies or their flesh that makes meat-eating, in contrast to eating eggs and dairy products, a particular wrong. If there is, that could ground a case for vegetarianism without also grounding a case for veganism in principle and perhaps – crucially – in practice. A short conclusion follows.

Arguing for Veg(etari)anism

There are plenty of cases for vegetarianism in the academic literature, but, on closer inspection, they turn out to be cases for veganism. By way of example, let us look to the arguments for vegetarianism that revolve around harm to animals. Other kinds of arguments – including environmental arguments – will be explored later.

Let us take Robert Nozick’s case for vegetarianism. This appears in Anarchy, State, and Utopia, which is one of the most-read works of the twentieth-century political philosophy. He begins by asserting – reasonably, we might think – that “[a]nimals count for something” (1974, p. 35). “Suppose,” he says, “that eating animals is not necessary for health and is not less expensive than alternative equally healthy diets.” Thus, the advantage of “eating animals is the pleasures of the palate, gustatory delights, varied tastes” (1974, pp. 35–36, emphasis Nozick’s). The question is whether “they [that is, the pleasures], or rather … the marginal addition in them gained by eating animals rather than only nonanimals, outweigh the moral weight to be given to animals’ lives and pain” (1974, pp. 36–37, emphasis Nozick’s). The death of these animals, Nozick says, is surely incidental to the pleasure gained by eating them – but that does not prove that the eating is permissible:

Suppose then that I enjoy swinging a baseball bat. It happens that in front of the only place to swing it stands a cow. Swinging the bat unfortunately would involve smashing the cow’s head. But I wouldn’t get fun from doing that; the pleasure comes from exercising my muscles, swinging well, and so on. It’s unfortunate that as a side effect (not a means) of my doing this, the animal’s skull gets smashed. To be sure, I could forego swinging the bat, and instead bend down and touch my toes or do some other exercise. But this wouldn’t be as enjoyable as swinging the bat; I won’t get as much fun, pleasure, or delight out of it. So the question is: would it be all right for me to swing the bat in order for me to get the extra pleasure of swinging it as compared to the best available alternative activity that does not involve harming the animal? (1974, p. 37, emphasis Nozick’s)

Nozick can see no way to justify the eating of meat that does not also justify the swinging of bats in the above case. It could be that an omnivore could accept this – eating meat and swinging bats are acceptable. Nozick doesn’t say this explicitly, but he presumably holds that such a judgment flies in the face of the intuitive notion that animals count for something, or at least the unstated additional assumption that this “something” is non-negligible (cf. Milburn 2017, 2018b). Instead, in Nozick’s view, “the extra benefits Americans today can gain from eating animals do not justify doing it. So we shouldn’t” (1974, p. 38, emphasis Nozick’s).

In this argument, Nozick talks about meat and, in other places, speaks of his vegetarianism. But it should be clear that the argument actually works far better as a case for veganism than as a case for vegetarianism (Milburn 2017, fn. 2). Eggs and milk are not necessary for human health, and alternatives are accessible to many in the West. The farming of eggs and milk involves the infliction of a great deal of death: male chicks are killed shortly after birth, while male calves are killed as unnecessary or are raised for meat (and thus killed). Meanwhile, both hens and cows are killed at a fraction of their lifespan as their productivity drops. These practices – and many others common in the industries – lead to a great deal of suffering. Such practices are required if we want to have access to eggs and milk at prices anywhere near as cheap as those currently commonplace. So, exactly the same argument that Nozick offers for vegetarianism can be run for veganism. Indeed, Nozick’s arguments resemble those of Anna Charlton and Gary Francione (2016), though the latter explicitly focus on veganism, rather than vegetarianism. Charlton and Francione paint a picture of a man named Fred, who keeps animals in his home in pain and then kills them, simply because doing so offers him pleasure. But we recognize that this is not a good reason for him to torture and kill these animals. When it comes to eating animals, “we are all Fred” (2016, p. 296). As such, unless we wish to jettison our view that animals – to parrot Nozick – count for something, we should be vegans. (Here, Charlton and Francione are offering their “common-sense” case for veganism, and not defending their “abolitionist” position.)

Nozick’s case for vegetarianism thus sounds more like a case for veganism. The same sort of observation can be made about the respective cases for vegetarianism of Peter Singer (1980) and Tom Regan (1975), the twentieth century’s foremost animal ethicists. The precise details of these arguments need not concern us, in part because they are not dissimilar to Nozick’s, but especially as – perhaps with a few caveats – neither Regan nor Singer would have too much trouble admitting that they were ultimately talking about veganism. The word veganism, it is perhaps worth noting, was little used in the USA in the 1970s.

A final example is offered by the arguments of Carol Adams (1990), whose The Sexual Politics of Meat offers, according to its subtitle, “a feminist-vegetarian critical theory.” In the book, Adams links the eating of meat to sexism, tying together the treatment of animals in a meat-eating culture and the treatment of women in a patriarchal culture. However, this link holds no less with milk and eggs than it does with meat. Indeed, eggs and milk are the key examples of “feminized protein” – the production of protein from plants via the female reproductive system. Thus, Adams’s critique of patriarchal eating practices is far from complete when addressing only meat-eating; there is room for feminist critiques of dairy and eggs to be at least as fervent as feminist critiques of meat. Adams now describes her position as “feminist-vegan” (2010) and has written that though The Sexual Politics of Meat “was subtitled A Feminist-Vegetarian Critical Theory, in terms of [her] understanding of feminized protein, and its use and abuse of female bodies, it could have been A Feminist-Vegan Critical Theory” (2017, p. 23).

Let us take this as indicative: though the philosophical literature is ostensibly replete with arguments for vegetarianism, these are actually arguments for veganism. This is due to the very real existence of death, suffering, and exploitation in the egg and dairy industries.

Personal Motivations for Vegetarianism

We may think that all of this is by the by and that – philosophers’ arguments for vegetarianism aside – we can easily come up with reasons to be vegetarian that are not reasons to be vegan. One set of arguments will drop the focus on the death and suffering of animals as the motivating concern for vegetarians and look to other factors. Many of these reasons – however important they are for individual vegetarians – should not be taken to be particular good reasons for other people. They are thus not, no matter how good or bad, the right kind of “reasons” for vegetarianism. Ethicists are concerned with all reasons for action, but they are most interested in reasons for action that do not depend on idiosyncratic beliefs. The “ethical” perspective taken by this chapter seeks reasons for vegetarianism that are not unique to a particular person, culture, or people. Instead, it seeks reasons that should be applicable generally. This does not mean that they will not rest upon potentially controversial claims about science or ethics – but these claims may be ones that should be generally accepted, whether or not they are generally accepted. The reasons explored in this section are not like this. They are personal, or rest upon beliefs that are not the kind of beliefs that an ethicist would typically call on all people to hold.

One reason for vegetarianism could be health. Some vegetarians may hold that vegetarianism has health advantages over omnivorism and veganism. Equally, of course, many vegans hold that their diet has advantages over vegetarianism and omnivorism, and plenty of omnivores hold that their diet has advantages over veganism and vegetarianism. Were it the case that these vegetarians were unambiguously correct, this might – along with some claims about the ethics of healthy eating – form the basis of a moral argument for vegetarianism. There is, however, likely some truth to the claims of the vegetarians, the vegans, and the omnivores (especially when atypical dietary needs are considered). The healthfulness of a diet is not a simple matter. Well-planned, balanced vegan, vegetarian, and omnivorous diets can all be healthful, though – crucially – all can carry risks that are not associated with the others. For example, while iron deficiencies can be associated with both vegetarian and vegan diets, these diets are also associated with lower risks of heart disease (Melina et al. 2016). So, while some vegetarians may be motivated by health, it remains to be seen that this argument should convince others – there is no evidence of a consensus among dietitians about the all-things-considered “best” diet for health, and different kinds of health concerns are going to pull particular people in different directions.

Some vegetarians might be motivated by religious concerns. So, for example, many Jains, Buddhists, Rastafari, Sikhs, Seventh-day Adventists, and Hindus follow vegetarian diets. There are also minority vegetarian traditions in other faiths, including Christianity and Judaism, as well as individual believers of a range of faiths (and none) who see religious or spiritual significance to their vegetarianism. Other religious practices may blur the lines between veganism, vegetarianism, and omnivorism. For example, Hindu vegetarians will frequently abstain from eggs as well as meat, while restrictions on particular meats are typical among even omnivorous Jews and Muslims. Naturally, if someone holds that their religious doctrine (or personal spiritual development) demands, encourages, or endorses vegetarianism but does not demand, encourage, or endorse veganism, that will be a very good reason for them to be vegetarian, but not vegan. It is true that veganism is less typical as a religious practice – though certainly not uncommon of among many of the broadly vegetarian groups, including Jains, Rastafari, and Seventh-day Adventists. Meanwhile, members of some new religious movements – such as the followers of “Supreme Master” Ching Hai – are overwhelmingly vegan. Reasons for being vegetarian that are grounded in religious beliefs are not transferable to people who do not share these religious beliefs. Nonetheless, healthy internal debates in particular religions’ ethical discourses on veganism and vegetarianism should not be overlooked by scholars interested in the ethics of eating meat (see, e.g., Linzey and Linzey 2018) – they should not, however, be taken to be universalizable in the way that arguments from philosophical ethics can be.

Similarly, some vegetarians might stick to their vegetarianism, even in the face of a vegan alternative, because vegetarianism is part of their cultural inheritance in a way that veganism is not. So, for example, someone who was raised as a vegetarian may not have any attachment to the goal of reducing harm to animals. Vegetarianism need not even be something that they have particularly thought about. There is nothing fundamentally irrational about this. Again, though, we should not take a given person’s cultural practices to be normatively significant for the rest of us – beyond, potentially, as something to be respected, within reason.

We can note a final reason for being vegetarian that can be quickly dispensed with as entirely personal: some vegetarians may have an aesthetic objection to meat, but not to eggs and dairy – they do not like meat’s taste, texture, or similar. Thus, these people have a perfectly good reason to be vegetarian but not vegan. However, unless this aesthetic reason is tied to something deeper, there is no reason to think that their vegetarianism is something that anyone else should take up.

These, to repeat, are not the kinds of reasons that should serve to convince others. Perhaps they could become the right kind of reason. For example, if there really was a God commanding us to be vegetarian rather than vegan, that could be normatively significant. But, short of proving that there is such a God, the religious vegetarian should not expect to convince skeptics. What is more, the person who is vegetarian for one of these reasons should be ready to acknowledge that there may be ethical reasons that override their more personal reasons. For example, while someone belonging to a culture that has a given practice has a good reason to engage in that practice, they should be ready to admit that, if the practice is unethical or unjust, their cultural practice should give way to good moral sense.

Vegetarianism as a Middle Way

Are there any arguments for vegetarianism (and not veganism) that are not deeply personal? Surely some arguments – in everyday dialogue, if not the philosophical literature – will point toward vegetarianism’s status as a “middle ground” between veganism and omnivorism. So, we could acknowledge the positive arguments for veganism related to the moral status of animals, or the environmental damage associated with animal agriculture, but nonetheless hold that we are doing “enough” if we switch from omnivorism to vegetarianism and argue that an imperative to switch to veganism is too demanding.

This kind of argument is vulnerable to at least two kinds of counters. The first observes that vegetarianism need not be thought of as a suitable compromise position between omnivorism and veganism, insofar as meat need not be more harmful (in environmental, animal-welfare, or all-things-considered terms) than eggs and milk. Mark Budolfson (2015, 2018) is one philosopher who has collated empirical evidence to support this claim by calculating “harm footprints” of different foods. Much of this supports the claims of the vegetarian motivated by environmental concerns. Take Budolfson’s figures related to beef production. He suggests greenhouse gas emissions of 102 kg CO2eq (i.e., the release of greenhouse gases doing the equivalent damage of 102 kg of carbon dioxide) per kg of protein produced and 93 kg CO2eq per 10,000 kcal produced. Cabbage, on the other hand, sees 25 kg CO2eq per kg of protein and 13 kg CO2eq per 10,000 kcal. When it comes to water usage, beef requires 75,969 liters to produce 1 kg of protein and 60,645 liters to produce 10,000 kcal of energy. Cabbage requires 21,875 liters and 11,200 liters, respectively. Cabbage significantly wins out. As the vegetarian might expect, milk ranks somewhere in the middle, producing 60 kg CO2eq per kg of protein (which needs 25,270 liters of water) and 31 kg CO2eq per 10,000 kcal (needing 13,049 liters of water).

The trouble is that once we start adding in other possible products, the omnivore < vegetarian < vegan picture starts to fall apart. One particularly disruptive example is mussels, which produce a mere 6 kg CO2eq per kg of protein and 8 kg CO2eq per 10,000 kcal. (Budolfson provides no data on how much fresh water they require, but, given that they are farmed hanging in the sea, one imagines very little.) If the vegetarian is motivated by finding a “middle ground” between omnivorism and veganism, it is not obvious that they should be dropping mussels rather than milk. (It is worth acknowledging, incidentally, that Budolfson’s picture also troubles the vegan argument.)

What if the motivation for vegetarians is not environmental but animal-focused? The simple fact is that – counterintuitively or not – it is not clear that there is less harm in the farming of eggs and milk than there is in the farming of meat.

Consider, by way of example, milk farming. Karin Kolbe (2018, pp. 472–474) identifies four key areas of suffering resulting from just one practice in the dairy industry: the separation of cow and calf. This is something practiced even on organic, high-welfare farms. It leads, Kolbe argues, to deprivation to the mother, who would prefer to be with the calf; deprivation to the calf, leading to health problems; welfare problems associated with the other farms to which the calves are sent; and the slaughter of calves at a very young age. Kolbe also offers detailed statistics for the number of calories of milk produced per death of a male calf – neither deaths of female calves nor the cows themselves are included in this. Within the European Union, this varies from 473,668 per male calf in Slovenia to 1,175,992 per male calf in Denmark, but the average is 804,712 per male calf. While this is greater than the calories produced by killing smaller cattle bred for beef – a female highland cow (the smaller sex of a small breed) provides some 500,000 calories – even the highly efficient milk production in Denmark cannot compare to meat production in terms of calories per death when it comes to larger breeds of cattle raised for beef. A male Charolais produces some 1,750,000 calories of meat, while a Blonde d’Aquitaine bull produces 1,625,000 (Kolbe 2018, pp. 474–476). Kolbe thus concludes that the production of dairy is a greater ethical problem than the production of beef: “suffering of animals and death per calories created are higher in dairy farming than in meat production” (2018, p. 469).

Perhaps something similar is true of eggs. Chickens are very small, and though they lay lots of eggs in their lifetimes, they are unable to produce anywhere near the amount of food through laying eggs as a cow can through being slaughtered. Thus, a lot of chickens living short lives full of suffering and then being killed when their productivity drops might seem to be a greater ethical problem than a single cow being killed – even if she, too, has led a life of suffering followed by an early death. More animals means more suffering and more death. This leads Nick Cooney (2014, pp. 3–9) to argue that it is much more important for those concerned with reducing animal death and suffering to eliminate eggs than eliminate beef. Consuming eggs leads to more animal deaths a year (2 chickens per omnivorous American versus 1/8 a beef cow per omnivorous American), more days of animal suffering a year (a year per omnivorous American for egg-laying hens, 23 days per omnivorous American for beef cows), and indeed – Cooney argues – more intense suffering.

Finally, we can add that mussels once again create a puzzle – at least if Budolfson’s assessment of them is right:

…mussels have essentially no animal harm footprint at all – partly because mussels are not conscious and so harvesting them does not involve animal harm that has any important weight, and partly because the land and water footprint of mussels is very small[.] (2018, p. 91)

The point of these quantitative explorations of harm is the following: if vegetarians are genuinely interested in identifying a “middle ground” between veganism and omnivorism, it is not clear that vegetarianism (as we understand the term) is it. Now, perhaps we can quibble with the numbers or assumptions of Kolbe, Cooney, or Budolfson. Regardless, vegetarians must do some careful arguing to show that their position is a suitable middle ground. We should not jump to any conclusions about (particular) meats being so much worse than milk and/or eggs.

But this leads us to the second worry. Even if vegetarians can justify vegetarianism as a suitable “middle-ground” approach between veganism and omnivorism, they will need to justify favoring the “middle ground” rather than the “extreme” – i.e., if these vegetarians acknowledge worries about harms to animals or the environment, why be vegetarian rather than vegan? Two possibilities present themselves. One will be that adopting the vegetarian position is more likely to be effective as a political matter. This is something that will be touched upon in the next section. The second is that veganism is too demanding – i.e., that it calls for a high level of change in our lives, one that is disproportionate to the harm it aims to prevent (or speak against).

No doubt it is sometimes true that switching to a vegan diet is difficult and, for individual vegetarians, that will explain their decision not to go “all the way.” The (putative) relative difficulty of following a vegan diet, of course, is highly contingent (and the same is true when it comes to political expediency). It is going to be far easier to follow a vegan diet in some places than in others, and – in much of the Western world, at least – it is quickly becoming far easier to follow a vegan diet than it was. And, of course, vegans may challenge the moral relevance of this difficulty. Even relatively difficult changes – they might say – are required to combat the horrors of animal agriculture and/or climate change.

There is, however, an important contextual argument that is open to the vegetarian in response to these vegan challenges. This is that vegetarianism is an established, recognized dietary identity in a way that other “demi-vegan” diets simply are not. Thus, the vegetarian could say, given that (in a given context) veganism is difficult for them and given that (in the context in question) vegetarianism is a recognized identity in a way that other possible dietary identities are not, vegetarianism is the appropriate diet for them. This recognition of vegetarianism – where it exists – is significant for three interrelated reasons.

First, it simply makes things practically easy for the individual in question: to put it in blunt terms, they can ask for the vegetarian menu in restaurants and look out for “suitable for vegetarians” messages on food packaging. But there is another side to this: there is a clear, bright line for the individuals to follow. They can tell relatively easily whether something is “suitable” for them – if it contains dead animals, it is not. If it does not contain dead animals, it is fine. This can help deal with the kinds of biases that, as individuals, we face. Clear and precise rules are easy to follow; vegetarians are less likely to waver in their commitments than people following more complicated diets for ethical reasons (see Rothgerber 2015).

Second, in belonging to a movement – the vegetarian movement – the individual vegetarian is more likely to be practically (including politically) efficacious than if they chose to follow their own idiosyncratic diet. This is true even if their own idiosyncratic diet would be, were it adopted by all vegetarians, more efficacious than vegetarianism. This efficacy is not limited to the point about the numbers involved. It relates to its communicability: the vegetarian cares about animals, so she does not eat them. On the other hand, the adherent of an unnamed demi-vegan diet cares about animals, so she does not eat some animal products and some plant products (or perhaps does not eat some products produced in certain specified ways). It is not hard to see whose message is easier to transmit – not only is the vegetarian’s message simpler, but it can be transmitted in a single, familiar word.

Third, vegetarianism is already recognized as a less extreme form of veganism – what is more, vegetarianism is likely far more recognizable than is veganism. Thus, whether or not the vegetarian is right to think that vegetarianism is really a good middle ground between veganism and omnivorism, they will be perceived as treading that middle ground, and that may be what matters, if their vegetarianism is intended – in part or in whole – as a symbolic sociopolitical statement.

We seem to have reached a good argument for vegetarianism, and not veganism. This argument, however, depends on contingent, wholly changeable (putative) facts – facts, indeed, that will vary geographically and historically. These concern veganism’s relative inaccessibility, vegetarianism’s relative accessibility, the relative familiarity of vegetarianism, and the relative unfamiliarity of other possible “part-way to vegan” diets. Though this kind of argument has some potential, the vegetarian may want a less contingent argument for their position. For that, we will have to look elsewhere.

Zamir’s Case for Vegetarianism

The only philosopher to argue at length for vegetarianism in contrast to both omnivorism and veganism is Tzachi Zamir (2004, 2007). Zamir argues that a vegetarian utopia is a more compelling vision – for both humans and animals – than a vegan utopia. The benefits to humans include, primarily, access to eggs and milk. The benefits to animals are assessed on three grounds:

  1. 1)

    Quantitative. More animals will exist in a vegetarian world than a vegan world. Now, this speaks in favor of a meat-eating world over a vegetarian world. This leads to…

  2. 2)

    Qualitative. Many animals in the contemporary meat industry do not have lives worth living. This, however, is a contingent point. Why could we not have a meat-eating world in which animals lived happy lives? After all, Zamir’s vegetarian world would be one much kinder to animals kept for eggs and milk – as he accepts, their lives are often very bad on contemporary farms – so why could he not propose a world much kinder to animals kept for meat? This leads to…

  3. 3)

    Teleological. “[I]t may be the case,” Zamir writes, “that a pleasant life should not be lived if it ends in a way that is immoral” (2004, p. 372). When we bring a being into the world with a plan – such as killing and eating them – we must ask whether such a plan “constitute[s] a misrecognition of what having a life means” (2004, p. 372). In what we might call a humane-farming utopia, where animals are bred for meat but are raised in humane ways, animals may live in a way that “is better than not living: but it is not hard to imagine someone saying that such a life should not be lived” (2004, p. 373).

Zamir’s conclusion is that a vegetarian utopia is able to benefit both animals (in quantitative terms) and humans relative to a vegan utopia, is able to avoid the qualitative harms replete in contemporary forms of industrial agriculture, and is able to avoid the teleological wrongs involved in farming for meat (“humanely” or otherwise). Thus, the vegetarian utopia is to be preferred both to the vegan utopia and the humane-farming utopia.

If Zamir’s conclusion was simply one about how ethical and political systems fostering high levels of respect for animals could, in principle, support ultra-humane forms of egg and milk production, then he would find himself in good company. For example, many recent approaches theorizing ideal forms of human-animal relationships in political philosophy – despite being written by vegan animal-rights advocates – have left room for the consumption of eggs and dairy (see, e.g., Cochrane 2012, pp. 86–89; Donaldson and Kymlicka 2011, pp. 134–132; Wayne 2013). Admittedly, these philosophers and political theorists likely see much less room for the consumption of eggs and milk in an ideal state than does Zamir. But they seem to be clear examples of what Zamir calls “tentative vegans” (2004, p. 367) – people who accept that, in principle, the consumption of eggs and milk is permissible but who argue that, currently, given the harms perpetuated against animals in the milk and egg industries, we should be vegan. But Zamir wants to go a step further than these theorists and advocates for vegetarianism (and not veganism) in practice, and not merely in theory.

His argument for this is thoroughly pragmatic: “selective consumption, rather than a total ban, allows pro-animal people to financially support institutions that take steps in the right direction” (2004, p. 374). He proposes three criteria to judge whether supporting an institution (e.g., eating eggs from this farm) is commendable. These are worth examining at some length.

First, the step forward must be substantial; not just any improvement justifies support for a flawed institution. For example, though Zamir holds that the institution of child labor (like the institution of egg production) is not inherently unjust, one should not buy from factories using child labor just because they, unlike their competitors, offer the children a break and a cup of tea (2004, p. 375). If Zamir is right, though, there is presumably some level of improvement that would justify supporting such factories. Zamir holds that, despite the continued killing of unproductive females and practically all males, egg and milk farms that allow their animals to roam freely are making a large step forward, and so should be supported (2004, p. 375).

Tentative vegans may well want to dispute this. Francione and Charlton – who, it should be noted, would be principled, rather than tentative, vegans – say that “free-range” policies “are to animal ethics what padded water boards for use at Guantanamo Bay” would be to prison reform (2017, p. 300). There are two sides to this critique. One is that the steps taken by these institutions are not large. The other is that, even if they were, supporting steps “forward” might miss the point, insofar as the institutions remain fundamentally unjust. (To be clear, on Zamir’s own framework, the killing of these unproductive animals is still a serious problem – he does not hold that the lack of freedom enjoyed by cows and chickens is the only, or even largest, injustice in the contemporary egg and milk industries.) To adapt Charlton and Francione’s example, it would surely be a large step forward for an institution which tortured innocent prisoners to stop torturing them, but we might think that we should never be supporting institutions that imprison the innocent – whether or not they torture them. (Zamir could respond that imprisoning the innocent is always wrong, while egg farming is not always wrong. We will get to this shortly.)

Second, Zamir holds that a choice to support egg and milk producers should consider the effectiveness of the action. Veganism is typically harder than vegetarianism, meaning that many potential converts could be lost. Zamir holds that this means that veganism is counterproductive, but this is actually far from clear. To justify that claim, one would have to weigh the impact of a smaller number of vegans against a larger number of vegetarians, and, in any case, one would need to clearly tie the impact of these groups to a particular goal; what matters is not just short-term reform, but – Zamir’s own words – “the overall political goal” (2004, p. 377). It is not simply a matter of judging how much impact one has over the next week, but a case of judging the extent to which the collective actions bring us closer to the ideal. Even if Zamir is right that veganism is counterproductive by (say) leading to greater harm to animals at the present time – and that is far from clear, as it is ultimately an empirical issue – it may be that it is more productive in the long run. The most effective route to the top of a mountain sometimes takes one downhill.

There are, Zamir thinks, limits to this strategic prudence. Zamir rejects demi-vegetarianism. Though prudential (i.e., though he takes it that demi-vegetarianism will attract more converts than vegetarianism or veganism), demi-vegetarianism, “like occasional molesting,” involves “occasional participation in a morally wrong act and is hence unjustified” (2004, p. 376). He rejects the obvious vegan response that support for the egg and milk industry is wrong. He is uninterested in assessing the merit of competing descriptions of vegetarian consumption – e.g., of the merits of describing actions “as supporting reform [versus] as supporting fig-leaf exploitation” (2004, p. 376). He justifies his rejection of the obvious vegan response by again referring back to the value of cooperating with reforming (but imperfect) institutions (2004, p. 376). This is insufficient. This point does not differentiate cooperation with a reforming egg or milk industry and cooperating with a reforming meat industry, which is what he needs for his argument to hold. All three industries involve, by his own admission, great wrongs – so why is he committed to supporting two of the industries, but not the third? (Note that, in any case, differentiating between these industries may not be as easy as Zamir seems to assume.)

Zamir seems to differentiate between the meat industry (on one hand) and the egg and milk industries (on the other) by arguing that the meat industry is inherently unjust, while institutions of milk and egg farming are not. He opposes the consumption of meat on the grounds that it “complet[es] a temporally extended wrong” – that is, it completes the wrong started by the farmer, thus bringing together the ethics of consumption and the ethics of production – and that there is a “conceptually distinct wrong of participating in a wrong practice, even when one’s consumption does not increase suffering” (2007, p. 48). (The importance of tying together consumption and production is that causal-impotence objections to ethical eating will contend that this act of eating has no impact on that act of production and thus that refusing to eat meat on a given occasion will have no impact on the animals harmed by the meat industry. This is discussed at length in the reviews of arguments about meat-eating cited in this chapter’s introduction, so will receive no further examination here. [See also Nefsky 2018.]) In an unpleasantly evocative passage, Zamir compares the eating of meat, even when the act of consumption does not contribute to harm, to the use of a child prostitutes when “the pain or harm done to the children involved” will not be increased by the abuse, either because the prostituted child will not notice “one more indistinguishable client” or because the “client” can make the situation better for the victim – for example, by “tipping generously or behaving nicer than other clients would” (2007, p. 48). Zamir, however, refuses to extend the same analysis to the consumption of eggs and milk:

…avoiding eggs and dairy because of the immoral production practices these rely on cannot be conceptualised in terms of avoiding the completion of or participation of a wrong in the same sense of the prostituted child or the killed animals example. Unlike eggs or milk, no reform to a child-prostitution establishment will justify participation. (2007, pp. 49–50)

The problem here is that child prostitution and meat-eating are not analogous in the way Zamir makes them out to be, even within his own framework. Thus, while Zamir has adequately distinguished child prostitution from eating eggs and milk, he – once again – has not adequately distinguished eating eggs and milk from eating meat.

Why are child prostitution and meat-eating not analogous in the relevant sense? Zamir actually repeatedly accepts that meat farming could be reformed to be consistent with his position: his “formulation of vegetarianism allows for eating and using animals that have not died from planned killing for the purpose of eating them” (2007, p. 49). Thus, crucially, his “position does not prescribe a ban on raising animals for the purpose of eating them after they die on their own” (2007, p. 49). Thus, while “no reform to a child-prostitution establishment will justify participation” (2007, p. 50), there is reform to a meat-producing institution that will justify participation. Consequently, the in-principle distinction between farming meat (on the one hand) and farming eggs and milk (on the other) dissolves. In both cases – according to Zamir’s own position – the practices are real-world problematic while ideal-world plausible. He thus cannot appeal to ideal-world possibilities to permit the consumption of eggs and milk in the real world unless he is also willing to appeal to ideal-world possibilities to permit the consumption of meat in the real world.

Perhaps the distinction is found in the third of Zamir’s criteria to identify farms worthy of support. He asks animal advocates to judge “the magnitude of the loss experienced by the exploited entity as part of obtaining a particular product” (2004, p. 377). Crucial here is the claim that chickens do mind having their eggs taken and cows do not mind having their milk taken. We can grant this (contentious) claim for the sake of argument, but its relevance for our actions here and now is questionable. The loss to the unproductive chickens and cows, and to male calves and chicks, is absolute. They are killed. “[T]aking eggs and milk does not create suffering and loss,” Zamir writes, and “[b]oycotting products, the taking of which does not create suffering, seems extreme” (2004, p. 378). If Zamir is arguing that contemporary egg and milk farming are free of suffering and loss, his claim is false. Calves are separated from their mothers, udders are afflicted with mastitis, and cows undergo the terror, injury, and deprivation associated with transport and slaughter whether they are given a degree of freedom or not. And the male calves and spent cows who are killed lose everything. (Similar could be said about egg farming.) And, what is more, Zamir holds that it is “excusable” to purchase eggs and milk from less humane farms if products from more humane farms “are implausibly difficult to obtain” (2004, p. 378). So even if more humane forms of farming were free from suffering and loss, Zamir would still excuse vegetarians who support those institutions that do cause suffering and loss. So it cannot be that the moral distinction between eggs and milk on the one hand and meat on the other is that the former’s acquisition does not in fact result in a “high magnitude of … loss” (2004, p. 377).

The distinction instead seems to be that their acquisition does not necessarily result in a high magnitude of loss. I have already shown that Zamir cannot commit to this being the distinction, as he is open to an ideal-theoretic meat industry in which animals kept for meat are not killed. However, let us imagine that things were otherwise and thus concede that eggs and milk do not necessarily involve the infliction of a high magnitude of loss even while meat does. If one is designing ideal institutions, this conceptual fact is important. But if one is talking about our actions here and now – and let us recall that this is a condition that Zamir has introduced to talk about our “non-ideal” actions – it is surely the practical fact that matters. Compare: shooting is a hobby that need not entail a loss for anyone. A highly responsible shooter could engage in her hobby only at carefully managed clay-pigeon shoots. There is nothing essentially harmful about shooting. But that fact seems to have absolutely no bearing on the condemnation we would rightly direct at someone who shoots people for sport – and the condemnation that Zamir does direct at those who shoot animals for sport (2007, p. 11). Equally, that the harm in some harmful animal agriculture is not essential to the form of animal agriculture cannot justify or excuse those who support the harmful animal agriculture in question.

Tentative vegans are thus going to have a lot of concerns with Zamir’s arguments for vegetarianism here and now, even if they are sympathetic to his claim that a vegetarian utopia is preferable to a vegan one. Zamir has a dilemma: given that he has failed to distinguish between the ethics of eating meat (on the one hand) and the ethics of eating eggs and milk (on the other), his argument leads him either to tentative veganism (i.e., veganism in practice, whatever the ideal-world diet) or conscientious omnivorism (i.e., omnivorism favoring putatively humane farms). Either way, his arguments do not lead to vegetarianism here and now.

It is worth closing our engagement with Zamir by questioning the “vegetarian” status of his utopia (2007, pp. 54–56; 104–106) – respectable though his vision may be, it just is not clear that it is really a “vegetarian” utopia at all. Now, it is, as he carefully argues, not a vegan utopia. But that does not make it a vegetarian utopia. We have already seen how Zamir is open to eating the bodies of animals who have died naturally and, indeed, of farming animals so that we might eat their bodies when they die naturally. Meanwhile – something Zamir does not address – technological advances have opened the door to the possibility of growing meat without killing any animals (see Donaldson and Carter 2016; Milburn 2016). Cells taken from animals can be grown in a laboratory environment into safe and edible meat, with commercialization not far away. There is no obvious reason that this “clean” meat would not be consistent with Zamir’s position. Presumably, then, there are a wide range of ways that Zamir’s “vegetarian” utopia is, ultimately, omnivorous. Not only, then, are there important worries about Zamir’s vegetarianism in practice, but it is unclear whether he even advocates vegetarianism in theory. The vegetarian seeking a grounding for her position would be well advised to look elsewhere.

Alternative Cases for Vegetarianism: Consuming Flesh, Consuming Bodies

Perhaps a vegetarian seeking principled arguments should move away from appeals to the wrong of producing meat. She could look instead to the wrong of consuming meat or animals’ bodies. I say meat or animals’ bodies as these are two different arguments, and, in practice, they will pull in different directions – for example, those who take that it is wrong to eat meat may be troubled by clean meat or even highly realistic plant-based “meats,” while those who object to eating animals’ bodies may not be.

Arguments pushing in this direction are present but underdeveloped in animal ethics. As such, this section is best understood not as an argument in favor of vegetarianism but as an indication of how such an argument could be built.

Concerns about the eating of animal flesh – rather than with the suffering and death necessary to acquire it – appear in work criticizing clean meat and plant-based meat analogues. So, for example, John Miller (2012) and Matthew Cole and Karen Morgan (2013) critique clean meat for reinforcing a problematic vision of the place that meat has in Western cultures – though their arguments may not be unique to Western cultures – thus pointing toward an objection to eating meat regardless of the presence or absence of suffering and death. Such critiques are not unique to critical theory. Bob Fischer and Burkay Ozturk (2017) argue by analogy to imitation-human-skin lampshades that there is something morally dubious about consuming meat analogues. By extension, perhaps they could argue that there is something wrong with consuming “real” meat independently of any contribution to death and suffering of animals. And Susan Turner (2005) argues against the production/consumption of meat (“fake” or otherwise) regardless of the suffering and death involved, arguing that animals may have a right “not to be represented as a mere resource” (2005, pp. 4–5).

Crucially, for current purposes, these kinds of arguments are about meat specifically – they need not generalize to other animal products. They thus provide the seed of an argument for vegetarianism, but not veganism. For example, it is clear that Turner’s argument does not extend straightforwardly to the use of eggs and milk, while it is an open question whether Fischer and Ozturk’s does. Others exploring these questions have explicitly affirmed that their arguments against meat do not extend to milk and eggs. For example, Rebekah Sinclair (2016) – drawing upon Adams’s vegetarian/vegan critical theory – challenges “meatless meats,” rejecting the idea that meat is food, but does not extend her to challenge to products seeking to mimic milk and eggs. This is because milk and eggs “do not imply a necessary animal death” (2016, pp. 231–232). This claim is vulnerable to similar counters to Zamir’s arguments, but the central insight – about the difference between meat on the one hand and other animal products on the other – is plausible. It makes sense to reject the idea that meat is food in a way that it does not to challenge the idea that, for example, milk is food:

Milk exists solely as food; in this sense, it is different from flesh/meat, which exists first as the body of an animal. To deny that milk is food seems to suggest that infants, human and nonhuman, who drink their mothers’ milk are consuming something that is not food. This seems to be straightforwardly incorrect. (Milburn 2018a, p. 272)

This passage is a response to a “metaphysical” challenge to the idea of milk as food. The same paper – though a contribution to the literature on animal rights – defends the status of milk as food against four other arguments: ethical, disgust-based, health-based, and racial (see Milburn 2018a, pp. 271–274). Where eggs fit in this kind of split is an interesting question – and one that anyone seeking to defend vegetarianism using this approach should address.

Let us turn to respectful treatment of animals’ corpses. Chloë Taylor argues that “the dominant Western worldview is deontological with respect to dead humans and utilitarian with respect to dead animals of other species” (2013, p. 95). What this means is that (we assume that) we respect an animal when we use (especially eat) as much of their corpse as possible, while using (especially eating) a human’s corpse is (seen as) the height of disrespect. Perhaps we can understand vegetarians as seeking to challenge this – and they can challenge it independently of concerns about the death and suffering of animals used in agriculture, thus providing a seed of an argument for vegetarianism, but not veganism. Sue Donaldson and Will Kymlicka are two theorists who challenge the disrespectful treatment of animals’ bodies and who thus oppose the eating of (some) animals’ bodies independently of any belief in the wrongness of killing animals or making them suffer. (To be clear, Donaldson and Kymlicka certainly do condemn the killing and hurting of animals.) They write that ideas of respectful corpse treatment

are culturally (and religiously) variable, marking the boundaries of community. This could mean that while there are some ways in which we should never treat a corpse – human or animal, citizen or foreigner – there are special obligations we owe to members of the community. … Perhaps, then, we ought to treat the bodies of domesticated animals the same way as human bodies in any given society or community, but the same obligation does not apply for corpses of those from outside the community. (2011, p. 151)

This means that the bodies of domestic animals cannot be used to produce meat, but does not commit the authors to any claim about milk or eggs. This argument – as indicated by Donaldson and Kymlicka’s talk of “members of the community” – is tied up in their particular “zoopolitical” theory. But similar arguments need not be.

Cora Diamond (1978) characterizes part of what it means to see someone as a person or, indeed, an animal as a “pet,” as seeing them as “not something to eat” (1978, p. 469). Someone drawing upon this kind of approach could argue that appropriately seeing a being as an animal – as a “living creature, or fellow creature” (1978, p. 474) – and appropriately relating to that animal would involve refusing to recognize their body as a resource and as food. This is the direction in which Diamond moves when criticizing Singer for apparently being “perfectly happy to eat the unfortunate lamb that has just been hit by a car” (1978, p. 472). Diamond admits that it “does normally, or very often, go with the idea of a fellow creature, that we do eat them” (1978, p. 475), but this is exactly what the vegetarian seeks to challenge. For Diamond, this idea of a “fellow creature” offers a real possibility for an argument for vegetarianism, though she does not fully develop it:

I introduced the notion of a fellow creature in answer to the question: How might I go about showing someone that he had reason not to eat animals? I do not think I have answered that so much as shown the direction in which I should look for an answer. (1978, p. 477)

Crucially, for our purposes, rejecting the idea that the “fellow creature” is an edible thing offers us the beginning of an argument for vegetarianism that does not quickly become an argument for veganism. That the fellow creature is not an edible thing does not preclude the fellow creature being a source of edible things.

Where do these half-arguments – arguments about the wrongs of eating flesh or eating corpses – leave us? Let us imagine that one or both can be expanded into a full argument for the wrongness of eating meat that says nothing about any wrong in eating eggs or dairy. On the one hand, such an argument could be wholly freestanding, in which case the vegetarian has a perfectly coherent argument for vegetarianism against both the omnivore and the vegan. But it would be a curious vegetarian indeed who was motivated not because of the suffering and death of animals but because of a relatively abstract concern with the wrong of eating animals’ flesh or desecrating their corpses. More likely, this argument could be combined with an argument about the wrong of inflicting death and suffering on animals. It is a wrong/bad involved in killing or hurting animals to acquire food, and, in addition, there is a wrong in consuming meat/corpses, but not in eating eggs and milk.

Where does this position leave us? Let us draw upon the terms introduced by Zamir. This argument gives us a compelling vision of a vegetarian, rather than vegan, utopia. Depending precisely what is meant by “meat” or “corpses,” our vegetarian utopia will certainly not permit Zamir’s corpse farming, may or may not permit clean meat, and perhaps will not even permit plant-based “meats.” But it will allow eggs and milk. As such, the vegetarian here and now would be justified in rejecting “principled veganism,” leaving them with the options of tentative veganism, vegetarianism, and omnivorism. This new argument (remember that we are yet to develop this argument – we are assuming that it can be developed) gives the vegetarian a clear reason to reject omnivorism, even in more humane forms, as they hold that there is a wrong in eating meat.

They are thus left with vegetarianism and tentative veganism. Given that they object to death and suffering, they are presumably not going to want to support the egg and milk industry – though perhaps, to again echo Zamir, they may hold that supporting relatively humane egg and milk farming will help the industries transition toward more just forms. But they certainly are not going to object to, say, eggs from backyard chickens (see Fischer and Milburn 2017) or milk produced in particularly humane ways, whether this is from no-harm farming or technological means (see Milburn 2018a). So, the “tentative vegans” imagined might actually be – to coin a phrase – “particularly selective vegetarians” in many real-world cases. They will not, though, be “particularly selective omnivores.” No matter how free from suffering and death meat production is, these vegetarians will likely reject it: roadkill (Bruckner 2015), and Zamir’s corpse farming are out. Clean meat, meanwhile, may be out.

To summarize: If these arguments about the wrong of eating meat/corpses can be made to work, they offer us a freestanding argument for vegetarianism, but not veganism. It would be an odd argument, though, as it makes no reference to the death and suffering of animals. But if it is combined with more standard arguments, it can still give us an argument for vegetarianism, and not veganism. Granted, this argument would permit vegetarians only to support the most humane forms of acquiring eggs and milk (or, at a minimum, relatively humane forms of acquiring eggs and milk) – but it would provide something that, to date, has been lacking in the literature in animal ethics and food ethics: a consistent, principled argument for vegetarianism.

Concluding Remarks

It is very hard to argue for vegetarianism. Most arguments for vegetarianism, on closer inspection, are revealed to be arguments for veganism. This chapter has reviewed the one developed argument for vegetarianism in the philosophical literature and found it wanting. However, it has offered two other routes that one could take to argue for vegetarianism. One is thoroughly pragmatic and highly contingent. This would see vegetarians defend their position on the ground that vegetarianism is a good “halfway house” to veganism and veganism is too demanding. The other, however, is more principled. It argues that eating meat or utilizing the corpses of animals is wrong, independently of any wrong involved in supporting harm to animals. This can provide a principled (if bizarre) freestanding argument for vegetarianism but, if combined with more standard arguments about the wrong involved in harming (or supporting harm to) animals, could offer vegetarians a principled basis for supporting some limited forms of egg and milk production while wholeheartedly rejecting farming animals for meat.