If we are to measure how much overall reduction in animal suffering would be achieved by the different approaches proposed (e.g., abolitionism/veganism, welfarism/sobriety, or sentientism), we must not forget the residual or new suffering that these methods are likely to cause, especially if these methods result in massive disruption of agricultural production.
1. The cultivation of plants also generates animal suffering. Although this suffering is generally much less than the suffering caused by livestock production (because most forms of livestock production themselves use specially grown plants, and with a yield <1), some forms of livestock production cause significantly less suffering than others, and according to some authors may even cause less suffering than plant crops that provide the same nutritional service. Although the latter claim is poorly supported, there is considerable uncertainty about the suffering actually generated by mass plant agriculture, and thus about the residual suffering resulting from an abolition of livestock farming.
2. Removing the large areas used by extensive livestock farming would mean returning these areas to wildlife. This could be positive for biodiversity (although this is not certain everywhere). But it could also increase net suffering, as the suffering of wild animals can be greater than that of farmed animals, especially if they are farmed according to good animal welfare standards.
3. Abolishing livestock farming could increase human suffering, if it turns out that a vegan diet causes health problems for at least a fraction of humanity, if the abolition process generates unemployment, rural exodus, food risk, disruption of eco-socio-systems and interference by states or international organizations at the local level.
4. The situation could even generate backlash, for example if the weakening of food systems leads some populations to increase predation on wildlife in regions where it is already threatened, or if the disappearance of areas currently dedicated to livestock increases the risk of mega-fires in a context of climate change.
PLANT PRODUCTION AND ANIMAL SUFFERING
Among the arguments proposed against a vegan approach, it was suggested that plant-based diets also cause a significant amount of suffering: many animals are injured or killed in the various operations associated with plant cultivation: animals killed or maimed in the cultivation process, by harvesters, to protect crops (from insects to wild boars…), and crops. It has been proposed that an omnivorous diet including large animals and based mainly on grazing may ultimately cause less overall suffering than a vegan diet. A first attempt was made in 2003 by Stephen Davis, suggesting that the principle of lesser suffering defended the consumption of large animals rather than veganism. The weaknesses of Davis’ argument have been widely pointed out by Gaverik Matheny, 2003, or by Lamey, 2007. However, the argument has generated thought about the potential suffering posed by plant-based agriculture, and challenges to the argument do not necessarily take into account Davis’ omissions of many species, or a possible underestimation of the number of wild animals present per unit area (which will be highlighted later by Fisher & Lamey). Another publication (Archer, 2011), attempted to repeat this argument for Australia. But all these discussions are based on rather fragile data, or valid only in a local context. To date, despite numerous theoretical discussions and publications on the subject, there is a serious lack of comprehensive field studies that include the full range of species likely to be affected by agricultural practices, both in terms of crop and livestock production. It is therefore difficult to reach precise conclusions about the suffering caused by different types of agricultural production (see for example the very wide range proposed by Fisher & Lamey, 2018, with figures potentially much lower but also potentially much higher than those proposed by Davis or by Archer, depending on the estimates, the number of agricultural operations considered, but also on the type of animals to be taken into account, this last point being more subjective: How does one evaluate the moral importance of invertebrates targeted by pesticides, should one consider that humans are responsible for the predation occurring as a result of the uncovering of large areas, etc.? ).
The nutritional value should also be corrected, as some estimates used in the public debate are based simply on calories (e.g. Middleton, 2009, at the bottom of this page).
However, it is certain that, as things stand at present, livestock farming generally causes a significantly greater number of animals killed and overall suffering than crop farming (notably because many forms of livestock farming, for example monogastric livestock farming, necessarily involve increasing the quantity of crop farming, in addition to directly killing animals), except possibly in certain situations that are not necessarily marginal but are in the minority today (e.g. livestock farms with large cattle that are essentially grass-fed, or using essentially grass + co-products of human consumption, and few or no products specifically cultivated for them (see the page [Livestock efficiency]), which it is doubtful can become the majority (except for a drastic reduction of other types of livestock, which would pose its own risks and difficulties), and whose increase would come up against ecological problems (greenhouse gas emissions from ruminants, deforestation) and would therefore be difficult to envisage.
It should be noted, however, that there is no production system that is free of animal suffering, and that the suffering generated by animal consumption varies greatly depending on the type of production and the efforts made by farmers to respect animal welfare.
Conclusion on this point: if the point is to recall that all production implies suffering and death, that the suffering generated by mass plant agriculture is far from negligible, and that there will therefore remain residual suffering in the event of the abolition of livestock farming, unless particular actions are taken to eradicate it too, the argument seems valid.
If it is to be argued that crop farming causes as much suffering and animal mortality as livestock farming, this does not seem generally valid, except in situations that are unlikely to become the majority in the near future. However, the question should probably be extended to all sentient animals (especially insects), to productivity for a set of nutrients (not only proteins, and even less calories) and to non-food productions associated or associable with meat production (wool, leather, etc.). In particular vs. problems related to cotton (note: only 30Mha, but high pesticide use), synthetic textiles, and various plant alternatives), and more broadly all plant co-products (plant co-products can also push the cursor back in favor of plant) and animals, to get a global view of the issue. Currently, as Fischer and Lamey, 2018 note, assessing animal suffering generated by plant crops is very difficult.
The consequences and feasibility of totally closed plant agriculture, as proposed by Fischer and Lamey among others, would also need to be assessed to definitively address the issue, or any alternatives that technology might make possible, as well as the possibilities for improved animal husbandry practices (less direct suffering and less use of plant products). It is important to note that knowing and recognizing the suffering caused by plant agriculture, even if it is less than that caused by livestock farming, is also a prerequisite for implementing actions to reduce this suffering as well.
Another important point is the fact that, within animal production, the total suffering caused is very variable. It may therefore be much more effective in reducing animal suffering to advocate a reduction in the consumption of small animals and monogastrics, which cause much more suffering (see also attempts to quantify this, e.g. Saja, 2012), and to work in parallel with a demanding welfarist and environmental approach, acceptable and understandable by the greatest number of people, than to pursue a general veganization objective that seems to have little concrete effect (cf. for example the 2021 report of France Agrimer), does not fight against the constant transfer of consumption from large animals to small animals that has been observed for several decades, and often abandons the welfarist approach.
From the point of view of this first point, the worst possible situation from the point of view of animal suffering would undoubtedly be a breeding system that has become solely industrial, totally dominated by the breeding of small animals in poor conditions (which is already the case in terms of the number of animals raised and killed, but not yet in terms of the volume produced), with a global consumption of meat that continues to increase (which is very likely to remain a reality in all the regions of the world where the consumption of animal products is currently low, and where the populations aspire to increase it). This is all the more likely because industrial animal products are more economically competitive, and it is not certain that the popularity of animal welfare products will always be sufficient to compensate for their higher price. Substitutes for animal products can undoubtedly play a role in mitigating this risk, as well as in the acceptability of the reduction in consumption in rich countries, but they will have to prove both their viability and their capacity to replace animal foods effectively from a nutritional point of view, and not become an element of ever greater market concentration.
This first objection of the suffering generated by plant production is therefore probably not to be neglected, but it cannot by itself give an argument exonerating livestock farming, even in its best forms.
RETURN TO THE WILD AND POTENTIAL ASSOCIATED SUFFERING
A second objection, raised by some antispeciesists themselves, is that of animal suffering in the wild (see in particular the synthesis of Estiva Reus in the antispeciesist notebooks). Abandoning animal husbandry would mean giving back vast spaces to nature. If among the abandoned farms are farms in which animal welfare is real (and this will probably be largely the case if we talk about the vast spaces occupied by ruminants), then what will replace them, with its predation and parasitism relationships, its periods of famine, drought or intense cold, the enormous mortality of young animals, the mortality by fire (potentially increased by the disappearance of buffer zones linked to livestock farming, see below), etc, The risk of suffering is much greater than in the case of livestock farming, where animals are indeed killed and exploited, but in exchange receive incomparably more comfortable living conditions, safe food and a potentially much less painful death than in the wild (especially since, if it is possible to act to improve livestock farming conditions, any intervention on the wild is much more problematic). Authors such as Fisher & Lamey even think that the life of some wild animals might represent a net suffering and is not worth living, others even propose that the destruction of wild habitats would reduce suffering and would therefore be desirable (Tomasik, 2017), other authors such as Matheny consider in reverse that, even if the life of some farmed animals is « worth living », it decreases by its presence the number of lives worth living, Matheny. I’m not sure how one would decide this point, but the moral issue related to this point is again not black and white.
Conclusion on this point: from a consequentialist point of view, this seems potentially stronger than the previous objection: welfare-conscious farms might cause less suffering and better maximize animal welfare than the wild world that might replace them, even though it is very difficult to assess both the suffering and the welfare experienced over the lifetime of each animal, wild and farmed.
From a deontological perspective, it is possible to argue that all animal exploitation is inherently unacceptable, even if it were to reduce overall suffering, but the two approaches could then conflict.
An approach that takes existing situations into account might be to leave existing wilderness areas untouched (a total ban on expanding the footprint of animal husbandry, both directly and indirectly), but also to retain farms with good (but strengthened) welfarist safeguards, especially in areas of the world that are ecologically at risk (fire, over-hunting (see below for this point), etc.).
HUMAN SUFFERING AND POTENTIAL BACKLASH
A third point concerns the potential human suffering associated with a massive offensive against the raising and eating of animal products. Despite the confidence of vegan activists and some researchers in the safety of an all-plant diet, it is not clear that this confidence is well founded. Many authors, and sometimes among the pioneers of research on vegan and vegetarian diets (e.g., Key & Appleby), point out that there is a lack of very long-term studies, or that existing studies are methodologically insufficient to assess the long-term risks of exclusively or almost exclusively plant-based diets (see [enough studies to conclude?]), particularly for certain specific populations: children, the elderly, people with chronic diseases…) Drastically reducing animal production, or seriously increasing the prices of animal products, could harm at least these fractions of the population, as well as the most financially and culturally fragile. Financially constrained plant-based diets are probably more likely to be poorly planned, and thus carry even greater risk, than those same diets freely chosen by people able and willing to assume the difficulties.
From a producer perspective, many regions of the world, including developed countries, rely economically on livestock production, which structures landscapes and communities, and potentially, in fire-prone areas, provides important assistance in preventing mega-fires (which themselves cause significant animal suffering). Brutal actions against these types of livestock production would likely lead to increased suffering for the communities that depend on them, and in the poorest regions of the world to an increased risk of food insecurity and malnutrition (See [ Livestock, Food Security, Malnutrition] page). The lack of farmed meat can then push communities, at least in some regions of the world, to rely more massively on wild meat, further increasing the pressure on wild species sometimes already threatened by abusive hunting (see for example [Ripple et al. 2016]). Yet it is precisely these regions that are now seeing their livestock footprints increase rapidly, and indeed, action should be taken quickly. A catastrophic situation would be that the most virtuous livestock farms in the most virtuous regions would disappear (see the Nordic example in [Herzon et al. 2022]), while regions where livestock farming is not very efficient would see the footprint of this farming explode.
But any change can only be made on the basis of the needs of these communities, by attempting to resolve the admittedly difficult equation between the need for food security and sovereignty and the risks associated with an expansion of livestock farming. In these regions, a path that seeks to improve productivity and reduce the impact of livestock farming methods, and the expansion of its consumption of land, would probably be more effective than a path that imposes an authoritarian reduction of livestock farming from the world authorities, which would then be tantamount to green colonialism, with dubious results (see, for example, [this Twitter account] for the problems linked to reforestation actions, or NGOs such as Survival International for the risks of expropriation of indigenous communities, notably). The issue of fire risk is also fundamental in a context of climate change, and livestock farming, when included in well thought-out fire prevention plans, is a very effective tool for reducing biomass and/or breaking up inflammable areas. Another point is that putting farmers in difficulty is not only a source of suffering for them and their families, but also potentially for the animals they raise. Many of the farms denounced by activists in recent years for the undignified conditions in which the animals were kept seemed to be mainly farms where the necessary investments had not been made, due to the financial (or other) difficulties of the farmers.
Conclusion on this point: if we place ourselves in an anti-speciesist perspective, it is appropriate to include human suffering in the global animal suffering, and to take into account the suffering that could be caused by a drastic reduction of animal husbandry, or strong actions to limit meat consumption. It seems obvious that humans must take their part in the necessary work of reducing the damage caused by livestock farming, and accept a share of effort and sobriety. But this effort must not be made at the cost of the suffering of fractions of the world’s population who are often the most fragile.
We should also be concerned about the consequences of actions carried out without discernment and without adaptation to the specificities of the different eco-socio-systems, which could turn against both humans and non-human animals.
GENERAL CONCLUSION
The ethical question of animal husbandry must take into account all the issues related to animal suffering, including human suffering, and ask itself how to reduce this suffering in the most effective and secure way, i.e. by including three criteria: the theoretical impact, the chances of acceptance and success, and the associated risks. In many situations, it seems that a welfarist approach is more realistic on these three criteria than abolitionist objectives whose theoretical impact on global suffering is far from being perfectly sure in all situations, but which moreover would come up against strong resistance, and would lead to risks of destabilization, a destabilization itself potentially generating badly foreseeable cascading consequences. In any case, it does not seem to me that there is a clear ethical situation in which the duo abolitionism/veganism would be obviously superior to the duo welfarism/sobriety. Let us note that the second option does not prohibit the abolition of the most problematic and least reformable forms of breeding.
Anthropogoniques 