GBS Switzerland

Do Vegetarians Cause Greater Bloodshed? – A Reply

on December 11. 2014’s Felix Olschewski has argued that vegetarians “(may) cause greater bloodshed than meat eaters”. This was spread with headlines such as “Vegetarians are also murderers”. The topic is trending, as evidenced by the English article Ordering The Vegetarian Meal? There’s More Animal Blood On Your Hands, which Olschewski cited and which went sort-of-viral.

The following reply is based on a reply in German to Olschewski, but it also addresses the core claims of the English article by Mike Archer.

The argument in both articles is that there exists a certain form of animal consumption (meat of grass-fed cattle) that harms less animals than a vegetarian diet. Vegetarianism loses out in this comparison, Olschewski’s claim goes, due to the fact that the cultivation of crops harms a greater number of animals than the rearing and slaughtering of pasture-raised cattle. This clarification alone leads to the first problem with Olschewski’s article. (The problems are described below and are consecutively numbered, with summaries – which may also be read separately – in bold.)

(1) Olschewski recognizes that vegetarians are empirically 90% right and ethically 100% right. The title is therefore profoundly misleading. A fundamentally pro-vegetarian framing would have been more appropriate. The anti-vegetarian framing is damaging to both vegetarians, and on balance, probably to Olschewski himself as well, since the Olschweski/vegetarian goals overlap almost to 100% relative to the status quo.

When it comes to the ethical assumptions, Olschewski completely agrees with the vegetarians. He writes: “The renowned ethicist Peter Singer says that, whenever we have several possibilities to feed ourselves, we should choose the method that causes the least unnecessary harm to animals. I agree with him. When it comes to nutrition – or actually in general – we should always behave in the manner that causes the least amount of suffering.”

The ethical consensus (i.e the overlap of goals) here would have suggested a clear pro-vegetarian framing. Then, a next question would be what this implies in practical terms, i.e. how this goal should be implemented, given all the real-world considerations. At this stage, despite an alignment of goals, practical differences can arise when two parties rely on different empirical information. And indeed: Olschweski’s view is that vegetarian diets are suboptimal because there are other diets that harm even fewer animals. However, Olschewski is also very clear about the advantages of vegetarianism over conventional forms of meat-production, he writes: “It is hardly debatable that industrial meat production harms the environment and makes inefficient use of resources and is thus not sustainable. It is also obvious that it causes much suffering. Particularly when (mostly related to the inherent pricing pressure) the standard of care taken whilst slaughtering is reduced. Addendum 23.1.2014: Look out, it is fascinating how many readers ignore this paragraph. To stress the point once again: Industrial meat production, colloquially also known as factory farming, is unsustainable and environmentally harmful.”

Well over 90% of the animal products consumed nowadays are industrially produced. (Free range animals, too, are being fattened with crops). In agreement with the vegetarians, Olschewski shares the opinion that this production is unsustainable and should be discontinued. However, Olschewski believes that the best solution does not involve giving up meat completely, because he recommends eating a certain amount of meat from large animals that lived on pasture. If this diet was to be provided to the entire world population, it would mean that meat would be served once a week at most, if not less (there isn’t more grassland). Therefore, according to Olschewski’s own views, vegetarians are not only 100% in right ethically, they have also achieved 90% of the best practical implementation of these ethics.

This in turn means that Olschewski should welcome and actively promote an increase in vegetarianism in society, because in a vegetarian world, he would have achieved 90% of his goal. Instead, the article he wrote is closer to anti-vegetarian propaganda. Hordes of meat eaters now use this article, which is in line 100% with the ethics of vegetarians and with 90% of their practices, as a trump card in discussions. This shouldn’t be surprising: Humans are quite irrational, as is demonstrated by cognitive psychology. “Motivated cognition”, specifically Wishful-Thinking-Bias, is responsible for substantial clouding of thought: We search for points of data and “arguments” that confirm our prejudices and desired conclusions (Confirmation Bias). The same appears to be happening in this case: many meat eaters are incapable of dealing with the issue of animal consumption objectively. They search for “arguments” to convince themselves that their behaviour is not a problem. Olschewski’s framing delivers such an “argument” – which is then used by meat eaters for rationalization and reassurance.

Olschewski could respond that it is not his responsibility when people misinterpret his text. But this reply is unconvincing. The ethical goals acknowledged in Olschewski’s article are, as a result of the anti-vegetarian framing, now more difficult to reach. The framing is therefore irrational. Everything we can affect and that concerns our goals is in our responsibility. Whether it is more or less difficult for today’s meat eaters to convince themselves that their consumer behaviour is unproblematic (or even causes less “bloodshed”) is highly relevant to the goal in question. Olschewski could (easily) have influenced this with his choice of framing.

(2) How does this affect Olschewski’s credibility regarding the ethical goal of minimizing animal suffering? Why is only vegetarianism mentioned? Milk, and particularly eggs, also cause animals great harm. If there is an argument in favor of eating meat, then it should preferably be for pasture-raised meat from large animals (cattle). Products from smaller animals cause many more victims per calorie. Poultry leads the casualty statistics, followed by eggs. Olschewski’s blog promotes eggs and lamb.

If Olschewski, rather than paying mere lip service, really agreed with the ethics he describes, then he should have addressed veganism (unless there are gaps in his information?). Milk causes animal suffering for a variety of reasons, including for the calves, whose mothers’ milk is used for human consumption and who are are separated systematically from their mothers. The consumption of eggs is quantitatively worse: Laying hens suffer greatly before being killed, and their brothers, all the male chicks, are painfully gassed straight after hatching. This affects over 250 million male chicks per year in the US alone. Whether or not a product is free range makes no difference to this.

The argument in favor of eating  animal products at best applies when we focus only on pasture-raised meat from large animals. (The larger the animal, the lower the animal suffering caused per calorie. Chicken consumption harms more animals than beef consumption. The same would be the case if we consider whales versus cattle – and it is far from obvious that whale meat is not considerably better than beef in ethical terms. However, this is later implied by Olschewski without basis.) Given this, it is incomprehensible that promotes the consumption of eggs and lamb. This undermines Olschewski’s credibility, at least partially, and makes it more likely that his argument in favor of pasture-raised meat is not based on an openly objective and rational process, but rather that it is also partially a result of “motivated cognition”. Specifically: wishful-thinking bias. This in turn makes the conclusion of the argument more likely to be incorrect.

(3) Does pasture-raised meat really harm fewer animals? Olschewski’s figures are suspect. Calculations suggest that the pasture-raised meat theory is based on the flawed assumption that crop products yield the same quantity of food per land surface as pasture-raised meat. The truth is that the former produces significantly more. carries out calculations on animal deaths that refute those cited by Olschewski: “In a 2003 article in the Journal of Agricultural and Environmental Ethics, Steven Davis advanced the argument that fewer animals would be harmed if we consumed a diet containing large herbivores (namely cattle) fed on pasture than if we consumed a vegan diet, based on his calculation that more wild animals would be killed in crop harvesting than in producing food from a ruminant-pasture-forage system [1]. Gaverick Matheny identified a crucial error in Davis’ calculation: it assumed that equal amounts of land will produce equal amounts of food from crops or from animals on pasture [2]. In fact, an amount of land will produce much more food when used to grow crops for direct human consumption than when used to raise cattle, provided it is suitable for growing crops. Once Matheny corrected the calculation, Davis’s argument made the case for, rather than against, a vegan diet, given an objective to cause the least amount of animal death.”

This divergence seems to stem from the fact that Olschewski’s figures are based on Australian circumstances, and in particular on regularly occurring mouse plagues, which are countered through poisoning and which massively increased the number of deaths in relation to crop production. This analysis is disputed and refuted here. The details of this discussion are not of great importance, since the argument barely holds out in light of point (6) (see below). It is however enough to show that Olschewski’s figures are questionable. Subsequent points (4) and (5) are, in light of (6), perhaps unnecessary as well. They have been included for the sake of completeness.

(4) Olschewski mentions the disadvantages of crop cultivation, but ignores possible negative consequences of pasture-raised meat production. continues: “Davis’s argument was also criticized by Andy Lamey, who pointed out that (…) the argument overlooks ways that humans can be harmed or killed by beef production but not vegetable production [3].”

The view of eco-political author George Monbiot is also noteworthy: “The argument seems, once more, decisively in favor of veganism. (…) While researching my book Feral, I also came to see extensive livestock rearing as a lot less benign than I – or Simon Fairlie – had assumed. The damage done to biodiversity, to water catchments and carbon stores by sheep and cattle grazing in places unsuitable for arable farming (which means, by and large, the hills) is out of all proportion to the amount of meat produced. Wasteful and destructive as feeding grain to livestock is, ranching appears to be even worse. The belief that there is no conflict between this farming and arable production also seems to be unfounded: by preventing the growth of trees and other deep vegetation in the hills and by compacting the soil, grazing animals cause a cycle of flash floods and drought, sporadically drowning good land downstream and reducing the supply of irrigation water. So can I follow Al Gore, and do it better than I did before? Well I intend at least to keep cutting my consumption of animal products, and to see how far I can go. It’s not easy, especially for a person as greedy and impetuous as I am, but there has to be a way.”

(5) Olschewski compares today’s average crop production to the best form of animal farming; which is misleading. The best form of animal production would be better compared to the best form of crop production. His reasoning is therefore incomplete and the conclusion invalid.

Olschewski writes: “Since the biggest criticism concerns conventional meat production , let’s take a look at the widespread, conventional model of vegetable foodstuff: How is soy cultivated, the most popular alternative to meat? In monocultures. Incidentally, this does not only affect soy, but practically all conventional farming systems: Single-crop farming is used due to other options not being commercially viable. Vegetables are also cultivated in single-crop farms. Even the implementation of crop rotation does not change the fact that this requires, at least temporarily, single-crop farming. Existing ecosystems are cleared to make space, grassland is ploughed up and turned into arable land. However, grassland is a natural ecosystem and home to a great variety of animal and plant species, who lose their habitat as a result of this type of farming.”

Olschewski suggests that he puts forward an adequate comparison, one between conventional meat production and vegetable farming. It is beyond question that conventional vegetable farming performs much better. Olschewski argues that the best form of meat production has an edge over conventional vegetable farming. If instead of assessing ordinary meat production, we focus on the least morally abhorrent kind, then we should probably do the same regarding vegetable farming, otherwise the comparison is misleading. As a result, Olschewski’s argument is incomplete and the conclusion incorrect. He has not proven that a diet based on pasture-raised meat beats a vegan diet in it’s ability to prevent suffering.

Notably, Olschewski also fails to mention that today’s conventional vegetable farming is set up the way it is precisely because of animal production. 85% of global soy is fed to animals. Consequently, vegetable farming would look very different were we not to consume animal meat. This is how organic vegan farms operate, which – in accordance with Olschewski’s interests – carry out small-scale production. If he had reasoned correctly, he would have compared pasture-raised meat to this model.

6) Olschewski grounds his thoughts on a highly doubtful premise – the refutation of which could turn the conclusion upside down. Generally, the premise goes “nature, specifically natural = good” and more specifically “Wild animals benefit if their natural habitats and population size are left untouched.” However, more than nine out of ten wild animals have a life which roughly resembles the following description: Birth; battle over far too few resources against far too many siblings; painful death (shortly after birth). This is the destiny a Darwinian nature provides its creatures. One would not wish this upon anyone. (Just how one would not wish life in a factory farm on anyone, which is why it would be better if they ceased to operate.) The following presentation about “Reducing Wild Animal Suffering” substantiates this assertion:

The argument that animals are benefitted by leaving Darwinian ecosystems untouched is very questionable. It could even be the case that wild animals would benefit from natural ecosystems being pushed back. According to this, the “disadvantages” Olschewski assigns to veganism, as opposed to pastured meat, could actually be advantages. And industrial meat production would accidentally be the most advantageous option in this regard.  – We can draw from this that everything in this area is much, much more complex than has been understood so far. The central issue now can only be one of furthering an ethically correct attitude towards animals (hence: veganism), and to advance relevant research to get a clearer picture on which diets (and in general: which conduct) most successfully minimizes suffering in the long run. Only this way do we have a chance of drawing the right empirical conclusions from our ethical goals, adapted to the complexities of the world.

(6.1) Natural bias seems to be substantially inspired, and (mis)directed, by the naturalness bias. Their site reads: “Eat naturally – live healthy”. However, the idea of “natural => good” and “unnatural => bad” is a fallacy. Examples of things that are natural and bad, as well as of things that are unnatural but good, are easy to find. Medicine or the entire welfare state for example go completely against Darwinian nature – and that’s a great thing! There is also scant reason to assume that when it comes to nutrition, natural and unaltered products are always optimal. (Who likes wild-type bananas?)

We can improve the evolutionary status quo by scientifically researching which nutrients provide us with what we need to be healthy, and from this develop the relevant products and “supplements”. (It is highly improbable that this evolutionary status quo has reached the optimum position relative to our goals.) The opposition towards “supplements” seems completely irrational. Why are only some things considered “supplements”, for instance? Why should a tasty sweet containing vitamin B12 be classified as a “supplement”? It is a source of nutrition just like any other and not any more “artificial” than others. Even if these products were “more artifical”, there would still be no reason to prefer “natural” over well-produced “artificial” products.

(6.2) Rural idyll

Nature bias makes us also believe that wild animals live great lives in their natural surroundings. Unfortunately, as explained by animal ethicist Oscar Horta in his article “Debunking the idyllic view of natural processes: Population dynamics and suffering in the wild”, the opposite seems to be true. The main reason for this is that most species’ reproductive strategies are based on creating the greatest possible number of descendants. (The other reproductive strategy, also successful in evolutionary terms but unfortunately a lot less common, consists of having a small, well looked after number of children, focusing on quality rather than quantity.) For example, turtles lay up to 10 nests every gestation period, consisting of up to 100 eggs each! This way, many thousands of hatchlings are born from a typical mother. Should the population remain stable – which must become the case at some point, due to resources being limited, only one hatchling per parent will survive and reach adulthood. In other words: 99.9% of all turtles die shortly after birth, after fighting with far too many siblings over far too few resources. Death in nature is often brutal and unimaginably painful (see video above, 6:15).

What we intuitively imagine as a natural idyll is, highly unfortunately, better described as “natural hell”. This raises the question to what extent it is ethically good to preserve Darwinian ecosystems. One of the impacts of egg consumption is a continuation of the farmed population of laying hens and chicks, however, this is not a positive thing in light of what befalls these chicks and hens. Therefore, how can it be positive to maintain something which harms countless turtles, not giving (most of) them the opportunity to grow old and enjoy the good sides of life? Life and death in nature is in many cases similarly bad, sometimes worse even, than life and death in factory farms. If we were living in these conditions, we would wish for these painful circumstances to not be upheld in both cases.

A “perverse consequence” of this could be that conventional meat production, which pushes back natural ecosystems the most, could provide the most positive outcome for today’s wild animal population. (Unfortunately, nothing guarantees that reality proves itself to be anything but perverse.) The extent to which this consideration is relevant for our dietary choices today is discussed in (6.4).

(6.3) “The most ecological way of being is not existing”: No

Olschewski writes: “The solution is relatively simple. It consists of accepting the fact that when it comes to eating, someone will always be disadvantaged. Whenever I pluck a blueberry, a bird can no longer eat it. If I dig out a lettuce or even just a wild herb, I am depriving a rabbit of a meal. If I build an acre, an ecosystem and a habitat for other animals is destroyed. Even the house where I live occupies land that could have been another animal’s habitat. That’s life, the circle of blossoming and decay, eating and being eaten, life and death. That’s reality.”

Firstly, this is not a solution, but a mere declaration of the problem. The solution would comprise of determining, in a scientific manner, which nutritional habits and behavior in general would lead to the smallest number of victims. Secondly, Olschewski’s framing suggests that the ethical issue consists in intervening as little as possible. In this case, there would be a clear solution: We should minimize our influence in as many areas as possible. And the optimal solution for this is clear, too: We should take ourselves out as quickly as possible. – The error in this reasoning stems from believing that not becoming actively involved is the only thing of importance to those who are affected by our decisions. This is not the case. What is important to the individuals affected by our decisions is their well-being, that their life in this world goes well.

For our decision-making, this means we should make decisions in a manner that is positive to as many (ideally all) sentient beings as possible, or at least take those actions that result in fewer beings being negatively affected. The solution to the profound suffering of sentient beings out there – in both factory farms and natural ecosystems – does not consist of doing as little as possible and staying out of the situation. This would not pay enough respect to the interests of others; it would violate our ethical goals. Instead, one must stay active in such a situation, trying to have a net-positive impact. This requires (compassionate!) intervention.

There is no reason to uphold the cycle of despair of eating and being eaten (and to even reinforce this), only because it corresponds to the evolutionary status quo that seems pretty much inevitable within natural ecosystems. Even if we – with technology from the year 2,500, 3,000 or 10,000 – could never undertake anything to change this: the fact remains that being eaten is a horrific tragedy for the victim, which poses a huge ethical problem.

Olschewski goes on: “Do numbers matter? Are suffering and death cumulative? If it really was relevant that for humanity’s nourishment, as “few” animals as possible should be killed, would we slaughter a blue whale and freeze its meat every time a new baby is born? This would certainly last a lifetime, and as such only one animal would have to die to feed a human. Self-evidently, this cannot be a solution.” That this would be worse than the consumption of a far greater number of cattle is not self-explanatory. And yes, numbers do play an indispensable role. If we ourselves were affected by our decisions, we would also wish for all individuals, or (should this be impossible) as many individuals as possible, to be saved. Because then, and only then, is the probability of being saved at its highest.

If we, before entering the world outside, come up with a decision-rule that saves as many beings as possible, then every individual gets the best outcome: the highest probability of being spared from suffering and death, and consequently the highest probability of being able to live a good life. – An evident application of this principle is that it is obviously better for “livestock” if we consume less animal products instead of more, and that it is better to avoid animal products that cause suffering to a lot of victims, instead of avoiding products that harm only few victims (casualty statistics: chicken meat > eggs > pork > beef > cheese > milk).

Olschewski concludes: “It is thus impossible to answer the eponymous question. Do we measure the litres of blood or the number of injuries? Do we only count direct fatalities, or also those caused indirectly? How do we measure suffering? It doesn’t matter.”

Of course it matters! Yes: The eponymous question is underdetermined. It has to be determined what the ultimate (dis)value is that matters ethically. (Obviously it is important to know what the precise goal is that we want to pursue!) We aren’t feeling around in complete darkness: The answer has something to do with the preferences and/or experiences felt by sentient beings. The fact that suffering is difficult to measure is not an argument against it mattering ethically. It is important to avoid the Evaluability Bias, which can leads us to dispose of our actual goals solely because they are difficult to measure. This would be as irrational as the decision of a drunk to look for his car keys beneath a street light – because that is where he can see – instead of in the place where he remembers losing them. Suffering certainly seems to be what matters ethically, or at least an important part of it, because to not suffer would be of great importance to me, should I be affected. The rest seems secondary only.

(6.4) What to do?

The situation is much more complicated than previously thought: Vegan nutrition leads to less suffering of “livestock”. It could, however, lead to more suffering caused to wild animals as compared to meat-based diets, which more strongly contain natural ecosystems. It may be that the latter dominates due to wild animals being extremely numerous. However, the causal chain does not end here: The ever greater retreat of nature caused by meat-based diets contributes to more climate change. Climate change could have further effects, such as an increase in the probability of global instability, which would have devastating long-term effects. The long-term consequences could also be negative for wild animals: A warmer planet might contain greater amounts of biomass and therefore larger populations of wild animals (whereby a miserable fate would await almost 100% of the individuals involved – as described above). Speculative effects stemming from climate change that would be positive for wild animals are certainly conceivable as well. In short: The current uncertainties are enormous.

In situations like these, two practical actions are of clear value: firstly, the large-scale promotion of relevant research to obtain a clear overview on these issues, and secondly, the promotion of the correct ethical attitude to ensure a higher probability of this research being correctly applied and produced relatively quickly, if at all. This attitude is expressed by veganism, which recognizes the ethical problem of what befalls the individuals in “livestock”. If this is not recognized, then we will never reach the goal of a society that takes the suffering of wild animals adequately into account. It is therefore of great importance to reduce carnism and promote plant-based diets.

How can wild animals be helped systematically? One example: herds of wild animals are already vaccinated – yet not because of altruism, but out of human self-interest. We want to prevent being infected by them. Some wild animals are also fed in the winter – yet the goal is to kill and eat them in the summer. Nevertheless, these measures could be used to help wild animals. They could be combined with appropriate birth control, such as immunocontraception, which has already been tested as a non-violent alternative to culling. On the more speculative side, something like a “welfare state for elephants” could be considered.

However, for the time being, those species that bring thousands of descendants into the world, whose lives will more than likely be miserable on average, can only be helped by the curtailing of natural ecosystems. This would have to be implemented in a manner that doesn’t jeopardize global social stability and cooperation, something which seems to be of great long-term significance for all value systems and actors. It is not clear if this could at all be implemented at this stage.

What we can do today in order to have the largest possible impact on the course of the world is to promote an ethical (and cooperative/pragmatic) attitude towards ending suffering, and at the same time, to advance the research that will be needed.

Consciousness research will play an important part in this. Olschewski writes: “Veganism only concerns itself with sentient beings. An often comfortable yet arbitrary line is drawn, which leads to grasshoppers being usable as a food source – this of course also includes lactobacilli from fermented food such as Sauerkraut. What is the source of this certainty? Do spiders and insects really have no feelings or consciousness? The now outdated view that presented animals as mere automata and machines, incapable of feeling anything, was common just a few centuries ago. This has changed. Are we to rule out that spiders and insects possess the capacity to feel?”

It seems totally appropriate to only focus on sentient beings. Something which cannot feel emotion, and never had the capacity to do so – such as a stone – cannot suffer or possess wishes that can be frustrated. Things which cannot and could never possess feelings cannot be harmed, since no circumstance can be good or bad for them. This is the case for stones, for “me” as an embryo, and also, it likely is the case for plants: plants do not possess the required behavioural repertoire characteristic of conditions of sentience, nor do they possess a central nervous system (or analogous structure), a seeming necessity for consciousness. (Nevertheless: Here is a discussion on the possible relevance of the extremely unlikely suffering of plants, and here a TED-contribution on “plant intelligence”.)

Our degree of certainty should be much lower regarding insects. Olschewski suggests that vegans dogmatically reject the notion of insect sentience. This is not the case: Some vegans’ abstention from honey underlines the opinion that bees are ethically significant – or at least that we cannot be certain that they are not sentient and therefore insignificant. Presently, not enough is known about sentience in general, let alone in relation to insects, to make assertions with a high degree of probability. Here is a collection of scientific publications and thoughts on the topic of “insect sentience”. The differences between distinct insect species are considerable, and it seems that at least in some, the probability of possessing consciousness might not be negligible. Due to their astronomical numbers, they could be highly ethically relevant even if we only grant them a minimal probability of consciousness (maximization of expected value). For example, a “more humane insecticide” that kill insects within a short time frame instead of leaving them to slowly (and painfully) waste away could be developed through applied research and could reduce an enormous amount of suffering in expectation.

In conclusion, how can it ever be possible to seriously tackle these questions in an openly rational and scientifically competent manner, if society doesn’t even see the great damage done to sentient beings in the name of culinary enjoyment? Certainly not with eggs, lamb and pasture-raised beef. Veganism shows that what befalls “livestock” is not acceptable, and that we should pursue the ideal of guaranteeing a life free of suffering to all beings brought into this world. The exact details of the effects (on wild animals, climate change, etc.) different diets have is not the most important factor. The rational approach at the current time consists of

(a) the recognition of the existence of huge factual uncertainties (Overconfidence bias is one of the prime sources of error in human cognition),

(b) trying to reduce this relatively quickly through relevant research,

(c) promoting and cultivating an ethical stance that seeks to prevent suffering, thereby increasing the probability that a responsible application of research will take place quickly/at all. The suffering of “livestock” is the first problem in the long causal chain, it “simple” enough to understand (as compared to the issues involving wildlife), and it can be stopped easily enough by abolishing animal exploitation. Thereby, we promote an ethical stance, respect towards the interests of other animals, that will be essential when it comes to solving all the other more complex problems.