If You Eat, You Harm Animals

No matter what we eat, we still kill animals. Even in plant production, field mice are crushed by tractors. Birds’ nests fall victim to combines. Agricultural runoff poisons fish. And pesticides destroy insects in untold numbers. This paper aims to collect what we know about animal deaths related to plant production and to show the challenges inherent in determining a plausible estimate of those deaths. Depending on the result, could a plant-based diet be worse for animals than a diet containing animal products?

Without knowing how widespread and pervasive animal harm is, we don’t know what it will take to fix it. Existing empirical data is weak. The two most applicable studies, one from 2003 and the other from 2011, both have methodological issues. The first averages mouse deaths from farm machinery during grain harvesting in 1993 and rat deaths during the sugar cane harvest in Hawaii in 1971. The second looks at mouse deaths in Australia from poisons applied to grain fields to curb mouse plagues. Two other data sources look at avian deaths from pesticides in Canada and fish kills from agricultural water runoff in the U.S. but neither provides reliable answers. No data is available for reptiles and amphibians or for so-called secondary deaths where an animal eats another animal that has been poisoned. 

Extrapolating what data we do have, we come up with a total of over 7.3 billion animal deaths each year. However, there are several reasons to be skeptical of this number. First, it’s not reasonable to apply a “one size fits all” model to death rates of animals across crops and regions. Different crops attract different animals, harvest technology varies widely, and animals themselves may migrate in and out of fields, so population changes may have nothing to do with mortality from farm equipment. Studies in Argentina and South Dakota in the U.S. demonstrate these points. Data about kills from pesticides and insecticides do little to clarify the issue. Many pesticides, because of their documented effects on birds and fish, are now banned, so historical death rates, if known, are irrelevant.

As it turns out, many of the deaths from plant agriculture are not caused directly by machinery or poisons. Instead, human activities expose animals to predators who do the actual killing. As an example, the harvest of a field removes sheltering cover for mice, exposing them to raptors such as owls, hawks, and eagles. Given this, are humans responsible for these deaths? Or are they morally legitimized by being an unintended side effect? To further muddy the issue, there’s what the authors refer to as the “line-drawing problem”. Are all lives of equal value? Should we concern ourselves only with mammals, or do bird and fish lives count as well? And how do we factor in recent evidence on the sentience of insects? If they count in the death toll, the figures would be magnitudes higher. 

Finally, there is the issue of wild animal suffering. While we humans may think of wild animals as living in a natural paradise, field studies show a different reality. The scarcity of food and water, predation, disease, and intra-species aggression all cause wild animals to suffer. For example, many rodents produce large litters but offer little care to them. Mortality and suffering are high. Philosophers term these “net negative” lives. In this tenuous position, things are so bad that it would be better for them not to exist. 

As this discussion makes clear, philosophical assumptions matter a great deal as we try to determine how many deaths that plant production is responsible for. If we assume moral responsibility for predation, if we discard the net-negative argument, or if we include insects in the count of animals harmed, our kill rate skyrockets. This in turn makes arguments for veganism less supportable. But assumptions to the contrary for each of these issues brings the numbers down dramatically.

For animal advocates, this study raises more questions than it answers. About all we can say is that we want to reduce the animal harms created by plant production. While a vegan diet seems to be the best for animal welfare, even eating plants costs animal lives. As such, this information should catalyze efforts to develop less harmful farming practices such as no-till agriculture, non-lethal hazing practices for avian pests, rodent contraceptives, indoor farms, and improved sensing methods in agricultural machinery. And it might cause us to think more deeply about the value of all animal lives. While many advocates feel a strong moral high-ground by being vegan, it’s important to recognize that terms like “cruelty-free” are relative, and a perspective of “harm reduction” may be more coherent when engaging with some portions of the public.