The Case for Reducing Wild-Animal Suffering^[1]

This is the draft of a piece written by two of my friends. It was never finalized. I’ve uploaded it in case its contents are useful for others even though the writing isn’t fully polished. --Brian Tomasik, 15 Feb. 2015

1. Introduction

While animal activists and moral philosophers have often – and with good reason – voiced concern about the treatment of factory-farmed animals, the welfare of wild animals has been unjustly neglected.^[2]

In this paper, we propose that there are strong reasons to take the welfare of wild animals seriously. Our argument is very simple:

1. Most of the world’s suffering is experienced by animals in the wild.
2. We have a duty to alleviate suffering (absent defeaters).

Therefore,

3. We have a duty to alleviate the suffering of animals in the wild (absent defeaters).

Section 2 cites ample empirical evidence in support of the first premise. Section 3 surveys what various moral philosophers have say about the normative significance of suffering, and concludes that there is wide agreement that, absent defeaters, we have a duty to alleviate suffering.

We realize that our conclusion will strike many as controversial or even absurd, and that critics can cite various potential defeaters. Thus the third section of our paper considers several common objections, and argues that none of them is sufficiently strong to constitute an actual defeater.

In the fourth and concluding section, we tentatively discuss what course of action should be taken if our argument is sound. We propose that, given our current state of relative ignorance and technological limitation, the most promising route is to raise awareness of the seriousness of wild animal suffering and to promote research on ways to reduce the suffering of wild animals.

2. Most of the world’s suffering is experienced by animals in the wild

To answer the question of how much suffering is experienced by wild animals, we must answer three further questions:

1. How many wild animals are there?
2. Which animals are capable of experiencing suffering?
3. How much does an average wild animal suffer?

This section answers each of these questions in turn.

How many wild animals are there?^[3]

As far as we have been able to determine, there aren’t any readily available estimates of the world’s total population of wild animals. There are, however, some attempts to estimate the densities of various kinds of animals in different types of biomes. If we combine these densities with data about the total area of these biomes, we can arrive at estimates of the total number of wild animals.

The estimates for biome areas that we have used are taken from table 2 of Matthews (1983). Gaston et al (2003) give the following estimates for wild breeding bird densities: 300 individuals per km2 on cropland, 375 on pasture, 450 on grassland, 800 on temperate mixed forest, and 1250 on warm mixed forest. We assume that woodland and shrubland have the same bird densities as grasslands.  We also assume that the tropical rainforest density is 1250 and other forest density is 800.  As no figures are given for deserts and tundras, we conservatively assume them to have zero density.

Table 1. Estimated biome areas (Matthews 1983) and estimated bird density (Gaston et al 2003).

Based on studies by Gaston and Evans (2004) and Harris et al. (1995), Matheny and Chan (2005) estimate that wild mammal densities are 2.25 times higher than those of wild birds.

For land reptiles, Ishwar, Chellam, and Kumar (2001) examined 25 m2 quadrats in a tropical rainforest, and found that there were an average of 0.2559 reptiles per quadrat, equivalent to 10,236 reptiles per km2. This is roughly eight times the tropical rainforest density of birds. In calculating the total number of land reptiles, we assume that this ratio holds in other biomes as well.

For land amphibians, Vasudevan, Kumar, and Chellam (2001) found average densities of 0.79 individuals per 25 m2 quadrat in a tropical rainforest, equivalent to 31,600 individuals per km2. Vasudevan et al (2008) give estimates of 14,900 individuals per km2  on the rainforest floor, and in excess of 30,000 individuals per km2 along streams. Huang and Hou (2004) give an estimate at 35,000 to 102,000 individuals per km2 in a monsoon forest in southern Taiwan. If we assume, conservatively, that the right number is 14,900 individuals per km2 in tropical rainforests, that is equivalent to roughly 12 times the bird density in the same biome. As with land reptiles, we assume that this ratio also holds in other biomes.

Putting all of this together, we get:

World population of birds: 6.0645 × 1010

World population of land mammals: 1.36451 × 1011

World population of land reptiles: 4.8516 × 1011

World population of land amphibians: 7.2774 × 1011

Moode and Brooke (2010) estimate that between 0.97-2.74 trillion wild fish are caught every year. Conservatively, for our analysis we assume the lower bound, which we round up to one trillion. That same study estimated that 77,388,322 tons of fish corresponded to 0.97 - 2.74 trillion fish. That gives us roughly 13,000 fish per tone. Another study by Wilson et al (2009) gave the first ever estimate of fish biomass, at 0.8 - 2 billion tonnes. Again, we assume the lower bound, which we round up to one billion. If the same conversion rate holds, these estimates imply that at any point in time there are about 13 trillion fish in the oceans worldwide.

According to the Entomological Society of America, it has been estimated that there are around one quintillion (1018) individual insects in the world at any given time.^[4]

Most zooplanktons are copepods, and Cerullo (2000, p. 14) estimates that the world population of copepods is 1018.

All of this is summarized in the table below. For comparison, we’ve also included figures for the number of human animals, the number of animals in research labs and the number of animals raised for food.

Table 2. The livestock estimate is taken from 2010 FAOSTAT data (appendix A). The estimate of animals in research labs is taken from Orlans (1998).^[5] 

Which animals are capable of experiencing suffering?

Of course, the question of which of these animals are capable of suffering remains. It is very unlikely that zooplankton suffer, whereas it is very likely that non-human mammals suffer.  But for many types of animals, whether they suffer is unclear.  Given that self-reporting is the standard measure of pain in humans, measuring pain in non-human animals presents obvious methodological problems.

Nociception is the ability to detect adverse tissue damage, and it is sometimes followed by reflex reactions. Pain, on the other hand, is the negative subjective experience that typically accompanies tissue damage.  Basic nociception is present even in invertebrates, and possibly fish as well (Sneddon et al 2003). Nociceptive signals are processed in several primitive areas of the brain: the medulla, the thalamus and the limbic system. Vertebrates, but not invertebrates, possess these areas. In humans, the cortex plays an important role in pain perception. However, as we move downward in the evolutionary tree from humans, the cortex becomes smaller in size. (Sneddon, “Can Animals Feel Pain?”).

Of course, we don’t know whether nociception is always accompanied by pain experience. Many authors propose that we examine how closely a species resembles Homo sapiens when determining whether it is capable of suffering. Using this approach, Bateson (1991) suggests that all vertebrates are capable of suffering. This approach is useful as a first approximation. Soon enough, however, we enter a gray area where we cannot say with any confidence whether organisms of a particular species are or aren’t capable of suffering. Thus Bateson (1991) further notes that cephalopods such as octopuses have very complex nervous systems, and are also likely to experience pain. Looking for other signs of sentience, Elwood, Barr and Patterson (2009, p. 128) offer the following list of candidates:

1. a suitable central nervous system and receptors,
2. avoidance learning,
3. protective motor reactions that might include reduced use of the affected area, limping, rubbing, holding or autotomy,
4. physiological changes,
5. trade-offs between stimulus avoidance and other motivational requirements,
6. opioid receptors and evidence of reduced pain experience if treated with local anaesthetics or analgesics, and
7. high cognitive ability and sentience.

[discussion of how various species fare with regard to these criteria, perhaps?]

From an evolutionary perspective, the capacity to feel both pleasure and pain is clearly an adaptive trait. Pleasure and pain act as signals that influence behavior. Bateson (1991) offers several other related evolutionary explanations of pain. [more on Bateson here?] Ng (1995) also approaches the question of animal pain from this angle, and reaches the tentative conclusion that all plastic animals, i.e. those capable of learning, are also capable of suffering.

Suffering encompasses not only physical pain, but also stress and fear. Grandin and Johnson suggest that fear is much worse than physical pain for animals. Grandin and Deesing (2002) note that fear operates in more primitive areas of the brain than pain does. Hence animals that are too primitive to experience pain may nevertheless be able to experience fear, and with it the subjective experience of suffering.

We are no experts on animal suffering, so we decided to ask those who are. We sent out a brief survey to about 30 specialists in pain perception and related fields. We received 12 responses. The results are summarised below.

All (100%) of researchers rated the hypothesis that mammals can suffer as “extremely likely (>99%)”.  A large majority of researchers also …. As we have seen both in the cited literature and in the survey (?), many are willing to extend this to birds and certain invertebrates as well. [... More on insects and zooplankton here.] Roughly the same picture is painted in the Cambridge Declaration on Consciousness (Low et al 2012), which was signed by several prominent scientist during the Francis Crick Memorial Conference on July 7th 2012.^[6]

How much does a wild animal suffer?

As with humans, there are likely to be large variations within a species. Furthermore, there could also be significant differences between species.

Many of us have a very rosy picture of life in the wild. Is this picture accurate? Dawkins (1995, pp. 131-132) writes that

[t]he total amount of suffering per year in the natural world is beyond all decent contemplation. During the minute it takes me to compose this sentence, thousands of animals are being eaten alive; others are running for their lives, whimpering with fear; others are being slowly devoured from within by rasping parasites; thousands of all kinds are dying of starvation, thirst and disease.

As the above quote makes clear, predation isn’t the only source of suffering in the wild.

Ng (1995) argues that several popular evolutionary strategies have as a consequence that for many individuals, net welfare is probably negative. One such strategy maximizes the number of offspring, with the consequence that comparatively few individuals survive to mating age. If we assume that the negative welfare of an unsuccessful individual is on par with the positive welfare of a successful individual, it follows that welfare for the species as a whole is negative, given how many additional unsuccessful individuals there are. (One complication with this argument  is that it isn’t clear how many of the individuals that die early in life are actually conscious and capable of suffering at the time.)

What are the sources of wild animal suffering?

• Being eaten alive
• Fleeing in terror from predators
• Parasites
• Starvation, thirst, disease
• Sexual frustration

Predation: how long does it take for various predators to kill their prey? Tomasik offers some quotes. Not only distress, but also long-term psychological trauma.

Can we find some data on how animals die? What percentage of a given species die by predation/starvation etc?

The point here is not to wallow in misery, but simply to get clear on the various sources of suffering that are present in the wild.

When judging the welfare of wild animals, one possible strategy is to look at guidelines for the ethical treatment of domesticated animals.

I think it would be a good idea to mention some of the studies Brian refers to here, to make the point poignant.

However, we shouldn’t expect that a life in the wild is a life of constant misery. As Ng (1995) points out, pain is metabolically costly, and hence a life of undisrupted pain wouldn’t make much sense from an evolutionary perspective. Of course, such evolutionary reasoning no longer applies after the animal has passed reproductive age.

So how much wild animal suffering is there?

In this section, we have estimated the number of wild animals in the world, xxx, and discussed the numerous causes of wild animal suffering.

Even if mammals are the only animals capable of suffering, that would arguably be enough to establish the main claim of this section, that most of the world’s suffering is experienced by wild animals. According to the estimates of table 2, non-human mammals outnumber humans by roughly one order of magnitude.

3. We have a duty to alleviate suffering (absent defeaters)

4. Objections

It would be hubristic / We shouldn’t “play God”.

1. Distinguish between empirical and normative version of this objection. [When people raise this objection, I’m often confused about whether they refer to something empirical (e.g. given our ignorance, this might have bad consequences) or something normative (e.g. even if we’re sure that the hubristic act doesn’t have bad consequences, it’s nevertheless morally bad).]
2. On the empirical: we agree that it would be foolish to implement some drastic intervention in our current state of ignorance. What we should rather do is to research the matter further, and only intervene when we can do so in a safe way.
3. On the normative one – does this collapse into objection #2?

1. Bad track-record of this objection.  Many interventions that we now regard as unobjectionable were once objected on similar grounds.
2. No clear distinction between what counts as playing God and what doesn’t.
3. No good reasons against “playing God” if there is actually no God.
4. Even if there is a God, it is unclear that he would object.

We have an obligation not to interfere with “nature”.

1. We already do interfere.
2. Even if there is such an obligation, it isn’t sufficiently strong to justify not interfering. [How should we argue for this point?]

5. Again, we are already interfering to some degree.  Why doesn’t the objection apply in this circumscribed case?
6. Moreover, humans have causal impacts on “nature” in all sorts of ways. This means that even if the principle of non-intervention is correct, it probably has little or no applicability in practice. (There is a parallel debate in political theory about what the “libertarian” principle of non-intervention implies in a world where that principle has been violated repeatedly throughout history.)
7. Do we really think this? Reader’s comments to this piece (see esp. the best rated ones) provide anecdotal evidence that, at least in some cases, we don’t.
8. What’s so good about nature anyway?  The process that created the living creatures that now inhabit our world didn’t optimize for anything that could plausibly be regarded as morally valuable.

1. Dawkins’s quote.
2. “All species reproduce in excess, way past the carrying capacity of their niche. In her lifetime a lioness might have 20 cubs; a pigeon, 150 chicks; a mouse, 1,000 kits; a trout, 20,000 fry, a tuna or cod, a million fry or more; an elm tree, several million seeds; and an oyster, perhaps a hundred million spat. If one assumes that the population of each of these species is, from generation to generation, roughly equal, then on the average only one offspring will survive to replace each parent. All the other thousands and millions will die, one way or another.” (Hapgood 1979, 34, cited in Sagoff, 1984, pp. 302-303)

9. Even if there was some value in preserving the species that now exist (cf. Dworkin’s notion of “sacredness”, discussed by McMahan), should this value trump the enormous disvalue that this huge amount of suffering instantiates?

Life in the wild is actually pretty good. 

• Apart from citing the various causes of suffering in the wild, we could also respond by comparing the life of wild animals with the life of “wild” humans––if life in the wild is so good, then why not go back to the Stone Age, or even earlier?
• Try to dispel this notion by citing relevant literature.  We can transition to this objection from reply (b) to the preceding objection.  These evolutionary considerations show that life in the wild is, indeed, “red in tooth and claw”.

It is practically infeasible

1. Interventions needn’t be all or nothing: we can intervene without trying to eliminate suffering entirely. Given that some minor interventions could be implemented already today, this shows that the proposal is in fact practically feasible.
2. Again, we aren’t proposing that we implement some radical intervention now. Rather, we think it would be worthwhile to focus more on researching promising interventions.
3. Given how little effort has been invested, it would be premature to conclude that further research couldn’t reveal much more effective and reliable interventions than the currently available ones.
4. We should also say something about unpredictable large-scale effects on ecosystems...

1. There are perhaps two main versions of this objection: (i) it cannot be done at all, and (ii) it cannot be done without wide-ranging effects on the ecosystem, the results of which are bad or unforeseeable.

Statements of the objection:

• “for practical purposes I am fairly sure, judging from man's past record of attempts to mold nature to his own aims, that we would be more likely to increase the net amount of animal suffering if we interfered with wildlife, than to decrease it. Lions play a role in the ecology of their habitat, and we cannot be sure what the long-term consequences would be if we were to prevent them from killing gazelles. [...] So, in practice, I would definitely say that wildlife should be left alone.” (Singer 1973)
• “if propagation of the fittest genes contributes to the integrity of both predator and prey species, which is good for the predator/prey balance in the ecosystem, which in turn is good for the organisms living in it, and so on, then the very ecological relationships that holistic environmentalists regard as intrinsically valuable will be valued by animal welfarists because they conduce ultimately, albeit indirectly and via complex causal chains, to the well-being of individual animals.” (Everett 2001)

Replies:

10. We are already intervening with wild animals to some degree (“habitat destruction, captive breeding programs for big cats, rewinding, etc”), so at least in these cases the objection fails. As Cowen writes, “In [some] cases we are interfering with nature, whether we like it or not. It is not a question of uncertainty holding us back from policing, but rather how to compare one form of policing to another. Humans change water levels, fertilize particular soils, influence climatic conditions, and do many other things that affect the balance of power in nature. These human activities will not go away anytime soon, but in the meantime we need to evaluate their effects on carnivores and their victims.” (Cowen 2003) [add more examples by Cowen]

• Example 1: “culling” of kangaroos in Australia (Clarke & Ng 2006).
• Example 2: general use of pesticides (Tomasik 2009): “Humans spray 3 billion tons of pesticides per year, and whether or not we think this causes more wild-animal suffering than it may prevent, large-scale insecticide use is, to some extent, a fait accompli of modern society. If, hypothetically, scientists could develop ways to make these chemicals act more quickly or less painfully, enormous numbers of insects and larger organisms could be given slightly less agonizing deaths.” (Tomasik 2012)
• Example 3: “Human changes to the environment -- through agriculture, urbanization, deforestation, pollution, climate change, and so on -- have huge consequences, both negative and positive, for wild animals.” (Tomasik 2012) Also see Matheny & Chan 2005.
• Example 4: “pest” control. Singer makes an interesting suggestion: “It would not, presumably, be beyond modern science to produce a substance which, when eaten by rabbits or coyotes, produced sterility instead of a drawn-out death.” (Singer 1979)
• Example 5: “The government is spending hundreds of millions of dollars to store millions of tons of grain. Why not lay out this food, laced with contraceptives, for wild creatures to feed upon? Farms which so overproduce for human needs might then satisfy the needs of animals.” (Sagoff 1984, p. 303)        

11. It is true that we currently lack the knowledge (and the technology?) to abolish all wild animal suffering.  But the conclusion to draw from this is that we should be actively trying to obtain this missing knowledge.  In general, when we have strong reasons to do X but cannot currently do X due to contingent limitations, we still have strong reasons to overcome these limitations.  Perhaps a good way to accomplish this is to conduct a “pilot case” with a particular species of animals, such as elephants (cf.  Dave’s ‘A welfare state for elephants’).
12. Rebut knee-jerk charges of ecological illiteracy: show that advocates of compassionate stewardship fully grasp the thermodynamics of a food chain / "trophic levels" etc  - and to set out how cross-species fertility control via immunocontraception, behavioural-genetic tweaking, in vitro meat provision, the exponential growth of computer power leading to micro-management of every cubic metre of the planet (etc) can in principle deliver a significant reduction in suffering.
13. List specific ways in which we could intervene, in addition to those listed above (i):

• “One may modestly propose the conversion of national wilderness areas, especially national parks, into farms in order to replace violent wild areas with more humane and managed environments. Starving deer in the woods might be adopted as pets. They might be fed in kennels; animals that once wandered the wilds in misery might get fat in feedlots instead. Birds that now kill earthworms may repair instead to birdhouses stocked with food, including textured soybean protein that looks and smells like worms. And to protect the brutes from cold, their dens could be heated, or shelters provided for the all too many who will otherwise freeze.” (Sagoff 1984, pp. 303-304)

We don’t have any obligation to help wild animals.

1. Should we simply restate material from premise #3?

Statement:

• “Some feel that while humans have duties to treat well the animals that they use or live with, they have no responsibility to those outside their sphere of interaction. I find this unsatisfying; if we really care about animals because we don't want fellow organisms to suffer brutally -- not just because we want to keep our moral house clean -- then it shouldn't matter whether we have a personal connection with wild animals or not.” (Tomasik 2012)

Replies:

14. We may not have any agent-relative duty to help wild animals, because we are not responsible for their suffering. (Compare this with the case of animals in factory farms, for whose suffering we are responsible.) But we still have an agent-neutral duty to alleviate suffering.  Given the huge amount of suffering experienced by animals in the wild, this is a strong duty.

Helping wild animals is not the most cost-effective intervention.

15. Relative to what?  Extinguishing predators would mean a huge reduction in suffering not only for all animals currently alive, but also for all future prey.  It’s hard to think that such an intervention, given the astronomical values involved, wouldn’t be cost effective relative to mainstream interventions (the case might be different if the comparison class includes anti-extinction measures or benevolent meme-spreading efforts; but discussing these would be beyond the pale for an academic paper, I think).

• We need to take into account, however, that prey that don’t die from predators will still die in some other way.  So the reduction in suffering is the difference in the pain involved between these two deaths.

Since humans also prey on other beings (and each other), the argument implies that we should make ourselves extinct.

1. Predation isn’t the only (perhaps not even the major?) source of suffering in nature.
2. In contrast with predators, humans have the option of naturally following a vegetarian or vegan diet.
3. Our argument does not require that all predators be made extinct. They could be made to eat artificial lab-grown meat, which wouldn’t require any suffering to be produced.

16. McMahan lists this objection as one of the four most popular reactions to his piece.

5. Conclusion

We have argued that...

If our conclusion is correct, what course of action should be taken?

References

G. A. Bradshaw, Allan N. Schore, Janine L. Brown, Joyce H. Poole and Cynthia J. Moss

(2005). “Elephant Breakdown,” Nature 433:807

Mary M. Cerullo (2000). Ocean Detectives: Solving the Mysteries of the Sea, Austin:

Steck-Vaughn Company

Richard Dawkins (1995). River Out of Eden: A Darwinian View of Life, New York: Basic

Books

Robert W. Elwood, Stuart Barr and Lynsey Patterson (2009). “Pain and stress in

crustaceans?,” Applied Animal Behaviour Science 118:128–136

Encyclopedia Smithsonian, “Numbers of Insects (Species and Individuals)”, available at

http://www.si.edu/encyclopedia_si/nmnh/buginfo/bugnos.htm

Entomological Society of America (2010). “Frequently Asked Questions,” available at  

http://www.entsoc.org/resources/faq

FAOSTAT, http://faostat.fao.org/

Kevin J. Gaston, Tim M. Blackburn and Kees Klein Goldewijk (2003). “Habitat conversion

and global avian biodiversity loss,” Proc. R. Soc. Lond. B 270:1293–1300

Kevin J. Gaston and Karl L. Evans (2004). “Birds and People in Europe,” Proc. R. Soc. Lond.

B 271:1649-1655

Temple Grandin and Mark Deesing (2002). “Distress in Animals: Is it Fear, Pain or Physical

Stress?,” available at http://www.grandin.com/welfare/fear.pain.stress.html

Stephen Harris, Pat Morris, Stephanie Wray and Derek Yalden (1995). “A review of British

mammals: population estimates and conservation status of British mammals other

than cetaceans”, available at http://jncc.defra.gov.uk/page-2759

Ching-Yu Huang and Ping-Chun Lucy Hou (2004). “Density and Diversity of Litter

Amphibians in a Monsoon Forest of Southern Taiwan,” Zoological Studies 

43(4):795-802

N. M. Ishwar, Ravi Chellam and Ajith Kumar (2001). “Distribution of forest floor reptiles in

the rainforest of Kalakad–Mundanthurai Tiger Reserve, South India,” Current Science 

80(3):413-418

Gaverick Matheny and Kai M. A. Chan (2005). “Human Diets and Animal Welfare: The Illogic

of the Larder,” Journal of Agricultural and Environmental Ethics 18:579–594

Elaine Matthews (1983). “Global Vegetation and Land Use: New High Resolution Data Bases

for Climate Studies,” Journal of Climate and Applied Meteorology 22:474-487

A Mood and P Brooke (2010). “Estimating the Number of Fish Caught in Global Fishing Each

Year”, available at http://fishcount.org.uk/published/std/fishcountstudy.pdf

Yew-Kwang Ng (1995). “Towards Welfare Biology: Evolutionary Economics of Animal

Consciousness and Suffering,” Biology and Philosophy 10:255-285

Philip Low (2012). “The Cambridge Declaration on Consciousness,” available at

http://fcmconference.org/img/CambridgeDeclarationOnConsciousness.pdf

F. Barbara Orlans (1998). “History and Ethical Regulation of Animal Experimentation: An

International Perspective,”  in Helga Kuhse and Peter Singer (eds.), A Companion to

Bioethics, Oxford: Blackwell

Lynn U. Sneddon. “Can Animals Feel Pain?,” Pain, The Wellcome Trust, available at

http://www.wellcome.ac.uk/en/pain/microsite/culture2.html

Lynne U. Sneddon, Victoria A. Braithwaite and Michael J. Gentle (2003). “Do fishes have

 nociceptors? Evidence for the evolution of a vertebrate sensory system,” Proc. R.

Soc. Lond. B 270:1115–1121

Brian Tomasik (2012). “How Many Wild Animals Are There?,” available at

http://www.utilitarian-essays.com/number-of-wild-animals.html

Karthikeyan Vasudevan, Ajith Kumar and Ravi Chellam (2001). “Structure and composition

of rainforest floor amphibian communities in Kalakad– Mundanthurai Tiger Reserve,”

Current Science 80(3):406-412

Karthikeyan Vasudevan, Ajith Kumar, Barry R. Noon, and Ravi Chellam (2008). “Density and

Diversity of Forest Floor Anurans in the Rain Forests of Southern Western Ghats,

India,” Herpetologica 64(2):207-215

R. W. Wilson, F. J. Millero, J. R. Taylor, P. J. Walsh,2,3 V. Christensen, S. Jennings, M.

Grosell (2009). “Contribution of Fish to the Marine Inorganic Carbon Cycle,” Science

323:359-362