








































Bangladesh Journal of Bioethics 2017; 8(1): 1-6  
 

1 

Original Article 

 

Wild Fish and Expected Utility 

 

Bob Fischer 

Department of Philosophy, Texas State University, San Marcos, Texas, U.S.A. 

Email: fischer@txstate.edu 

 

Abstract: It’s difficult to process the number of fish killed annually by the fishing industry. 

Nevertheless, governments are encouraging people to eat even more fish—see, e.g., the 

USDA dietary guidelines—and although animal advocates certainly don’t concur with this 

advice, they generally haven’t prioritized fish in their lobbying efforts. Given the influence of 

utilitarianism on animal advocacy, the odds are good that this is motivated by an expected 

utility calculation. For those concerned about fish, is there any way to defend them against 

this calculation? I argue for an affirmative answer: once you factor in an asymmetry between 

fishing and terrestrial animal agriculture, the expected utility calculation comes out in favor 

of devoting resources to reducing fishing. 

 

Keywords: fish; animal advocacy; expected utility 

 

Introduction: It’s difficult to process the number of fish killed annually by the fishing 

industry. Based on FAO statistics from 1999–2007, Mood and Brooke put the global number 

between 970 billion and 2.74 trillion (excluding illegal fishing, bycatch, fish that escape from 

nets, etc.)
1
. What’s more, their estimate is probably low. They base their calculations on 

average global catches of 77 million tons, but Pauly and Zeller make the case that the FAO’s 

statistics were far too conservative
2
. They contend that global catches peaked at 130 million 

tons in 1996, and have only been declining at a rate of 0.38 million tons per year since then—

not the 1.2 million tons per year that the FAO claimed. If Pauly and Zeller are right, then a 

more realistic estimate is 1.64–4.63 trillion fish per year. But even if we stick with the most 

conservative estimate—i.e., 970 billion—we should note that it’s nearly fourteen times larger 

than the number of terrestrial animals killed annually for food, which is around 70 billion
3
. 

The fishing industry is responsible for a staggering loss of life. 

 

Nevertheless, governments are encouraging people to eat even more fish—see, e.g., the 

USDA dietary guidelines—and although animal advocates certainly don’t concur with this 

advice, they generally haven’t prioritized fish in their lobbying efforts. The Humane Society 

of the United States (HSUS), for example, which is the largest animal welfare organization in 

the U.S., spends a considerable amount on its Farm Animal Protection team. However, none 

of those funds are specifically devoted to fish, and at present, the majority of resources go 

toward relieving the suffering of layer hens. In such circumstances, can animal advocates 

justify focusing on terrestrial species? Elder and Fischer (forthcoming) argue that they can’t, 

at least when it comes to farmed fish
4
. But what about wild fish? 

 

mailto:fischer@txstate.edu


Bangladesh Journal of Bioethics 2017; 8(1): 1-6  
 

2 

To answer that question, we need to consider the framework that guides much U.S.-based 

animal advocacy. Granted, there are many considerations relevant to animal advocates. 

However, whether they are welfarists or abolitionists, nearly all U.S. animal advocacy 

organizations have a pragmatic bent, demonstrated by their willingness to collaborate with, 

and sometimes even celebrate, retailers and producers of animal products. Instead of insisting 

on ending the farming of a particular species, or devoting all efforts to vegan education, 

they’re willing to lobby for relatively small welfare improvements. So, even if they 

ultimately hope to end the human use of animals, they seem willing to think like utilitarians 

in the present. This, of course, is one of Gary Francione’s longstanding complaints about 

these organizations. I take no stand here on whether this particular complaint is justified. 

However, given that utilitarian reasoning does seem to guide U.S. animal advocacy, we can 

approach the question of whether they ought to be more concerned about wild fish within that 

framework. What’s more, this assumption is a charitable one, as it appears to make sense of 

their priorities. After all, suppose we run an expected utility calculation. Some wild-caught 

fish suffer for hours or days on long-lines or gill nets, but most don’t: their swim bladders 

explode as a result of coming to the surface too quickly, or they suffocate on the deck, or they 

are crushed under the weight of other fish. So let’s just suppose, quite generously, that the 

average wild-caught fish suffers for an hour before dying. Then, it’s relatively easy to 

calculate the life years of suffering produced by the fishing industry—roughly 110 million, 

using the most conservative estimate, and 529 million, using the most liberal one. Globally, 

however, six billion layer hens are killed each year, each of whom lives between 12 and 24 

months in very unpleasant conditions
5
. So, we don’t even need to factor in broiler chickens—

which easily outnumber layers by a factor of six or seven—to see that the life years of 

suffering due to the chicken industry are much, much higher than those produced by the 

fishing industry. HSUS, PETA, and other organizations look to be reasonable in emphasizing 

the plight of layers, and it makes sense that they are now beginning to shift their attention 

toward broilers
6
. 

 

Discussion: Is there any way to defend fish against this calculation? I think so: as I’ll argue, 

expected utility considerations break in favor of fish. Granted, that alone won’t show that 

advocacy organizations ought to shift resources away from chickens—or any other terrestrial 

animal—and toward fish. However, it will be enough to shift the burden onto animal 

advocates: they will need to explain the other factors that outweigh the consideration of 

expected utility. 

 

The key move is to consider an important difference between fishing and chicken production. 

If you reduce chicken consumption, you reduce chicken breeding, and the absolute number of 

chickens will drop. But if you reduce fish consumption, you don’t reduce the absolute 

number of fish. You reduce the number of fish in aquaculture, but you extend the lives of 

some portion of the billions (or trillions) of wild fish who would otherwise have been caught. 

Indeed, you probably get a significant gain in overall utility, as many of those now-uncaught-

fish have long lifespans. Salmon, for instance, can live 3-8 years; tilapia, 9-11 years; Atlantic 

cod, as many as 25 years; and Bluefin tuna, up to 30 years. To see how this goes, let’s just 

suppose that you’re able to reduce global chicken and fish production by just 1% each. That 



Bangladesh Journal of Bioethics 2017; 8(1): 1-6  
 

3 

would save 700 million chickens from an awful life, but even if we use the most conservative 

kill estimate, it would save 9.7 billion fish, some of whom can live for thirty years or more. 

This dramatically shifts the expected utility calculation in favor of fish. 

 

There are two objections that someone might make at this juncture. The first involves 

skepticism about fish sentience. I think the evidence clearly undermines this maneuver—see, 

e.g., Victoria Braithwaite’s work
7
—but let’s ignore this. To make this objection work, you 

need to specify the probability you’re willing to assign to fish sentience relative to chicken 

sentience. Suppose you have a very dim view of fish: you think that fish are—averaging 

across species—only a tenth as likely to be sentient as chickens. Still, fish will win by a huge 

margin: effectively, we’ll be running the expected utility calculation with 970 million fish as 

opposed to 9.7 billion, which is still greater in absolute numbers, and we haven’t yet factored 

in lifespans. 

 

A better move is to deny that the lives of fish are, on balance, worth living
8
. If wild fish 

generally experience lives in which suffering predominates, then it becomes much harder to 

make a utilitarian case for saving them. 

 

How might someone justify this stance on the lives of fish? The standard line seems to be 

this. First, we underestimate how much suffering occurs in the natural world: many fish 

suffer due to disease, injury, lack of food, pollution, and as a result of predation. Second, the 

vast majority of fish live short, painful lives, where there are few opportunities for pleasure. 

This thought is motivated by reflections on the distinction between K and R reproductive 

strategies. Humans employ a K strategy: we have few offspring and we invest a great deal of 

resources in keeping the ones we have alive. Most fish, however, pursue an R strategy: they 

have hundreds, thousands, or even millions of offspring, invest very little in them, and only a 

few of them survive. The unlucky offspring starve or are eaten relatively soon after birth, and 

on the assumption that these beings are conscious, this suggests that they experience 

considerable suffering before death. Third and finally, we tend to be subject to various biases 

that make it hard for us to appreciate just how bad things are for most sentient beings. For 

instance, an availability bias leads us to focus on large, mature animals instead of small, 

young ones; wishful thinking leads us to predict that things are better than the evidence 

suggests; and a powerful cultural narrative involves idealizing natural environments, despite 

evidence to the contrary. 

 

This pessimistic view isn’t relevant to calculations about the fish that are actually saved, 

since those fish are the relatively successful ones. (After all, fishing boats aren’t after 

hatchlings.) However, the worry isn’t primarily about those individuals anyway, but about 

their offspring. Nevertheless, things may not be as bad for the expected utility argument as it 

might seem. First, whatever the suffering in the lives of unsuccessful offspring, we can’t infer 

that suffering predominates until we know how much pleasure these animals get from various 

sources—e.g., having novel experiences, or feeding. To my knowledge, no one who defends 

the pessimistic view takes up this issue. Second, it’s worth noting that the reproductive 

strategy argument is based on the assumption that suffering is associated with all failures—



Bangladesh Journal of Bioethics 2017; 8(1): 1-6  
 

4 

i.e., unsuccessful attempts to feed, mate, etc.—while pleasure is associated with successes. 

Since most animals aren’t successful most of the time—as R strategies involve sacrificing 

most offspring to create the mature individuals in the next generation—it follows that 

suffering predominates in nature. However, it isn’t clear that the assumption about suffering 

and failure is justified. Suffering is likely to be associated with certain failures—e.g., those 

that result in organisms crossing some threshold of unsatisfied need—but since organisms 

can be motivated to satisfy their needs without suffering, it isn’t clear why we should think 

that suffering will be associated with all failures. And if it isn’t, then the short lives of most 

organisms may be relatively pleasant until, suddenly, they aren’t—e.g., when a predator gets 

them—which means that pleasure may dominate. 

 

Third and finally, the hypothesis that a huge number of animals live net negative lives is 

bound to have dramatic moral implications. It’s hard to see why we should make a concerted 

effort to preserve species with R reproductive strategies if we are, thereby, ensuring that the 

world contains much more suffering than it would otherwise. Nevertheless, this is what many 

of our conservation efforts involve. So, the standard of evidence here should be high: it isn’t 

enough that there’s a reasonable argument for the pessimistic hypothesis from some general 

considerations about reproductive strategies, the prevalence of disease and injury, and so on. 

We need more detailed ethological studies before reaching this conclusion. The upshot here is 

that it isn’t clear that the pessimistic view is true, and until we have more evidence in favor of 

it, it seems reasonable to set it aside in our deliberations. So, it seems to me that the expected 

utility argument still favors devoting more resources to relieving the plight of wild fish. 

 

At this juncture, perhaps the best move for chicken advocates (or those otherwise interested 

in focusing on terrestrial animals) is to abandon utilitarianism for painism—i.e., the view that 

we ought to minimize suffering, rather than the view that we ought to maximize net utility
9
. 

This fits neatly with the rhetoric of many animal advocates, who are, in general, much more 

vocal about reducing suffering than promoting happiness. More importantly, it will get the 

desired result: if we discount the good in the lives of fish, then the suffering of chickens will 

dominate.  

 

This is not the place to assess the merits of painism. For present purposes, then, let’s grant the 

chicken advocate this particular moral framework. Given that concession, we have to make a 

more speculative case on behalf of fish. Wild fish populations are in rough shape due to 

overfishing: the FAO estimates that 89.5% of wild fish stocks are either fully exploited or 

overexploited
10

, and Worm et al. argues that, absent intervention, we may face a fishless 

ocean by 2048
11

. That, of course, would result in widespread destruction of marine 

ecosystems, and it seems likely that such extensive oceanic changes would have ramifications 

for life elsewhere on the planet. What are the odds that the consequences will be good? 

Presumably, not very high. What seems more likely is that collapsing fish populations would 

cause immense suffering among the incredible number of other animals who depend on fish 

for their survival. So, even if our focus is solely on preventing suffering, there is a reasonable 

case for devoting attention to fish: efforts to scale back fishing are, inter alia, efforts to save 

the marine ecosystems on which so much life depends. 



Bangladesh Journal of Bioethics 2017; 8(1): 1-6  
 

5 

 

Conclusion: As mentioned earlier, I concede that the above doesn’t show that animal 

advocacy organizations ought to redirect resources toward fish. The above does show, 

however, that those organizations can’t rely on an expected utility calculation to justify their 

current priorities. Instead, they’ll need to argue that there’s simply no hope of convincing 

governments or consumers to limit fishing, or that market forces will check fishing without 

advocacy work, or what have you. Perhaps those arguments will be successful, though I have 

my doubts. Regardless, those are the arguments we now need to hear. 

 

 

Author Contribution: The author developed the conceptual idea and wrote the manuscript. 

 

Conflict of Interest: Declared none. 

 

Acknowledgement: Thanks to Max Elder, who got me interested in this issue, and to an 

anonymous referee, who made many helpful suggestions. 

 

References 

 
1
 Mood A Brooke P. Estimating the number of fish caught in global fishing each year. 2010 

http://fishcount.org.uk/published/std/fishcountstudy.pdf (accessed 9/17/2016). Such estimates 

are bound to be imprecise, as the FAO doesn’t report head counts. So, head counts have to be 

generated based on reports about (a) total catches, given in tons, not individual heads, (b) the 

distribution of species caught, and (c) information about the mean weights of members of 

those species. 
2
 Pauly D Zeller D. Catch reconstructions reveal that global marine fisheries catches are 

higher than reported and declining. Nature Communications 2016; 7. 
3
 See: http://www.ciwf.org.uk/media/3640540/ciwf_strategic_plan_20132017.pdf (accessed 

9/23/2016).  
4
 Elder M Fischer B. Focus on fish: a call to effective altruists. Essays in Philosophy 

forthcoming. 
5
 See: http://www.poultrytrends.com/ (accessed 9/23/2016). Of course, U.S. advocacy 

organizations are primarily concerned with U.S. production. However, the relative numbers 

don’t change if we limit ourselves to the U.S. market, so I’ll keep things simple by using 

global numbers, as this spares us some complications in estimating how many wild fish are 

killed to serve American consumers. 
6
 There would be no problem for fish if we were supposed to factor in the sheer number of 

lives, and not just life years of suffering. If we were, then as long as lives got some non-

trivial weight in the calculation, it would be likely to come out in favor of fish. However, this 

isn’t likely to be compelling to the relevant organizations: it’s a feature of utilitarianism—not 

a bug—that it ignores lives per se. If utilitarian reasoning is behind the decisions of animal 

advocacy organizations, we’ll need to defend fish in those terms. 
7
 Braithwaite V. Do fish feel pain? Oxford University Press, New York; 2010. 

http://fishcount.org.uk/published/std/fishcountstudy.pdf
http://www.ciwf.org.uk/media/3640540/ciwf_strategic_plan_20132017.pdf
http://www.poultrytrends.com/


Bangladesh Journal of Bioethics 2017; 8(1): 1-6  
 

6 

 
8
 Horta O. Debunking the idyllic view of natural processes: Population dynamics and 

suffering in the wild. Telos 2010; 17: 73-88. Ng Y-K. Towards welfare biology: evolutionary 

economics of animal consciousness and suffering. Biology and Philosophy 1995; 10: 255-85. 

Tomasik B. The importance of wild-animal suffering. Relations 2015; 3(2): 133-152. 
9
 Less dramatically, they might adopt some variant of utilitarianism that significantly 

discounts pleasure relative to pain. For independent considerations in favor of this maneuver, 

see: Shriver A. The asymmetrical contributions of pleasure and pain to animal welfare. 

Cambridge Quarterly of Healthcare Ethics 2014; 23(2): 152-162. 
10

 FAO. The state of world fisheries and aquaculture 2016: Contributing to food security and 

nutrition for all. 2016 http://www.fao.org/3/a-i5555e.pdf (accessed 9/28/2016). 
11

 Worm B Barbier E Beaumont N Duffy J Folke C Halpern B et al. Impacts of biodiversity 

loss on ocean ecosystem services. Science 2006; 3(314): 787-790. 

http://www.fao.org/3/a-i5555e.pdf

