Are seafood substitutes good for wild fish?


I conducted preliminary research into the effects of seafood substitute products on reducing the suffering of wild sea animals. This was suggested by Scott Weathers, who is considering launching a clean meat start-up.

Seafood replacements products (which include plant-based products or those made from cell culture) are gaining popularity and traction across the world. Fish, meanwhile, are the most numerous vertebrates killed by humans. Between one and three trillion wild fish, and 40-120 billion farmed fish, are slaughtered each year. If more people eat seafood substitutes, will wild animals suffer more or less?

There are numerous considerations here from: the quality of wild fish lives; the effects on wild animal populations; the replacement of fish consumption; the quality of farmed fish lives; and the suffering experienced during death. In this post, I describe the possible ramifications of seafood substitutes on wild animals, and give a brief qualitative assessment of which types of substitutes are likely to do the most good for wild sea animals.

Population considerations

Changing consumer demand for fish products is likely to have some kind of impact on wild animal populations, and thus on overall suffering, but it’s somewhat difficult to tell what the net sign of that effect is.

The population effects of farmed versus wild fish depend largely on the product. For example, canned tuna is produced almost solely from wild fish while almost all catfish is farmed. Salmon is harvested from both farmed and wild fish. The introduction of plant-based meat might also have indirect replacement effect,s e.g., the introduction of a plant-based cod product might motivate people to eat it rather than tuna or tilapia as well as cod, or vice versa. I don’t consider these replacement effects here, since this is primarily an economic question, but its impact is certainly relevant.

If I consider only wild-caught species, the impact on animal populations is still complicated to assess, as the vast majority of wild fish we eat are predators. Tuna populations have declined over time due to fishing. However, populations of their prey might have increased, and then their prey’s prey (smaller fish or plankton including copepods and other small invertebrates) will also experience population ramifications. Brian Tomasik has explored these ecological ramifications from fishing in more detail (Tomasik, 2015a, 2015b).

In addition to direct population shifts, we must bear in mind that most fish lay many eggs that hatch and die at early stages in their lives. Therefore, increasing the number of breeding adult fish will introduce larger numbers of larval fish who will not survive to adulthood. Fewer larval fish probably die in fish farming operations, since food is reliably provided.

Lastly, we might consider that farmed fish are fed fishmeal from wild fish as a protein source, mostly from small and numerous forage fish. Farmed salmon consume the most fishmeal. Carp and tilapia are also fed substantial amounts of fishmeal, although about five-fold less than salmon (the rest is plant-based) (“Fishmeal and fish oil facts and figures,” 2016, p. 32).

A straightforward consideration is that if fish lives are net-negative, and small animals are more sentient per unit biomass than large animals (see the section “Shrimp fishing” below), suffering is most reduced by maximizing the numbers of large animals at high trophic levels (Tomasik, 2015a, 2015b).

For now, bycatch and gradual population decline aside, the simple assumption I make is that for every wild fish caught, resources are freed that result in one new fish.

Farmed fish welfare

The welfare of farmed fish has been described elsewhere e.g., (The Humane Society of the United States & Index, 2008). Suffice to say, farmed fish are kept in close confinement, usually stressed, not allowed to exhibit natural behaviors, and are susceptible to disease. Their welfare is practically unregulated compared to other animals. I would guess that they have net-negative lives. However, they are also free from nonhuman predation, and have fewer large parasites than wild fish (see later down).

Farmed fish are generally killed without attention to humane slaughter, via processes similar to those used on fishing boats. Tilapia are fasted for 1-5 days, then netted and put in ice, water, or dry crates to be moved to the point of slaughter (Lines & Spence, 2012, p. 2). They are then chilled to immobilize them for exsanguination/evisceration. Catfish are fasted for 1-5 days prior to slaughter, which is not thought to cause significant suffering. They are then crowded and netted and moved into dry containers, where they are chilled and then die by asphyxiation or decapitation (Lines & Spence, 2012, p. 3).

Overall quality of lives of wild fish

Tentative conclusion: The lives of wild fish are better in some ways and worse in others than farmed counterparts. They are not clearly better or worse off overall.

I lean towards wild fish lives being somewhat negative, with great uncertainty. It would take a lot more knowledge about wild fish experiences and behavior for me to be more certain about this. On the other hand, this isn’t clearly a relevant question since, as I mentioned above, the simple assumption is that each fish death will result in the freeing of resources to create one new fish, or perhaps several more zooplankton, or a combination of the two. I definitely don’t know enough about zooplankton experiences and behavior (or capacity to suffer) to determine the net-sign of their existence, though zooplankton show some evidence of certain kinds of avoidance learning and thus may be able to feel pain. With these considerations in mind, I focus here on comparing wild and captive fish lives, and later, on the kinds of death faced by natural means versus human fishing.

Wild fish are unlikely to suffer from cramped conditions, as their farmed counterparts do, and are more able to perform natural behaviors. On the other hand, wild fish are exposed to predators, starvation, and other suffering caused by nature. For instance, Dutil and Lambert (2000) found that on average, 38.6% of adult wild cod found at any given time appear to have poor body conditions associated with starvation (Dutil & Lambert, 2000). Wild fish tend to have more parasitic worms than farmed fish do, for instance, up to 90% of populations of some wild fish species have nematode infections while almost no farmed fish (which are treated to prevent infections) do (Zukin, 2016). Parasitism probably decreases animal quality of life substantially (Ray, 2017).

I believe there are studies looking at comparative levels of cortisol or other stress hormones in wild versus farmed fish, but I haven’t explored these for this post.

What do wild fish most commonly die from?

My naive guesses for the commonly-preyed-upon species are, in order:

  • Predation;
  • Starvation;
  • Disease;
  • Fishing;
  • Reproductive mortality (dying on the way to / due to energy expenditure from reproduction which overlaps with some of the above causes); and
  • Old age.

I’m not sure what large fish, like sharks or tuna, usually die from (if not fishing). It may be the case that tuna are commonly predated by whales or sharks.

How much do fish suffer when eaten by predators?

Tentative conclusion: Painful but likely to be quick. Fish who are swallowed whole probably suffocate in the predator’s stomach. Fish who are torn beforehand experience more pain but die much sooner. The available research on this topic is quite scant.

Most large fish swallow smaller fish whole or in chunks. They have a muscular and elastic esophagus and intestine, and teeth on both jaws and their tongue. Food is further processed in the intestine. Digestion only happens in the gizzard and stomach, no digestion occurs in the mouth (as it does in humans).

The gizzard – not present in all fish – grinds food up and creates digestive enzymes. The stomach is acidic. Some fish don’t have stomachs, but these are generally ones who eat algae or similar foods rather than live fish. Predatory fish tend to have a stretchy Y-shaped stomach. They probably use hydrochloric acid in their stomachs, though I’m not sure in what concentration, and how much that expands or shortens time until death for prey swallowed whole.

Fish like the Black Swallower seem to be able to kill prey just by enveloping it, after which, presumably, the prey asphyxiates.

How much do fish suffer from asphyxiation?

Tentative conclusion: Asphyxiation is probably bad and somewhat extended. Asphyxiation is likely involved in most human- and predator-caused deaths.

Asphyxiation can be lengthy. A Humane Society report suggests that it takes 2-10 minutes for a trout to die from asphyxiation in air, depending on the ambient temperature (more quickly at higher temperatures) (Yue, 2008). Other studies report apparent loss of sensation around 15 minutes, and up to 200 minutes until death (The Humane Society of the United States & Index, 2008).

Naively, I would expect this is similar to the amount of time it would take a swallowed-whole fish to die in a predator’s stomach. (This is assuming a swallowed fish dies of asphyxiation, and not considering possible changes from chewing, digestion, or acid exposure. The presence of water may also influence how intense the suffering is, though I’m not sure in which direction).

If a fish is chewed prior to swallowing, I expect they die either immediately or much more quickly than one swallowed whole. This would involve, presumably, more pain (from both crushing and asphyxiation) over a much shorter amount of time. I wasn’t able to find research on the behavior or survival of fish swallowed alive, so these are speculations.

How much do fish suffer from death by slaughter?

Tentative conclusion: Fish caught by predatory animals are likely to suffer less than many fish caught using standard fishing methods, particularly in trawl nets where fish suffer for an extended amount of time in the water. Asphyxiation is a major direct cause of death in fish caught by humans.

A fish caught for food may suffer more than a fish being eaten alive. Most predatory fish do not chew their prey, so fish are swallowed alive and probably die in the stomach from asphyxiation and possibly acid. That said, I’d guess that even fish who are chewed alive will be in excruciating pain more briefly than the extended process involved in commercial fish slaughter.

Both fishing boats and predators have to chase their prey. During catching food, schools of swordfish “have potentially invested hours into injuring and exhausting their prey” (Herbert-Read et al., 2016), which is a comparable timespan to, say, trawling or purse seining (Yue, 2008). I would expect these chases to be comparatively stressful to prey as evading a fishing net. That said, the fish probably dies fairly quickly after the swordfish catches up, so there is no painful build-up of fish as in the back of trawl nets, and the asphyxiation or freezing period is not extended as we’d expect in a ship.

With regard to methods of fishing, I’m not sure if fish caught by purse seine suffer significantly before the net is pulled tight – they may be able to behave fairly naturally while the net is being laid out. Due to the extended suffering involved in trawl nets, cod-fishing probably increases suffering greatly versus natural cod lives. Purse-seining (as to catch anchovies) is also worse than wild deaths, but not as bad as trawling, since the duration spent in the water is shorter. Purse-seining does, however, affect many more fish – consider that the Atlantic cod usually weighs 40 kg as an adult, and an anchovy weighs less than 30 g. One ton of caught anchovies contains over 100 times more fish as one ton of cod.

Shrimp fishing

Shrimp have smaller brains and probably fewer cognitive capacities than equivalently-sized fish, and thus, possibly, experience less suffering. But shrimp for food contains more individuals (and probably more neurons) per pound than larger fish. Tomasik proposes several other mechanisms than simple neuron count by which small animals might matter more per gram – faster subjective experiences (Tomasik, 2017a), efficiency, and behavior, among other factors (Tomasik, 2017b). I think these considerations are important and highly relevant. As one might imagine, there are no regulations regarding the humane slaughter or treatment of shrimp.

More surprisingly, shrimp also has an astonishing amount of bycatch. There’s an average 85% discard rate by mass from shrimp catches (Clucas, 1997), much of which is fish (and, more rarely, sea mammals and turtles). These animals are caught and suffer for extended periods in trawl nets.Thus, between the bycatch rate and the potentially enormous amount of suffering entailed in catching such small animals in quantity, I think shrimp replacement products are a promising area for reducing suffering. Additionally, the bycatch and potential harm to rare and showy species e.g., sea mammals and turtles, might make it more appealing to conservation-minded consumers.

Tentatively promising opportunities for seafood substitutes

  1. Shrimp products: Catching shrimp likely involves causing large amounts of suffering to a truly enormous number of animals.
  2. Farmed salmon products: It’s hard to tell whether the lives of farmed fish are worse or better than the lives of wild fish, but their deaths likely involve considerably more suffering than natural wild deaths. Reducing number of salmon caught also reduces the enormous number of small wild fish killed for fishmeal to them. This has a negative impact by shortening those lives, although I am uncertain about the sign of those lives, and about whether death from fishing in these cases involves more suffering than their eventual death in the wild.
  3. Cod and tuna products: Cod and tuna, as well as other fish who are trawl-caught, experience extended suffering before death that is likely significantly worse than what they would experience in the wild.
  4. Catfish, and tilapia products: It’s hard to tell whether the lives of farmed fish are worse or better than the lives of wild fish, but their deaths likely involve considerably more suffering than natural wild deaths. These are fed wild-caught fish products like salmon, although many fewer – their diets are mostly plant-based.


Clucas, I. (1997, October). Discards and bycatch in Shrimp trawl fisheries. Retrieved January 16, 2018, from

Dutil, J.-D., & Lambert, Y. (2000). Natural mortality from poor condition in Atlantic cod (Gadus morhua). Canadian Journal of Fisheries and Aquatic Sciences. Journal Canadien Des Sciences Halieutiques et Aquatiques, 57(4), 826–836.

Fishmeal and fish oil facts and figures. (2016, December). Retrieved January 16, 2018, from

Herbert-Read, J. E., Romanczuk, P., Krause, S., Strömbom, D., Couillaud, P., Domenici, P., … Krause, J. (2016). Proto-cooperation: group hunting sailfish improve hunting success by alternating attacks on grouping prey. Proceedings. Biological Sciences / The Royal Society, 283(1842).

Lines, J. A., & Spence, J. (2012). Safeguarding the welfare of farmed fish at harvest. Fish Physiology and Biochemistry, 38(1), 153–162.

Ray, G. (2017, November 22). Parasite Load and Disease in Wild Animals – Wild-Animal Suffering Research. Retrieved January 16, 2018, from

The Humane Society of the United States, & Index, A. (2008). An HSUS Report: The Welfare of Animals in the Aquaculture Industry. Retrieved from

Tomasik, B. (2015a, December 22). Which Marine Trophic Level Contains the Most Total Suffering? Retrieved January 16, 2018, from

Tomasik, B. (2015b, December 23). Trophic Cascades Caused by Fishing. Retrieved January 16, 2018, from

Tomasik, B. (2017a, October 24). Do Smaller Animals Have Faster Subjective Experiences? | Essays on Reducing Suffering. Retrieved January 16, 2018, from

Tomasik, B. (2017b, November 3). Is Brain Size Morally Relevant? | Essays on Reducing Suffering. Retrieved January 16, 2018, from

Yue, S. (2008). An HSUS Report: The Welfare of Farmed Fish at Slaughter. Retrieved from

Zukin, N. (2016, July 25). Almost Every Kind of Wild Fish Is Infected with Worms. Retrieved January 16, 2018, from