I’d tend to agree that wild-caught small forage-feeding pelagic fish like anchovies/sardines might be among the animal products that humans could consume with the lowest counterfactual harm and the greatest nutritional benefit to people on otherwise largely plant-based diets.
From an animal advocacy perspective, I think its useful to have specific suggestions on what to eat for people who are resistant to going vegan or would find it very hard to eliminate all animal products, and this post has prompted me to consider small pelagic fish more seriously.
A small nitpick is that I believe the duration of stress experienced by small pelagic fish in European purse seine fisheries is likely to be in excess of the 10minutes you suggest. I’ve only encountered one study relevant study here (Marcalo et al (2006)), which involved looking at stress markers in sardines in Portuguese fisheries sampled throughout the capture process. That study suggests the capture process can take 90-120minutes from encircling through to the end of transfer fish on board, and and most stress markers seemed to increase throughout time in the net. Although this point in itself would not change the high-level conclusions (as the key crux for me is how stress of capture compares to the counterfactual death), I do worry your claim here was overconfident and that other arguments you’ve made in this post might not stand up to scrutiny.
One item I’d like to explore more is whether similar arguments might also apply to slightly larger pelagic forage fish like Mackerel and Herring. My understanding is by-catch is also relatively low for these species (though they are often captured by pelagic trawlers), but they have a much higher aggregate value and value per fish. The possibility of developing and implementing humane capture technologies (e.g. electrical stunning) seems greater for these species than for sardines/anchovies, and fewer would need to be consumed to achieve a given level of calories/nutrition.
Thank you so much for your thoughtful feedback, Sagar, and especially for pointing out the issue with the stress duration. That’s very helpful. We really aren’t experts on this matter. We’ve tried to research this issue as thoroughly as we could and made a real effort to seek out input from marine biologists and fish welfare researchers. That said, we weren’t able to find anyone with specific expertise on sardines and anchovies. As a result, there’s still a lot of question marks for us, which is why we are so keen to get the feedback of people like yourself. :)
It seems with net hauling we mistakenly based our numbers on a study citing net setting times, without realising that this did not include haulage. We looked into the study your shared as well as more recent studies and it seems the 90 to 120 minutes time from net set to loading on board is a good estimate to go for, though it seems it can sometimes even exceed this time if slippage is required or the boat is overloaded. We have updated our claims and would love to hear your thoughts on the uncertainty this may introduce in the trade off between deaths in the wild vs from fishing!
We personally think the direct suffering experienced by the sardines and anchovies may not be the most important consideration, especially as they appear to have a very low moral weight. Instead, the environmental impacts of different protein sources and their indirect effects on wild animal populations might dominate the overall animal welfare implications.
We also think it’s important to add that what we’re really trying to explore with this post is not only whether sardines and anchovies might be a good option for people who are resistant to going vegan, but also whether they could be a meaningful and defensible choice for people who already consider themselves vegan. That is, of course, a more controversial angle, and one where the nutritional considerations play an especially important role.
Regarding larger pelagic forage fish like mackerel and herring, we did not consider them for several reasons:
They are higher in the food chain and longer-lived, which increases the risk of heavy metal and microplastic accumulation.
Their more complex hunting behaviours likely indicate a higher welfare range, which increases the potential for direct suffering, though this might change if stunning technologies could be implemented (thanks for raising this interesting point!).
Unlike sardines and anchovies, which feed on rudimentary, brainless planktonic organisms, mackerel and herring prey on more complex animals, including small fish like sardines and anchovies. Building on this post (thank you, Bella!), reducing mackerel/herring populations could increase sardine/anchovy numbers, and if those species lead net-negative lives, this could worsen overall outcomes. We don’t think this consideration applies to planktonic organisms.
Due to their higher trophic level and longer life spans, mackerel and herring stocks may be less resilient to fishing pressures, raising concerns about scalability.
Pelagic trawling, which is more commonly used to catch mackerel and herring, likely results in higher GHG emissions and slightly higher bycatch than purse seining/ring netting.
Sardines and anchovies are heavily used as feed in salmon farming, which is less the case for herring and mackerel. So eating sardines/anchovies might simply shift them from feed to food use, potentially displacing part of the salmon industry (yay!), whereas the same effect is less likely with mackerel or herring.
That said, our thinking on mackerel and herring is based on limited exploration. We haven’t looked into these species in detail, and we’d be really curious to hear your thoughts on whether any of these assumptions seem off.
FWIW, I wouldn’t consider planktonic animals necessarily brainless or unworthy of moral consideration. Peruvian anchoveta eat krill, which I imagine to be sentient with modest probability, and copepods, which I consider worth researching more.
Thanks! Appreciate your thoughtful reply, especially on mackerel/herring.
I don’t think stress duration itself changes the overall conclusions of this work, especially as my best guess is that increasing human (food) demand for sardines/anchovies might not counterfactually increase catch volumes for these species (unless it increases political pressure for increased quotas).
That said, I remain pretty uncertain about what the consequences of pushing up demand for these species might be. On the one hand, it could push up prices for fishmeal, which in turn could raise input prices for aquaculture for carnivorous/omnivious species. But on the other hand, it could also push up prices for sardines/anchovies, which in turn could push omnivores/pesceterians who currently eat sardines/anchovies to eat other foods. Those second-round effects seem pretty hard to know with confidence, but seems pretty plausible those folks could switch from anchovies/sardines to farmed salmon/shrimp or land-based meats like chicken.
I agree with your point on catch volumes. Though I do hope I’ll find a fisheries expert to evaluate this quota claim properly, as it seems so critical.
Regarding your second point, I don’t share all your concerns about the second-order consequences. Sardines and anchovies are currently used primarily as low-value inputs for fishmeal. The margins of sardines and anchovies for human consumption exceed that of fishmeal, so reallocating catch to human diets would likely not change the price.
Frozen sardines in the UK are currently priced slightly below chicken per gram of protein while also providing valuable EPA/DHA (canned sardines are a bit pricier due to packaging and processing costs).
For salmon or shrimp, their dependence on fish meal would make them more expensive were sardines/anchovies to shift to human consumption, making them less appealing substitutes.
Thanks for all your input :) Really appreciate it!
Thanks for a very thought provoking post.
I’d tend to agree that wild-caught small forage-feeding pelagic fish like anchovies/sardines might be among the animal products that humans could consume with the lowest counterfactual harm and the greatest nutritional benefit to people on otherwise largely plant-based diets.
From an animal advocacy perspective, I think its useful to have specific suggestions on what to eat for people who are resistant to going vegan or would find it very hard to eliminate all animal products, and this post has prompted me to consider small pelagic fish more seriously.
A small nitpick is that I believe the duration of stress experienced by small pelagic fish in European purse seine fisheries is likely to be in excess of the 10minutes you suggest. I’ve only encountered one study relevant study here (Marcalo et al (2006)), which involved looking at stress markers in sardines in Portuguese fisheries sampled throughout the capture process. That study suggests the capture process can take 90-120minutes from encircling through to the end of transfer fish on board, and and most stress markers seemed to increase throughout time in the net. Although this point in itself would not change the high-level conclusions (as the key crux for me is how stress of capture compares to the counterfactual death), I do worry your claim here was overconfident and that other arguments you’ve made in this post might not stand up to scrutiny.
One item I’d like to explore more is whether similar arguments might also apply to slightly larger pelagic forage fish like Mackerel and Herring. My understanding is by-catch is also relatively low for these species (though they are often captured by pelagic trawlers), but they have a much higher aggregate value and value per fish. The possibility of developing and implementing humane capture technologies (e.g. electrical stunning) seems greater for these species than for sardines/anchovies, and fewer would need to be consumed to achieve a given level of calories/nutrition.
Thank you so much for your thoughtful feedback, Sagar, and especially for pointing out the issue with the stress duration. That’s very helpful. We really aren’t experts on this matter. We’ve tried to research this issue as thoroughly as we could and made a real effort to seek out input from marine biologists and fish welfare researchers. That said, we weren’t able to find anyone with specific expertise on sardines and anchovies. As a result, there’s still a lot of question marks for us, which is why we are so keen to get the feedback of people like yourself. :)
It seems with net hauling we mistakenly based our numbers on a study citing net setting times, without realising that this did not include haulage. We looked into the study your shared as well as more recent studies and it seems the 90 to 120 minutes time from net set to loading on board is a good estimate to go for, though it seems it can sometimes even exceed this time if slippage is required or the boat is overloaded. We have updated our claims and would love to hear your thoughts on the uncertainty this may introduce in the trade off between deaths in the wild vs from fishing!
We personally think the direct suffering experienced by the sardines and anchovies may not be the most important consideration, especially as they appear to have a very low moral weight. Instead, the environmental impacts of different protein sources and their indirect effects on wild animal populations might dominate the overall animal welfare implications.
We also think it’s important to add that what we’re really trying to explore with this post is not only whether sardines and anchovies might be a good option for people who are resistant to going vegan, but also whether they could be a meaningful and defensible choice for people who already consider themselves vegan. That is, of course, a more controversial angle, and one where the nutritional considerations play an especially important role.
Regarding larger pelagic forage fish like mackerel and herring, we did not consider them for several reasons:
They are higher in the food chain and longer-lived, which increases the risk of heavy metal and microplastic accumulation.
Their more complex hunting behaviours likely indicate a higher welfare range, which increases the potential for direct suffering, though this might change if stunning technologies could be implemented (thanks for raising this interesting point!).
Unlike sardines and anchovies, which feed on rudimentary, brainless planktonic organisms, mackerel and herring prey on more complex animals, including small fish like sardines and anchovies. Building on this post (thank you, Bella!), reducing mackerel/herring populations could increase sardine/anchovy numbers, and if those species lead net-negative lives, this could worsen overall outcomes. We don’t think this consideration applies to planktonic organisms.
Due to their higher trophic level and longer life spans, mackerel and herring stocks may be less resilient to fishing pressures, raising concerns about scalability.
Pelagic trawling, which is more commonly used to catch mackerel and herring, likely results in higher GHG emissions and slightly higher bycatch than purse seining/ring netting.
Sardines and anchovies are heavily used as feed in salmon farming, which is less the case for herring and mackerel. So eating sardines/anchovies might simply shift them from feed to food use, potentially displacing part of the salmon industry (yay!), whereas the same effect is less likely with mackerel or herring.
That said, our thinking on mackerel and herring is based on limited exploration. We haven’t looked into these species in detail, and we’d be really curious to hear your thoughts on whether any of these assumptions seem off.
FWIW, I wouldn’t consider planktonic animals necessarily brainless or unworthy of moral consideration. Peruvian anchoveta eat krill, which I imagine to be sentient with modest probability, and copepods, which I consider worth researching more.
Thanks! Appreciate your thoughtful reply, especially on mackerel/herring.
I don’t think stress duration itself changes the overall conclusions of this work, especially as my best guess is that increasing human (food) demand for sardines/anchovies might not counterfactually increase catch volumes for these species (unless it increases political pressure for increased quotas).
That said, I remain pretty uncertain about what the consequences of pushing up demand for these species might be. On the one hand, it could push up prices for fishmeal, which in turn could raise input prices for aquaculture for carnivorous/omnivious species. But on the other hand, it could also push up prices for sardines/anchovies, which in turn could push omnivores/pesceterians who currently eat sardines/anchovies to eat other foods. Those second-round effects seem pretty hard to know with confidence, but seems pretty plausible those folks could switch from anchovies/sardines to farmed salmon/shrimp or land-based meats like chicken.
I agree with your point on catch volumes. Though I do hope I’ll find a fisheries expert to evaluate this quota claim properly, as it seems so critical.
Regarding your second point, I don’t share all your concerns about the second-order consequences. Sardines and anchovies are currently used primarily as low-value inputs for fishmeal. The margins of sardines and anchovies for human consumption exceed that of fishmeal, so reallocating catch to human diets would likely not change the price.
Frozen sardines in the UK are currently priced slightly below chicken per gram of protein while also providing valuable EPA/DHA (canned sardines are a bit pricier due to packaging and processing costs).
For salmon or shrimp, their dependence on fish meal would make them more expensive were sardines/anchovies to shift to human consumption, making them less appealing substitutes.
Thanks for all your input :) Really appreciate it!