Español | 中文 | हिंदी | Bahasa Indonesia

Up to 1.3 billion farmed fishes were killed for food in the E.U. in 2017. Despite these numbers and an increasing amount of research suggesting that fishes are sentient beings, there has been little attention given to farmed fish welfare. This article outlines the failings of the current system and offers ideas for how a better system might be designed, mainly focusing on the United Kingdom.

At present, there are two systems for monitoring aquaculture: certification schemes and assessment frameworks. Certification schemes allow an external organization to judge what occurs on a farm according to set standards. Assessment frameworks are systems for judging the conditions on a farm. The author sees issues in all the certification schemes and assessments that currently exist, and gives four main criteria for how we can improve fish welfare in both.

First, the article highlights the need for ‘completeness.’ Existing certification schemes and assessment frameworks only look at part of the picture regarding fish farming. Most of the certification schemes barely mention fish welfare. Those that do tend to focus on ‘input measures’ that reflect farm conditions, including fish density or water quality. Further, they typically focus on small measurable aspects of a fish’s physical health such as the number of skin lesions.

To resolve this, the author suggests that both input and output measures need to be looked at when assessing fish welfare. In other words, how the fishes feel needs to be considered as well as the general conditions on a farm. Certification schemes and assessment programs should also include ways of measuring positive welfare, not just the absence of suffering.

The article suggests that we use ‘whole-animal measures,’ which reflect the overall well-being of the fishes in terms of both general physical and psychological health. Examples of whole-animal measures include:

• Whether a fish uses the brain hemisphere that processes pain more than the one used to process pleasure
• Whether a fish seems ‘happy’
• Whether a fish interprets ambiguous stimuli pessimistically, which tends to be associated with unhappiness
• The quality of skin mucosa (although this only reflects physical health)

Whole-animal measures for fishes are relatively new, so we don’t know which ones would work best to assess fish welfare. Certification schemes and assessment programs might also use a well-designed set of partial measures that address all the important welfare issues without bringing in extraneous information. Once whole-animal measures are developed, partial measures can also be used alongside them to highlight any specific issues.

Second, the author emphasizes that we need to ensure the validity of any welfare measures. The measures used need to actually track the welfare of fish, not other issues such as the productivity of the farm. Very few prospective measures of fish welfare have been shown to reflect what fish genuinely care about. The article warns against industry-run schemes, suggesting that expert panels would be best to ensure objectivity.

Third, the measures must be feasible. They need to be both simple and cost-effective. Non-invasive measures are better for feasibility (and for fish welfare). Digital technologies exist that would make it possible to track individual fishes over time in great detail. While the upfront costs may be high, they could provide precise information on welfare with little ongoing cost.

Fourth, and finally, is the issue of what counts as acceptable levels of welfare. Here, the author emphasizes that we need to base our welfare threshold on the experiences of the fishes themselves, not on arbitrary measures like the current industry situation. Animal welfare scientists, industry experts, and fish biologists should work together to decide on a threshold. We could also use ‘aspirational tiers,’ which include actively positive welfare for fishes (e.g., where they can explore their environment), rather than just limiting negative side effects.

With these four criteria in place, fish welfare schemes could help fishes significantly more than they currently do. However, there are some challenges. Different species of farmed fishes are very taxonomically different, so certifications and assessments need to reflect that. Fish welfare enforcement should go beyond losing a certification — some kinds of fish mistreatment should be illegal. Finally, any scheme would need to be transparent so that customers can understand what the label means. Confusing fish welfare labels are considered ‘humanewashing.’

The conclusions made here are useful for advocates. They give something solid for campaigners to work towards: compete, valid, feasible, and transparent systems that require a high standard of welfare, are customized for each species, and in some cases are legally enforceable. Animal advocates can demand the measures outlined here and criticize institutions that do not meet them.