Our study was conducted in compliance with the Norwegian National Committee for Research Ethics in the Social Sciences and Humanities (NESH).Footnote 4 Specific protocols ensuring ethical compliance were approved by the Norwegian Center for Research Data (NSD), who undertakes ethical clearance on behalf of the University of Oslo. Informed consent was obtained at two levels. First, we sought permission to attend courses and conduct the study from the institution(s) responsible for the welfare courses. Permission was granted with no obligations in relation to the institution(s). Secondly, informed consent was obtained for individual voluntary participants in our study, who were de-identified in our notes and subsequently anonymized, in accordance with NSD guidelines. Voluntary participation was achieved as follows: we identified ourselves as independent researchers to all course participants at each course before they began. Information about the study was made available to all participants in the form of an informational pamphlet that we handed out at the commencement of each course. Participants who chose to speak with us were assured that they could withdraw from the study at any time and were given appropriate information about how to do so. A similar protocol was applied for the visits to aquaculture facilities.
Welfare courses were selected as the main point of entry into research on fish farm workers’ perceptions of fish welfare in the Norwegian aquaculture industry, partly because these courses were precisely and explicitly about fish welfare. Moreover, due to their mandatory nature, welfare courses brought together aquaculture employees with different areas of expertise, levels of experience, professional backgrounds, and companies (albeit largely from the same geographic localities).
During 2017 and 2018, we (the first and second authors) attended five welfare courses, taking place over 2–3 days in various localities along the country’s southwestern coast, between the cities of Trondheim and Stavanger. We also conducted ten interviews with course participants, including well-boat captains, on-site aquaculture personnel, slaughterhouse employees, operations managers, and course instructors from several different companies, representing a diverse range of actors within the industry. Of the interviewees, whose ages ranged from approximately 20–50, the majority (80%) were male. All of our interlocutors were Norwegian, or had been living in Norway for several years, and all interviews were conducted in Norwegian. We did not collect individual data on educational attainment, but Norwegian fish farm workers typically have secondary education, and many also have vocational degrees relevant to maritime or aquaculture industries. Participants working in administration often hold degrees in business administration or similar.
The first author of this article also conducted two all-day field visits to aquaculture facilities to further familiarize themselves with the contexts discussed at the welfare courses, often held at locations (hotels, conference centers, schools) in close proximity to fish farms. The second author had conducted fieldwork at a fish welfare course approximately five years prior to this study (cf. Lien, 2015), and has extensive field experience from aquaculture facilities, which informed our questions and the interpretation of our data.
Our methodological approach was based on the anthropological method of participant observation, supplemented by semi-structured interviews and informal discussions with welfare course participants during lunches, breaks, and after courses. Our observations and data were gathered in field notes, which were subsequently parsed and analyzed. This methodological approach allowed us to be present during the entirety of the courses, learning and repeating the course material in the same way that other participants did, and participating in the same formal and informal discussions, workgroups, and activities such as quizzes. Participant observation also conferred us the benefit of being considered “one of the participants”, providing us access to discussions and conversations outside of the courses themselves, whilst maintaining a clarity about our role as independent researchers at these courses—in other words, what James Spradley (1980) terms “moderate participation.”
Informed participant observation enabled us to recruit interlocutors without much difficulty. Some course participants effectively recruited themselves, while we chose to approach others based on comments made during courses, and arrived at others through “snowball sampling”, where existing interlocutors help to recruit acquaintances. All of the interlocutors we spoke to were informed about the nature of our research, our adherence to national research ethics guidelines, and the fact that all participation was voluntary.
Welfare courses comprised advantageous field sites for our research. Based on common, government-mandated guidelines, they provided insight into how animal welfare in aquaculture was addressed by legislative bodies through regulations. For course instructors, who often had a veterinary background, they provided the opportunity to discuss scientific knowledge and governmental regulations about fish and welfare with a broad group of fish farm workers with hands-on experience with the practical implementation of regulations. For participants, courses provided information, competency, and values, and served as classrooms for a universal crash-course (or refresher) on animal welfare, but also as arenas for professional and social interactions between participants who encounter animal welfare in different parts of the aquaculture operation. Welfare courses were therefore valuable sites for exploring how fish welfare is mobilized by regulatory bodies and course instructors, but also how fish welfare concerns are encountered in the day-to-day labor of fish farm workers, thus bringing together theoretical and practical knowledge. In this article, we focus primarily on the perspectives of course participants,Footnote 5 paying close attention to the ways that questions, statements, and concerns were formulated as well as to the ways in which the contexts (formal course settings, informal group discussions, and informal individual discussions) might affect the data gathered. We strove to prepare our questions with care. In conversations with fish farm workers, we attempted first to ask questions that elicited descriptive responses of concrete fish welfare-related situations, before encouraging farm workers to make more normative statements regarding fish welfare at their workplace.
One potential drawback of our methodological approach is that less caring attitudes towards fish welfare, existing perhaps under a veneer of compliance, might have been difficult to discern (due to the courses’ explicit focus on fish welfare and because of their mandatory nature). However, focusing on how welfare regulations are presented, discussed, and even challenged, still elicits insight into such issues as compassion fatigue and burnout, which would have been difficult to obtain by other means.
The findings of this study are presented in two parts. The first part addresses Norwegian authorities’ legal norms and state-of-the-art knowledge on fish welfare, as seen through an analysis of the pedagogical content of the welfare courses we attended. Here, we suggest that the courses present participants with two models—one that describes ‘how things are’ (concerning fish, welfare, aquaculture, regulations), and another that describes ‘how things ought to be’. While the first model was primarily articulated as part of the course curriculum, the second model was developed through sharing experiences of best practice, formally and informally. In this way, welfare courses also constituted arenas where attitudes towards fish welfare were shaped, both ontologically and normatively, in terms of ‘how fish are’ and in terms of how aquaculture operations ought to be.
The second part of the findings concerns farm workers’ own experiences of animal welfare regulations in their everyday aquaculture operations, highlighting dilemmas as they are experienced by the workers themselves and recounted to us either during or between welfare courses. These are important, because we maintain that fish farm workers are particularly well-positioned to notice welfare challenges, as they experience contradictory demands, objectives, and regulations related to welfare in their work. They are also a group whose insight is seldom reflected in studies and guidelines, which tend to be developed and discussed mainly by other professions such as veterinarians and biologists. Meeting fish farm workers in the context of a welfare course is therefore an excellent opportunity to notice dilemmas that might not register on the radar screens of other professionals.
Findings: Welfare courses as constitutive
While the welfare courses we attended differed in some respects (e.g. location, duration, number of participants, gender balance), all shared a common core and structure. Every course we attended began with a statement of purpose, which was to ensure that all participants were ‘on the same page’ regarding the knowledge required to secure fish welfare for salmon, and to provide the attitudes, knowledge, and skills deemed necessary to maintain good fish welfare in aquaculture operations.
Grounding ethics, normalizing fish sentience
Following this statement of purpose, courses moved on to a discussion of ethics, drawing briefly on Pythagoras and Descartes to illustrate some of the ways that animals and animal welfare have been treated in Western philosophical discourse. At every course we attended, the course instructors made it clear that little doubt remains over the assertion that fish are sentient beings, capable of feeling pain. At one course, the instructor introduced this topic by asking for a show of hands from participants who believed that fish were incapable of feeling pain. No hands went up, although social pressures and participants’ knowledge of expected responses could well have influenced this unanimous response. On the whole, the question of whether fish are capable of feeling pain appeared to be settled, and less relevant than the more immediate question of “how can I avoid hurting the fish unnecessarily?” For this part, course instructors presented a review of practical procedures such as feeding, handling, stunning, counting, and killing, emphasizing not only the ways that things are normally done, but why they are done this way; in other words, how welfare considerations have contributed to shaping quotidian aquaculture practices.
While three of the courses moved quickly from introductions such as those mentioned above, to more recent or pragmatic welfare entanglements, two of the courses included anecdotes from the past, or from other parts of the world. These anecdotes appeared to have been included to highlight how a) perceptions of animal welfare in Norway have changed over time, and/or b) how these differ in various contexts or cultures. While the overall narrative on changing perceptions of animal welfare was one of progress (towards increased awareness of and compliance with animal welfare considerations), the cross-cultural comparisons tended to portray Norway as being at the forefront, with other countries or cultures lagging behind.
Anecdotes such as these included images and video clips of farm animals being subjected to blatant animal abuse and cruelty, eliciting appalled responses from participants. One video clip showed a group of men in stereotypically Arab garb throwing live sheep into the back of a truck, hitting, kicking, and berating those sheep who did not immediately comply. Another showed a live fish being prepared and eaten in an Asian restaurant. The course instructor commented that many of the images depicted seemed to be from Asia and the Middle East and asked rhetorically whether this meant that they were “bad people”. He asked, “Is there is big difference between these images and the way we do things in Norway?”, prompting a mix of responses from participants. The course instructor continued: “It is better here—but why?” After a short interlude, he answered his own question: “Because the Food Safety Authority is breathing down our necks—but also because we have attitudes and knowledge that these [people in the videos] haven not received. There is a need for information and education, because you must know what a fish needs—and that is something we have in Norway.”
Ethics as cultural and legislative process
Through teaching material such as these videos, participants were invited to distance themselves from certain attitudes towards animal welfare, and concomitantly to self-identify as part of a group or culture that treated animals in a more ethically sound fashion. If Norway is imagined as better than other countries, it is made clear that this is not a result of culture, but of national differences in legislation and state control. Furthermore, such anecdotes suggest that although animal welfare is perceived differently in other places,’we’ are at the forefront. Just as stories about animal welfare in the past mobilize progress to explain differences over time in people’s perceptions of animal sentience, cross-cultural comparisons give the impression that some countries, or parts of the world are ‘ahead’. Animal welfare is in this way inscribed into time, location, and moral universe according to which cultures are organized hierarchically (for a similar process in relation to the story of animal domestication as a story of progress, see Lien et al., 2018). Participation in welfare courses entails participation in the work of such inscription and offers a guide to appropriate (morally sound) attitudes and behaviors, and contributes towards underpinning a sense or moral obligation amongst the participants.
As part of a historical overview of farm welfare legislation, all the courses mentioned the Brambell Report of 1965. This is a historically important document for animal welfare in IFAP contexts, as are the closely linked “Five Freedoms”: freedom from hunger or thirst, freedom from discomfort, freedom from pain, injury, or disease, freedom to express normal behavior, and freedom from fear and distress (FAWC, 1979).Footnote 6 Course instructors often held these freedoms up as ideals (how things ought to be), before asking participants if these freedoms were present for the salmon with whom they worked (how things are).Footnote 7 Responses to this question varied, with some course participants raising the uncomfortable question of whether any of these freedoms are achievable in industrial aquaculture.
All of the welfare courses reviewed national regulatory demands for animal welfare as well as those of various “welfare certification schemes” such as the RSPCA and Global Gap. There were also questions about who defines good fish welfare: veterinarians, fish biologists, consumers, journalists, the EU, “the market”? Various indicators of fish welfare (and non-welfare) as well as systems for assessing welfare on fish farms, such as the Salmon Welfare Index Model (SWIM) (Stien et al., 2012) were also reviewed and discussed in relation to data on salmon (and cleaner fish) physiology.
In each course, course instructors emphasized that fish farm workers were responsible for ensuring good welfare, and for communicating their experiences of regulations and practices that affect welfare “upward” through the hierarchical tiers of the company, even to the authorities and to legislative bodies. In this way, fish farm workers’ responsibility for fish welfare (and for situations in which welfare is compromised) was emphasized, as were their roles. These roles can involve conflicting demands, but they also lay down the legal basis of empowerment when it comes to “speaking up” and possibly also resisting the pressure to implement practices that are seen as problematic.
“Speaking up”: participants’ agreements and concerns
Active participation in the courses varied greatly. On some occasions, some 15–20 participants remained silent, dutifully taking notes or studying the educational materials they were provided. During one course, hardly anyone volunteered responses to the instructor’s questions. At others, participants actively engaged in discussions with the course instructors and one another, and hands were frequently raised to compare best practices in the workplace and to raise concerns about fish welfare in their daily work. There was at times an underlying tension in these conversations, where participants seemed to want to relate specifics of their workplace, but perhaps felt apprehensive about doing so in front of their colleagues and, in some cases, their superiors.
Courses typically included participants from different areas of production from several different companies. Not surprisingly, we found that informal discussions in smaller groups—during lunch breaks, for instance—yielded more specific information on matters of concern, and issues that had been raised in the plenary sessions were articulated more sharply, with more affective engagement, in these more informal conversations.
Overall, the interviews and discussions during welfare courses suggested that there was a great deal of agreement about many of the common issues raised regarding fish welfare for farmed fish. Participants expressed this of their own accord or when they were prompted by course instructors to reflect on their own experiences related to animal welfare. For the most part, it seemed as though most fish farm workers agreed that the welfare of their fish was generally good—although, as one interlocutor said, “it could always be better”. However, there were also several instances of informal conversations when fish farmers spoke more frankly, voicing concerns about the welfare-related challenges that they experienced in their work, and sharing their frustration with a smaller group of colleagues (and the authors). In the following section, we look at these concerns in more detail and highlight some of the dilemmas that are articulated by fish farm workers.
Findings: Fish farm worker perspectives
Welfare challenges as perceived by participants were mostly related to the way that fish welfare was safeguarded—or not—in everyday practice. The most common concerns were the welfare implications of sea lice treatments on salmon (brushing, warm water treatments, water flushing, pumping), the welfare of cleaner fish, issues related to overcrowding, long-term stressing of the salmon, and the intensification of industrial measures to increase growth, pushed by both the government and by private aquaculture actors. These were frequently related to us in solemn and sometimes emotional accounts, and we were struck by the profound sense of occupational and personal or moral responsibility for fish welfare that mobilized these accounts.
Their accounts and the way that they were related indicated that some fish farm workers saw themselves as stuck between “the devil and the deep blue sea” with regard to fish welfare, unable to balance contradictory demands. This was particularly brought to the fore by fish farm workers who were directly responsible for fish welfare on a day-to-day basis at aquaculture facilities.
Economies of scale versus moral considerations
The contradictory demands that fish farm workers outlined can be described as a classic conflict between economic and moral considerations. On the one hand, economic demands and expectations of increased, intensified, and more efficient production instilled a sense of responsibility in fish farm workers towards contributing to these objectives. This sense of economic responsibility is supported by major economic rewards, strong competition between private aquaculture actors, and the Norwegian authorities’ stated aim of a significant increase in aquaculture production made public in a white paper in 2015 (Ministry of Trade, Industry & Fisheries, 2015; see also Lien & Law, 2019).
On the other hand, there is the personal and professional responsibility that fish farm workers harbor for the welfare of the fish they care for. It should come as no surprise that many fish farm workers acknowledge fish sentience and express empathy and affective relations for the fish they care for. This has been observed prior to the implementation of new fish welfare regulations, including mandatory welfare courses (Lien, 2015).Footnote 8
We suggest that the introduction in 2010 of welfare legislation specific to aquaculture acted to confirm concerns about fish welfare relative to other IFAP industries in Norway. While the welfare courses themselves may not sufficiently address the underlying reasons for some fish farm workers’ concerns, they act to assert the legitimacy of such concerns and simultaneously provide an arena for their more forceful articulation and discussion. Hence, we can hear remarks such as the following, recounted by a farm worker during a welfare course:
A bit ‘too quick for its own good’ [litt fort i svingene] is an apt characterization of several things going on in the industry … there’s too much money in it … it’s the growth, you know, the fucking demand for continued growth. When you have business leaders that made billions of crowns, it is clear that some corners will be cut. They have to be.
Furthermore, and somewhat paradoxically, complying with a range of different and sometimes contradictory welfare regulations can create unintended and sometimes problematic consequences, even for practices ostensibly set in place to safeguard fish welfare, which is another cause of concern for several of the fish farm workers with whom we spoke.
This dilemma of contradictory, sometimes irreconcilable considerations is illustrated in Fig. 1.
(Not) Caring for cleaner fish
We suggest that what fish farm workers experience as irreconcilable considerations are best understood as indications of a structural dilemma, reflecting contradictions in the way that aquaculture is organized (Størkersen et al., 2021). The intensification of production, for example, may well be expressed as increased numbers of fish in the fish farms as well as the density of aquaculture sea cages in Norwegian fjords—both which provide ideal conditions for sea lice infestations (Taranger et al., 2015). Consequently, they require more measures to combat sea lice, such as the use of cleaner fish or other delousing treatments. According to fish farm workers, these are the instances in which fish welfare was most often compromised. As one interlocutor recounted about the use of cleaner fish:
We’ve gradually gotten to the point of having three different [types of fish] in the cages, with three different physiologies, [and] swim bladders: wrasse [closed swim bladder], lumpfish [no swim bladder], and salmon [open swim bladder]. That means a lot more work for us, because we should of course treat each type of fish differently. At the same time, we bring wrasse up [to the surface] when we bring up salmon, and they turn themselves inside out [de vrenger seg]—sure, it’s uncomfortable. It’s horrible.
In their final sentence, the interlocutor is referring to barotrauma, and specifically to the way that fish with closed swim bladders (or none at all) have no means of quickly regulating pressure changes brought about by rapid changes in depth. When they are suddenly brought up to the surface (often along with salmon), the air in their swim bladders expands rapidly, often causing serious internal damage to the fish and sometimes forcing their innards to emerge from their mouth and other orifices.
Handling the “losers”
Another issue that farm workers brought up was what is referred to in the industry as “loser” fish—those which grow too slowly during their first months in the cages. These fish are systemically sorted away when they are spotted and either culled or relocated to a separate cage. Here, they wait to be “asphyxiated and ground up with the rest of the dead fish” [daufisk]. In this way they are made invisible through the ordering of the assembly line (Lien, 2015: 135). This suggests both a conceptual and physical separation from “healthy salmon” (those that have thrived and grown quickly in the first weeks in the cage). When asked what was upsetting about this, one interlocutor replied that sorting out the fish that do not fulfill the industry’s expectations for rapid growth was “very uncomfortable”, because they were basically sending them to an early death. Here we see how the requirements for standardization and efficiency in the highly industrialized production systems is at odds with what could be seen as natural variation of salmon in the cage. But while the “early death” was perceived as uncomfortable by one farm worker, it could also be understood as a way of preventing or alleviating pain in the future. This is in keeping with the way that the entire set-up is geared towards uniformity, a condition that inevitably marginalizes young fish who have not shown adequate rates of growth. At stake, then, is how morality changes with the change of perspective, from caring for fish as individuals (as in the comment above), to caring for a cohort, such as a cage (or tank). The latter is the perspective that guides most operations and indicates that a hierarchy is at work in which individuals less likely to survive or to grow to an economically profitable size are systematically downgraded and eliminated, in order to safeguard the uniformity of each cohort.