Present, not used (part 1): A spectrum of visibility

Sara’s M&M work has encountered animals “bred, not used” that are created for scientific research yet not utilised in an experiment, i.e. unsuitable, unneeded, or used in breeding more animals. Meanwhile, Ally’s work in S&S has featured animals “caught, not used” in wildlife research, i.e. animals accidentally caught in traps that are not wanted for research purposes due to their species, age, sex, or other characteristic (commonly known as “bycatch”).

In this post we argue that not all animals involved in UK research are equally visible, in a regulatory and ethical sense; instead, they fall along a spectrum of visibility. We illustrate this by discussing moves towards increasing the visibility of “not used” animals by making them count. The annual counting of animals by the Home Office, which oversees the Animals (Scientific Procedures) Act (A[SP]A), can be seen as a “political device” (Buller and Davies, 2013, p. 2) whereby regular, public reporting encourages regulatory and ethical accountability. Hence, making animals count appears to be mobilised to expand ethical concern to non-experimental animals by encouraging ethical accountability for them.

Bred, not used

Under A(SP)A’s definition, a regulated procedure refers to any procedure undertaken for an experimental, scientific, or education purpose, which is applied to a protected animal (vertebrates and cephalopods), and that may “caus[e] the animal a level of pain, suffering, distress or lasting harm equivalent to, or higher than, that caused by the introduction of a needle in accordance with good veterinary practice” (Art 2[1]). The breeding of a genetically altered animal who then comes to similarly suffer is also a regulated procedure (Art2[3]).

The Annual Statistics report only those animals used for “regulated procedures”, further divided into “experimental procedures” and “creation and breeding of GA [genetically altered] animals”: these are therefore visible to the public. Intriguingly, the way genetically modified animals have been categorised has shifted over time, opening up interesting questions about how the relationship between genetic alterations and suffering has been understood and represented; but these must remain for another occasion.

This focus on regulated procedures as the “unit” for counting animals makes other animals’ involvement in research less visible from a reporting perspective, even if they shared the same living space and regulatory protection. It means that, for instance, non-GA animals used for tissue collection only, or that were otherwise bred but not used, are not covered in the Annual Statistics, and until recently were not publicly accounted for.

Instead, the arrival of the 2010 European Directive and its translation into UK law via the update of A(SP)A in 2012 made these other “bred, not used” animals reportable for the first time. This new legislation mandated the five-yearly publication of an Additional Statistics report counting all animals bred for research and not otherwise reported in the Annual Statistics. These Additional Statistics were first reported for 2017, and returned a further 1.81 million animals (of which 1.4 million were mice), adding nearly 50% to the total number of animals involved in research. They have been welcomed as another step towards increased transparency in animal research reporting.

Yet, if counting brings animals into the public view, without further differentiation it can still occlude ethically significant differences. For instance, the report doesn’t account for all the reasons why an animal may be “bred, not used” nor the percentage of animals per reason, instead providing examples, e.g. they might have been the wrong age or sex for experiments. It also excludes farm animals not bred for research.

Most notably, as one of our interviewees noted, there is no demarcation in the statistics between which animals needed to be bred, and which did not. The statistics therefore group together so-called “unavoidable surplus” (e.g. animals bred as part of GA colonies but unusable due to having the wrong combinations of genes) with “avoidable surplus” animals, e.g. resulting from inefficient breeding practices. Statistics, then, may be a way to make animals visible, but defining how and why they count can point to deeply political and ethical questions of accountability.

Caught, not used

Unlike animals bred but not used, animals caught but not used are not counted at all by the Home Office, since capture is not a regulated procedure under A(SP)A. Trapped animals might be counted in other statistics, such as those collected (but not necessarily made public) for the Wildlife and Countryside Act (WCA). However, not all animals are covered by the WCA and other wildlife laws. As we’ve discussed in a recent paper, there’s plenty of oversight of bird capture for example, but essentially none for species like rabbits, foxes, and wood mice.

For many research participants (researchers, vets, and others), this presents a serious ethical problem, since trapping is widely reported to be the most stressful part of research for a wild animal, but the welfare of trapped animals may receive little regulatory attention. There was therefore a push to make trapped animals of all kinds more visible and to encourage greater attention to their welfare. For example, some participants proposed making capture a regulated procedure under A(SP)A, which would mean that “caught, not used” animals are counted in annual Home Office statistics and protected under A(SP)A. However, this idea also received push-back, for example from those concerned that the onerous licensing of A(SP)A would discourage important wildlife-focused citizen science from occurring.

Calls to account for trapping and “caught, not used” animals fit with a broader discourse about the need to consider the effects of wildlife research on non-target animals, and even on broader ecosystems. For example, philosopher Howard Curzer and colleagues have argued that the ‘3Rs’ framework typically applied to labs ought to expand to ‘9Rs’ in wildlife research, given the potential for such work to affect entire ecosystems, not just the welfare of individual study subjects.

While capture remains outside of A(SP)A’s remit (with little sign of changing), bycatch, trapping, and effects on non-target animals in wildlife research are now mentioned in wildlife-specific A(SP)A guidance released in 2016. Researchers sometimes talked about this as an important signal that the Home Office is starting to take a greater interest in animals whose welfare is affected by wildlife research, but are not the research subject.

However, there may still be a way to go before this ethical extension is fully realised: researchers sometimes indicated that they are not usually asked many questions about the effects of their research on non-target animals, although such questions may come up in other elements of oversight such as AWERBs. It therefore seems that the ethical scope of A(SP)A is expanding to consider those “caught, not used” via quasi-legal guidance, even though they still don’t have a formal place in the law and are not made visible via statistics.

A spectrum of visibility

Our examples here point towards what we might call a spectrum of visibility in animal research statistics. Animals caught but not used are wholly invisible in statistics reported by A(SP)A (and some are also invisible in other laws), while those bred but not used have partial visibility, which has increased over time.

Animals bred or caught but not used are by no means the only animals towards the invisible end of this spectrum; Rich and Reuben have both written about other cases when animals are not counted. Invertebrates other than cephalopods are excluded from protection under A(SP)A. As Rich has pointed out, this and similar laws in other countries result in a lack of ethical or legal consideration given to horseshoe crabs, whose blood is often extracted and used for endotoxin tests. Reuben has also written about how zebrafish embryos – which are nearly transparent and therefore useful for research in developmental biology – are not legally counted as animals under A(SP)A when they are below 5 days old, a pragmatic decision based on the principle of excluding “non-sentient” animals from A(SP)A’s oversight. Zebrafish embryos are therefore somewhat invisible both literally and metaphorically.

Counting can be a political move aimed at encouraging care for these less visible animals, since counting can “open the possibility and responsibility of care” (Buller and Davies, 2013, p.3). There will be little push to do this in some cases; for example, some may see little need to care more about zebrafish embryos. But in other cases, there clearly is a political motivation to count: those advocating for further regulation and counting of trapping do so out of concerns about animal suffering.

Of course, counting in statistics is just one way that animals can be more or less visible; some research animals may receive more affection than others, for example. And, as we saw in the trapping case, a key consideration is not whether animals are counted in statistics, but whether there are ways of ensuring that their welfare is considered and attended to. Counting in publicly available statistics also makes animals visible to a specific audience (the public), which is arguably not the most important group if the goal of counting is to improve or change animals’ lives (we’ll return to this topic in part 2 of this blog series).

As Buller and Davies (2013) have argued, counting “is one of the most basic ways and most common ways in which we cut [1] and make sense of and interact with non-human animals”, and it is also “about the act of being accountable” (p. 1). Defining how to count, which animals do and do not count, and how numbers are represented are therefore important aspects of doing animal research ethics.

Indeed, as we’ve seen through our examples, no matter how you count, statistics will inevitably highlight some things and obscure others, including potentially ethically important information like why an animal was bred but not used. Similar issues could arise with capture if statistics were to be collected; for example, should figures on bycatch include non-native invasive species, which many people might like to see trapped and killed? Would it be better to separate bycatch into native and non-native, and what would this achieve?

Numbers also “do not remove ambivalence or the potential for ethical contestation” (Buller and Davies, 2013, p.6). So what if some proportion of animals “bred, not used” did not need to be bred in the first place, for example? Should we care? If you believe healthy laboratory animals not used in experiments have perfectly good lives, and that “death is not a welfare issue”, you might say no, while others would disagree. The ethical questions brought about by animal research do not go away if more animals are included in statistics, but they are opened up in new, more nuanced and potentially useful ways. Using statistics to make animals visible can therefore be an important means to broaden ethical debates.

In this and a forthcoming post, then, we’ve sought to share some initial reflections on how these “present, not used” animals are made visible politically and ethically, how people care for and about these animals, and what this tells us about efforts to extend the scope of care and ethical concern beyond experimental subjects.

References

Buller H and Davies G (2013) Movements, metrologies and animal matterings. In: STaC Seminar, 2013.