The State Of Ethological Approaches To The Assessment Of Animal Suffering And Welfare

Applied Animal Behaviour Science 118 (2009) 105–107

Contents lists available at ScienceDirect

Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim

Editorial

The state of ethological approaches to the assessment of animal suffering and welfare§ Advances in the neurosciences, in ethology and animal cognition have told us to expect at least some of our fellow creatures to be capable of experiencing affective states such as pain, fear, anxiety, or pleasure. Although covering a wide range of discrete emotional experiences, affective states may all be categorized into states of suffering and well-being. From an ethological perspective what seems to be important in terms of animal welfare is whether or not an affective state is unpleasant enough for an animal to want to get out of it (Dawkins, 2008). If it is, it is considered to reflect suffering, if not, it is considered to reflect wellbeing. This conception provides us with a behavioural way of recognising animal suffering, and has put ethology at the front of animal welfare sciences. However, the assessment of animal welfare is not a matter of the natural sciences only. Thus, whether or not a particular behavioural measure is relevant to the assessment of animal welfare critically depends on the ethical criteria according to which animals are to be protected. These ethical criteria are developed by philosophers and agreed on by society as a whole. They determine the legal means by which animal protection is implemented, but also the kind of scientific evidence that counts in the assessment of animal welfare. Animal ethics is hotly debated, for example between utilitarian animal welfarists and deontological animal rights advocates (Singer, 1975; Regan, 1983). This debate is, however, not the focus of this special issue. Instead, this special issue is aimed to provide an overview of the state of ethological approaches to the assessment of animal suffering and welfare, based on the ethical principles underlying current animal welfare legislation. The focus is placed on concepts and methods that allow us to assess animals’ affective states in view of reducing animal suffering and improving animal well-being. For one thing: what animals need for their welfare needs to be justified biologically by the animals’ own values. Whether animals do indeed have the capacity to experience feelings of suffering and well-being, still ranks among the hardest and most controversial questions in

§ This paper is part of a special issue entitled ‘‘Animal Suffering and Welfare’’, Guest Edited by Hanno Wu¨rbel

0168-1591/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.applanim.2009.02.021

biology (Koch, 2004). Feelings of suffering and well-being are subjective experiences that cannot be assessed directly by scientific methods, since the natural sciences strictly adhere to a third person perspective. Therefore, biology has long dismissed subjective experiences as a target worthy of scientific investigation, which probably explains why we still know rather little about them. In addition, based on an epistemological fallacy – what cannot be investigated cannot exist – animals were simply denied the capacity for subjective experiences by many scientists. This is actually surprising, given that pain and suffering have long been assessed and successfully treated not only in humans but also in veterinary medical practice. Moreover, not even in humans is there a direct way of assessing subjective experiences. There is no fundamental difference between animals and humans with respect to the scientific assessment of suffering and well-being. Not even language makes a fundamental difference. Although verbal communication is a useful tool for medical doctors in the diagnosis of human pain or suffering, it is neither unambiguous, nor applicable to all patients. In September 2007, in an attempt to leave this arbitrary distinction between humans and other animals in the attribution of affective states behind, the International Society of Livestock Husbandry (Internationale Gesellschaft fu¨r Nutztierhaltung, IGN) together with the Chair of Animal Welfare and Ethology of the Justus-LiebigUniversity of Giessen (Germany) held a scientific symposium entitled ‘Animal Suffering and Well-Being—International Symposium on the State of Science’. The aim of this symposium was to present and discuss recent advances in the study of animal suffering and well-being. The present special issue is based on contributions to this symposium. The issue starts with a review by David Fraser on the history of the study of affective states of animals, showing how this concept varied throughout human history, but also between ethnic groups and – not least – between different schools of thought within the natural sciences. Based on pioneers such as Jane Goodall, Barbara Smuts and others, his paper concludes with a plea for the collection of qualitative, narrative data in addition to quantitative data in view of understanding the unique features of individual animals. He argues that such data almost require scientists

106

Editorial / Applied Animal Behaviour Science 118 (2009) 105–107

to involve affect in order to achieve plausible explanations of behaviour. Not only how animals ought to be protected, but also why we want to do this, may be an ethological question. After all, animal protection is human behaviour. In the first part of his paper, Hanno Wu¨rbel thus takes an ethological perspective on animal protection with a view to better understanding how our own behavioural biology shapes our attitude towards animals. This may be crucial in view of avoiding unjustified anthropocentrism and uncritical anthropomorphism. In the second part of the paper, he then reviews concepts and methods used to assess animal welfare in terms of their significance for ethical decisiontaking, and proposes ethologically meaningful measures of ‘behavioural integrity’ as potentially powerful indicators of animal welfare. Due to the sentientist conception of current animal welfare legislation, one of the main targets of animal welfare science is to determine the boundary between sentient and non-sentient animals. Most welfare regulations are restricted to vertebrates, and even among vertebrate scientists, there is debate as to where to draw that boundary. Bob Elwood and colleagues review evidence that crustaceans might experience pain and stress in ways that are analogous to those of sentient vertebrates. They conclude that there is considerable similarity of function, although different systems are used, and thus there might be a similar experience in terms of suffering. In light of their findings, the treatment of crustaceans – and possibly other invertebrates – may need to be reconsidered. Based on a similar case made for fish in the recent past, Nicola Brydges and Victoria Braithwaite examined whether environmental enrichment of housing conditions for fish can have similar effects on learning, memory and temperament behaviours as found in rodents and many other captive mammals. In contrast to invertebrates and fish, the assumption that pigs can experience pain has been unchallenged for a long time. Nevertheless, we still know very little about nociceptive processes in pigs. Therefore, Mette Herskin and colleagues set out to validate a laser-based method to measure thermal nociception in pigs. Their findings suggest that behavioural responses to nociceptive cutaneous laser stimulation provide valid measures of nociception that can be applied in group-housed gilts, with minimal disturbance of the daily routines. One of the most influential concepts in applied ethology is the concept of motivational strength based on measures of demand. The interpretation of measures of demand in the assessment of animal welfare is based on several underlying assumptions. Lucy Asher and colleagues have tested two critical assumptions underlying most studies using measures of demand to assess animals’ preferences for different resources. Using European starlings, they demonstrate that presenting costs on an ascending order may yield valid measures of demand, but that random order presentations should be used to check for order effects. In contrast, they found that the concept of a finite energy budget may not be useful in the measurement of captive animal’s preferences, and that a different approach might therefore be needed. The most exciting additions to the toolbox of animal welfare scientists since the introduction of measures of

demand are undoubtedly measures of cognitive or judgement bias. These measures are known from human cognitive psychology to reflect affective states, especially emotional valence. Together with colleagues, Mike Mendl and Liz Paul present a comprehensive review of studies on rats, dogs, rhesus monkeys, starlings and humans that have since been conducted, and discuss the strengths and weaknesses of different affect manipulation treatments, and other aspects that may affect the observed effects and their interpretation in terms of affective states. One problem with current cognitive bias studies is that they require extensive training to establish the necessary positive and negative associations. Brilot and colleagues present an experiment using European starlings’ (Sturnus vulgaris) responses to eyespot stimuli that are naturally aversive to many bird species, and require no explicit associative training. Although their present approach was unsuccessful in determining cognitive bias, it may provide important guidance for the future search of measures of cognitive bias that do not depend on extensive associative training. The past years of animal welfare research have also witnessed an extension of focus in the assessment of animal welfare from using measures of negative affect (i.e. suffering) to using measures of both negative and positive affect. Manteuffel and colleagues present a review of their development and applications of rewarded instrumental learning as a form of cognitive enrichment, which enables farm animals to acquire instrumental behaviour that may improve their well-being. They conclude that this approach may have the potential to enhance not only animal welfare but also product quality in animal farming. This is followed by a paper by Reefmann and colleagues who studied whether ear and tail postures in sheep can be used as measures of emotional valence from negative to positive states. They found that frequent ear-posture changes were most clearly associated with situations inducing negative states, while a high proportion of passive ear postures was associated with situations inducing positive states. Thus, ear-posture changes represent a promising indicator of emotional valence in sheep. To conclude this section on positive affective states, Jonathan Balcombe reviews the evidence in the literature for animal pleasure and discusses its moral significance with respect to animal welfare legislation. The final section of this special issue deals with an important new development in farm animal production, breeding for disease resistance. Even under the most favourable housing and management conditions, infectious diseases remain a constant threat to farm animals, and a serious one as infectious diseases are ranked among the major sources of suffering in animal production. In the first paper, Gerald Reiner introduces the approaches taken to genetic disease resistance in pigs. Using Sarcocystis miescheriana resistance in pigs as a parasitic disease model, he highlights the difficulties in identifying carriers of favourable gene variants, and novel tools and technologies emerging from recent advances in genomics to tackle these. In the final paper, Reiner and colleagues then present a detailed behavioural analysis of the course of this parasitic model disease. By relating the behavioural changes to various clinical and clinical–chemical measures, they were

Editorial / Applied Animal Behaviour Science 118 (2009) 105–107

able to identify behavioural indicators of disease even at sub-clinical stages, suggesting that even sub-clinical conditions may compromise the pigs’ welfare. Taken together, the papers compiled in this special issue reveal the high degree of sophistication by which ethologists are now able to infer affective states and make convincing scientific cases for feelings of suffering and well-being in animals. The ethological study of animal suffering and welfare has thus great potential to add objectivity to animal welfare debates, and may be crucial in facilitating the implementation of the ethical goals as stated in animal welfare legislation. However, the papers compiled in this special issue should also reveal existing gaps in our knowledge and research, thereby contributing to the future advancement of ethology applied to animal ethics and to the improvement of animal welfare.

107

References Dawkins, M.S., 2008. The science of animal suffering. Ethology 114, 937– 945. Koch, C., 2004. The Quest for Consciousness: A Neuroscientific Approach. Roberts & Company Publishers, Englewood. Regan, T., 1983. The Case for Animal Rights. University of California Press, Berkeley. Singer, P., 1975. Animal Liberation. Harper Collins Publishers, New York.

Hanno Wu¨rbel* Division of Animal Welfare and Ethology, Justus-Liebig-University of Giessen, 35392 Giessen, Germany *Tel.: +49 641 99 38750; fax: +49 641 99 38759 E-mail address: [email protected] Available online 17 March 2009