For many people, laypeople as well as animal scientists and philosophers, animal welfare involves animal feelings. Scientifically, however, animal feelings are problematic. In the concluding remarks of a conference about the welfare of domestic animals in 1994, for example, two questions for further research were proposed: (1) What is the nature of feelings? and (2) Why is it not possible to measure the occurrence of feelings in animals directly? This book intends to give a philosophical and scientific-theoretical answer to both (...) questions. The two questions are transformed into the following: (1) How can we conceptualize animal experiences such as feelings or emotions in a philosophically and theoretically sound way? and (2) Which method is appropriate to gain knowledge of animal experiences?These questions are answered first by examining two ethological animal welfare theories, namely that of Wiepkema and Toates and that of Dawkins. These theories have animal feelings of welfare as their subject matter. My conclusion after examining these welfare theories is that they do not conceptualize and obtain scientific knowledge of animal feelings at all. These theories only study animal behaviour and physical aspects of animals and assume that these aspects refer to animal feelings of welfare. Because they are applications of animal ethology, these theories stick to the natural-scientific method of research. The use of this method leads to two different conceptualizations of animal welfare feelings. The theory developed by Wiepkema and Toates conceives of animal feelings as unobservable, internal causes of animal behaviour. In Dakwins' theory, seen as a logical-behaviouristic theory, "animal feelings" are names for particular forms of law-governed animal behaviour. My comment on both theories is that animal experiences are solely theoretical concepts or designations. These ethologists simply assume rather than demonstrate that particular animal behaviour is caused by or associated with animal experiences.Wiepkema, Toates and Dawkins also acknowledge this. In order to say that animal behaviour refers to subjective animal feelings, they use the argument from analogy that states that because there are similarities between human and vertebrate animal behaviour and (neuro)physiological processes, it is plausible that animals have similar experiences as humans. However, like most animal welfare scientists, they correctly say that this is an unscientific argument. My final conclusion is that it is impossible to infer subjective animal experiences from objectively studied, physical and behavioural phenomena.This conclusion is in line with the main thesis of philosophical hermeneutics. This thesis states that subjects cannot be studied as such in a natural-scientific way because the concept of 'subject' demands a different, non-objectivating method of knowledge. Philosophical hermeneneutics can perhaps answer the question of a proper concept for subjective animal experience, although animals are not the explicit subject matter of this discipline.Two founding fathers of 20th Century hermeneutics, namely Dilthey and Gadamer, are studied concerning the concepts of 'subject' and 'experience'. My conclusion is that the subject-philosophy of these two philosophers is primarily a philosophy for human subjects, thereby more or less excluding the possibility for animal subjectivity. Since Gadamer does not consider animals linguistic beings, he seems to exclude animal experience. However, he opens up the way to a philosophy about animal experiences as bodily experiences. Dilthey, in his later writings, follows the same line of reasoning as Gadamer emphasizing the common, cultural-historical meaning of experiences and their expression and understanding, which is typically human. In his early writings, however, Dilthey underlines the individual, subjective aspects of experiences in humans and higher animals. Although he later does not reject this psychological foundation of experience, he simply fails to ask: what about animal experience? thereby evading the question of how we as cultural and historical humans can understand non-cultural and non-historical animal experience. The line of reasoning these two philosophers have in common is that they say: if animals have experiences, then these experiences must be similar to human experiences. Actually, their subject-philosophy is a philosophy of human subjectivity which can hardly cover animal experiences.Rather than trying to find a philosophical concept of human experience that can cover animal experience, a third possibility between human experience and the absence of experience is looked for. The philosophies of the phenomenologists Merleau-Ponty and Plessner which emphasize the bodily character of human and animal experience offer this possibility.Merleau-Ponty's philosophical phenomenology seems to promise a conceptualization of bodily animal experiences. The conclusion of his analysis of human perceptions is that not the linguistic but the bodily way of human being is the foundation of human perceptions and experiences. He sees the human body as ambiguous: both physical and conscious (i.e., experiencing) and he assumes this philosophical analysis adequate to cover animal relationships with their environment. Inspired by the difference that Merleau-Ponty makes between the bodily comprehension of humans and the hermeneutic understanding of person, I propose the concept of the impersonal meaning of animal experience as an alternative of the personal, cultural and historical character of human experience.Because Merleau-Ponty repeatedly says that all living beings are intertwinings of physical body and consciousness, he seems to say that plants and micro-organisms are also experiencing beings, thereby making no distinction between animals and other non-human organisms.Plessner offers the required specification. He asserts that human as well as animal relationships to their environment are mediated by a self. To him, the human relationship to the environment is mediated by language, personality, culture and history. What Plessner calls a "double human self" - namely a bodily bound self and a reflective "I" - is the foundation of this relationship. He further states that human personality, culture and history give form to bodily experiences. Animals, by contrast, cannot distance themselves from their own body and bodily-bound self. Animals do not have the capacity of reflective, linguistic, personal or cultural-historical experience; they have only bodily and environmentally tied, here-and-now experiences. Thereby, Plessner provides an elaborated, philosophical concept of animal experience that is not similar to the concept of human experience.Buytendijk, who worked as an animal psychologist around the middle of this century, can be seen as applying Plessner's concept of bodily and environmentally tied animal experience to research. He fully adopted Plessner's philosophy of animals and humans and in his own animal experiments and in discussions of those performed by others, he adopts the view that animal behaviour is an expression of their experiences which are bound to the present Umwelt and bodily possibilities of the animals involved.When comparing the method that Buytendijk used with the hermeneutical and phenomenological method of understanding human experiences, one can see that:Unlike the meaning of human experiences, the meaning of animal experiences is not personal within a cultural-historical context. This is a vital, impersonal meaning that is bound to momentary bodily perceptions and actions in the present Umwelt.We can understand this meaning within a species-specific context.This species-specific context is not given beforehand as a standard for interpreting animal behaviour. We attain knowledge of this context by interpreting the meaning of expressions of particular animals of a species under various circumstances.Because animals, contrary to humans, are not open to others, we cannot share with them our knowledge of the meaning of their experiences.The conclusion of these comparisons is that our interpretation of the meaning of animal expressions always remains, conceptually and methodically, our human interpretation. Whether our interpretations of the meaning of animal expressions are more or less adequate depends on whether they meet the usual standards of hermeneutical understanding: coherence between interpretations and accordance with biological knowledge of the animals involved.Finally, two contemporary animal welfare debates are discussed. The first (a conference about welfare of domestic animals in 1994) is a philosophical and theoretical debate about the concept of animal welfare and the method for measuring it. The two main issues at that conference were: are feelings a fundamental aspect of animal welfare? and how can we measure animal welfare? The second question turned out to be the most important. Some participants said that feelings, although important, should not belong to the scientific concept of animal welfare because they cannot be measured; physical indicators of welfare are sufficient for speaking of welfare. Other participants held that we can indirectly know feelings by measuring physical and behavioural indicators. However, most of these indicators are the same as those used by animal welfare scientists who claim that they measure only the physical aspects of animal welfare, irrespective of associated animal feelings. Animal welfare scientists who stress feelings of welfare simply add that these observations refer to the animals' feelings of poor welfare. One participant of the conference argued that physical indicators of animal welfare cannot be used as indicators of feelings of welfare too; these feelings have to be demonstrated independently of physical indicators of welfare.In the second debate about welfare of farm mink, the validity of the argument that feelings have to be demonstrated independently has been exemplified. The question in this debate was whether some particular behaviour of farm mink, called "stereotypic behaviour", counted as expressions of poor welfare. All participants in this debate tried to demonstrate or contest this by means of natural-scientific experiments. A critical reading of this debate demonstrates that scientists only agree upon the designs and results of such experiments if they also agree upon the interpretation of the meaning of the animals' behaviour. Almost all the conclusions from experiments regarding mink's welfare can be contested from the point of view of another interpretation of the mink's behaviour. This debate shows the primacy of the interpretation of animal behaviour as an expression of experiences over results of natural- scientific experiments. This primacy requires an explicit method for interpreting animal behaviour in order to reach an agreement about various interpretations. The type of research into welfare of farm mink as used by one of the participants in this debate is considered as containing elements of such an interpretative method. Contrary to her own intention, the researcher does not see the mink's behaviour only as a causal effect of fulfilled or unfulfilled needs, but primarily as meaningful behaviour. By carefully looking at and comparing the form of the behaviour of farm mink and wild mink, she tries to interpret their behaviour. The aim of these interpretations is not to look into the animals' heads in order to see experiences as causes of animal behaviours. I state that the aim is to attain a coherence between the assumed meanings of various animal behaviours. This coherence provides us with a background for natural-scientific explanations of these behaviours, experimental results and other physical data.At the end of this book, the four principles of animal welfare research as an integration of interpretative and natural-scientific research are evaluated:In order to study animal welfare as subjective experience, a view is needed that conceptually and methodically maintains animal welfare as subjective experience. Saying that animal welfare is not only physical but primarily a matter of feelings, requires something other than physical measurements of welfare. The debate about farm mink shows that scientists studying animal welfare from a natural-scientific angle already take this view, albeit implicitly. The method developed above can make the interpretation of animal behaviour a matter on which ethologists and other animal scientists can attain an argumentative consensus.Although the concept of animal welfare is about what matters to the animal, the animal's point of view should be abandoned as the criterion of knowledge of animal experiences of welfare. Animals are not able to agree upon interpretations of their behaviour. Hence, these interpretations always remain human interpretations.The species-specificity of animal behaviour cannot function as the criterion of animal welfare. Whether a particular animal behaviour is species-specific or not depends on one's interpretations of various behaviours of animals of a certain species.Interpretations of animal behaviour as indications of good or poor animal welfare should not be in contradiction with natural-scientific research data that also refer to good or poor welfare. They should be brought into coherence with each other. Natural science studies the physical aspect and interpretative science the expressive aspect. These two types of research are not independent of each other; they both rely upon background knowledge regarding what counts as good or poor animal welfare. In some cases the results of both types of research can contradict and challenge each other. I argue that one cannot say beforehand which of these views is correct; this can only be decided upon by trying to attain a new consensus about interpretative and natural-scientific assessments of the animals' welfare. (shrink)