Document Type


Publication Date



Many questions about animal welfare involve the affective states of animals (pain, fear, distress) and people look to science to clarify these issues as a basis for practices, policies and standards. However, the science of the mid twentieth century tended to be silent on matters of animal affect for both philosophical and methodological reasons. Philosophically, under the influence of Positivism many scientists considered that the affective states of animals fall outside the scope of science. Certain methodological features of the research also favoured explanations that did not involve affect. The features included the tendency to rely on abstract, quantitative measures rather than description, to use controlled experiments more than naturalistic observation, and to focus on measures of central tendency (means, medians) rather than individual differences. Much animal welfare science has dropped the philosophical stance but retained most of the methodological features. Thus, animal welfare scientists attempt to understand affect through quantitative measures, often in controlled experiments, with relatively little focus on individual differences. An alternative paradigm, seen in the work of Jane Goodall, Barbara Smuts and others, made a fundamental departure from these methodological features. These scientists collected qualitative, narrative data as well as quantitative; they described complex behaviour rather than measuring selected abstract features; and they attempted to understand the unique features of individual animals rather than averages for a species or type. Data produced by this alternative paradigm almost require scientists to involve affect in order to achieve plausible explanations of behaviour. Suitably developed, the alternative paradigm may provide a useful tool for fundamental studies relevant to animal affect and animal welfare.


In compliance with the publisher’s copyright and archiving policies, this is a post-print version of the document. Post-print materials contain the same content as their final edited versions, but are not formatted according to the layout of the published book or journal.