The assessment of the health and welfare of farm animals is steadily gaining importance, not only from a human health perspective, but also for the sake of the animals themselves. Consumers are increasingly concerned with the effect of industrial farming practices on animal well-being, and so there is a growing need to develop scientifically validated methods which address the welfare experience of animals on farms. Qualitative Behaviour Assessment (QBA) is a novel method developed at Scottish Agricultural College (SAC), designed to make an animal’s welfare experience accessible through a ‘whole animal’ approach. From a ‘whole animal’ perspective, what animals feel is an integral part of what they do, and so their experience speaks to us through their behaviour, as an expressive ‘body language’. By integrating details of an animal’s posture, behaviour and of the context in which these occur, we can assess that body language, and judge whether the animal is calm, anxious, enthusiastic or agitated, whether it is generally content or in distress. The question is whether such judgements are reliable, and can be a legitimate part of scientific welfare assessment. Traditionally scientists are taught to be distrustful of applying terms such as ‘content’ and ‘distressed’ to animals, regarding use of such terms as ‘subjective’ or ‘anthropomorphic’. However, several years of Defra/SEERAD funded research at SAC have shown qualitative judgements of behaviour to be unfailingly reliable, repeatable, and well-correlated to quantitative behavioural and physiological welfare indicators, both for individual animals and for animals kept in groups (eg Wemelsfelder et al., 2000, 2001; Wemelsfelder & Lawrence, 2001; Rousing & Wemelsfelder, 2006; Wemelsfelder, 2007). Thus, when facilitated by an appropriate methodology, qualitative judgements of behaviour can and do attain scientific validity.