New areas of endeavour succeed either because they make a compelling case for their worth within an existing paradigm or because a new generation is raised for whom a once heretical paradigm is now the norm. It is difficult to tell which route will lead to the acceptance and growth of veterinary behavioural medicine, but the latter seems likely and, perhaps, necessary. As one of the newest specialities in veterinary medicine, behavioural medicine may suffer more than other specialities from an unclear identity because so many disparate groups who are not rooted in veterinary medicine have participated in its evolution. There is also the issue of data. Behavioural medicine is often cited as a field for which there are very few hard data; however, it is also a field for which data collection can be difficult and time consuming for non-trivial questions. This is because behaviour is the ultimate integrator of all organ system responses, and as such, is a dynamic outcome resulting from the interactions of complex mechanisms. Understanding such systems is difficult, but progress can occur if an attempt is made to understand all the mechanistic levels that contribute to behavioural patterns and behavioural conditions. For this to happen we need a paradigm shift that moves us away from the medical paradigmatic model and towards a new paradigm that is based on hypothesis formulation and testing between interacting levels of mechanism. Clinical impression and expertise based on outcome must be replaced by a scientific method that provides for phenotypic definitions that are coupled to putative underlying, interacting mechanisms that can be evaluated. Only in this manner can phenotypic description and its clinical application in veterinary behavioural medicine keep pace with advances in molecular biology and genetic epidemiology that deal with genetic liabilities and vulnerabilities inherent in multi-gene and major gene effects. This paradigmatic shift has the potential to revolutionize the way we view diagnoses in medicine, in general, but has particular implications for the care needed to characterize problematic and truly abnormal behavioural conditions. This approach also has the advantage of alerting us to when we do not know something, and forcing us to consider the needs of our patients on their own terms. A more rigorous scientific approach in this field could also go a long way in making us more humane.