Morgan's canon is a very widely endorsed methodological principle in animal psychology, believed to be vital for a rigorous, scientific approach to the study of animal cognition. In contrast I argue that Morgan's canon is unjustified, pernicious and unnecessary. I identify two main versions of the canon and show that they both suffer from very serious problems. I then suggest an alternative methodological principle that captures all of the genuine methodological benefits that Morgan's canon can bring but suffers from none of its problems.
Available at: http://works.bepress.com/simon_fitzpatrick/2/