Human Resource Management (HRM), as a sub-discipline of management science, is in its infancy. HRM practices are often Utopian in expectation and fail to incorporate a realistic view of existing knowledge bases in the psychological, social, and biological sciences. The HRM discipline relies upon theoretical approaches (eg theories of motivation, satisfaction, and performance) which are: (1) almost invariably linear in conceptualisation and depend largely upon correlational evidence, (2) frequently validated within nonrepresentative contexts that are overly constrained by researchers and (3) overly simplistic in that the constraints and patterns imposed by our biological, psychological and social systems are frequently ignored or assumed to constitute random error within the models. This frequently translates into HRM practices which map reasonably well onto theory yet fall short of yielding expected outcomes. The theories do not match the realities observed. We point to nonlinear dynamics and chaos theory as a way of conceptualising how common HRM practices may translate into observable outcomes. Such an approach will force managers to pull back from simple reliance on linear predictions and realise that truly effective HRM practices should be sensitive to the unique, complex and less systematically predictable patterns of human behaviour.