The traditional model of motor weakness derived from the works of Hughlings Jackson is that of a negative phenomenon following a cortical injury such as stroke, in contrast to the positive phenomena of brisk reflexes and the release of primitive, complex reflexes. The physiology of orderly motor neuron recruitment and de-recruitment elaborated by Henneman, in which the smallest, weakest, and slowest motor neurons are recruited first followed by increasingly large, powerful, and fast motor neurons, provided a physiological explanation for weakness following a cortical insult. This was especially so with the later work of Evarts who demonstrated that the frequency of firing of Betz cells (upper motor neurons) in the cerebral cortex specified the strength of isometric contractions. Maintaining a contraction is underpinned by pools of upper and lower motor neurons being recruited in rotation by relay as it were. Reduced populations and sizes of such upper motor neurons would result in poorly sustained firing of neurons and hence weakness. However, in cerebral palsy (CP), the injury is seldom a focal lesion in the motor cortex but more usually bilateral lesions affecting subcortical white matter. Moreover, the brains of children with CP are immature and still developing. The work of Sutherland demonstrating persistence of the wrap-around electromyogram discharges throughout the gait cycle, and subsequently of Leonard, support a model of ‘developmental co-contraction’ as the default muscle activation pattern out of which, in the uninjured state, refined, selective motor strategies emerge to provide fluency and elegance of movement. As Leonard and colleagues demonstrated, children with CP never shake off this co-contracting pattern which results in clumsy, poorly executed, and energy-intensive movements. According to the landmark study by Tedroff et al. (p 789) children who have a hemiplegia or diplegia frequently select the wrong muscles when attempting to perform a maximum voluntary contraction. This miss-selection often involves an antagonist or remote muscle from the intended joint of action. A consequence of these findings necessitates a revision of the essence of the motor disorders traditionally referred to as ‘spastic’.