In this paper, in line with Carstens, Hornstein, and Seely (2013), I advocate the traditional concept of head movement that can be maintained in terms of category and relevant features of the moving item and the target, contra Chomsky (2013) and Cecchetto and Donati (2010). This is made possible by incorporating the notion label into the definition of extension and command relation. The major results of the discussion are: A moved head can bind its trace in terms of label-command, head movement may not violate the Extension Condition under the revised extension defined in terms of label, locality can be measured in terms of label-crossing. Thus, there remains little anomaly with head movement.