Although there is strong evidence that Broca's area is important for syntax, this may simply be a by-product of greater working memory and/or cognitive control demands for more complex syntactic structures. Here we report an experiment with event-related transcranial magnetic stimulation (TMS) to investigate whether Broca's area plays a causal role in morphosyntactic processing when both working memory and cognitive control demands are low. Participants were presented with word pairs that could either agree or disagree in grammatical number or gender while receiving stimulation to Broca's area or to the right intraparietal sulcus (a control site). Stimulation of Broca's area significantly reduced the advantage for grammatical relative to ungrammatical word pairs. In contrast, stimulation of control site left this grammaticality advantage unchanged. The interaction between grammaticality and stimulation was specific to Broca's area, suggesting a clear involvement of the region in morphosyntactic processing. (Contains 3 figures.)