We analyze the comprehension data in Broca's aphasia, pooled together by Berndt, Mitchum, and Haendinges (1996). We show that once analyzed properly, these data have statistical structure that is very similar to that revealed by the analysis in Grodzinsky, Pinango, Zurif, and Drai (1999). The suggestion that the latter authors doctored the data to obtain a desired outcome is as false as the claim that the data in Berndt et al. show no regularity. Comprehension scores in Broca's aphasia do have statistical structure, which correlates with syntactic structure. Thus, the role of Broca's area and its vicinity in language processing can be made more precise.