In this case study of an aphasic patient with difficulties in numerical processing, the patient responded to a series of linguistic and numerical tasks designed to assess efficiency levels in processing various linguistic components. In addition, the patient completed a series of transcoding tasks that were directed at isolating whether the problems were associated primarily with arabic numerals or with other modalities (spoken or written). Data were analyzed using chi-square goodness-of-fit tests. Statistically significant differences were obtained between spoken verbal and written verbal outputs and between arabic and spoken verbal outputs. Based upon an analysis of errors, it was tentatively concluded that the disorders were associated with two types of dissociation operating together, one between spoken verbal and written verbal outputs at the syntactical level and the other between lexical and syntactical components in the spoken verbal output. A revised model is proposed to provide a tentative explanation for these observations.