Purpose: Adults with language-based disabilities are known to have deficits in oral language; however, less is known about their written language skills. Two studies were designed to characterize the writing of adults with language-based disabilities. Method: In Study 1, 60 adults, 30 with language impairment and 30 with typical language, completed written narratives. Forty-one written language measures were analyzed. In Study 2, the measures that had the most potential for reliably indexing deficits were analyzed in an additional 77 adults. Results: Three measures that showed significant between-group differences and had robust effect sizes in Study 1, total number of verbs, 1-part verbs, and errors, were applied to the samples in Study 2. A group difference for percentage of errors was replicated in the second sample. A discriminant analysis identified 75% of the adults with language impairment and 30% of the adults with typical language as having an impairment based on the percent of written errors. Conclusions: The writing task revealed consistent group differences in written errors and is clinically applicable in describing a client's writing. However, the number of written errors was not robust enough to identify whether an adult had a language impairment or not.