In three structural priming experiments, we investigated whether deaf and hearing writers differ in the processes and representations underlying written language production. Experiment 1 showed that deaf writers of Mandarin Chinese exhibited comparable extents of structural priming and comparable lexical boosts, suggesting that syntactic encoding in written language production is similarly sensitive to prior lexical-syntactic experience in deaf and hearing writers. Experiment 2 showed that, while hearing writers showed a boost in structural priming when the prime and the target had homographic or heterographic homophone dative verbs compared to unrelated ones, deaf writers showed a homophone boost only with homographic homophone verbs but not with heterographic homophone verbs. This finding suggests that while hearing people develop associated lemmas for homophones due to phonological identity, deaf people do so due to orthographic identity. Finally, Experiment 3 showed no boost in structural priming in deaf writers or hearing writers when the prime and the target had the same verb of the same orthography (i.e., in the same script) than of different orthographies (i.e., between Simplified and Traditional Chinese), suggesting that neither hearing nor deaf people use orthographic identity to reactivate the prime structure. In all, the findings suggest that syntactic encoding in writing employs the same syntactic and lexical representations in deaf and hearing writers, though lexical representations are shaped more by orthography than phonology in deaf writers.