Aphasia is a significant cause of disability and reduced quality of life. Two speech pathology treatment approaches appear efficacious: multimodal and constraint-induced aphasia therapies. In constraint-induced therapies, non-verbal actions (e.g., gesture, drawing) are believed to interfere with treatment and patients are therefore constrained to speech. In contrast, multimodal therapies employ non-verbal modalities to cue word retrieval. Given the clinical and theoretical implications, a comparison of these two divergent treatments was pursued. This systematic review investigated both approaches in chronic aphasia at the levels of impairment, participation and quality of life. After a systematic search, the level of evidence and methodological quality were rated. Meta-analysis was conducted on 14 single case experimental designs using Tau-U, while heterogeneity in the four group designs precluded meta-analysis. Results showed that high-quality research was limited; however, findings were broadly positive for both approaches with neither being judged as clearly superior. Most studies examined impairment-based outcomes without considering participation or quality of life. The application and definition of constraint varied significantly between studies. Both constraint and multimodal therapies are promising for chronic post-stroke aphasia, but there is a need for larger, more rigorously conducted studies. The interpretation of "constraint" also requires clearer reporting.