This case study analyzed data from an asynchronous written mediated exchange between a native speaker and nonnative speaker of English. Three methods of analysis-error analysis, propositional complexity (idea units), and qualitative analysis were used to analyze learner performance over time. The results from these methods were visualized and interpreted according from Complex Dynamic Systems Theory (CDST) perspective, which allows longitudinal language production data to be interpreted from multiple perspectives so that change can be understood at both the micro and macro levels. Results from the analysis revealed patterns of linguistic change both within and between participants as they interacted over time. Several instances of continuity in language production were observed between the participants, including a drop in error rate that coincided with an increase in minor idea units during one turn of the exchange. Results from a qualitative analysis offer several explanations for this finding and are related to the implications this holds for CDST.