The paper introduces four types of models of translation:
1. Comparative model, commonly used in contrastive research, and it aligns translations with source texts of parallel text to explore correlations between them.
2. Process model, represented by communication approaches, and it maps different phases of the translation process over time.
3. Causal model, where translations are seen as both affected by precursor conditions and as inducing effects on readers and cultures.
The paper also provides a discussion of for types of hypotheses traditionally used to understand the phenomenon of translation: interpretive, descriptive, explanatory and predictive.
According to Chesterman, only a causal model enables us to generate all four types of hypotheses, therefore, we should 'explicitly seek to develop in translation research'. Furthermore, a causal model can provide a comprehensive empirical research programme for translation studies, and operate as a basis on which to construct a translation theory or theories. Another practical consequence of research based on a causal model would be its applicability to translator training and quality assurance. If we can illustrate particular ties between causal conditions, translation profile features, and observed effects, this might lead to a greater perception of how to produce translations that have more desired effects and less undesired ones.
This paper is a required reading for the Ph.D. in Translation Studies candidacy exam (Comps) at the Institute for Applied Linguistics at Kent State University. This Xmind map (you can view and download by clicking the Xmind icon) provides a summary of the main concepts presented in this paper:
Chesterman, A. (2000). A causal model for translation studies. Intercultural faultlines, 15-27.