The acquisitional sequence Strategy C-Strategy D would then replicate stages in the early development of language. Strategy C would have had to be developed in order to interpret case roles in the stage in which topic-comment ordering was dominant. Strategy D would have succeeded it as soon as sentence-order stabilized and became the primary marker of case relations. Note that, originally, adoption of Strategy D would have had none of the dysfunctional side effects that it does nowadays with children acquiring English since, at the time, there were no passives and no clefts; Strategy D would have given the right answer every time.
- Bickerton (1981), a pag.273-274

Experiments carried out by Bever and his associates indicate that children between two and three rely on Strategy C to comprehend sentences but that a little later they switch to Strategy D. This serves to explain the otherwise quite baffling fact that children’s performance with regard to sentence types which involve nonagentive initial NPs (passive, clefts) actually deteriorates rather than improves between three and four.
- Bickerton (1981), a pag.273