The Effects of Overt and Covert

In the past twenty years, considerable research has examined various aspects of syntactic complexity. Indeed, our most widely used indices of language development are syntactic (Words/T-unit and Words/clause, see Hunt, 1965); and instruction in sentence-combining (SC) methods has produced consistent gains in syntactic complexity from pretest to posttest and even to delayed posttests (Mellon, 1969; O'Hare, 1973; Combs, 1976; Callaghan, 1977; Swan, 1977). Clearly, research has established SC activity as a useful tool for increasing the syntactic complexity of students' writing. In addition, the gains from SC practice positively influence teachers' judgments of quality of writing (O'Hare, 1973; Combs, 1976; Pedersen, 1977). While the effectiveness of SC has been established, there is reason to believe that the impact of instruction has not been adequately tested. In all of the SC studies, "normal" writing samples were elicited for pre- and post-tests. Conclusions about the effects of SC were based on these samples. However, in none of the studies were the students overtly told to write long, complex sentences. It was inferred that gains due to SC instruction could be determined from these "normal" samples. While this may be true, two possible faults seem readily apparent. First, all the studies assume that "normal" writing samples reflect what the students are capable of doing. No study examined whether students who were asked to write more complexly on the pretest would "naturally" use all of the structures which were included in the SC treatment. The research on language development indicates that post-elementary students know the structures of English. What distinguishes among them is the frequency of use and the recursive use within a single sentence. This leads directly to the second possible flaw. SC may not teach the use of the structures, but simply provide the students with a cue, a covert cue, that they should write longer, more complex sentences. When teachers spend extended time on SC, they underscore the value of syntactic complexity. Thus, students may not be learning how to construct longer, more complex sentences from SC practice, but rather that they should do so.