| For students of English as a second language (ESL) learning to write expository essays, a major stumbling block is often the use of cohesion, the set of grammatical and lexical devices signaling the meaning of relationships and the links between individual segments of a text. The objective of this study was to investigate the ways in which ESL students in a pre-freshman composition course use cohesive devices to link segments of text which function significantly in the development of their expository essays, and to specify the problems they still have in the use of these devices.; For segmenting the essays, a new analytic unit was developed: the functional unit of discourse, or F-unit, comprising the set of clauses and clause equivalents serving an identifiable rhetorical function in written discourse. Objective linguistic criteria used to characterize the F-units differentiate them from similar structures providing restrictive modification or filling an essential syntactic slot in a matrix clause, which do not qualify as F-units. For identification of the cohesive devices, the outline of cohesion in English presented by Halliday and Hasan (1976) was modified to include all devices providing ties between F-units, including items which function intra-sententially. Five major categories of devices, each with several subcategories, were used: reference, substitution, ellipsis, conjunction, and lexical cohesion.; Five sets of essays, written during a 14-week semester, were analyzed. The essays were segmented into F-units; the cohesive devices, together with their referents appearing in the text, were then identified. In order to check the suitability of any conjunctive devices, the functional role played by each F-unit was identified. Finally, the grammatical and semantic errors in cohesion were identified and grouped according to error type. This error analysis specified the major problems in the use of cohesive devices and provided a tentative identification of some causes for the errors.; Findings of the Study. Although the compositions were written over a 14-week period, there was little change in the relative degree to which devices from the different categories and subcategories were used to achieve overall cohesion. The students relied most heavily on lexical devices. They also used a considerable number of reference items--particularly pronominals and demonstratives--and, to a lesser extent, conjunctions. Ellipsis and particularly substitution served only a minor role in cohesion.; Errors were most common in all types of reference devices and in adverbial conjunctions, although there were also errors in the other types of devices. The most frequent type of error was the use of a device with no identifiable referent in the text. Other major error types included the substitution of an inappropriate item for a required cohesive device, problems in number agreement, and the use of devices with ambiguous referents.; Certain patterns in the use of the devices and in the cohesive errors can be explained by differences in the time available for writing and/or any prior preparation for the essays, transfer of or interference from native-language forms, overgeneralization of the use of familiar items, and the influence of teaching procedures and the presentation of material in the students' text.; In the analysis, the F-unit proved to be a flexible research tool for objectively isolating the rhetorically significant units in the essays and for dealing with the occasionally nonstandard sentence and clause structures found in ESL writing. |
