Cues People Use to Paragraph Text.

This paper reports the results of three studies on the paragraph. In Study 1, subjects were asked to paragraph a text from which paragraph indentations had been removed. Results indicate that readers can consistently paragraph unparagraphed text, thus supporting Young and Becker's (1966) assertion that the paragraph is a psychologically real unit of discourse. Results also reveal that readers rely heavily on breaks in text cohesion (e.g., topic shift) and on paragraph length as paragraphing cues. In Study 2, four new subjects were asked to paragraph the same text used in Study 1, and to "think aloud," giving their reasons for paragraphing, as they did so. Analysis of these thinking aloud protocols reveal that, in the absence of a strong paragraphing cue, readers will read ahead in a text, sometimes flagging weaker paragraphing cues as they go. If they feel the unparagraphed text is too long, they will go back and paragraph at these weaker cues until all paragraphs in the text are an acceptable length. Based on the results of Studies 1 and 2, a model of how readers paragraph was devised. The model was tested in Study 3 on new subjects who were asked to think aloud as they paragraphed the same text used in Study 1 , and another, longer text. The model predicted the new data quite accurately. Deciding whether paragraph boundaries are psychologically real or arbitrary is very much like deciding whether geographical boundaries are psychologically real or arbitrary. Typically, state boundaries are not psychologically real because travelers cannot find them without the help of signs. Coast lines, on the other hand, are very real. People who miss them fall into the ocean. In the same way, we would consider paragraph boundaries artificial if people could find them only with the help of paragraphing marks, and real if people could consistently find them in texts from which paragraphing marks had been removed. Some linguists such as Hodges (1941) have viewed the paragraph as an arbitrary device used by the writer to "give the reader a breathing spell" (p. 311). Similarly, Rodgers (1967) has suggested that a section of text "becomes a paragraph not by virtue of its structure, but because the writer elects to indent" (p. 182). However, Young and Becker (1966) and Koen, Becker, and Young (1969) have provided strong evidence that paragraphs are indeed psychologically real. They asked readers to paragraph text from which all paragraphing markers had been removed and found that their readers Research in the Teaching of English, Vol. 18, No. 2, May 1984