From Conversations to Stories: An Analysis of Differences Between Oral and Written Language Comprehension

FROM CONVERSATIONS TO STORIES: AN ANALYSIS OF DIFFERENCES BETWEEN ORAL AND WRITTEN LANGUAGE COMPREHENSION

Ann D. Rubin Bertram Bruce Bolt Beranek and Newman Inc. 50 Moulton Street Cambridge, Mass. 02138

A traditional view of reading comprehension has been that it is the sum of oral comprehension skills and decoding skills; the educational corollary to this belief has been a heavy emphasis on the teaching of decoding. As part of our work at the Center for the Study of Reading, we are examining some important differences between oral and written language~--we hypothesize that, while many of children's highly-developed oral language skills serve them well in the context of reading comprehension, additional processes are necessary for their understanding of text. Children may even exhibit "bugs" which derive from a too-general transfer of oral language processes to the reading situation.

Some of the more obvious differences between oral and written language are the following: the presence of stress and other intonation features in speech, presence vs. absence of shared spatial and temporal context, the predominantly interactive mode of children's oral language, paragra~h structure in text, the permanence of written language and different syntactic and discourse structure of conversations and text. Much of our work to date has consisted of the development of a detailed taxonomy of the differences between the language experience a child is familiar with, e.g., conversations, and the new one (reading) he or she is attempting to master. We have divided the distinctions into medium and message-related differences---ana distingulshed-several dimensions in each of these categories. The result is a view of language experiences which designates children's conversations and stories as extremes along several dimensions. It also suggests many intermediate points which share some characteristics with both extremes. These intermediate points represent tasks requiring a partial transition from oral conversation skills to reading skills.

Two of the dimensions in this analysis are spatial and temporal commonality. Children's conversational experiences generally assume a shared spatial and temporal context. As a result of this consistency, children may develop comprehension (and production) strategies that take advantage of the fact that they and their listener are in the same place, at the same time and can see the same objects, as well as each other. Children's language experiences contain many instances of deictic expressions, terms whose interpretation depends upon the temporal and spatial context of the utterance as well as its speaker.

we hypothesize that children's reliance on shared context in their language experiences may be a source of trouble in reading comprehension. To this end, we are investigating the following two issues:

1. Do children have problems interpreting deictic terms in text? Notice that correctly understanding "the next day" in:

Esmerelda told King Arthur she would visit his castle the next day.

requires the child to create a time point that exists solely in terms of the text (the time of Esmerelda's comment to King Arthur) and interpret the deictic phrase in terms of that manufactured entity.

1. What devices in text perform the function of deictic phrases in speech? Do children have trouble with such mechanisms? The function of deictic expressions such as "this chair" or "that doggie" (often accompanied by pointing) is to designate a particular object which is in the shared context of speaker and listener. In text, the designation of objects is often accomplished by noun phrases ("The blue chair," "Jack's doggie"). However, the objects often do not actually exist; rather, they (or a description of them) must be constructed by the child without a physical referent.

we are investigating whether these of difficulty for children learning to

added levels of complexity may be sources _ read.