Reading Skill and Language Skill

To learn to read is to acquire a visual language skill that systematically maps onto extant spoken language skills. Some children perform this task quite adeptly, while others encounter much difficulty, and it has become a question of both scientific and practical merit to ask why there exists such a range of success in learning to read. Obviously, learning to read places a complex burden on many emerging capacities, and in principle, at least, reading disability could arise at any level from general cognition to visual perception. Yet since reading is parasitic on spoken language, the possibility also exists that reading disability is derived from some subtle difficulty in the language domain. In this article, my intent is to review some of the many studies that have explored the association between reading skill and spoken language skill. These reveal that when certain spoken-language skills of good and poor beginning readers are critically examined, considerably many, though perhaps not all, poor readers prove to possess subtle deficiencies that correlate with their problems in learning to read. LINGUISTIC SHORT-TERM MEMORY DIFFERENTIATES GOOD AND POOR BEGINNING READERS One of the more compelling reasons to view reading deficiency as the derivati ve of a language deficiency is that success at learning to read is associated with the adequacy of certain linguistic short-term memory skills. In our work at Haskins Laboratories, my colleagues and I have found clear indications of this association in a variety of different studies of good and poor beginning readers. For the moment, however, let me put aside a discussion of those studies in order to consider first the short-term storage requirements of normal language processing, and to summarize some recent findings as to how these requirements are met by the mature language user. These considerations pertain to both written and spoken language and provide a +Also Bryn MavIr College. Acknowledgment. Much of the work described in this paper, along with preparation of the manuscript itself, was supported by NICHD Grant HD-01994 and BRS Grant RR-05596 to Haskins Laboratories. I am grateful to the many members of the reading research group at Haskins, particularly to Isabelle Liberman and Donald Shankweiler for their comments on earlier drafts of this paper, and to Alvin Liberman and Ignatius Mattingly for their insights into the psychology of language. [HASKINS LABORATORIES: Status Report on Speech Research SR-69 (1982)]