ongoing research but to participate in an intense, 5-day dialogue about
the research in reading acquisition and the implications of this work for
early intervention. Researchers were asked to address not only what they
have learned from their research programs, but also to discuss unsolved
problems. This dialogue prompted numerous questions of both a theo-
retical and applied nature, generated heated debate, and fueled our op-
timism about the important gains that have been made in our scientific
understanding of the reading process, especially our understanding of
the critical role played by phonological abilities. The contributions to this
volume are sometimes different from the original presentations, in part
because they reflect the exchange that took place among conference par-
ticipants.

This book is divided into the following four sections: Theoretical Foun-
dations, Subtypes of Dyslexia, Beginning to Read and Spell, and Impli-
cations for Intervention. It should be noted that the separation between
sections is somewhat artificial, and several of the chapters could have
been presented in more than one of the sections

The first section of the book includes four chapters that explore theoretical
foundations and begins as did the conference, with a presentation by
Alvin Liberman, based on his eloquent keynote address. Liberman first
describes the relevance of a theory of speech for understanding reading
acquisition, and then explores the implications of opposing views of the
biology of speech for understanding what makes reading and writing
hard to acquire. As Liberman explains, the constituents of speech are
articulatory gestures (not sounds) managed by a biologically specialized
phonologic module that coarticulates the gestures and makes it possible
for the speaker-hearer to produce and perceive speech at a rate that would
be impossible otherwise. These gestures evolved with language to serve
only a phonetic purpose--no other. Unlike the visual percepts evoked by
the letters of the alphabet, articulatory gestures are phonetic by their very
nature, requiring no cognitive translation to make them so. But, as Liber-
man told his audience at the conference, a consequence of coarticulation
is that

Awareness of phonologic structure [needed to understand how print rep-
resents speech] is hard to come by: The module spells for the speaker and
parses for the listener, leaving both in the dark about phonologic process;
coarticulation destroys all correspondence in segmentation between acoustic
and phonologic structures, making most consonantal segments unavailable

-xii-