Our contribution to this special issue on reading theory questions the effectiveness of the prevailing interventions intended to improve word-reading and reading comprehension skills in children with reading disability (RD). Our hypothesis is that we as a field may have inadvertently diluted reading theory in ways that compromise the power of intervention programs. For both word reading and reading comprehension we argue that current intervention programs target instruction at a knowledge level below that which is necessary to foster reading skill development that is "generative" in children with RD. Further, we contend that current interventions for children with RD fail to mimic and promote the inductive learning mechanisms that characterize typical reading development. Thus, we return to reading theory in an attempt to identify ways that current interventions may be reconceptualized to treat word-reading and reading comprehension deficits. In doing so, we call for the development of a new generation of reading interventions that target the fundamental knowledge structures and learning mechanisms known to support typical reading development.