ABSTRACT
Theory is important. A good theory, so I was told at the University of Minnesota, is one that synthesizes the past, explains the present, and predicts the future. In other words, a good theory accounts for previous and present findings from data collections and predicts or projects what future findings will be. Good theories are explanatory and hypothesis-generating. They are also efficient. Extensive numbers of parameters and dependencies indicate that the theory either is not stated sufficiently directly or is not nuanced adequately to enable it to account for the majority of observations of whatever phenomenon. There are many theories of first-language reading. Grabe (2009) and Hudson (2007) rehearse most of these L1 theories in their volumes. Yet, these volumes never draw direct links between the assumptions made in the theories, the growing second-language reading data base, and second-language reading theory development. In order to make progress in the field, rehearsal of old theories and recapitulations of studies without reference to a theory render little service. This chapter, in contrast to other writings on L2 reading, poses questions about whether the assumptions made in certain theories actually fit or describe the second-language reading process and/or whether a theory based in a first-language process can ever adequately capture the second. Interestingly, Hudson fails to refer to any theory of second-language reading while Grabe dismisses what exists: “In L2 reading only one general descriptive model of reading has been proposed (Bernhardt, 1991; 2000) and it is somewhat vague in its specifications of component abilities and implications for reading development” (Grabe, 2009, p. 104). In contrast, Hedgcock and Ferris (2009) describe the situation differently by referring to the “complex factors” (p. 35) outlined in Bernhardt’s (2005) compensatory model.
