The widely endorsed practice of activating relevant knowledge 1 prior to reading is challenged in this paper. It is argued that the endorsement is derived from a god's eye point of view that fails to acknowledge that relevance can be determined only by readers with respect to and in the course of making their own interpretations. To sustain this conclusion, the paper first outlines a perspectival (perspective-relative) view of reading that includes theoretical positions on the relation of reading to inferring meaning, on the importance of understanding firstperson intentionality, on the intelligibility of the concept of text information, on the distinction between literal and inferential interpretations, on what it means for readers to integrate their knowledge with text information, and on why the concept of interpretation is more suitable than the concept of comprehension for theorizing about reading. This perspectival view of reading provides a means to make coherent sense of interpretations, at one and the same time, being justifiably relative to readers' different beliefs and purposes for reading, and also con-1. We are using the term "knowledge" as it is used in the reading field. According to this usage, some pieces of a reader's knowledge may be false. On certain philosophical accounts of knowledge, in order for a proposition to form a part of someone's knowledge, it must be true. If a reader holds to a proposition that is false, the proposition would be considered part of the reader's system of beliefs, but not part of the reader's knowledge. That is, within philosophy, to ascribe knowledge to someone is to make an epistemic judgment. Either "knowledge" as used in the reading field, or "belief as used in the philosophy field captures all that readers rely on. However, in failing to make the distinction as made in philosophy, the terminology used in the reading field creates the awkward position of having to distinguish between reader's knowledge that is true and reader's knowledge that is false, which can be counterintuitive. Fenstermacher (1994) has discussed this issue in greater detail.
392Journal of Reading Behavior strained by universal interpretive standards of adequacy. The paper thence proceeds to show that, as a consequence of the nature of inference, the relevance of a reader's knowledge to a text interpretation is founded on a relation that is created as part of the interpretative act, and is not something to be sought in its own right. The paper concludes by arguing that having specific knowledge is not the main desideratum in interpreting texts. Rather, the main desideratum is using effectively the knowledge one has.Much has been learned over the last 30 years about the role of readers' knowledge in reading. In particular, theories developed within that time have examined more closely the connection between readers' knowledge and such factors as the social and political contexts of reading, readers' intentionality, and the construction of multiple meanings for the same text. In some...