Comprehending implicit meanings in text without making inferences

There are two thin lines that separate descriptions of text comprehension. The first is the line between what a text says (i.e., its explicit or literal meaning) and what is inferable from the text (i.e., its implicit meaning). The second thin line is one that separates two kinds of implicit meaning processes. On one side of this line is what text researchers refer to as inferences. Although these inferences can be further sorted into various types (e.g., Graesser, Singer, and Trabasso, 1994), they have in common the idea that the meaning that is obtained is not in the words and syntax of the text but is constructed by the reader based on knowledge that is largely independent of that particular text. On the other side of the line are implicit meaning processes that are closely bound to the language of the text, the meanings of words, and the grammar of the language. These lines are “thin” because they divide meaning processes only approximately and are subject to encroachment in difficult cases. There is good company for such thin lines in closely related distinctions in linguistic descriptions, especially semantics versus pragmatics and meaning versus interpretation. In text comprehension research, attention has been diffusely distributed across both lines, treating inferences as a set of differentiated types that contrast with literal meaning. Indeed, analyses of the variety of inferences have led to taxonomies that differentiate among nine different types (Pressley and Afflerbach, 1995) and even thirteen or fourteen types (Graesser et al., 1994). These multiple varieties cross various dimensions of mixed types, some reflecting linguistic devices in the text (thematic role assignment, referential anaphora), but most lacking linguistic sources, dependent on reader knowledge (e.g., character emotion, causal consequence) or coherence strategies (e.g., causal coherence). Taxonomies are of limited value when they are not well defined according to some classification hierarchy, however. A mature taxonomy, for example, a biological taxonomy, serves long-lasting functions (in between modification) because it is organized hierarchically around structural or functional principles (e.g., biological