The philosophical and empirical study reported here developed from the observation that information science has had no comprehensive understanding of nonuse of information. Without such an understanding, information workers may use the words "nonuse of information" while referring to very different phenomena. This lack of understanding makes the job of the information professional difficult. For example, the model presented here reduces hundreds of theories of information behavior to a conceptually manageable taxonomy of six conditions that lead to nonuse of information. The six conditions include: 1) intrinsic somatic conditions, 2) socio-environmental barriers, 3) authoritarian controls, 4) threshold knowledge shortfall, 5) attention shortfall, and 6) information filtering. This dissertation explains and provides examples of each condition. The study of a novel area that had no prior theory or model required a novel methodology. Thus, for this study, I adopted the pragmatism formulated by Charles Sanders Peirce, a method of evaluating concepts by their practical consequences. This pragmatism applied in two ways to the study of nonuse of information. First, because nonuse of information is a behavior, pragmatism helped me to limit the psychologic implications of the study to behavior, rather than to expand the discussion to psychodynamics or cognition, for example. I justified this limiting on the basis that behavior reflects the use or nonuse of information, and behavior is more observable than other aspects of psychology, such as cognition. Second, Peirce's concept of pragmatism supported another of his contributions to philosophical inquiry, retroduction, sometimes referred to as abduction. To study nonuse of information through retroduction, I created a fivestep "definition heuristic," based on the writings of Spradley and McCurdy. I then created a nine-step "retroduction heuristic" based on the system of logic identified and termed "retroductive" or "abductive" by Peirce. I used this heuristic to identify examples of nonuse of information and applied the examples to a second corpus of research reports that contained examples of compelled nonuse of information. The taxonomy of this study resulted from this second application and represents a descriptive model of compelled nonuse of information. / text