Abstract

The dissertation considers the concept of hierarchy and the ways that it has been used and misused in recent literature. It appraises the work of Herbert Simon and his claims for the primacy of hierarchy in the organization of systems. It also discusses the ways in which the work of Bertrand Russell, specifically the theory of types, and Alfred Tarski, specifically the object language--metalanguage distinction, have been misunderstood as being relevant to questions of systems organization. Claims of hierarchical organization in computer systems are also considered. In addition, it is pointed out that systems can be conceptualized in different ways and that although some of those conceptualizations are hierarchies, others are not. Different definitions of "hierarchy" that are currently used are discussed. Finally, several non-hierarchical complex systems are considered. The dissertation concludes that Russell and Tarski are relevant to discussions of hierarchy only in the most general way--namely that the ontology of Logical Atomism supports both hierarchical organizational forms and the systems developed by Russell and Tarski. It also points out that an a priori commitment to hierarchical organization may result in a failure to recognize other, heuristically valuable organizational models.