Fulvio Mazzocchi

This paper considers how hermeneutics and other related theories may bring new insights into KO. They provide a most realistic representation of the complexity of knowledge and meaning according to which new forms of KOSs could be designed. Computational and conceptual aspects of these issues are discussed taking into account a number of case studies. Introduction The development of knowledge organization (KO) has strongly been influenced by empiricism and logical positivism. However, an increasing attention to other theoretical approaches, and in particular to hermeneutics, can be found in many theoretical studies in KO and related fields (Chalmers, 2004; Cole & Avison, 2007; Fonseca & Martin, 2005; Hansson 2004; Hjørland, 2007). With its basic idea of the ‘situatedness’ of understanding, contemporary hermeneutics has the potential to become a theoretical background on the basis of which new forms of knowledge organization systems (KOSs) are conceived. Wittgenstein’s late philosophy, which shares some similarities with parts of hermeneutics, is also considered for such a purpose. The possibility of applying these approaches to KO issues is then investigated, both in their computational and conceptual aspects, by analyzing some case studies. 1. Towards hermeneutic approaches in KO: computational applications 1.1 The heuristic approach for the development of IS methodology Information systems (IS) using thesauri, classifications and other KOSs adopt from their beginnings a heuristic model of reasoning deeply influenced by logical positivism. This includes, for example, attempts to codify human reality into formal definitions. Accordingly, representations of phenomena are fixed and based on logics and exclude representations of individuals’ context and situation (Chalmers, 2004). The concept of heuristics refers to a reasoning that advances learning in basic principles to progress in knowledge through successive stages, as to enable the accumulation of successes and learning from errors. The reasoning counts on fundamental concepts that enable coding the well-known. This codification provides the possibility of extending, strengthening, or revealing what is known, thus creating the foundations of new knowledge in relation to the already known. The development of IS with a heuristic model follow two knowledge representation methodologies: the Procedural form and the Declarative form. In the former, knowledge is integrated into the computer program. The knowledge representation is independent from the computational process: the computer program is seen as a collection of functions or instructions given to the computer. In the latter, a computer program is as a set of interacting individual units or objects, where any object can manage its own state and operations. Both methodologies, although implying different implementation systems, have three logical steps in common: