Faculdade de Letras da Universidade do Porto - OCS, 15th INTERNATIONAL ISKO CONFERENCE

Font Size: 
Gregory H. Leazer, Robert D. Montoya, Jonathan Furner

Last modified: 2018-02-21


“Knowledge organization system” (KOS) and related terms (such as “classification,” “taxonomy,” “ontology,” and “vocabulary”) pick out members of a family of concepts whose referents share characteristics of several kinds. The referents vary from Internet search engines like Google, proposed systems like the Semantic Web, international systems for bibliographic control, and systems like Facebook which serve as platforms for personal interaction and as sources of shared reading.  KOSs have similar functions: they are created by similar kinds of agents, for similar kinds of reasons, using similar kinds of methods.  Increasingly characterized as a “knowledge infrastructure” that mediates access to—by various formulations—information, knowledge or documents.  They comprise and are constituted by various kinds of standards, operating in multiple kinds of inter-institutional practice, from local practices to highly integrated global arrangements.

The last century of development of classifications, and of KO systems more generally, has proceeded with two major assumptions:

--KO systems should be modeled after scientific knowledge, using scientific processes, attentive to expert consensus and evidentiary practices to develop their conceptual content and structural arrangement.  We call this the theory of scientific classification.

--Advances in KO system research is due to a functional experimental program that has relied upon technical advances.  These techniques include the development and use of controlled languages, computer-based search and retrieval, and more recently on increasingly sophisticated statistical approaches.  We call this the theory of technological progress.

In this paper we explore the limits of both of these theories.

Firstly, the theory of technological progress had H. E. Bliss as one of its most significant early advocates, particularly with his book of 1929, Organization of Knowledge and the System of the Sciences.  However, later theorists have emphasized the lack of consensus in scientific exercise, such as Kuhn’s The Structure of Scientific Revolutions (1970) is an example of the myth of consensus; Hull (1988) and Gould (1981) demonstrate the difficulties and mismatches between human categorical modeling of the natural world and scientific taxonomy.  The scientific approach is idealized, and cannot avoid the kinds of distinctions that are rooted in cultural systems.

Secondly, the field of KO system development is most often conceptualized as a technological endeavor, has shifted from a field where documents were represented by metadata gathered through manual processes and browsed in relatively simple manual systems to a field where documents are represented by automatically generated metadata and searched in computational systems.  Key to this transformation has been the implementation of the computational algorithm and increasingly sophisticated statistical approaches.

This technological change has occurred in a period of rapid social change: not just technological, but also social.  Globalization has further imperiled indigenous cultures and languages; increased social stratification; and digitization has not only multiplied documentary genres and formats but also enabled concentration of media ownership.  These trends, along with increasingly onerous copyright and intellectual property regimes, have restricted fair and open access to knowledge resources.  KOS, once the exclusive purview of universities, libraries, archives, museums and learned societies, with some commercial support, now is significantly shaped by major commercial players like Google, Thomson Reuters, and Reed Elsevier.

However, in a period of rapid technological and social change, key assumptions regarding the nature of the human mind and of social/cultural relations, of classificatory structures and of linguistic phenomena remain relatively under-investigated.  A lack of understanding the social and cultural practices that underpin KO and human interactions with information and KOS has led to undesired consequences in the design, development and use of KO systems.  KO systems, by failing to account for various kinds of social difference and variation, incorporate social bias, and reinforce it.

Of increasing significance, then, is that KOS itself must become the locus of critical reading and literacy.  Users must not only develop an information literacy, defined as “the ability to locate, evaluate, and use effectively the needed information” (American Library Association, 1989) but develop a critical capacity for understanding how the various systems of knowledge organization permit, limit or shape their means of access to information.  Common conceptualizations of critical literacy serve as a point of departure for developing a notion of critical information literacy, especially as it pertains to KOS.  Critical literacy, as derived from Freire (1970), emphasizes the reader’s ability to understand and analyze a text’s ideological meanings and its role in various discursive formations, and that the reader and the text are both historically constructed.  We seek a similar understanding not just of texts or of information, but of the methods of access to texts and information as well, and that the information seeker is an agent within a larger cultural milieu.

The consequence of enabling a more critical engagement with KOSs entails exposing their functions, design assumptions and biases to critical use.  How to accomplish this task is an open and relatively unexplored research question, but has received some attention, for example, in Drabinski 2013.  More positively, we find substantial recent interest in basic rational questions within classification, including the principal nature of classificatory structures, and their historic, linguistic and cultural bases.

Works Cited

Bliss, Henry Evelyn. The organization of knowledge and the system of the sciences. 1929.

Gould, Stephen Jay. "What, if anything, is a zebra." Natural History 90.7: 6 1981.

Hull, David L. Science as a process: an evolutionary account of the social and conceptual development of science. University of Chicago Press, 1988.

Kuhn, Thomas S. The structure of scientific revolutions, 2nd enl. ed. University of Chicago Press, 1970.