This formulation of the function of language is interesting for Library and Information Science (LIS) in that it directly affects research on classification and the organization of knowledge. The meaning of terms is what constitutes the basis of ”subjects” and relations in knowledge organization systems. If we study the epistemological discussion in relation to bibliographic classification, for instance in the erudite writings of Elaine Svenonius, we find that the ”late Wittgenstein” strikes down on most of the underlying epistemological assumptions of classification throughout history. No small thing. After centuries of Aristotelian classification where each class must be both exclusive and exhaustive, with each subject in one and only one place, the door was suddenly open for other interpretations of the possible organiziation of knowledge. Subject terms could have different meanings in different areas, contexts and situations. Svenonius offers a good and simple example: ”mercury” – metal, planet, car, Greek god etc etc. Classification had to face, as a consequence of Wittgenstein’s thought, such practical issues as how to treat homonymes and polysemes – ”Polysemy abounds”, Svenonius exclaimes.
The turn from the Aristotelian coloured picture theory of the Tractatus, where a subect in a classification scheme mirrored ”reality”, to a view on subjects seen in analogy with family resemblances led to, among other things, fuzzy set theory and use of ambiguity operators in indexing systems.
Reading the literature on knowledge organization one sometimes get the feeling that the end now was reached. If language used in indexing and classification is not absolute, neither in relation to truth nor to reality, but instead relating to discourse and context, we do indeed have a huge set of interesting challenges to deal with - and so has been done.
Philosophical investigations was first published in 1953, two years after Wittgenstein’s death. Even though influential visions of new modes of organizing knowledge had been formulated by Paul Otlet in the 1930’s and Vannevar Bush in the 1940’s, the environment of these thoughts were firmly set in the pre-computorized world – a world, some would say, much simpler. Aristotle reigned still.
Today we have a somewhat different situation. New metaphores and new modes of thinking to intellectually grasp a technology we could not conceive of just a few decades ago seem to be needed. The vast dissemination and lack of organization of knowledge in today’s digital document environments challenge us. It seems to be the very practice of post-modernism (which owe so much to Wittgenstein), that is in front of us. It is in this context that the concept of rhizome has come to relevance in LIS. It was formulated by Deleuze & Guattari in the mid 1970’s. Most people today try as good as they can to avoid French post-modern philosophers, and Deleuze has mostly gone unnoticed in the LIS literature, but in an interesting way the concept of rhizome has stuck, and started its own (rhizome-like) life in the literature of the discipline. Lyn Robinson and Mike Maguire has made a good overview of the relevance and use of the concept.
Rhizome? Well, it is a term taken from botany. There, it is a subterranean stem structure consisting of ”nodes” where roots may spring out of virtually every place of it, in every direction. Most of us encounters rhizomes in the kitchen, as ginger. The rhizome is thus seen metaphorically as the way in which information and knowledge connect on the world wide web and in digital document environments. From every place, anytime, in any direction. The language theory of Wittgenstein is not seen as sufficient anymore. The Internet is an anarchic mass of documents which links to each other in unpredictable ways, making systems for information and knowledge organization face completely new challenges. The question is, of course, how revolutionary this is. Has it not always been like this? Perhaps technical tools have just made the chaos of reality explicit? How do the ”old” metaphores (the tree of knowledge, nets of knowledge) hold up against a dynamic rhizome-like structures of documents? We cannot say. What is clear though is the the need for new metaphores and modes of analysis in knowledge organization are considered to be acute. We may lean on Aristotle, and we may lean on Wittgenstein, and in many aspects they of course still hold sway. Knowledge is not its presentation, the Internet is not our knowledge – it is merely a technical tool which we in many ways still have to figure out what to do with, both in terms of epistemology and in oragnizational practice. Those who say we have already figured it out are not worth believing. Many should – like Wittgenstein – withdraw into silence, think hard and come back when they have something to say on the "novelty" of knowledge today. Perhaps the rhizome metaphor will show to be stronger than more established structures of knowledge classification. Perhaps it does best to keep in the kitchen – the health benefits of ginger seem, at least for now, indisputable.
Svenonius, Elaine (2004) The epistemological foundations of knowledge representations. Library Trends, Vol. 52(3), pp. 571-587.
Robinson, Lyn & Maguire, Mike (2010) The rhizome and the tree: changing metaphors for information oragnisation. Journal of Documentation, Vol. 66(4), pp. 604-613.