The concept of “facet” is immensely popular today, much more so than in Ranganathan’s own time. This popularity, confirmed by the extended use of faceted analysis within various types of information systems, does not mean, however, that a definitive characterization of this abstract concept has been established, or that all those who talk about or use facets are entirely clear as to the nature and function(s) of such a handy device. In his article titled: “Des classifications aux thésaurus: du bon usage des facettes”[1], published in 1999, our French colleague J. Maniez adopts a Linguistics perspective to describe what he perceives as a weakness of the faceted approach. Ranganathan first allocated concepts to one of five fundamental facets according to their essential nature; he then proposed for each discipline one or more suitable indexing formulas, assigning specific functions to each facet in the “sentence” thus created.  Maniez implies that Ranganathan himself was not able to, or for some reason decided not to, dissociate the semantic (nature) and syntactic (functional) dimensions of facets in knowledge organization and information representation. This fusion/confusion of semantics and syntax is observed in the diversity of definitions and of applications in contemporary information systems. Such ambiguity is far from uncommon in Information science (IS); Hjorland (2013) suggests that it does weaken IS theoretical foundations, and may even compromise the efficiency of applications.

In the context of a research project whose main objective is to develop a faceted indexing and retrieval system for legal decisions, our team has had to confront this ambiguity while looking for appropriate facets around which to structure the planned system. Through an analysis of recent literature, with particular attention given to the writings of specialists V. Broughton, K. La Barre, C. Gnoli, and B.C. Vickery, we have identified four ways of looking at facets; these have influenced the selection and naming of facets in our own project, as well as their projected role in the information system under development. In this paper, we will first illustrate these four ways of looking at and of exploiting facets, with examples discussed in the literature or observed in current applications. Our focus will be on the following:

-          facet as process / facet as product: Facets may be defined as criteria, as a basis for allocating objects or concepts to mutually exclusive classes. Facets may also be defined as the resulting classes themselves. In the former instance, facets are essential to the dividing process; in the latter, facets are autonomous entities, the product of the dividing process.

-          facet is nature / facet is function: Facets may represent the very nature or essence of a concept or object (living being, activity, attribute, space, etc.). Facets may also be used to specify the role (agent, product, instrument, context, etc.) of the concept or object in a sentence, a classificatory structure, the representation of a subject, etc. Maniez suggests that the former be called categorial facets, of which there will exist very few; the latter should then be seen as structural facets; these could be numerous and vary from one field of application to others.

-          facet of an object / facet of a subject: Facets are commonly used to describe the characteristics of physical objects such as commercial goods, museum artefacts, library documents, etc. Facets are also used to represent the various abstract components of a subject, as in a subject heading or a decimal class number.

-          facet for structure / facet for navigation: As one of the foundational elements of classification systems, facets have traditionally been used to structure document collections, disciplines, even the whole of knowledge. In contemporary information systems, they more often act as filters to facilitate navigation and retrieval, and to improve the presentation of results.

The several potential combinations of these conceptualizations, for example facet as function used to describe a subject to facilitate navigation, add even more complexity to the description and theorizing of an essential IS concept.

In the paper, we will also present examples related to our own project to demonstrate that, in at least two of the four previously described groupings, that is nature / function and structure / navigation, it is possible for a single facet to assume a double identity and to play a dual role in an information system.

The ambiguity between the semantic and syntactic role of the facet, first introduced by Ranganathan himself, was later maintained by the Classification Research Group (1955) and in the many subsequent sets of facets based on the original list proposed by the CRG. Our experience suggests that there is no clear way around the semantic/syntactic confusion, and that the very nature of the concept of facet will remain ambiguous. As suggested by Maniez, terminological refinements could, however, clarify the contemporary discourse around the concept of facet.

Bibliography

Broughton, V. 2013. Faceted classification as a general theory for knowledge organization. SRELS Journal of Information Management, vol. 58, no 1, 49-72.

Classification Research Group. 1955. The need for a faceted classification as the basis for all methods of information retrieval. Library Association Record, vol. 57, 262-268.

Gnoli, C. 2008. Facets: a fruitful notion in many domains. Axiomathes, vol. 18, no 2, 127-130.

Hjorland, B. 2013. Facet analysis: the logical approach to knowledge organization. Information Processing & Management, vol. 49, 545-557.

La Barre, K. 2011. Traditions of facet theory, or a garden of forking paths? Facets of knowledge organization: Proceedings of the ISKO-UK Second Biennial Conference, 4-5 July 2011, London. Bingley, UK, Emerald Group, 2011, 95-105.

Maniez, J. 1999. Des classifications aux thesaurus : du bon usage des facettes. Documentaliste – Sciences de l’information, vol. 36, no 4-5, 249-262.

Vickery, B.C. 2008. Faceted classification for the Web. Axiomathes, vol. 18, no 2, 145-160.

[1] “From classification to thesauri : making good use of facets”