DYNAMIS. Acta Hisp. Med. Sci. Hist. Illus. 2002, 22, 551-609.

Geoffrey C. BOWKER; Susan Leigh STAR. Sorting things out: Classification and its consequences, Cambridge-Massachusetts, MIT Press, 2000 [1999], 377 pages. ISBN: 0262522950

Classification is an «old» issue in modernity, as many historians of science have pointed out. But as Star and Bowker suggest in Sorting Things Out, it is also a newly contemporary issue under conditions of flexible capitalism and its particular modes of globalization, particularly with regard to global information systems. Star and Bowker locate their discussion in the wider field of communication and information studies, which address the emergence of changing systems and forms of communication from human speech to electronic telephone and Internet technologies.

Like other works that consider taken-for-granted dimensions of technology as objects of study in their own right, Sorting Things Out considers classification as a central feature of information systems and their associated technologies. This work also draws on science and technology studies, an interdisciplinary field of research characterised by its critical engagement with scientific and technological knowledges, practices and artefacts. Rather than assuming that science and technology are either positive or negative features of the world, scholars in this field ask how and why particular forms of science and technology come into being, and what part they play in the making of our worlds. A central feature of this research is its critique of scientific objectivity and technological neutrality, and its insistence on the fundamentally social and cultural nature of science and technology. The turn to everyday technologies and knowledges within science and technology studies has also made seemingly mundane aspects of «high-tech» subject to analysis.

Bowker and Star ask the same basic questions about classification that science and technology studies scholars might ask about any other scientific or technological invention, namely: what is classification, in what terms should it be described, and what is its social or cultural significance and power? The authors answer these questions using a two-part approach. First, they consider classification as a general category, addressing the methodological and theoretical issues it raises. They link this general discussion to specific cases in the second part of their approach, which analyses four classification systems, including the International Classification of Disease (ICD), the classification of tuberculosis among patients and physicians, the South African system of racial classification before and during apartheid, and the Nursing Interventions Classification.

Working across these systems, the authors consider how classification carries meanings and information across time and space, and how it encodes histories. of moral and political values. So, for example, the authors suggest classifications of death can be found in the ICD categories, but that the entries for these classifications record historical shifts in the definition of death, and are thus introduce an element of ambiguity within the text itself that amplifies the scope of interpretation in practice as well. The authors also consider how classification can exert a twisting force, or «torque» on individual biographies. They argue that in the South African system of racial classification, for example, the lack of fit between mixed-race people and the classification system has impeded access to education, housing and other necessities. At the same time, Bowker and Star suggest, this lack of fit exposes the kinds of exclusions that such systems can produce, and the limits of any classification system. Bowker and Star also show how, as part of an effort to legitimate nurses’ invisible labor, the design of the Nursing Classification System necessarily requires political decision-making, compromises, the reduction of flexibility, and contestation as well as resistance. Classification emerges from this two-part approach as a fundamentally human (and universal), useful, and contingent mode of knowledgemaking whose political and moral effects are always inseparable from the design and use of any such system.

Star and Bowker’s account of classification is unnecessarily universalising and suffers from the lack of any historical dimension. A more located approach, both in time and place would enhance the power of their arguments. Still, for me the book is compelling because of the authors’ pragmatic approach. From the outset, Star and Bowker align themselves with the designers and users of classification systems, rather than as external critics. The question that follows from this positioning is not «what are the limits and failures of classification», but rather «how does classification work, and how might classification systems be designed in politically and morally effective ways?» By framing their project in this way, the authors enact a critique from «inside» the practice of classification that does not begin from an idealised position of purity, but instead theorises classification as a practice that will entail necessary limitations and compromises. The object of their theoretical work is to specify ways of actually making classification systems that are more responsive to historical and cultural changes and differences. Since classification systems have material and social consequences, this pragmatic approach to theory seems most appropriate, as compared to what otherwise goes by the name of «radical critique».

Their pragmatic approach is particularly important, since as Star and Bowker point out, classification is a fundamental component of increasingly powerful transnational information structures. Their approach undercuts any claim to classification as objective practice, and highlights the political dimensions of every classification system. At the same time, their pragmatic approach shows both what particular classifications exclude or fail to capture, and how exclusion can be politically effective in specific cases. By insisting on the inescapability of classification, and then examining how classification works (and doesn’t), their work points towards practical ways of creating and using classification systems such as those currently operating in transnational information structures, in ways that are more attentive to both local and global relations of inequality.

CLAUDIA CASTAÑEDA
Universidad de Lancaster, Reino Unido