Toward a Sociological Theory of Information

Citation

Cibangu, S.K. (2010), "Toward a Sociological Theory of Information", Journal of Documentation, Vol. 66 No. 2, pp. 297-299. https://doi.org/10.1108/jd.2010.66.2.297.3

Harold Garfinkel's (1952/2008) long‐awaited manuscript on information theory is out. And Anne Warfield Rawls – an engaged critical expert of Garfinkel's work – brings the manuscript into sharper relief by providing in appendix Garfinkel's prior research as far back as July 1942. Garfinkel's manuscript (referred to in the book as memo 3) brings to bear the developments of the information age that took place in the first half of the twentieth century. Widely known for his seminal work on ethnomethodology, Garfinkel as a writer on information theory “will come as a surprise to most scholars – even many who are familiar with his work. But, the argument is entirely consistent with his overall position” (pp. 17‐18). Although Garfinkel made “a more social treatment of information” (p. 32), he furnished one of the most comprehensive examinations of both information concept and information theory.

Presented in a seminar Garfinkel taught during the Spring of 1952 at Princeton University (USA), the manuscript is divided into 15 parts, covering themes such as conceptions, properties, conditions, kinds, factors, and problems of information. The manuscript spans a wide range of disciplines: psychology, communication, business, philosophy, mathematics, economics, and linguistics, among others. Perhaps most interesting to information scientists, in Part I, Garfinkel examines six information theories developed by: Claude E. Shannon; Norbert Wiener; George A. Miller; Karl W. Deutsch; Jurgen Ruesch and Gregory Bateson; and John von Neumann and Oskar Morgenstern.

Part I constitutes the core of the manuscript, with Shannon's information theory receiving the bulk of Garfinkel's critical attention.

Perhaps with the exception of Shannon's and Wiener's information theories, the rest of theories are not quite as familiar to a general readership. Garfinkel's inclusion of these theories is explained by his assertion that “all [these information theorists] treat information as an existent that is subject to clear, empirical, physical, and, in some cases, mathematical manipulations” (p. 102). And Rawls explains that “these theories treated the social aspects of information as relatively unimportant, focusing instead on the symbolic, cognitive and logical aspects of information” (pp. 12‐13). As with mathematical representations and classification of information, focusing on the symbolic aspect of information cannot help but display the relationship between ideas and concepts. By contrast, according to Garfinkel, information is that which is lived and experienced in social order, unexpectedly and contingently. It means that, unlike entropy tenets, an increase in unexpectedness, chaos, and anomaly indicates higher information content. Garfinkel's view counters the classic information theory based on noise‐ and meaning‐free information. Information becomes not the logical descriptor of ideas and referents, or the abstract predictor of effects, but the milieu of mutually and locally shared actions and processes. To a great extent, Garfinkel took issue with the logico‐deductive perception of information, which also implies that information classification needs to be life‐, experience‐, and context‐centric. “It [uinformation] should be capable of being an object of the experiences of love, hate, respect, fear, judgment, and so on … It must somehow be found,” Garfinkel went on, “not only in the wide‐awake attitudes of everyday life and be acquired through the senses, but must be found in the cloud‐cuckoo worlds of dreams, fantasy, scientific theory, the theatre, children's play, etc.” (pp. 110‐11).

According to Rawls, “Garfinkel argues that information only exists in and through the ways in which it is constituted and apprehended cooperatively in social institutions, according to mutually oriented processes of sequential order production” (p. 13). For this reason, information is situated, relational, and non‐redundant. Rawls clarifies, “as Garfinkel points out (and elaborates later, as well as in the two earlier memos), things that are repeated are not necessarily redundant … repeating always says something new” (p. 36). In part II of memo 3 Garfinkel outlines the nuanced character of information, noting, “it must make sense to talk of its clarity or ambiguity, of its uniqueness and typicality; of its private, public, personal, impersonal, anonymous or identified character” (p. 111). When Garfinkel discusses six major information theories in part III of memo 3, he focuses within the school of phenomenology. Two theories lastingly retained Garfinkel's attention: Parsons' and Schuetz' phenomenologies, under Kant's and Husserl's leads, respectively. Garfinkel's phenomenological discussion centers on the relation between the perceiver (informant) and the perceived (information), and in light of this discussion, the dependence or independence between the perceiver and the perceived appear to be problematic because “for a theory of information”, Garfinkel reasons, “the observer is the final arbiter over the question of whether or not and with what degree of coloration of ignorance, error, and myth, the actor has information about the world” (p. 117). Garfinkel maintains, “most particularly, different decisions on these problems yield different theories of information” (p. 111). Upon further examination, the section reveals that Garfinkel's thoughts take shifting stances on information theory. At various points in the manuscript, Garfinkel describes information as a contradictory phenomenon, perfect and imperfect, complete and incomplete, or primary and secondary. Further research might engage these discrepancies. In addition, Garfinkel's (1948/2006) communication theory and information theory (Garfinkel, 1952/2008) call for further research given that while at times both theories appear interchangeable, at other times they contradict each other.

While the manuscript represents a tremendous wealth of research involving information theory, its editor's comments leave the reader puzzled on more than one account. For instance, “in the manuscript,” Rawls states, “Garfinkel launches a serious critique of classic information theory and the semantic and linguistic theories of communication it assumed – a critique that was further elaborated by his later research and also by his students and colleagues” (p. 2). At another point, though, Rawls observes, “he [Garfinkel] does not criticize them [information theories/theorists]. Nor is he accepting their positions” (p. 33). Also more than once, the book mentions “five major theories” (p. 30) or “five information theories” (p. 32) whereas the manuscript lists six information theories (see pp. 106‐9). Finally, the manuscript's title is somewhat misleading, for the manuscript covers concepts and authors far beyond the sociological realm and the editor's comments. It is hoped that the lost section, part XIV: Summary of the Theory, will be found and published at some point in the future. Future information research is needed to identify information as a social and a human phenomenon insofar as Garfinkel's phenomenological arguments leave aside the social nature of information. The book is lacking a pertinent Journal of Documentation (Ditmas, 1948) article, let alone information science sources. Perhaps most frustrating for readers wishing to immerse themselves in the text, Rawls' hundred pages of notes overwhelm the manuscript, taking up roughly the same number of pages as the manuscript itself.