World I, World II, World War II, and World III

Brookes, B. C. (1980). The foundations of information science. Part I. Philosophical aspects. Journal of Information Science, 2, 125–133.

Bush, V. (1945). As we may think [Electronic Version]. The Atlantic Monthly, 176, 101-108. Retrieved Nov. 10, 2005 from http://www.theatlantic.com/doc/194507/bush.

Saracevic, T. (1999). Information science. Journal of the American Society for Information Science, 50(12), 1051–1063.


Vannevar Bush (1945) found himself, as an academic and a scientist (one who had been instrumental in the U.S. War effort), swamped by the amount of research that academic specialization had caused to be produced. The body of published work far outstripped the ability of a single scholar to access and assimilate it. He imagines some fanciful aids to research—a walnut-sized, forehead-mounted camera, for instance—but also called for the creation of something very much like the modern computer, attached to a searchable, hyperlinked Internet. “For mature thought,” Bush writes, “there is no mechanical substitute. But creative thought and essentially repetitive thought are very different things. For the latter, there are, and may be, powerful mechanical aids.” For processing observations and data, and adding them to the body of common knowledge, Bush proposes a machine he calls a “memex,” a desk in which a researcher’s books, records and communications would be stored and easily accessed via microfilm for the purpose of association—which is how Bush writes that the human mind works. The memex would create “associative trails” which could themselves be accessed, and even shared with colleagues. The present-day realization of Bush’s dream need not be elaborated upon (or even named, though I do above) for the connection to be clear.

Bush’s thought that the goal of research is to add to a common body of knowledge meshes well with Popper’s world of objective knowledge, as outlined in Brookes (1980). This is Popper’s “World III,” the first two being the physical world (World I) and the subjective worlds of individual human perception (World II). World III encompasses the sum of “human thought embodied in human artefacts, as in documents of course but also in music, the arts, the technologies” (Brookes 1980, p. 127). Brookes argues that this World III should be the basis for the theoretical and practical work of information science. The practical work would be the collection and organization of the knowledge in World III (he writes that librarians are have only worked with archiving the physical documents, not in assimilating and combining their content), and the theoretical work would be to study the interactions between World III and World II, the subjective world of perception. Brookes proposes his own hypothetical machine (though unlike Bush’s, his is a theoretical tool—a metaphor) called a “perceptron” (p. 132). Since it is a machine that can be tuned to pick up certain types of information, what it gathers can be called objective—though as soon as it is transmitted to a human researcher, it becomes subjective. This, to Brookes, is an ideal and a problem for further study. Brookes ends, quite inspiringly, actually, with a call to information scientists to recognize Popper’s World III as a basis for research. He points out that it is the only one created by humans, showing that there is something special about humans and returning us to some of the anthropocentric glory that Copernicus and Darwin (rightly) diminished.

Saracevic (1999) takes these philosophical foundations (those of Bush and Popper) and places them at the beginning of an outline of information science as a discipline. He points out the discipline’s three “powerful ideas”: information retrieval, relevance, and interaction; and he defines “information” in the broadest sense: cognitively processed messages that appear in a relational context (p. 1054). Saracevic writes that information science divides into two clusters: the domain cluster, which includes information analysts of various sorts, and the retrieval cluster, which is mainly interested in applied usages. Within the information retrieval (IR) cluster, there is a paradigm split between those who study systems exclusively and those who acknowledge a human user of the IR systems. Saracevic argues that the important problem here is to find a theory that can encompass both ends of the spectrum. Educational and professional divisions follow similar philosophical lines. On the topic of relevance, the important problem is finding a way to make systems relevance (responses to queries) more compatible with other types of relevance (what users actually search for). Saracevic ends with this as the most important topic for further research in the discipline.

I find it quite instructive to read these essays in chronological order and to watch a discipline move from dreamy theorizing to practical solutions to real problems. I’m personally more attracted to the theory in information science, and I can find great personal utility in Popper’s three worlds, especially in one of my research interests, which is the process of canonization of written works—how a book or an article becomes a “classic.” I also like the idea, which Brookes mentions in passing, that a small change in information can lead to a huge change in understanding. It’s an idea like that that makes the incremental work of scholarly research worthwhile—and tolerable.