Review of The
Philosophy of Information, ed. by Ken Herold. [Special issue] Library
Trends 52(3) Winter 2004, 373-670.
A revolution is afoot in theorizing
information. This issue of Library Trends was stimulated by Luciano
Floridi’s 1999 book Philosophy and Computing[1],
his website www.wolfson.ox.ac.uk/~floridi/, and his quest for a “Philosophy of
Information.” Issue editor Ken Herold invited several individuals to write
something related to “philosophy of information” with or without reference to
what Floridi has said and, more especially, has not said, about Library and
Information Science (LIS).
The first paper, by Ian Cornelius,
politely explains that Floridi’s account “is too innocent of LIS practice to be
accepted without revision.” Then Bernd Frohmann, with barely a mention of
Floridi, argues for a radically different approach drawing on his own important
new book Deflating Information: From Science Studies to Documentation [2].
Following Wittgenstein and work on the social practices of science, Frohmann
denounces mentalistic pictures of information as “occult, noble, and
intentional substance by virtue of which a document is informing.” He argues
cogently for “a shift from theories of information to descriptions of
documentary practices.” Ron Day offers a radical critique of some
underlying assumptions about information, arguing, inter alia, that although
political agencies constitute community through reason and laws, language
itself is constitutive of those same agencies. Jonathan Furner then
demonstrates that there is no need to use the word “information” at all, since
satisfactory terms are already established that cover all that we need to
discuss. Already it is clear that this issue is not a homage to Floridi.
John Budd argues that formal notions
of relevance are too restrictive, Don Fallis picks up on Patrick Wilson’s work
on the problems of verifying and evaluating testimony, Birger Hjørland argues
for philosophical “realism,” and Torkind Thellefsen suggests that a
“fundamental sign” characterizes each knowledge domain.
Elin Jacob defines categorization as
the sorting of objects into sets, and classification as dealing with
relationships between categories. Since the work of Elinor Rosch and others,
Jacob explains, classical categorization can now be seen as imprecise (fuzzy,
non-binary), in strong contrast to classification. (Aristotle also thought that
categorization was imprecise outside of mathematics and it is not explained why
classifications should not now also be seen as imprecise).
Two outstanding contributions
deserving a much wider readership occur in the second half: Jack Mills provides
an excellent, expert and experienced discussion of modern classification theory
and practices in LIS; and Elaine Svenonius presents a useful tutorial, which
could usefully have come earlier in the issue, on different ideas about
knowledge and how these ideas have influenced the design of knowledge
representations embodied in retrieval systems.
Stephen Paling shows that writers
from very different backgrounds have converged in their examination of the
interconnectedness of classification, rhetoric, and the making of knowledge.
Hope Olsen comments on the ubiquity and origins of an emphasis on hierarchy.
Amanda Spink and Charles Cole assert that Human Information Behavior should be
part of any philosophy of information.
A lengthy paper by Søren Brier draws on his many years of work
developing a grand unifying theory of “cybersemiotics” that could adequately
include library and information science. He writes that “The problem with the
now-classical functionalistic information processing paradigm is its inability
to encompass the role of the observer” and “Meanings are the result of a
coupling process based on joint experiences.”
The most disappointing paper is the last, a “Reappraisal” of LIS as
applied philosophy of information, by Floridi, a professor of logic and
epistemology. It is entirely self-referential, citing only his own writings,
and with no indication that he has read any of the preceding fifteen papers.
What are we to make of all this? The image evoked is one of resolute Ptolemaist being politely ignored by a group of Copernicans vigorously engaged in developing a more complex view of the universe. There is a valid and respectable field of formal information theory based on propositions, algorithms, uncertainty, truth statements, and the like, but its formal strengths are also its limits and make inappropriate and inadequate for the concerns of library and information science.
MICHAEL K. BUCKLAND
School of
Information Management and Systems
University of
California, Berkeley