Journal of the American Society for Information Science
40(3):152. 1989

Back to H. Small and E. Garfield's letter

Response: Absence of Citations Can Be Valuable Clue


Henry Small and Eugene Garfield have contributed’ an interesting suggestion for using co-citation analysis to explore multiple-link indirect paths that might connect two literatures.

They also note that, in my article, I had failed to find citation and indexing analysis useful in revealing logical connections [ 1]. However, in the light of my subsequent Brief Communication concerning an exploratory trial-and-error search strategy, that failure itself can now be placed in a new and more positive perspective [2]. Once a logical connection between two literatures is hypothesized (perhaps with the aidof the idea proposed by Small and Garfield), it is still necessary to determine whether that hypothesis is already known, our goal being to find previously undiscovered connections. A co-citation analysis of the two literatures is then helpful. If there are many co-citations, the hypothesis can be discarded for our purposes on the ground that other people have already brought the two literatures together (by cociting them) and therefore almost certainly must have noted the connection. A negative result (i.e., few or no co-citations) is even more interesting for it suggests that the hypothesis has not’ before been proposed, and so is eligible for further study. Thus what I called a "paradoxical quest for the absence of retrieval clues" [1, p. 233], was neither as whimsical nor as paradoxical as it might have seemed. The route to successful hypotheses is first to make a lot of guesses then get rid of those that are wrong.

Don R. Swanson
University of Chicago

1.  back to text Swanson, Don R. "Two Medical Literatures that are Logically      but not Bibliographically Connected." Journal of the American Society for Infor~nation Science. 38(4):228—233; 1987.

2.  back to text Swanson, Don R. "Online Search for Logically-Related Noninteractive Medical Literatures: A Systematic Trial-and-Error Strategy. Journal of the American Society for Information Science. (in press).