Search for a command to run...
Some of us make modest use of citation analysis in our work, 1 but remain radically skeptical of the claims of those who devote more prime time and energy to the elaboration of such methods. Why do we not accept the faith? Why can we not do the proper Kuhnian thing and let the “paradigmatic achievements” of the new quantitative methods define the field for us—posing our fundamental problems, laying down agreed techniques, prefiguring acceptable answers, and unrolling a “progressive research program “? 1 suggest that what is at issue here is essentially a dl~ference of aim. My conception of “doing the sociology of science” allows citation analysis, at best, only a very peripheral role. I will try to outline my position as succinctly as possible. 1) Let me first identify and reject a claim that seems to me to lurk, if only implicitly, behind these quantitative methods: essentially, the claim is that, in transcending the “limited, subjective and biased” perspective of individuals, and in giving some “public, aggregated, objective and unbiased” account, these measures have, as it were, “a preferred logical status”. They are more “objective”, more “reliable”; they can be used to “correct” participants’ accounts; they can define “what really is (or was) the case”, and can arbitrate between conflicting accounts; and so on. These quantitative procedures are often Iabelled ‘‘scientific”, and the sociology (or history) to which they give rise is ‘‘scientific sociology’ ‘—as opposed, presumably, to qualitative, individualistic and “biased”, ‘‘incomplete” sociology. Garfteld, Sher and Torpie, for inst ante, in their pioneer 1964 paper, state: