ment when the collective is defined no longer by a nature but by ex-
perimentation, it is going to be necessary to have access not to a global
scenarization, now, but to an experimentalprotocol(task no. 7). It is
going to be necessary to keep track of trials, to record their results, to
archive and preserve them. Administrations ensure the continuity of
public life, as we all know. This continuity becomes even more indis-
pensable when it is necessary to retain the entire set of hypotheses,
the propositions that have been accepted and rejected that are gradu-
ally going to compose the common world.
This competency will be found again, for example, in the function
of perplexity (task no. 1): How can we detect new phenomena at the
extreme limit of the sensitivity of instruments, without a meticulous
accumulation of data over a very long time? No one has the ability to
keep track of these except administrators. How can we proceed to
comb and rank various incommensurable entities (no. 3), if no one has
archived the set of choices already made and carefully retained the
more or less solemn engagement of the parties? How can we render ir-
reversible decisions (no. 4) without the multiplication of procedures—
votes, signatures, consensus-building meetings—that allow us to sta-
bilize the collective provisionally? How can we ensure that that con-
sultation (no. 2) takes place according to due process without a persis-
tent verification of the qualifications that allow the various parties
involved to participate in it? How effective would the ethics of discus-
sion be, without being followed up by bureaucratic processes? Admin-
istrators are going to have responsibility for distinguishing all the
functions (no. 5) and for coordinating the various professions; yet they
will not be able to ensure this coordination unless they can prevent
themselves from shifting from forms to content. All the other callings
are substantive; administration alone is, if not proceduralist, at least
procedure-driven.^26
Once the skill of administration is added to those we have deployed
in the preceding chapter, it becomes possible to define more precisely
the learning curve on which the good articulation of the collective
rests from now on, by asking the various callings to collaborate in a
single function. Scientists know perfectly well how to characterize the
learning curve: they call it aresearch front.More than all the others,
they are sensitive to the difference between cold, acquired Science on
the one hand, and hot, risky, dynamic, competitive research on the
EXPLORING COMMON WORLDS
205