In order to realize the potential of historical computing, a formal-
ized research process is essential. Already the establishment of
computer data bases requires much systematic planning. Even more
than linguistic analysis, the quantitative exploration of such files
in tables and graphs yields its full benefits only when based upon
a structured and controlled research design. Though in principle
similar to hermeneutic or philological interpretation of texts, for-
malized research demands a more systematic and self-conscious ap-
proach, not leaping instinctively from document to insight, but
rather carefully considering the implications of each investigative
step. That is not to say that an occasional quotation or table cannot
be used as illustration without much methodological awareness.
But data bases and quantitative methods can only make their full
contribution to historical understanding when they are applied to
appropriate questions in a formal manner. Since reams of computer
printout do not automatically yield compelling results, historical
computer users must be willing to risk confrontation with some of
the central elements of social scientific thinking. According to the
recent introduction by Kenneth R. Hoover ( 1988), the importance of
such a structured research design can hardly be overstated. Hence
the uninitiated may well ask: What are quantifiable questions, how
does formalized research proceed, and what are its implications for
project planning?

At the beginning a researcher often wonders what constitutes a
quantifiable historical problem. In order to make a wise decision, a

-25-