background image
S u m m a r y
When we determine what we are going to say about the pictures in emblem books we have
to make decisions in the following areas:
how to connect the descriptions of emblems to descriptions of other kinds of pictures;
how to connect the descriptions of one emblem to other emblems in a series and of one
emblem book to other emblem books in a collection;
how to use the sometimes extremely detailed information supplied by the epigram, motto,
and commentary;
how to adapt descriptions to the changed meaning of a picture due to a change in usage.
Th e ro l e o f I c o n c l a s s
The question must now be posed about where Iconclass fits in the broader context of the deci-
sions listed in the first part of this paper. Actually, and perhaps surprisingly, it does not.The deci-
sions concerning which parts of a picture to describe and how many descriptors to use for that
purpose are independent of the decision to use Iconclass to express those descriptors. It is only
after decisions about the goal and the depth of our indexing have been taken that Iconclass is
brought in. It then serves a simple, yet crucial, purpose: it is there to help us be more systematic
in our descriptions.
By limiting the variety of our vocabulary and applying a coherent structure to it, we automat-
ically cause the pictorial elements we select for indexing to be grouped in a systematic way.With
almost 30,000 concepts present in the schedules, many of which can be expanded and adapted to
individual pictures, the limitations of Iconclass certainly do not prohibit a rich subject access.
Rather than writing about this in an abstract way, I will attempt to demonstrate how this works
with the help of a number of examples. Some of these examples are taken from our own proj-
ects. Other examples were supplied to us from other projects, in particular from the German
Emblem Project, Digital Emblematica, at the University of Illinois at Urbana-Champaign, for
which I thank Marshall Billings and Mara Wade.
Th e To m - Th u m b p a ra d i g m
When we start to think about describing and indexing historical pictures, it is essential to
decide on the purpose and the status of our descriptions.We have to ask ourselves whether it is
our ambition to give a precise and detailed ekphrasis of each picture, either to evoke it for those
who do not have it in front of their eyes, or to provide the viewer with a historical explanation
of what he sees.This exercise, intended to sharpen the eye, is a standard feature of most art his-
torical curricula and to have mastered this art of description, is considered a hallmark of the art
historian.The alternative is that we identify and label the pictorial elements which we consider
important enough to act as access points in a retrieval procedure. Clearly, this ambition is more
modest. It does not aim to produce descriptions that are the final result of research. On the con-
trary, it aims to produce iconographical data that will facilitate and stimulate new research.
How do we create descriptors that are as effective as the pebbles Tom Thumb dropped behind
him to find his way back again and how does Iconclass help us to accomplish that? In answering
this question, I shall look first at how the hierarchical structure of Iconclass helps us if our descrip-
tors are of divergent levels of precision. I will then pay attention to the way problems of ambi-
guity are solved in a systematic classification. Next, I will discuss the "economics of indexing,"
when I look at the benefits of composite descriptors. How the use of a published standard allows
the end user access to the same facilities as the original indexer will be my last topic.
DC_Emblemsbook_180204 19.02.2004 11:25 Uhr Seite 35