Artefact – image – text

The challenge of multimodality

Cylinder seals provide us with a wealth of information inherent in their materiality, pictorial and textual contents. A colourful and shiny stone cylinder worn close to the body of its owner, a well chosen combination of written and pictorial signs rolled out as perpetual bands on a lump of clay, the careful arrangement of seal impressions on a cuneiform tablet form a complex system of multimodal communication and perception. Any attempt to unravel their social and cultural functions must confront this complexity, first, by taking it apart, and second, by reconciling it.

Segmentation (first), building relations (second)

A strict differentiation between “artefact” and “image” forms the basis of our approach since one seal image could be impressed on several clay artefacts, and an artefact, i.e. a stone cylinder, a cuneiform tablet, or a simple lump of clay, could bear the impression(s) of one or more seals.

Each seal image consists of visul components which we call “picture elements” including inscriptions. We segment them into boxes and polygons, annotate picture elements in detail through a controlled vocabulary and inscriptions according to TEI standards. Much stylistic and iconographic scholarship depends on comparing pictorial details across time and space. Thus, ACAWAI-CS offers help to anyone interested in the specificities of ancient hairstyles, seating furniture, lions, deities, monsters, etc. Similarly, annotated seal inscriptions can be analysed according to the gender, profession or institutional affilitation of the seal owners (prosopography) and in regard to the deities invoked in them.

While segmentation comes first, interaction and semantic relations between picture elements, and between picture and text, are what ultimate interests us. We find them visually expressed in the figures’ arrangement in groups, in their gesture, movement, gaze, and handling of objects. Modelling relations between entities is ideally done in a graph-based manner. Therefore, a neo4j database serves as backbone and working tool for ACAWAI-CS, helping us to capture and analyse the complex network existing between picture elements and textual components, and ultimately between the artefacts and the people who used them.

From image to word, from word to LOD: The annotation challenge

“A picture is worth a thousand words…” but: How to capture its essence in less than twenty? Creating a concise and consistent vocabulary that turns the complexity of ancient West Asian imagery into a searchable dataset is quite a challenge. Over more than a century of cylinder seals research, scholars applied many different terms to the same compositions and picture elements. ACAWAI-CS seeks common ground, keeping its wording as simple as possible but as complex as necessary. We refer our terminology back to established thesauri like the Getty Art & Architecture Thesaurus or Iconclass whenever possible. However, it remains a work in progress, subject to change with the growth of the endeavour. More information will become available soon.


Contact    Social Media:    Impressum    Privacy Policy