Subproject 4.1: Atthidographers

project director: Prof. Dr. Ch. Schubert
project collaborator: André Bünte, M. A.
Lehrstuhl für Alte Geschichte / Historisches Seminar / University of Leipzig

The Atthidographers are a group of historians who wrote down the local history of Athens. The historical work of Hellanicos of Lesbos concering the history of Attica, which has been published after 407/6 B.C., is commonly taken as the beginning of the era of writing of the Atthides, whilst the work of Philochoros from the early 3rd century B.C. marks its end. Those documents are characterized by their annalistic form consisting of sections referring to years preceded by the names of Archonts. The sometimes comprehensive local stories got entirely lost. The remaining sources are quotes and paraphrases that have been found in the works of later authors (refer [Jac03]). Since 2005 the 'Fragments of the Greek Historians' are available in digital form (Brill, Leiden 2005). Furthermore the September 2007 Brill edition offered the digital online version of the 15 volumes of the 'Fragments of the Greek Historians', by Felix Jacobi (1923-1958).

The Atthides which focus laid on myth, religion, history as well as culture, literature and the topography of Attica are offering important information about the history of Athens. There are rarely other sources regarding for instance the development of democracy in Athens. The Atthidographers are mainly concerned with the period reaching from the mythical origins to their own time. In face of these facts the loss of their texts does much damage to historical studies, and the collection of fragments by Jacobi gains great importance.

Here is the starting point of our subproject 'Atthidographers'. The aim of the project is to gain more information about the Atthidographers by the help of modern text mining methods. The tools available for the digitalized corpora so far allow to search in the complete Greek literature for names of the known Atthidographers in order to possibly find more fragments or paraphrases. The problem of this method is that it depends on the occurrence of the original names of the Atthidographers in the searched documents by later authors. This kind of explicit marking of quotes or paraphrases though is not the usual practice in the context of this literature. These facts make the problem of identification of text segments related to the Atthidographers a difficult, laborious, and risky venture (compare to subproject 4.2.).

An advanced method using text mining strategies would be based on a comparison between the statistical parameters of the Atthidograph data on the one side and the whole Greek corpus on the other. New text passages related to the Atthidographs could possibly be identified with the help of such statistical semantic fingerprint.

Beside this, the methods established in the context of the project 'Wortschatz' at the University of Leipzig allow an extended search, not only comprising the actual key word but semantically related terms too. Furthermore, the use of co-occurrences reveals a new multidimensional view on the semantic of a word. The following graph gives an example.


for a bigger version please click on the picture