Towards a database of Homeric nominal compounds

In the study of Ancient Greek language the classification and analysis of compounds is a challenging matter. In particular, Homeric language provides us with a very ancient and problematic corpus of compounds, as it comprises items that date back to different stages of formation and often show peculiarities due to the artificial character of epic poetry. Within this framework, an interesting possibility is to create a digital database of the Homeric compounds, that on the one hand paves the way to deepen our knowledge of compounds in Greek and of the process of compounding in general, and on the other hand provides a useful analysis tool for both linguists and philologists. This database, currently under development by us, will first of all list all the compounds occurring in the Homeric texts, along with their principal meanings. Then, the database will provide the reader with information about the lexical and morphological status of both the compound (compositional and lexical category) and its members (lexical category and etymology). To offer exhaustive information for each compound, all its occurrences in the texts will be listed (possibly with links to online corpora), together with the most relevant bibliographical references. Moreover, since Homeric compounds are often found in later stages of the development of Ancient Greek, the research will investigate whether semantic shifts took place in diachrony, and what kind of them. At the present stage of research, we have arranged the material as raw data in a MS Excel chart. For the construction of an online database, a preliminary study is to be done on the dependency relationships between the various fields of the db, since the structure of a relational db fits more our needs. So far, we have listed all the compounds in the Homeric corpus (ca. 1700 items) with the indication on the form of their First Members (FM) and Second Members (SM), alongside with a provisional hypothesis of classification. The major issues with the elaboration of the db involve the adequate account of the frequent allomorphy of both FM and SM and the fact that in some cases a compound may be listed under two different classes. Moreover, it is worthwhile to note that the goal of our db is not to do the linguistic analysis,. The latter is indeed a linguist’s task, given that the analysis necessarily implies a thorough survey of the occurrences of the compounds in the texts. As a consequence, our aim is to make the data available in the most clear way, after their manual upload by the linguist. This tool will therefore allow for comparative analyses and cross queries and thus will help the user easily pinpoint similarities and differences between compounds sharing common features. As a final remark, the classification of the compounds will imply reflecting on what classification criteria to propose. To this extent, the project of this digital database of Homeric compounds is conceived of as part of a wider research on composition in its general issues, not only in Greek but possibly in other ancient Indo-European languages.