OA Database "Adjective-Adverb Interfaces in Romance"

The project Open Access Database Adjective-Adverb Interfaces in Romance shall be carried out by the Research Group The Interfaces of Adjective and Adverb in Romance at the University of Graz. Since 2002, this Research Group has published approximately 40 articles and books on this topic. All publications were consequently based on corpora; three of them are already available in an open access modality via a graphical user interface, the first one since 2005. The project thus stands in the tradition of a corpus based linguistics that since the very beginnings aims at warranting the scientific principle of verification of the analyses by the scientific community, which requires free access to all data. Cooperation with the Centre of Information Modelling (Austrian Centre for Digital Humanities, ZIM-ACDH) in this project will turn it possible to make a big qualitative step ahead, insofar as the tagging of the data according to the state of the art in linguistics (responsible: Martin Hummel) and information technology (responsible: Johannes Stigler) can be updated or newly operated in this way. For this purpose, both cooperation partners have already prepared two fully tagged pilot corpora, which have been used for first publications. The project will additionally provide open access to the underlying research data in XML/TEI P5 format and offer further conversion into standardized formats proposed by CLARIN ERIC in order to ensure optimal re-usability to the scientific community. The project will apply these principles to several, mostly already existing databases, including the data of international partners such as Adam Ledgeway (Cambridge) and Adrian Chircu (Cluj-Napoca) on the dialects of Italy and Old Romanian. Thus a common platform for databases will be elaborated which at the end of the project will cover major parts of the Romance languages, and still more parts on the long run. This means that the traditional methodology of historical-comparative linguistics will receive a modern basis which will be used in the future by the Research Group, e.g. for promoting the career of young researches (already planned: one habilitation thesis, one dissertation), and hopefully by other linguists as well.

The Open-Access Database: Adjective-Adverb Interfaces in Romance" (2017-2020) was a cooperation project between the Department of Romance (https://romanistik.uni-graz.at) and the Austrian Centre for Digital Humanities (https://informationsmodellierung.uni-graz.at). The project created or added a series of corpora to a joined database providing common research tools for all sub-corpora: "Open-Access Database: Adjective-Adverb Interfaces in Romance" (http://gams.uni-graz.at/context:aaif). In sum, over 35.000 historical and present-day attestations for adjectives with adverbial functions in several languages (French, Spanish, Portuguese, southern Italian varieties, Romanian and Latin) have been documented and tagged with the same tagging categories and tools. This includes cases such as Fr. voler haut 'to fly high' or aller droit 'to go direct(ly)', as well as prepositional phrases such as Pt. de novo 'again, newly' or Sp. a ciegas 'blindly'. The project has updated or developed an annotation model including categories related to word formation, syntax, morphology and semantics. Each attestation cited has been manually tagged according to seven different categories and has been attributed to a lemma. For the cross-linguistic purpose, the project has engaged cooperation with international partners, especially for Italian varieties, Portuguese, Romanian and Latin. In addition, three previously existing databases on French and Spanish have been updated and thus saved for the future in terms of sustainability. All the sub-corpora have been tagged with the same categorization and annotation principles. They allow for cross-linguistic parsing, e.g. for several Romance languages at the same time. The research data can be acceded openly and sustainably via the GAMS system, which operates the long-term storage of digital data at the University of Graz. The annotated sub-corpora are freely available and downloadable in reusable formats (XML/TEI, XML/RDF or Excel). The annotation model has been conceived as an Ontology of the Web of Data, thus constituting the conceptual foundation of the technical solutions applied in the research mask. It will allow in future to link the highly structured data with other data sources. The website of the ORD database includes a clear and transparent description of the metadata, the sources as well as the annotation tools. The corpora description further contains a list of the lemmata processed in each of them and the number of attestations. The database is open for adding new sub-corpora in the future. The annotation model is freely available for this purpose. Due to its original cross-linguistic conception, the database provides a useful tool for further research in the framework of the long-term project cluster "The Adjective-Adverb Interfaces in Romance" (https://adjective-adverb.uni-graz.at/en/). Several publications have already derived benefit of the data. The international research community's recent increase of interest in the domain of prepositional adverbial adds additional and certainly crucial value to the database.