Theory of reduplication and construction of a database

Reduplication refers to a grammatical procedure in which one word, or a part of a word itself figures as the formative for the expression of a grammatical category. Reduplication is not common in European languages but is one essential means of word formation and inflection in most other parts of the world. Research on reduplication has assumed a rather central role in the linguistic literature of the past 20 years (esp. in morphology and phonology, cf. the different autosegmental approaches, prosodic and templatic morphology, optimality theory, and others). Recent findings in cognitive linguistics propose that utterances should better be analyzed in larger strings than the traditional segment. As reduplication frequently can be analyzed as being based on a prosodic unit (mora, syllable, foot) it constitutes an ideal testing ground for asegmental theories. The scope of the Graz Project on Reduplication is twofold: a) One part of the work will be the installation of a Webserver on Reduplication with the database on reduplication in its center. Besides other functions (complete bibliography, online publications, etc) the database will enrich the theoretical discussion by offering - according to different search functions - information on all relevant aspects connected with reduplication. The aim is to support information on a range of around 200 reduplicating languages. This database is integrated into a European network of typologically oriented databases (LTRC and TypeWeb) and as such will be able to contribute to a more complex correlation of linguistic categories and their formal means, as well as to a better understanding of grammar and language. b) The other part of the project will be the theoretical research on reduplication and related issues, using the collected cross-linguistic data to test hypotheses on the issue. As we have already started with the Graz Conference on Reduplication in 2002, an occasion where most relevant researchers on reduplication came to discuss their work. We intend to establish Graz as one center for typological research on phonology and morphology in general, and reduplication specifically in Austria.A pilot-project of the actual proposal has been funded by the Jubiläumsfonds der Österreichischen Nationalbank. During this period we succeded in establishing and programming the essential lines of the database. We had broad international response to our work and have been invited to take part in the above mentioned European projects. The actual state of the Website, including a preliminary version of the database will be found at: http://ling.uni-graz.at/reduplication/.

As a process of morphology reduplication is either a method of forming new words, or a means of adapting existing words to specific grammatical contexts. In the overwhelming majority of occurrences its semantics are highly iconic, in the sense that reduplicated forms frequently denote grammatical meanings related to intensity or augmented quantity. Because of its specific formal procedure reduplication is very well suited for non-linear, prosodically based models of grammatical processes, as the chunk of the base-word (stem, root, etc.), consisting of more than one segment, easily lends itself to a description in prosodic terms. Non-linear procedures have been in the centre of attention of linguistic (and other cognitive) studies in the past two decades. Therefore, also reduplication has been the matter of an increasing number of publications, frequently disregarding the empirical basis. The aim of the Graz project on reduplication was to construct an electronic database in order to allow a.) a more critical handling of data; b.) to formulate own questions which crucially involve reduplication and verify possible answers empirically; c.) the advance of typological knowledge by not randomly comparing items or concepts, but by checking hypotheses through a sampled selection of languages of the world. The database has been constructed taking into account all accessible criteria to be found in treatments on reduplication in the languages of the world and the most advanced studies in the field; the programming turned out to offer two basically distinct portals to approach the database: a traditional search tree function and a query builder where the display of the results reflects the specific interest a user may formulate according to his scientific goals. The website moreover offers a series of additional services, like the vastest searchable bibliography on the topic, classical 19th century literature in pdf-format, and utilities like the reduplication finder, a tool for screening material in order to find possible reduplicative patterns. Besides the database which is the heart of the project, the project members have contributed to the discussion both regarding the technical side (as to workshops and publications on the use of databases and its technical specificities), but also on theoretical linguistic issues (e.g., to the discussion of the origin of reduplicative patterns), as well as descriptive matters such as Veronika Mattes` treatment of reduplication in Bikol. Moreover, during the period of the database activities there took place two international conferences: one on general matters of reduplication, and one on the specific topic of diachrony and productivity of reduplication. Both conferences were edited at renowned publishers (Hurch 2005 and Hurch & Mattes 2009).