Documenting variation across Europe and the Mediterranean: the Pavia Typological Database

This paper describes the Pavia Typological Database (PTD), a follow-up to the MED-TYP database (Sansò 2004). The PTD is an evergrowing repository of primary linguistic data (words, clauses, sentences) documenting a number of morphosyntactic phenomena in the languages of Europe (and including in some cases languages from the Mediterranean area). Its prospective users are typologists wanting to access primary, typologically uninterpreted (but glossed) data, but also anyone interested in linguistic variation on a continental scale. The paper discusses the background and motivation for the creation of the PTD, its present coverage, the techniques used to annotate the primary data, and the general architecture of the database. 1. Background: Areal typology and Standard Average European The last three decades have witnessed an increasing popularity of areal studies within linguistic typology (see Campbell et al. 1986; Nichols 1992; Haspelmath 1998, 2001; Dahl 2001; Koptjevskaja-Tamm and Wälchli 2002; Ramat and Stolz 2002; Haspelmath et al. 2005, among others). The primary object of these studies is not (or, perhaps, no longer) the identification of Sprachbünde (or linguistic areas, as they are often referred to in English). As Dahl (2001: 1456) puts it, “areal patterns [of distribution of linguistic features] are of interest irrespective of whether they can be described in terms of linguistic areas in the traditional sense”: in fact, hardly any typological variable is evenly distributed over the world, and most reveal systematic areal skewings. For example, inclusive/exclusive distinctions, or numeral classifiers show frequency peaks around the Pacific, clicks are found in substantial frequency only in Africa, relative pronouns only in Europe, etc. As a result of this renewed interest in areal issues, a brand-new discipline has come to the fore, which combines insights from both traditional typological studies and areal linguistics (i.e., the discipline concerned with similarities between geographically contiguous languages which has its roots in traditional dialectology and sociolinguistics). This new discipline, areal typology, can be defined as the study of the areal distribution of typologically relevant features of languages. Many European languages (Romance, Germanic, Balto-Slavic languages, but also, more marginally, the westernmost Finno-Ugrian languages, as well as Maltese) share a number of structural features which give these languages a very peculiar profile and make them stand out among the world’s languages. This “exotic” character of European languages is a relatively new insight, as only recently “linguists realized how peculiar the core European languages are in some ways when seen in the worldwide context” (Haspelmath 2001: 1492). The EUROTYP research program, launched in the early nineties by the European Science Foundation, has produced a critical mass of typological and areal studies dealing with dozens of grammatical features that are characteristic of the core European languages, and that together define the Standard Average European area. 2. The Pavia Typological Database (PTD) Such typological findings have renewed a strong interest in the study of Europe as a linguistic area. In 2003, the research program Europa e Mediterraneo dal punto di vista linguistico: Storia e prospettive (“Europe and the Mediterranean from a linguistic point of view: History and perspectives”), sponsored by the Italian Ministry of Education (FIRB – Fondo per gli Investimenti della Ricerca di Base), was launched under the direction of Paolo Ramat (Università di Pavia). The main aim of this research program has been the typological documentation of morphosyntactic phenomena in languages belonging to the EuroMediterranean area, in order to describe the distribution of various structural traits within this area and to uncover phenomena of both microand macro-areal convergence. This research program is a follow-up to another research program, called MEDTYP and sponsored by the Italian National Research Council (1997-2000), which was concerned almost exclusively with languages in the Mediterranean area. The PTD has been created within the framework of the FIRB research program. Its main goal is to make the amount of data collected within this and the previous research program available to the community of typologists. These data were only partly in electronic form, and were at risk of evaporating in the short term. 2.1. Typological databases Electronic databases are increasingly popular tools in typological research. Despite the advantages of such tools, there are problems connected both with their construction and with their standardization (see below, §4). For instance, there is generally a considerable gap between the information stored in typological databases and primary data: primary morphosyntactic data are much more difficult to handle computationally than typological generalizations. Over the last few years, it has become increasingly clear that the task of typological databases is not only the electronic storage and automatic retrieval of the data, but also the determination of the range of potential variation