• variationist linguistics
• dialectology and dialectometry
• varieties of English world-wide
• historical linguistics, in particular: syntactic change
• language/dialect typology
• discourse and conversation analysis
• psycholinguistics

• Corpus approaches to geolinguistic and genetic patterns
  of aggregate morphosyntactic variation in varieties of English
• Morphosyntactic complexity in varieties of English
• Predicting syntax in space and time

• the Freiburg Corpus of English Dialects Sampler (FRED-S)

Corpus approaches to geolinguistic and genetic patterns
of aggregate morphosyntactic variation in varieties of English

Taking advantage of the uniquely extensive pool of naturalistic corpus resources documenting a huge number of geographic varieties of English, the research agenda outlined here seeks to marry corpus-linguistic methodologies to dialectometrical and biometric analysis techniques. With a primarily synchronic interest in the ''big picture'' (i.e. aggregate morphosyntactic similarities between dialects and varieties of English), the overarching goal is to deepen our understanding of how, and to what extent, geographical distance and genetic/historical relatedness bear on the representation of linguistic knowledge in authentic linguistic data. Crucially, the investigation will encompass not only geographically neighboring low-contact dialects, but also colonially transplanted, high-contact L1 and L2 varieties of English all around the world. The project will consist of two main studies, one being concerned with dialectometrical analysis (Study 1) and the other one with bioinformatic analysis (Study 2). The line of research proposed here will open up venues for interdisciplinary exchange with historians, geographers, biologists, ethnographers, and population geneticists.

Recent publications: [15], [23], [a3], [a4], [a5], [p5].

This research project applies a set of morphosyntactic complexity metrics to a large number of varieties of English (including traditional dialects, high-contact L1 varieties, L2 varieties, and English-based pidgins and creoles, stylistic varieties, and historical varieties). The project draws on survey data as well as on corpus data. The resulting complexity indices are being investigated in terms of their correlation with a number of pertinent language-external factors (language contact, size of the speaker community, functional pressures, and so on).

Recent publications: [16], [19], [20], [21], [22], [a1], [a5].

Probabilistic models of corpus data can be used both to predict higher-level grammatical choices and to quantify changes in such choices across different speaker groups in geographic or social space and in historical time. The present project will combine these uses of probabilistic models with a third: to measure and compare the syntactic predictive capacities of speakers of different varieties of the same language in parallel psycholinguistic tasks. The aim is to establish theoretical and empirical links between variation at the levels of the group and the internalized linguistic knowledge of the individual, as developed in exemplar-based models of grammar.

Using resources available through FRIAS, we will compare syntactic word order choices in two semantically conditioned constructions of English -- the dative alternation and the genitive alternation -- in historical time (1650-1990) and space (across America and Britain). The directions of change discovered will provide the anchoring context for studies of contemporary variation at the group and individual levels in post-Colonial English varieties in Australia, New Zealand, Canada, and the United States.

This project is part of a larger research endeavor ("The development of syntactic alternations") where we collaborate with Sali Tagliamonte, Marilyn Ford, Anette Rosenbach, and Jennifer Hay.

Recent publications: [8], [a2].

The Freiburg Corpus of English Dialects Sampler (FRED-S) is a 1-million word corpus sampling a subset of those texts in the 2.5-million word Freiburg Corpus of English Dialects (FRED) that are not subject to copyright restrictions. FRED-S is  publicly accessible (please contact me if you're interested). Read up about FRED-S here and here.