September 7-9 2017
Université de Neuchâtel, Switzerland

The 2017 SLE summer school will take place in September 2017 at the University of Neuchâtel in Switzerland. The summer school will offer a range of short theoretical and methodological courses that will introduce the participants to linguistic topics of high current relevance. The courses will be complemented by three keynote lectures.

The deadline for applications has passed, many thanks to everyone who has applied! We have received applications from 28 different countries, and we are currently reviewing them. Notification of the outcome will be given by February 1, 2017. We wish all applicants success and the best of luck!

The summer school is competitive, with a total of 30 places. Tuition, accommodation and lunches will be provided free of charge for successful applicants, but we ask delegates to pay for their own travel to and from Neuchâtel.


Damian Blasi (Zurich)
Data science applied to language: a bag of methods to wrangle with linguistic data

It is commonplace to note that the 21st century came with unprecedented amounts of data and statistical methods relevant for the study of language. Data-oriented science doesn't only provide ancillary tools to traditional research on language, but it can also serve as a source of new hypotheses, as a principled way of discovering complex patterns in data and as a safeguard against some of the biases that affect qualitative research. On the other hand, inferences with data come with their unique set of limitations and opportunities for misuse. In this course, I'll provide a brief introduction to data science, revisiting classic statistical topics as well as recent methodological developments from the last decade. I'll illustrate the concepts with case studies taken from my work as well as from others, ranging from linguistic typology to language acquisition, historical linguistics and psycholinguistics.

Session 1: General introduction to statistical modelling and machine learning
Session 2: Regression and classification, causal inference and model evaluation
Session 3: Clustering, mining and learning from the data

Francesco Gardani (Zurich)
Advances in contact linguistics

Language contact is one of the fundamental factors of language change. This course will begin with a survey of prominent approaches to contact linguistics, such as Thomason & Kaufman (1988) and van Coetsem (2000), and of the possible effects of language contact on all levels of linguistic processing - phonology, syntax, and morphology. After this introductory session, the course will focus on types of borrowing and degrees of borrowability. In the final session, we will consider a selection of case studies on a world-wide scale. The course pursues three main goals: First, to introduce the most recent developments in this area of research; second, to make the students acquainted with the array of data that has become available in recent years; and, third, to lead the students to a full understanding of how data from language contact informs linguistic theory, in terms of the architecture of grammar.

Session 1: Contact-induced language change
Session 2: Typology of borrowing
Session 3: Case studies

Natalia Levshina (Leipzig)
Semantics in space: semantic maps, MDS, correspondence analysis and graph-theoretical approaches

Spatial metaphors are ubiquitous in semantic research. Linguists speak about semantic maps, semantic vector spaces, semantic distance, semantic narrowing and broadening, etc. This practical course shows how one can use R, a free statistical environment, to create spatial semantic models, which allow the linguist to compare the semantics of different linguistic units, as well as to investigate the semantic structure of a specific word or construction. One can also visualize the development of semantics in time or explore intra- and cross-linguistic semantic variation. The statistical procedures and visualization techniques discussed in this course include traditional methods (Multidimensional Scaling, Cluster Analysis, Correspondence Analysis), as well as more recent ones (graph-theoretic approaches and motion charts). The methods are presented in a linguist-friendly way, with a focus on the essential link between linguistic theory and statistical methods. For each of these methods, I will discuss the data format, main steps of the analyses, available diagnostics tools, practical caveats and interpretation of the visual output. The course includes real-life case studies and practical exercises. R code will be provided for all steps of the analyses.

Session 1: Implicational and probabilistic semantic maps based on cross-linguistic data: graph theory and Multidimensional Scaling.
Session 2: Distributional models of semantics: Semantic Vector Spaces and motion charts.
Session 3: Exemplar-based models of semantic structure: Correspondence Analysis and Multidimensional Scaling.

Florent Perek (Birmingham)
Experimental methods and behavioral data

This course will introduce participants to the use of behavioural methods in linguistic research, in the form of controlled experiments tapping into the latent linguistic knowledge of speakers and/or their cognitive processing of language. Traditionally the province of psycholinguistics, such experiments typically involve measuring the effect of certain variables (either language-internal or language-external) on the behaviour of speakers in language production or comprehension, which is meant to reveal aspects of the cognitive representation of language. The course will briefly review the use of experiments in linguistics, and describe several common experimental designs, such as self-paced reading, priming, lexical decision, sorting tasks, cloze tests, etc., with a particular focus on the kind of problems they can be applied to, the type of data that they produce, and how this data is interpreted. Participants will be taught how to design experiments, and will receive a brief tutorial on the PsychoPy software package for implementing them.

Session 1: A review of experimental methods
Session 2: Designing experiments
Session 3: PsychoPy: a crash course

Karsten Schmidtke-Bode (Leipzig)
Usage-based linguistics and language typology

The course provides an introduction to the so-called usage-based approach to language structure and its application to linguistic typology. After a general outline of the historical roots and core assumptions of the theory, we will look more specifically at the ways in which usage-based linguists have brought their specific conceptual apparatus to bear on the explanation of cross-linguistic patterns in morphosyntax. As it turns out, many important phenomena in typological research (e.g. argument marking, word-order correlations, clause-linkage patterns) have been explored from several, apparently competing avenues of usage-based explanation, and a central goal of the course is to discuss and to systematize these different strands of usage-based thinking. The course does not presuppose any prior knowledge of either the framework or linguistic typology.

Session 1: Introduction: Usage-based linguistics, linguistic diversity and typological patterns
Session 2: Explaining linguistic universals (I): Conventionalized processing efficiency
Session 3: Explaining linguistic universals (II): Historical persistence; Towards a synthesis

Elena Smirnova (Neuchâtel)
Diachronic construction grammar: Tenets, recent developments and open questions

When the framework of Construction Grammar (CxG) was originally developed, its main objective was to develop a theory of language that would most accurately describe speakers' linguistic knowledge. CxG was thus born as a synchronic theory of language. In was only in the second step that it proved to be a useful descriptive tool for the analysis of language change. As a usage-based and basically cognitive approach to language structure, CxG lends itself very well to modelling gradual, incremental, bottom-up change resulting from language use. Such concepts as frequency effects, analogy, chunking, and entrenchment find their natural place in the new field of linguistic research meanwhile known as Diachronic Construction Grammar (DCxG). The course provides an overview of recent developments in DCxG research and of central questions that drive current research in this area. We will focus on both theory and application, covering conceptual fundamentals of DCxG and their application to particular instances of language change. We will be looking at case studies from different domains of grammar and from different languages. Strengths and limitations of the DCxG approach as well as questions for future research will be debated at the end of the course.

Session 1: Introduction: Construction Grammar and diachrony
Session 2: Overview of the current research field of DCxG (case studies)
Session 3: Open questions and areas for further research


More detailed information about the keynote lectures will appear in due course.

Susanne Michaelis (Leipzig)

Marianne Mithun (Santa Barbara)

Johan Van der Auwera (Antwerp)


The applications will be ranked primarily on the basis of academic merit. In keeping with the goals of the Societas Linguistica Europaea, we particularly encourage applications from weak-currency countries. The organizing committee will strive for a balance of different countries, different genders, and a broad range of linguistic interests in the pool of accepted applicants. Notification of the outcome will be given by February 1, 2017. We wish all applicants success and the best of luck!


Martin Hilpert (Neuchâtel)
Francesca Masini (Bologna)
Andrej Malchukov (Mainz)
Michela Cennamo (Naples)
Jeanette Sakel (Bristol)



The summer school will end on the afternoon of September 9, so that participants who plan to attend the SLE conference in Zurich (Sep 10-13) can seamlessly combine the two events: Zurich is a 90-minute train ride away from Neuchâtel.

Last update: Jan 9, 2017.

New speech disorder linguists contracted discovered!