Stability and Instability in Grammar
ICL20 Workshop | International Congress of Linguistics, July 2–6, 2018, Cape Town, South Africa

Theme of the workshop

What can cross-linguistic frequency of grammatical structures tell us about their diachronic (in)stability? The workshop starts from the null hypothesis that diachronically stable properties are those that appear as the typologically most frequent ones, and that cross-linguistic rarity correlates with diachronic instability.

By ‘stability’ we mean the resistance against forces of language dynamics and change. Historical linguistics can reveal how strongly a particular grammar has changed for a particular language. This kind of information from a large number of languages makes possible more general statements about the tendency of specific grammatical properties to stay stable for a long period of time or to change rather quickly (cf. Harris & Campbell 2000; Fleischer & Simon 2013). On the other hand, recent progresses in quantitative typology enable us to make substantive generalizations about the cross-linguistic frequency of language structures. Taken together, historical and typological data allow us to assess the explanatory power of the above mentioned null hypothesis.

To take word order as an example, 76% of all languages (WALS) show either SOV (41%) or SVO (35%) as their dominant word order, which would suggest that SOV and SVO are more stable and resistant against language change than the other four possible systems. However, there is a much stronger diachronic tendency for SOV to change towards SVO than vice-versa (Vennemann 1974; Newmeyer 2000). This suggests that SVO is the most stable system whereas the superior number of SOV systems might be due to other factors, but this does not reflect the stability of SOV (Gell-Mann & Ruhlen 2011). As the word order example shows, in order to make better predictions about the stability of grammars, a fruitful synthesis of typological and historical data is indispensable. Statistical evidence for recurrent patterns might be a first indicator for its stability aspects, but it needs to be controlled against common diachronic pathways in the languages of the world.

The workshop aims at bringing together typologists, historical linguists and evolutionary linguists who are studying stability aspects of grammars, which might non-exclusively include: word order (cf. Sornicola, Poppe & Shisha-Halevy 2000), case marking (cf. Berg 2014), aspect and tense (cf. Herring 1993), consonant and vowel inventories (cf. Jäger 2008), etc. We invite for empirically well-grounded contributions that discuss the relationship between (in)stability and cross-linguistic frequency and its possible theoretical consequences, e.g. whether common pathways of change shape the typological landscape (Blevins 2004) or whether typological preferences constrain the direction of change.

We believe this theme serves to highlight a growing interest to test linguistic theory against increasingly larger data sets (extensive fieldwork and documentation, corpus data, typological databases). We envision that this theme captures several areas of emerging interests such as the breakdown of a longstanding fundamental distinction between the cross-linguistic quantitatively motivated research and diachronic qualitatively motivated research.

This Workshop is held at the 20th International Congress of Linguistics in Cape Town, South Africa (July 2-6 2018). It is organized by Dankmar Enke, Guido Seiler (LMU Munich) and Roland Mühlenbernd (Ca'Foscari University Venice).

The scientific committee includes:

Juliette Blevins (The Graduate Center, CUNY)
Heather Burnett (Université de Paris 7 - Denis Diderot)
Ashwini Deo (The Ohio State University)
Regine Eckardt (University of Konstanz)
Jürg Fleischer (University of Marburg)
Michael Franke (University of Tübingen)
Elly van Gelderen (Arizona State University)
Olav Hackstein (LMU Munich)
Martin Haspelmath (Universität Leipzig, MPI Jena)
Agnes Jäger (University of Cologne)
Gerhard Jäger (University of Tübingen)
Johanna Nichols (University of California, Berkeley)
Eva-Maria Remberger (University of Vienna)
Gerhard Schaden (Université de Lille)
Horst Simon (Free University Berlin)
Mary Stevens (LMU Munich)