Approaches to linguistic areas have largely focused either on purely qualitative investigation of area-formation processes, on quantitative and qualitative exploration of synchronic distributions of linguistic features without considering time, or on theoretical issues related to the definition of the notion 'linguistic area'. What is still missing are approaches that supplement qualitative research on area-formation processes with quantitative methods. Taking a bottom-up approach, we bypass notional issues and propose to quantify area-formation processes by (i) measuring the change in linguistic similarity given a geographical space, a sociocultural setting, a time span, a language sample, and a set of linguistic data, and (ii) testing the tendency and magnitude of the process using Bayesian inference. Applying this approach to the expression of reflexivity in a dense sample of languages in northwestern Europe from the early Middle Ages to the present, we show that the method yields robust quantitative evidence for a substantial gain in linguistic similarity that sets the languages of Britain and Ireland apart from languages spoken outside of Britain and Ireland and cross-cuts lines of linguistic ancestry.*

Démotion du sujet et de l’objet direct avec imm-(a-N) en irlandais médiéval. L’irlandais médiéval est susceptible d’intégrer dans son complexe verbal un morphème imm-(a-N) avec des propriétés syntaxiques et sémantiques très particulières. Dans cet article, nous nous proposons de fournir une description de l’emploi d’imm-(a-N), tout en tenant compte des différents changements que ce morphème induit sur la réalisation des arguments, et d’explorer la manière dont imm-(a-N) modifie la sémantique du verbe. La chronologie des données à notre disposition suggère qu’à l’origine, les constructions comportant ce morphème imm-(a-N) ont servi à désigner des événements réciproques et que, à partir de là, elles sont rapidement devenues un moyen pour exprimer des événements d’action collective simultanée (plusieurs participants effectuent la même action dans le même contexte temporel et local, mais pas réciproquement) ; du point de vue de la sémantique, ces constructions se comportent comme des verbes actifs. En ce qui concerne la réalisation des arguments, elles combinent à la fois des caractéristiques du passif (démotion du sujet) et de l’antipassif (démotion de l’objet).

In many cases of apparent contact-induced change the contribution of genealogical correlation in the language sample and its interaction with processes such as matter and pattern replication are difficult to specify. In order to get a better sense of the relevance of shared ancestry, we quantify the change in similarity since the late Middle Ages in a sample of Romance and Germanic languages with data from a selected grammatical domain (expression of reflexivity). We compare their dynamics to patterns of change of similarity in two contact zones in Europe, namely the British Isles (Dedio et al., 2019) and the Balkans. Concerning the genealogical signal, the results indicate a maintenance and gain of similarity in Romance as opposed to a loss of similarity in Germanic. This hints at the importance of the inherited states, the time since the split from the common ancestor, and subsequent developments. We presume that these factors are likely to be at the origin of the maintenance and increase in similarity observed for the sampled Romance varieties. While this result cannot be generalized beyond the specific case study presented here, the basic approach will contribute to a better understanding of how contact, genealogy and culture interact in shaping the dynamics of linguistic similarity.

No abstract

The—for European languages—large amount of bound elements in the older Insular Celtic languages and the array of phonological interactions within morphological and phrasal structures have lead several researchers to conclude that individual words play a lesser role in the grammars of those languages. Based on current typological research on wordhood and a thorough discussion of the problems and limitations of studying wordhood in corpus languages, this article gives an in‐depth investigation of the morphological and phonological word‐like domains in the Middle Welsh verbal complex. There are several structures that could be labelled ‘word’ on the basis of the findings presented here, which leads to the conclusion that the term is not very useful for the synchronic description of this language. Die für indogermanische Sprachen große Menge gebundener Elemente in den älteren inselkeltischen Sprachen und die Prävalenz morphologischer und phonologischer Interaktion in morphologischen und syntaktischen Strukturen haben einige Forscher zu der Schlussfolgerung gebracht, dass die einzelnen Wörter in den Grammatiken dieser Sprachen eine untergeordnete Rolle spielen. Auf der Grundlage aktueller typologischer Forschung zu wordhood und einer Diskussion der Schwierigkeiten bei der Untersuchung von wordhood in Korpussprachen bietet dieser Artikel eine detaillierte Untersuchung der wortartigen Domänen im mittelkymrischen Verbalkomplex. Da es mehrere Strukturen gibt, die auf der Basis dieser Untersuchung als ‘Wort’ bezeichnet werden können, erscheint dieser Begriff für die synchrone Beschreibung dieser Sprache wenig hilfreich. [German]