Journal of Historical Linguistics - Volume 1, Issue 1, 2011

This article is a contribution to the long standing issue of identifying directionality in semantic change. Drawing on evidence from a sample of morphologically complex terms in basic vocabulary for 149 globally distributed languages, it is argued that cross-linguistically preferred synchronic relationships of word-formation provide clues to likely directions of diachronic semantic developments. The hypothesis is tested against diachronic data from Indo-Aryan languages, and, in spite of a number of counterexamples, a correlation is found. In addition, it is shown how these data can be applied to semantic reconstruction, and a scenario of semantic change which involves morphological complexity in an early stage of semantic development is sketched.

For good reasons, linguists are highly skeptical when it comes to predicting linguistic change. As has been argued in Heine (2003: 598–599), based on observations on some regularities of grammatical change made within the framework of grammaticalization theory, however, it seems possible to propose at least some probabilistic predictions on what is a possible grammatical change and what is not. In the present article it is argued that this also applies to grammatical change that takes place in situations of language contact. As more recent research has demonstrated (Heine & Kuteva 2003, 2005, 2006, 2008, 2010, Heine forthcoming), contact-induced grammatical change is essentially subject to the same principles of grammaticalization as grammatical change not induced by contact. The data analyzed in this article concern case marking in Slavic languages, more specifically expressions for comitative and instrumental participants.

The change of the Standard Arabic interdentals, [θ] > [t] and [s], and [ð] > [d] and [z], in some Egyptian and Levantine Arabic dialects was traditionally explained in terms of two historical changes. The first change was a complete merger of [θ] with [t] and [ð] with [d]. The second change, i.e. [θ] > [s] and [ð] > [z], began with borrowing words with [θ] and [ð] from Standard Arabic, replacing them with similar sounds from speakers’ native phonology — [s] and [z] respectively. The data set used in this study comprises of naturally occurring speech of fifty-two Christian rural migrants to the city of Hims in Syria. The findings suggest that there is no variation in the use of [t] and [s] instead of [θ]. There is rather a synchronic puzzling STABLE LEXICAL SPLIT PHENOMENON: some words are specifically used with [t] and others are specifically used with [s]. The same applies to [d] and [z]. The study shows that although the two historical changes play a role, they are not the sole explanation of this stable lexical split phenomenon. Hence, the role of frequency is examined. Quantitative analyses reveal that highly frequent words are produced with the stops, whereas less frequent words are produced with the fricatives. This phenomenon is explained in terms of usage-based theory and two opposing frequency effects. The first frequency effect led to the merger with the stops. The second frequency effect made highly frequent words produced with the stops resistant to the second change that introduced words with the fricatives.

This article examines changes in the valency marking in the history of English. I start with a discussion of the typological literature on measuring basic valency and point out the problems with such an approach. A sample of 18 Old English verbs provides no basic valency pattern for Old English; this makes Old English different from the other Germanic languages. I then review the evidence, presented in, for instance, Visser (1963), that there is an increase in transitivity in the history of English and argue that this increase is partly due to verbs ceasing to mark Theme-preserving alternations, between anticausative and causative. I also examine Theme-changing alternations, between intransitive and transitive, and argue that, due to the changes in aspect marking, objects become licensed by a light verb, v. I conclude by suggesting a syntactic structure that accounts for the various stages of English and argue that the main changes are due to an increase in morphological intransparency.