G Arcodia

Towards a typology of relative clauses in Sinitic: headedness and relativisation strategies

The study of relative clauses (RCs) in Sinitic has attracted considerable attention in typological literature, especially because of the exceedingly rare correlation of Verb-Object order and prenominal relatives (Tang 2006; Dryer 2013b). As to the structural differences of RCs within Sinitic, although previous studies have brought to light several key issues (e.g. Liu 2005), there is much work yet to be done in this area. Basing on data from a convenience sample of 44 Sinitic languages, in this paper we will try to sketch a typology of relativization in Sinitic languages according to two parameters, namely the position of the head and relativization strategies. Also, we will highlight some meaningful correlations and implications constraining diversity in the domain of relativization in Sinitic.

The Construction Morphology Analysis of Chinese Word Formation

The lexicon of Modern Chinese is characterised by a preponderance of multimorphemic words, which are typically built from lexical morphemes, either bound or free. Compounding, broadly understood as the combination of two or more lexical morphemes, is by far the most common word formation device in the modern language. While drawing a sharp boundary between compounding and derivation for Chinese has proven difficult, there are indeed a number of items which possess derivation-like features, including bound status, fixed position, and a stable, often bleached meaning. Moreover, bound items, sometimes without morphemic status, may acquire the meaning of a word as part of a construction, and generate new words and constructions with that acquired meaning. In this chapter, we will apply the principles of CxM to the analysis of Chinese complex words, showing how a constructional approach may best explain several phenomena which are characteristic of Chinese word formation, including the genesis of new meanings for lexical morphemes as part of word formation schemas, rather than in isolation. Also, we will show that the parameter of headedness in compounding may not be set for the language as a whole, but is rather specified in schemas.

More on the morphological typology of Sinitic

Chinese is often defined as a ‘textbook example’ of an isolating language, with comparatively few affixes that are usually etymologically transparent (Sagart 2004). After ‘deconstructing’ the notion of the isolating morphological type, I shall discuss data from a number of Chinese dialects spread over the Shanxi, Shaanxi, Henan, Hebei, and Shandong provinces. I will show that there seem to be some areal clusters with productive morphological phenomena not expected to occur in isolating languages, which can be explained both by the cross-linguistically widespread tendency towards the reduction of certain items in speech production and, arguably, by processes of convergence among dialects. (This article is in English.) 提要 學者們通常認為漢語是典型的孤立語，詞綴較少，由來知根知底 (Sagart 2004)。本文首先對孤立語的概念進行“解構”。接著，基於山西、陝西、河南、河北以及山東方言的語料，說明漢語方言中似乎存在著地域上成片的具有能產性的形態現象，而這種現象似乎不該出現在孤立語中。本文認為，產生這類形態現象的動因包括語法化標記語音侵蝕的普遍傾向以及方言之間的互相影響。

ON THE PRODUCTIVITY OF THE CHINESE SUFFIXES − 兒 −R, −化 −HUÀ AND −頭 −TOU *

The notion of ‘productivity’ is an essential one in the study of linguistic morphology, but its definition is indeed challenging, and there are different ways to measure different aspects of the productivity of a morphological process. In this paper we shall adopt Baayen’s P measure of productivity for a corpus-based study of the productivity of three Mandarin derivational suffixes, namely the nominalizer/diminutive −兒 −r, −化 −hua ‘−ise, −ify’ and −頭 −tou, a ‘dummy’ nominal suffix (Lin 2001:82), in order to assess how this index relates to our received knowledge about the productivity of such forms, and, also, to compare our results with a previous study by Nishimoto (2003) on a small corpus of Modern Chinese. Moreover, in a diachronic perspective, we shall compare data from the Academia Sinica Tagged Corpus of Early Mandarin Chinese and from the Academia Sinica Balanced Corpus of Modern Chinese. We shall show that our P values mostly reflect what descriptive works tell us about the productivity of the affixes considered here in two different periods of the history of the language; when corpus data for previous stages of a language are available, they appear as a better basis for assessments on the profitability of a morphological process than dictionary data.