Volume 43, Issue 2

In many areas of linguistics, international publication of one's research in English has become the norm. At the same time, one may find linguistic material from any language quoted in order to support a theoretical argument. This article discusses the use of examples from Norwegian in modern theoretical linguistic literature written by non-native linguists. It shows that these examples are often incorrect and that defective documentation is reused. This seems to be related to a fundamental lack of understanding of the language at issue, and to many linguists' inability to base their research on the linguistic competence of others. Linguistic publishing thus has a problem, yet the situation is all the more serious for linguistics as a discipline. In principle, defective documentation means defective data, which, in turn, may lead to defective analysis and correspondingly to ill-founded theories. A critique of linguistic documentation represents a necessary step in order to re-establish a sound empirical base of linguistics, whatever the theoretical approach.

In the recent literature (Adamson, A lovely little example: Word order options and category shift in the premodifying string, John Benjamins, 2000; Paradis, Reinforcing adjectives: A cognitive semantic perspective on grammaticalisation, Mouton de Gruyter, 2000; Athanasiadou, Language Sciences 29: 554–565, 2007; Traugott, Journal of Historical Pragmatics 8: 295–309, 2007a, Cognitive Linguistics 18: 523–557, 2007b), the development of ‘emphasizing’ meanings by prenominal elements of the English NP is generally envisaged along a path leading from descriptive, propositional meaning to emphasizing meaning (e.g. from a complete sentence to complete nonsense ; cf. Paradis, Reinforcing adjectives: A cognitive semantic perspective on grammaticalisation, Mouton de Gruyter, 2000: 245). However, earlier Bolinger (Degree words, Mouton, 1972: 59, 60) had posited an alternative path for elements that also express determining meanings, i.e. from determining to emphasizing meaning (e.g. from the complete period to a complete fool ). This paper will assess these opposed claims about the origin of emphasizing uses by means of a data-based study of the development of two adjectives, particular and specific , that have descriptive, determining, as well as emphasizing uses. Reconstruction of the various meaning extensions and structural reanalyses that have affected the adjectives indicates that the now mostly neglected determining uses are an essential step in the subjectification process to emphasizing meanings and they can shed new light on both the chronology and the conceptual mechanisms involved in the syntactic-semantic shifts.

In the Papuan language Iatmul, the subordinate predicate in a cleft construction has been desubordinated; synchronically, the resulting focus construction shares biclausal and monoclausal characteristics. What sets the Iatmul case apart from similar constructions in other languages is the possibility to resubordinate the predicate. This constitutes an alternative way to complete the path from bi- to monoclausal status. Rather than continuing the ‘de-marking’ of the originally subordinate verb in the direction of main clause functions, the available marking is exploited so that the previously desubordinated verb can be used in a cross-linguistically unusual type of focus relative clause. There, focus marking is lost on the obligatorily overt subject when the latter is a lexical noun, but is maintained on pronouns. In this case, both nominal and verbal morphology are formally those of a cleft construction, but syntactically the structure is monoclausal.

In this paper we investigate the linguistic phenomenon of transcategorization , that is, the categorial shift of a lexical item with no superficial marking, resulting from its employment in a new (morpho)syntactic environment. Our overall aim is to contribute to the description of transcategorization processes from a typological perspective and to highlight their synchronic consequences on the structure of the lexicon. We analyse paradigmatic instances of transcategorization from typologically different languages and discuss the notion of transcategorization with reference to related notions such as conversion, precategoriality, flexibility and polifunctionality. We argue that transcategorization, understood as a diachronic shift from a source to a target category, is more characteristic of languages with clear-cut parts-of-speech distinctions, such as fusional languages. By contrast, isolating languages, where lexical categories are not clearly marked formally, are better characterized as languages with precategorial lexemes. Our main goal is to stress the role that transcategorization plays in shaping the parts-of-speech systems of languages and to highlight its relevance in parts-of-speech theories and models.

Insights from cognitive psychology indicate that the zero value on a scale is an important reference-point phenomenon in the processing of relative adjectives. However, linguistic evidence for the reference-point status of the zero value has not been provided hitherto. In this paper, I search for such evidence by analyzing the use of dimensional adjectives denoting vertical size in two languages – English and Russian. Three types of linguistic implications of the zero are discussed. First, the zero point is crucial to the account of markedness asymmetry between antonymous adjectives. Second, the reference-point status of the zero value motivates differences in the compatibility of relative adjectives with maximality adverbs in languages such as Russian. Third, the location of the zero point and the position of the entity vis-à-vis the zero point have implications for adjective-noun combinability.

Head marking and dependent marking are considered to be major parameters of syntactic diversity, and are traditionally related to the geographical distribution of languages. Some areas such as Standard Average European favor dependent marking, while head marking is preferred in the New World. However, head and dependent marking may also occur in the same language, either in different domains of grammar or – more interestingly – in competing constructions that are used to convey the same propositional content. In Italian, for example, the choice of head or dependent marking is strongly conditioned by pragmatic factors. Head marking mainly expresses topical information, while the focus is usually conveyed by dependent marking. Moreover, head marking is preferably used in the spoken informal language, while dependent marking prevails in the written formal register. This indicates that structurally different language types may be similar in the organization of pragmatic information.