Volume 57, Issue 6 -

In polysynthetic West Caucasian languages, the morphological verbal complex amounts to a clause with all kinds of participants cross-referenced by affixes. Relativization is performed by introducing a relative affix in the cross-reference slot that corresponds to the relativized participant. However, these languages display several crosslinguistically rare features of relativization. Firstly, while under the view of the verbal complex as a clause this affix appears to be a relative pronoun, it is an unusual relative pronoun because it remains in situ . Secondly, relative affixes may appear several times in the same clause. Thirdly, relative pronouns are not expected to occur in languages with prenominal relative clauses. Fourthly, in the Circassian branch, relative pronouns are identical to reflexive pronouns. These features are explained by considering relative prefixes to be resumptive pronouns. This interpretation finds a parallel in the neighboring East Caucasian languages, where reflexive pronouns also show resumptive usages. Finally, since in some West Caucasian languages the relative affix is a morpheme with a dedicated relative function but still shows properties of a resumptive pronoun, our data suggest that the distinction between relative pronouns and resumptive pronouns may not be as clear as is usually assumed.

Numerous studies have dealt with the coding of information sources within and across languages. These studies have shown that despite the significant differences in the number of formally distinct evidentiality categories, languages tend to have grammaticalized markers for certain information sources, but not for others; different kinds of sensory perception, inference, assumption and hearsay evidence are among those information sources that receive explicit coding. In this paper, another evidence type, namely general knowledge, will be examined. It will be shown that general knowledge differs from other information sources in its nature, but it also has features in common with them. In addition, a formal-functional typology of general knowledge coding will be proposed based on the nature of the element used for this purpose. Finally, the rationale behind the discussed types and the central theoretical implications of the paper will be discussed. The attested types either stress the peculiar nature of general knowledge providing evidence for its independent information source status, or they emphasize the common features shared by general knowledge and other information sources, most notably reliability of information.

Predicate nouns in German, as well as in other languages, may occur bare or with an indefinite article. This alternation is possible with role nouns, which refer to well-established aspects of individuals such as professions and nationalities. Bare NPs differ from indefinite NPs in that they have restricted meaning, are number neutral and are restricted in modifiability. In the literature, these peculiarities received different explanations. The new account combines previous analyses and is based on the following assumptions: the noun that projects an indefinite NP denotes a kind, while the noun projecting a bare NP denotes a capacity. This difference corresponds to the difference in predication: indefinite NPs predicate about the whole individual assigning it membership in a certain kind, while bare NPs predicate only about one social aspect of the individual, identifying it with a certain capacity. Since bare predication concerns only one aspect of the individual, it is partial. Bare predication can now be considered under a broader view of partial predication, a phenomenon very common in argument alternations, and can be analyzed with the tools that have proved effective in this domain. The approach to bare predication taken here thus has a larger empirical coverage.

This paper aims to bridge the relationship between metalinguistic if you like as a non-propositional discourse marker and its conditional counterparts. This paper claims that metalinguistic if you like is polysemous between a hedge that denotes the speaker’s reduced commitment to some aspect of the main clause, and an optional yet potential conditional reading that interlocutors can legitimately draw on in interaction which is brought about due to the ‘if p , q ’ sentence form. That is, although the metalinguistic reading is most likely obtained automatically by default, it also carries an available conditional reading that is akin to other metalinguistic conditional clauses such as if you see what I mean . Next, a semantic representation of metalinguistic if you like is developed that takes on board a characterization of conditionality that departs from lexico-grammatical conventions, such that conditionals of the form ‘if p , q ’ no longer bear a one-to-one correspondence with “conditional” truth conditions. Employing a radical contextualist semantic framework in which the unit of truth-conditional analysis is not constrained to the sentence form, utterances employing metalinguistic if you like are given a semantic representation such that the if -clause does not contribute propositional content, yet they also maintain their status as conditionals as the sentence form gives rise to a potential conditional secondary meaning.

The acceptability of many syntactic constructions varies across speakers. In some, but not all cases, such interspeaker variation is due to speakers coming from different dialect regions. However, since neither dialect regions nor acceptability judgments are discrete, it is not always straightforward to determine when variation in a construction’s acceptability is geographically constrained. In this paper, I present one method of making this determination using the G i * statistic. I apply it to a variety of syntactic constructions, including presentative datives, so don’t I , be done my homework , intensifiers like wicked and hella , and do -support with have yet to sentences. Each shows a different geolinguistic profile, revealing different kinds of issues that quantitative analysis can shed light on. The results show fine-grained geolinguistic differences in syntactic acceptability judgments, demonstrating that the G i * statistic can be fruitfully applied to further our understanding of geographically-based variation in acceptability judgments.

Against a multidisciplinary background this contribution explores the areal typology of western Middle and South America. Based on a new language sample and a typological questionnaire that is specifically designed to bring some of the poorly documented and extinct languages into the debate, we explore the areal distribution of 77 linguistic traits in 44 languages. While one of the goals of the present article is to provide a general up-to-date view of the areal patterning of these traits on a large scale, we also explore a number of specific questions in more detail. In particular, we address the relationship between known language areas like Mesoamerica and the Central Andes with their respective peripheries, the possibility of detecting an areal-typological signal that predates the rise of these linguistic areas, and, finally, the question of linguistic convergence along the Pacific coast. We find that, while the languages of the Mesoamerican periphery are rather diffuse typologically, the structural profiles of the Central Andean languages are embedded organically into a more general cluster of Andean typological affinities that alters continuously as one moves through geographical space. In different ways, the typological properties of the peripheral languages may reflect a situation that goes back to time depths which are greater than that of the emergence of the Mesoamerican and Central Andean linguistic areas. Finally, while we can confirm typological affinities with Mesoamerica for some languages of coastal South America, we do not find support for large-scale linguistic convergence on the Pacific coast.