Volume 53, Issue 3 -

Previous research has shown that givenness influences adults’ syntactic choices with alternating dative verbs like give . For example, adults show a preference for given-before-new ordering with the two postverbal arguments of dative verbs. This paper considers whether givenness similarly affects child datives. In two elicited production studies with four-year-olds and a control study with adults, the discourse that participants heard before producing a dative construction was systematically varied. These studies indicate that (i) like adults, preschoolers tend to produce given-before-new argument ordering, (ii) givenness does not affect all verbs and arguments equally, and (iii) the givenness effect on word order can be largely, but not fully, attributed to the influence of givenness on referring expressions. In other words, participants showed a strong tendency to pronominalize given information and to mention pronouns first (short-before-long). These findings point to a high degree of continuity between child and adult language. Like adults, children attend to and integrate cues from form, function, and discourse as they process language for production.

This paper offers the first empirical and theoretical account of an NP construction referred to as that noun thing (TNT) in English. I argue that the construction is a Langackerian reference point construction, with the basic use of referring in situations in which speakers find it difficult to characterize the referent in question. I also show that speakers systematically rely on this relational means of referring in specific contexts to convey a range of Gricean conversational implicatures such as speaker disapproval, which are above and beyond the conventional meaning of the construction. The TNT thus offers a clear space from which to view the interplay of conventional meaning in cognitive and construction grammars with classic Gricean pragmatics.

In this paper we argue that the notions generally grouped together under the heading of evidentiality actually belong to four different evidential subcategories, which are different from one another in terms of their semantic scope. The hierarchical, scopal architecture of Functional Discourse Grammar is used to define these four categories. After giving our arguments for this new classification, we test a number of predictions that follow from it concerning the coexistence of evidential subcategories within a language and the co-occurrence of evidential markers in a single clause. We investigate our predictions in a sample of 64 native languages of Brazil. The data from these languages show that the presence of one or more of the four evidential subcategories can be systematically described in terms of an implicational hierarchy.

Words in German show several instances of a seemingly optional schwa-zero alternation, both in relation with inflected forms as well as in the final position of stems and simplex words, as in des Tag(e)s ‘the day, gen. sg.’, or gern(e) ‘gladly’. The present paper proposes that the (non-)appearance of schwa is partially governed by a hitherto unknown prosodic parallelism: the schwa-containing form (a branching trochaic foot) is preferred whenever a neighboring word is of the same trochaic shape; and vice versa: the schwa-less form is found adjacent to another monosyllabic form. In other words, adjacent feet are required to have identical structure, binary branching (bisyllabic) or non-branching (monosyllabic). Large-scale corpora are used as the main source of evidence for the verification or falsification of the hypothesis. A diverse set of nouns and adverbs involving schwa-zero alternations were studied in appropriate phrasal contexts, both from present-day Standard German and from Early New High German. Based on comprehensive corpus counts, these phrases are tested for the hypothesis of prosodic parallelism. A series of chi square tests and a generalized linear model with mixed effects demonstrate statistically that the prosodic shapes of the target word and its adjacent form are not independent of each other. The focus of the paper is on empirical evidence for Prosodic Parallelism as a new type of prosodic constraint. The relevance of this constraint for the persistence of variation over a long period of time is discussed as well.

Most theories of scalar quantifiers, of whatever persuasion, assume a lexical lower-bound-only, ‘at least’ meaning for scalar quantifiers, offering pragmatic or grammatical mechanisms for deriving the upper bound (Carston 1990; Chierchia 2004; Horn 1972 and onwards). I have challenged the lower-bound analysis in Ariel (2004), proposing instead a circumbounded analysis for quantifier most , where the upper bound too is lexically specified. I here extend the analysis to some . The most important feature of the circumbounded analysis is that it splits into two what are commonly considered one and the same interpretation of ‘less than all’. The first is a lexeme-level upper bound which asserts the speaker’s commitment to the existence of some proper subset , the reference set, for which the predicate holds. The second is pragmatic, an exclusion of the complement set from the predication. Based on new questionnaire data, my main argument here is that even in contexts which clearly militate against a ‘not all’ interpretation, a nonmaximal upper bound is understood. More generally (and tentatively), I will cast doubt on the whole Gricean idea that linguistic semantics should be reduced to logic.

The article deals with the phenomenon of lability (ambitransitivity), in other words, the ability of a verb to be either transitive or intransitive. I analyze the historical development of verbs which are currently labile in modern Russian. The main group of Russian labile verbs includes verbs of motion. On the basis of corpus and dictionary data, I conclude that the behavior of the lexemes under analysis is far from being uniform. However, interestingly, for most of them, e.g., lit’ ‘flow/pour’, gonjat’ ‘run, drive/chase, pursue’, and kružit’ ‘go round, turn, roll’, the proportion of the intransitive use grows throughout the period under analysis, though for the verb kapat’/kapnut’ ‘drop, pour in drips’, in contrast, the transitive use becomes more and more frequent. In the cases when the intransitive use becomes more frequent, the semantic change matches the statistical one. In the beginning, verbs of this subtype were only used intransitively in a restricted type of contexts (e.g., for gonjat’ ‘run, drive/chase, pursue’, hunting contexts represent this restricted class), where the intransitive use might be a result of object omission. Later on, the semantic range of the intransitive use became wider and the lability was no longer semantically related to object omission. I conclude that the emergence of an intransitive use of a strictly transitive verb or the increase of the number of intransitive uses is often accompanied by semantic changes: first, a verb is used intransitively in a very restricted range of contexts (for the verb gonjat’ these include hunting contexts). Then, the range of contexts and the semantics of the intransitive use becomes wider, the number of intransitive uses grows and the verb becomes canonically labile with both uses being equally or almost equally frequent. Importantly, the borderline between A- and P-lability is not as strict as it is put sometimes: P-lability (the causative/non-causative alternation) can in some cases be traced back to A-lability (object omission).