This article discusses naming constructions in Estonian dialects, which express what something, linguistically realized as an object argument, is called. The construction consists of a naming verb in the impersonal voice, an object, which is named somehow, and a complement, which expresses what the object is called (i.e., Something [OBJ] is called [V] something [COMPL]). All the three elements of the construction can vary in spoken dialect material, where the object argument and the complement can also be omitted in some cases. The goal of the article is to ascertain whether the variation in the naming constructions is influenced by dialectal differences or rather by intraconstructional factors, which in turn may be linked to the spoken form of communication. The dataset, consisting of 905 naming constructions, was extracted from the Corpus of Estonian Dialects, annotated for verb, object, complement, the occurrence of the connector et ‘that’, and the dialect area, and analysed by using conditional inference trees (Breiman et al. 1984; Hothorn et al. 2006) and correspondence analysis (Greenacre 2006, 2007). The combination of the two methods was chosen to reveal as many of the associations and interactions between the variables as possible. The results showed that a surprising amount of the variation can be explained by the dialect area in which the construction is used. The lexical variation in the construction, i.e., the choice of the finite verb, displayed the clearest link to areal distribution: the southern dialects (Mulgi, Tartu, Võru, Seto) and the adjacent Eastern dialect showed a clear preference for the verb kutsuma ‘to call; to invite’, whereas the northern dialects preferred the verb hüüdma ‘to call; to shout’. The verb hõikama ‘to call (out); to shout (out)’ was almost exclusively used in the Mulgi dialect (spoken in southern Estonia), and the verb nimetama ‘to call, to name; to mention’ associated more with Alutaguse dialect (the dialect spoken in the north-easternmost part of Estonia, next to the Russian border). Not surprisingly, lexical variation is where the traditional division of Estonian dialects to northern and southern group applies. The choice of the object argument marking was also primarily dialectally determined, but the dialects were divided rather on the eastern-western axis: the eastern dialects (Alutaguse, Coastal, Eastern, Võru, Seto) were more prone to mark the object in the translative case, which is also the canonical case for the object in the naming construction in standard written Estonian, whereas in the western dialects (Insular, Western, Mid, Mulgi) and in Tartu dialect, the most frequent case for object marking was the nominative. This is in accordance with previous studies on Estonian dialects (Lindström forthcoming), according to which in the western dialects the nominative tends to acquire the functions of the partitive case in the contexts where the opposition of the two cases is not semantically relevant. The omission of the object argument does not depend on the dialect area, but is rather connected with information structure and the universal principle, according to which the information that is known and/or can be predicted does not need to be explicitly expressed (Givón 2017: 3). Complement marking does not show clear dialectal differences, but is instead affected by the choice of object argument marking. Although it is the nominative case that is, somewhat surprisingly, by far the most frequent option for complement marking in the dialect corpus data, the expected translative case is slightly more prominent, when the object is also marked as one would expect it to be in standard Estonian, i.e., in the partitive case. The abundant use of the nominative case can be explained by the specifics of spoken language, where the complement is typically used as a citation and is therefore less integrated to the verb’s argument structure. Compared to the omission of the object argument, the omission of the complement was rare, which is not surprising: the complement entails new and therefore the most important information and omitting it is only possible in somewhat exceptional cases, such as self-repairs. It is up to future research to determine whether or not the variation in the form of the naming constructions is also tied to additional factors not taken into account in this study. Such factors include word order (i.e., the position of the complement and the object in relation to the finite verb), and the principles of information structure, which might have an even greater effect also on the omission of the arguments.
The focus of the article is the dialectal-typological analysis of Perfect Simple/PS (Ro. ‘perfectul simplu’) in Romanian dialects, i.e. Bobiceşti/Olt/Wallachia and Almaş/Arad/Transylvania. Paralleled by the main literature-based characteristics of the tense, the data do not only emphasize the surprisingly preponderant distribution of PS in Transylvania but they also reveal the key status of the Perfect Simple as part of the Romanian dialect use as well as its typologically relevant double status (Standard Romanian versus dialect use). The comparison between the ‘general purpose’ Latin-to- Romance PS evolution and the ‘Balkansprachbund’ Greek-Aorist characteristics ends with commentaries on the double perspective of Romanian: Wallachian (generally reflected in Standard Romanian) manifests Balkan-oriented phenomena and Transylvanian preserves Latin/Romance-oriented features.
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.