- Formal studies in Slovenian syntax: In honor of Janez Orešnik ed. by Franc Lanko Marušič and Rok Žaucer
Linguistics is a prism: any given language in its entirety can be refracted through it to reveal an astounding spectrum of properties. And these properties may prove to be unique or universal, but our understanding of how this prism must work is enhanced by every finding. Few have understood this better than Janez Orešnik, to whom this volume is dedicated, and whose work largely focused on Slovenian and Icelandic. He explored many facets of these languages: diachronic development as well as synchronic phenomena, from Icelandic epenthesis to Slovenian verbal aspect. Professor Orešnik's work is far-reaching in its scope and widely influential. This volume is a fitting tribute for a scholar with broad interests whose work has influenced and inspired several generations of linguists.
The first paper of the volume, "On Second Position Clitics Cross-Linguistically" by Željko Bošković, exemplifies the intentions of Marušič and Žaucer (as stated in their introduction) in that it is both informed by and a contribution to the study of Slovenian syntax but that it also draws from a much wider body of knowledge, illustrating clearly the integration of Slavic (and Slovenian) linguistics into the broader scheme of general linguistics.
In this paper, Bošković examines one of the many generalizations related to the NP/DP parametric division (Bošković 2008, 2012), namely that only NP languages can accommodate second-position clitic systems. Central to his argument, of course, is the nature of the difference between NP and DP languages.
DP languages are defined as those that contain a definite article, which are characterized by three basic properties: (1) they are unique (i.e., morphologically distinct from demonstratives and other determiners), (2) they occur maximally once per traditional NP, and (3) they confer a definiteness interpretation onto their complement NP. This precise definition forms the basis of exclusion of many languages that were previously analyzed as having definite articles, casting them instead as NP languages. For example, this exclusion extends to colloquial Slovenian, which has been argued (notably by Marušič and Žaucer (2014)) to contain a definite determiner ta. According to the definition [End Page 143] given here, ta violates both the morphological uniqueness condition and the single-instance-per-NP condition and thus does not qualify as a definite article.
An NP language is distinguished from a DP language by two related features: (1) while it may have an indefinite article, it must not have a definite article (as defined by the conditions above), and (2) bare NPs can be interpreted as definite.
This paper is an investigation of the intersection of the DP/NP parameter and another parameter, namely, the second-position clitic parameter, which Bošković sets up as dichotomous between languages that have second-position clitics and those that do not. He readily admits that the latter category is diverse, but he focuses in on languages that have verb-adjacent clitics. The reader must then keep in mind that the analysis that is given, while capable of handling the data from these two language types (2nd position and V-adja-cent), may not extend to all clitic types cross-linguistically.
Bošković offers diachronic evidence from Greek to support the generalization that only NP languages accommodate second position clitics. In Homeric Greek, an NP language, clitics occurred overwhelmingly in the second position. But by the time Koine was being spoken, by which time a definite article had emerged, there was no longer a system of 2nd position clitics. Perhaps even more compelling is the variation between two dialects of Ossetic, Iron and Digor. The dialects differ in part due to the presence of a definite article in one (Digor) and its absence in the other (Iron). This correlates with another difference: Iron displays 2nd position clitics while Digor does not.1
In the subsequent sections of the paper...