In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • Comparative Syntax of Balkan Languages
  • Olga Arnaudova
María Luisa Rivero and Angela Ralli, eds. Comparative Syntax of Balkan Languages. In the series Oxford Studies in Comparative Syntax. Cambridge: Cambridge University Press. 2002. Pp. 240. US$40.00 (softcover).

The eight papers comprising this volume bring together a considerable body of research concerning the syntax of languages traditionally argued to comprise the Balkan Sprachbund (Bulgarian, Greek, Rumanian, and Albanian). With this volume, the field of comparative Balkan linguistics is reexamined from the perspective of generative grammar (GB) including its recent developments known as the Minimalist Program (Chomsky 1995 and later).

Brian D. Joseph's paper, "Is Balkan Comparative Syntax Possible?", discusses the dichotomy 'language of the Balkans' (a geographic designation that takes into account any language on the Balkan peninsula) and "Balkan language" (a subset of languages distinguished by common features and displaying parallels traditionally attributed to language contact). This dichotomy (which draws on Schaller 1975) allows Joseph to distinguish between "comparative syntax of Balkan languages", which refers to general typology and can include any language in the Balkans, and "comparative Balkan syntax", which utilizes the concept of "Balkan language", initially developed by the proponents of the Sprachbund theory. Joseph further illustrates this difference on the basis of Balkan negation. The author discusses m-negators in Greek and Albanian (Modern Greek mi(n)and Albanian mos) and argues that they are contact-induced and thus should be treated as a 'syntactic Balkanism' in the traditional sense, while negative fusion (the fusion of the negative marker with the auxiliary, as in the Greek dhen(similar to n'tin English won't) is an epiphenomenon, best described in his view from the perspective of "comparative syntax of the Balkan languages".

The next three articles in the volume are dedicated to Balkan subjunctives (or "conjunctives", as labeled in traditional grammars of Bulgarian and Rumanian). Subjunctives have historically replaced infinitival structures in all Balkan languages and sometimes display an empty (null) subject also in the complement clause as in Prospθo na fiγo(Greek) ('I am trying to leave'). In some cases the subject of the lower clause obligatorily co-refers with the subject of the matrix clause as in a situation reminiscent of French and other Romance languages where the subjects of both clauses co-refer (a case of "control") when the complement clause is an infinitival. In other cases they are ambiguous between two readings as in O janis θeli na fiji(Greek) ('Janis wants to leave'/'Janis wants him to leave'), [End Page 107]or a disjoint interpretation (a case of 'obviation') when the complement clause is in the subjunctive.

While all the three authors agree that the so-called subjunctive particle ( nain Modern Greek, in Romanian, in Albanian, and dain Bulgarian) is the realisation of a mood head (cf. also Rivero 1994), they differ in their concrete proposals about the nature of this head and about the exact realisation of co-reference and control in Balkan. It emerges that Balkan subjunctives display some puzzling properties and pose a challenge to the traditional GB theory of control, subject raising and obviation. Dobrovie-Sorin notes that verbs in Balkan subjunctive complements are inflected for number, person, and tense and suggests that null subjects in subjunctive complements cannot be possibly identified as PRO but are rather comparable to reflexive pronouns bound by an antecedent. In her view, the selectional properties of the main verb will determine whether the empty element is a contextual or intrinsic anaphor. This hypothesis is formally supported by the claim that subjunctive particles incorporate to the verbal cluster and form a complex X constituent with the other elements in the verbal cluster, providing a "transparent" structure for co-reference. Anna Roussou develops a different line of analysis. Following Manzini and Roussou (1998), she approaches the peculiarities of Balkan subjunctives from a theory which views finiteness as a property of the complementizer system (cf. also Rizzi 1997) and thematic roles as features. According to her, in control and raising structures, the overt NP is base generated (i.e., directly merged) in the matrix clause (hence no movement from the embedded clause is...

pdf

Share