Language Research

Abstract The definition and typological description of the category of voice can be provided in two ways: i) at the level of semantic arguments (or semantic roles), ii) at the level of grammatical relations (or syntactic functions). The most important theoretical concept that is determined in terms of these two levels of representations and enables one to capture the rich variety of voices is that of diathesis. Diathesis is defined as a pattern of mappingsemantic arguments onto syntactic functions. In the present research, we explore the typology of diathesis and voice in Kurdish language on the basis of Kulikov’s work (2011). The results of the research show that Kurdish has these diatheses: passive (agentless, absolute and impersonal), conversive, 2/3 permutation, causative (transitive and intransitive), antipassive, reflexive, reciprocal, and anticausative. Of these, only passive, causative and conversive have morphological realization on the verb. Therefore, based on this analysis, one can say that Kurdish has only three voices: passive, causative and conversive.

Abstract This study explores the nature of relatively free word order of verbs and their arguments in Persian. Based on the theory of phases and multiple spell-out which is an instantiation of Minimalist Program, we attempt to account for the restrictions on these word order variations. Having a direct interaction with the syntax-phonology interface rather than a purely syntactic one, these restrictions, which have no interaction with information structure, are accounted for using linearization principle (Fox & Pesetsky 2005) and sentential stress rule (Kahnemuyipour 2004). The main generalization is that no argument that bears the sentential stress can appear on the right side of the verb. As such, nonspecific object of transitive verbs and nonspecific subject of unaccusatives and passives cannot follow the verb. This specific argument imposes no restriction on verb fronting which is in fact VP fronting in Persian.

Abstract The present study aims to examine the analysis of Persian adjective phrases (APs) within Role and Reference Grammar (RRG) and to appraise the merits and demerits thereof. In RRG, there is no independent AP; therefore, predicative adjectives are considered as PREDICATE inside NUCLEUS of the CLAUSE and attributive adjectives are regarded as modifiers within the periphery of NUCLEUS_N. Each type of adjective can, in turn, have its own modifiers as arguments or non-arguments. On the basis of this analysis, attributive APs are represented as predicative APs in a form which is very similar to relative clauses. The exclusion of AP from syntactic analysis enjoys prominent methodological and psycholinguistic advantages as such a theory demonstrates the principle of economy and offers more plausible explanations for first language acquisition as well as easier language processing. The application of RRG analysis to APs in Persian indicates that this theory can well account for such phrases in this language. This is best reflected in the effective analysis of split predicative APs, which serves another advantage for RRG. However, there are still such cases as the representation of split attributive APs which remain controversial thorough investigation.

Abstract This paper addresses the validity of two hypotheses concerning changes of vowel quality in Persian. The theory of centralization claims that there is a strong correlation between vowel duration and vowel quality. As vowel duration is shortened in unstressed contexts, the size of the vowel inventory reduces toward the center, leading to considerable overlap between different categories and a reduced, crowded acoustic space. In contrast, the dispersion theory of speech perception maintains that there is no correlation between vowel duration and vowel quality, and that vowels are arranged in the acoustic vowel space so as to minimize the potential for perceptual confusion between the distinct vowel categories. Changes in vowel quality were examined in unstressed syllables using duration and F1 and F2 spectral values. Results indicated that, contrary to the prediction of the centralization hypothesis, there was no strong correlation between duration and vowel quality differences, and that vowels displayed no significant spatial reduction in the direction of the center of the vowel space. The results are more consistent with the dispersion theory as they show that differences in quality in Persian are limited to small positions around the area in the acoustic space occupied by vowel categories. In light of our findings, we may well classify Persian with languages which do not possess the process of vowel reduction.

Abstract Deictic expressions belong to a category of linguistic expressions which seemingly require an extra-linguistic context for their referents to be recognized. They have referential, anaphoric and free functions. The aim of the present article is to answer a key question about these expressions: is it possible to offer one single explanation or model for all these functions? To do this, the paper first refers to the main perspectives and analyses suggested on the semantics of deictic expressions and discusses the advantages and disadvantages of each. It is shown that the major flaw in these analyses is their inadequacy in covering all uses of deictic expressions. Then, to solve the problem, a cognitive-discoursal model is proposed based on a concept called "discourse referent". It removes the borders between referential, anaphoric and free functions of those expressions. It is concluded that it is the introduction of these referents to the discourse and their establishment in the discourse that makes them referable by using deictic expressions.

Abstract This paper presents an attempt to analyze the verb phrase ellipsis in Persian in terms of the questions of identity and licensing. Assuming a minimalist framework for the analysis, it is argued that verb phrase ellipsis in Persian is structurally licensed by a governing head which takes vP as its complement, through an algorithm of E(llipsis)-feature deletion at PF. It is also shown that structural identity of the elided constituent and its antecedent is crucial for the derivation of verb phrase ellipsis in Persian.

Abstract Adopting a discourse-pragmatic approach, this study examines the synchronic multifunctionality of the discourse marker hâlâ from the perspective of (inter)subjectification. The findings obtained from the synchronic corpus of colloquial Persian indicates that hâlâ has proceeded along the (inter)subjectification cline identified by Traugott and Dasher (2002), Der (2010), and Degand and Vandenbergen (2011). In other words, the semantic change of hâlâ involves increased subjectivity and intersubjectivity. However, semantically, it is more complicated than other discourse markers because it has acquired presuppositional meaning which is beyond the (inter)subjective meanings related to the conventional components of discourse situation.