The causative alternation is a transitivity alternation between a causative verb and its anti-causative counterpart. Concerning the relationship between the two variants, two general approaches have been taken by linguists: derivational and non-derivational. In the derivational approach, it is supposed that one of the variants is basic and the other one derived. In the intransitive base approach (causativisation), the anti-causative is considered basic and the causative derived. In the transitive base approach (anti-causativisation), however, the causative is considered basic and the anti-causative derived. In the non-derivational or common base approach there is no direct relationship between the two variants but both come from a third source or root. The aim of this article is to critically evaluate the above-mentioned approaches in the light of morphologically marked alternating verbs in the Khū ri dialect __ a Northwest Iranian language spoken in the Central Desert of Iran. The morphological evidence in Khū ri shows that the derivational approaches to the causative alternation do not have a general analytic strategy and can only account for the morphological marking of the causative or anti-causative variant. On the contrary, the non-derivational common base approach covers the areas of variation in Khū ri (one of the two variants or both of them could be marked) and can explain this morphological variation inclusively.

According to causal hypothesis, the main factor in argument realization is causal relationship between participants. Based on this hypothesis, alternations can be accounted for, by difference in verbal profiles. In causative alternations, the difference between two constructions is due to the initial point of verbal profile: In causative variant, causer is located at the initial point but in anticausative variant, it is occupied with cause, because anticausative variant describes a situation in which the focus is on the change of causee. The manual study of 3602 data taken from Hamshahri Corpus II survey revealed that focusing on the change of causee is related to topicality. Causee which appears as subject is more topical than causer. data analysis, also revealed that in causative constructions, the topicality of causer and consequently focusing on the causality of it, induce causer to reveal as subject.

Focusing on the argument alternation of some Light Verb Constructions (LVCs) in Persian, we will show how a frame-based description of verbal polysemy may be used to answer some questions about syntactic behavior and argument structures associated with such verbs. In Persian, a number of LVCs consisting of a light verb and a preverb (PV) participate in causative alternation. The causative variant is formed with the LV ændaxtæn 'cause to fall', and the anti-causative variant is formed with oftadæn 'fall'. But in some cases, the anti-causative variant is not acceptable by native speakers of Persian, whereas the causative variant is acceptable. The peculiar behavior of these verbs in different contexts is explained in the framework of Frame Semantics (Fillmore & Atkins, 1992). We will argue that these verbs are associated with two semantic frames. Contrary to Karimi-Doostan (2005) who claims that the argument structure of LVCs is determined by PVs, and Folli, Harley & Karimi (2005) who propose that both PVs and LVs compositionally contribute to the A-structure of LVCs, using the notion of Frames (Fillmore & Baker, 2010), we propose that the whole construction including the LVC determines its A-structure.

The aim of this article is to study the holistic effect in locative alternation of Persian. In Persian a number of verbs denoting dislocation of a substance (locatum) to a surface or a container (location) participate in locative alternation (bar zadan ‘load’, pashidan ‘spatter’) The locative alternation in Persian is characterized as a special relationship between two syntactic realizations of the same verb, which are called the locative variant and the objective variant.In locative variant the locatum is realized as NP accompanied by “ra” and location as a prepositional phrase (PrepP). In objective variant the location argument appears as NP accompanied by “ra” and locatum as a bare NP just before the verb. It is shown that contrary to Roshan (1998) and Lazard (1982), realization of location as NP accompanied by “ra” is not necessarily associated with holistic reading. When location argument appears with “ra”, it is considered to be “more affected” than PrepP. Contrary to Anderson (1988) who claims that the partitive/holistic effect is associated with the alternate realization as NP or PrepP, we propose that the choice between coding the location argument as a NP or a PrepP amounts to a choice between coding it as the patient or not. The patient stands in a privileged relation to the process, and it is encoded as “primary affected” entity (NP). The entity realized as a PrepP is encoded only as a secondary affected – hence the location is less ‘affected’ (Langacker 1991). So holicity is seen as an instantiation of the ‘primary affectedness’ semantically associated with the patient in general and “ra” marks the argument which is coded as patient. Count locatums obligatorily appear with “ra” and are associated with holistic and partitve reading. Mass and generic locatums can appear with or without “ra”. When appeared without “ra” holistic reading is not possible since they are not bound in their instantiation domain. When appeared with “ra”, the “quantificational” function of “ra” bounds them in their instantiation domain- hence the possibility of holistic reading.

Object alternations involve different realizations of arguments, and are represented in three forms, including locative alternation, dative shift, and object-oblique alternation. Objects in object-oblique alternation are expressed in two different structures without fundamental changes in meaning. The second participant, in a transitive predicate, is either stated in a direct object position or in a preposition construction. The present paper investigates object-oblique alternation in Persian Aktionsart, based on Role and Reference Grammar. The objective of this study is twofold; firstly, to determine the im/possibility of object-oblique alternation in different classes of verbs, and secondly, to specify various readings of this alternation in verb classes consisting of states, activities, achievements, and accomplishments. The data are extracted from Persian newspapers, websites, and daily conversations. Data analysis denotes that some state verbs, including cognition, emotion, and possession verbs can be represented in object-oblique alternation. The alternate forms of cognition and possession verbs, refer to the kind of object in question. Alternation in sub-classes of activity verbs, that is, performance and consumption verbs, display total-partial interpretation. The alternative forms of creation verbs in this class, in the case of having a non-resultative verb, denote the kind of objects in question. Directed perception verbs, regardless of the parameter of object individuation, are represented in object-oblique alternation. Some achievement verbs may also be used in this alternation. Accomplishment verbs are rarely used in object-oblique alternation, except in cases where depending on the context, total-partial interpretation is intended. Generally speaking, the predicates that undergo object-oblique alternation share some characteristics; they are mostly non-state ones, their subjects are mainly agentive, and the objects are mostly low or non-affected. Furthermore, the verbs, which are more prone to object-oblique alternation, are chiefly non-resultative, and do not lead to fundamental changes in the object.

Being Causative/non-Causative is a kind of contrast in verbs that is reflected in the syntactic structures and changes their meanings. This article investigates the three different categories of Causative construction in Tāti and decides, based on the findings of Functional Linguistics, that being Causative/non-Causative is not a definite classification, but a continuum in which each verb finds its own place according to the agent of the event and its being intentional or unintentional.

Our purpose in this article is to investigate the formal mechanisms of causative constructions of Modern Persian language based on Dixon (2012) typology. In this diachronic study, a corpus of more than 330000 words and 4151 causative constructions extracted from 30 books of 4th to 7th and 13th and 14th AH centuries. The analysis of Persian causative constructions shows that this language uses three known lexical, morphological, and syntactic mechanisms in order to express the causative meaning so that lexical mechanism has the highest occurrence frequency among them. The data shows that syntactic causative application is restricted to the recent centuries so that no construction of this type has been found in early centuries. Our findings also prove that 4th century has the most frequency of causatives and we can see a decreasing rate of these constructions so that 13th century has the lowest position and again we observe an increasing rate in the current century. The results also indicate that morphological causatives have had a decreasing frequency so that some verbs like "faramushandan", "shanavandan", "agahandan" and… have been omitted from speakers’ vocabulary. Another result is that although existence of a causative verb is necessary for a construction to be regarded as a causative, it is not enough.