---

This research examines the lexical processing of head-first endocentric compounds in Persian based on semantic transparency. This study shows whether the processing is related to the morpheme-based, whole-word, or dual-root models. Moreover, this research concludes whether the processing is associated with sublexical or supralexical models. Two separate masked priming tasks are conducted. In the first experiment, the compound word is the primed word and the modifier is the target. In the second experiment, the prime is the compound word and the target is the head. Three conditions are considered: transparent, opaque, and orthographic-overlap. In order to analyze the result, mixed models are used. The results show significant priming effects for transparent and opaque conditions. While no priming effect is shown for the orthographic-overlap condition. It can be concluded that the lexical processing of head-first endocentric compounds is independent of semantic transparency. The processing is morpheme-based and it is on the basis of the sublexical model. Furthermore, the semantic processing of these words is supralexical. Moreover, the lack of priming effect for the orthographic conditions is seemingly associated with orthographic complexity in the Persian writing system.
 

---

A number of linguistic studies on compounding have acknowledged that due to the existence of head element, endocentric compounds are semantically compositional and transparent. The current study aims to focus on the semantic aspect of Persian endocentric compounds to show that: 1) in some cases, the existence of head element does not entail the semantic compositionality and simplicity of conceptual structure in endocentric compounds, and 2) compound words which are categorized as endocentric compounds differ in terms of the complexity of conceptual structure. Considering that the ability of Conceptual Blending Theory (Fauconnier & Turner, 2002) to describe meaning construction in compound words has been previously approved by some cognitive linguists, this theory has been applied to analyze meaning construction in a number of Persian endocentric noun-noun nominal compounds. A close analysis reveals that although some endocentric compounds are semantically compositional and prompt for simplex networks, there are endocentric compounds which are not semantically as transparent and compositional. In this kind of endocentric compounds, metaphor or metonymy has affected the modifier element, thus meaning construction triggers single-scope networks which are neither as complex as double-scope networks nor as simple and compositional as simplex networks. This result implies the diversity of conceptual structure in words which are defined as endocentric compounds..
 
 1. Introduction
Compounding is a common word-formation process almost in all languages (Dressler, 2006: 23), for forming new lexical units by merging two or more pre-existing lexical units. Based on the most pervasive and traditional semantic categorization of compound words, as proposed by Bloomfield (1933), there are two main groups of compounds, namely endocentric and exocentric compounds. In endocentric compounds, one of the constituents plays the role of head and a hyponymic relation is observable between the compound and the head element. Numerous prior studies have acknowledged that, due to the existence of head element, the semantics of endocentric compounds is transparent and compositional. However, no study has been made yet to specify whether the presence of head element leads necessarily to the semantic simplicity of conceptual structure in endocentric compounds.
Given that conceptual blending theory can account for the meaning construction process of compound words (cf. Coulson, 2001; Fauconnier & Tuner, 2003; Benczes, 2006; Schmid, 2011), the current study applies the network model of conceptual blending to explore the semantics of Persian noun-noun nominal compounds in order to show that: 1) in some cases, the existence of head element does not entail the semantic compositionality and simplicity of conceptual structure in endocentric compounds, and 2) compound words which are categorized as endocentric compounds differ in terms of the complexity of conceptual structure.
 
2. Literature Review
The semantics of compounds has been one of the most challenging aspects of compound words being investigated in numerous studies. These studies cover many areas of investigation including semantic transparency and opacity (e.g., Afrashi, 2000; Sabzevari, 2013; MirEmadi & Majidi, 2007), headedness and the position of head (e.g., Tabatabaei, 2004; Khabbaz, 2008), the relation between constituents (e.g., Tabatabaei, 2013; Sabzevari, 2012, 2018), and the role of metaphor (or metonymy) in the meaning of compounds (e.g., Estaji & Ghanun, 2009; Torabian, 2013).
Besides, through the development of conceptual blending theory, several studies have analyzed compound words within the integration network model of conceptual blending (e.g., Turner & Fauconnier, 1995; Sweetser, 1999; Coulson, 2001; Fauconnier & Turner, 2003; Benczes, 2006). 
 
3. Methodology
To the aim of collecting noun-noun nominal compounds in Persian, Sokhan Comprehensive Dictionary (2003) was consulted and a total number of 694 compounds were extracted from the dictionary. As a next step, endocentric compounds were taken out from the collected data (a number of 372 endocentric compounds) and the possible action of metaphor and/or metonymy upon their meaning was examined. This procedure led to 216 non-metaphorical and non-metonymical compounds, 137 compounds with metonymical modifier, and 19 words with metaphorical modifier. Following this, considering the frequency of identified patterns, we randomly analyzed a number of compounds from each pattern based on the network model of conceptual blending theory.  
 
4. Results and Discussion
The blend analysis of Persian endocentric nominal compounds reveals that, besides semantically compositional endocentric compounds, such as âb-anbâr, which trigger simplex network and their modifier is linked through a role-to-value relation to their head element, there are several endocentric compounds that show a slightly more complex conceptual structure. In the meaning construction of non-compositional endocentric compounds, such as taxte-sang and sofre-mâhi, while the input space corresponded to the head element contributes literally to the blended space, the input space corresponded to the modifier is projected metaphorically or metonymically. This group of endocentric compounds prompt for single-scope networks which are neither as complex as double-scope networks nor as simple as simplex networks.
5. Conclusion
The current study found that the existence of head element does not necessarily guarantee the semantic simplicity and compositionality of endocentric compounds. In a number of cases, metaphor or metonymy acts upon the modifier element of endocentric compounds, therefore the meaning construction triggers single-scope networks which are not as simplex and compositional as simplex networks. In other words, not all endocentric compounds are semantically simple and compositional. Instead, there are several endocentric compounds that are not fully compositional, that is, they have a slightly more complex conceptual structure. This implies that drawing a strict boundary between endocentric and exocentric compounds, based on compositionality and semantic transparency, is not reasonable, because there are endocentric compounds which, like exocentric compounds, are metaphorical or metonymical. This study suggests a continuum to show different degrees of complexity of compounds. While non-metaphorical and non-metonymical endocentric compounds lie at the one end of the hypothetical continuum, endocentric compounds with metaphorical or metonymical modifier are nearer to the other end of the continuum where compounds with metaphor- and/or metonymy-based head and modifier lie.
 

---

In the present article, different types of semantic relationships between the components of endocentric compound words in Gahvarehei Kurdish are investigated. The corpus of the research data includes endocentric compounds that have been obtained from the everyday speech of native speakers of Gahvarehei Kurdish, which is one of the dialects of the southern Kurdish branch and is common in Gahvareh section in Dalahoo County, Kermanshah province, Iran. The results of the research show that endocentric compound words in Gahvarehei Kurdish in terms of structure include: noun-noun head-initial, noun-noun head-final, noun-adjective, adjective-noun and preposition-noun. Various semantic relationships between the components of these structures can be found, such as origin, material, gender, content, possession, use, type, dependence, similarity, instrument, kinship, manner, time, amalgamation, taste, physical condition, age, function, production tools and place. The research findings, which include a list of semantic relationships between the components of endocentric compound words in Southern Kurdish, can be a complement for other lists provided by researchers in other Iranian languages ​​as well as other languages ​​in the world and lead us to a comprehensive classification of different types of semantic relations in compound words.

1. Introduction
Compounding is one of the most important word-formation processes in many languages of the world. In Kurdish, the process of composition, along with derivation, is one of the two main and most widely used processes of word formation. The discussion of semantic relationships between the constituent components of compound words and the presentation of lists or regular classifications of them is one of the important topics in the studies of the field of word formation, especially in the case of endocentric compound words. Our goal in this study is to provide a comprehensive classification of the types of semantic relationships between the components of the noun-noun, noun-adjective, and adjective-noun compounds in Gahvarehei Kurdish. Accordingly, the main question of this study is what are the types of semantic relations in endocentric compound words in Gahvarehei Kurdish?
The research data consists of endocentric compounds collected from the daily speech of Kurdish speakers residing in the city of Gahvareh (Dalahoo County, Kermanshah Province, Iran). The common language type in this region is considered to be one of the southern Kurdish dialects in terms of linguistic characteristics, and it is closely related to the Kalhori dialect, but it has differences with it in terms of phonology and vocabulary.

2. Literature Review
To date, some studies have been conducted on the processes of word formation, and in particular the discussion of compounding and compound words in various Kurdish dialects and varieties, among which the following can be mentioned: process of repetition in Kalhori (Jahanfar, 2017); compound verbs in Kalhori dialect (Afshar & Zarei, 2017); comparison of word-formation processes in Persian and Kurdish languages (Jendo Ismail, 2018); and word-formation processes in Ilami Kurdish dialect (Sherafat, 2020). However, the discussion of the semantic relationship between components of compound words has been neglected and no study has been conducted in this field. Therefore, we will review studies that have been conducted on the endocentric compound words in Persian and other Iranian languages.
Sabzevari (2010) based on CARIN theory (Competition among Relations in Nominals) has investigated the relationship between the components of Persian compound nouns and the role of this relationship in the semantic classification of these words. He points out that the semantic interpretation of a compound word depends on the relationship between its components and the descriptor of the compound noun (i.e. the non-head element) determines the type of relationship between the head and the dependent. Asi and Badakhshan (1389), after examining the various characteristics of compounding that distinguish compound words from syntactic phrases, have analyzed the types of compounds in Persian. After discussing the types of existing classifications, they finally considered Scalise & Bisetto’s classification (2009) as the most complete classification, and based on that, they divided the types of Persian compounds into three main groups: subordinate, attributive, and coordinate, each of which can be exocentric or endocentric. Following Costello and Keane (2000), Sabzevari (2019) considered three types of semantic relations in Persian nouns: 1) relational compounding; 2) Property compounding; 3) Conjunctional compounding.

3. Methodology
The present research was carried out using a descriptive-analytical method and the data collection method was such that the researcher wrote down every compound word that he encountered in his family environment, work environment, and surrounding community in three months (in total 820 cases). Then, based on a criterion called the ‘hyponymy test’, endocentric compound words were distinguished from other types of compound words. In total, 345 of the collected compound words were of the endocentric type. Finally, the obtained words have been analyzed and classified in terms of the type of semantic relationship between their components based on the approach of Adams (2001) and Moldovan et al. (2004).

4. Results
The results of the study show that all compound words in Gahvarehei Kurdish are formed around the core axis of the noun category but other lexical categories such as verbs, adjectives ، adverbs and so on cannot be in the head position of these words. Dependents can be a noun, adjective, or preposition, and somehow modulate the meaning of the head and limit its inclusion. Therefore, these types of combinations are structurally divided into four types: noun-noun, noun- adjective, Adjective – noun, and preposition – noun. The diversity of semantic relationships between the components of compounds leads to a range of meanings in them. The types of relationships identified in each of the structures examined in this study are as follows:

Table 1
Semantic Relationships in Endocentric Compounds of Gahvarehei Kurdish

Head-initial NOUN- NOUN	origin, gender, content, ownership, usage, type, dependence
Head-final NOUN- NOUN	similarity, instrument, kinship, origin, place, manner, time, content, type
NOUN- ADJECTIVE	color, taste, relationship, physical state, age, gender, sex, function, means of production, origin
ADJECTIVE - NOUN	gender, size, age, physical state
PREPOSITION - NOUN	location

The findings of the research, which includes a list of semantic relationships between components of endocentric compound words in Gahvarehei Kurdish, along with the findings of Iranian researchers such as Sabzevari (2011) and Karimidostan and Vahid (2015) can lead us to provide a comprehensive classification of the types of semantic relations in compound words in Iranian languages. These findings can also complement the lists provided by researchers of other world languages (such as Adams, 1973; Ryder, 1994; Costello and Keane, 2000; Adams, 2001; Moldovan et al., 2004; Estes and Jones, 2008; Szubert, 2012).
 

---

Compounding is one of the most common and also very productive word formation processes among languages in the world. Because of its productivity and high frequency, it has precisely been studied by many Iranian and non-Iranian linguists. However, this productive process has not yet been studied in Mazandarani. In the present study, first Mazandarani compound words taken from Tabari dictionary, have been analysed in the form of a table and based on the criteria such as syntactic category, semantic classification of compound types (endocentric, exocentric and dvandva), the position of head and being verbal or nonverbal. By taking into consideration the above mentioned criteria and based on three fundamental concepts, namely, head, dependent (non-head) and feature percolation, the resulting compounds have been analysed separately. At the end of the article, after studying and evaluating the hypotheses, appropriate explanations have been presented. The high frequency of “N+ N → compound noun construction” refers to their predictability in meaning compared to other constructions. It means that there is a direct relationship between high frequency of a word formation process and its semantic predictability. Also, the reason why endocentric compounds are more frequent than others is due to the fact that both constituents of the compound are used in their original meanings. Hence, endocentric compounds are unmarked elements and are easier to learn in the process of language learning. The above mentioned findings are in line with markedness theory. That is why the frequency of occurrence of unmarked elements is higher than marked ones which is related to language universals.