Analyzing Polysemy in "zædæn" Light Verb Constructions: A Constructional Morphology Perspective

 Introduction
Compound verbs represent a cornerstone of modern Persian, serving as a primary mechanism for lexical innovation and coping with the need to express novel concepts. These structures, typically formed by combining a non-verbal element (a noun, adjective, or prepositional phrase) with a semantically light verb, are highly frequent and productive. Among these light verbs, "zædæn" (prototypically meaning 'to hit' or 'to strike') stands out for its extraordinary versatility and frequency. Its semantic range is vast; foundational dictionaries like the Sokhan Dictionary (Anvari, 2002) list up to 71 distinct meanings for the verb. Its core, prototypical meaning involves the application of physical force, conceptually built around a triad of elements: an agent (the hitter), a patient (the entity hit), and an instrument or manner (the type of contact).
This study moves beyond traditional, purely lexical analyses to adopt the framework of Construction Morphology (Booij, 2010a, 2010b). This theoretical approach allows for a holistic analysis where the compound verb is treated as a construction—a stored pairing of form and meaning. The model posits a network of schematic constructions that license the formation of new words, utilizing key concepts such as constructional schemas, subschemas, default inheritance, and motivation. Our primary objective is to leverage this framework to systematically categorize the extensive array of "zædæn" compound verbs, map the intricate polysemous network that connects them, and analyze the cline of semantic bleaching that "zædæn" undergoes, from its concrete core meaning to highly abstract and grammaticalized functions. This analysis aims to provide a more coherent understanding of Persian word formation, with potential applications in pedagogical contexts for learners of Persian and in computational linguistics for natural language processing tasks.
Background
Construction Morphology is a word-based, constructionist model that sits within the broader paradigm of cognitive linguistics. It challenges strict modularity by arguing that morphological patterns are best represented as constructions—conventionalized form-meaning pairings—organized in a hierarchical lexicon (Jackendoff, 2007; Culicover & Jackendoff, 2005). Its formal foundations were significantly advanced by Booij (2010b), building on earlier type-based morphological work by Riehemann (1998).
In Persian linguistics, the application of Construction Morphology has gained considerable traction over the past decade. Following its introduction by Rafiei (2012), scholars have successfully deployed it to analyze a wide range of phenomena, including derivational suffixation (e.g., Bamshadi & Ghatreh, 2017; Bamshadi et al., 2018), prefixation (Bahrami-Khorshid & Namdari, 2019), and various compounding patterns (e.g., Karimi et al., 2023; Hasanpouran & Alizadeh, 2023). Crucially, this framework has proven particularly effective in handling polysemy, a central feature of the Persian verbal system. Previous work has already demonstrated its utility in analyzing other high-frequency light verbs like "gereftan" (to take) (Razmdideh & Hassan-Shahi Raviz, 2021). This study seeks to extend this successful line of inquiry to the complex and pervasive case of "zædæn".
 
Method
This research is grounded in a descriptive-analytical methodology. Data was systematically gathered from a combination of authoritative lexical sources and modern language corpora. The primary sources included the Sokhan Dictionary (Anvari, 2002), the Zansou Dictionary (Keshani, 1993), the Dictionary of Syntactic Valency of Persian Verbs (Rasooli et al., 2011), and the specialized Corpus of Persian Light Verb Constructions (Es'haghi & Karimi-Doostan, 2022). To capture neologisms and contemporary usage, online Persian dictionaries and language resources were also consulted. This process yielded a robust dataset of 330 distinct compound verbs featuring "zædæn".
Following collection, the data underwent a purification phase to remove non-compound entries. The remaining verbs were then subjected to a conceptual categorization, informed by existing dictionary classifications (e.g., Anvari, 2002; Family, 2006) and refined by native speaker linguistic intuition. The core of the analysis was conducted within the Construction Morphology framework. We posited a general constructional schema for Persian compound verbs, [X]I [zædæn]j[X]i[zædæn]j_Vk, and proceeded to identify the specific macro-schemas and their subsequent subschemas that are instantiated by the light verb "zædæn". The resulting network of schemas, organized from more abstract to more specific levels of description, is presented through detailed diagrams that visualize the hierarchical relationships and polysemous links within the data.
Results
Our analysis reveals that the semantic landscape of "zædæn" compound verbs is not chaotic but highly structured, organized into six primary macro-constructional schemas. Each schema is formally represented by a constructional template that pairs a specific morphological form with a core semantic value. The variable X in the nominal/adjectival slot (N/A) and the semantic variable SEM_i are co-indexed, meaning the specific meaning of the compound is a function of the element that occupies the X position.
The six identified macro-schemas are as follows:

The Impact Schema:
This schema generates verbs where an agent volitionally acts upon a patient, causing a change of state. Its formal representation is:
<[[X] N/Aᵢ [zædæn]ⱼ]Vₖ ↔ [Putting an effect related to SEMᵢ]ₖ>
This overarching schema branches into several subschemas, including:

Damaging/Attacking (e.g., šamšir zædæn → to strike with a sword; hile zædæn → to deceive),
Transition/Modification (e.g., jowš zædæn → to boil),
Trimming (e.g., riš zædæn → to shave),
Partitioning (e.g., parde zædæn → to curtain off),
Adding (e.g., namak zædæn→ to salt).

The Emission/Spreading Schema:
This category covers verbs meaning to emit, disseminate, or spread something, often sensory outputs. Its formal representation is:
<[[X] N/Aᵢ [zædæn]ⱼ]Vₖ ↔ [Spreading related to SEMᵢ]ₖ>
Key subschemas under this macro-schema include:

Visual Emission (e.g., barq zædæn→ to flash),
Auditory Emission (e.g., sut zædæn → to whistle),
Onomatopoeic Sounds (e.g., qor-qor zædæn → to gurgle),
Human Vocalization (e.g., faryad zædæn → to scream),
Musical Sound Production (e.g., gitār zædæn 'to guitar-hit' → to play the guitar).

The Transfer/Penetration Schema:
Verbs in this schema involve using an instrument to transfer a substance into an entity or to penetrate a surface. Its formal representation is:
<[[X] N/Aᵢ [zædæn]ⱼ]Vₖ ↔ [Transferring/Penetrating related to SEMᵢ]ₖ>
This schema includes subschemas for:

Substance Administration (e.g., mādd zædæn→ to do drugs),
Injection (e.g., āmpul zædæn 'to ampoule-hit' → to inject),
Loading/Refueling (e.g., bensin zædæn 'gasoline' → to refuel).

The Rapid Action Schema:
This schema captures quick, instantaneous, and often momentary actions, which are highly prevalent in colloquial speech. Its formal representation is:
<[[X] N/Aᵢ [zædæn]ⱼ]Vₖ ↔ [Quick action related to SEMᵢ]ₖ>
Subschemas under this macro-schema include:

Creation/Establishment (e.g., māgāze zædæn → to set up a shop),
Preparation (e.g., māye zædæn → to add yeast),
Quick Measurement (e.g., tæxmin zædæn → to estimate),
Communication (e.g., telefon zædæn → to telephone/call).

The Automatic Movement Schema:
This category describes dynamic, often repetitive and involuntary motions. Its formal representation is:
<[[X] N/Aᵢ [zædæn]ⱼ]Vₖ ↔ [Automatic movement related to SEMᵢ]ₖ>
This is realized through subschemas such as:

Rhythmic Movement (e.g., par zædæn 'to feather' → to flap wings),
Excitatory Movement (e.g., čarx zædæn → to spin; zānu zædæn → to kneel).

The Theft Schema:
This is a small but distinct and productive schema with no further subschemas. It specifically generates verbs meaning to steal or take illicitly. Its formal representation is:
<[[X] N/Aᵢ [zædæn]ⱼ]Vₖ ↔ [Stealing related to SEMᵢ]ₖ>
Here, X is typically a noun denoting the source or location of the theft (e.g., jib zædæn→ to pickpocket; bānk zædæn → to rob a bank).

A key finding that cuts across all schemas is the cline of semantic bleaching that the light verb "zædæn" undergoes. It can retain a strong sense of its original meaning of force and impact (e.g., in the Impact schema) or be almost entirely delexicalized, serving primarily to convert the non-verbal element into a predicate while often imparting a nuanced meaning of suddenness, speed, or efficacy, as is most evident in the Rapid Action and Automatic Movement schemas. This continuum confirms earlier observations (e.g., Mansouri, 2013) and highlights the grammaticalization path of this high-frequency light verb.
 
 
 
Figure 1: "Six macro-constructional schemata of Persian compound verbs with the light verb 'zædæn'"
 
Discussion
This study underscores the powerful explanatory capacity of Construction Morphology for tackling complex morphological phenomena in Persian. The framework successfully accounts for the extensive polysemy of "zædæn" compound verbs not as a list of arbitrary meanings but as a structured, motivated network of related constructions. The identified schemas and subschemas, linked by inheritance hierarchies, reveal a systematic architecture behind what might otherwise appear to be an erratic set of expressions.
The pervasive features of suddenness and speed across many schemas can be interpreted as semantic traces inherited from the core meaning of "zædæn", even in its most bleached forms. This grammaticalization process is a fundamental driver of linguistic productivity, allowing the language to efficiently create new verbs, particularly for adapting loanwords (e.g., email zædæn, tag zædæn) and conceptualizing new actions.
The implications of this research are twofold. Theoretically, it provides strong evidence for construction-based approaches to morphology, demonstrating their superiority over strictly lexica list models in dealing with productive and polysemous systems. Practically, the detailed schema-based description offers a valuable resource for pedagogy, providing a clear, structured framework to help learners master this challenging aspect of Persian vocabulary. For computational linguistics, these schemas can be formalized and implemented to significantly improve the performance of NLP systems in tasks involving parsing, generating, and translating Persian compound verbs.
 
 
References

Anvari, H. (2002). Farhang-e Bozorg-e Sokhan. Tehran: Sokhan.
Bahrami-Khorshid, S., & Namdari, A. (2019). “The Study of Prefixed Verbs in Persian within the Framework of Construction Morphology”. Research in Western Iranian Languages and Dialects, 7(2): 35-55. doi: 10.22126/jlw.1970.1072
Bamshadi, P., & Ghatreh, F. (2017). “The polysemy of the Persian suffix '-i': A Construction Morphology approach”. Linguistic Investigations, 8(2): 265-289.
Booij, G. (2010a). Compound construction: schemas or analogy? In S. Scalise & I. Vogel (Eds.), Cross-disciplinary issues in compounding(93–108). John Benjamins.  https://doi.org/10.1075/cilt.311.08boo
Booij, G. (2010b). Construction Morphology. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199274269.001.0001
Culicover, P., & Jackendoff, R. (2005). Simpler syntax. Oxford: Oxford University Press.
Es'haghi, M., & Karimi-Doostan, G. (2022). The Corpus of Persian Light Verb Constructions. Journal of Linguistic Research, 14(1): 173-198. https://literature.ut.ac.ir/compound-verb
Family, N. (2006). Exploration of Semantic Space: The Case of Light Verb Construction in Persian[PhD dissertation, EHESS]. https://tel.archives-ouvertes.fr/tel-00011891
Hasanpuran, Z. and Alizadeh, A. (2024). Investigating the Compound Words Ended to “Negar” Stem in Persian Language: A Cognitive Approach. Language Research, 14(2): 79-104. doi: 10.22059/jolr.2024.363973.666852
Jackendoff, R. (2007). A parallel architecture perspective on language processing. Brain Research, 1146: 2–22. https://doi.org/10.1016/j.brainres.2006.08.111
Karimi, K., Najafian, A., Mahdi-Beyraghdar, R., & HosseiniMaasoum, S. M. (2024). “Redefinition of Compounding in Sorani Kurdish Morphology: An Investigation on Nominal Compounds in Construction Morphology Approach”. Research in Western Iranian Languages and Dialects, 12(1): 73-94. doi: 10.22126/jlw.2023.9211.1704
Kashani, K. (1993). The Zansu Persian Dictionary. Tehran: Center of University Press.
Mansouri, M. (2013). The light verb in Persian compound verbs. Studies of Languages and Dialects of Western Iran, 1(1): 77-104.
Rafiei, A. (2012). “The concept of agent in Persian derived words”. Linguistic Researches, 2(7): 19-32.
Rasooli, M.‎‎, Moloodi,  A.‎ S.‎, Kouhestani, M.‎, Minaei Bidgoli,  Behrouz.‎ (2011)‎.‎ A Syntactic Valency Lexicon for Persian Verbs: The First Steps towards Persian Dependency Treebank.‎ In 5th Language & Technology Conference (LTC): Human Language Technologies as a Challenge for Computer Science and Linguistics, Poznań, Poland,pp. 227-231.
Razmdideh, P., & Hassan-Shahi Raviz, M. (2021). Analyzing the polysemy of Persian compound verbs with the light verb 'gereftan' based on Construction Morphology. Iranian Journal of Language and Linguistics, 6(2): 213-252.
https://doi.org/10. 22099/jill.2022.42693.1294
Riehemann, S. Z. (1998). “Type-based derivational morphology”. The Journal of Comparative Germanic Linguistics, 2(1): 49–77. https://doi.org/10.1023/A:1009713009957