Analysis of the Types of Complete Reduplication Processes in Sarhaddi Balochi Dialect in Optimality Theory Framework

Introduction
Balochi is a language from the Northwestern branch of the Iranian languages (Indo-European family) (Jahani, 2001: 23). This language is divided into two main dialects: Eastern and Western (Elfenbein, 1989: 359-360). According to Jahani and Korn (2009: 637), the Western dialect (Rakshani Balochi) includes the Sarhaddi variety and is spoken in various regions, including Merv in Turkmenistan, small areas of Khorasan in eastern Iran, Golestan, Sistan in southern Iran, northern parts of Balochistan around Zahedan and Khash, as well as in southern Afghanistan and southwestern Pakistan. Optimality Theory, proposed by Prince and Smolensky (1993/2004), provides a suitable framework for analyzing the phonological and syllabic complexities of Sarhaddi Balochi dialect. McCarthy (2008: 13) believes that Optimality Theory emphasizes the interaction of constraints in language production and perception and categorizes the constraints into two groups: markedness and faithfulness. This theory has been particularly applied in related studies on reduplication (Katamba and Stonham, 2006: 213). Wilbur (1973) defines reduplication as a morphological rule related to the phonological aspects of language. Marantz (1982) refers to it as affixation and identifies complete and partial reduplication as two forms of copying. Each type of reduplication has its own subcategories. In this regard, complete reduplication can be divided into three subtypes: augmented, non-augmented, and echoic.
To analyze the reduplication process, McCarthy and Prince (1995) proposed the Correspondence Theory within the framework of Optimality Theory. In the Correspondence Theory, three main faithfulness constraints for reduplication including Maximality Input-Output (MAX I-O), Dependency Input-Output (DEP I-O), and Identity Input-Output Feature (IDENT I-O (F)) have been introduced by McCarthy and Prince (1995: 16) and Kager (2004: 205). 
 
Materials & Methods
This study employs the Correspondence Theory of Optimality Theory to identify and rank the universal constraints governing reduplication and to explain its structural nature in Sarhaddi Balochi dialect. Additionally, this research aims to address key questions, including the types of complete reduplication processes in Sarhaddi Balochi dialect and how these processes can be explained based on the principles and constraints of the Correspondence Theory of Optimality Theory. To fulfill this descriptive-analytical research, the linguistic data comprising 112 reduplicated words with diverse syllabic structures were collected from 10 speakers (6 men and 4 women, aged 60–75, with low literacy) in villages of Khash county. The data were gathered through free speech recordings (stories, proverbs, interviews) and library resources, particularly Jamalzehi (2011). The data were transcribed using the International Phonetic Alphabet (IPA) and analyzed in tableaux to determine the constraints and their ranking.
Some studies have been conducted on reduplication in Iranian languages by Allahyari and Razinezhad (2016), Heidari (2016), Tafaroji Yeganeh and Jahanfar (2017), Rahmatinezhad et al. (2019), and Sayyad and Kord-e Zafaranlu Kambuziya (2021). However, regarding reduplication in Sarhaddi Balochi dialect, only Jamalzehi (2011) and Karimi Dadkan (2012) have provided synchronic descriptions of instances of this process. Therefore, a theory-based research in this area appears necessary.
 
Results & Discussion
This section analyzes various types of complete reduplication in Sarhaddi Balochi dialect, including: Complete Non-Augmented Reduplication, Complete Augmented Reduplication, which comprises Complete Medial Augmented Reduplication with interfixes such as [-ɑ-], [-pa-], [-be-], [-ke-], and  [-ta-], Complete Final Augmented Reduplication with suffixes like [-iǝn], [-ok], and [-agɑ], as well as Complete Echoic Reduplication, in turn, divided into two subtypes: Suffixal Complete Echoic Reduplication with the initial consonant or base vowel change with the insertion of conjunctive [o], and Suffixal Complete Echoic Reduplication with the initial consonant or base vowel change without the insertion of conjunctive [o]. These processes are analyzed with a focus on the interaction of faithfulness and markedness constraints.
At first, Complete Non-Augmented Reduplication, such as [send.-send] "separated, separated", involves the full repetition of the base word. Key constraints in this process include (MAX I-O), (MAX B-R), (DEP I-O), (DEP B-R), and (NO CODA). The optimal candidate is [send.-send], as it only violates (NO CODA) without any fatal violations. In Complete Medial Augmented Reduplication, such as [ʈuh.-ɑ.-ʈuh] "the largest," an interfix like [-ɑ-] is inserted between the base and the reduplicant. This process is governed by constraints such as (MAX I-O), (DEP I-O), (DEP B-R), (ONSET), and (ALIGN-R (B-σ)). The optimal candidate is [ʈu.hɑ.-ʈuh], as it only violates (ALIGN-R (B-σ)) and (NO CODA) constraints. In Complete Final Augmented Reduplication, such as [narm.-narm.-ok] "slowly, slowly", a suffix like [-ok] is added to the reduplicate. Key constraints include (MAX I-O), (DEP B-R), (ONSET), and (ALIGN-R (B-σ)). The optimal candidate is [narm.-nar.m-ok], as it only violates (NO CODA) and (ALIGN-R (B-σ)). The Suffixal Complete Echoic Reduplication with initial consonant change is divided into two subtypes: First, the initial consonant change and with the  insertion of conjunctive [o], such as [lirtʃ.-o.-pirtʃ] "badly crushed", which involves constraints (MAX I-O), (DEP B-R), (ONSET), and (ALIGN-R (B-σ)), and the optimal candidate is [lirtʃo.-pirtʃ], as it only violates (ALIGN-R (B-σ)) and (NO CODA). Second, the initial consonant change and without the insertion of conjunctive [o], such as [pas.-mas] "sheep and the like", which involves constraints like (MAX I-O), (*Cor), (*Lab), and (MAX B-R). The optimal candidate is [pas.-mas], as it only violates (*Lab).  Suffixal Complete Echoic Reduplication with the base vowel change is classified into two types: in the type with the  base vowel change and with the insertion of conjunctive [o] such as [ʃarr.-o.-ʃerr] "good and bad", constraints like (MAX B-R), (IDENT B-R (back)), (IDENT B-R (high)), and (ONSET) are involved and the optimal candidate is [ʃar.ro.-ʃerr], as it only violates (NO CODA) and (IDENT B-R (high)). But in type with the base vowel change and without the insertion of the conjunctive [o], such as [ʤɑk.-ʤik] "shouting and crying," constraints like (MAX B-R), (DEP B-R), (IDENT B-R (back)), and (IDENT B-R (high)) apply and the optimal candidate is [ʤɑk.-ʤik], as it only violates (NO CODA), (IDENT B-R (back)), and (IDENT B-R (high)).  In all these processes, faithfulness and markedness constraints play a decisive role in selecting the optimal candidate.
 
Conclusion 
The present study analyzed 112 reduplicates in Sarhaddi Balochi dialect, identifying three types of complete reduplication consisting of: non-augmented, augmented, and echoic. The data analysis demonstrated that each type of complete reduplication has its own specific constraints and ranking. In Complete Non-Augmented Reduplication, faithfulness constraints (MAX and DEP) and the markedness constraint (NO CODA) play a significant role. In Complete Medial Augmented Reduplication with the interfix [-ɑ-], constraints such as (MAX I-O), (DEP I-O), (DEP B-R) and (ONSET) have higher priority. For Complete Medial Augmented Reduplication with interfixes [-pa-], [-be-], [-ke-], and [-ta-], the constraint ranking is: MAX I-O; MAX B-R >> IDENT(F) B-R >> NO CODA. In Complete Final Augmented Reduplication, the constraint ranking is: MAX B-R; DEP B-R >> ONSET >> IDENT(F) B-R>> NO CODA. In types of  Suffixal Complete Echoic Reduplication, the constraint ranking varies depending on the insertion or non-insertion of the conjunctive [o] and the initial consonant or base vowel change. In the type with the initial consonant change and the insertion of conjunctive [o], constraints like (MAX I-O) and (ONSET) have higher priority, while in the type with the initial consonant change and without the insertion of the conjunctive [o], constraints such as (*Cor) and (MAX B-R) are more prominent. In Suffixal Complete Echoic Reduplication with the base vowel change, both types emphasize faithfulness constraints (MAX B-R) and markedness constraints (IDENT B-R).