Compensatory Lengthening in Sanandaji Dialect: A Diachronic Approach

 The main goal of this study is to understand the mechanism of compensatory lengthening in words’ historical evolution in Sanandaji dialect. This research, conducted within the framework of Iranian languages’ phonological evolution, focuses on vowel lengthening due to deletion of consonant. Through understanding this phonological process, it would be possible not only to determine whether words are native or borrowed but also to identify trace and timing of borrowing, alongside with reconstructing their previous forms and evolution. To achieve this purpose, words with known previous forms have been examined in order to realize the compensatory lengthening process. These words include Arabic loanwords and shared words with Persian language, which may be loanwords or words with a common root. As preliminary results, lengthened vowels and their corresponding short forms are identified, along with the six consonants whose deletion would at first sight led to the lengthening of the adjacent vowel. Finally, it has been shown that it is only the deletion of ‘h’ that led to the compensatory lengthening, while other consonants were changed to ‘h’ before deletion. 
 
Introduction
Understanding the characteristics of the compensatory lengthening process in the dialect of Sanandaj can provide insight into the historical evolution of words in this dialect and provide a basis for reconstructing their earlier forms. This study aims to examine the compensatory lengthening of vowels caused by the deletion of a following consonant within the same syllable in the Sanandaji dialect. The findings are used to determine the track and time of borrowing for loanwords and also to reconstruct the evolution of original lexical items.
The Sanandaji dialect is a variety of Kurdish language spoken by people living in the city of Sanandaj for their daily conversations. This dialect is classified within the central group of Kurdish dialects. The “Kurdish languages” are among the northwest group of New Iranian languages (Abolghassemi, 2018, p. 212), and Sanandaji, as one of the Kurdish dialects, belongs to this group of Iranian languages. As a result, with a chronological approach, it is assumed that both Sanandaji and Farsi are derived from Proto-Iranian and became distinct varieties in the periods that followed (Middle, New, and Modern periods). However, it is expected that Sanandaji has many similarities with the Persian language, either because of their common origin or due to linguistic borrowings.
Theoretical Framework
"In phonology, compensatory lengthening is an effect in which the deletion of a segment is accompanied by an increase in the length of another, usually adjacent to it, thus preserving syllable weight" (Crystal, 2008, p. 91). The deleted segment can be a consonant, vowel, or semi-vowel, and the effect of this deletion, if any, can be either the lengthening of a vowel or the intensification of a consonant or semi-vowel. The term compensatory lengthening has sometimes been used only for the lengthening of a vowel due to a consonant deletion (see Kager, 1999, p. 380). This process has been described and formalized through various linguistic theories. "The mora conservation hypothesis is perhaps the most influential and widespread contemporary treatment of compensatory lengthening" (Kavitskaya, 2002, p. 5). This theory, whose final form was completed by Hayes (1989), considers the preservation of syllable weight as the cause of compensatory lengthening and explains it using the Moraic model. Many researchers have also used Optimality Theory (together with moraic theory or independently of it) to explain this process.
In the study of compensatory lengthening from a diachronic perspective, the historical evolution of words is studied and their earlier forms are reconstructed. This research investigates the process of compensatory lengthening in the evolution of Sanandaji words within the framework of the phonological evolution of Iranian languages, which is based on philological theories.
Examining compensatory lengthening in the Sanandaji variant of the Kurdish language is the aim of the present paper. The focus here is on the lengthening of vowels due to the deletion of consonants. Lengthened vowels were discussed first, followed by the study of deleted consonants one by one. Six consonants have been recognized whose deletion, at first glance, has led to compensatory lengthening of vowels in the Sanandaji dialect. In all discussions, in addition to the main form of loanwords in the source language and the written form of the equivalent words in Persian, the intermediate forms of words and the forms of Persian words in the predecessors of the Persian language (Dari Persian, Middle Persian, Old Persian, and Proto-Iranian) are also studied when necessary. The possibility of discontinuous compensatory lengthening has also been investigated.
Finally, in addition to tracing the paths of loanwords and reconstructing historical word forms, the findings indicate that five of the six identified consonants were transformed into ‘h’ before deletion.
Results and Conclusion
A summary of the results is presented below:
Lengthened vowels resulting from the compensatory lengthening process have retained the quality of the original short vowels. Considering that there are short and long vowels of the same quality in the Sanandaji dialect as distinct phonemes, and by comparing them with Persian and Arabic vowels, the probability and trajectory of borrowing were determined for each word. For original words, the evolutionary path and its relation to the evolution of equivalent Persian words were also determined.
Compensatory lengthening has occurred in a syllable where another consonant followed the deleted consonant, either in the same or the next syllable. The only exception is ‘řā,’ where the lengthened vowel is placed at the end of the word.
The discontinuous compensatory lengthening that has occurred in modern spoken Persian in some monosyllabic words, especially Arabic loanwords, has not occurred in Sanandaji.
The consonants whose deletion seems to have led to compensatory lengthening are ħ, ʕ, h, ʔ, x, and š. In words where ʔ, h, x, and š existed in their previous forms and were deleted in their present forms due to the compensatory lengthening process, they were converted to ‘h’ before deletion. In other words, these consonants were first transformed into ‘h,’ and in the next stage of evolution, the deletion of ‘h’ caused compensatory lengthening. For the pharyngeal consonant ‘ʕ,’ we cannot be sure of such a transformation occurring in a single step. This consonant may have changed first to the pharyngeal ‘ħ’ and then to the glottal ‘h,’ after which the deletion of ‘h’ caused compensatory lengthening.
    According to the findings of this research, it is necessary to consider the historical background and evolution of words to investigate the compensatory lengthening process. The study shows that after borrowing and before the application of compensatory lengthening, words may have undergone other transformations. Additionally, original words may be the result of compensatory lengthening applied to an older form of the word than its counterpart in modern Persian. One of the most important findings is the conversion of the consonants ʔ, ħ, ʕ, x, and š into ‘h’ before deletion. In fact, in the final stage of evolution, the deletion of ‘h’ alone caused the compensatory lengthening of the vow
Ethical Considerations
Not applicable
Funding
Not applicable
Conflict of interest
The authors declare no conflict of interest.