An Emergent Grammar Model for the Linguistic Notion of Diglossia in Arabic and Faifi

Abstract

The study of the grammar in ‘diglossic speech communities’ in which two or more varieties of the same language are used in different contexts, raises important questions: How do these varieties interact? Are there clear-cut boundaries between language varieties in these contexts? Can we posit that speakers, in certain circumstances, may possess more than one internal grammar of their language that is associated with different social contexts? If so, how do speakers develop knowledge of this type of grammar? More importantly, how can we systematically account for the massive scale of variation that is part of any diglossic speech grammar? This dissertation demonstrates how the structural phenomena of diglossia can be derived from domain-general cognitive processes, including memory, attention to frequencies and similarities, categorization, establishing connections between similar categories, making generalizations from observed patterns, etc. These processes operate in multiple instances of language use and create the grammar of diglossic speech. This dissertation uses the diglossic grammar of Faifi, an Arabic diglossic speech community located in the Faifa mountains in southwestern Saudi Arabia, to demonstrate the emergent grammar properties. The same principles should apply to the other diglossic speech communities of Arabic. The analyses are grounded in the core of a relatively new framework called Emergent Grammar (EG), which drifts away from using the concept of Universal Grammar and genetic endowment specific to language as central in driving research questions and the way of interpreting data (Archangeli & Pulleyblank, 2012a,b,c; 2013; 2014a,b; 2015a,b,c; 2016; 2017; 2018 among others). Instead, EG allows abstract symbolic grammatical structures to emerge from the experience with the language, using the general cognitive capabilities of the language learner. This idea is not entirely new to the field, considering the long-standing emergentist approach to language. What is new and appealing about the EG framework is that it accounts for language through a set of bottom-up regulatory principles (i.e., emerging conditions and relations among morphemes) that filter out unwanted combinations, resulting in the optimal output. The dissertation demonstrates how speakers build a holistic grammar system from different speech contexts such as home, public spaces, school, formal and informal situations that govern word formation processes (i.e., what is allowed and disallowed regarding the linguistic output). The dissertation also discusses why Faifi grammar (as a diglossic grammar) must be viewed as a constantly-updating system shaped by language experience.