Saudi Dialect Research Articles

Linguistic Convergence and Divergence in the Use of the Third Person Singular Feminine Suffix among the Qassimi Speech Community in Riyadh

This study examines the linguistic impact of dialect contact on Qassimi Arabic (QA) speakers residing in Riyadh. Focusing on the morphophonemic feature of the third-person singular feminine suffix (-ah/-ha), the research investigates patterns of convergence and divergence through a variationist sociolinguistic lens. Data from sociolinguistic interviews with 32 QA speakers in Riyadh were analyzed using SPSS. Findings indicate a prevalent use of the non-standard variant [-ah], suggesting a strong preservation of traditional linguistic features among QA speakers despite exposure to other Saudi dialects. This was explained by the concentration of QA speakers in particular neighborhoods creating ethnic enclaves that restrict exposure to other Saudi dialects and thus preserving pre-migration speech patterns. The suffix’s morphophonemic function and lack of social or symbolic meaning also had a role in its preservation.

The Social and Cultural Representation of Saudi Dialects in the Media: A Sociolinguistic Study of Najdi and Hijazi Dialects

The Social and Cultural Representation of Saudi Dialects in the Media: A Sociolinguistic Study of Najdi and Hijazi Dialects

Advancing AI-Driven Linguistic Analysis: Developing and Annotating Comprehensive Arabic Dialect Corpora for Gulf Countries and Saudi Arabia

This study harnesses the linguistic diversity of Arabic dialects to create two expansive corpora from X (formerly Twitter). The Gulf Arabic Corpus (GAC-6) includes around 1.7 million tweets from six Gulf countries—Saudi Arabia, UAE, Qatar, Oman, Kuwait, and Bahrain—capturing a wide range of linguistic variations. The Saudi Dialect Corpus (SDC-5) comprises 790,000 tweets, offering in-depth insights into five major regional dialects of Saudi Arabia: Hijazi, Najdi, Southern, Northern, and Eastern, reflecting the complex linguistic landscape of the region. Both corpora are thoroughly annotated with dialect-specific seed words and geolocation data, achieving high levels of accuracy, as indicated by Cohen’s Kappa scores of 0.78 for GAC-6 and 0.90 for SDC-5. The annotation process leverages AI-driven techniques, including machine learning algorithms for automated dialect recognition and feature extraction, to enhance the granularity and precision of the data. These resources significantly contribute to the field of Arabic dialectology and facilitate the development of AI algorithms for linguistic data analysis, enhancing AI system design and efficiency. The data provided by this research are crucial for advancing AI methodologies, supporting diverse applications in the realm of next-generation AI technologies.

MF-Saudi: A multimodal framework for bridging the gap between audio and textual data for Saudi dialect detection

Detecting variations in dialects within a language can be challenging, particularly in regions with rich linguistic diversity like Saudi Arabia. To our knowledge, no prior attempts have been made to develop a multimodal, audio–textual framework for Saudi dialect detection. The current approaches often concentrate on detecting dialects only based on audio or textual data, which fails to capture the complex relationship between both modalities. In this paper, we propose a novel Multimodal Framework, called MF-Saudi, for Saudi dialect detection. The framework consists of three main components: (1) a pretrained BERT encoder for extracting and encoding textual information; (2) an acoustic model for representing audio signals and fusing them with textual information via the fusion layer; and (3) an alignment learning module to develop meaningful representations that capture the complexities of audio–text relationships, resulting in improved dialect detection. We conduct empirical evaluations on a real-world dataset, demonstrating that our solution outperforms some of the state-of-the-art baseline methods. The experiment’s code can be found here: https://github.com/raed19/MF-Saudi.

Salient sociophonetic features, stereotypes, and attitudes toward Jazani Arabic

Far more attention has been paid to attitudes toward Standard Arabic than colloquial varieties, especially Jazani Arabic. This study aimed to identify Saudis’ stereotypes, beliefs, and attitudes toward Jazani, as well as salient sociophonetic features that might have an impact on those attitudes. An additional goal was to determine whether Saudi speakers of other dialects could identify Jazani. The study asked the following questions. Can speakers of different Saudi dialects identify Jazani speakers from sociophonetic input alone? What are Saudi speakers’ attitudes toward Jazani Arabic? What are the salient sociophonetic features of Jazani Arabic? A total of 183 Saudi Arabic speakers participated in an online attitudinal survey. Qualitative and quantitative analyses were conducted on the collected data. Findings showed that respondents could identify the speakers as Jazanis based solely on sociophonetic characteristics. Non-emphatic /r/ and word-initial consonant clusters were pointed out as salient features of Jazani Arabic. In addition, participants’ responses included general impressions, cultural associations, and social status. Overall, speakers of other Saudi dialects had more positive attitudes than Najdi speakers, who held more negative attitudes toward Jazani speakers. The results reflected the social stratification in Saudi Arabia and confirmed beliefs, stereotypical views, and misrepresentations of Jazani speakers’ social status.

Sentiment Analysis of Students’ Feedback on E-Learning Using a Hybrid Fuzzy Model

It is crucial to analyze opinions about the significant shift in education systems around the world, because of the widespread use of e-learning, to gain insight into the state of education today. A particular focus should be placed on the feedback from students regarding the profound changes they experience when using e-learning. In this paper, we propose a model that combines fuzzy logic with bidirectional long short-term memory (BiLSTM) for the sentiment analysis of students’ textual feedback on e-learning. We obtained this feedback from students’ tweets expressing their opinions about e-learning. There were some ambiguous characteristics in terms of the writing style and language used in the collected feedback. It was written informally and not in adherence to standardized Arabic language writing rules by using the Saudi dialects. The proposed model benefits from the capabilities of the deep neural network BiLSTM to learn and also from the ability of fuzzy logic to handle uncertainties. The proposed models were evaluated using the appropriate evaluation metrics: accuracy, F1-score, precision, and recall. The results showed the effectiveness of our proposed model and that it worked well for analyzing opinions obtained from Arabic texts written in Saudi dialects. The proposed model outperformed the compared models by obtaining an accuracy of 86% and an F1-score of 85%.

A study of phonological and lexical systems in Al-Oyoun’s dialect

The present study aims to cover some aspects of the dialect of Al-Oyoun city in Al-Ahsa and whether there is a kind of reciprocal relationships among different Saudi dialects. The present study covers a number of issues, such as the geographic details about Al- Oynoun city in Saudi Arabia, the main features of AlQyoun dialect. The oral data used in the present study is collected from the natural speech of 20 people who were grown up in both Al-Oyoun and Al-hoota. The 20 participants were interviewed and their natural speech was recorded. Phonological and lexical aspects of the representative samples have been linguistically analyzed. The study has reached a number of important findings that there is a kind of mutual relationship among Saudi dialects especially when there is a close contact between the people of two different dialects.

Detecting Qassimi Saudi Dialect in Saudi Digitally-mediated Communication: A Linguistic Perspective

This paper argues that there are some phonological, morphological and semantic spoken features that are faithfully transmitted to written texts in written interactions via social media. The study adopts a mixed methods approach to detect whether there is a match between the spoken Qassimi Saudi Dialect and its written counterpart. The data for this study comes from WhatsApp textual chats exchanged between 103 Qassimi Saudi speakers. In addition to the linguistic analysis of chats, an online questionnaire was employed to examine the perception of Qassimi as well as non-Qassimi Saudi speakers on whether words elicited from the data are from the Qassimi dialect or not. The findings indicate that some spoken features of Qassimi Arabic have been found in the written digitally-mediated communications of Qassimi speakers. Qassimi Dialect spoken features are found at the phonological, morphosyntactic and semantic levels. The findings of this study have a number of practical and methodological implications for linguists and dialecticians

Are You Emojically a Mom? Age, Social Role, and Emojis in Saudi Mother-daughter WhatsApp Groups

This descriptive linguistic study aims to check the effect of age (old/young) and social role (mother/daughter) on the kind of emojis and their intended meanings that Saudi mothers use while chatting with their daughters. The data has been collected from twenty Saudi mother-daughter WhatsApp groups’ chat interactions. The study is qualitative in its approach and has used computer-mediated discourse analysis (CMDA) of ‘coding and counting’ (Herring 2004) on (58 screenshots) and has revealed (42 emoji types) of Saudi mothers used in different frequencies. A combination of multi-choice and open-ended questionnaires has been tailor-made to arrive at the subjective perspectives of ninety Saudi young females. The findings show that there are variations in the kind of emojis used and their intended meanings, signaling the two generations and social roles apart. The results signaled (7 Saudi mom emoji types) that are never used by Saudi young females at all. These emojis were found to be age-specific and social role specific. The rest of the emojis (35 emoji types) were found to be used by both groups but with differences in usage and intended meanings, marking again the differences in age and social role. Despite variations of emojis and their meanings, there were many factors that helped youngsters to understand the intended meanings of their mother’s emojis: the accompanying text, frequent chatting with their mothers, and the literal representation of the emojis their mothers used. Even though accommodation is the process Saudi young females developed in order to understand their mother’s emojis and their intended meanings, there were minor misunderstandings found. The reason has been attributed to what has been termed in this study as the ‘emoji gap’ that marks the effect of age and social role (Evans 2015, 2017; White 2017; Marko 2022). Saudi mother-daughter WhatsApp conversations revealed that they are approximate to the structure of face-to-face conversation in relation to their back-and-forth quasi-synchronous messages, topics discussed, the informal style, the intimate relation, and the Saudi dialects used accompanied with related emojis that agree with the content of the digital texts. In this study, emojis are proved to be a visual language that mirrors its users in relation to their age and social role in digital text.

Subtitling Arabic humour into English

This article examines how humour in Arabic stand-up comedies is translated into English in an audio-visual context. The study uses a case study of Arabic stand-up comedies streamed on Netflix, including Live from Beirut by Adel Karam and Comedians of the World/ Middle East. The shows which are subtitled into English involve a variety of Arab comedians speaking different dialects, including the Levant dialect (Lebanese, Jordanian, and Palestinian) and Gulf dialects, particularly the Saudi dialect. While several studies were conducted to examine the translation of English humour into Arabic, very few explore the translation of Arabic humour into English, especially in the realm of audio-visual translation. Arabic and English are two different languages reflecting different norms and cultures and, therefore, many linguistic and cultural challenges are expected to arise in the process of translation between them. The study draws on Pederson’s (2005) strategies for translating cultural references and Díaz-Pérez’s (2013) strategies for translating wordplay and puns. The study identifies two types of humour used in the Arabic stand-up comedies, namely language-restricted jokes (wordplay, puns, language variation, and taboo language) and culture-restricted jokes which require knowledge about the concept or character being referred to. Several translation strategies were used by Netflix subtitlers to render these types of jokes into English, including paraphrasing, generalizing, specification, substitution, and omission.

Nunation and the L2 acquisition of English indefinite articles by L1 Najdi and Hijazi speakers

Whether nunation is a marker of indefiniteness is a controversial issue in the literature of Arabic grammar. To date, no study has investigated whether the use of nunation in a speaker’s L1 can affect their second language (L2) acquisition of English articles. This paper reports the results of a study on the use of English articles by speakers of two Saudi dialects, Najdi and Hijazi, of which only the former – Najdi – grammaticalises nunation. The study recruited 56 participants (24 Najdi speakers, 24 Hijazi speakers and 8 native speakers of English). The experimental groups were third-year secondary school students whose English was at the elementary level, according to the Oxford Quick Placement test. The participants completed a 48-item multiple-choice test that examined their use of the English indefinite articles a and ø. It was found that a) the participants used ø more accurately than a; b) Najdi speakers were more accurate in using a than ø due to the presence of nunation in Najdi; and c) because Hijazi does not use nunation, Hijazi speakers were more sensitive than Najdi speakers to the semantic properties of nouns modified by articles.

Applying a Character-Level Model to a Short Arabic Dialect Sentence: A Saudi Dialect as a Case Study

Arabic dialect identification (ADI) has recently drawn considerable interest among researchers in language recognition and natural language processing fields. This study investigated the use of a character-level model that is effectively unrestricted in its vocabulary, to identify fine-grained Arabic language dialects in the form of short written text. The Saudi dialects, particularly the four main Saudi dialects across the country, were considered in this study. The proposed ADI approach consists of five main phases, namely dialect data collection, data preprocessing and labelling, character-based feature extraction, deep learning character-based model/classical machine learning character-based models, and model evaluation performance. Several classical machine learning methods, including logistic regression, stochastic gradient descent, variations of the naive Bayes models, and support vector classification, were applied to the dataset. For the deep learning, the character convolutional neural network (CNN) model was adapted with a bidirectional long short-term memory approach. The collected data were tested under various classification tasks, including two-, three- and four-way ADI tasks. The results revealed that classical machine learning algorithms outperformed the CNN approach. Moreover, the use of the term frequency–inverse document frequency, combined with a character n-grams model ranging from unigrams to four-grams achieved the best performance among the tested parameters.

Semantic Negation on Verbal Clauses with Verbs of Imminence in Saudi Arabic Varieties

Negation in languages is a phenomenon, which receives researchers' attention for a long period. Negation occurs on different type of clauses by which it shows to have morphological and syntactic changes over the structure of words, phrases and sentences. Verbs of imminenceare used sometimes to denote negation in verbal clauses in Arabic. ka:d is one of those verbs which carries the negation of the clause without using any negative particles. Saudi dialects use the verb baʁa to perform the same function. This paper discusses baʁa in verbal clauses illustrating its syntactic and semantic impact on the structure of the clauses. It requires the imperfective form of the verb to follow. Negating a clause, with the only negative particles ma: and/or la:, does not carry the negative sense because it is delivered through the use of baʁa. Thus, the negation of verbs of imminence does not require the presence of any negative particles. Otherwise, the positive meaning is delivered.

The Analysis of Coda Clusters in Jizani Arabic: An OT Perspective

This paper explains how native speakers of Jizani Arabic (henceforth, JA) treat final consonant clusters in superheavy syllables (CVCC) using a parallel Optimality Theory (Prince & Smolensky, 1993, 2004) to show how the theory can account for the cross-linguistic variations of coda clusters through the ranking of different constraints. JA is a Saudi dialect spoken in the southwestern part of Saudi Arabia in Jizan city. It is common among many Saudi Arabic dialects like Najdi, Hijazi, Taifi and Qassimi that rising sonority in coda clusters is avoided by using vowel epenthesis to comply with the Sonority Sequencing Principle (henceforth, SSP), where there is no difference between nasals and liquids. However, in JA, we observe that vowel epenthesis occurs only if the last segment in CVCC is a liquid (/l/ or /ɾ/); for instance, /tʕifl/à [tʕifil] ‘child’ and /ħibɾ/à [ħibiɾ] ‘ink’. The vowel has been epenthesized because the last consonant in both examples is more sonorous than the preceding obstruents. However, the vowel will not be inserted if the final consonant is a nasal preceded by an obstruent; for instance, /laħm/à[laħm] ‘meat’ and /ɡutʕn/à[ɡutʕn] ‘cotton’. Although the universal sonority scale ranks nasals as more sonorous than obstruents, nasals in JA behave as they are equally sonorous as obstruents. In other words, nasals in this dialect group with stops and fricatives in the sonority scale.

MULDASA: Multifactor Lexical Sentiment Analysis of Social-Media Content in Nonstandard Arabic Social Media

The semantically complicated Arabic natural vocabulary, and the shortage of available techniques and skills to capture Arabic emotions from text hinder Arabic sentiment analysis (ASA). Evaluating Arabic idioms that do not follow a conventional linguistic framework, such as contemporary standard Arabic (MSA), complicates an incredibly difficult procedure. Here, we define a novel lexical sentiment analysis approach for studying Arabic language tweets (TTs) from specialized digital media platforms. Many elements comprising emoji, intensifiers, negations, and other nonstandard expressions such as supplications, proverbs, and interjections are incorporated into the MULDASA algorithm to enhance the precision of opinion classifications. Root words in multidialectal sentiment LX are associated with emotions found in the content under study via a simple stemming procedure. Furthermore, a feature–sentiment correlation procedure is incorporated into the proposed technique to exclude viewpoints expressed that seem to be irrelevant to the area of concern. As part of our research into Saudi Arabian employability, we compiled a large sample of TTs in 6 different Arabic dialects. This research shows that this sentiment categorization method is useful, and that using all of the characteristics listed earlier improves the ability to accurately classify people’s feelings. The classification accuracy of the proposed algorithm improved from 83.84% to 89.80%. Our approach also outperformed two existing research projects that employed a lexical approach for the sentiment analysis of Saudi dialects.

The development of L2 interactional competence in Arabic: The case of multiple requests

AbstractThis study explored the development of L2 learners' interactional competence in Arabic when making requests. Combining a longitudinal and cross‐sectional design, it closely examined how learners' language proficiency affected their use of multiple requests in daily‐life interactions. Participants included L2 learners (n = 40) and Saudi Arabic native speakers (NSs) (n = 10). L2 learners were evenly divided into two groups (low‐level and high‐level). They each completed three interactive role‐plays in Phase 1 and three similar scenarios in Phase 2 6 months later. The findings showed that multiple requests can be vertically (e repair) and horizontally (i.e., recapitulation) organized in the talk. Multiple requests in a vertical arrangement occurred due to L2 learners' violation of sequence organization rules. L2 learners with lower proficiency levels were more likely to make multiple requests vertically. However, as their proficiency improved, their use of vertically arranged multiple requests reduced. On the other hand, multiple requests in a horizontal arrangement occurred with increasing L2 learners' linguistic and interactional competence. At native‐like linguistic and interactional competence levels, L2 learners tended to decrease this interactional practice and adhered firmly to turn‐taking and adjacency pair rules. Furthermore, it was found that L2 learners, no matter what their proficiency level, relied heavily on Modern Standard Arabic (MSA) while Arabic NSs tended to use the Saudi dialect. Therefore, it was concluded that learners' proficiency plays a crucial role in the production of multiple requests in interaction and their reliance on MSA in daily interactions reflects their incomplete pragmatic competence.

The Vowels of Qassimi Dialect

This paper describes the vowels of the Qassimi dialect and explores the characteristic features of those vowels. To achieve this goal, the researcher has compiled a list of Qassimi words. Each word represents a major allophone of each vowel phoneme. These words were then repeated by Qassimi female speakers from Buraidah and recorded by the researcher. The data was subjected to acoustic analysis, and the results were compared to the acoustic results of English vowels. The analysis showed many similarities in the characteristics of both Qassimi and English vowels. However, it also showed some significant differences that distinguish the vowel system of Qassimi from the English vowel system as well as other Saudi dialects. This study sheds light on those differences, however it is left to future studies to investigate further and possibly compare other aspects of these dialects. This study contributes to the theoretical description of Saudi dialects, an area that needs many more contributions. In addition, this paper could be a part of the contrastive studies of Arabic and English, second language acquisition studies, or various other applied studies.

A taxonomy of antonymy in Arabic: Egyptian and Saudi proverbs in comparison

AbstractThis study has set out to identify, quantify, typify, and exemplify the discourse functions of canonical antonymy in Arabic paremiography by comparing two manually collected datasets from Egyptian and Saudi (Najdi) dialects. Building upon Jones’s (2002) most extensive and often-cited classification of the discourse functions of antonyms as they co-occur within syntactic frames in news discourse, the study has substantially revised this classification and developed a provisional and dynamic typology thereof. Two major textual functions are found to be quantitatively significant and qualitatively preponderant: ancillarity (wherein an A-pair of canonical antonyms project their antonymicity onto a more important B-pair) and coordination (wherein one antonym holds an inclusive or exhaustive relation to another antonym). Three new functions have been developed and added to the retrieved classification: subordination (wherein one antonym occurs in a subordinate clause while the other occurs in a main clause), case-marking (wherein two opposite cases are served by two antonyms), and replacement (wherein one antonym is substituted with another). Semicanonical and noncanonical guises of antonymy are left and recommended for future research.

AraCust: a Saudi Telecom Tweets corpus for sentiment analysis.

Comparing Arabic to other languages, Arabic lacks large corpora for Natural Language Processing (Assiri, Emam & Al-Dossari, 2018; Gamal et al., 2019). A number of scholars depended on translation from one language to another to construct their corpus (Rushdi-Saleh et al., 2011). This paper presents how we have constructed, cleaned, pre-processed, and annotated our 20,0000 Gold Standard Corpus (GSC) AraCust, the first Telecom GSC for Arabic Sentiment Analysis (ASA) for Dialectal Arabic (DA). AraCust contains Saudi dialect tweets, processed from a self-collected Arabic tweets dataset and has been annotated for sentiment analysis, i.e.,manually labelled (k=0.60). In addition, we have illustrated AraCust’s power, by performing an exploratory data analysis, to analyse the features that were sourced from the nature of our corpus, to assist with choosing the right ASA methods for it. To evaluate our Golden Standard corpus AraCust, we have first applied a simple experiment, using a supervised classifier, to offer benchmark outcomes for forthcoming works. In addition, we have applied the same supervised classifier on a publicly available Arabic dataset created from Twitter, ASTD (Nabil, Aly & Atiya, 2015). The result shows that our dataset AraCust outperforms the ASTD result with 91% accuracy and 89% F1avg score. The AraCust corpus will be released, together with code useful for its exploration, via GitHub as a part of this submission.

A semiautomatic annotation approach for sentiment analysis

Sentiment analysis (SA) aims to extract users’ opinions automatically from their posts and comments. Almost all prior works have used machine learning algorithms. Recently, SA research has shown promising performance in using the deep learning approach. However, deep learning is greedy and requires large datasets to learn, so it takes more time for data annotation. In this research, we proposed a semiautomatic approach using Naïve Bayes (NB) to annotate a new dataset in order to reduce the human effort and time spent on the annotation process. We created a dataset for the purpose of training and testing the classifier by collecting Saudi dialect tweets. The dataset produced from the semiautomatic model was then used to train and test deep learning classifiers to perform Saudi dialect SA. The accuracy achieved by the NB classifier was 83%. The trained semiautomatic model was used to annotate the new dataset before it was fed into the deep learning classifiers. The three deep learning classifiers tested in this research were convolutional neural network (CNN), long short-term memory (LSTM) and bidirectional long short-term memory (Bi-LSTM). Support vector machine (SVM) was used as the baseline for comparison. Overall, the performance of the deep learning classifiers exceeded that of SVM. The results showed that CNN reported the highest performance. On one hand, the performance of Bi-LSTM was higher than that of LSTM and SVM, and, on the other hand, the performance of LSTM was higher than that of SVM. The proposed semiautomatic annotation approach is usable and promising to increase speed and save time and effort in the annotation process.