Arabic Corpus Linguistics

Prague Arabic Dependency Treebank (PADT) consists of refined multi-level linguistic annotations over the language of Modern Written Arabic. The kind of morphological and syntactic information comprised in PADT differs considerably from that of the Penn Arabic Treebank (PATB). This paper overviews the character of PADT and its motivations, and reports on converting and enhancing the PATB data in order to be included into PADT. The merged, rule-checked and revised annotations, which amount to over one million words, as well as the open-source computational tools developed in the project are considered for publication this year.

Idioms represent a fascinating linguistic phenomenon that has captured the attention of many linguists for decades. The ubiquity of these expressions in language use, the wide range of functions they perform in discourse, the problems they often cause in domains such as foreign language learning and translation, and, very importantly, their typical divergence from the normal rules of grammar and semantic compositionality are among the main reasons for this scholarly interest.

This book is a corpus-based study of idioms in Modern Standard Arabic. Examining the data with regard to their semantic, discursive, lexical and grammatical properties, the author sheds light on their intricate nature, establishes the major patterns of their linguistic behaviour, and provides explanations for these patterns.

I verify a chronology in which seven groups of passages represent consecutive phases. A proposed chronology is verified if independent markers of style vary over its phases in a smooth fashion. Four markers of style follow smooth trajectories over the seven phases: The first is average verse length. The second encompasses the 28 most common morphemes in the Qurʾān. The percentages of these morphemes in a text constitute its stylistic profile. The thus-defined stylistic profile is shown to vary in a smooth fashion over "time", i.e. over the proposed chronological sequence of phases. Third, a similar thing holds for a profile based on the frequencies of 114 other common morphemes. Fourth, similar results are obtained for a list of 3693 relatively uncommon morphemes. In addition to establishing a relative chronology in seven phases, this essay demonstrates the stylistic unity of many large passages. It also shows that the Qurʾān has one author.

摘要：
最近几十年，语料库语言学已成为现代应用语言学的支柱。因此，本文的宗旨是更深入地探讨语料库建设的一些认知性和操作性的步骤，以便把语料库观念向广大的研究人员推广。本文主要分为三个部分：
1. 语料库建设：理论与实践
2. 语料文本的加工层面
3. 语料格式属性的标注
第一部分探讨语料库语言学作为新学科发展的必要性，即从讨论学科的理论观念阶段进一步发展到应用的阶段，容纳最近该学科在工作方法论上所发生的新变化。在这个范围内，本文先探讨语料库语言学的定义和语料库的类型，目的是想确定一些更细致、可靠的语料库分类标准，以及所引出的语料库建设需求。
第二部分探讨语料库建设筹备阶段的语料加工层面，这部分又分三个部分：其一，语料文本的来源。详细地讨论文本的两种来源——纸质文本和电子文本；其二，现存的电子文本形式；其三，语料规范的不同层次。主要讨论四种层次：1-语料收取和记录语料库编纂有关的信息，2-文本储存阶段，3-文本转换阶段，4-选取语料样本阶段 。
第三部分探讨语料库建设中的语料格式属性标注，集中讨论双语语料库的建设，因为该语料库是应用语言学的基石，把实际运用的语言作为出发点来进行不同方面的应用研究，如翻译学、辞典编纂学、外语教学等。

关键词：语料库、语料加工、计算机语料库、计算机语料库规范问题、语料格式属性标注、XML语言与语料格式属性标注、汉阿对齐语料库、汉阿对应语料库。

هدف الدراسة:
عرض توجهِ أنشطةِ أقسامِ اللُّغاتِ الأكاديمية نحو استثمارِ المخزونِ اللغويِّ العربيِّ؛ بهدفِ استخراجِ رصيدٍ إلكترونيٍّ من أبنيتِه وتراكيبِه وأساليبِه، ومن ثَمّ توظيفَهُ في إنتاجِ مشاريعَ علميةٍ وتطبيقاتٍ لغويةٍ، وبيان أهمية فهم هذا النوعِ من النشاطِ اللغويِّ الحديث من قِبل باحثينا.
استعراض نماذج توظيف تلك الأرصدة في إنتاج مدونة/بنك لغوي بغرض تحليل أسئلة ونتائج اختبارات قياس مهارات اللغة العربية لغير الناطقين بها "اختبار اللسان العربي "اسْعَ".

منهج الدراسة:
اعتمد الباحث المنهج الوصفي التحليلي.

نتائج الدراسة:
إنتاج “مدونة "صوامت اللغة العربية" تخدُمُ الفصل الأكاديميَّ بتقريب أحد جوانب اللغة العربية للجهات والمعلمين والطلبة غير الناطقين؛ حيث تراعي الوحدات الصوتية العربية التي لا يوجد لها مقابل في اللغات المختلفة محل الدراسة "الصينية، الأسبانية، الإنجليزية، الروسية، الألمانية، الإندونيسية، الفرنسية، الأردية، التركية والهوسا" من حيث نطقها وذلك من واقع رموز الالف بائية الدولية IPA.

توصيات الدراسة:
العمل على تقييم المدونة وتقويمها؛ للوقوف على جوانب القصور، ومحاولة معالجتها بطريقة منهجية تتفق وأساليب صناعة المدونات والأرصدة اللغوية، وطبيعة اللغة العربية.

Study aims: To encourage the departments of academic languages towards investing in a linguistic resource which portrays clearly Arabic grammar, syntax and morphology. Then, to employ it in the production of scientific projects and linguistic applications, demonstrating the importance of understanding this kind of modern linguistic activity to researchers. The study also aims to provide a review of models for the production of linguistic corpus, for the purpose of analysing the questions and results derived from the Arabic Lingua Test (ALT) – a test specifically designed for assessing Arabic language skills in non-Arabic speakers.

Approach to the study: The researcher used a descriptive analytical method.

The results of the study: Production of the Arabic Consonant Corpus (ACOCO) which can assist academic stakeholders in bringing facets of the Arabic language to non-native speakers, teachers and organisations. ACOCO utilises Arabic phonetic units, which have no equivalent in some other languages. Those languages considered were Chinese, Spanish, English, Russian, German, Indonesian, French, Urdu, Turkish and Hausa. The corpus thus addressed Arabic pronunciation, in terms of the International Phonetic Alphabet (IPA) symbols.

Recommendations of the study: Further work on evaluating and refinement of ACOCO, to identify any shortcomings, and to address them in a systematic manner consistent with the methods of corpus processing and corpus linguistics. This would include observing further the modern structure and use of the Arabic language.

http://daadacademy.org/arabic-consonants-corpus-ACOCO/

[…]orandum est ut sit mens sana in corpore sano. Saturae X, v. 356 Decimus Iunius Iuvenalis (ss. I -II d. C.) RESUMEN: Tras una introducción a los corpus y su desarrollo en el campo de la lingüística moderna, presentamos un panorama general sobre la situación actual de los corpus del árabe centrándonos en el funcionamiento del arabiCorpus. Seguidamente exponemos los beneficios del uso de los corpus aplicados a la enseñanza de lenguas extranjeras subrayando que, sin embargo, se carece hasta el momento de estudios o materiales que apliquen el conocimiento sobre los corpus a la enseñanza del árabe. Finalmente sugerimos una serie de actividades basadas en el manejo del arabiCorpus que esperemos aporten a los profesores de árabe ideas innovadoras para sus clases de ALE.

Verbs in Maltese can occasionally be negated with the nominal negator mhux. Analogous negation with reflexes of muš has been observed in Egyptian Arabic, with those researching it speculating that it is a recent phenomenon. Yet its earliest attestation is in the Egyptian Arabic of the 15th century. Drawing data from observations of Levantine usage and corpora of Egyptian and Tunisian Arabics and the Korpus Malti, we document the same type of construction in eastern Mediterranean Arabics and Maltese. This suggests that the phenomenon arrived in Malta with the first Arabic speakers to inhabit the islands, and it adds to Mifsud's (2008) " curious similarities with the Eastern dialects " and another " striking parallel " to Borg's (1997) occurring " in the realm of syntax " giving cause to revisit Stumme's (1904) contention that the antecedents to Maltese are Levantine.

Despite the notion that written Arabic is invariable across the Arab world, a few researchers, using large corpora to discover patterns of usage, have demonstrated regional differences in Arabic writing. While most such research has focused upon the lexicon, this corpus-based study examines a syntactic difference between Egyptian and Levantine writing: the treatment of object pronouns. A search of an entire year of writing in regional newspapers found that Levantine writers tend to use the free object pronoun [iyyā-], placing the direct object after the indirect, about twice as often as Egyptian writers do, who for their part prefer to place the direct object before the indirect. A proposed reason for this is that the free object pronoun is used to mark the direct object in spoken Levantine vernaculars but not in Egyptian. This seems to indicate that local spoken vernaculars exert a fundamental influence on writing.

The study explores the process of using Arabic websites for Arabic language learning, utilising the Arabic Corpus Linguistic approach. This approach enables data-mining out of websites, systematically compiling the mined data, as well as processing the data for the express purpose of Arabic language teaching including its clusters, such as Arabic pragmatics, Arabic linguistics, and Arabic translation teaching as well. The research is written descriptively and utilises qualitative methods used for analysing the process and step-by-step procedures to be executed to make good use of the data. This study is conducted based on the theory of source-based teaching, while the process of utilising the websites is systematically elaborated through the corpus linguistic mechanism. The research concludes that almost all Arabic websites can be employed to be authentic, reliable teaching sources. The sources can be made good use of for teaching the four language competencies, for being the object of linguistic studies and for translation through the particular use of websites whose contents are bilingual or multilingual. The utilisation of the Corpus for teaching and learning has still been needing wide-spreading and promoting either among practitioners or among researchers of the Arabic language in Indonesia. This study highlights that almost Arabic-language websites are one of the richest sources of learning. These learning resources can be used for language learning and various other dimensions of scientific Arabic. Corpus linguistics has many benefits for learners and teachers in Arabic language learning. This study gives the new approach of Arabic teaching-learning using website resources, and the dynamic of Arabic learning using technology.

تُعرّف الكلمات المميزة ضمن إطار لسانيات المدونات اللغوية بالكلمات التي يختلف تكرارها في المدونة اختلافا واضحًا عن تكرارها في مدونة أخرى وذلك من منظار إحصائي. ولاستخراج الكلمات المميزة للمدوة فوائد واستخدامات متعددة تبدأ من توجيه اهتمام الباحث لما يمكن دراسته وليس انتهاء بتحديد الأخطاء والفروق الإملائية بين المدونات. ويتطرق هذه الفصل بشكل أساسي إلى التعريف أولا بالكلمات المميزة واستخداماتها وفوائدها وكيفية استخراجها من المدونة من جهة. كما يتطرق إلى تأثير المدونة المرجعية reference corpus من حيث حجمها وموضوعاتها وزمنها ومنطقتها الجغرافية على نتائج الكلمات المميزة من جهة ثانية. ويؤصل لأهم المقاييس الإحصائية المستخدمة في استخراج الكلمات المميزة مع تبيان تأثيرها على الكلمات المميزة وأوجه الشبه والاختلاف بين هذه المقاييس من جهة ثالثة. ومن جهة رابعة: يشرح بشكل مبسط ما يمكن استبعاده من الكلمات المميزة اعتمادا على نوع الدراسة المستخدمة وكمية الكلمات المميزة التي يمكن اعتبارها في الدراسة. وقد درستُ التأثيرات المذكورة آنفا باستخدام منهج تجريبي، مُركزا على الجانب الكمي من موضوع الكلمات المميزة، ولم أتطرق إلا بشكل بسيط جدا إلى تحليل الكلمات المميزة المستخرجة أثناء التجارب لتشعب هذا الموضوع وارتباطه بنوع الدراسة المعتمدة على الكلمات المميزة. وأهم ما خلصت إليه التجارب هو أن التأثير الأكبر للمدونة المرجعية يتعلق بموضوعاتها ومدى تنوعها بينما حجمها وزمنها ومنطقتها الجغرافية قليلة الـتأثير. كما أوضحت التجارب أن المقاييس الإحصائية يمكن أن تصنف إلى مجموعتين، كل مجموعة منها تضم مقاييس تتشابه بشكل كبير جدا في الكلمات المميزة الأكثر تميزا، وتكشف عن نمط معين من الكلمات المميزة. وتتضمن المجموعة الأولى معامل الاحتمال اللوغاريثمي، ومربع كاي، ومعامل زد، ومعامل تي. بينما تتضمن المجموعة الثانية معامل الغرابة والمعلومات المتبادلة. وفي نهاية الفصل ذكرت بعض الملاحظات والاعتبارات التي يجب النظر إليها بعد استخراج الكلمات المميزة.

The earliest Maltese grammars of the 18th and 19th-centuries attest a polar interrogative enclitic –š. An early 20th-century grammar shows it retaining an interrogative function and marking indirect questions. By the late 20th century, grammars either do not address it or they attest it as being largely restricted to an erstwhile 3rd-person copular pronoun hux in polar interrogatives and indirect questions. Nevertheless, the MLRS Corpus shows it functioning in all personal pronouns and in verbs in indirect and rhetorical questions. Its preservation in Maltese provides a clue to the origin of a similar feature in Levantine, North African and Andalusi Arabics.

This paper investigates the constructional behaviour of three of the most frequent GO verbs in Modern Standard Arabic: ḏahaba, maḍā, and rāḥa. These verbs are considered somewhat synonymous according to many classical and modern dictionaries of Arabic. Nevertheless, each verb has a distinctive profile manifested in its constructional behaviour, which explains why these verbs are not easily interchangeable in various contexts of use. In this paper, I will examine the prototypical uses of the three MSA GO verbs based on corpus data (extracted from arabicorpus.byu.edu) by highlighting the lexico-syntactic frames they each associate with. This is achieved by annotating a large number of contextualized uses (per verb) for a variety of lexico-syntactic features. The data frame is subsequently probed with the help of Hierarchical Configural Frequency Analysis (von Eye, 1990; Gries, 2004) as a means of highlighting recurring and significant patterns of variable co-occurrences. The quantitative analysis is followed by a qualitative analysis that further explores the lexico-syntactic frames that pertain to different aspects a deictic motion event. The results obtained from both the quantitative and qualitative analyses highlight the idiosyncratic constructional properties that characterize the use of each verb in various physical and figurative motion event construals.

Corpora have made many significant contributions to our understanding of how language works and how it is used in the society. In particular, they have become an indispensable tool in almost any systematic investigation of the lexicon. This chapter gives an overview of some key aspects of Arabic lexis that can be examined straightforwardly with corpora. Among the topics addressed are word frequency, clusters, synonymy, polysemy, and semantic prosody. Sample corpus analyses are provided to show how corpora can help uncover major patterns in the linguistic behavior of Arabic lexical items. Finally, the chapter discusses some implications for teaching Arabic as a foreign language.

In the present paper we look into the different ways in which one of the Arabic modalities of proximity, viz. kāda, is rendered in Hispanic Qur´anic translations ranging from the 16th century and up to the early 17th one. The analysis is thus based on an assortment of jewels2 from the Mudejar-Morisco cultural
heritage, a corpus of manuscript texts that offer a valuable testimony to the Islamic hiero-variety of Romance in the context of an Islamic hiero-Sprachbund, as postulated by Casassas (2010: 368-396).
We follow in our analysis the classification of Arabic modalities provided by Nadia Anghelescu (1999: 130-134) and we look into the different mechanisms used by the Morisco translators in order to transpose Arabic structures into the Romance vernacular. We also include a revisited version of the calque taxonomy put forth by Soha-Abboud (2000: 141-148), by taking into account the semic analysis of some
peculiar Romance structures that exhibit a marked Arabic influence.

Modal auxiliaries in spoken Arabic usually embed verbs in the unmarked imperfect. Yet, Brustad (2000) has documented modals embedding perfective verbs in the speech of an informant from a village near Latakia, Syria. This study demonstrates that rather than its being a geographically isolated rural peculiarity, this type of construction is a feature of coastal Levantine dialects, rural and urban, from northern Syria to southern Lebanon and inland as far as Aleppo and the hinterlands of Homs and Damascus. What is more, Arabic varieties allowing modals to embed perfect verbs are widely dispersed outside the Levant, attested in the Arabian Peninsula, the Sudan, and North Africa. Perfect modals are semantically meaningful, introducing a time-bounded dimension into discourse.

Prague Arabic Dependency Treebank (PADT) consists of refined multi-level linguistic annotations over the language of Modern Written Arabic. The kind of morphological and syntactic information comprised in PADT differs considerably from that of the Penn Arabic Treebank (PATB). This paper overviews the character of PADT and its motivations, and reports on converting and enhancing the PATB data in order to be included into PADT. The merged, rule-checked and revised annotations, which amount to over one million words, as well as the open-source computational tools developed in the project are considered for publication this year.

الهدف الأساسي لهذا المبحث هو تقديم نماذج تطبيقية، وعينات توضح طرق استخدام ’لسانيات
ا الحاسوبية في رصد وتحليل الظواهر اللغوية. وأما المادة اللغوية المستخدمة 􀄓 المدونات‘، وسبل استثمار إمكانا
في التحليل فهي ’العربية الفصحى المعاصرة‘ المستخدمة في لغة الصحافة المكتوبة في كافة الأقطار العربية.
ولقد اخترنا هذا الموضوع لربط لسانيات المدونات كمنهجية في الدراسة والتحليل بالإشكال النظري المتعلق
ببعض ظواهر العربية المعاصرة لكي يتسنى لنا مناقشة ما يتعلق بذلك من قضايا ومسائل مع التمثيل
المناسب المعتمد على ’لسانيات المدونات‘. وينقسم المبحث إلى أربعة أقسام رئيسية. فبعد أن نتناول
مفهوم ’الفصحى المعاصرة‘ وما يتعلق به من إشكالات نظرية ومنهجية في القسم الأول، سننتقل في القسم
الثاني إلى وصف المدونة وأداة التحليل المستخدمة في الدراسة الحالية. أما القسم الثالث فيستعرض أهم
النتائج المستخلصة التي توضح طرق استخدام المدونات في التحليل اللغوي. والقسم الرابع يتضمن مناقشة
عامة وموجزة تربط النتائج المستخلصة بالإطار النظري، وتوضح الإضافة المنهجية التي تقدمها لسانيات
المدونات، وما تفتحه من آفاق جديدة في مجال البحث اللغوي.

Text classification (TC) is an essential field in both text mining (TM) and natural language processing (NLP). Humans have a tendency to organize and categorize everything as they want to make things easier to understand. Therefore, text classification is an important step to achieve this goal. Arabic text classification (ATC) is a difficult process because the Arabic language has complications and limitations resulting from the nature of its morphology. In this paper, a proposed approach called the Master-Slaves technique (MST) was used to improve Arabic text classification. It consists of two main phases: in the first phase, a new Arabic corpus of 16757 text files was collected. These text files were classified into five categories manually. In the second phase, four different classifiers were implemented on the collected corpus. These classifiers are Naïve Bayes (NB), K-Nearest Neighbour (KNN), Multinomial Logistic Regression (MLR) and Maximum Weight (MW). Naïve Bayes classifier was implemented as Master and the others as Slaves. The results of these slave classifiers were used to change the probability of the Naïve Bayes classifier (Master). The four classifiers used were implemented individually and the simple voting technique was implemented among them too on the collected corpus to check the effectiveness and efficiency of the proposed technique. All the tests were applied after the pre-processing of Arabic text documents (tokenization, stemming, and stop-word removal) and each document was represented as vector of weights. For the reliability of the results, 10-fold cross-validation was used in this paper. The results showed that the Master-slaves technique gives a good improvement in accuracy of text document classification with accepted algorithm complexity compared to other techniques.

As opposed to its earlier counterparts written in Aljamiado – i.e. Romance transcribed in Arabic letters –, the late Morisco Ms. 235 (Toledo, Biblioteca de Castilla-La Mancha) was copied in Latin script. Produced in early 17th century, in the eve of the Morisco expulsion, the codex contains the sole complete vernacular Koranic translation from the Mudéjar and Morisco era to survive until the present day (López-Morillas, 2011: 113). On these grounds, linguists designated it in several occasions a veritable ‘jewel’ of the Aljamiado literature (Cotto Andino and Fernández Martín, 2001; López-Morillas, 2011; Ayerbe Betrán, 2012).
The current paper aims to validate the hypothesis brought forth by López-Morillas (1994: 17-24), according to which the Aragonese prefix des- was used in this Romance vernacular not only with an adversative role, but also in order to intensify the meaning of the base, as a consequence of the translator’s attempt to render into Romance the Arabic ḍll root and its semantic load. Therefore, we identify throughout our corpus the Morisco translations of this root and we contrast them with the renderings of two other roots that carry same meaning, devoid, however, of the former’s emphatic nuance (i.e. ġwy and ṭġy). We analyse the samples (two hundred fifty-four of them) and we interpret the data by means of statistical coefficients χ2, p and φ.

Like in other languages, Arabic distinguishes between conditional sentences on the one hand, represented by the operators of supposition ʾiḏā, ʾin and law applied to a clause, and indefinite conditional sentences on the other, represented by the nouns man (« whoever »), mā (« whatever »), matā (« whenever »), ʾayna, ʾayna-mā (« wherever »), mah-mā (« whatever »), etc. The vast majority of recent grammars of Contemporary Written Arabic, according to Classical Arabic grammar treatises, present the syntax of these indefinite conditional sentences as equivalent to that of the ʾin conditional structures in Classical Arabic, that is to say either ʾin faʿala … faʿala or ʾin yafʿal … yafʿal. But a detour by the contemporary Arab press shows many other realities. This study exhibits and analyzes this great syntactic diversity that affects not only the apodosis but also the protasis of these conditional sentences as well as it shows an innovation: the temporal de-neutralization within the protasis of faʿala which can then be interpreted as a formal and semantic past, following the fact that yafʿalu and sa-yafʿalu can be used in it as present and future.

Linguistic argumentation is defined by Arab grammarians to mean a formulation of grammatical rules from primary sources using anomaly, consensus, measurement, an argumentation based on circumstances and other rules. The anomaly sources are considered the most authentic references which are the holy Quran, Hadith al-Nabawi and Arabic poetry. This research aims to highlight the different views of Arab grammarians in devising the grammatical rules of the hadith. The authenticity of the Quran is indisputable as its verses are reliable and consistent. However, is there a difference between reliable reading and anomalous reading? Why are grammatical rules which are derived from anomalous Qira ʾat accepted, while the linguistic argumentations resulting from Hadith al-Nabawi are refuted? Why it is accepted in unknown poetry? And why are the argumentation from Quran and Arab poetry not merited and taken as evidence in linguistic argumentation? These questions are answered thoroughly by presenting the evidences and arguments from scholars in the field. An analytical study is being used as an approach to discuss the topic.

The Aljamiado-Morisco literature consists mainly of religious writings produced mostly in the 15th and the 16th centuries by a cryptic Iberian Muslim community. With twenty seven codices known up to this date, the Koranic translations are of great linguistic interest.
Recent studies paid special attention to the textual criticism of the manuscripts and attempts to classify them in families of texts were made for twenty six of them. However, given that the studies in question didn’t rely on the correct original Arabic text, some of the
scribal errors might not be actual errors. If that is the case, the finding will call for a review of the above-mentioned classification. In the present paper I aim to identify the original Arabic text for eleven of the codices. The novelty of this study resides in the analysis
of the J 101 manuscript, as well.

Paronomasia, used as a grammatical device, is a construction much favoured in Semitic languages. It comes as no surprise to encounter such structures in calque languages.
However, considering that those translations are very literal, it’s unusual to come across paronomastic constructions when they don’t appear as such in the original texts. That is the case of the Romance translations of some superlatives in the Aljamiado-Morisco Koranic manuscripts. Hence, the present paper aims at addressing the following question: Does a
translation such as la más ermosa figura ḏ e las figuraš (manuscript J 39, 150 r) – meant to render the superlative ’aḥ sani taqwīmin (Koran, 95:4) – rely on a certain “oriental” mental logic of the translator, as linguist Galmés de Fuentes (2009: 118) noted, or is actually the result of the translator’s efforts to render the original text as accurately as possible?

This paper would like to introduce the reader into those aspects of the Arabic language which require some special treatment compared to languages Europeans are more familiar with. In spite of having fresh experience in building the Prague Arabic Dependency Treebank, the authors try to take a broader view of the problems encountered under way. The topics discussed include linguistic data retrieval, morphology and morphotactics modelling, and description of the language on the analytical level.

The Arabic tour kāna … sa-/sawfa yafʿalu is frequently encountered in contemporary Arabic press as well as in novels. Recent grammars of this state of language seem to oscillate between three distinct values which are conditional (« il ferait/he would do »), periphrastic imperfect (« il allait faire/he was going to do ») and past conditional (« il allait faire/he was going to do »). Yet it is the latter two interpretations that appear most often chosen with a counterfactual reading. This article proposes then to show, from a large corpus of contemporary Arabic press, but also from insights from French linguistics, how the Arabic tour kāna … sa-/sawfa yafʿalu, understood in Arabic linguistics as a future in the past, is not, like his French counterpart, only liable to a counterfactual reading. Specifically, this article will show how this Arabic structure is able to “say what did not happen”, to “say what irrevocably happened” and to “say what was possible to happen whether it happened or not”, with particular emphasis on the agreemental identity of such a turn, proving that many of its uses are in fact loans from European languages.

Le tour kāna … sa-/sawfa yafʿalu se rencontre fréquemment dans la presse arabe contemporaine et les romans. Les grammaires récentes de cet état de langue semblent osciller entre trois valeurs distinctes que sont conditionnel présent (« il ferait/he would do »), imparfait périphrastique (« il allait faire/he was going to do ») et conditionnel passé (« il aurait fait/he would have done »). C’est pourtant ces deux dernières interprétations qui semblent le plus souvent retenues avec une lecture contrefactuelle. Cet article se propose de montrer, à partir d’un large corpus de presse arabe contemporaine, mais aussi à partir de réflexions tirées de la linguistique du français, comment le tour arabe kāna … sa-/sawfa yafʿalu, compris en linguistique de l’arabe comme un ultérieur du passé, n’est pas, à l’instar de son homologue français, uniquement passible d’une lecture contrefactuelle. Plus précisément, il s’agira de montrer et de comprendre comment ce tour arabe est en mesure de “dire ce qui ne s’est pas produit”, de “dire ce qui s’est irrévocablement produit” et de “dire ce qu’il était possible qu’il se produise, que cela se soit produit, ou non”, en insistant notamment sur l’aspect concordanciel d’un tel tour, prouvant que nombre de ses emplois seraient en fait des calques depuis les langues européennes.

I am writing my PhD dissertation on the elaboration of a vocabulary frequency list specific to the Arabic novel, based on a literary corpus that I am building. The purpose of this research is to come up a list of 2,000 most frequently used words in the Arabic novel so that it can be used as a guidance when simplifying modern and/or contemporary novels and designing supplementary pedagogical materials. The task requires designing a valid criteria to build a solid literary corpus so that a reliable frequency vocabulary list can be obtained.

This paper would like to introduce the reader into those aspects of the Arabic language which require some special treatment compared to languages Europeans are more familiar with. In spite of having fresh experience in building the Prague Arabic Dependency Treebank, the authors try to take a broader view of the problems encountered under way. The topics discussed include linguistic data retrieval, morphology and morphotactics modelling, and description of the language on the analytical level.

Purpose: The study explores the process of using Arabic websites for Arabic language learning, utilising the Arabic Corpus Linguistic approach. This approach enables data-mining out of websites, systematically compiling the mined data, as well as processing the data for the express purpose of Arabic language teaching including its clusters, such as Arabic pragmatics, Arabic linguistics, and Arabic translation teaching as well. MethodologyThe research is written descriptively and utilises qualitative methods used for analysing the process and step-by-step procedures to be executed to make good use of the data. Main Findings: This study is conducted based on the theory of source-based teaching, while the process of utilising the websites is systematically elaborated through the corpus linguistic mechanism. The research concludes that almost all Arabic websites can be employed to be authentic, reliable teaching sources. The sources can be made good use of for teaching the four language...

Collocation extraction from corpora, whether complete or according to specific criteria, plays a significant role in computational linguistics, corpus linguistics, and natural language processing. In this paper, we present Musaheb, an Arabic collocation extraction tool that has been designed and implemented to overcome the limitations of existing collocation extraction tools. Musaheb can extract n-gram collocations (for n ≤ 5) in addition to extracting the collocates of specific word types within a window size of zero to 15 words. Moreover, it provides eight collocation statistics that may be used to calculate collocation strength, and permits the input of various constraints during node selection (that is, the determination of the word or phrase whose collocates we wish to search for) and collocate extraction. Based on the user preferences for the node, concordance, and collocate selection, Musaheb saves all nodes and their associated collocates in an XML file, allowing easy conversion to different formats. Two language resources for Modern Standard Arabic developed via Musaheb are presented in this paper: 1) a 2-gram language model, and 2) a 2-gram collocation list based on an Arabic newspaper corpus comprising more than 20 million words.