To the editors:

As I was reading through Kees Versteegh’s essay, one book kept coming to mind: Luzūm mā lā yalzam (The Unnecessary Necessity), also known simply as Luzūmiyyāt. It is not a grammar book, but rather an unusual poetry collection that contains an introduction with an extensive analysis of the structure of Arabic as it pertains to rhyme. Its author is Abū al-ʿAlāʾ al-Maʿarrī, a poet, philosopher, and grammarian who lived near Aleppo around the turn of the first millennium. Al-Maʿarrī lost his sight at an early age and lived under self-imposed restraints that included being a vegetarian and not leaving his house for long periods. In one of his most well-known poems, he writes,

I see myself in three prisons
Don’t ask about the vile news
Because I lost my sight
I am confined to my home
And in the malicious body is trapped my soul¹

In Luzūmiyyāt, al-Maʿarrī imposed on himself extraordinary linguistic constraints, in denunciation of other poets whom he criticized for embellishing their poetry, not through grammatical and intellectual prowess, but by “describing women, qualities of horses and camels, types of wines, and the glorification of war.”² He, instead, opted to analyze the morphophonology of Arabic and generated all possible combinations for rhymes using the whole of the Arabic alphabet. He wrote some 1,600 poems that followed exacting rules: using all letters of the alphabet, each time with one of the four possible short vowel marks, or ḥarakāt, except for alif because it can only occur with zero-vowel. He used a different preceding consonant each time. This amounts to combining a letter, for example b, with the different short vowels—ba (fatḥah), bu (ḍammah), bi (kasrah), and zero-vowel b (sukūn or waqf)—and before each form adding the same letter, for example t, making the ending t-ba, t-bu, t-bi, t-b. It does not matter what vowel follows the preceding letter; t in this case can be followed by a, u, etc. The next step is using each combination as the rhyme for a different poem. Consequently, one poem will have the rhyme t-ba, another poem will have the rhyme t-bu, a third will have the rhyme t-bi, and the fourth the rhyme t-b. Then the letter t will be replaced by the following letter in the alphabet. Once the combinations are completed using all letters of the alphabet as preceding letters before the same final letter with the different ḥarakāt, he moves on to the next letter to be the final letter, and so forth. The result is an astonishing volume in which poetic artistry is combined with comprehensive knowledge of the lexicon, phonology, morphology, and syntax of Classical Arabic.³

Al-Maʿarrī did it to prove it can be done; to show he could write complex poems that rival the revered Muʿallaqāt, the Seven Suspended Odes that hang in Mecca; and to pose a linguistic challenge to both his contemporaries and predecessors.⁴ Grammarians and scholars of the Islamic Golden Age engaged each other in highly theoretical debates that centered on language but spilled over into their writings in other areas.

Al-Maʿarrī’s magnum opus, studied in high schools across the Arab world, is arguably not Luzūmiyyāt but Risālat al-ghufrān (The Epistle of Forgiveness), which is exactly that: a risāla, an epistle. But it is certainly a very long one, described by Geert Jan van Gelder and Gregor Schoeler as a “lengthy, mocking reply by a cantankerous maverick, obsessed with lexicography and grammar, to a rambling, groveling, and self-righteous letter by an obscure grammarian and mediocre stylist.”⁵ The letter is a response to Ibn al-Qāriḥ, who had criticized al-Maʿarrī in another epistle. In Risālat al-ghufrān, al-Maʿarrī depicts Ibn al-Qāriḥ as he travels through heaven and hell meeting many famed poets. He wrote this book in 1033, almost 300 years before Dante’s Divine Comedy.

Versteegh captures very well the inquisitive spirit of the time and the centrality of language study both at a macro-sociolinguistic level and at a micro-analytical level. He walks through the tasks of the Arab grammarians, appreciating the full complexity of their analyses and their prescience. Versteegh has an exquisite understanding of the history, structure, and social reality of the Arabic language. He manages in his essay, as in many of his other writings, to touch on issues affecting Arabic as a language of a major religion, one with a long history of grammatical analysis, and one that is highly esteemed by its users in its written form but not so much so in its colloquial forms. His essay is a journey through all branches of language study and related disciplines. He does not leave a prime mover unnamed or a major topic untreated.

The Macro-Sociolinguistic Level

The major points Versteegh explores are Arabic and Islam, the Arabic diglossia, acquisition of Classical Arabic, and the perceived superiority of this language. The advent of Islam ushered in a break with everything related to the Jāhiliyyah period, pre-Islamic times often referred to as the Age of Ignorance, except regarding the language. While the new religion repudiated atrocious customs such as burying baby girls alive, cruelty to enslaved people, and bloody tribal strife, admiration for the language used by al-Jāhiliyyah poets only grew. Muslim religious scholars agree that the major miracle of Islam is the Qurʾan and its language, known as ʾiʿjāz (inimitability). The Qurʾan itself highlights the close connection between Arabic and Islam. In Chapter 12, verse 2, it is stated, “We have sent it down as an Arabic Koran; haply you will understand.”⁶ Chapter 17, verse 88 challenges humans to come up with similar verses if they can: “Say: If men and jinn banded together to produce the like of this Koran, they would never produce its like, not though they backed one another.”⁷

The preservation of the Qurʾan, its tafsīr (exegesis), and correct tajwid (elocution) have played a major role in the maintenance of Classical Arabic. From early on, as Islam spread into linguistically diverse regions, the notion of laḥn, or corruption of the language, took hold among religious authorities and language scholars. Non-native Arabic speakers who had to learn the language soon discovered that excelling in the study of Classical Arabic is a level terrain since native speakers themselves have to learn the rules of the formal language as well.

Versteegh is correct when he mentions that “[r]eligion may constitute a major reason for investing so much energy in learning Arabic, but it is by no means the only one.” In the Islamic world, memorizing the Qurʾan is not only an act of piety but a scholarly act as well. Some of the renowned secular Arab authors of the twentieth century memorized the Qurʾan in their local kuttāb before being exposed to influence from world literature, as in the case of the influential Egyptian author Taha Hussein, who was deeply influenced by al-Maʿarrī, and the Tunisian existentialist author Mahmoud Messadi. Classical Arabic has been the language of religion, but it was also the language of every aspect of formal learning until new knowledge was produced in other languages and was no longer accessible in Arabic. Colonialism further broke the line of transfer of knowledge in Arabic, as with the drastic action taken by France in 1932 when it declared Arabic a foreign language in Algeria. Even today, acquiring advanced competence in Standard Arabic is still a matter of scholastic attainment; in these contexts, studying and memorizing religious and classical literary texts is a common practice. As aspirations for social mobility through science and technology-centered education increase, and these skills are often taught in English and French, the focus on high competence in Standard Arabic is faltering outside religious and literary circles. Scholars and ordinary people alike express doubts about general levels of competence in the language.

The second point that Versteegh highlights is the diglossic situation of Arabic. He writes that

Although the Arabic dialects are as different from Classical Arabic as the Romance languages are from Latin, there is an enduring fiction that the colloquial language is, in fact, identical with the standard language and that school children are native speakers of the learned variety. This is not the case. For explanations of the texts, students remain heavily reliant on the colloquial language.

I disagree with Versteegh that colloquial language is thought to be identical to the standard language. The perception is that the standard language is the real, proper language and that the colloquial languages are corrupt versions of it intertwined with foreign elements. There are efforts to use some national dialects in formal domains and in written form.⁸ But the overall impression is that the dialects are no match for the standard language. The important thing to keep in mind about the diglossic situation of Arabic, here in agreement with Versteegh, is that speakers of Arabic are native speakers of their local dialect, not Standard Arabic. Standard Arabic is neither acquired naturally nor used extemporaneously in everyday life. Even in formal contexts, influence from the speakers’ native dialect is usually present.

Historically, the methods of learning Standard Arabic added to the authority of those who were highly competent in it. Who studied under whom became crucial as scholars would go through unimaginable hardships to attend the lectures of known scholars and be associated with learning from them or from someone who studied under them. Early efforts at determining scientific lineage by naming each of the šuyūḫ under whom one studied represent the same idea as the silsila for sufi scholars or ʾisnād in the transmission of ḥadīth, or even initiatives today such as the Mathematics Genealogy Project.⁹ Al-Maʿarrī himself studied in Baghdad before settling back in his native town, which he put on the map as a destination for knowledge seekers to learn directly from him. Versteegh mentions Sībawayh, who studied under al-Khalīl al-Farāhīdī. The first Arabic dictionary, Kitāb al-ʿayn, and the development of the ḥarakāt are both credited to al-Farāhīdī. In the introduction to his edition of Sībawayh’s book al-Kitāb, ʿAbd al-Salām Muḥammad Hārūn mentions all the major scholars associated with Sībawayh: eleven scholars under whom he studied, with references to their mentors; his three colleagues who studied with al-Khalīl as well; and his three most prominent students.¹⁰

In a way, developing a high level of competence in Classical Arabic is a reclamation of a perceived glorious past. It also fills a need to establish a connection with a tradition that harkens back to the Islamic Golden Age. Even today, memorization plays a significant role in education. In elementary school, maḥfūḍāt are a list of texts to be memorized. In high school Arabic language classes, the teacher may call on one or two students every day at the start of class to recite what they were supposed to memorize, often parts of classical texts. I again agree with Versteegh that now as in the past, at least in Tunisia, the education system emphasizes memorization, not exclusively but significantly.

Versteegh mentions the advantage of learning Standard Arabic by rote. Undeniably, it has a cognitive effect, and it contributes to the construction of personal and social identity. But, as he puts it, it only works for “a select few.” Student success is mediated by and dependent on competence in Standard Arabic, not the native dialect, and yet written Arabic is learned slowly. Furthermore, in many Arab countries, a foreign language is introduced for teaching advanced math, science, and technology, making scholastic success a moving target. There is a price for dedicating so much effort to achieving high levels of competence in a learned language substantially different from the native dialects: many will underachieve. This is reflected in national exams for access to higher education in Tunisia, where the rate of success for the lettres stream rarely surpasses 20%. It was as low as 13% in the main session of the year 2018.¹¹ Many parts of the Arab world still have surprisingly high rates of illiteracy.

The Micro-Analytical Level

Versteegh highlights how Arab grammarians regarded the universal structure of Arabic as a rule-governed system from which real and hypothetical examples can be scrutinized. Although dedicated study of Arabic continues in elementary and high schools today, phonology is the least taught aspect of the language, in great part due to the substantial overlap with the dialects that is not observed in other parts of the language. One of the most complete phonological descriptions of Arabic was provided in Sībawayh’s al-Kitāb. He classifies sounds according to place and manner of articulation, and sonority. He also lists the different allophones and reports on regional phonetic variation that he argues can only be distinguished in oral speech, an early hint to modern acoustic phonetics. In total, he identifies 29 sounds but recognizes 42 realizations, as linguists today know that there are more allophones than phonemes in any given language. In addition, Sībawayh describes in detail the active and passive places of articulation and the different processes of assimilation. He also refers to processes of lenition and fortition, and the role of frequency in sound reduction and deletion, providing plenty of examples along the way and identifying specific tribes that exhibit certain variable features. 

Versteegh defines what the field of naḥwu encompasses and what most concerned the grammarians in syntax. He also reviews some important concepts and analytical methods. These include markedness, as attributed, for example, to diptosis where the -a ending is used both for the accusative and the genitive instead of -i in the latter case, as happens in proper nouns. The study of case inflections is, and has been, a central element in learning Arabic grammar. In instances where the case marking is not clear, taqdīr (supposition) was a way to infer meaning and determine syntactic roles. Parsing the constituents in a sentence represents another major exercise in Arabic language classes even today. A sentence like ʾiʿrab mā taḥtahu khaṭṭ (Parse what’s underlined) is a familiar command to school children.

Another challenging morphological exercise is plural formation, for which students are not only concerned with the dual and the plural markings but also how they apply differently according to the noun’s grammatical gender. While regular plurals use suffixation, broken masculine plurals require internal morphological changes that can be challenging. Other exercises include extracting the root for each word and producing the different forms according to syntactic function using the template faʿala (to do), as again mentioned by Versteegh. Both flexional and derivational morphology remain major components in Arabic instruction today.

With regard to semantics, many Arabic speakers are aware of the language’s large lexicon. Versteegh mentions the volume Tāj al-ʿarūs with 120,000 entries, though the vast majority of the classical vocabulary is unfamiliar to the average user. In addition, historically there has been a resistance to loanwords, whose equivalents are often calqued using native morphemes or derived from an existing root. This is one reason the dialects are drifting further away from the standard language; colloquial speakers adopt loanwords naturally while Standard Arabic words, provided by official organisms, rarely make it to everyday speech.

Grammar and Social Justice

Versteegh’s survey of the Arabic grammatical tradition is not about Arabic alone. It is about scholarly inquiry and debate about learned languages, diglossic languages, and their contribution to the evolution of language and the field of linguistics. The sample of exercises Versteegh mentions may appear futile and even a type of abstract intellectual activity for its lack of immediate applicability. Nonetheless, many types of inquiry do not have a clearly identifiable and immediate application beyond developing critical thinking skills in students. The objective is not to commodify learning, but to develop critical skills through precise linguistic analysis of a language that may not be used in extemporaneous speech.

An issue of social justice is briefly hinted at by Versteegh when he mentions that “a select few” can benefit from learning complex grammar rules of a language not spoken in everyday life. When access to the highly valued code is not widely available, opportunities become restricted to a select few indeed. The challenge in diglossic situations is the marrying of theoretical inquiries with access to education.

Institutions of higher education around the world are now dismantling humanities programs and embracing applied training as the main way forward. Universities rush to corporatize education instead of cultivating intellectual inquiries. Versteegh would appreciate here the use of the verb harwala naḥwa, which means rushing toward someone or something but losing composure and being overtaken by desperation in the process. Universities should be commended for engaging students in driving course offerings through demand, but at the same time questioned when they leave it up to the students to decide what they do not want to learn before they actually learn enough to make an informed decision. Many knowledge areas have begun to be considered disposable, as if knowledge and the cognitive advantage gained are wholly replaceable by hyper-specialized applied skills. Despite appearing unnecessary to many, the study of the grammar of a learned language such as Arabic is still an unnecessary necessity, even if another millennium has passed since the Golden Age of the Arabic grammatical tradition and al-Maʿarrī’s Luzūmiyyāt.

Lotfi Sayahi

Kees Versteegh replies:

In his thoughtful response to my essay on learning Arabic grammar, Lotfi Sayahi refers to the eleventh-century Arabic poet Abū al-ʿAlāʾ al-Maʿarrī. The reference is quite appropriate since al-Maʿarrī’s masterly manipulation of the Arabic lexicon in his poetry betrays his intimate knowledge of Arabic grammar. This reference reminded me of another treatise by al-Maʿarrī, the Risālat al-malāʾika (Treatise of the Angels), in which he pokes fun at the pedantry of grammarians. He pictures a group of them standing before the Gates of Paradise, trying to interest those who dwell there in the etymology of the names of angels and fruits of Paradise. The Guardian Angel rebukes them sternly, warning them that the faithful do not have the time to engage in such trifles. In this satirical treatise, too, al-Maʿarrī’s linguistic prowess shines through.¹²

A critical point raised by Sayahi concerns my use of the notion of diglossia, in particular my remark about a persistent fiction in the Arabic-speaking world that the colloquial language and the standard language are identical. He is quite right that structurally colloquial and Standard Arabic are very different and that most speakers of Arabic would never dream of equating their vernacular language (ʿāmmiyya) with the standard language (fuṣḥā). The general attitude toward what Charles Ferguson calls the low variety in a diglossic speech community is fairly negative. Its use is commonly rejected, and speakers tend to regard this variety as nothing more than erroneous Standard Arabic, even though it is the native language of all speakers. No one grows up speaking the high variety, which is a language learned at school.¹³

My point was that, even when speaking their vernacular dialect, speakers often insist that they are speaking Arabic; sometimes, they go so far as to deny the existence of a colloquial variety, claiming that there is only one Arabic language. When I was living in Egypt in the late 1980s, I had the occasion to interview both the Coptic Pope and the Sheikh of the Azhar. During these interviews my carefully rehearsed sentences in Classical Arabic turned out to be entirely unnecessary since both dignitaries spoke Egyptian Arabic throughout the interview. Yet, they obviously regarded themselves as speakers of Arabic as an abstract notion. This was in Egypt, where the attitude toward the local dialect was relatively positive, and where foreigners were encouraged to learn Egyptian Arabic as a variety distinct from Standard Arabic. In other Arabic-speaking countries, such a pride in the national vernacular is often lacking; some speakers may even claim that they themselves or other highly respected speakers only use pure Standard Arabic, when, in fact, they all use the vernacular dialect in everyday life. These speakers, one might say, perpetuate the myth that the colloquial and the standard language are identical, not in the sense that the colloquial is similar to the standard language, but in the sense that the standard language is all there is.