When, after the Muslin conquest, new vernaculars emerged which differed from the language of the Qurʾān, exegetic commentaries became necessary. Grammarians aimed at explaining features in this diglossia within a formal framework. Sībawayhi (eighth century) wrote a fully developed grammar of Arabic which became a highly influential model and included, especially, words analysed in terms of roots, a dependency type of syntax, verb-agent vs. topic-comment sentences.With the spread of Islam, Arabic acquired large numbers of speakers, and grammars of indigenous languages were written — with adaptations — using the Arabic model.

The aim of this book is to give a sketch of the historyof the Arabic language, mother tongue of more than200 million speakers. Since its earliest appearanceas a world language in the seventh century CE,Arabic has been characterised by an oppositionbetween two varieties: a standard language, whichoccupies a prestige position and is revered as thelanguage of religion, culture and education; and avernacular language, which serves as the mothertongue for most speakers and is the natural means ofcommunication throughout society. The oppositionbetween these two varieties constitutes the majortheme of the present book.

The set-up of the book is basically chronological:after an introduction on the study of the Arabiclanguage in Western Europe, Chapter 2 deals with theposition of the Arabic language within the group ofthe Semitic languages, and Chapter 3 with itsemergence in historical times. Then, the linguisticsituation in the Arabian peninsula in the periodimmediately preceding the advent of Islam isdiscussed (Chapter 4).

In the course of the Arab conquests, after the death ofthe Prophet Muḥammad, the Arabic language wasexported together with the religion of the Arabs toa large part of the Mediterranean and Near Easternworld. In the next two chapters, the development ofArabic into a literary standard is analysed. Chapter5 describes the role of Arabic as the language ofliterature and administration. Chapter 6 stepsoutside the chronological framework and discussesthe structure of the Arabic language from anunexpected perspective, that of the Arabgrammarians, who analysed their own language in away that differed in many respects from the Westernmodel.

The contact between the speakers of Arabic and theinhabitants of the conquered territories broughtabout a restructuring of the language, which led toan opposition between standard language andvernacular dialect. Chapter 7 attempts to explainthe emergence of vernacular varieties of the Arabiclanguage. In Chapter 8, the influence of thevernacular language in the so-called Middle Arabictexts is analysed.

Introduction

In the course of history, some speakers of Arabic havebecome isolated from the central area. Living amongpeople who speak a different language and in orderto function in society they have to use the dominantlanguage of their neighbours, while continuing tospeak Arabic among themselves as a home language.Usually, in such linguistic enclaves (called inGerman Sprachinsel)the home language has little prestige, and thespeakers depend on the official language in theireveryday dealings. Because of the frequentcode-switching and the many integrated loans, theirlanguage is subject to all kinds of linguisticpressure from the dominant language (adstratalinfluence). The linguistic enclaves thereforepresent an area of general interest for the study oflanguage contact.

The study of the Arabic linguistic enclaves alsocontributes to our knowledge of the history ofArabic. The classicising influence of the standardlanguage was considerably weaker here than in thecentral empire. In some respects, therefore, thestructure of these dialects reflects an earlier formof spoken Arabic in the areas to which they wereexported without the pressure from the Classicallanguage, which in the core area acted as a targetfor most speakers (cf. above, pp. 146–9). There is,however, no direct correlation between the date ofcut-off and the present structure of these dialects.Hardly any linguistic enclave, with the possibleexception of Maltese, was ever completely isolatedfrom the central area, and in most cases thespeakers maintained some form of contact with theprestigious centres of the Islamic empire, be itonly because of their adherence to Islam.

In this chapter, we shall briefly discuss thelinguistic situation of a number of linguisticenclaves in which Arabic is spoken, and we shallalso deal with the linguistic situation of the manyArab immigrants in Western Europe and America.

Maltese

When in 256/870 the Tunisian Aghlabids conquered theisland of Malta, the inhabitants were Christians whoprobably spoke some kind of Romance dialect. Duringthe period of Muslim domination, the entirepopulation took over the Arabic language. If webelieve the report by the eighth/fourteenthcentury-geographer Muḥammad ibn ʿAbd al-Munʿimal-Ḥimyarī in his Kitāb ar-rawḍal-miʿtār, for some 180 years after theconquest there were no inhabitants at all and theisland was repopulated with Arabic-speaking peopleafterwards (Brincat 1991).

The Arabs

We do not know when the first nomads came to theArabian peninsula, and we certainly do not knowwhich language they spoke. It is usually assumedthat the settlement of the peninsula took place inthe second millennium Bce. In the south, advancedcivilisations were established in the period betweenthe thirteenth and tenth centuries BCE. They tradedluxury goods such as incense across the Red Sea tothe north and developed extensive commercialrelations with the Near East. The languages used inthe inscriptions of these civilisations belong tothe group of South Semitic languages; they arerelated to Arabic, but did not partake in some ofthe innovations exhibited by Arabic (cf. above, pp.21f.). The script of the South Arabian civilisationsis related to some of the North Semitic scripts,such as Phoenician, and was probably imported fromthe Syro-Palestine region to the south. It is fromthe South Arabian script that the later NorthArabian scripts are derived. The language of theSouth Arabian inscriptions is usually calledEpigraphic (or Old) South Arabian, and is dividedinto several dialects or languages, the mostimportant of which are Sabaean, Minaean andQaṭabānian. These languages must have died out soonafter the Islamic conquests. The present-day ModernSouth Arabian languages such as Soqoṭrī and Mehri,that are still spoken in a few linguistic pockets inSouth Arabia, are related to Epigraphic SouthArabian, but do not derive from it directly (cf.below p. 44).

The inhabitants of the South Arabian empires did notcall themselves ‘Arabs’. Towards the end of thesecond century BCE, some of the South Arabianinscriptions mention nomads called ʿrb (plural ʾʿrb), who are contrasted tothe sedentary population of the south. The earliestattested use of this name stems, however, from adifferent region: in a cuneiform inscription datingfrom 853 BCE, the Assyrian king Salmanassar IIImentions as one of his adversaries Gindibu from theland of Arbi orArbāya. The name‘Arabs’ as a people's name is used somewhat later,for the first time by Tiglatpilesar III (745–727BCE), and then more frequently by his successors,under the form Arabu,Aribi.

Dialects of the Arabian peninsula

The Arabian peninsula, the homeland of the Arab tribes,remains the least-known dialect area of theArabophone world. In pre-Islamic times, there wasprobably a division into Eastern and Westerndialects (cf. above, Chapter 4), but subsequentmigrations have changed the geographicaldistribution of the dialects considerably. AllBedouin dialects in this area now belong to the newtype of Arabic, although generally speaking they aremore conservative than the dialects outside thepeninsula in the sense discussed above (Chapter 10,p. 185). In the urban centres of the Gulf themajority speak typologically Bedouin dialects,whereas the Shiʿite minorities speak sedentarydialects.

Recent attempts at classification by Ingham (1982) andPalva (1991) distinguish four groups:

In Chapter 10 (p. 187) we have seen that outside theArabian peninsula Bedouin dialects in general arecharacterised by a number of features that set themoff clearly from the sedentary dialects in the samearea (e.g., the voiced realisation of the /q/, theretention of the interdentals, and the genderdistinction in the second- and third-person pluralof the verbs and the pronouns).

In 632 CE, the Prophet of Islam, Muḥammad, died in thecity of Medina. The century of conquests thatfollowed brought both the Islamic religion and theArabic language to the attention of a world that upuntil then had possessed only the vaguest notion ofwhat went on in the interior of the Arabianpeninsula. Ever since this first confrontationbetween the Islamic world and Europe, the Arabs andtheir language have been part of the Europeanexperience. At first, the intellectual relationshipbetween the two worlds was unilateral. Greekknowledge and knowledge about Greek filtered throughinto the Islamic world, while the Byzantines did notshow themselves to be overly interested in thingsArabic. Although their military prowess was feared,the Arabs’ culture and language were not deemedworthy of study. Their religion was regarded aslittle more than the latest in the series ofChristian heresies that flourished in the East. Forthe Byzantines, the Greek heritage did not need anycontribution from the inhabitants of a desert, whoseonly claim to fame rested on their ability to harassthe Byzantine armies and contest Byzantine hegemonyin the eastern Mediterranean.

In Western Europe, for a long time information aboutthe Arabs and their religion remained limited tovague notions about paganism in the desert ofArabia. After the conquest of the Iberian peninsulain 711, the Arabs’ presence came to be seen as adirect threat to Europe and Christianity. Yet therewas another side to this confrontation, becausethrough the Arabs Western Europe got in touch with apart of its heritage that it had lost in the turmoilof the fall of the Roman Empire. Western medicineand philosophy became dependent on the Arab cultureof Islamic Spain for knowledge of Greek medical andphilosophical writings, which were practicallyunknown in the West. From the eleventh centuryonwards, after the fall of Toledo in 1085, thesewritings began to circulate in Latin translations ofthe Arabic versions. The Arabic language itself wasnot widely studied, since most scholars relied upontranslations that were made by a small group oftranslators, often Jews, who had familiarisedthemselves with the language either in Arab Sicilyor in al-ʾAndalus.

In the twelfth century, during the period of theCrusades, Western Europeans became acquainted withIslamic culture and Arabic on site.

The definition of ‘Middle Arabic’

In the preceding chapters, both the development ofliterary Arabic and the emergence of colloquialArabic have been discussed. A principal questionthat we should now address concerns the relationshipbetween these two varieties of the language in thewritten production, both literary and non-literary,in early Islam. The language of many of thepreserved written sources does not correspond to theform of Arabic as it was codified by the grammaticaltradition. This applies both to the formal literarylanguage of the later period and to the language ofthe early papyri. No grammarian, for instance, wouldever use a form such as yaktubū instead of yaktubūna for the indicative of theverb, but this form regularly turns up both in thepapyri and in some written texts. Since this is theform that the contemporary dialects use, the obviousconclusion is that this usage reflects thevernacular of the writer. In this chapter, we shalllook at the source of the deviations from theClassical norm in written texts.

In modern studies of Arabic, the collective name forall texts with deviations from Classical grammar isMiddle Arabic. The term in itself has led to a lotof ambiguity, and it is essential to explain firstwhat it does not mean. In the history of English,Old English, Middle English and Modern English aredistinguished as chronological periods, and it istempting to take the term Middle Arabic to mean astage of the language between the Classical periodand the modern period, say between 800 and 1800. Asa matter of fact, some writers use the term ‘MiddleArabic’ in this way. In his handbook of ChristianMiddle Arabic (1966–7: I, 36), Blau stated that‘M[iddle] A[rabic] constitutes the missing linkbetween C[lassical] A[rabic] and modern dialects’.In later publications, however, he has modified hisuse of the term, in order to avoid themisunderstandings that arise when Middle Arabic istreated as an historically intermediate stage. Incontemporary Arabic texts, mistakes may occur justas easily as in the Classical period, and it wouldtherefore be a mistake to assign any chronologicalconnotation to the term ‘Middle Arabic’. We shallsee below that when we analyse mistakes incontemporary texts, these turn out to be very muchlike those in Middle Arabic texts from the‘Classical’ period.

The linguistic situation in the Islamicempire

The period of the Islamic conquests immediately afterthe death of the Prophet in 10/632 constituted adrastic change in the history of the Arabiclanguage. Within a few decades, speakers of Arabichad spread over an enormous territory and imposedtheir language on the inhabitants of the conqueredcountries. Speakers of Arabic had been resident inSyria and Egypt before Islam (see above, Chapter 3),but their language had never been a language ofprestige outside the peninsula, and consequentlythere had never been an incentive for non-Arabs inthese countries to learn Arabic. In this chapter, weshall look at the consequences of the process ofArabicisation after the conquests for the structureof the language. First, we shall describe thelinguistic situation in the conquered territories;then the changes in the language will be discussed;finally, we shall look into the explanations thathave been advanced for these changes.

The historical details of the conquests are known fromthe detailed accounts of the Muslim historians, butmuch less is known about the process ofArabicisation. In the earliest stages of theconquests, the military efforts of the Islamicauthorities in Medina were aimed at the politicalcontrol of the Arabic-speaking tribes, first withinthe peninsula during the so-called ridda wars, and then outsidethe peninsula, where since time immemorialArabic-speaking tribes had roamed the Syrian desertand Iraq. The initial motive behind the conquestsmay have been the idea that all Arabic-speakingpeoples should be united under the banner of Islam,while the conquest of the neighbouring sedentaryareas occurred more or less as an afterthought.

Because of the lack of relevant documents, it is hardto tell what the rate of Arabicisation was, and inmost cases we can only make a guess as to the periodof time that was needed for Arabic to be adopted asthe main language of the empire. We do know thatArabicisation was much more complete, and possiblyeven progressed at a faster rate than the process ofIslamicisation.

Introduction

The diglossic situation sketched above in Chapter 13also obtains in North African countries, but therethe situation is complicated by the presence of asecond prestigious language, the language of theformer colonial power, France. In the literature onthe Maghreb, this situation is often described asone of ‘bilingualism’. In the original version ofFerguson's model this term referred to thecoexistence of two unrelated languages within onespeech community (cf. above, p. 242). In more recentliterature, ‘bilingualism’ is used as apsycholinguistic term referring to a speaker'sproficiency in two or more languages. Yet the bulkof the literature on the linguistic situation in theMaghreb uses ‘bilingualism’ (or its Frenchequivalent bilinguisme) in Ferguson's original sense.It seemed, therefore, best in the present chapter toretain the term ‘bilingualism’ for the coexistenceof Arabic and French (and to some degree of Arabicand Berber/Tamazight) in the Maghreb.

In this chapter we shall first look at the Frenchcolonial policy in North Africa (and Lebanon) andits effects on the linguistic situation in theseregions (section 14.2). Then, we shall discuss thevarious ways in which these countries implemented apolicy of Arabicisation after independence (section14.3). After that, we shall investigate theattitudes towards French, and the patterns ofcode-mixing and borrowing between Arabic and French(section 14.4). Finally, we shall deal with thelanguage policy with regard to Berber (Tamazight,Amazighe) and the present situation of this languagein Morocco and Algeria (section 14.5).

Whatever the term that is used to describe the presentlinguistic situation in the Maghreb (and inLebanon), its roots lie in the colonial period, inwhich the French language became indissolublyconnected with this region. Unlike British colonialpolicy, the French authorities aimed at theassimilation and integration of the indigenouspopulation. The official point of view was thatFrance did not colonise other countries in order toexploit them, but in order to bring them Frenchcivilisation. French officials presented thecolonial policy as France's mission civilisatrice, and some of themmay even have believed in this, but this attitudewas certainly not shared by most French colonists(colons), whoopposed any effort to educate the indigenouspopulation.

Pidginisation and creolisation

In Chapter 8 we have seen that in the course of theArab conquests large numbers of people in theconquered territories learned Arabic as a secondlanguage. Eventually, it became the first languagefor the majority of the inhabitants of the Islamicempire. Inevitably, the way these new learners spokeArabic affected the structure of the language. SomeArabic historians ascribe the changes – or as theycall it ‘the corruption of the language’ (fasād al-luġa) – explicitlyto this second-language learning process. Ibn Ḫaldūnstates that when the Arabs started to mingle withthose who spoke broken Arabic, even their ownchildren took on this way of speaking by beingexposed to it (Chapter 8, p. 138). A modernrephrasing of this idea is Versteegh's (1984)proposal to regard the shift from pre-Islamic Arabicto the new vernaculars as an instance of a processof pidginisation, a radical restructuring of alanguage as part of massive second-languagelearning. The assumption here is that nativespeakers when addressing foreigners use a simplifiedform of their own language. In a short period oftime, and without any formal teaching, the newlearners acquire this reduced variety as a secondlanguage in order to communicate with the nativespeakers and with speakers of other languages. Indoing so, they apply their own strategies oflanguage learning to the input. The result is areduced language variety or pidgin, which may remainin use for a long time as an auxiliary language.When speakers of different languages startintermarrying, they communicate in the reducedvariety and transmit it to their children. Thesethen acquire the pidgin language as their firstlanguage, and through expansion andgrammaticalisation they creolise it, that is,transform it into a new natural language or creole.Most known cases of pidginisation/creolisationinvolve Indo-European languages (English, French,Dutch, Portuguese, Spanish), which were learnt bythe slaves exported from Africa to the New World.Children born to these slaves were exposed to thepidgin from birth and creolised the language.

This model for the emergence of the modern Arabicdialects has been criticised on several grounds.

The Study of the Arabic Dialects

In the preceding chapters, we have concentrated on thefeatures that the Arabic vernaculars or dialectshave in common as against the Classical Standardlanguage. In that context, we have shown that theyrepresent a different type of Arabic, rather thanjust a modified version of the Classical language.In this chapter and the next the focus will be ondifferences between the dialects, in particular, thegeographical variation and the separation intoseveral dialect areas. The issue of thesociolinguistic variation between the dialects andthe standard language will be reserved for Chapter13.

The systematic study of dialect geography is a typicalinvention of Western European nineteenth-centurylinguistics. But it would be wrong to suppose thatthe Arabs themselves were not aware of the variationin speech in the Arabophone world. We have seenabove that grammarians accepted the variation in thepre-Islamic dialects and even collected thevariants, because in their view these belonged tothe corpus of pure Arabic speech (p. 45). They werenot interested, however, in the urban dialects thatarose all over the empire. In accordance with theirviews on the Arabic language, they regarded these aserroneous and refrained from mentioning them intheir writings. But those outside the grammaticaltradition did show an interest in the linguisticdifferences between the various parts of the empireand their causes. At an early date, al-Jāḥiẓ (d.255/868) informs us that ‘the people in the citiestalk according to the language of the Bedouinimmigrants that had settled there, which is why youfind lexical differences between the people of Kūfaand Baṣra and Syria and Egypt’ (wa-ʾahl al-ʾamṣār ʾinnamā yatakallamūna ʿalāluġat an-nāzila fīhim min al-ʿArab, wa-li-ḏālikatajidu l-iḫtilāf fī ʾalfāḏ̣ min ʾalfāḏ̣ ʾahlal-Kūfa wa-l-Baṣra wa-š-Šām wa-Miṣr)(al-Bayānwa-t-tabyīn, I, ed. as-Sandūbī, Beirut,n.d., p. 38). In Kūfa, he adds, the immigration ofPersians to the city brought in a number of Persianwords: the inhabitants of this city say jahār-sūj (Persian čahār ‘ four’ + sū(j) ‘ road’) where theBasrans say marbaʿafor a crossroads, and they use words such as ḫiyār instead of qiṯṯāʾ ‘ cucumber’, andwāzār instead ofsūq ‘market’(Persian ḫiyār,bāzār).

The language of the Arabs

The Qurʾān that wasdelivered to the Qurayš by the Prophet Muḥammaddescribes itself as being ʿarabiyyun ‘Arabic’ and mubīnun ‘ clear’. The twoattributes are intimately connected, as for instancein Q 43/2–3: ‘By theclear Book: We have made it an Arabic recitation inorder that you may understand’ (wa-l-kitābi l-mubīni: ʾinnā jaʿalnāhu qurʾānanʿarabiyyan laʿallakum taʿqilūna). Alllater generations have believed that its text wasthe best example of the ʿArabiyya, ‘the language of the Arabs’;in fact, that its style and language could not beimitated because of its clarity and correctness(ʾiʿjāz al-Qurʾān).In combination with the word lisān, the adjective ʿarabiyyun came to indicatea supra-tribal language that served as the bindingfactor for all those who lived in the Arabianpeninsula, as opposed to the ʿAjam, the non-Arabs who lived outsideit and spoke different languages. The Qurʾān does not use the wordʿArab, only theadjective ʿarabiyyun,but mentions the ʾAʿrāb. The latter term probablyindicates those who lived in the desert outside thecities and resisted the message of the Prophet, as,for instance, in Q9/97 al-ʾAʿrābu ʾašaddu kufranwa-nifāqan ‘ the Bedouin are the worst indisbelief and hypocrisy’.

Retsö (2003) believes that the distinction betweenʿArab and ʾAʿrāb goes back to theAssyrian and South Arabian inscriptions (see Chapter3, pp. 26f.). In these inscriptions, the ʾʿrb are mentioned as alliesand auxiliary troops of sedentary civilisations. Inthe Qurʾān, too, theʾAʿrābappear in amilitary context; they are chastised because theyneglect their duty in warfare. The ʿrb, on the other hand,appear to constitute a special nomadic group,residing in southern Syria. After the third centuryce they disappear from the sources; the an-Namārainscription of 328 ce (see p. 35) is one of the lastsources in which they appear, when Imruʾu l-Qayscalls himself ‘king of the Arabs’. Retsö connectsthe disappearance of the ʿArab from the sources with the absenceof this group from the Qurʾān, and their almost completeabsence from pre-Islamic poetry and Ḥadīṯ.

Yet the adjective ʿarabiyyun occurs as a characteristic ofthe revealed text, and this gives it specialsignificance.