Written by Afsheen Sharifzadeh, a graduate of Tufts University focusing on Iran and the Caucasus. The goal of this article is to explore the origins of Moroccan Darija through the lens of local sedentary and Bedouin dialects, and to examine historic contacts with autochthonous Berber languages and foreign languages such as Spanish and Persian.

Background

Northwest Africa (the “Maghreb”, from Arabic المغرب الاقصى al-maġrib al-aqṣa “the farthest West”; and, more recently, Berber: ⵜⴰⵎⴰⵣⵖⴰ Tamazɣa “Land of the Imazighen [Berbers]”) experienced two waves of Arabization separated in time by nearly five centuries. Following a classification made by Ibn Khaldūn, the first wave was composed of mainly urban soldiers while the second wave brought thousands of nomadic Arabian families to the region. The number of Arabophones who arrived in both waves must have been small beside the indigenous Berber-speaking population (still partly Latin-speaking in the towns of the old Roman coast), and whilst many Berbers shifted to the language of their conquerors over the centuries, this process has failed to result in the adoption of an exclusively Arab identity. 

Arabic vernaculars in Morocco, Algeria and Tunisia are referred to locally as Darija (lit. “the current speech; vernacular language”), while the literary Arabic standard used in education, broadcast and administration is called al-ʕarabiyya al-fuṣḥā (lit. “the most eloquent Arabic”). Darija varieties may in turn be divided into sedentary (pre-Hilali, or مديني mdīni lit. “urban”, or حضري ħadˤari “settled”) and Bedouin (Hilali or عروبي ʕarūbi, lit. “Arabian”) dialects. However, these descriptors technically refer to the medieval “Middle Eastern” vernaculars from which they descend rather than the lifestyles of their current speakers. While there is largely direct lineage between the progenitor Middle Eastern urban dialects and Moroccan urban dialects, in some cases, they do not correspond. For example, the Jebli Arabic dialect spoken in the northwestern rim of the Atlas mountains represents a “sedentary” (Pre-Hilali) dialect adopted by Berber mountaineers along the trade routes between Fes and Tangier, which therefore links it more closely with the old urban dialects of North Africa. Moreover, Moroccan Darija, which originated in 20th century Casablanca and is now the lingua franca of the country, is of Bedouin provenance (discussed below).

To further complicate the linguistic landscape in Morocco, sedentary and Bedouin Arabic varieties have interacted individually with neighboring Berber languages for over a millennium. The sedentary (Pre-Hilali) Arabic dialects, representing the first layer of Arabization, are the most innovative and bear the strongest morphosyntactic traces of these contacts. In contrast, Bedouin dialects were chiefly influenced by Berber phonetic and semantic interferences. The sedentary varieties are marked, further, by a stratum of Hispanisms, carried by the Muslim and Jewish refugees of the successive Iberian expulsions, culminating in 1492 and in the Morisco expulsion of 1609, who settled overwhelmingly in the cities; the Bedouin dialects of the countryside lay outside the refugees’ orbit, and the Spanish words accordingly stop at the city walls.

Moroccan Darija, which can be used interchangeably with “Casablancan Koine”, arose recently as a product of rapid urbanization around Casablanca in the 20th century. The dialect is of Bedouin provenance, owing to the origin of most Casablancans, but it features a strong admixture of Pre-Hilali elements brought by old urban elites who migrated from Fes, Rabat, Salé, Meknès and other northern cities. It is thus a modern koine, precipitated from many streams of migration, and the continuation of no single historically attested dialect. Although it is now the mother language of a plurality of Moroccans, there still exists a significant degree of variation within the speech community, owing to its rather recent genesis and the varied regional origins of its speakers. At times, such as in the case of Fessis, Koine speakers have purposefully transferred certain features of their ancestral lects (e.g. Fessi use of non-trilled [ɹ], emphasized occlusive laryngeal  [ʔˤ] or [q] in lieu of Bedouin [g], and gender-neutralized second person pronoun انتينا intīna in the perfective and imperfective aspects) because these features had long signified urbane refinement and, therefore, prestige.

Sedentary (pre-Hilalian) Arabic dialects:

The first wave of Arabs arrived armed on horseback, apparently without their families, in the 7th century AD. This band was composed of mixed urban Arabians (presumably Meccans) and “Middle Eastern” Muslim converts–soldiers hailing from Syria, Mesopotamia, Persia, Egypt and elsewhere–who introduced Perso-Arabian Islamic material culture and sedentary Arabic varieties to Roman Africa, the two Mauretaniæ and Hispania. In 670 A.D., the city of Kairouane (from Middle Persian 𐭪𐭠𐭫𐭥𐭠𐭭 kārawān “military column; caravan”) was established as a headquarters for the Muslims’ expansionist ambitions in the region, wherefrom they “plunged into the heart of the country, traversed the wilderness in which his successors erected the splendid capitals of Fes and…[Marrakesh], and at length penetrated to the verge of the Atlantic and the great desert” (Edward Gibbon). It deserves remark that the very seat of Islamization in the Maghreb bears a Persian name.

The newcomers, numbering, it seems, only in the few thousands, exchanged extensively with their Berber-speaking neighbors, who must have greatly outnumbered them. In the culinary sphere they took up local foods, couscous first among them (Berber ⵙⴻⴽⵙⵓ seksu → Arabic كسكسو ksksu “couscous”) and in return, they transferred tastes from the orient (particularly Persian cuisine, discussed below). However, owing to the very fact of their small numbers, apparently not enough correct Arabic input was available to language learners, which resulted in significant interference from Berber in the learning process. For example, it has been suggested that the agentive m- prefix arose due to overgeneralization of a corresponding Berber agentic prefix (ⴰⵎ am-) with a broader scope of use in the first stages of language learning (e.g. مزيان mzyān “good” from زين zayn; including exceptional cases that are actually derived from Berber roots; e.g. Shilha ⴰⵎⵥⵍⵓⴹ amẓluḍ “poor” → مزلط məzluṭ “poor”). The fəʕʕāl noun class tells the same story from the other side: unproductive in Classical Arabic, where it is confined to occupations (نجار nəjjār “carpenter”), it was seized upon by Berber bilinguals as though it were the general participle, and Darija now says xəwwāf “fearful” on the pattern of “carpenter”

This wave was followed by a period of intense Islamization of the “pagan” autochthones, whilst per Quranic prescriptions the ahl al-ḏh̲imma (Iberian Christians and Jews, and the Jews of the Berber lands) were largely left alone. For centuries thereafter Arabs and Berbers lived in uneasy communion, the latter frequently serving as clients to the former, until various Berber-led rebellions toppled the Umayyad Caliphate by the middle of the 8th century. Arabic vernaculars remained essentially confined to the cities the Arabs inhabited, while in the countryside Berber languages remained the means of inter-community communications until present times. The turning point came in the 9th century, after the foundation of Fes as the Idrisid capital: thousands of refugees from Andalusia and from Kairouan settled in the city, founding its great mosques and madrasas (al-Andalusiyyin “of the Andalusians” and al-Qarawiyyin lit. “of the Kairouanians” mosques, c. 859 A.D.). In a sociolinguistic sense the influx meant that there was suddenly abundant native Arabic input available to learners, as well as a social motivation to speak “pure” Arabic, which favored the leveling of salient Berber transferences. It probably served to reverse the process of creolization in urban dialects which had been underway before this period. However, the few exceptional items that survived were probably too frequent by that time to be leveled by prescriptive Arabic influences (e.g. مزيان mzyān “good”; جوج juj “two”; imperative اجي aji “come!” through elimination of irregular تعال taʕāl “come).

The urban Arabic dialects in Morocco which descend from the first wave of Arabization include those of Fes, Salé, Rabat, Meknès, Marrakesh, Tétouane, Chefchaouen, Tangier, Ouazzane, Taza, as well as the Jebli dialects in the Rif region. Across modern political borders, the dialects of most old cities in the Maghreb (Tlemcen, Constantine, Tunis, Kairouane, Oran, Tripoli etc.) are included in this group. Although a minority dialect, Fessi Arabic enjoys prestige in Morocco to this day, for it is the speech of the old urban patriciate, the merchant and scholarly families of Fes; and when that class migrated over the last century to Casablanca, drawn by the new economic capital and its French schools, it carried the dialect with it as a mark of station.The core vocabulary of the urban Arabic varieties in the Maghreb often corresponds to urban dialects further east, such as in the case of Fessi Arabic, vis-à-vis Casablancan Koine which is of Bedouin provenance:

English	Moroccan Darija	Fessi Arabic	Damascene Arabic
“He did”	dār	ɛamal	ʕamal
“I liked”	bġit	ḥabbit	ḥabbeyt
“heart”	gəlb	ʔalb*	ʔalb
“What’s your name?”	ašnu smītek	šnu ʔesmek	šu ʔesmek
“Listen!”	ṣnnit	smaʕ	ismaʕ
“Good”	mzyān	mliḥ, mzyān	mniḥ
“Here! (take this!)”	hāk	khūd	khud

Pre-Hilali dialects like Fessi and, through influence, Casablanca Koine, make the most use of analytic morphology such as the analytic genitive instead of the constructed genitive which is used in ʕarūbi lects. A few salient Pre-Hilali morphologic innovations are listed below:

1. In Koine, the construct genitive is no longer productive and is used only in certain relatively frozen constructions, having been replaced by dyāl or d– (e.g. L-malek d’l-mghrib “The King of Morocco”).
   1. Some ʕarūbi glosses with the constructed genitive are preferred in Koine (akhūti “my brothers”) in contrast to the more innovative Pre-Hilali forms used in the northern city dialects (khāwa dyāli “ibid.”)
2. Analytic elatives using ktar “more” have replaced the Classical Arabic afʕal construction : bnīn ktar “tastier” instead of *abnan. 
3. Indefinite singular nouns employ a gender neutralized prefix واحد ال waḥd l- in lieu of a gendered suffix (this development parallels the Medieval Baghdadi prefix فد fad “a” from فرد fard)
4. An analytic dual construction emerged which has supplanted the inherited suffixed dual. It uses the word جوج juj “two” (from Classical Arabic زوج zawj “pair; couple” < Greek ζεῦγος zeûgos “yoke” via Aramaic ܙܘܓܐ zawgā). 
5. The feminine singular agreement for inanimate plurals in Classical Arabic was replaced by masculine gender and plural number: bināyāt zuwīnīn “beautiful buildings” (cf. Classical Arabic bināyāt jamīla)
   

English	Moroccan Darija	Modern Standard Arabic
“Beautiful things”	Shiḥwāyej zuwīnīn	Ashyāʔ jamila
“Many things”	Bzzāf dyāl ḥwāyej	Ashyāʔ kaṯīra
“Tasty, tastier”	Bnīn, bnīn ktar	Laḏīḏ, alaḏḏ
“One man”	Waḥd r-rajel	Rajul wāḥed
“Two women”	juj d’l-ʕiyālāt	imraʔatān
“Our house is more beautiful than their house”	D-dār dyālna zuwīna ktar mn d-dār dyālhum	Baytuna ajmal min baytihim

As in the other dialects, new formations arose from native Arabic materials, albeit with little concordance with the eastern Arabic idioms: باش bāš “in order to”, from بِأَيِّ شَيْء‎ (biʾayyi šayʾ, “with what thing”); لاباس labās “well, good” from Arabic لَا بَأْسَ‎ (lā baʾsa), the verb noun of بَؤُسَ‎ baʾusa “damage, calamity”; كاين kāyin “there exists” (from كائن kā’en “existing”; cf. Medieval Baghdadi Arabic اكو aku from يكون yakūn). Some further innovations include:

1. Use of ka- or the variant ta- for progressive tense; e.g. ka-nktəb “I write”, ta-tbġini “You love me”
2. The future tense is constructed using ġādi or simply ġa- (cf. Classical Arabic ġadan “tomorrow”)
3. First person marker with n- and first person plural use of prefix n- + suffix -u; e.g. nmši “I go”, nmšiu “We go”. This reshaping of the inherited paradigm is the signature isogloss of western Arabic idioms, from Libya to Morocco to the Arabic of medieval Sicily and al-Andalus.
4. Gender has been leveled in the 2nd person singular perfect, with the feminine ending generalized (-i) (note: Pre-Hilali dialects use an immutable pronoun intīna with the masculine 2nd person perfect verb ending)

English	Moroccan Darija	Muslim Baghdadi Arabic	Damascene Arabic
“I just came to tell you that I love you very much”	Ġīr jīt bāš ngullək ka-nbġīk bzzāf (variant ta-nbġīk)	Bas ijeet ʕalamud agullak innu āni aḥebbek hwāye	Bas ijet ʕashān aʔullak innī baḥebbak ktīr
“There is or there isn’t”	Kāyen wla ma kāyenš	Aku lo māku	Fī aw mā fī
“I will dance too”	Ḥta ana ġādi nšṭaḥ	Āni ham raḥ arguṣ	Ana raḥ arʔuṣ kamān
“Because”	Ḥet, ḥetāš	liʔan	leʔannu
“We wanted to sleep there”	Kenna nbġīu nnəʕʕsu təmmāk	Činna nrīd ənnām əhnāke	Kān biddna nnām hunīk
“If you were a true Moroccan, you shouldn’t have sold your country to foreigners”	Kūn kenti maḡribi quḥḥ, kūn ma bʕtiš l-blād dyālek lgwar	Lo činit maghrebi ḥaqiqi, ma chān baʕet blādek lil faranj	Iza kent maghrebi ḥaʔiʔi, ma kān beʕt baladak lil faranj

On the Persian Elements in Morocco

Old urban dialects, such as that of Fes, have imbued Moroccan Darija with words of early imperial Islamic origin which recall the Orient, including: بالزاف bzzāf “very; a lot” (from Classical Arabic بالجزاف bil jazāf, itself from Persian گزاف gazāf “exaggerated, excessive”) and ڭاوري gāwri (from Persian گاور gâvor “Zoroastrian from Mesopotamia”; later adopted by the Muslims meaning “infidel”), كفتة kafta (from Persian كوفته kūfte, كوفتن kūftan “to beat, grind, shatter”), شوربة shurba “soup” (from Persian شوربا šurbâ “a kind of stew”, lit. “salty/sour stew”); مݣانة magāna “clock, watch” from Persian پنگان pengān “water clock” (cf. Tunisian Arabic منڨالة mungāla), شرجم،سرجب sarjam/sharjab “window”  from Persian چارچوب čârčub “frame”; note Libyan Arabic  روشن‎ rōšan, “window” from Persian روزن rowzan‎ “crevice, hole”); بزطام bizṭām (cf. Algerian Arabic تزدام tezdām) from Persian جزدان jozdān “wallet”; سطوان siṭwān “tiled courtyard surrounded by peristyle within the house” (from Persian ستون sotun “column”); خرشوف xarshūf “artichoke” (from Middle Persian 𐭤𐭠𐭫𐭰𐭥𐭯 *xār-čōp, lit. “thorny stick”), خمم xemmem “to think” (irregular variant with -m final of Classical Arabic خمن xammana “to guess, suspect”, maybe ultimately an assimilated root from Persian گمان gumān “supposition; speculation”). Urban dialects like that of Salé have preserved other Abbasid archaisms, such as مارستان māristān “hospital”, from Persian بيمارستان bimārestān. A later Ottoman-mediated Persian stratum is also discernible: طربوش ṭarbūsh “fez hat” (from Persian سرپوش sar-pūš “head-cover”), beside the much older بابوش bābūsh “slipper” (from Persian پاپوش pā-pūš “foot-cover”).

(1) A Moroccan man clad in traditional striped jeleba fills a bottle of water at Fontaine Nejjarine, Fes, Morocco (c. 18th century) (2) Horseshoe arches with polychrome glazed tiles decorate the Bab Bou Jeloud (built by the French in 1913), with a view of the Bou Inania minaret and old medina, Fes, Morocco (3) Polychrome glazed tiles with geometric motifs, resembling those in Persian architecture, at Dar el Bacha museum, Marrakesh, Morocco (4) The author (right) and his brother in Fes, Morocco (c. 1998)

The presence of Iranian loans might at first seem unremarkable, given the many hundreds of Persian words that entered Arabic itself, directly or through Aramaic, after the Islamic conquest of Sasanian Persia. Persian, Greek, and Aramaic are the principal donor languages of Classical Arabic, though only the latter two are commonly emphasized; and Persians played a seminal role in the standardization of Arabic as a literary medium, a primacy the Arab historiographer Ibn Khaldūn himself recorded in detail. On that account, a Persian word in Moroccan speech should be no more surprising than one in Cairo or Damascus. But the words assembled below resist that explanation: many are unattested in the Arabic vernaculars further east. Had they traveled inside common Arabic, they would be found wherever Arabic is found; their confinement to the Maghreb indicates instead that they arrived by their own road, and, taken together with the material and onomastic evidence detailed below, points to a small but influential Iranian element in early Islamic North Africa.

In light of the drastic shortage of historical records on these migrations, the presence of Persian names among intellectuals and dynasts of the Islamic period in North Africa lends credence to that idea. Apparently their descendants founded two dynasties in the Maghreb–the Rustamids (777–909 A.D.) were Persian Ibāḍī imāms, founded by ʿAbd al-Raḥmān ibn Rustam, whose very name preserves that of the Shāhnāmeh’s great hero; their capital Tahert, in modern-day Algeria, was known to contemporaries as ʿIrāq al-Maghrib, “Mesopotamia of the West”, and even Balkh al-Maghrib, “Bactria of the West”, and it is telling that when men sought a comparison for the city, they reached for the Iranian east. A century and a half later, the Khurāsānid dynasty (Bānū-Khurāsān, 1059-1148 A.D.) founded by ‘Abd al-Ḥaqq ibn ‘Abd al-‘Aziz ibn Khurāsān and, as the name suggests, of reputed Khurāsānī descent, ruled an independent principality in modern-day Tunisia. Persian immigrants or those with Iranian nisbas appear further afield in Andalusia (e.g. Ziryāb), detailed below. 

It is notable that west of Baghdād (from Middle Persian 𐭡𐭢𐭣𐭲‎ (bgdt /bag-dād/, “given by god”; compare Russian given name Богда́н Bogdan “ibid.”), nowhere in the medieval Islamic world did Iranian forms gain such currency as in the Maghreb. Perhaps this owed in part to the nature of urbanity in Anatolia, Egypt, Palestine and Greater Syria, where for centuries the materials of Greco-Roman civilization reigned supreme among a Romanized Christian citizenry. Indeed Roman churches and civic buildings in these regions were frequently dismantled by the Muslims for reusable materials or converted to mosques, sometimes by little more than the addition of a mihrab and minaret (e.g. the Umayyad Mosque of Damascus which was formerly the Basilica of Saint John the Baptist; Greek: βασιλική του Αγίου Ιοάννή του Βαπτιστή Vasilikí tou Agíou Ioánni tou Vaptistí). Islamic architectural forms in these regions employed nearly identical building materials (limestone ashlar) and ornamentation schemes (marble veneer decorations, Greek order columns) to the forms that preceded them, which were Roman in character. 

The Maghreb was by contrast comparatively virgin ground for the Muslim immigrants, who founded a significant number of new cities which they cohabited with Berbers. These new spaces invited the transmission and local innovation of eastern architectural vocabulary, fashions and culinary stylings. For example, the adoption of the quadripartite Persian garden form (Persian: چهار باغ chahār bāgh; Arabic: رياض riāḍ), itself an ancient Iranian inheritance, was most marked in the Maghreb, and it became a prominent feature in Moorish palaces in Spain (such as at Madinat al-Zaḥrā, the Aljafería, and the Alhambra). In the 9th century, the potters of Iran had learned to fire colored tiles with metallic pigments over a glaze of tin oxide, which heralded a renaissance of majolica tile production in the Persianate world and reached, it seems, as far as the Maghreb. The technique behind Maghrebi zellīj, the firing of glazed and lustered tile, belongs on this evidence to the Persianate tradition, whatever the local genius that later shaped it into cut-tile mosaic; and the rival derivation, that the figural Roman mosaic of tesserae depicting humans and animals somehow fathered the abstract Islamic zellīj, strains belief. The latter hypothesis demands, further, that Persian and Maghrebi tilework, strikingly alike in composition and design, arose twice over, at one moment, and each in ignorance of the other. And it must then explain why the art was absent in Syria, Palestine or Egypt, where Roman and Byzantine mosaics stood in the greatest abundance, and appeared instead at the far end of the Mediterranean, where they were fewest.

Evidence of Persian craftsmen and architects exporting their crafts abounds in the Arab world; for example, the glazed tiles on several Cairene buildings, including the bulbous stone minarets of the mosque of Nāṣer Moḥammad in the citadel (1318-35 A.D.), indicate that craftsmen from Tabrīz operated a workshop in Cairo in the 1330s and 1340s. Double-shelled, ribbed domes ending in muqarnas corbels (stalactites) with high drums banded with calligraphy are Persian features to be seen on the Mausoleum of al-Ṣultaniyya in Cairo, and which are later encountered in Timurid Samarqand. What the Maghreb took from this eastern current, however, was selective. The characteristic Persian four-ayvān plan, which conquered the madrasas of Egypt and Anatolia, was not adopted in the far Maghreb, whose madrasas kept the single courtyard; it was instead the Persianate surface arts that took hold. For example muqarnas, an occasional imported grace-note in Cairo, became in the west a native industry, carved in stucco to a refinement that culminated in the honeycomb vaults of the Alhambra. Moreover the Maghreb built, as Persia built, predominantly in baked earth and brick, surfaces that invite glazed tile and carved plaster, whereas Egypt and Syria built in limestone ashlar dressed with marble veneer, surfaces that refuse them. Thus Persian elements appeared throughout Maghrebi architecture, and by no mysterious route: men who had seen Persian buildings, and men trained by them, must have been present in North Africa to build from memory.

Moroccan culinary tradition attributes the widespread use of nuts like almonds, pistachios and walnuts, dried fruits such as raisins, prunes, dates and apricots, pickled lemons (حامض مصير ḥamudh mṣir) and saffron to Persian influence. Indeed, the two cuisines share an idiosyncratic palette, sweet, savory and sour in one dish, that sets each apart from its immediate neighbors. Moreover, it is known that Muslims introduced saffron from Persia to North Africa in the 10th century. The pattern is the same one the loanwords show: forms that are faint or absent in Egypt and Syria appear full-strength in the Maghreb, as though the current had passed through the intervening lands without settling there; and the likeliest carriers are, again, the few but influential Iranians among the early Muslims of the west.

It appears, then, that a thin but consequential Iranian thread runs through the foundation layer of Islamic North Africa, legible at once in the name of its first city (Kairouane), in two of its dynasties (the Rustamids and the Khurāsānids), in a loan stratum, in the Persian garden form, in architectural vocabulary and a tile tradition, and in cuisine. Indeed, the thread was taken up again, centuries later, when Ottoman rule brought a second wave of Persianate civilization to Algiers, Tunis, Tripoli, and other urban centers.

Andalusian Influence

From the 15th century onward, the old Maghrebi cities (particularly Tangier, Fes and Rabat, and more broadly Constantine, Tlemcen, Kairouan, Tunis, Tripoli and Bizerte) received a massive influx of Muslims and Jews expelled from Islamic Spain (Arabic الاندلسيين al-Andalusiyyīn; the final wave, the nominally converted Moriscos, arriving after the expulsion of 1609), who migrated “back” to a Maghreb most had never seen and settled among their distant dialectal cousins. This second urbanite wave further shaped the urban dialects, while leaving the ʕarūbi speech of the countryside largely untouched. Andalusian families, still distinguished by their surnames–Jorio, Fenjiro, Chkalante, Guedira, Gharnaṭa, Qorṭoubi, Andalousi, Prado, Vergas, Lavra–brought features of Spanish and Andalusian Arabic to the local vernaculars. The ​​طرب أندلسي ṭarab ʾandalusī (and by extension غرناطي gharnāṭi “Granadan” in Tlemcen, from the Arabic name for Granada) is attributed to a certain Ziryāb, a freed Persian slave-turned-courtier and musical pioneer who introduced Persian instruments and musical modes to Muslim Spain. The Andalusian repertoire, with its foundations in classical Persian and Spanish music, is thus distinguished from ʕarūbi (what has been termed shaʕbi “folk”), Gnāwa and Berber folk music, which employ different instruments, formats and modes. The Gnāwa repertoire in particular, which features the guembri lute and the call-and-response format of vocal performance, links it with the musical systems of the Songhay in Timbuktu, Gao and Djenné.

A number of old Iberian Romance words entered the Fessi dialect, the legacy of large numbers of Andalusian Muslim and Jewish refugees who migrated to Maghrebian cities after their expulsion from Islamic Spain, some of which have been adopted in Casablancan Koine. These include سيمانة simāna “week” from Spanish semana; Fessi كوشينة kušina “kitchen” from Spanish cocina (doublet with darija كوزينة kuzina, probably a later borrowing during the Spanish protectorate); اشكولة iškuila “school”; from Spanish escuela; قرة qirra from Spanish guerra “war”; فورنو furnu “oven” from Spanish forno; شلية šelya “chair” from Spanish silla; طابلة ṭabla “table” from Spanish tabla; فيشطة fishṭa “party”, from Spanish fiesta; كريلو grīllu “cockroach” from Spanish grillo “cricket”; قميجة qamija “shirt”, from Spanish camisa; بسطيلة basṭila from Spanish pastilla “puff pastry with various fillings”; فابور fābor “free [of charge]; a favor” from Spanish favor; rwina from Spanish ruina “mess, wreck”, qīmrūn “shrimp” from Spanish camarón. In cities like Salé where the Andalusians settled, Spanish words penetrated even the core vocabulary, though these items have not passed into the koine, owing to their salient foreign nature:

English	Salawi Arabic (Salé)	Spanish
“Good”	buinu	bueno
“Wrong”	falṣo	falso
“Luck”	as-suirti	suerte
“Alone”	ṣulu	solo

The descendants of the Andalusian Jews, who called themselves megorashim, “the expelled”, retained their Judeo-Spanish vernacular (Ḥaketía and Ladino) in the northern cities until the community’s dispersal, to Israel, France and the Americas, in the mid-20th century.

Bedouin (Hilalian) or ʕarūbi Dialects: the Basis of Moroccan Darija

The second wave of Arabic speakers were Bedouins of the Banu Maʕqil, Banu Hilal, Banu Sulaym and Bani Ḥassan tribes who migrated en masse in the 11th and 12th centuries from the Najd region of Arabia to the Maghreb, where they took home in the arid plains and hautes-plateaux. The impetus for this forced migration was the Fatimid Caliphate’s desire to confine the unruly Bedouins in the south of their dominion. As the Bedouins were unaccustomed to urbanity, the Banu Hilal settled in areas where they continued their nomadic lifestyle, and this wave brought significantly greater linguistic arabization and spread of nomadism in areas where agriculture had previously dominated. Camps (1983) estimates the number of Bedouins to have been relatively small; only around 80,000 at the outset, and their migration extended over two centuries and not all of them settled in Morocco. Local Berbers (most likely belonging to the Masmuda confederacy and thus speaking a language related to Shilha, given the Bedouins’ settlement in the lowlying plains between Souss and Rabat), who shared a similar lifestyle of pastoralism, were quickly assimilated to the language of the incomers. The nature of their contact was such that, over time, the Arabs adopted the modified Arabic of their Berber neighbors and started to speak it amongst themselves. The whole community then passed this speech to its children, and the learners’ variety became the mother tongue of all. With the establishment of Casablanca, it was the ʕarūbi migrant workers and their dialects, rather than the old sedentary dialects such as Fessi, that became the basis of Moroccan Darija.

In contrast to the sedentary dialects which use the local innovation ka- or ta-, ʕarūbi dialects particularly in northeastern Morocco and Algeria developed a progressive aspect marker and by extension copula -را rā– (lit. “he saw”), probably from Najdi ترا tarā “to be seen”– : e.g. rāni nkteb = “I am writing” (lit. “He saw me writing”); rāk mn wjda “you are from Oujda”. Some comparisons between Moroccan Darija and Najdi Arabic are made below:

English	Moroccan Darija (Oujda)	Najdi Arabic
“Youth, young generation”	drāri	dhrāri
Interrogative particle	wəš	weš
“What my heart wanted (f.)”	Lli bġaha gəlbi	Illi bġaha galbi
“Take this”	hāk	hāk
“She is going”	Rāha tmši	Tarāha temši

With regard to Berber elements in ʕarūbi (Bedouin) dialects, examples of both metatypy and phonological restructuring exist, which is possible in extended and complex contact situations. The Arabization process presumably started with the Arabs’ neighbors, and then these Arabized Berbers served as intermediaries to their next neighbors, etc.  A few examples of Berber metatypy in ʕarūbi are listed below:

1. The Berber feminine/diminutive circumfix ta….t used mainly with nouns of occupations in Moroccan Darija (e.g. bənnay “mason” –> tabənnayət “profession of masonry”) but also for abstract noun derivations, e.g. dərri “child” –> tadərrit “childhood”
2. The Berber genitive marker n “of”, though limited to some kinship terms like bb°a n Sufiān “father of Soufiane”

On Berber (Amazigh) influences in Darija

Berber (endonym: ⵜⴰⵎⴰⵣⵉⵖⵜ tamazīɣt; Ar. الامازيغية al-amazīġīyya; الشلحة aš–šilḥa) is a distinct language family belonging to the Afro-Asiatic phylum. Blench (2018) notes that Berber is considerably different from other Afroasiatic branches, and indeed its genetic relation to Arabic–of equal magnitude to its relation with Ancient Egyptian and Cushitic–is vanishingly remote. Proto-Berber probably split from a common ancestor with the other known Afro-Asiatic branches between ~10,000-9,000 years before present, and then this group underwent a putative linguistic bottleneck as recently as ~3,000 years ago, which apparently eradicated the internal diversity of the family and left a contracted number of idioms from which all of the modern Berber languages descend. Proto-Berber then appears to have spread across a vast area from the western banks of the Nile river to the Atlantic coast of North Africa. Today, an estimated 30-40% of Moroccans, 20-30% of Algerians, 5-10% of Libyans and 1% of Tunisians use a Berber language as a mother tongue, but these numbers would have been higher only a century ago. 

The region was visited from an early date by seafaring Phoenician merchants from the Eastern Mediterranean (but, apparently, few Greeks), who established important port colonies (e.g. Shilha ⴰⴳⴰⴷⵉⵔ Agadir, from Phoenician 𐤂𐤃𐤓‎ gādīr “wall;compound”, cognate with Arabic جدار jidār “wall”; Rusadir from Phoenician 𐤓‬𐤔𐤀𐤃𐤓‬ rushādir “powerful). These immigrants later established the Carthaginian empire and perpetuated Punic language and culture, eventually subjugating the Romanized para-Berber kingdom of Numidia. The presence of Punic borrowings in Proto-Berber points to the diversification of modern Berber language varieties subsequent to the fall of Carthage in 146 B.C.; only Guanche and Zenaga lack Punic loanwords. Berberophones along the Mediterranean coast later became subjects of the Roman Empire under administrative divisions of Tripolitania, Africa, Mauretania Caesariensis and Mauretania Tingitana, but commercial and cultural exchanges between Romans and Berbers along the Līmes Mauretaniæ persisted for centuries through the Byzantine period. 

During this age of contact, Roman innovations including the ox-plough, camel (Kabyle: ⴰⵍⵖⵯⵎ a-lɣəm via metathesis from Latin camēlus or Coptic ϭⲁⲙⲟⲩⲗ (camoul)?), and orchard management were adopted by Berber communities. Latin loanwords in Berber include, notably, ⴰⴼⵓⵍⵍⵓⵙ afullus “chicken” (from Latin pullus); ⴰⵙⵏⵓⵙ asnus “donkey” (from Latin asinus), ⵓⵔⵜⵉ urti “garden” (from Latin hortus); ⵜⴰⵖⴰⵡⵙⴰ taɣawsa “thing” (from Latin causa “reason/case; motive”); ⴰⵏⵖⴰⵍⵓⵙ anɣalus “angel, spiritual entity” (from Latin angelus); ⵉⴳⴻⵔ iger “cultivated field” (from Latin ager); ⵜⴰⴼⴰⵙⴽⴰ tafaska “feast/religious celebration” (from Latin pascha “Easter; Passover”); ⴰⴼⴰⵍⴽⵓ afalku “bearded vulture” (from Latin falcō “falcon”), ⵜⴰⴼⴻⵙⵏⴰⵅⵜ tafesnaxt “carrot” (from Latin pastinaca “parsnip, stingray”), ⵢⴻⵏⵏⴰⵢⴻⵔ yennayer “January; first month of the Berber New Year” (from Latin ianuarius), inter alia.

Berber is the substrate language for the Arabic varieties spoken in the Maghreb. Nonetheless, it has left few lexical borrowings in Arabic varieties, with the most notable exception of toponyms. The overwhelming majority of place names in Morocco are of Berber origin, e.g. ⴰⵎⵓⵔ ⵏ ⴰⴽⵓⵛ amur n akush “Land of God”, whence Marrakesh and Morocco; ⵉⵎⵉ ⵏ ⵜⴰⵏⵓⵜ imi n tanout  “mouth of a small well”, ⵉⴼⵔⴰⵏ ifran “caves”; ⴰⵎⴽⵏⴰⵙ amknas whence Meknes “the Miknasa Zenata Berber tribe”; ⵜⵉⵟⵟⴰⵡⴰⵏ tiṭṭawan “springs” (with the Ghomara Berber feminine plural suffix ⴰⵏ –an) whence Tetouane; ⵜⵉⵏⴳⵉ tingi “marsh”, whence Tangier; ⵡⵔ ⴰⵣⵣⴰⵣⴰⵜ ur azzazat “without clamor; silent” whence Ouarzatate, etc.

Other ancient local words may include ݣناوة gnāwa “Sub-Saharan Africans and their musical repertoire in Morocco; a caste of largely endogamous black Africans living among both Arabophones and Berberophones”, probably from a Shilha gloss ⴰⴳⵏⴰⵡ agnaw “deaf-mute” (pl. ⵉⴳⵏⴰⵡⵏ ignawen) and later “black people” (semantic development resembles Arabic عجم ‘ajam “confusing or unclear way of speaking” → “Persians”). This word is probably also the origin of “Guinea” and perhaps “Djenné”.

In the domestic and culinary spheres, there exist a few important loanwords: كسكسو ksksu “couscous” from Berber ⵙⴻⴽⵙⵓ seksu; ساروت sārūt “key”, from Central Atlas Berber ⵜⴰⵙⴰⵔⵓⵜ tasarut; and صيفط ṣīfṭ “to send” (cf. Kabyle ⵙⵙⵉⴼⴻⴹ ssifeḍ). Some local flora and fauna have also retained their Berber names: زايس zāyis “octopus” from Berber ⴰⵣⴰⵢⵣ azayz; مشّ mušš “cat” (cf. Central Atlas Tamazight ⴰⵎⵓⵛⵛ amušš); لالة Lalla “lady; respected woman”, (related to Central Atlas Berber ⴰⵍⴰⵍⵍⵓ alallu “dignity”, from ⵜⵉⵍⴻⵍⵍⵉ tilelli “freedom”); دشار dšār from ⴷⵛⴰⵔ dšar “region”; شلاغم šlāḡem “mustache” from Berber ⴰⵛⵍⵖⵓⵎ ašlɣum. 

It is possible that the use of feminine gender language names in Darija is Berber substrate: الريفية al-rifiyya “Riffian language”, cf. ⵜⴰⵔⵉⴼⵉⵜ Tarifit “Riffian language”. Additionally, Moroccan Darija exhibits a preference for broken plurals, particularly for ethnonyms or nationalities endemic to the region. By contrast, geographically distant peoples use the regular jarr plurals: مغاربة mġārba  “Moroccans” (cf. Modern Standard Arabic مغربيين maġrebiyiin “Moroccans”) , تونسة twansa “Tunisians”, فواسة fwāsa “Fessis”, ريافة riyāfa “Riffians”, شلوح šluḥ “Shilha Berbers”, سواسة swāsa “people from the Souss region; frequently used synonymously with Shilha Berbers”, جبالة jbāla “Jbelis”, but مصريين maṣriyiin “Egyptians”, عراقيين erāqiyiin “Iraqis”, etc. It is interesting that Berber too uses complex and apophonic broken plural formations, like other Afro-Asiatic branches: ⴰⵎⵖⵔⵉⴱⵉ amɣribi → ⵉⵎⵖⵔⴰⴱⵉⵢⵏ imɣrabiyen “Moroccans”; ⴰⵛⵍⵃⵉ ašllḥi → ⵉⵛⵍⵃⵉⵢⵏ išlḥiyen “Shilha Berbers”, ⴰⴳⵍⵍⵉⴷ agllid “king” → ⵉⴳⵍⴷⴰⵏ igldan “kings”; ⵜⴰⴳⵍⴷⵉⵜ tagldit kingdom → ⵜⵉⴳⵍⴷⵉⵏ tigldin “kingdoms”.