Navigation – Plan du site
Parlers contemporains

Rural Dialect of Egyptian Arabic: An Overview

Manfred Woidich
p. 325-354

Texte intégral

  • 1 See Vollers, 1890, Nallino, 1900, Willmore, 1905, and the dictionary by Spiro, 1895.
  • 2 See for instance Gairdner, 1926, Elder, 1927, Mitchell, 1962, Salib, 1981.
  • 3 See Woidich, 1989a.

1In the last 30 years of the 19th century, European Arabists felt a need to describe and analyse the Arabic colloquials in order to use them for comparative and historical linguistic goals. It was the time when the Middle East opened for Europeans, and Egypt in particular became accessible to thousands for trade and other professions and, last but not least, research of all kinds. This is why the first comprehensive and, by contemporary standards, scientific grammar of an Arabic Colloquial appeared as a description of the colloquial of Egypt, i.e. the Grammatik des arabischen Vulgär dialects von Ägypten by Wilhelm Spitta-Bey (1880), the former director of the Khedivial Library in Cairo. Others soon followed and by the beginning of World War I, a considerable amount of literature on the so-called Egyptian Colloquial had appeared 1. After World War I, the interest focused mostly on textbooks and teaching material 2. Today, a host of publications on Egyptian Arabic is available and a, certainly not exhaustive, bibliography of Egyptian Arabic collected by me some years ago contains no less than 680 books and articles 3.

  • 4 That the Cairo Arabic of the 19th century showed many features which today are considered rural is (...)

2Most of these works claim to describe the Arabic dialect of Egypt, thus giving the impression that there is only one. In fact, they all describe a variety of the elevated colloquial of Cairo which was certainly in the 19th century, as it is today, the standard Egyptian dialect. Spitta was aware of this and readily acknowledges in his introduction that the title of his grammar Grammatik des arabischen Vulgär dialects von Äegypten is too unprecise, given that he describes only the language of educated Cairo, and not the many Arabic dialects spoken in the countryside outside the capital. All the early grammarians such as Spitta, Vollers, Nallino and Willmore give some hints on the peculiarities of the speech of tne fellahin and Bedouins without going into any detail 4.

  • 5 Winkler, 1936, and for the village Kimàn near Guff in Upper Egypt, see Winkler, 1934.
  • 6 Tomiche, 1962, is dedicated to the same topic, but does not go beyond Winkler, 1936.

3The first real account of rural Egyptian Arabic we owe to somebody eise. In the early 1930s, the German anthropologist Hans Alexander Winkler did his research on the material culture of the fellahin in some thirty villages all over Egypt, and in particular in Upper Egypt. He published his findings in his monumental Ägyptische Volkskunde 5. Winkler had a good command of both spoken and written Arabic and registered faithfully in a rather precise transcription all the terms for the tools the peasants use, in particular the parts of the plough, different kinds of sickles and hoes, harrows, pottery, parts of the oven and so on. He even collected an exhaustive list of the terms for the parts of the human body He did not restrict himself to recording the data, but tried to describe the distribution of the dialects of the fellahin throughout Egypt in order to corroborate his ethnographic findings. The features he based his classification on were primarily lexical. He considered only one phonological feature and no morphological features at all. Nevertheless, he managed to give a basically correct and still useful first account of some very important isoglosses which separate The Arabic dialects of Egypt 6.

4It took some thirty more years for a new attempt to be made to investigate the basic features of a group of rural Egyptian dialects and, for the first time, to publish the results in the form of maps. Fahmi Abul-Fadl, an Egyptian scholar and a student of the well-known Hans Wehr, describes in his doctoral dissertation of 1961 the speech of the fellahin of his native Sarqiyya province in the east of the Delta. He offers a profound phonological analysis and a series of maps together with samples of texts in transcription. As to Upper Egypt, a Sa'îdi grammar was published by Khalafallah in 1969, and a short account of B’eri Arabic was given by the present writer in 1974.

5Fahmi Abul-Fadl’s dissertation is the forerunner of the dialect atlas of Egypt which was planned and carried out by Peter Behnstedt and myself. After a period of field work which lasted from 1976 to 1983, a map was published in 1984 in the Tübinger Atlas zum Vorderen Orient (TAVO), and the atlas itself eventually appeared in 1985 together with a collection of transcribed and translated texts (1987,1988). This atlas covers most parts of Egypt and contains 561 maps which are based on data gathered from approximatively 800 villages. A comprehensive Arabic-German glossary of rural speech followed in 1994. The reverse, a German-Arabic glossary,

6a comparative Grammar, and some grammars of local dialects are in preparation. We may now say, that the most important features of the rural dialects of Egypt and their distribution throughout the country are at least rudimentally known.

The main regions


7By “rural dialects” we mean the dialects of the peasants in both northern Egypt (fallahîn) and Upper Egypt (sa'ayda}, as well as those of the inhabitants of the oases in the Western Desert.

  • 7 See Winkler, 1936, p. 391, and the word list p. 343-338, and Littmann, 1920.
  • 8 See Winkler, 1936, p. 388, 391, and personal communication by Georg Krotkoff.
  • 9 Winkler, 1936, p. 382 f., Vycichl, 1953.

8We do not consider here the varieties spoken by true Bedouins who arrived recently in Egypt or still live their nomadic live style because these forms a separate subject and will be treated elsewhere. Neither shall we deal here with the dialects of some marginal paria groups such as the Halab and the Nawar in Middle and Upper Egypt of which but little is known. The speech of the Nawar, though basically Arabic, contains many foreign words, some apparently of Indo-Aryan origin: mans ~ minis, “man, father”; burdi, “brother”; astra, “star”; ag, “fire”; dandi, “teeth” 7. The Nawar are said to be of Gipsy origin; as to the Halab, Winkler assumes that they descend from Upper Egyptian peasants expelled by 'Abàbda and Ga'âfra tribes from their original homelands. For them a kind of “pig-latin” is reported: nouns are prefixed with ma- and suffixed with -is, -âis for instance ma-wdàn-is (widân), “ear”, muxx-âis (muxx), “brain”, me-snâb- is (sanab), “moustache” 8. Another group which will not be considered here are the 'Abàbda in the Eastern desert. They speak a Sudanese type of dialect, with rigi'lha, “her leg” (stress on inserted vowel) and with a stressed final vowel in habli, “my rope”, and hamrâ (f.), “red”. Their lexicon is interspersed with cushitic words: ba'asùb, “jackal”, dangar, “plain”, kîsi, “heel”, helèkîb, “young donkey”, kakar, “viper” 9. Presumably they were originally speakers of a cushitic language who adopted Arabic some generations ago.

Dialect groups

9By means of certain phonological, morphological and lexical features (isoglosses), some of which will be discussed below, we are able to distinguish the following main dialect areas in Egypt (see following page).

Nile Delta (map I)

Nile Delta (map I)

Western dialects (WD)

WD 1 eastern Bihêra, Kr. išŠêx, parts of ilĠarbiyya

WD 2 Burullus

WD 3 western Minufiyya

WD 4 ilBihêra, western Minufiyya, western Ġarbiyya and parts of northern Gîza

Northeastern dialects (NED)

NED 1 idDaqahliyya, eastern Kr. išŠêx, parts of ilĠarbiyya

NED 2 Manzala dialects

Central dialects (CD)

Centre of the Delta : ilMinufiyya, ilĠarbiyya, ilQalyubiyya

CED: eastern Qalyubiyya, parts of southern Šarqiyya

Eastern dialects(ED)

ED 1 central, northern and eastern Šarqiyya

ED 2 northern frontier area of ilŠarqiyya, parts of idDaqahliyya

ED 3 southwestern frontier area of ilŠarqiyya, southern parts of idDaqahliyya, southeastern parts of the centre of the Delta

Nile Valley (map II)

Nile Valley (map II)

{H}: bundle of isoglosses

Middle Egypt

Northern (NME)

NME 1 southern Gîza, northern Bani Swèf, Fayyûm

NME 2 southern Bani Swèf, northern Minya

Southern (SME) southern Minya.

Upper Egypt (UE)

UE 1 from Asyût (Abu Tiğ) to Nag' Hammâdi, further on the east bank from Gûs further south to the altitude of Armant, on the west bank from ilBallâs to ilGurna

UE 2 mainly the Gina-bow on the east bank approximately from Nag' Hammàdi to Gûs and on the west bank to ilBallâs

UE3 on the west bank from ilQurna ilBi'irât to Isna (B'êri)

UE 4 on the west bank from Isna to Ġarb Aswân, on the east bank from the altitude of Armant to Aswân

Western Desert

BAH ilBahariyya, three varieties: east, centre, west FAR alFarâfira

DAX adDâxila, three varieties: east, centre, west

XAR ilXârga, two varieties: north, south

  • 10 For more details, see Behnstedt-Woidich, 1985 ff, vol. l. See in particular the introductions to th (...)

10Needless to say, the geographic delimitations in this table give only an approximate idea of the situation in the core areas, i.e. the centers of each dialect region. These are relatively homogeneous, show little or no criss-crossing isoglosses, and exhibit most features of the area 10. Although the dialect groupings suggest that there are sharp distinctions between the different regions, this is only occasionally the case - as for instance the border between UE 1 and UE 3 at the west bank of Lugsur. In most cases, however, the transition is gradual. As is true elsewhere in the world, the majority of the dialects described here exist in a continuum and not in precisely delimited geographical areas.

11The transitional areas thus show features of more than one core dialect area. Historical interpretation of their data is difficult, as these areas can go back either to internal linguistic development or to dialect mixing. Which interpretation is appropriate can only be decided by a thorough study of the region's settlement history. As we generally know very little about this, research on such settlement histories remains a desideratum.

  • 11 See Behnstedt, 1979.

12There are relic areas with one or more features which apparently were more widespread in earlier times. These include in particular the region around Abu Girg in northern Middle Egypt (NME 2) 11, which shows a close relationship to the northern Delta (WD 1, NED) in its verbal morphology (see features 18-b, 19, 20 below), and the region of southwestern Sarqiyya and southern Daqahliyya (ED 3), which shows the development of k to c (a unique feature for Egypt), and also has affinities with northern Middle Egyptian verbal morphology (see feature 18-c below). The coastal WD 2 (Burullus, Burg Migîzil} dialect also exhibits unique features such as retention of q and the diphthongs, and stress on the penultimate syllable of darâbu, 'they hit", which give it an extraordinary position within the Egyptian Arabic dialects.

  • 12 See Behnstedt-Woidich, 1982.
  • 13 See Behnstedt-Woidich, 1982, and Woidich, 1993.

13The dialects of the four Western Oases are closely related to ME and the Delta dialects by the bukara-syndrome (feature 9 below) and by syllable structure (feature 7 below), but deviate in many other respects 12. All are different dialects, however, which can be further divided into subdialects. There is no isogloss peculiar to the oases as a whole which would separate them from the Nile Valley. Although their isolated position at the periphery of the Egyptian mainland would make us expect them to be quite conservative, they display a variety of linguistic innovations. These are presumably to be explained by the mixing of groups speaking different dialects (or even different languages, in the case of Berber) in the oases with strange cases of dialect adaptation as a resuit. Indeed, many of the innovations can be explained as interdialect forms 13. Generally speaking, the varieties of iIXârga and ilBahariyya are closer to those of the Nile Valley than adDàxila and alFarafira, while the latter show a greater influence from dialects of the North-Atrican type.

Bundles of isoglosses

  • 14 The letters marking the bundles are writlen between {} brackets. The features mentioned here and in (...)

14The dialect groups listed above are delimited by bundles of isoglosses distinguishing between the features described below in D (see maps 554-559 in Behnstedt-Woidich, 1984), On maps I and II they are represented in the form of thick lines which run through certain areas 14.

  • 15 The numbers of the maps refer to Behnstedt-Woidich, 1985, Vol. 2.

15In the Delta, there are several such bundles which distinguish dialectal areas. An important one runs from the east of Lake Burullus southwards until it reaches the western branch of the river Nile, thus separating WD from NED (A) and from CD (B). Another bundle (C), running southeast from the south of Lake Idku until it joins (A, B), subdivides WD into an eastern and a western group, WD 1 and WD 4 respectively. NED and ED are separated by a bundle (D) which stretches from the Manzala-region southwest to the eastern branch of the Nile (see map 557) 15. Bundle (E), finally, forms the boundary between ED and CD/CED (see map 555) in the eastern part of the Delta.

  • 16 Already isSa'âra, some 25 km to the south of igGiza, shows all traits of an NME 1 dialect.
  • 17 The Hawwâra are attested there from the 14th c. on, see EI III p. 299b.

16In the Nile Valley, there are two regions through which a considerable number of isoglosses run (see maps 554 and 559). The first {F} covers the southern part of the province of igGîza, which is traditionally seen as the old frontier between Lower and Upper Egypt 16. The other large bundle of isoglosses (1) runs immediately to the south of Asyùt and marks a part of Upper Egypt which was not densely populated in the Middle Ages (see map 553) and which was settled by bedouins from the west 17. Not surprisingly, the dialects spoken there (UE 1, UE 3) show strong bedouin substrate. We can thus distinguish two major linguistic areas: Middle Egyptian (ME) from the outskirts of igGîza to Abu Tîg (some 25 km south of Asyùt) and Upper Egyptian (UE) proper from Abu Tîg to Aswàn. Both isogloss bundles coïncide with characteristic ethnographic borderlines found by Winkler (1936) (see below). Another larger bundle (H) in the Nile Valley separates the northern part of Middle Egypt (NME) from the southern part (SME) (around ilMinya, see map 558). Afourth (G) separates the Fayyum and northern B. Swayf areas (NME 1 ) from the southern B. Swayf area (NME 2) in northern Middle Egypt, with a transitional area around the, town of B. Swayf (see map 558).

  • 18 It is not the place here to go into the problems of weighting the isoglosses, i.e. the question as (...)

17As to the relevance of the bundles 18, (1,17) south of Asyùt seems to be the most important with 17 features from the selection in D (see below). Nearly as strong is (F, 13) south of igGîza which separates Middle Egypt from the Delta. In the Delta, (D, 11) and (E, 12) we see that the east (isSarqiyya) differs quite a lot from the centre and the north east. Even stronger is (G, 13), which divides northern Middle Egypt into two regions (NME 1 and NME 2). (H, 10) separates the northern part of Middle Egypt (NME) from the southern part (SME) in the ilMinya region, (B, 9) marks the border area between WD and CD in the south of the Delta. Less marked are the lines separating WD from NED (A, 5) and WD 1 from the western group WD 4 (C, 3). Adducing more features would certainty not change this general picture.

A selection of the most important isoglosses delimiting the dialect areas

18The most important linguistic variables used to delimite the dialect groups are the following:

Phonological features

1) g/ğ(*ğ)

19What we hear in Cairo as g [g] as in gamal “camel” (class. ğ) is pronounced as an affricate ğ or dj in Upper Egypt and the eastern Delta, as d [d] in many villages between Asyùt and Gina, as a sibilant z in the western part of the Delta, in the oases of ilBahariyya (but more prepalatal in ilBawîti), and in alFarâfira (maps 10-14). Going southwards from igGîza, the ğ-region starts some kilometers after the town B. Swayf with the villages of Tansa and B. 'Atiyya [G). In the northern Nile Valley and the Delta, the distribution of g/ğ forms a characteristic image with g covering an area which extends from south of B. Swayf, igGîza following the eastern branch of the Nile in a broad strip to idDumyât. The Delta is thus divided into three areas: the centre CD and north-east NED (with g), the east ED with ğ and the west WD (with ğ ~ ž). The bundies involved are {A, B} and {D, E}.

  • 19 For the distribution of trade posts along this line, see map 551 and Behnstedt, 1980, p. 65.

20The g-area follows the eastern branch of the Nile, i.e. the ancient trade route from Cairo to Dumyàt 19. Its shape suggests that it is either an innovational area with g spreading from Cairo along the trade route, or a relic area, where g, supported by the trade route, was able to resist the influence of bedouin ğ. For a discussion see below. See further feature 2.

2) ’/g (*q)

  • 20 q is pronounced glottalized in ilXârga.

21*q is pronounced as a glottal stop ' or g [g] ('alb - galb, “heart”), although a realization as q [q] is still retained systematically in Burg Burullus and Burg Migîzil at the extreme ends of the Delta and in the oases of alFarâfira. In ilXârga and eastern Dâxila q and g occur in an irregular distribution side by side in different roots (see maps 6-8)20.

  • 21 This is not true for the oases, since alFarâfira oasis has q but z.

22In the Delta and in the Nile Valley, the variables 1 and 2 are related to each other by implication: the reflex g of *q implies the reflex ğ of *ğ, and vice versa. On the other hand, the reflex ' or q of *q implies g for 21. Consequently, the distribution of '/g here shows the same geographical pattern as that of g/ğ (see map 15 and {A,B,D,E, G}) and is to be explained on the same historical grounds.

3) Secondary emphasis

23The western (WD) and eastern parts (ED) of the Delta show a characteristic pharyngealization of r (r) and m (m) {A, B, D, E}: farxa, “chicken” firâx, “chickens”, kamân, “too”, gumàš, “cloth” (comp. Cairo farxa, firàx, kamân, 'umàš. see map 34). This is due to the impact of bedouins who settled in the area.

4) Vowels

24ā: in the eastern part of the Delta, ā is split into clearly marked allophones with a strongly backed [a:] in emphatic environments [ifra:x], “chickens”, and a fronted [e:] or even [e;] elsewhere ([be:b] {E}), the same is true in the south of ilXàrğa. Moving southwards in the Nile valley, ā gradually loses its inclination towards [a:] so common in Cairo in non-emphatic environments, becoming more open as in [ba:b] in the Sohâg region.

25ē and ō : both tend to be diphthongized as [e:"], [o:"] in the Nile valley south of the town Of ilMinya {H}: bēt [be:’t] “house”, yōm [jo:'m] “day”.

5) Diphthongs

26The diphthongs ay and aw are changed to ē and ô respectively as in Cairo b ē t, “house”, môt, “death”. They are retained - at least in certain contexts - in Kr. išŠēx along the western branch of the Nile, in Burg Burullus and Baltîm and in certain relic areas in the Delta (i. e. the north eastern and the north western parts of the centre): e. g. bayt, mawt. For the whole region of B. Swayf and Fayyûm (NME 1) {F, G}, and for alFarâfira and adDâxila, the diphtongs are also characteristic (see maps 1,2).

  • 22 SeeWoidich,1991,p.1643ff. . .
  • 23 SeeBehnstedt,1979,p.122ff.

27The reduction of ay to a before a cluster of two consonants is a shibboleth for NME 1 dialects: ma‘alahš “never mind!” (<*ma‘alayhš, cf. Cairo ma‘alešš), itnan ginayh, “two pounds” (< *itnayn ginayh) 22. Due to morphophonological changes by suffixation in many of these areas, the diphthongs alternate with other vowels, for example (western adDàxila): bayt, “house”, versus bitè, “my house”, and bîtihôm, “their house”; (Burg Migîzil) fawq, “above”, versus fùqi, “above me, and fuqna, “above us” 23. This is not the case in the Fayyùm and B. Swayf areas, where diphthongs remain stable in these cases.

6) Pausal forms

28A feature typical for rural dialects is pausal Imàla of -a which varies between [e], [i] and [ih] (see map 35): kalba, “bitch” [kelbe], [kelbih]. A pausal [e] is present in northern B. Swayf even in emphatic environments: batta (batte), “duck”.

29Pausal lowering for ê, î, û takes place in the Manzala region of the Delta (NED 2): bèt [be:t], “house”, fallahîn [fellehe:n], “farmers”, sû' [so:], “market”.

7) Deletion of i and u

30This very important morphophonological feature is considered to be a distinguishing mark of the neo-Arabic language type. In Egypt, unstressed i and u in open intermediate syllables are elided in the following environments (see maps 62-66):

31(a) implies (b) and (c), (b) implies (c). Regions with (a, b, c) are the Šarqiyya dialects (ED 1) (D, E) and most dialects south of Asyût {l}, starting from Abu Tîg some 25 km to the south of Asyût (UE 1,3,4, less so in UE 2). In the same dialects, i in pretonic unstressed syllables is elided (as in hinàk, “there”, which becomes ihnâk, see map 68).

32(b) and (c) apply without (a) in the centre of the Delta (CD, ED 3), Cairo, and Middle Egypt between ilMinya and Asyût (SME): miskit, mâska ~ maska, yiktîbu –yiktibu {H,I}.

  • 24 These are “parlers non differentials” in another sense than in the Syro-Lebanese area since both i (...)

33(c) applies without (a) or (b) in a closed region which begins with northern Middle Egypt in the south and includes the whole west and north east of the Delta (WD, NED, NME) {B, D} (see map 64) as well as in the oases ilBahariyya and ilXàrga: yiktîbu ~ yiktibu, misikit, wihiša, “ugly” (f.) 24.

34There is no elision of i or u at all in the oases of alFaràfira and adDàxila: kàmil+a > kàmila, “complete” (f.), yâkulô, “they eat”.

8) Insertion of i

35Elision of i in environment (a) of [7] results in a cluster of three consonants which is resolved by the insertion of i before the second consonant from the right (ED 1, UE): yiktib+u > yiktbu / yikitbu, “they write” (see map 56). Elision of i in (a) -vCC_Cv thus implies insertion in -vC_CCv {D, E, l}.

9) Bukara-syndrome

36Insertion of a vowel in clusters -Crv is a hallmark of NME {F, H} and oases dialects. The inserted vowel has the same quality as the vowel following r : midara, “winnowing fork”, midere (in pausa), midiriti, “my winnowing fork” (see map 47).

10) Shortening of long vowels before two consonants

37This is typical for the Delta with some exceptions for the Šarqiyya (ED, see map 74): bàb+ha > babha, “her door” {D, E}. In the Nile Valley vowel length is retained {F}: bâbha.

11) Stress pattern

38The most important variable is the position of stress in words which contain a consonant cluster. It falls either on the vowel following the cluster, as in madrâsa, or on the vowel preceding it, as in mâdrasa. The first pattern, a salient feature of Cairo Arabic, dominates the Delta with some exceptions in the Šarqiyya and NED 2 {D, E}, the second the Nile Valley to the south of igGîza ({F}, see map 59).

39In the oases adDâxila, alFarâfira and western Bahariyya, we find a western (North-African) stress pattern as in libân / libàn, “milk” (alFarâfira, adDâxila), widin, “ear” (western Bahariyya).

Morphological features

12) Demonstr. -t -n etc

40This variable divides the Delta in a western and an eastern part with -t prevailing in the west and -n in the east: dahuwwat, dahuwwan (see map 148) (A}. In the Nile Valley we find mostly -ti as in huwwati, “he”, hiyyati, “she” (NME 1) {F} without lengthening of the final vowel. Further south, lengthening takes place: huwwâti, hiyyâti (SME) {G}.

13) Genitive particle

  • 25 Spitta,1880, §7a p.25.

41Besides bità', which covers all Egypt up to Asyùt and is used in all superregional varieties, there are some other particles of this kind (see map 189): šuġl 25 between Asyùt and Nag' Hammàdi (UE 1) as in ilbagara šuġulti, “my cow” {l}; ihnîn from Nag’ Hammàdi to Lugsur (UE 3) and allîlallèl, further to the south.

14) sarab / sirib

42The distribution of the a- and the i-type of the first measure verbs provides us with a clear-cut isogloss in the Delta (see map 196). The western part (WD) prefers the a-type with certain roots, whereas the i-type is common in the eastern and central parts: šarab, faham, rakab, sama’, etc., versus širib, fihim, rikib, simi’, in NED, ED, CD, CED {A, B}.

43In the Nile Valley between Asyùt and Nag’ Hammàdi (UE 1) the i-type receives stress on the final syllable due to the impact of sedentarized bedouins, thus yielding šrib ~ išrib, “he drank” (see maps 190-94) {I}. On the west bank of Lugsur (UE 3) stress falls on the inserted vowel: išrib.

15) 3. sg. f. perf. at or -it as a suffix

44The centre (CD), the west (WD) and the north east (NED) of the Delta only show -it for the, perf. of strong verbs: darabit, “she hit”, libsit, “she put on”. The Šarqiyya dialects (ED) in contrast, show a vowel harmonic distribution of -it with i- bases and –at with a-bases: darabat, but libsit (D, E). The same distribution can be seen in the UE 1 and UE 2 dialects south of Asyût (I). UE 3 on the west bank of Lugsur only has -at: durubat, libsat (see map 203).

16) 3. sg. f. perf. weak verb

45The perf. of the final weak verb has -at (as in banat, “she built”) in most rural districts, in contrast to banit in Cairo (see maps 286, 287) and some smaller areas of the Nile Valley.

17) and pl. imperfect

46There are three different types of paradigms (see map 210):

(a) aktib

(b) aktib

(c) niktib

“I Write”




“we Write”

47The original system (a) developed into (c) by paradigmatic levelling. In Egypt, (a) is the common type in the Delta and is characteristic for the centre (Cairo, CD), the east (ED) and the north east (NED) {B, C}, as well as for Middle Egypt (NME, SME). (c) dominates the west of the Delta (WD 4) and the Nile Valley from Asyùt southwards as far as Nağ’ Hammâdi (UE 1) and the west bank of Lugsur (UE 3) (11). Paradigm (b) seems to stand halfway between (a) and (c), and indeed, we always find it in areas which lie between (a) and (c) {C}. In the Delta, for instance, the area with (b), i. e. the WD 1 dialects, forms a bridge between the west (c) and the north east (a). In this way, (a), (b) and (c) represent what is called a terrace landscape (see map 211) where (b) can either be seen as the intermediate stage of an internal development of paradigm (a) to (c) via (b), or as the result of dialect contact. Since the existence of (c) in Egypt presumably goes back to the impact of sedentarized Bedouins returning from North Africa, the latter explanation is more likely.

18) Measures lla and lli

  • 26 SeeWoidich, 1989.

48Here the morphological distribution of /a/ and /i/ in the last syllable of the verbal base (d, g) must be distinguished from the phonological one (a, b, c, e, f, see map 226) 26. .

49measure II. perf., imperf.

  • (a) kassar, yikassar ; kallim, yikallim

  • (b) kassar, yikassar; killim, yikillim; 'allim, yi'allim

  • (c) kassar, yikassar; kallam, yikallam

  • (d) kassar, yikassir ; kallam, yikallim (morphological)

50measure III. Perf., imperf.

  • (e) sâfir, yisâfir

  • (f) sâfar, yisâfar

  • (g) sâfar, yisâfir (morphological)

51(a), with a in a backing environment and i in a fronting one, is standard for Cairo, the west (WD 3,4) and the centre of the Delta (CD) {C, E, F}.

52In (b) not only the vowel of the last syllable follows the rule but the first vowel as well: killim. yikillim, “he spoke”, but 'allim, yi'allim, “he taught”. This pattern we find at the periphery of the Delta, i.e. far in the north (WD 1, NED 1) {C, D} and again in the south between ilFasn and Abu Girg, in the middle of an area which belongs to the NME 2 dialects. We find a similar distribution of the vowels in ilBahariyya and alFaràfira oases. The geographical division of this feature into several unrelated peripherical areas makes it likely that (b) is a relic feature.

53(c) invariably displays an a in the last syllable and is characteristic for the whole region south of igGîza, i. e. ilFayyum, B. Swayf to ilFasn, and again from the south of Abu Girg to ilMinya {F, H}. In the Delta we find (c) in an area forming a narrow strip between the inner Sarqiyya and the centre of the Delta (ED 3, CED) {E}. Again, the geographical division suggests that (c) is a relic feature, too.

54(a), (b) and (c), all imply (e), i. e. in all these areas measure III is formed with an i in the last syllable: sàfir, yisàfir, “he traveled”. The only exception is in the southern area of the oasis ilXàrga which has an a both in measure II and III: (c) kallam, yikallam and (f) sàfar, yisàfar.

55In (d) and (g), the final vowel is morphologically determined, i.e. a in the perfect, i in the imperfect: kallam, yikallim and sàfar, yisàfir, (d) covers al the Nile Valley south of ilMinya (SME, UE) {I}, the eastern Delta (ED 1, 2) {D, E} and the oases of adDàxila and central Xârga. (d) implies (g) and vice versa.

56Measures t-II (V) (see map 237) and t-III (VI) (see map 240) can be derived by adding a prefix it- to measures II and III respectively. Dialects without a morphological distribution of the vowels in II (a, b, c) do not change these when the prefix it- is added in either the perfect or the imperfect. t-ll and t lll are derived thus directly from II and III: it+kallim > itkallim, yitkalîim. il+ 'àbil > it'àbil, yit'àbil.

57Dialects with a morphological distribution of the vowels as in (d, g), however, invariably have a in the last syllable of the imperfect of t-II and t-III and i in measure II and III:it+kallam > itkallam, yitkallam (II: yikallim); it+gâbal > itgâbal, yitgâbal (lll:yigàbil).

19) Measure t-I prefix (VII) (see map 242)

58The prefix of the passive of measure I is it- or in-: gasal, “he washed” > itgasal - ingasal, “he was washed”. it- is productive in Cairo and in nearly the entire Nile Valley except for the regions between ilFasn and Abu Girg and Nag’ Hammàdi and Gina where il (<*in-) occurs. In the Delta, in- is productive in the Sarqiyya (ED) (D, E) and parts ofthe Bihèra, northwestem Minufiyya, and northern Gîza (WD 4 and their periphery) {b}. All these regions have been influenced by Bedouins (see feature (3) above which shows nearly the same pattern).

20) Measure t-l. perf., imperf. (see map 244)

59Types: (a) misik - itmasak. (b) misik - itmisik, “he was seized”.

601. The perfect of measure t-I derives from both the a-type and the i-type of the perfect of measure I through the addition of the prefix it- (or in-, see 19):it+gasal > itgasal, “he was washed”. In most areas, the i-type is treated like an a-type and receives a: (a) it+misik > itmasak. This is the case in Cairo, the centre and the east of the Delta (ED, CD, see map 248) {B, D, F}, in the Nile Valley (except between ilFasn and Abu Girg), and in the oases of adDàxila and ilXàrga.

612. The a-type and the i-type retain their vowels as in (b) in the western and the northeastern parts of the Delta (WD, NED) (B, D), the Nile Valley between ilFasn and Abu Girg, and the oases of ilBahariyya and alFarâfira: (misik) it+misik > itmisik, “he was seized”, (darab) it+darab > iddarab, “he was hit”.

623. With type (b), the imperfect always shows the same vowels as the perfect: yitmisik, yiddarab. With type (a), the vowels of perfect and imperfect differ: either the last a of the perfect becomes fin the imperfect (as in (a) itmasak, yitrnasik), or both the final and the penultimate a become i as in (p) itmasak, yitmisik.(a) is characteristic of the Šarqiyya (ED) {D, E} and southern Middle Egypt Irom ilMinya to Asyût (SME) {H, I}. (a) can further be found in the area between Nag ' Hammàdi and Gina, and from Idfu on southwards (UE 4). (p) is the norm for Cairo and the centre of the Delta {A, B, D, E}, and northern Middle Egypt from igGiza to ilMinya {F} apart from the region between ilFašn and Abu Girg (see maps 244,245,252).

21) gêt~gît

63The perfect of gih. gà, etc. follows either the a-type or the i-type of the weak verb (see map 305); 1. sg. perf, gêt or gît, “I came”. The variable presents a clear isogloss separating Cairo and the Delta (with get) from the Nile Valley from igGîza on southwards (with gît) {F}.

64Parallel to the gêt ~ gît isogloss are those of the plurals bàb. Ibwâb ~ bibàn; fàs, fùs /fisàn (see map 347) {F}.

22) Present marker

65bi - is the dominant present marker throughout the Delta and northern Middle Egypt (NME 1) as far as ilFašn {G}. Further south a variety of markers are in use which are apparently derived from 'ammàl. These include amma- (NME 2. SME), ‘amma- (UE 1) (I), 'a- (UE 3), imm-, um- (ilXârga, ilBahariyya), ba- (UE 1, UE 4) and less frequently 'ama-, 'am-, ma'-, ma- (see map 219 f. for more detail).

23) Negation of participles

66From ilMinya southwards, participles and even nouns can be negated with ma-... -s, as in ma-gadirs, “he is not able”; in the north only miš is possible, i.e. miš gàdir (see map 316) {H}.

Lexical features

24) “here”

67hina covers the Delta and the Nile Valley as far as Asyùt (I). From Asyùt to Aswàn ihnà [ihnih in pause) prevails. Less common forms are hèna (DE 3), hàna, hunî [adDàxila], and hawn (alFarâfïra) (see map 174).

25) “when ?”

68Forms which go back to *aymatà cover the whole Delta and northern Middle Egypt (NME): imta (Delta), aynta ~ aymta (NME 1) {G}, èmta (NME 2). Further south meta is the most common {H}, and forms such as mita, matin can be found sporadically in southern Upper Egypt. In the eastern Delta and ilXârga, imta is more often than not replaced by the newly formed wagtè ~ waktè, “which time?” (seemaps 185, 186).

26) “where?”

69fèn (fayn where diphthongs still exist) is the common form all over the Delta and the northern Nile Valley as far as the region south of Asyùt, sometimes in shortened forms such as fa (see maps 183,184). From this region on wèn is used {1} except in the areas between Nag' Hammàdi and Lugsur, and Isna and Aswàn, where fèn shows up again.

27) “how?”

70The Delta, Cairo, igGiza and the northern part of B. Swayf are covered by izzayy - izzây, a back formation of izzayyak (< *išzayyak). Forms derived from *kayf and revealing Bedouin influence begin as far north as the Fayyum and Beni Swayf areas with kif (besides izzayy} {G}. Further to the south we find kîf and kèf, and from Asyût on, kèf- kè (see map 187) (l). South of Asyût, we sometimes find being used as a preposition: hamra kè ššày, “red as tea” (UE 3) {I}.

28) “now”

71Items denoting “now” are generally derived from *waqt, “time”, with a prefixed demonstrative di. dilwa't ~ dilwa'ti is the usual form in Cairo and the central Delta, while dilwagt(i), dilwaxt cover the eastern and the western parts. The -ti in dilwa'ti can be related to a demonstrative element common in many pronominals and adverbials: dilgèti, dahuwwâti, ikdèti, ba'ati, ilwa't(i) without di occurs in the north eastern part of the Delta.

72In the Nile Valley, in particular south of Asyùt, a host of phonologically changed, reduced, or otherwise reshaped forms of *dilwaqt occur (see maps 180,181). Among others we find: dilwaxt (NME 2, SME) (G, I). dilwak (UE 2). dilwaxît. dilwakît (UE 1, 2), lakîti (UE 1) dilgê(ti) (UE 3), dôk, dak, dalag (UE 4), and dolag (Assuan), see further Behnstedt-Woidich (1994) sub wqt. The fricative x (< *q,k) in place of the expected glottal stop in dilwaxti etc. may occur even in regions where *q was generally replaced by a glottal stop, as in the B. Swayf area (NME 1). Whether this is a relic form or an innovation by spreading is uncertain.

73Forms deviating from the general scheme are: halwagit which is recorded for some villages in the Šarqiyya (see map 179), and derivatives of 'an such as dilwân. “now”, and lillân, “until now” which occur in some places south of Asyût.

29) “yesterday”

74imbàrih (reduced from *yôm imbârih < *yôm inbàrih < *yôm ilbârih, compare 'amnawwil, “last year”) covers most of the Delta except the western part (WD 4) where ilbàrih prevails (B), imbàrih is common between Suhâg and Lugsur and as ambàrih south of Lugsur (UE 4) (see map 403).

75Middle Egypt from B. Swayf and Fayyum to Suhàg is dominated by 'asiyya, originally “eve”, but here “yesterday” by semantic extension (G). In the Fayyum and B. Swayf areas, both imbàrih and 'asiyya occur side by side.

30) “good”

76kawîs is limited to the east of the Delta (ED 1) {D, E} whereas the diminutive kuwayyis covers the Delta and the northern B. Swayf area. milîh [G], and further south, malîh (SME) {H} dominates as far as Asyût, South of Asyùt (UE 1,2,3) zèn is common together with its counterpart sen, “bad” (l). These both reveal a bedouin influence. The area south of Lugsur (UE 4) prefers sabi for “good” (see map 398).

31) “piece”

77To the common northern word hitta for “piece” corresponds dinsa - dansa south of Asyùt (DE, see map 396) {I}. In some places in Upper Egypt we find hitta (Idfu region.UE 4), in others gut’a or habâbi.

32) “chicken”

78A clear-cut isogloss runs immediately south of the city of igGîza {F} with farxa, “chicken”, in the north and farrûga - farruğa in the south. In the south farxa means “young pigeon” which is zaġlùl in the north (see map 429).

33) “egg”

79bèda is generally used for “egg” throughout the Delta except in the west (WD 4), where dahya is found (see map 432) {B,C}. There, as in many dialects influenced by bedouins, bèda is a taboo word meaning “testicle” (see map 435). Both words occur in the Fayyûm and B. Swayf areas whereas south of B. Swayf to ilMinya bèda prevails again {G, H}. Between ilMinya and Asyût, again, dahya is common. South of Asyùt {I}, bèda is often replaced by new words derived from roots meaning “to roll” such as kahrùta, kahrèti, dahrùg. dahrùdi, ğahrùti (see maps 433,434).

34) “frog”

80The common word for “frog” is dufda'a in the Delta except the Sarqiyya, where it is gufda 'a. Forms derived from dufda'a are also used throughout the Nile Valley down to its most southern part (UE 4); dubda'a in the Fayyûm and B. Swèf areas {F} to ilMinya {H}; from there to Lugsur, du'dufa and du'udfa are common. South of Lugsur mainly words of onomatopoeic origin are found: ga'ôra, ğarôra, ga'rùra, ğarôna,.ga’ù’a, ga'ôya, gu'wayya. These remind us of the coptic krogr, from which bagarôra (ilBahariyya), bagarùr (alFarâfira), buglul, baglùl, biglâl, buglôl (adDàxila) are certainly derived in combination with the article pe-(see map 440,441). Other words for “frog” are laglôg (ilXàrga, Barîs) and ğifda'a (ilXàrğa town).

35) “shadoof”

81The name for this irrigation tool has to be studied in combination with the word for the pole from which the bucket hangs (see map 527). Throughout the Delta and Northern Middle Egypt šadûf is used for the tool. Šayyâla, 'irg, zâna and other words are used for the pole in the Delta, while ‘ûd is found in the B. Swayf area {G}. From ilMinya (SME) {H} on, ‘ûd denotes the implement as a whole as well as the pole. 'ûd is thus clearly a pars pro toto form, - as indeed is sadûf, which originally means “basket” and is used in this latter meaning in the oases. Baskets were used in place of the modern buckets in earlier times.

36) “yoke”

82The common word for “yoke” in the Delta is nàf. Northern Middle Egypt from igGîza {F} to approximately Biba {G} (NME 1) has dand while further to the south only karab is used (see map 502).

37) “ploughshare”

83In the Delta and northern Middle Egypt as far as approximately Abu Girg, the word for “ploughshare” is silàh; further to the south sikka is used (see map 482).

38) “house”

84There are two words for house, bèt and dàr (see map 456). In the north dàr means the traditional farmer's house built of mud brick while bèt is used for a modern house of burned brick, bèt is used for both types of houses in the south, while dàr is not used at all. As in (37), the isogloss runs approximately through the area of Abu Girg.

39) “stable”

85For “stable”, the northern word is zirîba in contrast to southern hôš (see map 462). The isogloss runs directly behind igGîza {F}, hôš “stable” is also used in the Delta, however, namely in a small area at the northern end of the Rosetta branch of River Nile(WDl).

40) “hoe”

86The characteristic northern word for “hoe” is fâs, whereas in the south tùrya is used (see map 5211.). The isogloss runs somewhere through the south of igGîza. fàs in the south means “axe”, for which the north has balta {F}.

41) “duck”

87In the Delta, only batta is used. bahha is found in the northern part of B. Swayf from igGîza on, and in the Fayyûm for the local variety of duck (see map 436). From B. Swayf {G} to Suhâg, bihha is used and further to the south we find buhha.

Syntactical features

42) Circumstantial clauses

88One of the rare syntactical variables involves the structure of the circumstantial clause {I}: the type wi humma mašyîn, “as they were going” shows up south of Asyût as humma w mašyîn, i. e. with the same structure as in some syro-lebanese dialects.

Dialect geography and human geography

89Now we have examined a variety of linguistic features separating the various Egyptian dialects, it is time to consider the extent to which these dialectological facts coincide with other facts of human geography. That is to say, whether the regions marked by these isoglossic bundles coincide with those of anthropological features of material and spiritual culture, and the role played by other phenomena - such as trade routes and religions affiliations - in the spreading of linguistic features.

Material and spiritual culture

90Winkler [1936] p. 455 concludes that the dialect borders coincide only to some extent with those of material and spiritual culture. According to him, the prominent border of material culture south of Asyùt is only paralleled by a weakly marked linguistic border. This does not correspond to the findings of the dialect atlas. Winkler distinguishes only two main dialects, A and B, and does so primarily on the basis of a single variable, i.e. *q ~ ' ~ g (see feature 2), and some lexical features. This is why he places the main boundary between dialects in the Nile Valley south of B. Swayf. Using many more morphological and lexical variables, and taking a much broader approach, the dialect atlas instead places the most important bundle of isoglosses south of Asyùt {I}, which closely coincides with the most significant division of material culture, South of Asyùt we find (Winkler, p. 454):

  • the bell-shaped oven (furn) with two chambers (north of Asyùt it is of cubic shape);

  • loaf-shaped bread ('eš šamsi) (north of Asyut unleavened bread is used);

  • a two-fingered rake for the threshing-floor (midra) (north of Asyùt it has five fingers),

  • the use of a handmill (rahàya) fixed into an earthenware tub (north of Asyùt it is placed on a woven mat);

  • a wicker work seat for the threshing-carriage (nôrag) (north of Asyût, a wooden chair);

  • that people claim to see a palm-tree in the full moon. .

91Moreover, the very important isoglossic bundle {F} south of igGîza also coincides with a significant cultural boundary as well. Seen from the north (see Winkler, loc. cit.), this marks the point at which the use of the battâna / battâma (a tool used to make earthen walls), and the two handled plough stops, and the point at which people begin to say that the corona of the moon has the form of a bowl.

92Bundle {G} which divides the dialects of northern Middle Egypt into NME 1 and NME 2 (south of B. Swayf) is paralleled by the cultural border where the half-circle shaped sickle of the Delta stops. Farther to the south sickles are hook shaped (see Winkler, 1936, plate 56).

  • 27 SeeRiad, 1961.

93In the Delta, only one feature is known to coincide with linguistic border:the Šarqiyya which is sharply separated from the rest of the Delta by the bundles {D} and {E}, is also marked by a different type of plough 27.

Trade routes

  • 28 For the concentration of trade posts along the Dumyât-branch of the Nile in the late Middle Ages se (...)
  • 29 So the traditional interpretation, see Behnstedt, 1978, p. 65, Behnstedt-Woidich, 1984, 1, p. 31f.

94- The most conspicuous example of the impact of a trade route on the distribution of a linguistic variable is that of the reflexes of *q and - see map 15; feature (1) and (2)-, i.e. '/g and g/ğ respectively in the Delta. The picture shown by the map is a dynamic one and can be seen -scenario [1]- as the spreading of the Cairene features '/g along the medieval trade route from Cairo to Dumyàt 28 in a corridor through a region with g/ğ 29. But we can imagine yet another scenario [2]: the resistance of '/g to the impact of bedouinized dialects with g/ğ from western and eastern parts of the Delta, with '/g being backed by the existence of this trade route and its Cairene influence. In this way, the bedouinized dialects were possibly prevented from covering the Delta entirely.

  • 30 See as a summary Singer, 1994.

95[1] starts from the assumption that the Delta had g/ğ and '/g spread from Cairo along the trade route, whereas [2] starts from '/g (or q/g] in the Delta and its partial replacement by g/ğ due to Bedouin influence outside the area along the trade route. Scenario [2] is certainly supported by the existence of the relic area WD 2 at the periphery of the Delta (mediterranean littorial) - with q and g. An early spread of q and g from a capital (al-Fustât) along the branches of the Nile to the littoral of the Mediterranean and the important harbors Alexandria, Rosetta and Damiette would further correspond to what we know of the Arabic settlement history in North Africa after the conquest in the 7th century 30. The corridor-like form of the distribution of '/g - g/ğ corresponds further to the distribution of another feature of bedouinized speech, i.e. secondary emphasis (map 34). Other maps can be interpreted in the same way: dukhum ~ dukham (map 166), masak ~ misik (map 194), yuhrut ~ yihrit (map 216), prefix measure Vll in- ~ it- (map 242). 'akl ~ wakl (map 267).

  • 31 See Blanc, 1981.

96In order to decide which of the two scenarios is more likely,we need a more detailed knowledge of the region's settlement history. The existence of this corridor makes clear that the pronunciation g (*ğ) must be an old one in any case. If we assume scenario [1], we have to conclude that it harks back at least to the late Middle Ages when this trade route between Cairo and Dumyât was active. Since Alexandria regained its economic importance and the old trade route was reactivated, only at the beginning of the 19th century, a change of ğ to g in the 18th century as proposed by H. Blanc 31, can hardly explain today's geographical distribution of g (*ğ). If Cairo had had ğ in the late Middle Ages, this would have spread along the trade route, not g. By the same token, scenario [2] presupposes g at an even earlier stage, presumably immediately following the conquest. Either way, g in this area must be a relic.

97- On the east bank of Middle Egypt, there is no transit traffic at all. We find here the genitive exponent šugl much further north than on the west bank, where the standard form bità' follows the highway far to the south, see map 189 in Woidich-Behnstedt(1980) p. 183.

98- In the western Delta, the common word for a water jar used for transport is zal'a (see map 443 and Woidich-Behnstedt, 1980, p. 182, map 5). Along the route Alexandria-Cairo (ittarî’ izzirà'i), however, we find the standard ballàs progressing southwards into the region with zal'a.

Religion and dialect

  • 32 See Björnesjö, 1994.

99It is not uncommon in the Middle East for religious groups to differ with regard to dialects. These differences go back to the different geographic origins of the respective groups. In Upper Egypt, on the west bank of Lugsur some 5 km south of ilGurna, there is a Coptic village, 'Izbat ilBasîli, situated in an otherwise Muslim environment. It belongs to a nearby, now deserted, monastery. The dialect spoken in ilBasîli is the same as that spoken on the east bank (UE 1). An explanation for this may be that the surrounding muslims - in contrast with the Coptic population-mixed with bedouins and in this way developed the UE 3 dialect. The Copts thus preserved their original colloquial. An argument against this scenario might be that place names with ‘izba are in general of recent origin dating from the 19th c. 32. Therefore ‘izbat ilBasîli may be too young a foundation for the development described above. An alternative explanation, therefore, is that Copts from the east bank were transplanted to the west bank not too long ago in order to serve the monastery. Whatever is true, those scenarios make clear that dialect divergences do not depend on the religion itself so much as on the history of settlement.

Sedentarization of Bedouins

  • 33 Some data on bedouin settlement (state 1884) are to be found in Hartmann, 1900, p, 220-222.
  • 34 See Woidich-Behnstedt, 1980, p. 182, Woidich, 1997.

100The presence and settlement of Bedouins 33 in Egypt since the Arabic Conquest (and perhaps before) is of paramount importance for our understanding of the linguistic history of Egyptian Arabic (see the discussion of trade routes above). Bedouinization can be detected by certain linguistic features, be they in their original or an adapted (inter-dialect) form 34. For example:

Phonological features

101g for *q; strong secondary emphasis (farxa firàx, kamân, gumàš, see feature [3]); glottalized pronunciation of r, stress patterns: Maghrebinian patterns in adDâxila and alFaràfira (libàn ~ libân, see feature [11]); stress on the possessive suffix -î, -nî', stress on article; stress on the in-prefix in indarab (in certain villages in ED); gàhawa-syndrome (in UE 3 nâxala, “palm tree”. 'ala mâhalak, ”slow down!”); elision of front vowels after -CC-(see feature [7,8]); vowel alternations [masak+at> misikat, balah+a > bilaha);

Morphological features

102fem. pl. intin - intan, hinna; -kin ~ kan, -hin - -han, yimšan, rayhàt, etc.; verbal measures II and III show morphological distribution (see feature [18]); paradigmatic levelling in the imperfect niktib-niktibu (see feature [17]); perfect types šrib, išrib; plural types šnatt, “bags” and bnitta, “girls”;

Lexical features

103kèf/kîf for izzayy, “how?” (see feature [27]); zèn, “good” and šên, “bad”; from Asyùt on, Arabic metaphors are used for the “pin” of the plough (Griessaule): muftâh, 'iyâr, xabur, watad, witt, gifil, dul', instead of balanğa ~ bilinğa (apparently a latin loan) in the Delta, and bagrum ~ bağrûm in Middle Egypt (probably coptic) - see Riad (1961) map (a) on p. 252 and map 483 in Behnstedt-Woidich (1985); also the replacement of the taboo word bèd, “eggs” (testicles) by dahya (western Delta, see map 435) or a word derived from a root meaning “to roll”, such as kahrûti, dahrùg (south of Asyût, see feature [33]).

  • 35 See Björnesjö, 1994, p. 32a.
  • 36 See Woidich-Behnstedt, 1980, p. 181 f and map 4 p. 187; Björnesjö, 1994, p. 32b, is more hesitant i (...)
  • 37 See Björnesjö, 1994, p, 33a.
  • 38 See Björnesjö, 1994, p, 33ab.
  • 39 See Björnesjö, 1994, p. 34a.
  • 40 See Woidich-Behnstedt, 1980, p. 181.

104Another feature related to the settlement of Bedouins are certain types of place name. Toponyms starting with Nazlit... (Nazlit 'Abille, etc.), for instance, probably correspond to the recent arrival of groups of Arabs. Nearly all of them appear in Middle Egypt and might be connected with the Ğuhayna-tribe 35. Others are those combined with Nağ' (Nağ’ Hammàdi etc.), which are concentrated in the provinces of Suhâğ and Asyùt 36. Combinations with Awlâd...(Awlâd Badr, etc.) mainly appear in the eastern Delta and at the east bank between Asyût and Suhàğ. They point to Bedouin settlement 37, as do toponyms starting with Banî... (Bani 'Adi etc.). These mainly occur in Middle Egypt and to the south as far as Suhàğ, but also in some cases in the eastern Delta 38. Toponyms with Zâwiya... occur in the western part of the Delta and further south exclusively on the west bank of the Nile Valley as far as Qena. This suggests the settlement of Arabs of Libyan origin 39. Plurals of personal names such as al-Fa'àlil, al-Fa'alîl, al-Fulànàt [ilMahamîd, il'Atiyyàt) -again a type of toponym which can be ascribed to Bedouins- are common in the eastern part of the Delta and in southern Upper Egypt 40. With all necessary reservations in mind, we may state broadly that bedouin place names concentrate in areas such as the western and eastern fringes of the Delta and (the area) south of Asyût, where the varieties of Egyptian Arabic are spoken show greater or lesser evidence for the impact of Bedouin dialects, be they of western or of eastern type.

  • 41 The Fazara tribe, a fraction of the Dubyàn and belonging to the Qaysites, is well-known in Egyptian (...)
  • 42 See Woidich, 1995.

105Another hint of Bedouin origin, i.e. oral history, has unfortunately not been investigated systematically. The B'èri speakers (UE 3) on the west bank of Lugsur trace themselves back to the well-known alFazâra tribe (EI II, s.v.) which has always been present in Egypt, Libya, and Chad areas 41. As they tell us, they came to Upper Egypt from the west following the droppings (ba'ar) of their camels and guided by their ancestor Hasan ilFazâri. By the same token, many of the inhabitants of the oases report that they arrived there as Bedouins, partly from the west and partly from the Nile Valley 42.

  • 43 For the problems which such a research is confronted with, see the introduction to Bjôrnesjô, 1994.
  • 44 See Behnstedt-Woidich, 1985, vol. I, p. 26.

106Needless to say, much more systematic research is needed on both the toponyms themselves and the history of Bedouin settlement in general 43. A factor which has to be taken into account here is the density of the population in Upper Egypt through the Middle Ages. Map 553 (based on Halm (1982)) shows that this area was sparsely populated south of Asyùt, a fact which certainly facilitated the settlement of the Bedouins there in great numbers 44.

Future activities

107The dialect atlas such as it is now offers but a rough collection of the linguistic material and a preliminary overview of the Egyptian rural dialects. It has its deficiencies: even with data collection, it was not possible to keep the parameters of the informants stable. So the sources are somewhat divergent. And it has of course its lacunae: the southern part of Upper Egypt between Idfu and Aswân, the Ababde, the Manzala-region certainly need more research. But, nevertheless, it gives us an idea of the kind of variables which occur and shows us the areas in which the main bundles of isoglosses delimit certain dialect groups. For further research with more refined methods of sampling, which should lead eventually to regional atlases, these areas deserve a closer look.

108Another important desideratum is describing the main areas by exemplary reference grammars. The (present) writer hopes to present in the near future some local grammars of this kind on B'èri (UE 3) and the Oases, as well as a comparative grammar of the rural dialects of Egypt, partly based on as yet unpublished material. And last, but not least, the whole atlas should be redrawn in a more consistent manner than it is now. The computer should make this possible.

109As to the historical interpretation of the linguistic facts, we are in desperate need of a history of settlement in the Nile Valley based not only on medieval writers but also on Ottoman sources and data from later centuries.

Haut de page


ABUL-FADL F., Volkstümliche Texte in arabischen Bauemdiatekten der ägyptischen Provinz Šarqiyya mit diatektgeograpnischen Unter-suchungen zur Lautlehre, Dissertation, Munster, 1961.


- « Zur Dialektgeographie des Nildeltas », ZAL 1, 1978, p. 64-92.

- « Die nord-mittelâgyptischen Bukara-Dialekte », ZAL 3,1979, p. 62-95.

- « Die Dialekte des Burullus-Sees: Burg Migîzil, ein ägyptischer Fischerdialekt », Comptes rendus du Groupe linguistique d'études chamito-semitiques XVIII-XXIII (1973-79), p. 115-168.

- « Weltere koptische Lehnwôrter im Âgyptisch-Arabischen », Die Welt des Orients, XII, 1981, p 82-98.


- « Die àgyptischen Oasen - ein dialektologischer Vorbericht », ZAL 8, 1982, p. 39-71.

- Ägyptischer Dialektatlas, Band 1 und 2, Wiesbaden, 1985. Band 3:1.- Delta-Dialekte, Wiesbaden, 1987; II.- Niltaldialekte; III.- Oasendialekte, Wiesbaden, 1988. Band 4: Glossar Arabisch-Deutsch, Wiesbaden, 1994.

BJÖRNESJÖ S., « Réflexions préliminaires sur la toponomie de l'Egypte », Lettre d'information de l'Observatoire urbain du Caire contemporain n° 38, Cedej, Le Caire, octobre 1994, p. 30-38.


- « Egyptian Arabic in the Seventeenth Century, Notes on the Judeo-Arabic Passages of DARXE NO'AM (Venice 1697) », Studies in Judaism and Islam, Jerusalem, 1981, p. 185-202.

- EI II: The Encydopaedia of Islam, New Edition, Volume II C-G, Leiden, 1991,

- EI III: The Encydopaedia of Islam, New Edition, Volume III H-IRAM, Leiden, 1986.

ELDER E., Egyptian Colloquial Reader, Cairo, 1927.

GAIRDNER W. H. T., Egyptian Colloquial Arabic, Cairo, 1926.

GARCIN J.-C., Un centre musulman de la Haute-Egypte medievale : Qûs, Le Caire, 1976.

HALM H-, Ägypten nach den mamlukischen Lehensregistern, Band I: Oberägypten und das Fayyûm, Wiesbaden, 1979.

HARTMANN M., Lieder der Libyschen Wuste, Leipzig, 1900:

KHALAFALLAH A., A Descriptive Grammar of Sa'idi Egyptian Arabic, The Hague, 1969.

KIPPEL E., « Etudes sur le folklore bédouin de l'Egypte », Bulletin de la société khédiviale de géographie, VIIe série (1911), n° 10, p. 571-616.

LABIB S. Y., Handelsgeschichte Agyptens im Spatmittelalter, Wiesbaden, 1965.

LITTMAN E, Ziegeuner Arabish, Bonn Leipzig, 1920.

MacMICHAEL H. A., A History of the Arabs of Sudan, Cambridge, 1922.

MITCHELL T. F., Colloquial Arabic, The living Language of Egypt, London , 1962.

NALLINO C. A., L'arabe parlato in Egitto, Milano, 1939 (2nd ed. Milano 1913, 1st ed.1900).

RIAD M., « Native Plough in Egypt », Bulletin de la Société de Géographie d'Egypte XXXIII, 1961, p.241-277.

SALIB M., Spoken Arabic of Cairo, Cairo, 1981.

SINGER H.-R., Die Beduinen als Trager der Arabisierung im islamischen Machtbereich, Gedenkschrift Wolfgang Reuschel, Akten des III, Arabistischen Kolloquiums, Leipzig, p. 21-22, November 1991, Stuttgart, 1994, p. 264-274.

SPIRO S., An Arabic-English Dictionary of the Colloquial Arabic of Egypt, 1st. ed. Cairo 1895, Repr. Beirut 1973.

SPITTA BEY W., Grammatik des arabischen Vulgärdialectes von Äegypten, Leipzig,1880.

TOMICHE N., « Les parlers arabes d'Egypte - matériaux pour une étude de géographie dialectale », Etudes d'Orientalisme dédiées à la mémoire de Lévi Provençal, tome 2, Paris, 1962, p. 767-779.

VOLLERS C., Lehrbuch der ägypto-arabiscben Umgangssprache, Kairo, 1890.

VYCIHL W-, « Zur Sprache und Volkskunde der 'Abbàdi’ », Anzeiger der Österreichischen Akademie der Wissenschaften, Phil.-Hist. Klasse, Jahrgang 1953, p. 177-184,

WILLMORE J. S., The Spoken Arabic of Egypt, 2nd London 1905, 3d ed., London, 1919


- Bauern zwischen Wasser und Wüste, Stuttgart, 1934.

- Ägyptische Volkskunde, Stuttgart. 1936.


- « Die Perfekt im Dialekt von il-Bi'rât' », Mélanges de l'Université Saint-Joseph, tome XLVIII, Beyrouth, 1973-74, p. 355-372.

- « Ein arabischer Bauerndialekt aus dem südlichen Oberägypten », ZDMG 124,1974, p. 42-58.

- « Zur Bildung der Verbalstämme in den Ägyptisch-Arabischen Dialekten: Der II. und der III. Stamm », XXIII, Deutscher Orientalistentag, Ed. by Einard Schuler, Stuttgart, 1989, p. 200-210.

- Bibliographie zum Àgyptisch-Arabischen, IMNO, Universiteit van Amsterdam, 1989.

- « Short /a/ in Cairo Arabic Morphophonology », Alan S, Kaye (ed.), Semitic Studies (In honor of Wolf Leslau on the occasion of his eighty-fifth birthday), Volume II, Wiesbaden, 1991, p. 1632-1651.

- « Die Dialekte der ägyptischen Oasen: westliches oder ôstliches Arabisch?  », ZAL 25, 1993, p. 340-359.

- « Das Kairenische im 19. Jh.: Gedanken zu Tantâwi’s Traité de la langue arabe vulgaire », Dialectologia Arabica (A Collection of articles in Honour of the Sixtieth Birthday of Professor Heikki Palva), Studia Orientalia ed, by the Finnish Oriental Society 75, Helsinki, 1995,p.271-287.

- « Upper Egyptian Arabic and Dialect Mixing in Historical Perspective », to appear in Festschrift Georg Krotkoff, 1997.

WOIDICH M. and BEHNSTEDT P., « Zum Sprachatlas von Àgypten », ZAL 5, 1980. p. 212-242.

Haut de page


1 See Vollers, 1890, Nallino, 1900, Willmore, 1905, and the dictionary by Spiro, 1895.

2 See for instance Gairdner, 1926, Elder, 1927, Mitchell, 1962, Salib, 1981.

3 See Woidich, 1989a.

4 That the Cairo Arabic of the 19th century showed many features which today are considered rural is a different issue, see Woidich, 1995.

5 Winkler, 1936, and for the village Kimàn near Guff in Upper Egypt, see Winkler, 1934.

6 Tomiche, 1962, is dedicated to the same topic, but does not go beyond Winkler, 1936.

7 See Winkler, 1936, p. 391, and the word list p. 343-338, and Littmann, 1920.

8 See Winkler, 1936, p. 388, 391, and personal communication by Georg Krotkoff.

9 Winkler, 1936, p. 382 f., Vycichl, 1953.

10 For more details, see Behnstedt-Woidich, 1985 ff, vol. l. See in particular the introductions to the areas in Behnstedt-Woidich, 1987, 1988, and EI VIII s. v. al-Sa’îd (Dialects).

11 See Behnstedt, 1979.

12 See Behnstedt-Woidich, 1982.

13 See Behnstedt-Woidich, 1982, and Woidich, 1993.

14 The letters marking the bundles are writlen between {} brackets. The features mentioned here and involved among others in the formation of the bundles are the following: (A)= 1,3,12,14,20; (B)=1,3,7,14,17,19,20,29,33; (C)=17,18,33; (D)=1,3,7,8,10,11,15,18,19,20,30; (E)=1,3,4,7,8,10,11,15,16,19,20,30; (F) = 5, 9,10,11,12,18, 20,21, 32,34,36,39,40; (G) =1,5,12,22,25,17,28,29,30,33,35,36,41; (H)=4,7,9,18,20,23,25,30,33,35; (I)=7,8,13,14,15,17,18,20,22,24,27,28,30,31,33,34,42.

15 The numbers of the maps refer to Behnstedt-Woidich, 1985, Vol. 2.

16 Already isSa'âra, some 25 km to the south of igGiza, shows all traits of an NME 1 dialect.

17 The Hawwâra are attested there from the 14th c. on, see EI III p. 299b.

18 It is not the place here to go into the problems of weighting the isoglosses, i.e. the question as to whether we should give one isogloss more importance than the other. Here all features are given the same value and only the number of isoglosses is taken into account.

19 For the distribution of trade posts along this line, see map 551 and Behnstedt, 1980, p. 65.

20 q is pronounced glottalized in ilXârga.

21 This is not true for the oases, since alFarâfira oasis has q but z.

22 SeeWoidich,1991,p.1643ff. . .

23 SeeBehnstedt,1979,p.122ff.

24 These are “parlers non differentials” in another sense than in the Syro-Lebanese area since both i and a are here retained and not elided.

25 Spitta,1880, §7a p.25.

26 SeeWoidich, 1989.

27 SeeRiad, 1961.

28 For the concentration of trade posts along the Dumyât-branch of the Nile in the late Middle Ages see map 551 (based on map 4 of Labib, 1965).

29 So the traditional interpretation, see Behnstedt, 1978, p. 65, Behnstedt-Woidich, 1984, 1, p. 31f.

30 See as a summary Singer, 1994.

31 See Blanc, 1981.

32 See Björnesjö, 1994.

33 Some data on bedouin settlement (state 1884) are to be found in Hartmann, 1900, p, 220-222.

34 See Woidich-Behnstedt, 1980, p. 182, Woidich, 1997.

35 See Björnesjö, 1994, p. 32a.

36 See Woidich-Behnstedt, 1980, p. 181 f and map 4 p. 187; Björnesjö, 1994, p. 32b, is more hesitant in this case.

37 See Björnesjö, 1994, p, 33a.

38 See Björnesjö, 1994, p, 33ab.

39 See Björnesjö, 1994, p. 34a.

40 See Woidich-Behnstedt, 1980, p. 181.

41 The Fazara tribe, a fraction of the Dubyàn and belonging to the Qaysites, is well-known in Egyptian and Sudanese history, see MacMichael, 1922, p, 144 f. p. 293; Kippel, 1911, p. 557. The Fazàra have been present there throughout the Middle Ages. For an account of Bedouin stems in Upper Egypt in the XIVth century, see further Garcin, 1976. p.359ff.

42 See Woidich, 1995.

43 For the problems which such a research is confronted with, see the introduction to Bjôrnesjô, 1994.

44 See Behnstedt-Woidich, 1985, vol. I, p. 26.

Haut de page

Table des illustrations

Titre Nile Delta (map I)
Légende Western dialects (WD)
Fichier image/png, 21k
Titre Nile Valley (map II)
Légende {H}: bundle of isoglosses
Fichier image/png, 9,5k
Fichier image/jpeg, 19k
Haut de page

Pour citer cet article

Référence électronique

Manfred Woidich, « Rural Dialect of Egyptian Arabic: An Overview », Égypte/Monde arabe, Première série, 27-28 | 1996, mis en ligne le 08 juillet 2008, consulté le 22 mars 2017. URL :

Haut de page


Manfred Woidich

Amsterdam University

Haut de page

Droits d’auteur

© Tous droits réservés

Haut de page
  • Logo Centre d’études et de documentation économiques juridiques et sociales (CEDEJ)