Perceptual Dialectology of Egypt. A View from the Berber-Speaking Periphery

Valentina Serreli


Perceptual dialectology is a sub-branch of folk linguistics first systematized by Dennis Preston in the 1980s (e.g. Preston 1989, 1999). Through the technique of mental mapping, borrowed from cultural geography, perceptual dialectologists seek to discover the perceived distribution of speeches, populations, and prevailing ideologies. In the Arabic-speaking context, map task experiments were conducted by Theodoropoulou & Tyler (2014) among students at Qatar University, and by Hachimi (2015) among Moroccans of different age, gender, social class and origin, to identify dialect boundaries and labels across the Arabic-speaking world and to understand the underlying ideologies.

This paper discusses lay speakers’ perception of the linguistic boundaries in Egypt, based on a map-drawing experiment submitted to secondary school students, in the Egyptian Oasis of Siwa. Siwa is a Berber enclave situated 50 km away from the Libyan border, where the beginning of mass Arabization dates back to the 1980s and whose geographical and social peripherality limited dwellers’ mobility and contact with outsiders. The map task, which asked the students to locate on a map of Egypt languages and dialects spoken in the country, was part of a questionnaire administered in 2011, with the aim of understanding the patterns of language use in the oasis, through the study of speakers’ attitudes, beliefs and ideologies (Serreli 2011). The results show that the respondents are aware of the major linguistic boundaries within Egypt, although they did not pay the same attention to all areas: Siwa, the area around Marsa Matruh, the Nile Delta region and Upper Egypt were identified quite clearly by a great number of students, while the Sinai Peninsula was taken into account to a lesser extent and the oases were largely ignored. The paper also presents the respondents’ labelling choices, which tell us something about their ideas and judgements of Egyptian communities and their reciprocal relationships.

1. Introduction

1Drawing on an attitudinal and perceptual dialectological survey, the paper presents the dialectology of Egypt, according to the perception of high school students from Siwa Oasis, a peripheral and non-native Arabic-speaking region of Egypt. Our aim is to answer two main questions: first, we want to individuate whether or to what extent linguistic boundaries drawn by professional dialectologists correspond to those drawn by non-specialists; secondly, we are interested in the way speakers define the languages, varieties, or ways of speaking, whose existence they are aware of.

2The paper is organized as follows: section 2 introduces the field of perceptual dialectology, and the studies conducted in the Arabic-speaking context. Section 3 presents the scope of the research and, more specifically, the map-drawing experiment used to collect the data discussed in this paper. Section 4 presents the results obtained, concerning dialect boundaries and metalinguistic labels. Short concluding remarks are given in Section 5.

2. Perceptual dialectology

3Perceptual dialectology is a sub-branch of folk linguistics, the discipline that studies non-specialists’ ideas, beliefs, overt comments and subconscious responses about languages and linguistic facts (e.g. Preston 1999). Perceptual dialectology focuses on the beliefs about geographical variation, that is dialect distribution and boundaries. It starts from the consideration that “nonlinguists know not only that people in different parts of the world speak different languages, but also that people in different regions speak the same language differently” (Preston 2010: 89). The discipline began in the Netherlands and Japan in the 1920s, but it is with the works of Dennis Preston on the varieties of American English that it came to prominence starting from the late 1980s (e.g. 1989, 1999, 2010). One of the techniques used within this approach is mental mapping, borrowed from cultural geography, in which respondents are asked to draw linguistic boundaries on a blank or semi-blank map, according to their own awareness and beliefs. It is assumed that ordinary speakers’ beliefs and perceptions of the distribution of linguistic varieties are important as well as their language productions as they help to understand sociolinguistic trends within a community and to explain linguistic variation phenomena.

2.1 Arabic perceptual dialectology

4To my knowledge, only two studies using the map-drawing task in the Arabic-speaking context were recently published.

5Theodoropoulou and Tyler (2014) present a perceptual dialectological study conducted in Qatar, using a map task, innovative in the Arabic-speaking context. Forty female undergraduate students of the Qatar University were given a blank map of the Arab world and asked to provide information about linguistic boundaries and labels. They clustered Arabic varieties into five groups: Maghreb, Egypt and Sudan, Levant, Gulf, and Somalia, and they also referred to an intermediary group identity (as “Shammi or Khaliji"), standing between the Arab speaking community as a whole and the national one. The authors obtained information about the way Arabic dialects are grouped, about hierarchies among Arabic dialect groups and about the way those groupings related to the labels applied.

6Hachimi (2015) presents the results of a map-drawing and labelling task administered to fifty-two Moroccans in different cities of the country (Casablanca, Marrakesh, Meknes, and Agadir), diversified for age, social class, and gender. The study sought to explore perceptual linguistic boundaries of the Arabic speaking world and the ideologies that render varieties of Arabic good or bad. The participants identified five dialect areas: Arabian Peninsula, Iraq, the Levant, Egypt, and the Maghreb, and evaluated Syrian as the best Arabic, followed by Levantine and Egyptian, and, in turn, by Maghrebi and Gulf Arabic. Hachimi’s twofold analysis of both map-drawing and post-mapping discussion aptly shows the potentials of a perceptual dialectological approach in the study of the ideologies and cultural stereotypes about regional Arabic varieties that are dominant in Morocco; the choice to target a diversified sample rather than a sample made of students only is also much valuable.

3. Presentation of the study

3.1 Overview

7The research was conducted in the Oasis of Siwa, one of the five biggest oases of the Egyptian Western Desert and the only Berber-speaking area of the country.

8Siwa is inhabited by about 28 thousand people (2015, personal communication, Governorate of Marsa Matruh), the vast majority of which is native speaker of Siwi, the easternmost Berber language. The rest of the oasis’ population is Arabic-speaking: it includes the Shahibat tribe, who settled in the oasis at the beginning of the XX century and whose native variety is a Bedouin type of Arabic, and diverse Arabic speakers coming from different regions of Egypt together with their own native Arabic variety. The oasis population is hence constituted, according to the dwellers themselves, by “Siwans”, “Bedouins”, “Egyptians” and foreigners, who are distinguished by ethnicity, morality, traditions and language (see below for further information about this point).

9Siwi is maintained by its native speakers, by now bilingual, with the exception of few monolingual Siwi speakers among children in preschool age and elders. Bilingualism is not symmetrical: while almost all Siwi speakers learn Arabic, only some of the Arabic-speaking dwellers learn Siwi, because of its pervasiveness in the oasis daily life and its value as a solidarity bounding tool.

10The data presented were collected in 2011 and are extensively discussed in Serreli (2011). A written questionnaire in Standard Arabic was administered to 90 students attending the last two years of secondary school in the city of Siwa. Students were chosen because of the need to find educated speakers who could read and write Standard Arabic, be familiar with completing a questionnaire and, at the same time, constitute a homogeneous group in terms of age, education and cultural background, easy to reach within our specific conditions. The questionnaire consisted of 23 questions, including personal information, questions about sociolinguistic data (i.e. language use) and psycholinguistic data (i.e. the relationship between speaker and language), and two map-drawing experiments.

11Most of the respondents were born in the oasis in a Siwi-speaking family (80%) and are, most likely, Siwi native speakers who have learnt Arabic as a second language. They declared to be proficient in both Siwi and Arabic, but to use Siwi to a much greater extent than Arabic.

12Considering that the more one travels, the more (s)he is aware of language differentiation and is likely to associate a given speech to its region, respondents were asked about the frequency and destinations of their travelling outside the oasis. The sample proved not to be very prone to traveling, as half of the respondents stated that they travel outside Siwa less than five times per year and only one third does it more than five times a year. Moreover, the frequency of their traveling to a specific city is related to its geographical proximity to Siwa: the nearest city, Marsa Matruh, was visited by almost everybody, while the furthest, Cairo, by half of the respondents (Serreli 2011: 62-66; Battesti 2006: 147).

3.2 The map-drawing experiment

13The map-drawing task inquired respondents about ‘the dialects used in Egypt’. This question was put towards the end of the questionnaire (No. 19), and it was intentionally introduced after a question asking to give direct evaluations concerning given dialects of Egypt, in order not to influence the respondents’ ideas about the dialect areas. Two kinds of map were proposed:

1) Map1 is a map of Egypt with the indication of the main Egyptian cities, namely Siwa, Marsa Matruh, Alexandria, Port Said, Suez, Sharm el-Sheikh, Tanta, Cairo, Giza, al-Fayum, al-Minya, Asyut, Hurgada, Luxor and Aswan.

Map1 was supplied to 62 students, but 8 were left blank: the analysis is based on 54 maps.

2) Map2 is a map of Egypt with the indication of Cairo and Siwa only.

Map2 was supplied to 28 students, 8 were left blank: the analysis is based on 20 maps.

14Different maps were introduced to avoid biases related to the amount of information given in the map, for instance, the indication of a city can encourage the respondent to consider that area even if (s)he was not going to.

15This experiment aimed at obtaining information about the perceived linguistic boundaries within Egypt and about the perceived relationship between language varieties, based on the labels chosen.

Fig. 1. Example of Map1 compiled

Fig. 1. Example of Map1 compiled

Fig. 2. Example of Map2 compiled

Fig. 2. Example of Map2 compiled

4. Results and discussion

4.1 Perceived linguistic boundaries

16Results of Map1
In almost none of the 54 maps respondents traced boundary lines, but they either put signs or wrote labels to indicate a speech in the map.
9 maps were compiled putting a sign besides the name of the city: this only allows us to say that the most considered cities were Siwa and Aswan, followed by al-Minya and Marsa Matruh, but the lack of labels renders quite impossible to understand the connections between the varieties of different cities.
45 maps were compiled writing the name of a speech variety besides the name of a city. They show different degrees of detail, as the number of speech varieties indicated varies greatly. The resulting dialect areas are Siwa, the Nile Delta and the Sinai, Upper Egypt and Aswan.

1) Siwa was identified by 91% of the informants as a distinct dialect area speaking a Berber variety;

2) Marsa Matruh was mostly considered as a distinct dialect area, i.e. speaking a variety different from the one spoken in the Nile Delta (58%), but a few respondents put it together with the Nile Delta (16%);

3) The Nile Delta was considered a separate area in most cases, namely when the major cities of Cairo and Alexandria received the same label, mostly ʿāmmiyya (36%), and when only Cairo received a label, mostly ʿāmmiyya (27%); on the contrary, 11% of the respondents distinguished Cairene from Alexandrian by using different labels (e.g. maṣri vs iskandarāni);

4) Upper Egypt was identified as a distinct dialect area corresponding to the region around the cities of al-Fayum, al-Minya and Asyut by the majority of the respondents (49%) but percentages vary between one city and another. Some respondents (between 9 and 15 %) indicated the area as speaking the same variety as the Nile Delta;

5) In Aswan, many respondents indicated the presence of a different language (51%), while a small percentage (7%) included Aswan in the ʿāmmiyya-speaking dialect area with Cairo and the other cities of the Delta and Sinai, and another 7% included it in the ṣaʿīdi–speaking dialect area.

17Results of Map2
Out of the 20 maps analysed, one student put marks only without writing down any label referring to varieties. The remaining 19 respondents wrote labels referring to the speech varieties and, in some cases, also the cities or the regions where they believed them to be spoken. Again, maps provide different degrees of detail. The most considered areas were Siwa and Cairo (94%), followed by Alexandria and Upper Egypt (44%), and Marsa Matruh (39%); the least considered were the varieties of the Nile Delta and the Sinai Peninsula (28%).

1) Siwa is one of the most frequently considered areas and it is recognized as a distinct Siwi-speaking dialect area;

2) Marsa Matruh was mostly considered a Bedouin-Arabic-speaking area set apart from the Delta region;

3) The Nile Delta was considered in 28% of the cases as an ʿāmmiyya-speaking area. Cairene was considered by 94% of the respondents, but mostly labelled ʿāmmiyya as well: this tells us that, according the respondents, a type of Arabic based on Cairene is spoken in the Nile Delta. In fact, Cairene was distinguished only few times, as qāhirāwi or qāhiriyya. On the contrary, when Alexandrian was indicated, it was mostly to stress its specificity or difference from Cairene, by labelling it iskandarāni;

4) Upper Egypt was always recognized as separate ṣaʿīdi-speaking dialect area;

5) Sinai was mostly identified as a Bedouin-speaking area (22% out of 28% total labels); one respondent used the label fuṣḥā to indicate that a Bedouin dialect was spoken (see below).

Overall, there was a tendency to “assign” a dialect to the cities given on the map, that is why sometimes maps of the type Map1 appear more detailed than maps of the type Map2: this can explain why, for example, the areas around Luxor and Hurgada were never considered by respondents compiling the Map2 type. However, greater amount of information does not mean major accuracy or correctness, and, moreover, noticeable differences in the results between Map1 and Map2 appear in only in two cases: the nūbi-speaking region around Aswan, recognized only by students filling in Map1; and the Sinai Peninsula, included in the ʿāmmiyya-speaking region by the informant who filled in Map1, but recognized as a Bedouin-Arabic-speaking area by those who filled in Map2.

191) Non-Arabic languages

a) Siwi, indicated by the vast majority of the respondents, was correctly situated it in the oasis of Siwa and mostly labelled sīwi or (as-)sīwiyya. A few respondents used labels that reveal awareness of Siwi being a Berber language, that is barbar or ʾamazīġi(-yya).

b) The Nubian language, attested in Egypt in two main dialects, is spoken in the Aswan Governorate and by Nubian migrants to Cairo and other big cities (Miller 1996; Wilmsen & Woidich 2007: 1). The respondents who filled in Map1 correctly indicated the nūbi(-yya) speech besides the city of Aswan.

202) Arabic varieties

21a) The perceptual map that emerges from our data features a Lower Egyptian region roughly corresponding to the Nile Delta, where a dialect mostly labelled ʿāmmiyya or maṣri is spoken. The differentiation between Eastern and Western Delta dialects drawn by dialectologists (e.g. Wilmsen & Woidich 2007) is absent in the respondents’ mental maps. According to dialectologists, Cairene Arabic, which serves as Standard Egyptian, belongs to Central Delta group but some features peculiar to Cairene can be recognized, while Alexandrian modern dialect is close to Standard Egyptian but with a strong influence coming from the originally spoken Bihera dialect. In our maps, sometimes, the speech of Cairo and/or the speech of Alexandria are distinguished from the rest of the Delta region and between one another: this is done by choosing different labels, be they unambiguous ones, as qāhirāwi and iskandarāni, respectively, or simply labelling Cairene as maṣri while labelling the others ʿāmmiyya. In the evaluative task, the dialects of Cairo and Alexandria had the most positive judgements, being regarded as the easiest and more pleasant, useful and modern among the given dialects.

22b) Moving southward, the respondents individuate a unique region along the Nile Valley, with no mention to the distinction between Middle and Upper Egyptian made by dialectologists (e.g. Wilmsen & Woidich 2007). In our maps, the dialect spoken approximately between al-Fayum and Asyut is almost unanimously labelled ṣaʿīdi, aṣ-ṣaʿīdiyya. This comes without surprise, if one considers that the perceived differences between Cairene (or Standard Egyptian) and the Upper Egyptian dialects constitute widespread stereotypes in Egypt, in that Upper Egyptian dialects are stigmatized in popular culture and according to speakers’ attitudes (Haeri 1996; Miller 2005; Rosenbaum 2008). In the evaluative task of our questionnaire, the ṣaʿīdi dialect received the most negative evaluations, judged difficult, useless, and not pleasant; furthermore, interviews conducted in Siwa by the author revealed that Siwans are generally aware of the existence of a ṣaʿīdi dialect characterized by the pronunciation of the Classical Arabic qāf and ǧīm as /g/ and /j/, respectively, as opposed to the Standard Egyptian realizations as /ʾ/ and /g/ (Serreli 2011).

23c) Eastern Bedouin dialects are spoken in the Sinai Peninsula, while dialects pertaining to the Western Bedouin dialects are found in a couple of areas, among which the Western Desert and the Mediterranean coast to the west of Alexandria, where dialects of the Sulaymi type are spoken (Wilmsen & Woidich 2007: 2). The presence of a Bedouin type of Arabic around the city of Marsa Matruh was acknowledged by most of the respondents filling in both Map1 and Map2, through the labels badawi(-yya), ʿarabi(-yya), ʿarabāwi(-yya). On the contrary, only among the respondent who filled in Map2, some acknowledged the presence of Bedouin dialects in the Sinai Peninsula, and used the labels lahǧa al-badu, ʿarabiyya.

24d) The dialects of the oases are connected both to the dialects of the Nile Valley, especially Middle Egyptian dialects, and to the Western Libyan Arabic, but they do not form a single separated group (Wilmsen & Woidich 2007: 6-7). The oases were largely ignored by our respondents, as nobody indicated any of them in his/her map; accordingly, in the evaluation task, the speech of the oasis of Bahariyya resulted to enjoy no prestige nor was it considered at all. It is true that the oases were not indicated and this could have biased the results of the type Map1, but this is not true for the type Map2, where only Cairo and Siwa were given. Therefore, this result points to the scant importance of the oases in the respondents’ lives.

25Finally, a few respondents proposed a map characterized by the existence of a dialect labelled ʿāmmiyya or maṣri and spoken in almost all regions of the country, including Marsa Matruh, the Nile Delta, the Nile Valley and the eastern littoral and only excluding Siwa, or excluding Siwa and Marsa Matruh, which in this case is considered as Bedouin-speaking.

4.2 Metalinguistic labels

26The analysis of respondents’ labelling choices might reveal stereotypes and underlying ideologies. More precisely, “respondents’ map-labelling task shows the evaluative richness that can be obtained from the association from dialect and place” (Hachimi 2015: 48). However, although our respondents were not very creative while filling in the maps, space constraints prevent us from extending the analysis to data collected through interviews and discussions. Therefore, we will limit our analysis to the labels found in the maps.

27The labels that the informants assigned to the linguistic varieties proved to be unrelated to the type of map they were filling in. They can be classified into those that refer to a place, to an ethnonym, or to the status of the variety.

Labels referring to the place where the variety is spoken.

28They are neuter and mostly refer unequivocally to one specific region, be it a city or a region.

Some examples are:

- al-qāhiriyya, qāhirāwi, qāhrāwiyya, named after al-qāhira, Cairo; it is rarely used, and always to indicate that the speech of the capital is distinguished from the others of the region;

- iskandarāni, named after the city of Alexandria; it is used when there is a need to distinguish it from the other dialects spoken in the Delta region;

- ṣaʿīdi, aṣ-ṣaʿīdiyya, lahǧa ṣaʿīdiyya, after the Upper Egyptian region, usually referred to as ṣaʿīd maṣr.

Labels referring to the people who speak the variety

29Labels that come from an ethnonym are mostly used for linguistic minorities or for groups that are distinguished from the Egyptian-Arabic-speaking majority. There are three cases in our data:

- barbar, ʾamāzīġi(-yya), named after the Berber or Amazigh people; it is used to define the speech of Siwa Oasis, stressing its belonging to the Berber language family, as opposite to the Arab varieties;

- nūbi(-yya), named after the Nubian people; although Nubia is also a region, it is with reference to the people that this label is used by our respondents, who never mentioned “Nubia” as a place where the variety is spoken nor as a geographical reference;

- badu, badawi(-yya), ʿarabāwi(-yya), named after the Bedouin or Arab people. A distinction between “Bedouins” or “Arabs”, in the narrow meaning of the term, and “Egyptians” (see below, under maṣri) is sharply drawn by both Siwans and Bedouins, based on the language difference and on a different cultural identity (Abu Lughod 1986; Cole and Altorki 1998; Serreli 2016).

30Within this class also fall those labels that are used to refer to a people, but originally corresponded to a toponym:

- sīwi(-yya), named after the Siwan people and Siwa Oasis. Siwa is used to address the place, the people and the language. sīwi defines a well delimited cultural identity, it refers to the land, the people, the language, the habits, culture, traditions. For the oasis dwellers, a Siwan is not a resident of the oasis but a member of the Siwan group, that is one born in one of the ten Amazigh tribes; other people who were born and have always lived in Siwa are usually identified as badu, or maṣri, or any other way according to their family’s origin;

- maṣri, named after the city of Cairo or the whole country of Egypt, and after maṣriyyin, Cairenes or Egyptians; it can refer to Cairene Arabic or to Egyptian Arabic, as a synonym of ʿāmmiyya. The term maṣriyyīn is ordinary used in Siwa to refer to those people who are neither sīwiyyīn (in the meaning referred to above) nor badu; it has a sociocultural meaning, and by no means implies that the latter two are not Egyptian citizens (a similar use of ‘miṣriyīn’ is attested among the Bedouins of the northwest coast, see Abu Lughod (1986) and Cole and Altorki (1998)). The use of this label in the maps is quite ambiguous, as it can refer to Cairene, as opposite to the rest of the Delta region (here mostly labelled ʿāmmiyya); to Cairene, as opposite to Alexandrian (here labelled iskandarāni); to Egyptian Arabic (interchangeably with ʿāmmiyya), spoken almost everywhere in the country and contrasting with Bedouin Arabic (here labelled badu, ʿarabi, ʿarabāwi); to Egyptian Arabic (interchangeably with ʿāmmiyya), spoken in the Delta region, as opposed to both Bedouin Arabic and Upper Egyptian dialects (labelled ṣaʿīdi).

Labels related to the social value or use of the variety

31The last class is constituted by labels that refer to the sociolinguistic status of the variety which is defined, or that imply a value judgement. Examples are:

- ʿāmmiyya, lahǧa ʿāmmiyya is used with a double meaning. In some maps, it indicated any non-standard Arabic variety, that is every variety spoken by the Arabic-speaking Egyptian citizens; in this case it is only opposed to sīwi and, sometimes, nūbi. In other maps, it indicates only some of the non-standard Arabic varieties spoken: in most cases it refers to the urban varieties of the Delta region as opposed to both non-Arabic languages and to Arabic specific speeches as Bedouin and Upper Egyptian dialects;

- fuṣḥā is used only once, to define the dialect spoken in Sinai. The respondent is most probably indicating that a Bedouin speech is found there and uses fuṣḥā to define it, in line with the longstanding stereotype of the similarity of Bedouin Arabic to the Standard language (e.g. Ferguson 1968).

5. Closing remarks

32The paper presents a preparatory analysis of a small amount of data; moreover, due to space limits, we have chosen to forgo theoretical discussion and literature review and devote space to the data. A further publication based on a twofold analysis of map-drawing tasks and interviews, conducted with a wider and diversified sample, is expected.

33Besides providing a draft of the perceptual dialectological map of Egypt, an interesting finding of the study is the fact that places and population that one might assume to be somehow related or relevant in Siwa Oasis dwellers’ lives, as the closest oasis of Bahariyya, were ignored, while places and population which are geographically further, as Upper Egypt and its speech, are always there in their linguistic map of the country. Further analysis is needed to reveal the actual extent to which each region is relevant and why. As for the labelling patterns, it is interesting, for example, the attribution of different meanings to labels as ʿarabi, maṣri, or ʿāmmiyya, according to their context of use, and the denomination of people and languages in general; a deeper analysis of the reciprocal relation between labels chosen by each respondent is required.

34On a wider perspective, our aim is to show that a perceptual dialectological approach that makes use of different kind of data, has much to provide in terms of our understanding of cultural and linguistic ideologies, stereotypes, ideas about one’s own positioning within a wider context and about the relationships between one’s own and other populations - and therefore languages. Finally, it is a valuable tool for investigating sociolinguistic trends and variation patterns.


Abu Lughod, L. 1986. Veiled Sentiments: Honor and Poetry in a Bedouin Society. Berkeley and Los Angeles: University of California Press.

Battesti, V. 2006. “Pourquoi j’irais voir d’en haut ce que je connais déjà d’en bas?” Battesti, V. & Puig, N. (eds.), Egypte/Monde arabe, Troisième série, Terrains d’Égypte, anthropologies contemporaines. 139-179.

Cole, D. & Altorki, S. 1998. Bedouin, Settlers, and Holiday-Makers. Egypt’s Changing Northwest Coast. Cairo: The American University in Cairo Press.

Ferguson, C. 1968 [1959]. “Myths about Arabic”, J. Fishman (ed.), Readings in the Sociology of Language. Paris, New York: Mouton – The Hague. 375-381.

Hachimi, A. 2015. ““Good Arabic, Bad Arabic”. Mapping Language Ideologies in the Arabic-speaking World”, ZAL 61. 35-70.

Haeri, N. 1996. The Sociolinguistic Market of Cairo: Gender, Class and Education. London and New York: Kegan Paul International.

Miller, C. 1996. “Nubien, berbère et beja : notes sur trois langues vernaculaires non arabes de l’Égypte contemporaine”, CEDEJ: Égypte/monde arabe, Première série, Les langue en Égypte. Retrieved on 09/08/2011 from

Miller, C. 2005. “Between Accommodation and Resistance: Upper Egyptian migrants in Cairo”, Linguistics 43. 903-956.

Preston, D.R. 1989. Perceptual Dialectology: Nonlinguists’ Views of Areal Linguistics. Dordrecht: Foris Publications.

Preston, D.R. (ed.) 1999. Handbook of Perceptual Dialectology. Vol. 1. Amsterdam/Philadelphia: John Benjamins Publishing Company.

Preston, D.R. 2010. “Language, People, Salience, Space. Perceptual Dialectology and Language Regard”, Dialectologia 5. 87-131.

Rosenbaum, G.M. 2008. “Ṣaʿīdī and Fallāḥī versus Cairene Dialects: Use and Function in Egyptian Culture and Literature”, Prochàzka, S. & Ritt-Benmimoum, V. (eds.), Between the Atlantic and Indian Oceans. Studies in Contemporary Arabic Dialects. Proceedings of the 7th AIDA Conference, held in Vienna from 5-9 September 2006. 383-393.

Serreli, V. 2011. Language attitudes in Siwa (Egypt). A Sociolinguistic Investigation. Unpublished MA Dissertation. University Ca’ Foscari of Venice.

Serreli, V. 2016. Society, Language and Ideologies in the Oasis of Siwa (Egypt). Listening to people’s voices. Unpublished PhD Dissertation. University of Sassari and University of Aix-Marseille.

Theodoropoulou, I. & Tyler, J. 2014. “Perceptual Dialectology of the Arab World: A Principal Analysis”, Al-ʿArabiyya 47. 21-39.

Wilmsen, D. & Woidich, M. 2007. “Egypt”, Versteegh, K. (ed.) EALL 2. Leiden: Brill. 1-12.


Valentina Serreli

University of Bayreuth,