Druckansicht der Internetadresse:

Sprach- und Literaturwissenschaftliche Fakultät

Professur für Arabistik – Prof. Valentina Serreli

Seite drucken


​Project supervisorsEinklappen
  • Manfred Woidich, Professor for Arabic and Islamic Studies (emerit.), University of Amsterdam
  • Jonathan Owens
Research staff:Einklappen
  • Dr. Jidda Hassan, Reader, Dept. of Languages and Linguistics, Maiduguri University, Maiduguri, Nigeria
  • Smaranda Grigore, doctoral student
  • Claudia Wolfer, doctoral student
  • Dr. Robin Dodsworth, North Carolina State University, Sociolinguistics and statistical analysis of data Visiting researcher

Deutsche Forschungsgemeinschaft


Beginning in July 2011 - renewed until August 2015


This study brings together elements of general linguistics, corpus linguistic and Arabic linguistics in order to investigate central issues of language change. The ultimate goal of the project is to determine the extent to which idiomatic structures are maintained across varieties separated diachronically and geographically.

In brief, the main points of the study are as follows.

1. Idiomatic structure in Nigerian and Egyptian Arabic will be compared, based on a comparison of spoken corpora, i.e. the study is based on naturally occurring speech, not elicited material.
Arabs moved out of Upper Egypt into the Sudan, Chad and Lake Chad area in the fourteenth century. These are the ancestors of today’s 500,000 or so native Arabic speakers in Nigeria, who form one locus of the study. This forms the historical basis of the study.

2. Whereas in the fundamental domains of phonology and morphology show few fundamental differences between Nigerian and Egyptian Arabic, idiomatic expressions are strikingly different and display little overlap. This is apparent if typical idiomatic collocations for the word “heart” are compared (using Hinds and Badawi (1986: 713) as the Egyptian Arabic source).

Nigerian Arabic

1. ma natallif leyum galbuhum „we don’t spoil their hearts = we don’t anger them”

2. galbi raagid “My heart is lying down = I am at ease”

3. bukurubni fi galba “he holds me in his heart = he has something against me”

4. galba faar “his heart boiled = he got angry”

5. galba angaTa’ “his heart got cut = he got scared”

Egyptian Arabic

6. ħaTTa ’id-ha ‘ala ’albaha “she put her hand on her heart = she is apprehensive”

7. yigiilak ’alb “may your heart come to you = have courage”

8. ’albi ‘andak “my heart is at you = I sympathize with you”

9. ’albi ‘aleek “my heart is on you = I worry about you”

10. ’alb iš-šams “the heart of the sun = the sun’s core”

The Nigerian idioms do not generally occur in Egyptian, and vice versa. Furthermore, generalizing a schematic meaning over the NA and EA usage, the central metaphor or metonymy of “heart” in NA is different from that in EA, which can be represented as follows.

Heart as symbol of emotional equilibrium: NA

Heart as symbol of concerned attitude, particularly for another: EA (6-9)

Heart as symbol of physical center: EA (= 10)

Equally, the different idioms are expressed by different collocational profiles. EA (10) has a locative possessor noun. Since the meaning of “heart as a location” is lacking in NA, so too, it may be expected, will locative possessor nouns of the noun galb be lacking.

A first goal of the study emerges from these comparisons. The collected idioms will be generalized into schematic meanings, and their typical range of collocations will be documented.

​Cognitive linguistics Einklappen

Descriptively the study falls within the larger domain of cognitive linguistics. Initial inspection of about 50,000 words of Nigerian Arabic text suggests that for key words such as “heart, eat, grab”, metaphorical usage far outweighs literal. galb “heart” rarely refers to a physical organ in the body. By the same token, the metaphors and metonymies behind the idiomatic usage are conventionalized metaphors which are quite frequent in everyday speech. The current study concentrates on such conventionalized metaphors.

We use the term ‘idiomatic structure’ to refer to the constituent lexemes which collocate to form the idioms. Whereas schematic meanings are defined on the metaphorical meanings, the measurement of the degree to which NA and EA differ will be performed on the idiomatic structure of the two varieties, i.e. on the constituent lexemes themselves.

Historical linguisticsEinklappen

It is hypothesized that the idiomatic structures of EA and NA differ significantly, where significance will be, inter alia, defined in statistical terms. It will further be inferred that observed differences arose from an ancestral situation in which NA was closer to EA idiomatically than it is today. In order to provide a baseline of comparison, a sample of functions words such as “at, in front, from, there is, here …” will be compared between the EA and NA samples. The assumption on this point is that basic function words will show a lesser degree of collocational difference between EA and NA than do the idiomatic collocations.

It should also be noted that Egyptian Arabic in particular will be examined for its own internal differentiation. A country of 60 million speakers may be assumed to be heterogeneous in ways to be determined.

​Language contact Einklappen

It is hypothesized that significant differences between EA and NA will be due, in the case of NA, to contact with African languages. It has already been noted, for instance, that a large number of idioms which differentiate NA from EA are calque-equivalents of Kanuri. The meaning expressed in (2) above, for instance, is the following, in EA, NA and Kanuri, the dominant language of the Lake Chad region.

EA nafsi mistirayyaħa “lit. my soul is resting”,

NA galb-I raagid lit. “my heart is lying down”

Kanuri: kare-nәm bo-zә-na “his heart is lying down, lit. heart-his is lying down”.

Clearly, NA patterns with Kanuri here.


Corpora will be used from Egypt and Nigeria which have been collected over many years by the two PI’s. They will be tagged for relevant linguistic features. They will also be edited and posted on this website and that of the Bayreuth-based DEVA, together with audio files.



The DataEinklappen
  • The Collection (PDF)
  • Transcription conventions
  • Text inventory
  • Studies relevant to Nigerian Arabic


The transcriptions of the codeswitching texts are presented in part in two formats. In one the
texts are presented as straight transcriptions with nothing added. In the second the extensive
coding which was used as the basis of various publications (see bibliography) is included.
Those interested can contact Jonathan Owens for the coding key. Many variables were
tagged for, but among them is a code for identifying which languages are in play where. In
both formats many of the Hausa passages are translated (in purple), so that for those who
have a background in Arabic at least, the gist of the entire text can be followed. An English
translation for about half of CS 5 is provided, and a short translation passage from CS 7 will
be provided as well.
All of the CS recordings took place in Maiduguri, many of the speakers in these texts also
appearing in the Maiduguri interviews and group conversations.


  • CS 5 (transcription) PDF
  • CS 5 (tagged) PDF
  • CS 5 (audio file) Audio
  • CS 5 (translation) PDF
Group ConversationsEinklappen


Klicken Sie hier, um zur vollständigen Sammlung der Texte und Übersetzungen zu gelangen.



Maiduguri interviews
The Maiduguri interviews were the first to be collected. Their purpose was twofold. On the one hand a sample of 56 individuals was made with a basic 2 x 2 division of men/women and under 30 vs. over 50. Added to these two basic demographic variables, there were four areal concentrations, three in neighborhoods with a high Arab density (Gwange, Gambori, Ruwan Zafi) and one a sampling of Arabs living in largely non-Arab districts. As the sociolinguistics of Maiduguri (unfortunately, like too much of the Arabic-speaking world) was completely terra incognita, the survey established basic benchmarks. As far as the content of the interviews go, they were largely orientated towards factors which elucidate the dynamics of Arabic as a minority language in a growing urban area. Most interviews therefore entered around a fairly set protocol, though the questions were always introduced in as informal, conversational a manner as possible. After asking basic biographical information, questions included the language repertoire of the interviewees and their immediate family and circle of friends, whether they had studied formally either in the Quranic school system (sangaaya*) or western education and in which languages they were literate, their exposure to media Arabic, including non-Nigerian Arabic stations (BBC,Kuwait, Chad, Sudan etc.), how common inter-ethnic marriages were and what they implied for language use in the family, whether the interviewees could distinguish different dialects of Nigerian Arabic, and general questions about the social life and institutions of Arabs in Maiduguri. The senior author was always accompanied by an Arab colleague, who was encouraged to participate in the questioning as well. As the interviews progressed, they became more fluent, if slightly routine. Mr. Dana Allamin became particularly adept at guiding the Maiduguri interviews, as well as many of the village ones.
* Curiously, the same etymological origin as ‘synagogue’, probably from North Africa.




Introduction (PDF)


Klicken Sie hier, um zur vollständigen Sammlung der Texte und Übersetzungen zu gelangen.

For aficionados (transcriptions only)


Verantwortlich für die Redaktion: Juniorprofessor Valentina Serreli

Facebook Twitter Youtube-Kanal Instagram Blog UBT-A Kontakt