Back Home Up Next

The Nijmegen Arabic/Dutch Dictionary Project

 

AIM AND METHODS

by

Jan Hoogland

&

Kees Versteegh

 

this article was published in:
Revue de la Lexicologie
Mélanges Offerts au professeur Mohammed Rached Hamzoui
ed. 14-15,1998-1999
editors: Ibrahim Ben Mrad, Abdessattar Joober

This article was published about two years after the execution of the project started (and about three years before the execution was completed).

1. Introduction: The need for a dictionary of Arabic in the Netherlands

Summary of this section

2. The aim of the dictionary project

Summary of this section

3. The methods used in compiling the dictionary

Summary of this section

4. Bibliographical references

1. The need for a dictionary of Arabic in the Netherlands
- summary of this section

The study of Arabic in the Netherlands has had a long and almost uninterrupted tradition since the 16th century. Those who studied the language either had religious motives, because they wished to apply their knowledge of Arabic to the study of the Hebrew Bible, which was written in a related language, or - and in this respect the Netherlands were rather different from other countries - their motives were commercial. In line with the practical nature of the Dutch Republic such motives were usually regarded as entirely compatible with the scholarly study of Arabic. To give just one example: when he was appointed as professor of Arabic in the University of Leiden Golius (1596-1667) first decided to make a long journey to the Arab world - something quite unusual for a scholar in those days - not only in order to fetch Oriental manuscripts, but also to establish and confirm commercial contacts.

     Part of such missions was the collection of materials for the lexical knowledge of Arabic, an indispensable part of the study of Arabic. After his return Golius, for instance, published with his Lexicon arabico-latinum the first European dictionary of Arabic, which was to remain the most important authority on the language for several centuries. This tradition has never died since. In the 19th century Dozy's contributions to the lexicography of Arabic constituted a considerable advance on the existing lexicographical materials of the time (cf. Versteegh 1987).

     Obviously, the attitude towards the study of Arabic and the need for knowledge about this language has changed since the days of Golius. But the commercial motive is still there: about 6% of Dutch export concerns the countries in North Africa and the Middle East, and while many traders firmly believe that knowledge of English and French is enough for their commercial purposes, there is a growing conviction that for a successful interaction knowledge of the main language of the Middle East is essential.

     A new factor in Dutch attitudes towards Arabic as a language is the presence of a large number of Arabophones in the Netherlands. Unlike many other countries the Netherlands has never been a country of immigrants. Yet, in the course of its history sizeable minorities have come to the country. From the fifties onwards large numbers of immigrants were recruited from the Mediterranean countries to work as a cheap labour force. During the early stages of the process of migration it was generally thought that most of the foreign workers, as they were called, would work for a few years and then return to their country of origin. This meant that there was no need for them to integrate into Dutch society and they only had to learn Dutch to such a degree that they could communicate on the shop floor. After some years, however, it turned out that the Moroccans and Turks working in the Netherlands did not intend to return to their own countries at all. Many of them brought their family over and opted for the Netherlands as their new country. At the moment the number of Moroccan immigrants is approximately 200,000. This number includes children born of Moroccan parents in the Netherlands.

     When it became clear that the migrants in the Western European countries were there to stay, the government was forced to modify its policies. The policy of the Dutch government is marked by a series of shifts in aims and means. At first all efforts were directed at the acquisition of Dutch as a second language. Early on, however, a discussion arose about the need to provide for a curriculum for immigrant children in their own language, in line with current thinking about the importance of first language acquisition in the acquisition of a second language. The ideas about home language instruction (HLI, in Dutch Onderwijs in Eigen Taal, OET) crystallized in the establishment of a number of curricula for primary schools.

     In 1982 60% of the Moroccan schoolgoing children followed HLI lessons, in 1987 70%. Since 1987 HLI is also offered in secondary schools, starting with 23 schools, and increasing to 100 schools at the moment. Of the Moroccan children attending a form of secondary education 74% participate in these lessons, which deal with the standard language exclusively. Since 1990 it is possible for secondary schools to include Moroccan and Turkish in the final exams; the first such exams have already been taken.

     A basic problem arises when we look at the definition of what constitutes the 'home language' for the Moroccans involved. It is often assumed, naively, that the children concerned are native speakers of their home language. As a matter of fact, the majority of these children do not have Arabic as their first language at all. True, there is a constant influx of children who come from Morocco to Holland at an advanced age, having received schooling in Morocco. These children are not likely to lose their own language, whether it is Moroccan Arabic or Berber, and their proficiency in Standard Arabic is on the same level as their peers in Morocco. But those who were born in the Netherlands very often do not speak Moroccan well. In El-Aissati (1996) numerous examples are given of the structural deficiencies in the language of Moroccan youngsters. Although these children usually understand Moroccan Arabic reasonably well, their own production of the language is poor, and they make morphological and syntactic mistakes. The problem for many of these children is that they do not feel at ease in speaking the language of their parents. They speak haltingly, with an impoverished vocabulary, and when given the chance, tend to switch to Dutch, the language they habitually use with their siblings and their friends. Even though their comprehension of the language is still intact, at least on the level of everyday conversation, their own production falls far behind. Native speakers immediately recognize the difference between speech samples by immigrant children and those from children who speak the language as their native tongue. This underlines the need for teaching materials in Modern Standard Arabic, if indeed Modern Standard Arabic is the language to be taught to the children from these minorities (see below).

     An authoritative report by a committee of experts in the field of bilingual education came out in 1992. It was entitled Cedars in the backyard and presented a number of recommendations concerning the policy to be followed in education of linguistic minorities. With regard to the teaching of the language of origin of these minorities the committee criticized the successive aims of the current curriculum, paedagogical, transitional, emancipatory, and proposed a new approach to this form of instruction. In their view the languages of the linguistic minorities should be treated on a par with other foreign languages and occupy a similar position both in primary and in secondary schools. The value of the languages involved and their inclusion in the curriculum should be autonomous and not derived from any ulterior aims. Arabic and Turkish, no doubt the two most important minority languages in the Netherlands, should be taught as living foreign languages.     The solution of the Cedars report was to step away from the aims of HLI as it had been given until then. The authors no longer believed that HLI helped to improve the level of Dutch proficiency or the sense of identity of the immigrant children. Instead, they called for a new status of the languages of the minorities, for what they called autonomous reasons. In their view, there was an intrinsic importance in teaching the languages of the minorities, both for the minorities involved and for the state.

     This judgment of the committee may appear to be somewhat on the optimistic side. It is not certain that in the future Dutch-speaking children will join the children of the minorities in the lessons of Arabic and Turkish. Besides, so far there has been no concern for unification of the curriculum. If Arabic were to take a place alongside French and German (the situation of English is so patently different that any comparison becomes void) teachers should at least have at their disposal course materials that are geared to their special needs.

     As matters stand at the moment, however, almost no materials are available for these learners of Arabic. Specifically, there are no reliable dictionaries to assist them in their difficult task of learning Arabic. In the universities, most students are able to handle the English-language Arabic dictionaries such as Wehr. But using a dictionary in a foreign language is not a feasible option for learners in secondary school.

     All major European languages, French, German, English, Spanish, Italian, have at least a basic dictionary of Arabic; even some of the minor languages, especially in Central and Eastern Europe, have such a dictionary. In Dutch nothing of the kind is available. This is why in 1995 it was decided by the CLVV (Commissie voor Lexicografische Vertaalvoorzieningen), a commission financed by the Dutch and Flemish governments to initiate a project which was to lead to two sets of bilingual dictionaries, a smaller one of about 15,000 lemmas and a larger one of about 40,000 lemmas. The smaller project was entrusted to Mark van Mol (University of Leuven, Belgium), who intends to finish his project in 1999 (these learners' dictionaries were published in 2000, more information is available through this link. The larger project is being carried out by a team of the University of Nijmegen (The Netherlands), led by an editorial committee consisting of Prof. Manfred Woidich (University of Amsterdam) and both authors of the present paper. According to the time schedule followed by the team the dictionary will be completed in 2001.


2. The aim of the dictionary project
- summary of this section

Obviously, the children of the Moroccans in the Netherlands constitute the first target group for a Dutch/Arabic dictionary. But precisely in the case of this target group, when we speak about an Arabic dictionary we need to define more clearly what exactly we mean by 'Arabic', both in terms of the lexicographical materials available and in terms of the target group of the dictionary. The most immediate problem concerns the definition of the language choice for the potential users in the Moroccan minority in the Netherlands. In the case of other minorities it is relatively easy to determine the language that is regarded by the members of the community as their own language, but in the case of the Moroccan minority it is much more difficult to identify this language. The use of the term 'home language' in HLI has created a lot of confusion in the debates about the need for HLI and in the implementation of programmes, as well as in the training of teachers for the programmes. When asked about their own language many informants simply say they use Arabic or Moroccan, which may mean either Standard Arabic as the prestige language of the Arab countries, or Moroccan Arabic as the vernacular of Morocco, or Berber, the language of the majority of the Moroccans residing in the Netherlands (cf. Boumans 1998).

     The choice of language variety for a dictionary is directly connected with the question about the language variety to be taught in HLI. Especially in the case of the Moroccan minority, the different aims of HLI led to different choices, sometimes simultaneously. If HLI was to serve as a paedagogical or transitional instrument, the obvious choice was the language of the parents: children speaking Berber are not helped when offered a curriculum in Standard Arabic, since that is a completely foreign language for them, at least in the Dutch context. In the Moroccan situation all Berber children go to Arabic speaking schools and are expected to learn that language (cf. Wagner 1993).

     A second remark concerns the status of the Standard Arabic language. While the response to the curriculum in Standard Arabic has always been very positive in the Moroccan community, it is not certain that this response had anything to do with the implicit or explicit aims of the Dutch government. As a matter of fact, the main reason why parents were in favour of HLI seems to have been their wish to preserve the values of the Islamic community to which the Moroccans belong. These parents apparently believed that the only way to preserve these values and to help them raise their children as good muslims was to give them a proper training in Standard Arabic. When a choice must be made for the language to be taught in the HLI programmes, one is confronted - even apart from the problem of Berber - with the presence of two varieties of Arabic: vernacular and standard Arabic. Most of the official programmes for HLI for Moroccans do not select the vernacular as the language variety, but the Standard Arabic that is prescribed in the Moroccan school system. In other words, the variety taught does not conform to the explicit aims of the programmes for HLI, which is to teach the actual home language. In the Dutch system of HLI Moroccan Arabic is the language of instruction in the first two years of primary school, but in later years when the emphasis is shifted towards reading and writing Moroccan children are expected to learn Standard Arabic.

     The Moroccan children in the Netherlands, though they are the most important target group, are not the only people in need of a good dictionary. A second category is that of Dutch students of Arabic. Although the report Cedars assumes that Arabic could be introduced as a normal component of the school curriculum, there is not much chance that this will actually happen in large numbers. There are however ample opportunities for Dutch students who wish to learn Arabic: six universities offer a complete curriculum of Arabic and while the methods and aims differ, all curricula include Modern Standard Arabic. The total number of yearly enrollments is approximately 45. Besides, there is a formal translator training programme at the University of Maastricht. Apart from this a large number of so-called open universities offer courses in Modern Standard Arabic and although no exact numbers are known it may safely be assumed that the total number of people trying to learn Arabic at any given moment is more than 300.

     Apart from the above mentioned group of Moroccan children in Dutch secondary education there is a large group of Moroccan adult learners of Dutch, either those who have recently arrived or those who have never had the opportunity to learn Dutch. They attend Dutch lessons in a large variety of courses, both private and government sponsored. Their need for reliable dictionaries is perhaps even larger than that of the children since most of these adults are unable to consult dictionaries in other languages and are totally dependent on a Dutch/Arabic and Arabic/Dutch dictionary.

     A third group is that of the growing number of translators who are active in the field of translation and whose work has steadily become more important over the last decades. Their work covers both directions of translations. They translate Dutch government information into Arabic, or Arabic legal, marriage and divorce documents into Dutch for private persons, or commercial papers such as tenders for commercial companies.

     Finally we may mention here those people in the Arab world who wish to learn Dutch for whatever reason. Courses of Dutch are at present given in Egypt, Syria and Morocco and although the total number of new students rarely exceeds 30, the enthusiasm of these students is high and the public relations value of such courses should not be underestimated.

     The variability in aims of these groups is obvious, and this made it very difficult for the editorial team of the Arabic dictionary to opt for one language variety. In the end we decided to use Standard Arabic exclusively. For the purposes of the dictionary Modern Standard Arabic was defined as the written language of the media in contemporary Egypt and Morocco. If there are differences between the way Modern Standard Arabic is realised in these two countries this is marked in the dictionary. The choice for Egypt as one of the two reference points for the use of Modern Standard Arabic is based on the interests and needs of most Dutch-speaking Arabists and on the large cultural importance of Egypt and its supraregional function in the Arab world. The choice for Morocco is dictated, of course, by the presence of large numbers of Moroccans in the Netherlands and Flanders. In this connection it may be noted that the cultural treaty between Morocco and Netherlands explicitly mentions the creation of a Dutch/Arabic dictionary as one of the activities both countries are to support.

     The aim of the project is then the compilation of two bilingual learner's dictionaries Arabic/Dutch and Dutch/Arabic.
Both volumes of the dictionary deal with the language from a synchronic perspective and are geared towards contemporary Dutch and Modern Standard Arabic. For the lexicon this means that the dictionary aims at the inclusion of those terms that are essential for the understanding of modern society, e.g., social, political and economic circumstances. The language of literature and the Qur'an is included only insofar as it is needed to understand the language of the media. Since the dictionary is not meant as a technical dictionary, only general technical terms from such fields as economy, law, and so on, are included; for specific technical terminology reference will be made to technical vocabularies.

     The dictionaries will be bidirectional, i.e., they may be used in both directions for purposes of production and comprehension. The main argument for this dual functionality is that both volumes should be useful for users with both an Arabic and a Dutch linguistic background. The Dutch/Arabic part needs to be a production dictionary because it should cater for the needs of a comparatively large group of Dutch and Flemish students of Arabic, who need assistance when they wish to translate into Arabic from Dutch. As we have seen above, a large part of the Moroccans in the Netherlands (and something similar applies to North African immigrants in Flanders), cannot be regarded as fully competent in Standard Arabic, quite apart from the fact that in many cases it is doubtful whether they can be regarded as native speakers of any Arabic variety at all. Especially for those who were born in the mahjar contact with Arabic is limited to a minimum and while the language has an important symbolic function it tends to be used less and less in in-group communication. When these children take up Arabic in secondary school as a subsidiary subject, a production dictionary is as important to them as it is to those of their fellow students who do not know any Arabic at all but wish to learn the language. Such a production dictionary could play an important role both as a learning tool in second language acquisition and in the preservation of the home culture of these children, as well as in the improvement of the quality of government related translation for educational or instructional purposes

     At the same time the Dutch/Arabic volume should contain enough information to make it useful for those users in the Netherlands and in the Arab world who know no Dutch at all but are learning the language. For them the function of this volume is comprehension and the information included in the description and translation of the lemmas should be sufficient to enable them to use the dictionary profitably.

     The Arabic/Dutch volume, on the other hand, must be a production dictionary since it is intended specifically for Arabophones. They need this production function since it enables them to use Dutch at a high level. At the same time the Arabic/Dutch volume should be able to function as a comprehension dictionary for native speakers of Dutch or Moroccan children who have lost their own language.


3. The methods used in compiling the dictionary
- summary of this section

Although a number of bilingual Arabic/Dutch dictionaries are available on the Dutch market (e.g., Derwish 1984 Dutch/Arabic, Farouk 1995 Dutch/Arabic, Amien 1988 Dutch/Arabic, Amien 1998 Arabic/Dutch), these are better described as vocabulary lists, since they contain only very little grammatical information or no information at all, and examples with entry-words in context or expressions are almost completely lacking. No distinction is made between various meanings of words, or the semantic shades are entered in an unsystematic way. Most of these products seem to be intended solely for Arabic-speaking users, i.e., for the production or comprehension of Dutch. Dutch-speaking learners of Arabic in many cases will not find the necessary information in these dictionaries.

Through this link you can view examples from these different dictionaries.

     A number of requirements for the dictionary under construction in the present project may be derived from this. The outcome of the project has to cater for the needs of both Arabic-speaking and Dutch-speaking users. It should contain grammatical information in both languages, since both parts of the dictionary will have to function as a production dictionary and comprehension dictionary at the same time. And finally, in contrast with all existing products, both parts will have to contain a substantial number of expressions and examples presenting the entries in context. These examples should be taken from authentic texts in both languages involved.

     The present project started with a Dutch lexical corpus that had been made available by the funding organisation of the project, the CLVV. This corpus is called RBN (Referentie Bestand Nederlands, Reference File Dutch). Since the compilation of this corpus has taken place according to acknowledged lexicographical standards, the staff of the present project could concentrate on creating its counterpart, the Arabic lexicographical corpus (ALC). The editorial team decided to proceed along the following track. First of all we add Arabic equivalents to the words and expressions in the RBN. This constitutes the point of departure for the Dutch-Arabic part, and the first step towards the creation of the ALC.
The following link will show an example of the RBN data.

 

     The process of adding these Arabic equivalents to the Dutch data takes place along the lines of a system devised by the already mentioned Dutch/Flemish commission, which aims at constructing bilingual databases for dictionary compilation by 'smart reversion'. By using this system we are constantly able to reverse the existing database from Dutch with added Arabic equivalents to an Arabic-Dutch database. During the second stage of the project we shall concentrate on the completion of the ALC and the addition of Dutch equivalents to the words and expressions that have been added during this completion process. Presumably, the Dutch part will also increase in size during this next stage.
The following link leads you to the section about the working methods.

 

     Obviously there must be a quantitative balance between the RBN and the ALC. Therefore, the ALC has to contain a considerable number of expressions, examples and collocations, just like its Dutch counterpart, the RBN, which is a very rich source of representative language as it is being used both in spoken and written form. Since the RBN contains three levels of semantic units (lexical units, example units, idioms), the ALC will have to contain the same levels, with the same degree of richness.

     The process of compiling the Arabic lexicographical corpus (ALC) will be achieved in several stages, and on several levels. The first stage will be completed after we finish the process of adding Arabic equivalents to the Dutch entries in their various meanings, and to the Dutch expressions and examples. This process has not yet been completed at the time of writing, but it is  possible to give some figures on the basis of the present situation. The translation of around 20,000 Dutch nouns has resulted in a list of about 10,000 Arabic nouns. This result may at first sight seem surprising, but at least three explanations can be advanced. Firstly, Dutch is very productive in creating compounds, and many thousands of these compounds have been fully lexicalized and entered into the RBN as dictionary entries. Arabic, on the contrary, only allows the creation of compounds in exceptional cases; the usual equivalent of a Dutch compound is either a combination of two nouns in a construct phrase, or a combinations of a noun with a relative adjective (nisbe). We have decided to treat these construct phrases and noun-adjective combinations as examples and not as independent entries. As a result, thousands of Dutch entries will be 'hidden' as examples in the Arabic entries, thus causing the number of Arabic nouns to be substantially lower then the number of Dutch nouns. On the highest level of meaning units, i.e. the lexical unit, there will therefore be a certain quantitative imbalance between both languages: lexical units in Dutch will be translated with example units in Arabic.
The following link leads you to the section about concordancing.

 

     A second explanation for the difference in number of entries may be derived from the fact that for many Dutch nouns there simply does not exist an Arabic equivalent. In specific domains related to Dutch society, or modern industrial society or the welfare state, many terms and expressions exist for which it is impossible to present an Arabic equivalent. In such cases there is an option to describe the meaning of such words as an explanation or a definition. These descriptions are stored in the database in such a way that during the reversion process they will not be incorporated in the Arabic lexicographical corpus, since they are not derived from authentic Arabic texts, but invented as neologisms or calques on the basis of the Dutch terms. It goes without saying that in the next stage, when Arabic words will be excerpted from the Arabic text corpus, the reverse effect will take place as well: many Arabic words, related to specific semantic domains such as religion, administration, law, and even such domains as desert and bedouin life, will be entered into the corpus, without any Dutch equivalents being available.
The following link leads you to the section about descriptions.

 

     A third explanation for the discrepancy between the number of Dutch and Arabic entries may be found in the fact that Dutch is both a language being used for formal goals and a spoken language, since there is no diglossia in the Netherlands. The RBN, being a representative lexicographical corpus, also contains a considerable number of words and expressions taken from the spoken language, which in the RBN are labeled as informal, or even as slang. Since the present project aims to cover only Modern Standard Arabic, which is a language appropriate for more or less formal situations only, many of the informal Dutch words and expressions do not have an adequate equivalent in Modern Standard Arabic. For such words and expressions a unidirectional description is given.
 

     During the next stage, that of completing the Arabic lexicographical corpus we shall add, among other things, Arabic words belonging to a number of semantic domains which do not exist in the Dutch situation. The Islamic domain, for example, is only represented in the RBN by a limited number of terms. A corpus of texts on Islamic subjects will therefore be checked in order to add to the ALC typically Islamic terms.

     Another step to be taken, will be the comparison of the list of Arabic entries resulting from the first stage (translating from Dutch into Arabic) with a list of words derived from a corpus of authentic Arabic texts. Preferably, such a list should have been a frequency list of Modern Standard Arabic. However, in view of the scarcity of reliable frequency lists of Modern Standard Arabic (Fromm, Kouloughli, Abdu, Landau , and their quantitative limitations (they contain only the 3000 to 4000 most frequent words of the Modern Standard Arabic lexicon), neccesary information about the so-called upper segment of the lexicon is not directly available. The production of a reliable and accurate frequency list on the basis of our text corpus is not feasible without investing a huge amount of time. In order to be able to produce such a frequency list, the corpus should contain two types of information, which are absent in our corpus. In the first place, the corpus has been stored without vowels. An unvowelled corpus can hardly be used to extract reliable and accurate frequency data. The second type of necessary information is grammatical and morphological tagging.  Unvocalised Arabic texts contain a large number of homographs, which can only be disambiguated by means of grammatical tagging.
The following link leads you to the section about the Arabic frequency lists.

 

     In order to compensate for the lack of reliable frequency data in completing the ALC, two solutions have been chosen.
First of all we have created a very rough frequency list by processing the corpus with a number of search-and-replace operations to separate frequent prefixes and suffixes from words in the corpus. This has resulted in a list of many thousands of words, which will be used to check against the entries of the ALC after completion of the Dutch/Arabic translation stage.
The following link shows some fragments from this 'stripped' corpus.

 

Secondly, a comparison will be made between the prefinal ALC and other dictionaries. We still have to make the final selection of those modern dictionaries of Modern Standard Arabic that will be used for this purpose. Not many dictionaries have been published during the last years.
The following link shows information on the Dutch-Arabic and Arabic-Dutch learners' dictionaries of Mark van Mol.

 

     As for the second level of meaning units, i.e. examples and idiomatic expressions, it is to be feared that existing recently published dictionaries will not be a rich source of information. Especially the category of collocations is very poorly represented in existing dictionaries.

In 1993 one of the present authors carried out an analysis of a number of dictionaries  on the topic of rather frequent collocations in Modern Standard Arabic (Hoogland 1993). The conclusion drawn from this comparison was that learners of Arabic have at their disposal very few materials to consult in order to find collocations in Arabic. Bilingual and monolingual dictionaries turned out to be equally incomplete in helping the user find very frequent combinations of nouns and verbs. Because of the unpredictability of these combinations the user, being a learner or imperfect user of at least one of the languages involved, absolutely needs to find this information in the dictionary.

More on this topic can be read at the pages about collocation.
The following link leads you to the article about collocation in Arabic.

 

     Filling in the level of example units will therefore constitute one of the most important steps that remain to be taken. The ALC will be enriched with a considerable quantity of useful idiomatic expressions, collocations and illustrative examples. This process will be carried out in various ways. In the first place, the process of translating the Dutch/Arabic part has resulted in a considerable number of expressions and collocations. Secondly, all Arabic words in the ALC will be treated systematically in order to add examples in context. As a result of reading a vast amount of Arabic text, thousands of collocations and idiomatic expressions have been selected and stored in a database containing 'rough materials' which may be incorporated in the dictionary. Another very rich source of rough materials is the 3 million words text corpus in combination with a concordance program, which yields thousands of occurrences of words presented in a very structured way. More text is available, but for frequent words this size of the corpus is sufficient; for less frequent words we will be able to use over 8 million words. Finally, native speakers of Arabic will add collocations and other types of occurrences in context by introspection.

     It is our hope that this process will indeed result in the compilation of a rich lexicographical corpus of Arabic. The diverse approaches guarantee that the corpus compiled will be balanced and representative. We furthermore hope and expect that the final product will be both innovative, because of the tools being used, and useful, because of the inclusion of numerous idiomatic expressions and collocations.


4. Bibliographical references

Amien, Sharif. 1988. Amien's Nederlands-Arabisch Woordenboek. Rotterdam: Arabisch-Nederlandse Uitgeverij.
(Amien's Dutch-Arabic dictionary)

 

Amien, Sharif.1998. Amien's Groot Woordenboek Arabisch-Nederlands. Rotterdam: Arabisch-Nederlandse Uitgeverij.

(Amien's Large Arabic-Dutch dictionary)

 

Boumans, Louis. 1998. The syntax of code-switching: Analysing Moroccan Arabic/Dutch conversation. Ph.D. University of Nijmegen.

Derwish, H.H. 1984. Kramer's Woordenboeken, Nederlands-Arabisch. Amsterdam. (Kramer's dictionaries, Dutch-Arabic)

Cedars in the Backyard. 1992. Ceders in de tuin: Naar een nieuwe opzet van het onderwijsbeleid voor allochtone leerlingen. Den Haag: Ministerie van Onderwijs en Wetenschappen.

El Aissati, Abderrahman. 1996. Language loss among native speakers of Moroccan Arabic in the Netherlands. Ph.D. University of Nijmegen.

Farouk, Ibrahim A. 1995. Al Manhal Nederlands-Arabisch Woordenboek. Amsterdam: Arabesk Woordenboeken.
(Al Manhal Dutch-Arabic dictionary)

 

Hoogland, Jan. 1993. "Collocations in Arabic (MSA) and the treatment of collocations in Arabic dictionaries". Proceedings of the Colloquium on Arabic Lexicology and Lexicography, ed. by Kinga Dévényi, Tamás Iványi & Ariel Shivtiel, 75-93. Budapest: Eötvös Loránd University & Csoma de Körös Society.

Versteegh, Kees. 1987. "Nahwiyyuna wa-lughawiyyuna wa-mawqif Dozy iza' at-turat an-nahwiy al-`arabiy". Fi l-mu`jamiyya al-`arabiyya al-mu`asira, ed. by Ahmed El-Ayed & Ibrahim Ben Mrad, 401-413. Tunis: Dar al-garb al-Islam_.

Wagner, Daniel A. 1993. Literacy, Culture, and Development: Becoming literate in Morocco. Cambridge: Cambridge University Press.

summary Introduction

The study of Arabic in the Netherlands has had a long and almost uninterrupted tradition since the 16th century. Golius, professor of Arabic in the University of Leiden (1596-1667) first decided to make a long journey to the Arab world, in order to fetch Oriental manuscripts, but also to establish and confirm commercial contacts.

     Part of such missions was the collection of materials for the lexical knowledge of Arabic, an indispensable part of the study of Arabic. After his return Golius, for instance, published with his Lexicon arabico-latinum the first European dictionary of Arabic, which was to remain the most important authority on the language for several centuries. In the 19th century Dozy's contributions to the lexicography of Arabic constituted a considerable advance on the existing lexicographical materials of the time.     A new factor in Dutch attitudes towards Arabic as a language is the presence of a large number of Arabophones in the Netherlands. From the fifties onwards large numbers of immigrants were recruited from the Mediterranean countries to work as a cheap labour force. After some years, it turned out that the Moroccans and Turks working in the Netherlands did not intend to return to their own countries. Many of them brought their family over and opted for the Netherlands as their new country. At the moment the number of Moroccan immigrants is approximately 200,000. This number includes children born of Moroccan parents in the Netherlands.

In a certain stage, a discussion arose about the need to provide for a curriculum for immigrant children in their own language, in line with current thinking about the importance of first language acquisition in the acquisition of a second language. Since 1990 it is possible for secondary schools to include Moroccan and Turkish in the final exams; the first such exams have already been taken. This underlines the need for teaching materials in Modern Standard Arabic, if Modern Standard Arabic is the language to be taught to the children from these minorities (see below).

     As matters stand at the moment, however, almost no materials are available for these learners of Arabic. Specifically, there are no reliable dictionaries to assist them in their difficult task of learning Arabic. In the universities, most students are able to handle the English-language Arabic dictionaries such as Wehr. But using a dictionary in a foreign language is not a feasible option for learners in secondary school.

     All major European languages, French, German, English, Spanish, Italian, have at least a basic dictionary of Arabic; even some of the minor languages, especially in Central and Eastern Europe, have such a dictionary. In Dutch nothing of the kind is available. This is why in 1995 it was decided by the CLVV (Commissie voor Lexicografische Vertaalvoorzieningen), a commission financed by the Dutch and Flemish governments to initiate a project which was to lead to two sets of bilingual dictionaries, a smaller one of about 15,000 lemmas and a larger one of about 40,000 lemmas. The smaller project was entrusted to Mark van Mol (University of Leuven, Belgium), who intends to finish his project in 1999. The larger project is being carried out by a team of the University of Nijmegen (The Netherlands), led by an editorial committee consisting of Prof. Manfred Woidich (University of Amsterdam) and both authors of the present paper. According to the time schedule followed by the team the dictionary will be completed in 2001.

Summary of: 2. The aim of the dictionary project

Obviously, the children of the Moroccans in the Netherlands constitute the first target group for a Dutch/Arabic dictionary.

          The Moroccan children in the Netherlands, though they are the most important target group, are not the only people in need of a good dictionary. A second category is that of Dutch students of Arabic.  There are ample opportunities for Dutch students who wish to learn Arabic: six universities offer a complete curriculum of Arabic and while the methods and aims differ, all curricula include Modern Standard Arabic. The total number of yearly enrollments is approximately 45. Besides, there is a formal translator training programme at the University of Maastricht. Apart from this a large number of so-called open universities offer courses in Modern Standard Arabic and although no exact numbers are known it may safely be assumed that the total number of people trying to learn Arabic at any given moment is more than 300.

     Apart from the above mentioned group of Moroccan children in Dutch secondary education there is a large group of Moroccan adult learners of Dutch. Their need for reliable dictionaries is perhaps even larger than that of the children since most of these adults are unable to consult dictionaries in other languages and are totally dependent on a Dutch/Arabic and Arabic/Dutch dictionary.

     A third group is that of the growing number of translators who are active in the field of translation and whose work has steadily become more important over the last decades. Their work covers both directions of translations.

     Finally we may mention here those people in the Arab world who wish to learn Dutch for whatever reason. Courses of Dutch are at present given in Egypt, Syria and Morocco and the public relations value of such courses should not be underestimated.

     The variability in aims of these groups is obvious. For the purposes of the dictionary Modern Standard Arabic was defined as the written language of the media in contemporary Egypt and Morocco. If there are differences between the way Modern Standard Arabic is realised in these two countries this is marked in the dictionary. The choice for Egypt as one of the two reference points for the use of Modern Standard Arabic is based on the interests and needs of most Dutch-speaking Arabists and on the large cultural importance of Egypt and its supraregional function in the Arab world. The choice for Morocco is dictated, of course, by the presence of large numbers of Moroccans in the Netherlands and Flanders.

     The aim of the project is then the compilation of two bilingual dictionaries Arabic/Dutch and Dutch/Arabic.

Both volumes of the dictionary deal with the language from a synchronic perspective and are geared towards contemporary Dutch and Modern Standard Arabic. For the lexicon this means that the dictionary aims at the inclusion of those terms that are essential for the understanding of modern society.

     The dictionaries will be bidirectional, i.e., they may be used in both directions for purposes of production and comprehension. The main argument for this dual functionality is that both volumes should be useful for users with both an Arabic and a Dutch linguistic background.

Summary of: 3. The methods used in compiling the dictionary

     A number of requirements for the dictionary under construction in the present project can be formulated. The outcome of the project has to cater for the needs of both Arabic-speaking and Dutch-speaking users. It should contain grammatical information in both languages, since both parts of the dictionary will have to function as a production dictionary and comprehension dictionary at the same time. And finally, in contrast with all existing products, both parts will have to contain a substantial number of expressions and examples presenting the entries in context. These examples should be taken from authentic texts in both languages involved.

     The present project started with a Dutch lexical corpus that had been made available by the funding organisation of the project, the CLVV. The staff of the present project had to concentrate on creating its counterpart, the Arabic lexical corpus (ALC). The editorial team decided to proceed along the following track. First of all we add Arabic equivalents to the words and expressions in the RBN. This constitutes the point of departure for the Dutch-Arabic part, and the first step towards the creation of the ALC. During the second stage of the project we shall concentrate on the completion of the ALC and the addition of Dutch equivalents to the words and expressions that have been added during this completion process. Presumably, the Dutch part will also increase in size during this next stage.

     The process of compiling the Arabic lexicographical corpus (ALC) will be achieved in several stages, and on several levels. The first stage will be completed after we finish the process of adding Arabic equivalents to the Dutch entries in their various meanings, and to the Dutch expressions and examples.

     During the next stage, that of completing the Arabic lexicographical corpus we shall add, among other things, Arabic words belonging to a number of semantic domains which do not exist in the Dutch situation. The Islamic domain, for example, is only represented in the RBN by a limited number of terms. A corpus of texts on Islamic subjects will therefore be checked in order to add to the ALC typically Islamic terms.

     Another step to be taken, will be the comparison of the list of Arabic entries resulting from the first stage (translating from Dutch into Arabic) with a list of words derived from a corpus of authentic Arabic texts. Preferably, such a list should have been a frequency list of Modern Standard Arabic.      As for the second level of meaning units, i.e. examples and idiomatic expressions, it is to be feared that existing recently published dictionaries will not be a rich source of information. Especially the category of collocations is very poorly represented in existing dictionaries.

     Filling in the level of example units will therefore constitute one of the most important steps that remain to be taken. The ALC will be enriched with a considerable quantity of useful idiomatic expressions, collocations and illustrative examples.

     It is our hope that this process will indeed result in the compilation of a rich lexicographical corpus of Arabic. The diverse approaches guarantee that the corpus compiled will be balanced and representative. We furthermore hope and expect that the final product will be both innovative, because of the tools being used, and useful, because of the inclusion of numerous idiomatic expressions and collocations.

reactions to: j.hoogland@let.kun.nl
last updated 26/10/2003 15:16 +0100
Back Home Up Next