LINGUIST List 29.597

Mon Feb 05 2018

Review: Arabic, Andalusian; Historical Linguistics; Lexicography: Corriente, Pereira, Vicente (2017)

Editor for this issue: Clare Harshey <>

Date: 12-Sep-2017
From: David Wilmsen <>
Subject: Encyclopédie linguistique d’Al-Andalus Vol. 2
E-mail this message to a friend

Discuss this message

Book announced at

EDITOR: Federico Corriente
EDITOR: Christophe Pereira
EDITOR: Ángeles Vicente
TITLE: Encyclopédie linguistique d’Al-Andalus Vol. 2
SUBTITLE: Dictionnaire du faisceau dialectal arabe andalou
PUBLISHER: De Gruyter Mouton
YEAR: 2017

REVIEWER: David Wilmsen, American University of Sharjah

REVIEWS EDITOR: Helen Aristar-Dry


Reviewing a dictionary is different from reviewing other scholarly productions, because there is no underlying theme or argument to assess. There is, however, a huge amount of material to absorb and digest, such that a review could never do justice to all the information that a dictionary can provide.

In assigning dictionary reviews to my translation students, I alert them to the hallmarks of a good bilingual dictionary: front matter explaining the symbols used in the entries; a thoroughgoing introduction, guiding readers in the use of the dictionary; cross-referencing; and explications of usage in the entries themselves. Other information in the form of charts, maps, gazetteers, indexes of place or biographical names are always welcome, but a good dictionary must display at least the four features that I ask my students to address. Where Arabic dictionaries are concerned, cross-referencing means both references within the entries, but also a reverse glossary (at least). It is a peculiarity of bilingual Arabic dictionaries that they tend to be unidirectional, either Arabic to a target language or the opposite.

The dictionary under review, ‘Encyclopédie linguistique d’Al-Andalus Vol. 2,’ falls short in all four criteria. The last two specifications are not absolutely essential. It being a dictionary of an extinct variety of Arabic, its users will likely not need to acquire active command of the subtleties of usage in the lexemes within it, but will only need their meanings explicated as an aid to reading the texts from which they were culled. Nor are its users likely to trouble themselves over how exactly to translate a particular word or concept from their own languages into Andalusi Arabic. Nevertheless, the entries themselves do include reasonable but curtailed examples of usage and some cross–referencing, and a reverse glossary of sorts does close the book as an Index des termes par langue (pp. 1389–1495). That itself is a curious assortment, which we shall address presently.

The Dictionnaire is a compendious work, encompassing 1495 pages. Another twelve pages of front matter bring the tome to more than 1500 pages. Five of the pages in the front matter are taken up by titles, bibliographic details, and the TOC. Another blank page between that and the list of bibliographic acronyms leaves a scant 6 pages for the type of explanatory material that characterizes good reference works.

From there, the bulk of the Dictionnaire comprises a series of headings, listing words by their beginning consonant, according to the arrangement of the Arabic alphabet. A section begins with the Arabic letter in parentheses at the top of the page, followed by its name transliterated in Latin letters. So for example, the opening section is “(أ) Alif”. The second (p. 92) begins with “(پ و ب) Bāʔ et Pāʔ”, etc. The inclusion of the letter (پ) pāʔ, by the way, early on marks this as not an ordinary Arabic dictionary, inasmuch as Arabic does not possess the sound [p] except in borrowings, most of those not included in dictionaries unless they have been assimilated into the language, whereupon the consonant would be pronounced [b], for example, būr saʕīd ‘Port Said.’ A dictionary of Andalusi Arabic must include that letter, because of the many borrowings from Romance in the texts from which the entries are drawn.

The entries also follow the listing of the words in alphabetical order by tri-consonantal root in the manner typical of Arabic dictionaries, such that, for example, those wishing to look up the name Muħammad would search for it under the first letter in the root of the word {ħ} from the root {√ħ m d}, not {m}. The entries are arranged with the transliterated consonants of the root between curly brackets preceded by an asterisk and followed by the word in Arabic typeface between parentheses, as such (p. 442): *{ḪWF}(ﺧﻮف). This is followed by an abbreviation in upper case, referring to the primary or secondary source (generally an edited volume of a primary source) where an example of usage appears. Following that is a transliteration of various derived forms of the root. For example, IQ > ḫifta niḫāf ḫawf = tatḫawwaf <, GL > aḫāfu ḫawfun ḫāyifun /. The abbreviation with which this entry begins, IQ, refers to the 12th century Andalusian poet Ibn Quzmān, the list following > gives usages of that poet. For its part, GL refers to a work by Corriente referred to on the acronym pages as Corriente 1991a, which the bibliography section (pp. 1376–1387) reveals is another lexicographic work El léxico árabe estándar y andalusí del ‘Glosario de Leiden’. The transliterations indicate an example of the verb conjugated (in the 1st person indicative), the verbal noun, and the active participle. So, the entry for *{ḪWF} looks like this:

*{ḪWF} (ﺧﻮف)
IQ > ḫifta niḫāf ḫawf = tatḫawwaf <, GL > aḫāfu ḫawfun ḫāyifun / maḫūfun< + ID ʔym 2 >

After four lines of the usages of this root found in various works in or about Andalusi Arabic, the definition is given: craindre. With the help of the front matter, we can read most of it:

“ḪWF Ibn Quzmān deviant [the meaning of >] you feared [ḫif-ta fear.pfv-2ms] ‘I fear’ [ni-ḫāf 1s-fear.ipfv] ‘fear’ [ḫawf n] résulte de [the meaning of <] Corriente 1991a deviant ‘I fear’ [ni-ḫāf 1s-fear.ipfv] ‘fear’ [ḫawf-un n-nom] ‘fearing’ [fear.ptc] variation phonologique ou morphologique [the meaning of /] ‘fearful’ [maḫūf-un fear.ptc-nom] jointure ouverte ; ajout de préfixe ou suffixe ; mot rime [the meaning of +] ID [Jimenez’s edition of Ibn Danān]

When we come to ʔym 2, however, the meaning becomes indecipherable. It happens that {√ʔ Y M} is an Arabic root expressing a different semantic domain (“parmi les deux choices” p. 70). How that is related to Ibn Danān’s Sefer or it can be jointure ouverte; ajout de préfixe ou suffixe; ou mot rime to *{ḪWF} is not clear. Here is where a more detailed explanation of notation and cross-referencing would have been helpful.

The entire entry takes 12 lines, and is followed by and eight-line entry under *{ḪWL} (ﺧﻮل), referring to the oncle maternel and tante maternelle, their offspring, and a few other assorted meanings.

This proceeds for 1376 pages until finally ending in the Bibliographie, itself followed by the Index des termes par langue.


This is Volume 2 of the Encyclopédie linguistique d’Al-Andalus, Volume 1 having appeared in 2015. Entitled Aperçu grammatical du faisceau dialectal arabe andalou: Perspectives synchroniques, diachroniques et panchroniques, that volume covered the lexis, phonemics, morphology and syntax of the language, also providing some sample texts. It was, in that respect, a reiteration and revision in French of the 2013 Brill edition A Descriptive and Comparative Grammar of Andalusi Arabic, the authorship of which is accredited to the editorship of Institute of Islamic Studies of the University of Zaragoza, but a reading of the introduction of which will identify the author as Corriente. That volume is itself a return to Corriente’s pioneering 1977 work A Grammatical Sketch of the Spanish-Arabic dialect bundle. For its part, Volume 2, the present work, looks to be a revision of Corriente’s 1997 A Dictionary of Andalusi Arabic. As far as I know, it was Corriente who first coined the term Andalusi Arabic, just as he was the first to use the term “dialect bundle” in Arabic dialectology. Hence the term faisceau in the titles of the first two volumes in the Encyclopédie. According to the publisher’s web page, a third volume is planned, without that term in the title. That volume, Dictionnaire des emprunts ibéro-romans: Emprunts à l’arabe et aux langues du Monde Islamique, looks to be a revisiting of Corriente’s 2008 Brill release Dictionary of Arabic and allied loanwords: Spanish, Portuguese, Catalan, Galician and kindred dialects.

Two questions then arise: Why these revisitations, and what is new in all of these, especially in the volume under review? The compilers give an immediate answer to the first question in an otherwise spare two-page Avant-propos:

Il faut avouer qu’aucun dictionnaire dont les matériaux ne cessent de s’accroître ou d’être mieux compris ne peut se passer, après quelques années, d’une révision de son contenu, ainsi que de ses objectifs et de sa méthodologie. C’est pour ces raisons que nous avons décidé d’élaborer une nouvelle édition d’un dictionnaire d’arabe andalou (p. iii)
It should be the duty of the authors to answer the second question, as well. Because they do not, it falls to the reviewer to do so. What is new? Very little except the language of the text itself. The authors do explain why they chose to write in French, the native language of Pereira, rather than, say, Spanish, the native tongue of Corriente and his student Vicente, or English, the language of all the previous volumes upon which the Encyclopédie is meant as une révision de leur contenu, ainsi que de leurs objectifs et de leur méthodologie:

Nous avons choisi le français pour cet ouvrage, d’une parte en témoignage de reconnaissance du mérite immense des arabisants francophones au cours des derniers siècles, dont nous avons tous se bien profité, et d’autre part comme un appel, sur toute adressé à nous disciples à nos jeunes collègues de plusieurs pas, de la nécessité de ne pas s’astreindre à la connaissance et a l’utilisation d’une seule langue de culture.

This, too, is curious. It is by no means an idle appeal, but Corriente’s disciples, of which there are many – he having almost singlehandedly created the field of Andalusi Arabic studies out of only slightly (if skilfully) worked cloth – all are capable of presenting their contributions in French at international gatherings of Arabists. At least those who remain active in the field are. What is more, the arabisants francophones of the current generation are all capable of reading English and many of them write at least some of their contributions to the field in that language. Perhaps the appeal is to the Americans amongst their readership? Even then, the famously monolingual Americans, if they know any other foreign language at all well enough to engage in serious scholarship with it, it is Spanish. Much of Corriente’s prodigious output, especially his detailed examinations of the texts from which the entries from this dictionary are drawn is in Spanish. For example, of the sixty-four of his works cited in the bibliography, forty-five are in Spanish, as are his seminal studies of the colloquial Arabic poetic genres of kharja and muwashshah and his arduous work with the twelfth-century zajal poet Ibn Quzmān. Regardless, serious scholars of Arabic, Americans and otherwise, are tutored in reading French and German.

Rather, an appeal should be made to Arabists to pay closer attention to the output of their hispanohablantes colleagues, that they may gain the deep insights into the history of the language of their study that resides in the extensive scholarship about Andalusi Arabic that they produce. For, as Corriente is oft heard to lament, outside of Spain (he would say inside, too), the Andalusi Arabic dialect bundle is a neglected variety of the language, even while it is the best documented of the older dialects of Arabic. That documentation comes in the colloquial writings of al-Andalus that survive to this day, and it is largely Corriente and his disciples who have brought them to scholarly light. This dictionary and his early lexicographic works are invaluable aids in parsing the sometimes obscure meanings of those works, composed, after all, for readers of a bygone era who were more familiar with the idiosyncrasies of the language of the texts than are later generations, removed in distance and time from the civilization that produced them. Even native speakers of Arabic find a reading of the texts challenging in places.

As to what else is new in the present work, it seems that the best way to assess that is to choose a small section, featuring words beginning with a letter with which few words in Arabic begin. As very few words in Arabic begin with the first letter ظ, that appears to be a good place to conduct a test run of the dictionary. In the Dictionnaire, that entry consumes all of four pages (836–839), comprising but thirteen entries (not including the name of the letter, with which the sections characteristically begin). By contrast, the section covering the first letter of the Arabic alphabet (أ) requires ninety-four pages to cover its 712 entries. The analogous section in A Dictionary of Andalusi Arabic encompasses 34 pages and 841 entries. Its section on (ظ) also has more entriBeyond Language Boundaries: Multimodal Use in Multimodal Contexts, Marta Fernández-Villanueva and Konstanze Jungbluth. Berlin: de Gruyter. 2016, 254 pp.
es, but not by such a large margin, incorporating fifteen entries in all of about 1 and 3/4 pages (1997: 340-341). The compilers give no clue as to why the entries have been reduced in the latest work. Despite their fewer entries, the sections themselves take up more pages because of the larger typeface in the Dictionnaire than that of the Dictionary, which, with its greater number of entries, encompasses but 623 pages.

Aside from that, it lists the roots in both IPA transliteration and their Arabic characters, which its earlier counterpart has not done. Although the Dictionnaire often provides fewer definitions and examples drawn from the original texts than does the Dictionary, it makes up for that with an index listing borrowings from other languages, including borrowings or cognates from Semitic languages, from Akkadian to Ugaritic, that correspond to the root in question. This by itself makes it a monumental work, surely the outcome of a lifetime of study.

Nevertheless, this index listing roots is confusing. It truly does provide hundreds of cognates with Semitic languages, but we are not told whether those are borrowings or shared common Semitic lexis. The Arabic dialects of the Gulf do appear to contain some considerable borrowings from Akkadian, especially agricultural terms (Holes 2016: 8 & 10–19), owing no doubt to the extensive Assyrian and Aramaic presence in the Gulf until Islamic times. The Andalusi Arabic dialect bundle, as an amalgam of northern and southern dialects of Arabic, may indeed have retained borrowings from Akkadian, but are these what we find in the seven pages of roughly 1,000 Akkadian roots? Not likely. Likewise the twenty pages of Aramaic, are all of these also borrowings? Again, it is unlikely. The same might be asked of the Hebrew entries. And what about the Ugaritic? Were Arabic speakers ever in contact with Ugaritic? They may have been, but we have no evidence of it, unless the Andalusi Arabic is an attestation of that. If it were, the news should be broadcast far and wide. But the writers of the Dictionnaire do not explain why Ugaritic words are to be found in Andalusi Arabic. On the other hand, only about 150 words of Berber origin are listed. Surely speakers of Andalusi Arabic were in contact with Berber languages from the beginning of the Arab/Berber conquest of the Iberian Peninsula? There was often friction between Andalusi Muslims of Arab or native Iberian origin and those of Berber origins. Perhaps Andalusi writers of Arabic eschewed the usage of recognizably Berber words? Our compilers do not enlighten us on the matter.

Meanwhile, the revision of the earlier works, the need for which the compilers admit to, is apparent from the first page: “*{ʔBBWR} (أﺒﺒﻮر) dans Corriente 1997a: 1 semble être une erreur, car la graphie du manuscrit pourrait être correcte et il faudrait donc respecter > ʔnbwrh < (q.v.)”

Not every misinterpretation has been revised, however. In the same section, pp. 46–47, the entry *{ʔŠ) (اش) is defined by its basic meaning quoi, which is what its reflexes mean in all living dialects in which its reflexes appear, variously aš, ayš, ēš, aššu, šū, etc., all meaning ‘what?’ Yet in part 2 of the entry, we find this: “IQ, IA [Marugán’s 1994 edition of Ibn ʕĀṣim’s proverbs], et ZǦ [Bencherifa’s 1971 edition of al-Zajjālī’s proverbs] … « ne » IQ > iš nirīd naḥlaf < « je ne veuz pas jurer », > iš danb al+ḥadīd < « c’est ne pas la faute de l’acier » (p. 47).

This is a reversal, not a revision. In Corriente’s 1977 Sketch, we find the origin of the interpretation of aš as a negator: “/iš/ (segregated from the pronunciation with strong imāla [vowel fronting and raising] of the interrogative /aš/) begins its semantic transfer … being a clear negative” (1977: 145). Yet, the 2013 A Descriptive and Comparative Grammar of Andalusi Arabic, of which Corriente is the author, states, “in some cases, the interrogative rendering would be possible … ‘what is the fault of the iron?” (2013: 127). This is no doubt in response to Wilmsen (2014: 72–86), in which it is shown that in the reading of orthographic {aš} (however it was pronounced) in the two examples cited in the Dictionnaire and others adduced in the 1997 Dictionary (1997: 17) – where it is defined “iš not”, remains an interrogative. (Wilmsen and Corriente had corresponded about the matter in 2012.)

How to explain the reversal? It seems that the current Dictionnaire is, indeed, largely a reworking in French of the 1997 Dictionary, and the compilers simply undertook the arduous task of adapting that compendious work without considering others.

With that, it remains only to inquire after those who might benefit from consulting this work. The answer is first and foremost that they would be students of Andalusi Arabic and its texts who prefer that their first recourse be to reference works in French. Those scholars will be unable to avoid referring to Corriente’s works in English and Spanish, and, of course, to the works in Andalusi Arabic in all of its manifestations to which those invaluable references are a guide. It should be useful to Arabic historical dialectologists, too, if they were to take the time to consider the implications that Andalusi Arabic holds for the history of the Arabic dialects. Some, of course, have. Others should, and this dictionary will be an aid to them once they begin.

A small complaint in that regard: The Sketch more-or-less consistently refers readers to sections of the texts from which its examples are drawn; the Dictionary less so; and the Dictionnaire sporadically at best. This will oblige scholars to consult all the works in question. But they must, anyway.


Corriente, Federico. 1977. A Grammatical Sketch of the Spanish Arabic Dialect Bundle. Madrid: Instituto Hispano-Árabe de Cultura.

Corriente, Federico. 1991. El léxico árabe estándar y Andalusi del “Glosario de Leiden.” Madrid: Departamento de Estudios Árabes e Islámicos.

Corriente, Federico. 1997. A Dictionary of Andalusi Arabic. Leiden: Brill.

Corriente, Federico. 2008. Dictionary of Arabic and allied loanwords: Spanish, Portuguese, Catalan, Galician and kindred dialects. Leiden: Brill.

Corriente, Federico, Christophe Pereira, Ángeles Vicente (eds.). 2015. Aperçu grammatical du faisceau dialectal arabe andalou: Perspectives synchroniques, diachroniques et panchroniques. Berlin/Boston : de Gruyter

Corriente, Federico, Christophe Pereira, Ángeles Vicente (eds.). Forthcoming. Dictionnaire des emprunts ibéro-romans: Emprunts à l’arabe et aux langues du Monde Islamique. Berlin/Boston : de Gruyter

Institute of Islamic Studies of the University of Zaragoza. 2013. A Descriptive and Comparative Grammar of Andalusi Arabic. Leiden: Brill.

Jimenez Sánchez, M. 1996. Sefer ha-šarašim, par Sĕʕādyah ibn Danān. Granada: Universidad de Granada

Marugán, Marina. 1994. El refranero andalusí de Ibn ʕĀṣim al-Garnāṭī. Madrid: Hiperión.

Wilmsen, David. 2014. Arabic Indefinites, Interrogatives, and Negators: A linguistic history of western dialects. Oxford: Oxford University Press.

al-Zajjālī, Abū Yaḥyā ʕUbaid al-Dīn Aḥmad. 1971. amtāl al-ʕawām fī al-andalus (Proverbs of the Common People in al-Andalus), Muhammad Bencherifa (ed.). Rabat: Ministry of Culture.


David Wilmsen is head of the Arabic and Translation Studies Department at the American University of Sharjah in the United Arab Emirates. He has spent twenty-five years in the Arabophone world, previously at the American University of Beirut and the American University in Cairo. He is interested in the history and prehistory of the Arabic dialects and the development of their various syntactic features, especially object markers, interrogatives, negators, and existential particles. He is currently involved in documenting the Levantine Arabic features in the peripheral Arabic variety Maltese and in researching manifestations of the Croft’s negative existential cycle in Arabic.

Page Updated: 05-Feb-2018