Gallego (M. A.)_The languages of medieval Iberia and their religious dimension.pdf

© Koninklijke Brill NV, Leiden, 2003 Medieval Encounters 9,1Also available online – www.brill.nl

THE LANGUAGES OF MEDIEVAL IBERIA AND THEIR RELIGIOUS DIMENSION1

MARÍA ANGELES GALLEGO

Madrid

ABSTRACT

The aim of this article is to investigate language use as a social phenomenon in theperiod of eight centuries during which there was a Muslim state in the Iberian penin-sula—that is, more exactly, the period comprising 711 C.E. to 1492 C.E., which arethe dates of the Muslim arrival and conquest of the peninsula, and the end of thereconquest of the Muslim territory by the Christians, respectively. For this purpose,I will look into the contexts and registers in which the diVerent languages ofMedieval Iberia were used, as well as references to linguistic usages that the writtensources of this period contain. Religious aYliation will prove to be a decisive factorin the use of diVerent linguistic varieties and, moreover, the values ascribed to eachof them.

I. Background

The study and description of the inter-relations between language andsociety has been the object of a relatively new discipline, sociolinguis-tics, which has come to fruition within the past � fty years. Within thisperiod our understanding of the social organization of language behaviorhas considerably increased. What’s more, sociolinguistic studies have shedlight upon a range of phenomena previously ignored by theoretical lin-guistics by including social categories (social class, gender, social network)as variables of linguistic change. Sociolinguistic research has also broughtto light other phenomena of linguistic culture as important as diglossia(see below 2.1.3). The achievements of sociolinguistic theorizing, however,have had a limited echo in the study of the linguistic situation of theIberian peninsula in the Middle Ages.

A major hurdle in the way of incorporating the implications of the-oretical sociolinguistics in the case of medieval Iberia is the lack of data

1 The second part of this article relies heavily on the lecture I gave at the Centre ofMiddle Eastern Studies of the University of Cambridge (UK): “Language and Societyin Medieval Muslim Spain,” on 8 March 2001.

108 maría angeles gallego

collected from everyday speech and, more speci� cally, the lack of infor-mation on phonological variation. Phonological changes, representingprocesses of unconscious language change, have been established as amajor source of sociolinguistic information. Since the pioneer study car-ried out W. Labov2 on the English of New York, numerous studies onphonological variation have shown the close links existing between socialclass, gender, social network, and ethnicity (among other social vari-ables), and language use and variation.3 For the description of the Iberianlinguistic situation one obviously lacks this sort of data and, therefore,the variationist paradigm as developed by Labov cannot be applied.4

The understanding and description of language use achieved in varia-tionist studies, however, will be applied in the present study. In spiteof the general lack of data on naturalistic speech, there are realms oflanguage behavior relatively well documented and useful as a source ofevidence for a general description of language use in medieval Iberia.The sources and facts on which I base my study can be summarizedas follows:

anecdotes and references to linguistic usages preserved in medieval pri-mary sources;

names employed to refer to the diVerent linguistic varieties;literary evidence of the use of languages or, in other words, evidence

of what is written in what language; andhistorical circumstances, similar historical circumstances have given rise to

similar linguistic solutions. I will therefore use historical data to makeinferences about the kinds of language use in diVerent social contexts.

II. Language use in Christian Spain

1. Latin and Romance

When the � rst Muslim troops arrived in the Iberian peninsula in 711C.E., the linguistic situation they encountered was similar to that obtain-

2 See W. Labov, The Social Strati�cation of English in New York City (Washington, D.C.:Center for Applied Linguistics, 1966), and his later work Sociolinguistic patterns (Philadelphia:University of Philadelphia Press, 1972).

3 See, for instance, P. Trudgill, The Social DiVerenciation of English in Norwich (Cambridge:Cambridge University Press, 1974). The pressure of social networks and its eVects onphonological variation is the core of Milroys’ paper: L. Milroy and J. Milroy, “Socialnetwork and social class: toward an integrated sociolinguistic model,” Language in Society,21 (1992), 1-26.

4 See Labov, “Social Strati� cation of English.”

languages of iberia 109

ing in other parts of the former Roman Empire: the native populationspoke an incipient linguistic variety developed from Latin, known as“Romance,” although they continued to use Latin as their written lan-guage. Very little is known about the Romance of this period, apartfrom what can be deduced from some contemporary Latin writings inwhich the in� uence of the spoken language is clearly attested. This scantevidence, along with the reconstruction from later Romance texts, showsthat there already existed trends to regional or diatopic linguisticdiversi� cation.5 These linguistic centrifugal forces were the outcome ofthe deterioration of the Roman communication and administrative infra-structure that led to the isolation of the diVerent provinces, along withthe decay of the classical educational system.

The written Latin of this period mostly belonged to the categoryknown as late or “low” Latin. It diVered from classical Latin in severalaspects, notably the re� ection of an analytical rather than a syntheticlinguistic type (linear order of the elements of the sentence, loss ofdeclension, increased use of prepositions, etc.) and some phonetic changesas a result of the in� uence of the spoken language.6

The picture we get from the linguistic culture of this period is thatof a diglossic situation. But � rst it is necessary to clarify my use of theterm diglossia, since it will frequently be used throughout this study.The sociolinguistic condition known as “diglossia” was originally for-mulated in 1959 by Charles Ferguson7 in the following way: “Diglossiais a relatively stable language situation in which, in addition to the pri-mary dialects of the language (which may include a standard or regionalstandards), there is a very divergent, highly codi� ed (often grammati-cally more complex) superposed variety, the vehicle of a large andrespected body of written literature, either of an earlier period or inanother speech community, which is learned largely by formal educa-tion and is used for most written and formal spoken purposes but isnot used by any sector of the community for ordinary conversation.”8

The language of formal contexts is labeled H(igh) variety, whereas thecolloquial spoken language comes to be the L(ow) variety.

5 See R. Lapesa, Historia de la lengua española 5th ed. (Madrid: Editorial Gredos, 1986),par. 30 (123-28); R. Menéndez Pidal, Orígenes del Español: Estado lingüístico de la PenínsulaIbérica hasta el siglo XI (Madrid: Casa Editorial Hernando, 1926); and � nally R. Wright, LateLatin and Early Romance in Spain and Carolingian France (Liverpool: Francis Cairns, 1982).

6 Lapesa, Historia, par. 15-18 (68-81).7 C. A. Ferguson, “Diglossia,” Word, 15 (1959), 325-40.8 Ferguson, “Diglossia,” 336.

9 J. A. Fishman, “Bilingualism with and without diglossia; diglossia with and withoutbilingualism,” Journal of Social Issues, 23, no. 2 (1967), 29-38; revised and reprinted as“Societal Bilingualism: stable and transitional,” in Sociolinguistics: A brief introduction (Rowley,Mass.: Newbury House, 1970), 78-89.

10 C. A. Ferguson, “Epilogue: Diglossia Revisited,” Southwest Journal of Linguistics, 10(1991), 214-34.

11 H. Kassis, “The Arabization and Islamization of the Christians of al-Andalus:Evidence of their Scriptures” in Languages of Power in Islamic Spain, ed. R. Brann (Bethesda,Md.: CDL Press, 1997), 136-55 (138-39).


In his characterization of this linguistic situation, Ferguson depictedfour clear-cut examples: classical Arabic/colloquial Arabic, German/SwissGerman, French/Haitian Creole and literary Greek/modern Greek.Further research on diglossic situations carried out by Fishman9 haswidened the original scope of diglossia, by incorporating cases of genet-ically unrelated varieties. In Fishman’s “extended” de� nition, the keycharacteristic of diglossia is the social compartmentalization of functionand language. Despite Ferguson’s objections to the extended diglossia,10

I will use this concept (rather than “narrow” diglossia) in my descrip-tion of language use in medieval Iberia, in keeping with most researcherswithin sociolinguistics.

The use of Latin and Romance at the time of the Muslim conquest—and specially in the following centuries—corresponds to the diglossiccategorization. The domains of Latin were those of the High variety:education, religion, and literature, whereas Romance was used in informalcontexts and as an everyday spoken language. Another “diglossic” char-acteristic is that Latin was highly standardized whereas Romance lackeda norm of usage. Finally, Latin was highly valued within the linguisticcommunity, while Romance was considered “low” or “vulgar” speech.

The Muslim conquest brought about dramatic changes in the lin-guistic evolution of the Iberian Peninsula (see below p. 119 V.), but theRomance/Latin diglossic framework persisted until roughly the middleof the thirteenth century. In their relationships with Christians, Muslimsrecognized the current cultural scheme. It is worthy of mention thatthe Latin inscription which appeared in the � rst coins issued in the penin-sula by Muslims was the fundamental statement of Islam: “In the Nameof God, there is no god but God, alone, without compeer.” The pri-mary intention of these coins, as H. Kassis points out, was to informthe subjugated Christians of the contents of Islam.11 These Latin inscrip-tions show the acknowledgment by Muslims of the linguistic barrier thatseparated them from Christians at this early stage and their awarenessof the role of Latin as the language of formal contexts, including reli-


12 See J. Sánchez Belda, ed., Chronica Adefonsi Imperatoris (Madrid: Escuela de EstudiosMedievales, CSIC, 1950), 33, 79, 86, and Wright, Late Latin and Early Romance, 227-28.

13 See R. Wright, “Latin and Romance in the Castilian Chancery (1180-1230),” Bulletinof Hispanic Studies, 73 (1995), 115-28, and F. J. Hernández, “Sobre los orígenes del españolescrito,” Voz y Letra 10, no. 2 (1999), 133-66.

14 F. González Ollé, “El largo camino hacia la oficialidad del español en España,”in M. Seco and G. Salvador, eds. La lengua española, hoy (Madrid: Fundación Juan March,1995), 37-62 (38).

15 See R. I. Burns and P. E. Chevedden. Negotiating Cultures: Bilingual Surrender Treatiesin Muslim-Crusader Spain (Leiden: Brill, 1999).

gion. In later years, Islamic coins were inscribed in Latin and classicalArabic and, � nally, from 721 C.E. onwards only Arabic was used.

A more positive attitude among medieval scholars towards CastilianRomance is re� ected for the � rst time in the Chronica AdefonsiImperatoris, a medieval chronicle of c. 1150. Romance terms are intro-duced by the author with the sentence “as we say/as it is said in ourlanguage (“nostra lingua”)” rather than “vulgar speech” or vulgar lan-guage” as Romance used to be designated.12

Early in the thirteenth century, Romance varieties started to takeover certain domains of the High language, mainly in the area of lit-erary production. Poetic and narrative works were produced in Castilian,Galician-Portuguese, and Catalan varieties. SuYce it to mention theCastilian epic poem Cantar de mío Cid (pre-existing in an oral form sincethe middle of the twelfth century), the Galician-Portuguese lyric worksknown as Cantigas de amigo, and the Catalan Chronicle of James theConqueror or Llibre dels Feyts. In this period, Castilian started to sharewith Latin the juridical and diplomatic domains (pertaining to the Hlanguage) in the chancellery of the Kingdom of Castile and Leon.Although there was certain hesitation about its use until a later period.13

When King Ferdinand III of Castile conquered Córdoba (1236), forinstance, a juridical corpus to rule the city was prepared in CastilianRomance on the spot. However, as soon as the king returned to Toledo,a Latin extended version of the same legal dispositions was preparedby the royal chancillery, on the model of the Fuero Juzgo.14 In thetreaties between Christians and Muslims the languages used were nor-mally classical Arabic and Latin, as required by the formality of thedocument. This is the case of the surrender treaty between James theConqueror and the vanquished city of Játiva in 1244, written in Arabicand Latin. Only one year later, however, another treaty signed by thesame monarch and the Muslim rebel al-Azraq has (Aragonese in� uenced)Castilian as the language representing the Christian party.15


The Christian conquest (“reconquest”) of Muslim territories had reachedone of its � nal peaks by the middle of the thirteenth century and, aseach Christian kingdom made progress in this crusader enterprise, itsvariety of Romance spread in the conquered lands. This is the explanationfor the linguistic map of Spain, in which the dialectological variants runvertically from north (original Christian redoubt) to south (Muslim Spainor al-Andalus). Another linguistic fact of major importance is the preva-lence of Castilian over other Romance varieties, as an outcome of itsmilitary success and of its leading role in the campaigns against MuslimSpain. Had it not been for this historical circumstance, it is very unlikelythat Castilian would have become the most important Romance vari-ety and later the oYcial language of Spain. Other Romance varieties,especially Leonese and Aragonese, were closer to Latin and enjoyed ahigher prestige. Castilian, on the other hand, originated in the moun-tains of Cantabria, near a Basque speaking area, in a poorly Romanizedterritory of the Iberian peninsula. This inhospitable area was stronglyforti� ed (hence its name “Castilla” from “castillo” or castle). It has beenargued that these peculiar circumstances explain Castilian’s innovativefeatures and highly diVerentiating characteristics with respect to theother Romance dialects of the peninsula.

The de� nitive impulse to Castilian came from the scholarly activitiespromoted and championed by king Alfonso X the Learned (reigned1252-1284). In what is known as the “School of Translators” of Toledo,all kinds of literary works were translated from Arabic into Castilianand from Castilian into Latin. The Jewish reluctance to use Latin, alanguage closely associated with Christianity, played a signi� cant rolein the process of normalization of the use of Castilian: the usual prac-tice was that Jews translated Arabic texts into Castilian and Christianstranslated from Castilian into Latin. But many times the Castilian ver-sion remained, sometimes revised and corrected by a Christian.

There was a proli� c literary production in Castilian during this period,including not just belles lettres but historical, juridical, and scienti� c worksas well. These literary activities inevitably led to a process of linguisticnormalization or leveling, abandoning certain forms and favoring others.King Alfonso X placed special emphasis on the correctness of linguisticusage and encouraged writing in “straight Castilian” (“castellano drecho”)by giving example with his own writings and employing correctors(“emendadores”) of the language. The culmination of the standardiza-tion process was the publication in 1492 of the Gramática of Castilianby Antonio de Nebrija.


16 King Ferdinand of Aragon (Ferdinand the Catholic, reigned 1479-1516), for instance,employed Castilian to correspond with Muslim Andalusi rulers and with other Christiankings within the peninsula.

17 This is specially the case in Leonese and Aragonese speaking regions, whereas theGalician-Portuguese and Catalan varieties showed a stronger entity and participated inthe domains of written literature.

18 These documents were edited by A. González Palencia, Los mozárabes de Toledo enlos siglos XII y XIII, 4 vols. (Madrid: Instituto de Don Juan, 1926-30). I. Ferrando car-ried out a linguistic survey of this material in El dialecto andalusí de la Marca Media: losdocumentos mozárabes toledanos (Zaragoza: Universidad de Zaragoza, 1995).

In this way, Castilian became a language of culture at an earlierstage than any other Romance language in the Iberian Peninsula. What’smore, it was used as lingua franca and a language of culture in therest of Christian Spain, extending its in� uence to Galician-Portuguese,Basque, Aragonese, and Catalan speaking areas.16 But the eZorescenceof Castilian as a language of culture and diplomacy from the thirteenthcentury onwards did not completely put an end to the previousRomance/Latin diglossia: the realms of academia and religion remainedfaithful to Latin until a much later period (seventeenth and early twen-tieth century, respectively).

The predominance of Castilian in Christian Spain was not imposedby any planning policy (contrary to what happened in the eighteenthcentury). It was rather the outcome of the cultural and historical cir-cumstances. Castilian, in any case, was not the only language used inliterature: Galician was preferred for lyric compositions, whereas liter-ary works were produced in the other languages, although to a lesserscale than in Castilian. In those parts of the peninsula where a lan-guage other than Castilian was used, the functions of the High varietywere normally shared by Latin and Castilian,17 whereas the domains ofthe Low language were � lled by the respective local languages.

2. Christian Arabic

The extent of Arabization among Christians of Muslim Spain (knownas Mozarabs) has been the object of a long controversy (see below p. 119). That in certain parts of the Islamic territory Christians had Arabicas their written and spoken language by the eleventh century is attestedin the Arabic documentary evidence produced by the Mozarab com-munity of Toledo. The documents preserved consist of over 1200 legaldeeds,18 mostly dated to the twelfth and thirteenth centuries, that is, tothe two centuries that followed the reconquest of Toledo by king Alfonso


VI (1085). This is a singular case of maintenance of the Arabic languageby Christians under Christian rule. In Ferrando’s comprehensive analysisof this linguistic situation,19 three main reasons are given for the use ofArabic as a cultural vehicle of Christian society. In the � rst place, therewas no Christian language that could clearly compete with Arabic inful� lling the functions of a High variety. Latin, as signaled above, waslosing ground vis-à-vis Romance varieties, but no Romance variety couldclaim to be a language of culture as yet. A second factor that was deci-sive in retaining the use of Arabic was the arrival by the middle of thetwelfth century of a migratory � ux of Arabic speaking Christians � eeingfrom the intolerant North African Muslim rulers of al-Andalus. Finally,the wish of the Mozarabs to “mark a clear diVerence between themand other Christian groups” played a part in keeping this linguistic sta-tus quo.20

It was by the end of the thirteenth century when all three factorsfaded away: the emergence of Castilian as a language of culture andadministration, no signi� cant Christian migrations from the south, andthe integration of this minority group in the wider Romance speakingChristian community.

3. Muslim language use

As the Christian Kingdoms pushed southwards from the eleventh centuryonwards, major Arabic-speaking Muslim communities fell under Christianrule. These Muslim groups were known as Mudejars (or Moriscos whenconverted after 1492). The capitulation agreements between Muslimcommunities and Christian rulers generally allowed them to keep theirjuridical and religious organization, along with their language—in a sim-ilar situation as the Christian communities lived under Muslim rule. How-ever, the increasing pressure exerted by Christian authorities to ensurethe Muslims’ acculturation had an inevitable impact in their social andcultural network. These pressures evoked diVerent responses amongMuslim communities: on the one hand, linguistic assimilation to thewider Romance speaking community, as it was the case with the majorityof Muslim groups in the Peninsula; on the other, Arabic maintenanceas a means to preserve Muslim identity, as in the case of the Muslims

19 I. Ferrando, “The Arabic language among the Mozarabs of Toledo.” In J. Owens(ed), Contributions to the Sociology of Language: Arabic as a Minority Language (Berlin: Mouton,2000), 45-63.

20 Ferrando, “Arabic language,” 48.

of Valencia.21 These two diVerent patterns of language behavior are theexpected responses of an ethnolinguistic community under pressure. AsM. Brenzinger puts it: “Pressure on ethnolinguistic communities fromoutside may evoke language maintenance activities and resistance. Butit also may undermine self-perception, which can then result in changesin language use patterns. This is when the “downward spiral of reducedlanguage use and loss” may start its deadly circle.”22

But the eVort to preserve their Muslim identity is not only manifestin those communities that kept Arabic as a spoken and written lan-guage. Romance speaking Muslims and Moriscos in Castile and Aragondeveloped a kind of “Islamic” literature that tries to reconcile the useof Romance with their Muslim identity. This Islamic Romance litera-ture is known as Aljamiado.

For the correct understanding of Aljamiado, we need to bear in mindthe status of Arabic as the sacred language of Islam (see below p. 122)which has had a profound eVect in the way in which Muslims use Arabicand languages other than Arabic. This is a common phenomenon amongthose religious communities that believe in the sacrality of one particularlanguage: its “sacred” status has, � rst of all, an impact on the language´sdevelopment, since its speakers will show a highly conservative attitudetowards it (consider, for instance, the case of classical Arabic, the languageof the Koran, vis-à-vis dialectal or neo-Arabic). Furthermore, consideringone language “sacred” conditions the use of other languages by themembers of that religious group. The two main characteristics of theimpact of the sacred language on the “profane” one are the exporta-tion of its alphabet and certain vocabulary—mainly of the religiousdomain. If the members of the religious group in question are linkedby strong social ties (for instance marry and work within their own com-munity), reluctance to language innovation is bound to occur and there-fore “archaisms” will appear in diVerent linguistic realms.23 Finally, the

21 On the question of the Arabization of Muslims and Christians in Valencia, as wellas other issues regarding the use of languages in Muslim Spain, see C. Barceló (1984),Minorías islámicas en el País Valenciano. Historia y dialecto (Valencia: Universidad de Valencia,1984), and R. I. Burns, Muslims, Christians and Jews in the Crusader Kingdom of Valencia(Cambridge: Cambridge University Press, 1984), 172-192.

22 M. Brenzinger, “Language Contact and Language Displacement,” in The Handbookof Sociolinguistics, ed. F. Coulmas (Oxford: Blackwell, 1998), 273-84 (283).

23 As demonstrated in the sociolinguistic surveys carried out by James and LesleyMilroy, linguistic innovations are transmitted through weak social networks. When themembers of a social group are linked by strong social ties, there is a stronger resistanceto linguistic changes. See Milroy, “Linguistic change,” and L. Milroy, Language and SocialNetworks, 2nd ed. (Oxford: Basil Blackwell, 1987).



high appreciation and exceptional values ascribed to a sacred languagecan generate carelessness and lack of interest in the correct usage ofother languages on the part of the members of a religious community.

These characteristics � t the case of Islamic literature in Romance orAljamiado.24 Aljamiado writings are composed by and addressed toMuslims (or Moriscos). Its main diVerentiating features from the Romanceliterature of the rest of the population are the use of Arabic script, theinclusion of Arabic vocabulary (specially for religious concepts), and afreer re� ection of the spoken language than in the writings by Christianauthors. They show as well some “archaic” traits, mostly in the area oflexicon and phonetics.

The � rst preserved Aljamiado works are dated to the � fteenth cen-tury, notably the Breviario Sunni or Compendium of Islamic Law by Içede Gebir, composed in 1462. The expulsion of Muslims from Spain inthe seventeenth century put an end to Aljamiado literature since theRomance Morisco communities expelled from the Peninsula graduallyadopted the Arabic language of their North African homeland.

4. Jewish language use

The languages used by the Iberian Jews were basically the languages usedby the groups among which they lived, namely, Arabic and Romance.They also had Hebrew, the language of their own religious tradition(see below pp. 137-38), for liturgy and certain written spheres. Overall,Jews were characterized as a polylinguistic community, at least theirhigher strata.25 Their excellent knowledge of Arabic allowed them tohold important oYces in Christian Kingdoms, as secretaries and courtiers( Joseh Ferrizuel [“Cidellus”], Isaac ben Zadok [“Don Cag de la Maleha”],Sheshet Benveniste . . .).

24 See G. Wiegers, Islamic Literature in Spanish and Aljamiado: Yça of Segovia ( � . 1450),His Antecedents and Successors (Leiden: Brill, 1994), chs. 2, 3, 6; L. F. Bernabé-Pons, “Laasimilación cultural de los musulmanes de España: lengua y literatura de mudéjares ymoriscos,” in Chrétiens et musulmans à la Renaissance: Actes du 37e colloque international du CESR(Paris: Honoré Champion, 1998), 317-35; L. P. Harvey, “Aljamía,” in Encyclopaedia ofIslam (Leiden: Brill, 1960) 1: 404-5; O. Hegyi, “Language between Christianity and Islam:The case of Aljamiado Literature,” Scripta Mediterranea, 5 (1984), 29-38; A. Galmés deFuentes, “La lengua española de la literatura aljamiado-morisca como expresión de unaminoría religiosa,” Revista Española de Lingüística 16, no. 1 (1986), 21-38; M. J. Viguera,“Introducción,” in F. Corriente, Relatos píos y profanos del manuscrito aljamiado de Urrea deJalón (Zaragoza: Institución Fernando el Católico, 1990), 9-51.

25 The High varieties used by Jews in Medieval Iberia have been studied by E. R.Miller, Jewish Multiglossia: Hebrew, Arabic and Castilian in Medieval Spain (Newark, Del.: Juande la Cuesta, 2000).

They also had an active role as cultural intermediaries through theirtranslation activities. One the one hand, they were commissioned byChristian rulers to translate into Romance numerous Arabic works, asthose produced in the School of Translators of Toledo. The other maintranslation activity consisted of rendering into Hebrew the Arabic ( Judeo-Arabic, see below p. 139) writings of Jewish scholars living under Muslimrule. In this way, the impressive intellectual production of Arabic-speakingJews was transmitted to the rest of European communities that remainedinnocent of Arabic. One of the most highly reputed translators wasJudah ben Saul ibn Tibbon (c. 1120-c. 1190), the initiator of an activ-ity that would count with many followers among the members of hisfamily. After � eeing from Granada in the middle of the twelfth cen-tury, Ibn Tibbon got settled in the south of France, where he practicedas a physician and carried out his translation activities. In his famousEthical Will, addressed to his son Samuel ben Judah ibn Tibbon, heacknowledges the importance that knowing Arabic had for Jews: “youknow that the great men of our people attained their greatness andmany virtues only because of their ability in writing Arabic. You havealready seen what the Nagid, of blessed memory, said about the great-ness he achieved through it.”26

Judeo-Romance literature is another case of language use under theimpact of a sacred language, bearer of a speci� c religious tradition. Wedesignate Judeo-Romance literature those writings composed by Jews inRomance varieties,27 using Hebrew characters and in� uenced by thegrammar and lexicon of the Hebrew language.

There are Judeo-Romance texts dated to the pre-Expulsion period,including the Proverbios morales by Shem Tov de Carrión (fourteenth cen-tury) and the anonymous Poema de Yoçef (beginning of the � fteenth cen-tury). It is safer to label the language of these works “Judeo-Romance”rather than Judeo-Spanish since the existence of Judeo-Spanish beforethe Expulsion is still a controversial issue.28 What is clear is that Judeo-Romance literature of peninsular Jews did only clearly develop in exile,after 1492. It is after this date that all the Jewish Romance varieties

26 I. Abrahams, Hebrew Ethical Wills (Philadelphia: The Jewish publication Society ofAmerica, 1926), 57.

27 On peninsular Judeo-Romance varieties other than Castilian, see J. R. MagdalenaNom de Déu, “Judeorromances “marginales” de Sefarad,” Miscelánea de Estudios Árabes yHebraicos, 37-38 (1988-89), 41-53.

28 Defending the existence of a pre-expulsion Judeo-Spanish, see L. Minervini, Testi giudeos-pagnoli medievali, 2 vols., (Napoli: Liguori, 1992). The opposite view is held by R. Penny,“Judeo-Arabic Varieties Before and After the Expulsion,” Donaire, 6 (1996), 54-8.


29 See D. M. Bunis, “The Language of the Sephardim: A Historical Overview,” inMoreshet Sepharad: The Sephardi Legacy, ed. H. Bernart ( Jerusalem: Magnes, 1992), 2: 92-126; I. M. Hassán, “El español sefardí ( judeoespañol, ladino),” in La lengua española, hoy,ed. M. Seco and G. Salvador (Madrid: Fundación Juan March, 1995), 117-40; M. Lazar“Ladino,” Encyclopaedia Judaica. Jerusalém: Keter, 1972), 10: 1342-50.


came into contact as a result of Jewish groups from diVerent parts of theIberian peninsula coming together in the cities of the Ottoman Empire.Under these conditions of dialect contact and the subsequent process ofleveling, a new koiné was created that showed substantial diVerences withthe Romance varieties spoken in the peninsula. To this linguistic varietywe refer as Judeo-Spanish, Judezmo, Ladino or Sephardic Spanish.29

5. Conclusions

If we compare the sociolinguistic situation of the Iberian Peninsula atthe time of the Muslim conquest in 711 and that obtaining in 1492,after the surrender of the last Muslim state of the peninsula, the mainchanges could be summarized as follows.

The domains of Latin as a High language had been considerablyreduced by 1492 and taken over by Romance varieties, especially inthe area of literary production. Among the diVerent kinds of Romance,Castilian prevailed as a language of culture and administration. Thiswas the outcome of its early process of normalization and standardiza-tion, along with its political dominance as the language of the kingdomof Castile, leader of the reconquest campaigns against the Muslims.

During the period of eight centuries that there existed a Muslim statein the peninsula, Romance culture and society underwent a signi� cantin� uence of the more prestigious Arabic language and culture. Linguistic-ally, this in� uence resulted in numerous Arabic lexical borrowings andcertain grammatical in� uence on Romance varieties and even Basque.The � rst channel of transmission was arguably Arabized Christians whoemigrated from Muslim Spain to the Christian kingdoms. Languagecontact situations increased as the reconquest advanced and entire groupsof Arabic speaking population were absorbed within Romance Christiansociety. Arabic speaking Christians, Muslims and Jews gradually shiftedto the language of their environment, namely, Romance, with the excep-tion of Muslim groups in the eastern coast of the Peninsula, which keptArabic. The shift to Romance was complete in the case of Christians.Jews and Muslims, however, manifested a more complex language behav-ior: Hebrew and Arabic, the sacred languages of Judaism and Islam

respectively, occupied certain domains of the High language (the varietyhighly valued by the speakers.) Religious practice, as expected, was oneof them. Certain juridical and legal documents, related to social practicesregulated by religion (marriage, inheritance, etc.), were issued in theselanguages as well. In addition to that, Iberian Jews produced in Hebrewimportant literary works that included poetic, scienti� c and philosophicalworks. Finally, another speci� c linguistic behavior shared by Muslims andJews was the development at the end of the reconquest period (whenthat of “romancization” was stronger) of a variety of written Romance,for internal consumption, that incorporated elements of their respectivesacred languages and showed certain linguistic diVerences with respectto contemporary Romance writings by Christians.

III. Language use in Muslim Spain

The existence of a Muslim state in the peninsula (known as al-Andalus)had a profound eVect in the linguistic evolution of both Christian (seeabove p. 112) and Muslim Spain. As in other territories ruled by theemerging Islamic Empire, the Arabic language in al-Andalus acquireda central role as a language associated to the circles of power andadministration and as the sacred language of Islam. It has to be said,nevertheless, that the number of native Arabic speakers who took partin the invasion of 711 or arrived subsequently, as the Islamic rule cameto be established, was very limited. The army that crossed to the penin-sula was mostly composed of Berber males who must have spoken someform of Berber language and maybe some form of Latin.30 Despite thenumerical disadvantage of Arabic speakers vis-à-vis Romance speakers,Arabic gradually replaced Romance and Berber as spoken languages.A tendency to monolingualism with Arabic seems to have dominatedin Muslim Spain since the end of the eleventh century. Most scholarsagree on thorough Arabic monolingualism from the thirteenth centuryuntil the surrender of the last Muslim kingdom of Granada, and someclaim the complete extinction of the Romance language in Muslim Spainsince the end of the eleventh century.31

The extent of Arabization, the dates when it took place, and the pre-servation or disappearance of the Romance language in Muslim Spain

30 See D. Wasserstein, “The Language Situation in al-Andalus,” in Studies on theMuwashshah and the Kharja, ed. A. Jones and R. Hitchcock (Oxford: Oxford UniversityPress, 1991), 1-15 (4).

31 F. Corriente, Poesía dialectal árabe y romance en Alandalús (Madrid: Gredos, 1997), 338.



are linguistic issues controversial among scholars in the � eld.32 This lin-guistic discussion has taken place within a broader historiographical con-text. Continuous use of Romance language in Muslim Spain and alimited and late Arabization (alongside scarce Islamization) constitutesone of the core arguments of scholars claiming the continuity of onesingle Spain before, during, and after the Muslim invasion.33 The oppositelinguistic opinion, proposing the disappearance of the Romance languagearound the end of the eleventh century, along with a thorough linguisticand cultural Arabization of al-Andalus, is held within a substantiallydiVerent historiographical position. This linguistic evolution is contemplatedin a Muslim state and society whose political and social structures weresimilar to those existing in other parts of the Islamic world. MuslimSpain, according to this line of thought, represented a complete dis-ruption of social and cultural patterns of former Visigothic Spain.34

From a sociolinguistic point of view, one observes several shortcomingsin the way that evidence for the diVerent theories has been used. In the� rst place, there is insuYcient distinction between High language and LowLanguage in diglossic situations such as that of classical Arabic and col-loquial Arabic in Muslim Spain. Ignorance of this sociolinguistic situationmight lead to the wrong interpretation of facts. One of the argumentsgiven to demonstrate a merely super� cial Arabization of the populationof al-Andalus, is scarce knowledge of Arabic. “Evidence” for it is theabundant production of Arabic grammars and treatises for the correctuse of Arabic.35 The existence of these kinds of works only serves to

32 For a summary of the diVerent opinions, see O. Zwartjes, Love Songs from al-Andalus.History, Structure and Meaning of the Kharja (Leiden: Brill, 1997), 5-22; and more speci� callyabout the use of languages in Valencia, see Burns, Muslims, Christians and Jews, 172-92.

33 One of the clearest exponents of historio-linguistic attitude is F. J. Simonet who,in his study on the Romance dialect spoken in Muslim Spain, states for example: “TheArabs did not contribute to our civilization any substancial or formal element whose rel-evance can be measured in terms of bene� t and usefulness or by its long maintenance.As it happened with other barbar nations [. . .], settled in the middle of a highly civilisednation, they shone for some time on the remains of the vanquished”; F. J. Simonet,Glosario de voces ibéricas y latinas usadas entre los mozárabes precedido de un estudio sobre el dialectohispano-mozárabe (Madrid: Real Academia de la Historia, 1888), XLVI.

34 See P. Guichard, Structures sociales “orientales” et “occidentales” dans l’Espagne musulmane(Paris and The Hague: Mouton, 1977), and M. J. Viguera Molíns “Planteamientos sobreHistoria de al-Andalus,” in El saber en al-Andalus. Textos y Estudios, II, ed. J. M. Carabazaand A. T. M. Essaway (Sevilla: Universidad, Fundación El Monte, 1999), 121-32.

35 See, for instance, J. Vallvé, “Toponimia de España y Portugal. Las lenguas his-pánicas y su re� ejo en las fuentes árabes,” Boletín de la Real Academia de la Historia, 193,no. 2 (1996), 197-237: “It is reasonable to assume that the Arabic language in the Iberianpeninsula was impregnated by Latin and Romance languages to an incredible extent.This in� uence is re� ected in the phonetics and, specially, the Arabic vocabulary, and

corroborate, however, the Fergusonian characterization of a High varietyas a language with a “strong tradition of grammatical study”36 and alanguage that can only be acquired through an arti� cial learning process.In order to have a good command of classical Arabic, one has to learnlinguistic rules and structures that do not exist in the colloquial language.De� cient knowledge of the former or interference of the spoken varietyis the origin of deviations and mistakes in the use of classical Arabic. Theconcern for linguistic correctness in al-Andalus is similar to that existingin any modern Arabic country, whose speakers have an Arabic dialectas an everyday language and classical Arabic as a linguistic ideal.

Another characteristic of High is that speakers regard it as a languageof prestige and a superior language. The association of classical Arabicwith Islam, high culture, and power structures renders its knowledgehighly valued. It is therefore a prestigious thing to be mentioned in theaccount of somebody’s intellectual achievements that he was an expertor knew classical Arabic well. This is the case in the biographies ofsome Andalusi personalities. In the “traditionalist” school, however, thereading of evidence is substantially diVerent: the mention of good com-mand of Arabic in biographies points to the “rarity” of the event since,according to these scholars, knowledge of Arabic was very limited.37

Arguably, the most widely accepted conclusions with regard to theuse of languages in Muslim Spain are those posited by D. Wasserstein,38

belonging to the non-traditionalist position. In his analysis, Wassersteinuses the diglossia paradigm to explain the linguistic situation of MuslimSpain and its evolution. Notwithstanding, Wasserstein makes no dis-tinction between High and Low varieties.39

maybe the syntax [. . .] This special situation of Spanish Arabic led many grammariansto study the incorrections of the language” (213-14, 216).

36 Ferguson, “Diglossia,” 331.37 J. Ribera, “Épica andaluza romanceada,” in Disertaciones y Opúsculos I, (Madrid:

Maestre, 1928), 111-12: “When doing a careful reading of the biographies of SpanishMuslim ulemas, one observes an insistent and repetitive praise of the merit achieved bythose Spanish ulemas that were able to show their knowledge of the language of theArabs. It is clear that they had to learn it arti� cially, and the fact is that there were noschools until a later period. This would explain that in Andalusia there lived manyMuslims who could only speak Romance, in spite of the fact that they could read Arabicand translate it.” We observe here a clear inconsistency in Ribera’s analysis: he � rstargues little familiarity of Andalusis with classical Arabic (the language of the educatedclass [’ulamˆ"], the language that one learns arti� cially in the school), and then sur-prisingly concludes that this would account for the fact that they could not speak col-loquial Arabic (!).

38 Wasserstein, “Linguistic situation.”39 Wasserstein, “Linguistic situation,” 4: “when I use the word “Arabic,” I am not,



In what follows I will try to give a general picture of the sociolin-guistic evolution of Muslim Spain by, � rst, describing the two diglossicpairs that came into contact in the � rst period of Muslim presence inthe peninsula: classical Arabic/colloquial Arabic and Latin/Romance.For dealing with the development of these original diglossic pairs, I willfocus on language use in the diVerent religious communities: Muslims,Christians, and Jews.

1. Classical Arabic/Colloquial Arabic

The Arabs that came to the Iberian peninsula brought with them diVerentvarieties of spoken or dialectal Arabic, and one literary form, known asclassical Arabic. Classical Arabic evolved in the Arabian Peninsula as apoetic koine and super-tribal language in pre-Islamic times. But it waswith the rise of Islam that classical Arabic acquired its special status asa sacred language of all Muslims. Moreover, the Islamic conquests carriedwith them the use of colloquial and classical Arabic to the furthest endsof the Islamic empire.

Classical Arabic occupies a privileged position among Muslims as thelanguage of the Prophet Muúammad and, above all, as the languagethat God used to transmit His Message.40 The eloquence of the Arabicused in the Qur"ˆn (the written version of God’s Message) has been aliterary guide and linguistic authority for Arabic speaking Muslims fromthe Middle Ages until today. Its stylistic excellence is considered mirac-ulous within the Islamic community.

Linguistically, this variety of Arabic belongs to the synthetic linguis-tic type, contrary to the spoken or colloquial Arabic. The latter hadshown increasing analytical trends (loss of case endings and in� ectionsof the verb, � xed word order, etc.) since the pre-Islamic times. Thisincipient diglossia emerging in pre-Islamic times was accentuated anddeveloped in the Islamic period. Classical Arabic, in its “frozen” lin-guistic state, was the language of formal contexts including religion andliterature, whereas spoken Arabic � lled the functions of the Low lan-guage and its domains were those of informal contexts. As is typical ina diglossic situation, knowledge of classical Arabic had to be acquiredthrough formal tuition.

unless I actually say so, referring speci� cally to either spoken or written Arabic, but tothe two indiVerently. More signi� cantly, when I use the terms “Latin” and “Romance,”and also such expressions as “Late Latin” and “Spanish” or “proto-Spanish,” I do notmean to imply any signi� cant diVerences between the language forms behind them.”

40 See Qur"ˆn 16:103, 26:198, 41:44.

Classical Arabic was the language of administration and Islam inMuslim Spain, as in most of the Islamic world. Further, it was the vehi-cle of most of the intellectual production. The prestige imbedded inmastering Classical Arabic is evident in Andalusi society. It constitutesone of the qualities highlighted in the biographies of Andalusi person-alities.41 By the same token, lacking a good command of the literarylanguage, especially in the case of people who held important oYces,aroused severe criticism. This is the case of the Cordoban judge M�sˆibn Muúammad ibn Ziyˆd (tenth century C.E.) who is depicted by al-Khushan“ as an ignorant man for his lack of grammatical ability:

I heard from some ulemas that [the judge] M�sˆ ibn Ziyˆd was a manof good conduct and cultured, and that he had the outlook of a politeand respectful man. But the truth is that he was an ignorant and igno-minious man. It is told that in one occasion [. . .] this judge said thathe had fasted the whole month of Ramadan, including the day of al-’Arafa and then another day. [When he said that] he made two awfulmistakes: On the one hand, he thought that there was a day of ’Arafain the month of Ramadan as in the month of Dh� al-Hijjah and, on theother, he added the article al to the day of ’Arafa. I heard too that heused to put an alif at the end of the word “time” (marrah) and that heused to � nish the word “names” ("asmˆ") with hˆ" [rather than hamza]42

The pride in having a good knowledge of classical Arabic was a preva-lent notion in Muslim Andalusi scholarship. And a good way to proveone’s linguistic command was, no doubt, making a linguistic correctionto a real Arab. In al-Zubayd“’s biographies of Andalusi grammarians ofthe � rst Islamic period, he tells us meaningful anecdotes in this regard,including the argument between a native Andalusi and an Arab abouta grammatical question, in which the Andalusi ends up proving to havea deeper knowledge of the Classical language:

Muúammad ibn ’Umar informed me, learning it from more than oneperson that witnessed it, that Ab� Muúammad al-A’rˆb“ al-’Ëmir“ saidto ’Ibrah“m ibn Hajjˆj, thanking him for something that the latter didfor him: “By God, it was nothing but your right that led the Arabsto make you chief (sayyadatka)!. Ab� al-Kawthar al-Khawlˆn“ was pre-sent and said: “Ab� Muúammad, our experts in Arabic say “sawwa-datka.” Ab� Muúammad answered: “ ‘sawˆd ’ means ‘black,’ so they arewrong, they make a mistake.” Ibrah“m upbraided him saying: “Do you dare to challenge the Arabs in their language?” "Ab� al-Kawthar

41 See all the examples cited by Ribera, Disertaciones y Opúsculos, 111-12 n. 2.42 See J. Ribera, Historia de los jueces de Córdoba por Aljoxaní: Texto árabe y traducción española

por Julián Ribera (Madrid: Ibérica, 1914), 162 (Arabic text). My translation of the lastsentence diVers considerably from Ribera’s (p. 201).



wrote to Yaz“d ibn ÿalaúa [a reknowned Andalusi grammarian, ninthcentury C.E.], telling him about this matter and Yaz“d answered him:“what is known is “sawwadatka,” with waw. It might be that the formthat: Ab� Muúammad said is a variant used by the Banu ’Ëmir tribe.When Ab� al-Kawthar got the letter said: “Oh Ab� Muúammad, themaster has denied what you said” and told him what he had said. Al-A’rˆb“ then yelled and got agitated. Ibrah“m then called for Yaz“d andwhen Yaz“d arrived, he told him: “Do you dare to challenge this manin his language?” Ibn ÿalaúa answered: “Knowledge is not on the sideof the strife, but on the side of justice and truth. Let Ab� Muúammadanswer what I am going to ask him.” Ibrah“m said: “Ask!.” Yaz“dasked: “How do the Arabs say: “sˆda/yas�du” or “sˆda/yas“du,” and al-A’rˆb“ answered: “sˆda/yas�du.” Yaz“d said: “In this example, the wˆwis part of the verb. And how do the Arabs say: al-s�dad or al-s“dad?”andal-A’rˆb“ answered: “al-s�dad.” Yaz“d said: “This wˆw is part of thename.” He then asked: “What position has ’Umar ibn al-Kha ˆb—May God have Mercy upon him!—among you, with respect to purityof language?” Al-A’rˆb“ answered: “The highest.” Yaz“d said: “We givefor sure that he said: “Acquire knowledge before you are proclaimedmasters (tusawwad�),” and none of the experts in language doubts theveracity of this úad“th, contrary to other úad“ths that have proved tocontain errors.” Al-A’rˆb“ then persevered and said: “Oh people ofthe cities, what have you done with the language?43

A similar story, demonstrating the pride of local Andalusi scholars fortheir knowledge of classical Arabic, is found in al-Zubayd“’s biographyof ’Ufayr ibn Mas’�d (d. 919 C.E.):

When al-’Ajal“ came from Iraq, he withheld his books and was stingywith them. He summoned the people to come to him so that he coulddictate [his books] to them. People � ew to him, � ocked to his class,and al-Khushan“’s class was empty. ’Ufayr said: “al-Khushan“ told me:“How come you do not rush to where everyone else has rushed” andI said to him: “I do not seek a substitute for you.” He said: “I wantyou to go to that man and witness his class.” I went in the morningto al-’Ajal“ and was present when he dictated: “al-mirra means “enmity”and its plural is mirar.” One of the people who was copying down infront of him was Zayd al-Jayyˆn“. Then I said: “May God have Mercyupon you!, Ab� ’Ubayd stated in Al-Mu§annaf: “al-mi"ra [with hamza]means “enmity” and its plural is mi"ar.” It seemed to me that I sawthat Zayd had erased what he wrote. And I said: “This is the correctform.” Then I refuted him again about another word and yet a thirdone in that session and people left him and nobody went back again.

43 Al-Zubayd“, ÿabaqˆt al-naúwiyy“n wa-l-lughawiyy“n, ed. M. A. Ibrahim (Cairo, 1954)295-6. I am indebted to Dr. Manuela Marín for pointing me to the existence of thisanecdote.

The news reached al-Khushan“ and when I went to him, he cameclose to me and kissed me between the eyes, saying to me: “You cer-tainly are a depositor of knowledge!44

The cultivation of classical Arabic has its highest expression in the poeticart, which is considered the Arabic national art. Writing Arabic poetryentails a deep knowledge of the grammar and vocabulary of classicalArabic, along with a deep knowledge of the literary tradition of theArabs. Poetic works were produced in al-Andalus since a very earlyperiod and blossomed in the eleventh century C.E. More than two hun-dred poets were active during this period in al-Andalus. The numberand quality of these works attests to a profound cultural Arabization bythis date, at least in the high strata of society.

Regarding the use of colloquial Arabic, we must note that the Arabsthat came to the peninsula spoke a variety of Arabic dialects that grad-ually evolved into one main linguistic variety, known as Andalusi Arabic.The main characteristics of Andalusi Arabic had taken form by thetenth century.45 Linguistically, this dialect showed diVerences with classicalArabic similar to those manifested in other neo-Arabic dialects, includingan analytic grammatical structure and signi� cant phonetic shifts. Otherspeci� c elements of Andalusi Arabic vis-à-vis classical Arabic resultedfrom the in� uence of the Romance and Berber dialects. This latter phe-nomenon is obviously the outcome of the contact of the diVerent groupsof population of Muslim Spain: native Arabic speakers, on the one hand,and, on the other, the more numerous Romance population of thePeninsula, along with the Berber groups that arrived in diVerent waveswith the Muslim armies.

In a similar situation to modern Arabic dialects, Andalusi Arabic hadno recognition from the speakers as a diVerent linguistic variety. It wassimply regarded as “corrupted” or “vulgar” Arabic. Its domains werethose of the Low language: family, casual and intimate interaction.Colloquial Arabic was used as well for certain kinds of literary compositionpertaining to Low registers, including proverb collections and folk poetry.46

44 Al-Zubayd“, ÿabaqˆt, 298.45 F. Corriente, Árabe andalusí y lenguas romances (Madrid: Mapfre, 1992) 11. For a gen-

eral study on the development of Arabic in non-Arabic speaking territories, see K.Versteegh, “Linguistic Contacts Between Arabic and Other Languages,” Arabica, 48, no.4 (2001), 419-69.

46 The use of the colloquial language in these literary compositions has led F. Corrienteto a disproportionate evaluation of the relevance of Andalusi dialect, in one of his earlierworks on the topic: “In one respect SpAr [Spanish Arabic] is unique in its epoch andwould remain so for centuries amidst Arab lands, and it is by the fact that its speakers



Grammarians were not concerned with its description or study. Not-withstanding, they made numerous references to this colloquial varietyin order to prevent the cultivated class from being in� uenced by it andmaking mistakes when using the literary language. Compilations of these“vulgar” usages gave place to a linguistic genre known as Laún al-’ˆmmaor Treatises on the errors of language made by the common people.47

The linguistic ideas expressed in these works reveal the prevailing feel-ings of the Arabic speech community with regard to the colloquial vari-ety. The earliest work of this kind is the Laún al-’awˆmm, by the abovementioned grammarian al-Zubayd“ (928-989 C.E.). In the preface, theauthor expresses his views on the diVerences between the literary andthe ”vulgar” language: “It is a question of alterations, owed to our ’ˆmma[common people], which has modi� ed the pronunciation (of certainwords) or adapted the meaning, and has been followed in this practiceby a great many people, to the point where these incorrect usages havein� ltrated into the works of poets, and the most eminent scribes andfunctionaries include them in their correspondence and make use ofdepraved expressions in their conversations.”48

2. Latin/Romance

One can assume that the linguistic situation of al-Andalus in the � rstyears of Islamic rule was not very diVerent from that described forChristian Spain (see above), with the only, though crucial, diVerencebeing that the new ruling group spoke and wrote in a diVerent language.The initial sociolinguistic situation of Muslim Spain radically changedas the native population increasingly adopted the religion and languageof the conquerors. This is the expected evolution in a society in whichbeing a Muslim and having a good knowledge of classical and collo-quial Arabic are basic and almost indispensable factors of integrationin its administrative and power structures.

were aware of the personality of their dialect and not a bit ashamed of it, to the pointthat they sometimes preferred it over Classical Arabic for purposes such as folk poetryand proverb collections.” F. Corriente, A grammatical sketch of the Spanish Arabic dialect bun-dle (Madrid: Instituto Hispano Árabe de Cultura, 1977), 8.

47 See Ch. Pellat, “Laún al-’ˆmma,” in Encyclopaedia of Islam (Leiden: Brill, 1986), 5:605-10. Speci� c for al-Andalus, see G. KrotkoV, The “Laún al-’awˆmm” of Ab� Bakral-Zubayd“” Bulletin of the College of Arts and Sciences (Baghdad), 2 (1957), 3-15, and J.Pérez Lázaro, al-Madjal ilà taqw“m al-lisˆn wa-ta’l“m al-bayˆn (Introducción a la corrección dellenguaje y la enseñanza de la elocuencia) de HiÒˆm al-Lajm“ (m. 577/1181-1182) , 2 vols. (Madrid:CSIC, 1990).

48 I follow the English translation of Pellat, “Laún al-’ˆmma,” 606.

One of the best documented linguistic changes occurring in al-Andaluswas the displacement of Latin by classical Arabic as a language of cul-ture by the tenth century. No signi� cant Latin texts were produced inal-Andalus after this date.49 The demise of Latin among Christian pop-ulation forms the background to the famous complaints of Alvarus ofCórdoba in 854:

What trained person, I ask, can be found today among our laity whowith a knowledge of Holy Scripture looks into the Latin volumes ofany of the doctors? [. . .] Do not all the Christian youths, handsomein appearance, � uent of tongue, conspicuous in their dress and action,distinguished for their knowledge of Gentile lore, highly regarded fortheir ability to speak Arabic, do they not all eagerly use the volumesof the Chaldeans, read them with the greatest interest, discuss themardently, and, collecting them with great trouble, make them knownwith every praise of their tongue, the while they are ignorant of thebeauty of the Church and look with disgust upon the Church’s riversof paradise as something vile. Alas! Christians do not know their ownlaw, and Latins do not use their own tongue, so that in all the col-lege of Christ there will hardly be found one man in a thousand whocan send correct letters of greeting to a brother. And a manifold crowdwithout number will be found who give out learnedly long sentencesof Chaldean rhetoric.50

The history of Romance is more diYcult to trace given its nature as Lowlanguage and, as such, its neglect in the documentary evidence avail-able to us. As a case parallel to colloquial Arabic, Romance had notrecognition from its speakers as a language diVerent from Latin. It wasrather regarded as “vulgar” speech (see above p. 110). Christians’ per-ception of Romance as a “low” form of Latin51 is re� ected in one ofthe Arabic names given to Romance in Muslim Spain, namely, al-la “n“al-’ˆmm“ (“vulgar Latin”), found in Arabic pharmacological treatises.Consider the following examples, taken from the Commentary on Dioscoridesby the Andalusi pharmacologist Ab� Muúammad ’Abd Allˆh ibn al-Bay ˆr (� rst half of the thirteenth century):52

49 All the Latin texts produced in al-Andalus have been assembled by J. Gil, Corpusscriptorum muzrabicorum, 2 vols. (Madrid: Instituto Antonio de Nebrija, CSIC, 1973).

50 English translation by E. P. Colbert, The Martyrs of Córdoba (850-59): A Study of theSources (Washington: The Catholic University of America, 1962), 301. Latin text in Gil,Corpus, 1: 314-15.

51 See Lapesa, Historia, par. 40 (160-61).52 A. Dietrich, Die Dioskurides-Erklärung des Ibn al-Bai ˆr: Ein Beitrag zur arabischen

P�anzensynonymik des Mittelalters (Göttingen: Vandenhoeck & Ruprecht, 1991). “VulgarLatin” is also one of the names used by the tenth-century Andalusi botanist Ab� Dˆw�d



1. [murs“nis “mˆrus] . . . wa-thamratuhu huwa al-murdˆyˆnaj bi-l-suryˆn“yya wa-l-murtˆn wa-al-murta bi-l-la “n“ al-’ˆmm“ wa-zahruhutusamm“hu al-barbar bi-lughatihim "iqmmˆm wa-huwa al-raihúˆn bi-lugha ahl al-Andalus53

Translation: [murs“nis “mˆrus] . . . Its fruit is called in Syriac “mur-dˆyˆnaj” and in vulgar Latin “murtˆn” and “murta.” The Berbers callits � owers “ "iqmmˆm” and in the language of the people of al-Andalusthey are called “raiúˆn.”2. [q�mˆr�s] . . . wa-huwa ’a§“r al-dabb ’inda ahl al-Andalus wa-’ˆmmaal-Maghrib tusamm“hu "asˆsn� wa-bi-l-la “n“ al-’ˆmm“ ma r�niya54

Translation: [q�mˆr�s] . . . “the people of al-Andalus call it “ ’a§“r al-dabb,” while ordinary people of the Magrib call it “"asˆsn�.” In vul-gar Latin it is “ma r�niya.”3. [uqaiyˆqanÆus] . . . bi-lugha ahl al-Andalus zu’r�r al-awdiya wa-al-jabriy�l bi-l-’ˆmm“ min al-lisˆn al-la “n“55

Translation: [uqaiyˆqanÆus] . . . In the language of the people of al-Andalus [it is called] “zu’r�r al-awdiya” and in vulgar Latin [it iscalled] “jabriy�l.”

But the most normal way to refer to Romance in Arabic sources isusing the term ’ajamiyya (foreign, non-Arabic language). One must becareful, however, in the interpretation of this term since, as E. Lapiedrahas demonstrated, the original linguistic meaning of ’ajam developed intoseveral ethnic and religious senses, including “Christian,” and “non-ArabicMuslim,” as attested in Andalusi Arabic chronicles.56 An Arabic passageusually quoted with regard to the use of Romance among Muslims is,in my view, an example of the confusion created by the semantic evo-lution of ’ajam“. The passage in question appears in al-Khushan“’s accountof the appointment of a new judge in Córdoba by ’Abd al-Raúmˆn III(tenth century C.E.). Among the candidates to this oYce, al-Khushan“reports that one had Christian origins (interpreted by other scholars as“Romance speaking origins”): “At that time there was a candidate to

Sulaymˆn Ibn Juljul, when referring to Romance names, (see Simonet, Glosario de vocesibéricas, xxiv n. 4). Ibn Juljul´s references are preserved as well in a later anonymousbotannical treatise; see M. Asín Palacios, Glosario de voces romances registradas por un botánicoanónimo hispano-musulmán (siglos XI-XII) (Madrid and Granada: CSIC, 1943), xxxvi-xli. Foran Arabic edition, translation, and commentary see A. Dietrich, Dioscurides triumphans: einanonymer arabischer Kommentar (Ende 12. Jahrh. n. Chr.) zur Materia medica: arabischer Text nebstkommentierter deutscher Übersetzung/herausgegeben von Albert Dietrich (Göttingen: Vandenhoeck &Ruprecht, 1988), 2 vols.

53 Dietrich, Dioskurides-Erklärung, 375/115 (Arabic text/Translation and Commentary).54 Dietrich, Dioskurides-Erklärung, 373/131.55 Dietrich, Dioskurides-Erklärung, 378/95.56 E. Lapiedra, Cómo los musulmanes llamaban a los cristianos hispánicos (Alicante: Instituto

de Cultura “Juan Gil-Albert”, 1997), 258-85.

the judgeship who had Christian ancestors (kˆna f“ ubuwwiyyihi ’ajama).When Aslam was dismissed [from the judge oYce] and Al-îab“b wasappointed, Aslam started to say: ‘Thanks God, who made me be amongthose who say “There is no more god than God,” alluding to the can-didate whose parents were Christian (kˆna ˆbˆ"uhu ’ajaman).57 The reli-gious sense of ’ajam in this context becomes clear in Aslam’s remarkabout his being Muslim (“There is no more god than God”) in oppo-sition to the unsuccessful candidate, whose religious aYliation is not soclear. In Ribera’s rendition of the highlighted sentences, the original[but not appropriate for this context] linguistic meaning of ’ajam hasbeen adopted: “somebody descendant of a completely Latinised family,both his father and his mother” and “[the candidate] whose parentswere known to be Latinised.”58

The use of ’ajam in this passage is similar to its use in the MuqtabasII of Ibn îayyˆn, when the author narrates the episode of the surrenderof a rebellious group of Christians and Muslim converts (muwallads) tothe wizier HˆÒim ibn ’Abd al-’Az“z, in 875 C.E. HˆÒim gathered every-body in one place and then asked to each man: “muslim anta am a’jam“ ?”(are you Muslim or Christian?). If they answered “Christian” HˆÒimordered them to be killed immediately. To those who answered “Muslim”HˆÒim asked them to recite a sura of the Koran and then another one,and another one . . . until they hesitated or made a mistake, giving HˆÒiman excuse to kill them, saying “mˆ akhbartukum annahu a’jam“ ?” (did I nottell you that he was Christian?), until all the men were dead.59

The status of Romance as a Low language makes its use extremelyrare in belles lettres literature. One of the few testimonies of the use ofthis language in the literary works produced in Muslim Spain are the� nal couplets or kharjas of a strophic verse composition known asmuwashshaú. The norms of the genre dictate that these � nal verses should

57 Ribera, Historia de los jueces de Córdoba, 188.58 Ribera, Historia de los jueces de Córdoba, 234/188. Other scholars have interpreted

these sentences in same way. See, for instance, D. Hanlon “A Sociolinguistic view ofhazl in the Andalusi Arabic muwashshaú,” Bulletin of the School of Oriental and African Studies,60 no. 1 (1997), 35-46 (39). Hanlon has also erroneously identi� ed the candidate ofChristian origins with the one who is � nally appointed for the oYce [Al-îab“b]: “Theadoption of Islam as a way of life and Arabic as a means of communication similarlyenabled one al-îab“b, of Romance-speaking parents [. . .] to pursue a legal career andrise to the top of the Andalusi judiciary in the early tenth century” [p. 39].

59 See J. Vallvé, “Juniores, ‘fatas’ y jenízaros: una re� exión sobre la situación actualen Yugoslavia,” Boletín de la Real Academia de la Historia, 141 no. 1 (1994), 1-35 (21-22),and Lapiedra, Cómo los musulmanes, 264-65. Arabic text in M. A. Makki, ed., Al-Muktabasmin anbˆ" ahl al-Andalus (Beirut: Dar al-Kitab, 1973), 131-32.



constitute a ludic, frivolous element, in contrast to the poem in whichthey are inserted. The use of languages � ts this thematic distribution:classical Arabic is employed for the main poem and colloquial Arabicor Romance for the � nal verses. This linguistic distribution con� rms theperception of Romance and Colloquial Arabic at the same level, as Lowlanguages or languages of informal contexts.60 The two language levels(High and Low), operating in the kharjas attest to the “extended” diglos-sia that must have existed at one stage within the Andalusi speech com-munity, having classical Arabic as a language of culture and Romanceas a Low language.

The distinction between the two sociolinguistic levels should lead usto a reconsideration of the linguistic data available: the use of classicalArabic is not de� nitive evidence of the existence of an Arabic speakingpopulation, whereas Latin literature could have been produced by speak-ers of Colloquial Arabic. In fact, we have examples of both situations:after their territories were “reconquered” by Christians, Muslim com-munities kept on using literary Arabic, mostly notarial documents, forinternal purposes, at a time that they already had Romance as a spo-ken language.61 On the other hand, we learn from the Latin literatureproduced by the Christians of al-Andalus that some of the so calledmartyrs of Córdoba uttered their insults against Islam in vernacularArabic, although their instruction had been in Latin in most of thecases. Most scholars agree that when Latin stopped being used aroundthe tenth century and classical Arabic became the educated languageof everyone, Romance continued to be spoken among certain groupsof the population until the end of the eleventh century (if not later).The dates for its disappearance in Muslim Spain are a controversialissue beyond the scope of this article.62 In any case, we must bear inmind that little or lack of evidence in the written sources of that perioddoes not prove the disappearance of Romance, given its marginalizedstatus within Andalusi society.

The status of Romance as a Low language has other implications inour description of language use in al-Andalus: many scholars have sug-

60 See Hanlon, “A Sociolinguistic view of Hazl.”61 Viguera, “Introducción,” 19-20.62 See A. Galmés de Fuentes, Las jarchas mozárabes. Forma y signi�cado (Barcelona:

Grijalbo, 1994), 80-88, where the author argues the preservation of Romance in al-Andalus until the thirteenth century. In an erudite survey, Simonet compiled all the evi-dence from primary sources that could be used in favor and against the Arabization ofthe Christians of al-Andalus, see Simonet, Glosario de voces ibéricas, vii-xxxiii.

gested Romance/colloquial Arabic bilingualism in Andalusi society dur-ing the � rst two centuries of Islamic presence in the peninsula, withoutgiving much explanation of the terms of such bilingualism. We mustkeep in mind that we are speaking of two Low languages whose domainswere just the same: casual linguistic interaction. And by comparisonwith other bilingual situations, it is unlikely that speakers had main-tained the use of both linguistic varieties with the same functions forsuch a long period. As J. Fishman states: “Bilingualism without diglos-sia tends to be transitional both in terms of the linguistic repertoires ofspeech communities as well as in terms of the speech varieties involvedper se. Without separate though complementary norms and values toestablish and maintain functional separatism of the speech varieties, thatlanguage or variety which is fortunate enough to be associated with thepredominant drift of social forces tends to displace the other(s).”63 Inthe light of Fishman’s statement it seems more likely to think that thereexisted colloquial Arabic/Romance bilingualism simply as a previousand necessary step towards Arabic dominance. But it is unlikely that itconstituted a stable or permanent linguistic situation. Not all the groupsof the population, however, acquired Arabic as a spoken language atthe same time. We can assume that the Arabization of rural areas,women and marginalized groups of the population had taken place ata slower pace. This would account for the two or more centuries ofco-existence of Romance with Colloquial Arabic in al-Andalus and forthe transitional bilingual states.

3. Muslim Language use

The three spoken languages of the Muslims during the � rst period ofAndalusi history were Arabic, Berber, and Romance. This initial linguisticsituation gave way to the almost entire dominance of Arabic from theeleventh century onwards. The only language of culture used by theMuslims of al-Andalus was classical Arabic. No writings in the othertwo learned languages of al-Andalus, Latin and Hebrew, were producedby Andalusi Muslims.

The displacement of Romance and Berber by colloquial Arabic waspreceded by a period of language contact and bilingualism, which lefta signi� cant imprint in Andalusi Arabic.64 There are quite a few testimonies

63 Fishman, “Bilingualism with and without diglossia,” 36.64 See Corriente, Arabe andalusí, 125-53.



of the existence and use of Romance among Andalusi Muslims duringthe � rst two centuries of the history of al-Andalus. Some of them havebeen transmitted in al-Khusan“’s Book of the Judges of Córdoba, includingthe following two anecdotes having place in Córdoba, around the mid-dle of the ninth century C.E.:

At that time there was in the city an old man whose language wasRomance (a’jam“ al-lisˆn), called Yenayr. He used to go before thejudges to declare, [since] he was well known among the people for hisgood conduct and orthodox belief. The wazirs sent for him and askedhim about this judge. He answered in Romance: “I do not know him,but I have heard people saying that he is a bad man.” He said that[“bad man”] using a diminutive in Romance. When his words reachedthe emir—May God have Mercy upon him—the emir admired hiswords and said: “Nothing but truthfulness would have caused this manto utter such a word.” He then dismissed him as a judge.65

One day a woman came to the judge [Sulaymˆn ibn Aswad] andsaid to him in Romance (bi-l-’ajamiyya): “Oh judge, listen to this unluckywoman!” The judge answered in Romance: “You are not unlucky. Theunlucky one is Ibn ’Amar’s mule which is gnawing its rein at themosque’s door all day long”66

Certain episodes of Muslims speaking or showing knowledge of Romanceappear in Andalusi written sources of later periods, reaching the thir-teenth century. In Ibn êˆúib al-êalˆ’s chronicle of the Almohad Empire(Al-Mann bi-l-Imˆma), we read an interesting anecdote that involves anAndalusi knowing Romance and, further, a Christian king assuming hisknowledge of this language. This Andalusi, Ibn Waz“r, had come to thepeninsula from Marrakesh in the year 1170 C.E., accompanying theAlmohads in a military campaign. The Almohad leader Ab� Sa’“d, sonof the Almohad caliph ’Abd al-Mu"min, sent Ibn Waz“r to speak to theChristian King Fernando II of Leon and make a pact with him. IbnWaz“r’s account of this encounter is as follows:

When he [Abu Sa’“d] sent me, together with the oYcial ambassadors,to the above mentioned Fernando the Baboso (Fernando “the Drooler”),I entered upon the Baboso’s tent, and with him were his Christiancounts and his learned men. He spoke to me in his Romance tongue(bi-lisˆnihi al-’ajam“) and I said to his translator: “I do not understandRomance (lastu afham al-’ajamiyya).” But I only said that hiding the truthand as a trick so that I might understand from what he said whether

65 Ribera, Historia de los jueces, 96/118.66 Ribera, Historia de los jueces, 139/179.

he wanted good or evil. His translator translated on my behalf thepurpose for which I had come, as mentioned above. So I treated himkindly about the truce until it was completed.67

The linguistic reality that the appearance of Latin and Romance namesin Andalusi pharmacological treatises re� ects, is the object of contro-versial analysis. For scholars like A. Galmés de Fuentes, translating intoRomance the names of medicinal plants, as in the treatise of the abovementioned Ibn al-Bay ˆr, is evidence of the use of this language in al-Andalus until the thirteenth century C.E.68 Other researchers, like C. Barceló, claim that this Romance lexicon does not re� ect any live use,but that it is simply a transmission of older materials produced in theOrient in the ninth and tenth centuries.69

If the dates of the use of Romance are problematic, we have at leastclear testimonies of the attitudes and values ascribed to this linguisticvariety among the Muslims of al-Andalus: Romance is mostly associatedwith Christianity and non Arabic people. The use of the same term(a’jam“) in Andalusi Arabic sources with the two meanings of “Romancespeaker” and “Christian” (see above pp. 128-29) symbolizes this identi� ca-tion of language use and religious aYliation.

Further, Romance languages are linked with the people of the ChristianKingdoms, that is, with the enemies of the Muslim state. This is theidea underlying the complaints of the Toledan vizier Ab� al-Mu arrifibn Muthannˆ (d. 1066 C.E.), about hiring slave women from the Northfor the raising of Andalusi noble children:

67 The History of the Moroccan Empire in Maghrib, Andalusia, and Ifriqiya or volume II of“Alman, bil Imamah Al-mustahdha�n,” ed. Abdul Hˆd“ Attaz“ (Beirut: Dar al-Andalus, 1964),402. This text was translated into Spanish by A. Huici Miranda, Ibn êˆúib al-êalˆ: al-Mann bil-Imama (Valencia: Anubar, 1967). In Huici Miranda’s translation, however, thereis an important mistake (probably typographical): the sentence “lastu afham al-’ajamiyya”of the original text has been rendered as “I understand the foreign language” (“com-prendo la [lengua] extranjera”), that is, exactly the opposite of what the author meant.This wrong data has been incorportated in the work of several scholars, including N. Roth, Jews, Visigoths and Muslims in Medieval Spain. Cooperation and Con�ict (Leiden: Brill,1994): “During the Almohad period we have the contemporary evidence of Ibn êˆúibal-êalˆ, who reports [. . .] that when the former [Ibn Waz“r] went to negotiate withFernando II (“el Baboso”) of León, he told the Christian interpreter there that he under-stood the Spanish language” (my emphasis, p. 53) and in Zwartjes, Love Songs from al-Andalus,15 n. 18. It is from the entire story that we can deduce that Ibn Waz“r knew Romance,but not from that particular sentence which means the opposite of what is found inHuici Miranda’s translation.

68 Galmés de Fuentes, Las jarchas mozárabes, 88.69 Barceló, “Mozárabes de Valencia,” 271.



How far is high from low, the sky from the earth, the light from thedarkness, the eloquent expression from the barbarism [. . .] Aren’t we,people of this peninsula which is far away from the best nations, neigh-boring the Barbar (al-’ajam) masses, aren’t we the worthiest of excusefor [our] incorrect speaking (lukn) [. . .]? Because, isn’t it true that sinceone of the sons of your nobility starts to hear when he is born [. . . .],he does not hear but the words of a despicable, Romance speaking(a’jamiyya), simple minded slave-woman, and the baby does not suckle butfrom her breast, and does not acquire but her incapability of expression,and is not calmed but in her lap, and is not trained but under herdirection? To the point that when he becomes a man, culminating hisgrowth, he is in touch with the Christian kingdoms since he speaks tothem in their languages, he makes an eVort in keeping their language,he is concerned about their social classes and tolerates their habits.”70

The religious meaning of Arabic as the language of the sacred message,along with the special circumstances of al-Andalus, in constant belligerentstate with Romance speaking Christian kingdoms, made Arabic the onlylanguage viewed as the true language of Muslims. Even in the momentsof highest ethnic tension between Arab Muslims and native IberianMuslims, at the end of the ninth century, the meaning of the Arabiclanguage as the language of Islam seems to prevail over its ethnic con-notation as language of the Arabs.71 In the best known anti-Arab man-ifesto, known as Risˆla of Ibn García, Ibn García never rejects the useof Arabic. On the contrary, he asks the Arabs to be understanding withthe non-Arabs faulty literary style: “Oh you monopolizer of the art ofpoetry, and master of the pen in both prose and poetry [. . .] Even theelegant in speech may/sometimes lapse into error, so do not/accept themodest, well-meant work/other than with aVability.”72 Very signi� cantly,in the Arab refutations to this Risˆla, Arabs accuse Iberian Muslims of

70 Transmitted by Ibn Bassˆm, al-Dhakh“ra f“ maúˆsin al-Jaz“ra, ed. Iúsˆn Abbˆs, 8 vols.(Libya: 1978). I follow, with little modi� cation, the translation of Lapiedra, Cómo los cris-tianos, 274-75.

71 In an unpublished paper, M. Marín points to a very signi� cant event having takenplace at the time of these ethnic � ghts. The confrontation between the two groupsadopted a literary form in the town of Elvira: a poet was chosen as a spokesman ofeach party and was in charge of defending his group’s claims with his poetic composi-tions. The fact that the Iberian Muslims participated in this typical Arabic custom, pre-cisely to � ght the Arabs, indicates, their deep assimilation of Arabic cultural patterns;see M. Marín, “La arabización de al-Andalus: aspectos sociales y culturales,” paper givenin the colloquium “El siglo VIII: Islam y Occidente,” Madrid, 1997). I am indebted tothe author for kindly providing me with a copy of this work.

72 J. T. Monroe, The Shu’�biyya in al-Andalus. The Risˆla of Ibn García and Five Refutations(Berkeley and Los Angeles: University of California Press, 1970), 29.

ingratitude, for not acknowledging the gift of having the Arabic lan-guage instead of their previous uncivilized languages:73

Woe to you, O brave lion. O buttocks of an ass scratched by the crup-per, when you brayed; although it was in the language of the Arabs,your prisoners, that you acquired learning! [. . .] You muttered in theArabic language yet you wearied of the Arab’s wisdom; you utteredtheir war cry, imitated their poems, and brayed among their asses,though your brand is not like their brand. Did you not perfect yourintelligence after having spoken in your defective tongue and foreignbabble?[ [. . .] “Did you and your ancestors, O base one, O you whohave avoided the prescribed castigation, ever have a language to speak[. . .], or a grammar in your tongue for us to record, or a babble inyour previous condition to in� ect and decline?74

The religious signi� cance of Arabic is attested as well in later periods,after Muslim communities fell under Christian rule and subsequentlybecame Romancized. These Muslims groups developed their ownRomance literature, known as Aljamiado, whose main distinctive char-acteristics result from the incorporation of Muslim signs of identity,namely, the Arabic script and Arabic vocabulary pertaining to the reli-gious domain (see above p. 115).

4. Christian language use

The most distinctive linguistic trait of Christian population during the� rst Islamic period of al-Andalus was the preservation of Latin as a lan-guage of culture until approximately the middle of the tenth century. Theprogressive displacement of Latin by classical Arabic is the origin of PaulAlvar’s complaints in 854 (see above p. 127). In Alvar’s speech, there isa clear identi� cation of Latin as the language of Christianity versus Arabic,the language of the “Chaldeans.” The fact is that the links between Latinand Christiany had never been as strong as the links between Hebrewand Judaism, let alone those of Arabic with Islam. Latin was not thelanguage that Jesus Christ had used to transmit his message, nor wasit the language of the New Testament, originally written in Greek. Latin,however, had become the language of the Roman Church around thefourth century C.E., and it was the language praised by Christian per-sonalities as important as Saint Isidore of Seville and Saint Jerome.

73 Monroe, The shu’�biyya in al-Andalus.74 Response of Ab� Yaúyˆ ibn Mas’ada. See Monroe, The Shu’�biyya in al-Andalus, 32-

33.



It was probably the weak religious identity of Latin as the languageof Christianity, alongside the prestige of Arabic as a vehicle of culture,which led to the displacement of the former by Arabic in Christianwritings. This process was so extensive that by the time that Paul Alvarwas bemoaning the demise of Latin among Christians, somebody calledîaf§ ibn Albar al-Q� “ was translating the Book of Psalms from Latininto Arabic. Further, a canon law collection of al-Andalus was producedin Arabic by a priest called Vicentius around the eleventh century, ifnot before. The scant examples of Arabic literature of Andalusi Christiansincludes the work of two poets of the eleventh century: Ibn al-Mir’iz“and Ab� al-Qˆsim ibn Khayyˆt. No Arabic literature (belles lettres) byChristian authors is known to exist from the twelfth century onwards.75

With the exception of Paul Alvar’s statements, there are no clear tes-timonies on the part of the Mozarabs about their attitudes regardingthe use of languages. It has to be said, � rst of all, that their literaryoutput was considerably reduced by comparison with that of the Jewishminority or by comparison with Christian communities of the Near East.Their use of literary Arabic with an aesthetic purpose is only attestedin the case of the two poets mentioned above.

The identi� cation of Romance as the language of the Christians isattested in Arabic sources, as mentioned above (p. 128). We also learnfrom historical chronicles that the rulers of al-Andalus used to employAndalusi Christians as interpreters when they had to meet with Christianrulers of the North. This choice seems to be motivated not just by theknowledge that these Christians had of Romance and Arabic, but bythe role that they played as intermediaries, as people “in between” thetwo worlds.76

The identi� cation of Romance users with Christians can hardly havecorresponded to the Andalusi linguistic reality in the � rst years of Islamicrule, given that Islamization is a process that usually occurs at a fasterpace than Arabization.77 We can infer a sharpening of linguistic atti-

75 For a general study on the Arabic literature of Andalusi Christians, see P. Sj. VanKoningsveld “Christian Arabic literature from medieval Spain: An attempt at peri-odization,” in Christian Arabic Apologetics during the Abbasid Period (750-1258) , ed. S. KhalilSamir and J. S. Nielsen (Leiden: Brill, 1994), 203-24 and H. Kassis “Arabization andIslamization.”

76 See Lapiedra, Cómo los musulmanes, 269-70, where she gives an example of anAndalusi Christian dignatary who is dismissed from his oYce after translating a messageof the Northern rulers that specially bothered the caliph.

77 For seminal study on conversion to Islam in the Middle Ages, see R. Bulliet,Conversion to Islam in the Medieval Period: An Essay in Quantitative History (Cambridge, Mass.)

tudes from the end of the eleventh century, according to the historicalcircumstances: increasing advances of the Christian Kingdoms and arrivalin al-Andalus of the North African dynasties, of a stronger Islamic ortho-doxy. Not surprisingly, there is no trace of Romance language use inal-Andalus from the thirteenth century onwards, the period of highesttension between the two worlds. This is also the period when one ofthe Romance languages, namely Castilian, was used with the status ofoYcial language in most of the Christian North.

5. Jewish language use

The eZorescence of the Jewish community of al-Andalus, specially inthe tenth and eleventh centuries C.E., is unique in the history of Jewsin the Middle Ages: “No other Jewish community produced as manyJews who achieved positions of status and even power in the non-Jewishworld; and no other Jewish community produced such an extensive lit-erary culture re� ecting the deep impact of an intellectual life sharedwith non-Jews.”78 It is within this historical context that the most strik-ing linguistic phenomenon of the Andalusi Jews evolved: the revival ofthe Hebrew language for literary use outside the religious sphere.

Hebrew holds a special position in the Jewish tradition as a sacredlanguage: it is the language in which the Bible was written and the lan-guage of Jewish liturgy.79 The way Hebrew is referred to in Rabbinicalsources is precisely leshon ha-kodesh (the sacred language). Biblical Hebrewhad ceased being a spoken language after the fall of Jerusalem (587B.C.E.), with its domains as a written language limited since then primar-ily to religious writings.80 In al-Andalus, however, Hebrew acquired anew vitality as a literary language. In the wake of the concept of ’arabiyya,which claims the superiority of Arabic language and culture, prevailing

Harvard University Press, 1979). Conversion to Islam in al-Andalus has been studiedfrom a diVerent perspective by M. de Epalza “Mozarabs: an emblematic Christian minor-ity in Islamic al-Andalus,” in The Legacy of Muslim Spain, ed. S. Kh. Jayyusi and M.Marín (Leiden: Brill, 1994), 188-200.

78 R. P. Scheindlin “The Jews in Muslim Spain” in The Legacy of Muslim Spain, eds.S. Kh. Jayyusi and M. Marín (Leiden: Brill, 1994), 188-200 (188).

79 See A. S. Halkin, “The Medieval Jewish Attitude Toward Hebrew,” in Biblical andOther Studies, ed. A. Altmann (Cambridge, Mass.: Harvard University Press, 1963), 233-48. On the traditional ideas about the Hebrew language within Judaism, see R. Loewe.“Hebrew Linguistics,” in History of Linguistics, Volume I: The Eastern Traditions of Linguistics,ed. G. Lepschy (London: Longman, 1994), 97-163.

80 See A. Sáenz-Badillos A History of the Hebrew Language (Cambridge: CambridgeUniversity Press, 1993) 52.



among Andalusi Muslims, Jews made Hebrew an object of linguisticpraise and gave it a new life.81 Apart from extensively studying andanalyzing their language, as the Muslims were doing with Arabic, Jewishscholars in al-Andalus used biblical Hebrew for the composition of sec-ular poetry, in which so many excelled.

The high status of Hebrew within the Jewish community, along withits use as a language of “elevated” literary purposes, lead us to includeit as a type of High language. Notwithstanding, other linguistic vari-eties � lled the domains pertaining to this: classical Arabic, the languageof administration and general culture in al-Andalus, and Judeo-Arabic,a speci� cally Jewish variety of classical Arabic.

As the main vehicle of culture in al-Andalus, classical Arabic playeda key role as a language of learning among Jewish males of the upperclasses. Classical Arabic was also an indispensable tool for gaining accessto the Islamic power and administrative structures. Among the Jewishcourtiers employed by Muslim rulers, the most prominent case is thatof Isma’“l ben Naghr“la (known as Samuel ha-Nagid in Jewish tradi-tion), who served as a vizier for the ruler of Granada in the � rst halfof the eleventh century. His deep knowledge of classical Arabic isacknowledged not just by his correligionists (see above Ibn Tibbon’sremarks, p. 117) but also by Muslims scholars, including the Andalusihistorian Ab� Marwˆn Ibn îayyˆn, who portrays Ibn Naghr“la in thefollowing way:

This cursed man was a superior man, although God did not inform himof the right religion. [. . .] He was an extraordinary man. He wrote inboth languages: Arabic and Hebrew. He knew the literatures of bothpeoples. He went deeply into the principles of the Arabic languageand was familiar with the works of the most subtle grammarians. Hespoke and wrote classical Arabic with the greatest ease, using this lan-guage in the letters which he wrote on behalf of his king. He usedthe usual Islamic formulas, the eulogies of God and Muhammad, ourProphet, and recommended the addressee live according to Islam. Inbrief, one would believe that his letters were written by a pious Muslim.82

81 For an analysis of the factors and characteristics of the revival of the Hebrew lan-guage in al-Andalus, see A. Sáenz-Badillos “Philologians and Poets in Search of theHebrew Language” in Languages of Power in Islamic Spain, ed. R. Brann and D. I. Owen(Bethesda: CDL Press, 1997), 49-75.

82 Text transmitted by Ibn al-Kha “b in al-Iúˆ a f“ akhbˆr Gharnˆ a. I follow the Englishtranslation of A. Schippers, Spanish Hebrew Poetry and the Arabic Literary Tradition (Leiden:Brill, 1994), 54-55.

The literary work of Ibn Naghr“la represents the linguistic choices oflearned Jews in al-Andalus: Classical Arabic for interaction with theMuslim community and as the usual language of administration; Hebrewas the preferred language for poetry (the most praised art in the cul-ture of the time); and Judeo-Arabic for the rest of literature, mostly ofscienti� c character: grammar, medicine, theology, philosophy, etc.

The evaluation of Judeo-Arabic has gone through diVerent stages inmodern scholarship.83 When we refer to medieval Judeo-Arabic, wecould best de� ne it as a variety of classical Arabic used by Jews forinternal consumption, whose main characteristics are the use of Hebrewscript, a signi� cant input of Hebrew lexicon, and the use of registersclose to spoken language or neo-Arabic.84 The signi� cance of Hebrewas a sacred language within Jewish tradition is clearly in the origin ofthe distinctiveness of Judeo-Arabic vis-à-vis classical Arabic. The markedIslamic character of classical Arabic must have strengthened the attemptsof the Jewish community to establish its own distinctive character insuch a symbolic sphere as linguistic expression.

Conclusion

In sum, from the previous survey on the use of languages in MedievalIberia one observes the shaping eVect of religion in the users’ evalua-tive behavior of the diVerent linguistic varieties. This evaluation is lessovert in the � rst period of Islamic conquest but clearly de� ned fromthe eleventh century onwards. The aYliation of Classical and ColloquialArabic with Islam, that of Hebrew with Judaism, and Latin and Romancewith Christianity, fostered the prevailing linguistic attitudes in Muslimand Christian Spain. These attitudes are evident in the linguistic choicesof H domains, including belle-lettres literature and religious practice.Further, the development of a literature in Aljamiado, Judeo-Romance,and Judeo-Arabic responds to one general phenomenon of language useunder the in� uence of a sacred language by members of a social groupde� ned by religion.

83 The most extensive research on Judeo-Arabic has been carried out by J. Blau,whose opinions have changed over time. For a general study on medieval Judeo-Arabicand Blau’s updated opinions, see J. Blau, The Emergence and Linguistic Background of Judaeo-Arabic, 3rd ed. ( Jerusalem: Ben Zvi, 1999). For a general survey on Judeo-Arabic, seeB. Hary Multiglossia in Judeo-Arabic, with an Edition, Translation and Study of the Cairene PurimScroll (Leiden: Brill, 1996), 71-111.

84 See M. A. Gallego, “Factor religioso y factor lingüístico en el judeo-árabe medieval,”’Ilu: Revista de Ciencias de las Religiones, 2 (1997), 43-48.


Gallego (M. A.)_The languages of medieval Iberia and their religious dimension.pdf

Documents

written language

kinds of language use

social categories social

social variables

social phenomenon

spoken language

diverent social contexts

use of languages