B. Lukács

Matter Evolution Subcommittee of the Geonomy Scientific Committee of HAS

H-1525 Bp. 114. Pf. 49.

lukacs@rmki.kfki.hu

Medieval linguistic evolution in Hispania took 2 divergent paths because of different substrates and superstrates. A hypothesis is discussed that honorific title of Sir Rodrigo Diaz de Vivar, El Cid, comes not from Arab but Mozarab, sister language of Spanish.

Example 1

Ven, cydy, veni,

el querer es tanto bieni

Dexa al-zameni

con filyo d'Ibn al-Dayyeni

Kharja written by Y. Halevi [1]

Example 2

Odavan az ország,

Anon, from Upper Nógrád county [2]

Example 3

Des cuand mio Cidiello viénid

¡tan buona albischara!

com rayo de sol éxid

en Wadalachyara

Y. Halevy for Y. ibn Ferrusiel [1]

Example 4

Anon, collected by J. Kollár [2]

I do know that the Spanish part of the title is ungrammatical; still its meaning is unequivocal. I chose the mirror expression of the title of the famous epic "Poema de mio Cid". I try to answer the question, whose was the Cid, the hero of the poem.

According to common sense the answer is trivial. Sir Rodrigo was a hero of the Reconquista, therefore he is of the Spanish-Christian side. No problem.

And still... "Cid" is a honorific title: "Leader", "Lord" or such. But not in "Spanish". I mean, neither in Castilian, nor in Catalan, nor in Gallego; not even in Basque. It seems rather Arabic. (In French-Arab it is Sidi as we meet it in a lot of place names in North Africa.) Did the "Spanish", reconquering masses of freemen warriors & nobility call Sir Rodrigo "Leader" in Arabic? Sounds strange.

Hungary fought 150 years for her own Reconquista against the Ottoman Empire, another Southern Moslim state. During this one and half centuries interesting situations appeared which seemingly violated Aristotelian logic and obviously did the Western legal system. E.g. in Transylvania a semi-independent principality was formally the vassal of the Hungarian King (who was actually the same person as the German Emperor); while they paid the tax and sometimes the warriors to the Ottoman Empire, fighting then against the Christian Cause (and indirectly even against the Hungarian Cause), and still a Christian and Hungarian player.

It was possible that a Prince of Transylvania was a great name in "royal Hungary"; present Hungary honours the memories of all Transylvanian Princes (of some more, of some less) and there is a story recorded second-hand that one Archbishop and Primate told the Calvinist Transylvanian envoy, when only they two remained in the room, that "we have our good King and our status is honoured but did you not exist on the East, the German would spit under our collar" in which he was quite right. Nevertheless, while there were great Princes fighting for Hungary against Hungary, never anybody called them "my Effendi" (both in Hungarian & in Turkish "effendim"). That was a Turkish honorific and the hero was not a Turk. And Sir Rodrigo in the poem is "mio Cid". How, why?

In medieval Ages the grammar, and even more the vocabulary of a specific language was not strictly defined. (But obviously the French language is defined since 1637.)

Roman intellectuels used a lot of Greek words in Latin texts; what is more, then the Greek nouns were handled according to Greek grammar. Still, they are unequivocally Latin texts; if they are not pure Greek.

During Rinascimento a lot of authors wrote Latin; if not, they used the local vernacular. In France, where the transition from Latin to Old French was continuous, they recognised the difference in 813, when the Synod of Tours determined to translate some ecclestical Latin to "rustica Romana" for the use of local people, and after this the two languages did not mix. In Italy, where no substrate and not much superstrate influenced Vulgar Latin it took one century and a half more, and as a rarity even now occur hybrids. E.g. in 1962 Moretti calls an ancient Greek political body "il commune Boiotorum" [3]. While the meaning is unequivocal, it is clearly an Italian-Latin maccaroni; but it is exceptional.

Between Latin and a non-Neo-Latin such mixture is even more infrequent. True, both German and Magyar used ample Latin roots and I could cite Magyar texts from early XVIIIth century where the majority of verbal and nominal roots are Latin, still the grammar is pure Magyar. Modern European texts are generally written in one language, although Anglo-American roots are ample everywhere.

On the other hand, English is an explicit example that hybridization was possible even amongst different subfamilies. Present English is a hybrid of German Anglo-Saxon and Neo-Latin French. But note that this hybridisation happened in a way that the hybrid lost almost all its synthetic characters; most probably because the synthetic rules were different in the parent languages. According to a widespread idea (see e.g. [4]), when two communities of far languages mix then the language first pigdinizes, afterwards creolizes, and pidginization means the extreme simplification of grammar. And this sounds quite convincing.

Hungary is a country which is traditionally multiethnic. Therefore the Uralic language which is often called "Hungarian" is only the biggest language of Hungary, the Magyar. Since 1920 the ratio of non-Magyars is below 10 %, but this was always above 40 % between 1000 and 1920.

The official language until 1841 was Latin, and in some regions even peasants spoke some Latin; but if not, they spoke with each other as they could. Interestingly enough Pidgin languages never formed; people rather learned a second language. And Examples 2 & 4 at the very beginning are Hungarian hybrids. Both text is Magyar + Slovak; I slightly corrected the orthography to litterary norms, but the originals were simply "rural" ones. Some diacritical signs of the Slovakian may not go through.

Example 2 is linguistically simple but politically delicate. The short "poem" deals with the sad fate of the Hungarian Slavs caused by the breakdown of the power of King (in fact, Prince) Svätopluk. He died in 894, just before the Magyar conquest (895-896). According to the Magyar tradition Svätopluk did not recognise that the Magyars were to buy his country from him for a white horse, a crimson briddle and a golden saddle-cloth. Now, the Slovakian informer accepted the Magyar tradition, only resented it. The collector Ján Kollár was also heavily against Magyars, so much that he wrote his given name as Jan, because he wanted to be a Bohemian instead of a Slovakian. Now, I give the English translation, which is not at all surprising (I do not guarantee poetic quality):

Crimson briddle,

White horse,

The country is lost,

Oh my God!

However the third line is Magyar!

Lines 1, 2 & 4 are pure and correct Slovakian, Line 3 is pure and correct Magyar. And, while all the text may belong to a Slovakian (even to Prince Svätopluk), Lines 3 & 4 cannot be texts of a Magyar player in any case.

Then why to tell this line in Magyar between two Slovakians? We cannot know anymore; but my guess is that for syllable counting or, more generally, for metrum. So the composer took a Magyar line which fitted better amongst the Slovakian ones than any Slovakian he could have found.

Example 4 is more involved. In Line 1 words 3, 4, 5, 6 & 7 are Slovakian; word 8 is Magyar but with Slovakian orthography. In line 2 words 1-4 are Magyar and 5-8 are Slovakian. The translation (for word for word in order to keep the structure, so in English it will seem ungrammatical but in the original it is not):

Where wert, where wert thyself, thou shepherd shepherd?

When thou not wert yesterday evening at us?

Italic characters mean Slovakian words.

Now observe the structure. Line 1 tells everything twice, once Magyar, once Slovakian. Observe that declension is complete both in Magyar and in Slovakian, so the personal pronoun is not obligatory. Note also that the two "si"'s are homonymes in Slovakian. I do not know why to double; but for any case now "juhás" and "u nás" give a complete rhyme. Line 2 is complete and economic.

Observe also that at least the addressed (who is clearly male, not only being shepherd but also because Slovakian past tense shows gender) is surely Slovakian (in old Northern Hungary practically all shepherds were Slovakian); one may think that the female actress, the singer, might be Magyar, but it is hard to imagine Ján Kollár collecting Slovakian folk songs from a Magyar.

Obviously both informers' first language was Slovakian, they knew fluent Magyar too, and applied the mixture for poetic reasons. The result is not a pidgin speech: the speakers command full grammars. The texts are too short to check the amount of vocabulary; but pidgin or even Creole never evolved on the Magyar-Slovakian linguistic border.

Considering that Magyar and Slovakian belong to separate linguistic families this is interesting.

Examples 1 & 3 come from some 900 years ago, from Hispany. Both were composed by famous Yehuda Halévy (or Ha-Lévi). In Example 3 the addressed is Yosef ibn Ferrusiel, an influential Jew in the court of Alphonse VI of Castille (King of the Cid); ibn Ferrusiel is arriving at the city so the population goes to greet him with musical instruments, singing and such. So the event is situated on the Northern side of the border but definitely not in Toledo but in another city. We are here either contemporary with Sir Rodrigo, The Cid, or a few years after his death as the brief chronology shows:

And this is the time to note that the nickname of Yosef ibn Ferrusiel is Little Cid [5].

In Example 1 the addressed is an al-Hassan David ibn al-Dayyan, who may be either Muslim or Jew from the name. In Example 1 the girl calls Ibn-al-Dayyan "cydy", so "sidi", cca. "lord"; in Example 3 Yosef ibn Ferrusiel is "Cidiello", i.e. Little Cid. Both text is a very Spanish-like Neo-Latin matrix with some Arab words, written with Arab letters. Although so the vocalism is obscure, verbal declension and prepositions seem correct.

Now we do know a language spoken in Arabic Hispania, which was a conservative Neo-Latin language but enhanced with Arab cultural words, called so Mozarab [6]. Its vocalism was conservative, but short stressed [e] & [o] went into diphtong even before [j], intervocal voiceless stops, some initial consonant groups and initial F- remained unchanged. In Spanish (cca. already in the same time) short stressed [e] & [o] never become diphtongs before [j], the initial groups cl-, pl- and fl- gave ll- which is cca. [lj], the intervocal voiceless stops became voiced, and the initial F- became lost.

It seems as if the Mozarab evolution were rather Italian- than Spanish-like. The more conservative phonetics suggest less sub/superstrate influence and the retaining of F- has an obvious explanation. The loss of F- in Spanish and Gascon is routinely explained with a Proto-Basque substratum (observe the same name in Gascon than in Basque), considering that Basque is against initial F-. Now, you cannot expect too much Proto-Basque substrate in Southern Hispany ("Andaluz"); that territory from IIIrd c. BC was Phoenician & Greek, while on the North Basque had much more territory than now.

And there is a difference also in superstrate. Vandals ("Andaluz") were there only for decades. Then Western Goths arrived to the whole peninsula. However they were Arians, so they could not mix for a while with the Catholic Romans. Then came the Berbers & Arabs in 711, when the army was still dominated by Germanic Goths.

At this point I speak as an expert of Muslim occupations. Note that Hungarian state and Magyar language survived the occupation (the state of Hispany did not), so I can tell a whole story. If the army cannot stop the invaders, the professional warriors evacuate the occupied territory. In this way they lose the land property (even this was not exactly true in Hungary; landowners could take a reduced tax or rent from occupied Hungary), but they can get some new lands from the ruler; while peasants have no alternative; they either stick to land or become homeless. But, contrary to Hungary in 1526, in 711 in Hispany warriors were ethnically different from city-dwellers & non-warrior farmers. So, by any chance, the Germanic superstrate evacuated the Arabic territories and went North.

And then just after 711, with serious simplification, we get the following scheme for the initial conditions of linguistic evolution:

The next 350 years are quite enough for getting two very similar but distinct Western Neo-Latin languages on the peninsula. The southern one, being substrateless as the Italian, has in general also proto-Italian tendencies. This is conform with the facts. The superstrate (of three levels, Arab>Berber>Jew) gave high culture words. As for example Cid(i)=Sidi=Lord. Interestingly enough, in a lot of kharjas habib=friend, lover. (OK, in recent Magyar úr=Mr. was a milleneum ago the title of Heir Apparent, and asszony=Mrs. was the Queen.) And note: ibn Ferrusiel is "Cidiello". The root is an Arab loanword but the ending is Neo-Latin (quite Italian-like). The word Cidiello is good Mozarab.

When Sir Rodrigo took Valencia, the territory did not went to Castille. Connections between The Cid and King Alphonse VI were rather fluid. Sometimes The Cid was banished from Castille, sometimes he was a vassal of Alphonse VI and sometimes they were allies. At the reconquest of Toledo some 23 Arab/Berber statelets existed on the South

Let us see again a chronology [7], [8]:

1080 Vassals Sir Rodrigo and Sir García quarrel. King Alphonse banishes Sir Rodrigo and confiscates his lands.

1081 Sir Rodrigo with a few followers turns into a condottiere. County of Barcelona would not take him.

1082 So he takes service in Muslim Zaragoza against Castille.

1083 State of Zaragoza splits into 2.

1085 Alphonse takes Toledo and starts the siege of Zaragoza.

1086 Muslim statelets appeal for help from Morocco. Almoravids come in and defeat Alphonse at Deep Southern Badajoz. Aragonese forces join the siege of Zaragoza.

1087 Agreement of King Alphonse & Sir Rodrigo: the latter can hold all Muslim lands he takes. Zaragoza is taken; Alphonse VI sends there Sir Rodrigo as governor. Peace: Emir al-Mustain gets back the city as vassal of Alphonse VI.

1088 Siege of Aledo.

1090 Almoravids start to put down leaders of statelets Granada & Málaga.

1091 Almoravids put the end to Almería & Seville.

1093 Almoravids take Badajoz. Sir Rodrigo already has Zaragoza.

1094 Sir Rodrigo takes Valencia. Then he finds "good administrators from among the Mozarabs residing in his states" [8]; defence starts against Almoravids.

1099 Death of Sir Rodrigo; his successor is his wife.

1102 Castilian troops help the evacuation of Valencia; Lady Ximéné goes to Castille.

1110 Almoravids take Zaragoza.

The detailed story is much more complicated (note that in many cases Muslim emirs were vassals of Christian lords and so on) but even this suffices; now we can see that the many states of the peninsula have quite various goals & supporters. We get roughly the following scheme:

And then the picture is clear enough to conclude.

The scholars repeat endlessly that Sir Rodrigo got a Muslim title as Cid; see e.g. [7]. Other sources tell that the title is Spanish Arab (e.g. [8]). However there was another ethnic and another language which can be the source as well.

Imagine that you are a Mozarab in 1095 somewhere in mid-Hispany. Then you are a city dweller (farmers are mainly Muslim, at least Neo-Muslim), and mainly Christian, maybe Jew. (Although Mozarab poetry was also consummed by Muslims.) Either you live in or around Toledo (Alphonse VI), or you live in a northern Muslim city (say, Badajoz), or maybe you live in a statelet of Sir Rodrigo Diaz de Vivar, somewhere from Zaragoza to Valencia. Your goals, hopes &c. will be determined by your inherent situation (Mozarab) but also on your political one.

Assume that you live in the Toledo region. Situations turned better a few years ago as Alphonse VI liberated you. But you are moderately rich compared to Northern wildmen, and how will you keep your position? King Alphonse gets his main support from Spaniards. You have the good chance to be robbed in a few years, or, if better, heavily taxed, or, if better, suppressed by concurrents as proteges of the King &c. Just now life is good (we see that Yosef ibn Ferrusiel, the Little Cid has influence even without being Catholic); but for any case you are a Mozarab while the Spaniards liberated you, and already their grand-grandfathers were bleeding for the Northern and Spanish King. Your language is strange for them (albeit understandable); you pronounciate such an impossible and revolting sound as an initial F. Did you learn it from the pagan Muslims?

Assume that you are a Mozarab in Badajoz. Your city was just taken by some Moroccan fanatics. Maybe they robbed you; maybe they will. Maybe Alphonse VI will liberate Badajoz (henceforth as in the previous paragraph). OK, your situation will be better. Maybe now you are in the fourth layer of society and after liberation you will be in the second (first will be the Spaniards, whose grand-grandfathers already...). If during the siege an angry fanatic does not kill you; and if just after the siege an angry Spaniard does not rob and beat you. Briefly, a liberation would be good on long range, but not something you fanatically expect with unequivocal happiness. On the other hand, if Sir Rodrigo's army liberates you? There are Spaniards in his army, but those Spaniards are already familiar with the Mozarab language; there are Muslims too and the Muslims of the Catholic boss will not harm the Mozarab Catholics; and maybe there are Mozarabs too. Long live the Cid (will you shout in your own language, not in Arab if already they will be marching in, but even until that you can daydream)!

And if you are a Mozarab in the Cid's territory? Why, he is your Cid, and you are important for him. He has Spaniards in his army but not in his administration. You may have doubts in the Castillans, but he has too. Long live your own Boss! In Mozarab: Vivad el Cid!

Once more: there were two languages in the region where Cid meant Lord, Leader or such. There were two populations as candidates for conserving orally the Cid epics. Why should we guess just Muslims singing Sir Rodrigo's deeds if Mozarabs were much more interested in his success?

His success did not come, although there were chances. Consequently the Mozarabic language is extinct, except in some places in liturgy (and maybe at some places in Morocco; 1967 statistics knew about 80,000 autochtonous Catholics there, maybe some the descendants of Hispanic Catholics, maybe not). OK, maybe Mozarab and Spanish were too close for surviving both. But: it seems that the total Northern victory may have had some connection with the anomalous social situations just after as well.

[1] Z. Batzarov: Mozarabic Kharjas. http://www.orbilat.com/Modern_Romance/Ibero-Romance/Mozarabic/Mozarabic-Kharjahs_02.htm

[2] J. Kollár: Národnié zpiewanky. Král'ová Universická Tiskárn'a, Buda, 1834.

[3] L. Moretti: Ricerche sulle leghe greche. Rome, 1962

[4] R. A. Hall: Pidgin and Creol Language. Ithaca, 1966

[5] ***:Jarchas mozárabes. http://faculty.washington.edu/petersen/462/jarchas.htm

[6] Z. Batzarov: Mozarabic Language. http://www.orbilat.com/Modern_Romance/Ibero-Romance/Mozarabic/Mozarabic.htm

[7] L. H. Nelson: Some Thoughts on the Song of the Cid. http://www.ku.edu/kansas/medieval/108/lectures/cid-info.html

[8] J. Sedycias: História da Língua Espanhola. http://home.yawl.com.br/hp/sedycias/historia11a.htm

My HomePage, with some other studies, if you are curious.