Spanish "el", Arabic "al-"—why both?

[Cenzontle]

Medieval speakers of Castilian or Mozarabic adopted many Arabic nouns with the article "al-" attached ("el alcalde", "el almacén", "el algodón"...).
To us, it seems obvious that they were repeating the article. Why wasn't it obvious to them?

 

[fdb]

Either they did not realise that “al” is an article (in English we also say “the Alhambra” or “the Almagest”), or else they felt that the function of the Spanish article was different from that of the Arabic article. In Spanish you say “El alcalde de Zalamea”, but in Arabic you say “qāḍī madīnati X” without the article.

 

[Epilio]

According to the dictionary of the Spanish Academy the word alcalde (major) comes from the Spanish Arabic alqáḍi that in turn derives from the Classical Arabic qāḍī. It seems that the Mozarabs took the word from the dialect spoken in the peninsula, and not from the Classical version. For some reason the Spanish Arabic used more extensively the article.

 

[fdb]

No, you have misunderstood this completely. qāḍī is the form used in a possessive construction (construct state); al-qāḍī is the determined form. This is true both in classical Arabic and in all dialects.

 

[Abu Rashid]

Either they did not realise that “al” is an article (in English we also say “the Alhambra” or “the Almagest”), or else they felt that the function of the Spanish article was different from that of the Arabic article. In Spanish you say “El alcalde de Zalamea”, but in Arabic you say “qāḍī madīnati X” without the article.

It's a bit different when the article is effectively the same. It would be the equivalent of British English adding the to the phrase "The grand canyon", so it became "The the grand canyon". Of course these are two varieties of the same language, so it's not quite the same.

 

[Cenzontle]

If "al" and "el" were not so similar in form, I would not have asked the question.
Sometimes I see "the Rio Grande River", and that doesn't bother me (much).

 

[fdb]

I understand. The old Spanish did at least write “el” and “al-” with different vowels, so they clearly did not hear them as “effectively the same”, as my esteemed friend Abu Rashid puts it. The “Rio Grande river” is indeed a nice parallel. As is “the Guadalupe river” (guada = wādin = river).

 

[إسكندراني]

Wasn't Castillian Spanish introduce as a somewhat constructed language in the first place? That would make the way it came into being very different from what would have been the natural scheme of things.

 

[Cenzontle]

Wasn't Castillian Spanish introduce as a somewhat constructed language in the first place?

I would be very surprised if that claim were made. Can you say more about it?

 

[Cenzontle]

I meant to say, I assume that many Mozarabs were bilingual, and therefore understood the grammatical function of Arabic "al-".

 

[rbrunner]

I meant to say, I assume that many Mozarabs were bilingual, and therefore understood the grammatical function of Arabic "al-".

I tend to agree. The following theory, formulated in a slightly exaggerated way to make it clear, is hard to believe for me: In a land with intensive contact with Arabic and Arabs, and with many Arabic-speaking residents and bilinguals, people were not aware of one of the most basic facts of Arabic - that "al-" is an article - and in their ignorance and innocence borrowed the words with article attached.

I would be very surprised if that was already the whole (and therefore quite un-interesting) story.

By the way, are there any known counter-examples, i.e. Arabic words borrowed into Spanish without article?

 

[Hulalessar]

It all does seem a bit odd. The scale of borrowing presupposes that those who borrowed had more than an elementary knowledge of Arabic and therefore would have known that al was the article. It is though possible that it was Arabic speakers of Spanish who introduced the words and that the article remained attached because it was felt to be an integral part of the word. In any list of Arabic words in Spanish there are more words beginning with "a" than any other letter. The absence of "l" in many cases ( as in aceituna) is explained by the assimilation of the "l" in the following consonant if it is a "sun" letter. However, there are many nouns not beginning with "a", though in some cases that can be explained by the assimilation of sun letters because where that happens the vowel of al tends to disappear. That still leaves quite a few words beginning with moon letters that appear without "al-" and one has to wonder why some words have it and some do not.

It is interesting to note looking at a list of Arabic words in Spanish the almost complete absence of verbs.

 

[fdb]

It is interesting to note looking at a list of Arabic words in Spanish the almost complete absence of verbs.

I think you will find that this is the case with loan words in most languages.

The Indo-European language that has borrowed most heavily from Arabic is of course Persian, and here nouns were borrowed as a rule without their article (except in a very small number of fixed phrases). In Persia, Arabic was taught in schools for more than a thousand years, so literate Persians were perfectly aware of the rudiments of Arabic grammar, and I would imagine that the same was the case with the Muslims in al-Andalus. But perhaps words of the alcalde type were borrowed at an oral level, presumably by Christians, who had not had an Islamic school education and were consequently not aware that these words contained an article.

 

[Abu Rashid]

The “Rio Grande river” is indeed a nice parallel. As is “the Guadalupe river” (guada = wādin = river).

Add to that the "Sahara desert" (ie. the desert desert).

Might it be the case that because the article is attached to the noun in Arabic that it was taken to be part of the word?

 

[Hulalessar]

Might it be the case that because the article is attached to the noun in Arabic that it was taken to be part of the word?

It depends what you mean by "attached". The article is certainly attached in writing, but the question is whether the article is attached in writing because it is felt to be part of the word, or whether it is felt to be part of the word because it is attached in writing.

 

[Ben Jamin]

Medieval speakers of Castilian or Mozarabic adopted many Arabic nouns with the article "al-" attached ("el alcalde", "el almacén", "el algodón"...).
To us, it seems obvious that they were repeating the article. Why wasn't it obvious to them?

You make the presumption that the Mozarabs, who were bilingual, added the “el” to definite Arabic nouns, but what makes you believe that? Isn’t it more plausible to think that the Castilian article was added after the speakers forgot Arabic?

 

[francisgranada]

There are also words like alcohol or alchemy, which are medieval Latin borrowings from Arabic, so it is not exclusively a Spanish "phenomenon" to borrow the Arabic nouns with the article.

 

[Cenzontle]

are there any known counter-examples, i.e. Arabic words borrowed into Spanish without article?

Good question!
Lists of Spanish words of Arabic origin exist online, such as the Wikipedia list.
(Caution: In spite of the introduction's statement that "this list is relatively restrictive", it contains many words that Corominas does not attribute to Arabic.
Etymologists of Spanish can range from the caution of a Corominas to a tradition of "when in doubt call it Arabic".)
In looking for Arabic nouns that entered Spanish in the Middle Ages and which could have brought "al-" with them, but did not, you have to eliminate
(1) those not documented before 1492 (the expulsion of the Moors)—many of them being later, learned borrowings, and
(2) those beginning with "sun letters"—
apical consonants that bring about the assimilation of the "l" of "al"—e.g. 'oil' spelled al-zait but pronounced as if az-zait, for Sp. aceite).
Whether or not these sun-letter words entered Spanish with the vowel of "al-" is a different question, as Hulalessar points out.
So our search for examples is left with words beginning with the "moon letters" b, ca, co, cu, f, g, h, m, q, v.
The modern Spanish j corresponds mostly to an "sh" sound in Old Spanish, which is spelled with the sun letter shīn in Arabic, so Sp. jarabe 'syrup' (< Ar. sharāb) is not a telling example, and jaquequa 'migrane' (< shaqīqa) has two counts against it: the sun letter shīn and late documentation ("toward 1500" in Corominas).
Corominas traces jinete 'horseman' to an ethnonym transliterated with "Z-".
We are left with very few candidates, possibly including the following:
• barrio "neighborhood"
• bellota "acorn"
• jarra "jar" (with the Arabic "moon letter" jīm, not the sun letter shīn).
With some labor, perhaps a few more could be extracted from the Wiki list. Meanwhile, the list contains some 375 with "al-".
Ben Jamin's question

Isn’t it more plausible to think that the Castilian article was added after the speakers forgot Arabic?

is intriguing. I have to give it some more thought and try to imagine the concrete process.
As francisgranada points out, other languages borrowed words like alcohol and alchemy with the article—but they can be "forgiven" for not recognizing "al-" as an article.
The question remains for the Castilian-speakers, who had "el" in their language.

 

[fdb]

If you want to pursue this matter seriously you need to get the book by the Arabist Federico Corriente, Diccionario de arabismos y voces afines en iberorromance. (2nd expanded ed.; Madrid 2002) .

 

[rbrunner]

Isn’t it more plausible to think that the Castilian article was added after the speakers forgot Arabic?

Do you mean that maybe there was a time where people used Arabic articles for Arabic words in speak, and of course Spanish articles with Spanish words. Then, after some point, with knowledge about the Arabic words and their origin slowly degrading, people "lost" the ability to correctly use the the Arabic articles this way, permanently attached them to the words and added Spanish articles, because clearly the words needed (new) articles.

That would be a very interesting little tidbit about a particular stage of the development of Spanish, if true of course...

 

[Cilquiestsuens]

But perhaps words of the alcalde type were borrowed at an oral level, presumably by Christians, who had not had an Islamic school education and were consequently not aware that these words contained an article.

Most plausible explanation. I don't find this assimilation of a definite article in the noun itself so surprising. A few French words went through the same process. l'ierre > le lierre l'ingot > le lingot What is remarkable about these two words, if you'd bother give it a thought, is that this change apparently happened among native speakers, without any kind of foreign linguistic interference. Sounds to me more unlikely to happen than in the case of a borrowing from another language, in which you can expect some kind of confusion to take place. P.S.: As a native speaker of a romance language, I tend to think that to a Spanish speaker, these words just sound way better with -al in front of them.

 

[killerbee256]

Add to that the "Sahara desert" (ie. the desert desert).

Might it be the case that because the article is attached to the noun in Arabic that it was taken to be part of the word?

Another American example is "the la brea tar pits" "the the tar tar pits"

 

[Ben Jamin]

Do you mean that maybe there was a time where people used Arabic articles for Arabic words in speak, and of course Spanish articles with Spanish words. Then, after some point, with knowledge about the Arabic words and their origin slowly degrading, people "lost" the ability to correctly use the the Arabic articles this way, permanently attached them to the words and added Spanish articles, because clearly the words needed (new) articles.

That would be a very interesting little tidbit about a particular stage of the development of Spanish, if true of course...

Yes, you have made the hypothesis more precise. I myself can observe the way immigrants use Norwegian words with or without definite postposition (-en and-et) while speaking their own language. Those proficient in Norwegian use the words correctly in definite or indefinite form, according to the role of the word in the sentence, the non proficient people use them haphazardly.

 

[francisgranada]

Finally, the word Allah (from al Ilah, "the God", as far as I know) is also an example for the situation when the article becomes part of the noun.

I don't speak Arabic, but I can imagine also other Arabic words or terms where the article is (almost) "unseparable" from the noun, or at least, nouns used so frequently with article that this seems to be part of the noun for someone who doesn't speak Arabic. If so, then the Arabic words with the article "al" were adopted primarily by those Romance-speaking people who didn't speak Arabic.

For curiosity, in the so called kharjas (jarchas) we find some Arabic words both with and without the article. I think, the kharjas were written by people who spoke both Romance and Arabic (or/and Hebrew) and thus they might knew where to use the article and where not. For example:

Tanto amare, tanto amare, habib, tant amare!
Tan mal meu doler li-l-habib! Enfermo yed, cuánd sanarád?
Qué faré, mamma? Meu al-habib est' ad yana [puerta].

 

[Abu Rashid]

Finally, the word Allah (from al Ilah, "the God", as far as I know) is also an example for the situation when the article becomes part of the noun.

If this is a case of this, then it would probably be a solitary case. Also Allah cannot be prefixed with the definite article, so we'd never see doubling up of the article with this case.

 

[francisgranada]

If this is a case of this, then it would probably be a solitary case. Also Allah cannot be prefixed with the definite article, so we'd never see doubling up of the article with this case.

Yes, (hope) I understand. It's only an example, even if not the best, of course, because Allah is an excepcion from many points of view. For curiosity, see also an identical example in Italian: Iddio (from "il Dio" = "the God").

But, at least in theory, I can imagine a "doubled article" in a non-Arabic language, saying "the Allah". The reason for not using the "double article" in various (foreign) languages with the word Allah is due to the fact that Allah is intended as the name of the God (i.e. a personal name) and not as a generic word/term for "God". This is true not only for English, but e.g. also for my mother tongue: Allah is used without the article, but surely not because a "common" Hungarian person would be able to analyze the etymology of the Arabic word Allah ...... (at the same time, the world Isten in Hungarian is used with the definite article: az Isten = the God)

 

[Giorgio Spizzi]

Kids in pre-school age don't know how to write, therefore they can't say where the boundaries between words are located.
One classic example of the assimilation of the definite article in Italian we have in the case of names: My friend's name is Isotta, but as we in the north of Italy normally use a definite article in front of feminine first names, what her nephew hears is inevitably "L'Isotta"
/li'zɔt:a/. Consequently he will ask, for example "Dov'è la Lisotta?, Where is Lisotta?".
My father made the same mistake when naming his sister-in-law Isa, whom he would call Lisa (ie, L'Isa), which happens to be a real name, too.
My father, and many of his generation, used to be convinced that "the radio" (in Italian "la radio) were in fact "l'aradio". To make things worse — considering that the Italian word "radio" ends in an "o", which is typical of masculine nouns — he would stubbornly say things such as " Questo aradio è vecchio, bisogna comprarne uno nuovo" (This radio's old: we should by a new one) instead of the correct "Questa Øradio è vecchia, bisogna comprarne una nuova".
I could go on with innumerable examples.

GS

 

[Hulalessar]

Kids in pre-school age don't know how to write, therefore they can't say where the boundaries between words are located.

Older ones too sometimes. I asked a Spanish boy of about ten what his favourite food was. He said it began with "s". I tried every Spanish food item I could think of beginning with that letter and had to give up. The answer was "lasaña"!

There are a few words in English which used to begin with "n" but which shifted to the indefinite article viz: apron and adder.

 

[francisgranada]

An other example is the Hungarian word "lárma" (storm, noise, rumour ...) that comes from the Italian "all'arma" (lit. "to the weapon", finally also the origin the English "alarm") where the initial "l" (definite article in Italian) is intentended as part of the noun "arma" (weapon). I.e. spontaneousely analyzed as if it were "a larma".

 

[merquiades]

In looking for Arabic nouns that entered Spanish in the Middle Ages and which could have brought "al-" with them, but did not, you have to eliminate
(1) those not documented before 1492 (the expulsion of the Moors)—many of them being later, learned borrowings, and
(2) those beginning with "sun letters"—
apical consonants that bring about the assimilation of the "l" of "al"—e.g. 'oil' spelled al-zait but pronounced as if az-zait, for Sp. aceite).
Whether or not these sun-letter words entered Spanish with the vowel of "al-" is a different question, as Hulalessar points out.
So our search for examples is left with words beginning with the "moon letters" b, ca, co, cu, f, g, h, m, q, v.
The modern Spanish j corresponds mostly to an "sh" sound in Old Spanish, which is spelled with the sun letter shīn in Arabic, so Sp. jarabe 'syrup' (< Ar. sharāb) is not a telling example, and jaquequa 'migrane' (< shaqīqa) has two counts against it: the sun letter shīn and late documentation ("toward 1500" in Corominas).
Corominas traces jinete 'horseman' to an ethnonym transliterated with "Z-".
We are left with very few candidates, possibly including the following:
• barrio "neighborhood"
• bellota "acorn"
• jarra "jar" (with the Arabic "moon letter" jīm, not the sun letter shīn).
With some labor, perhaps a few more could be extracted from the Wiki list. Meanwhile, the list contains some 375 with "al-".
Ben Jamin's question

is intriguing. I have to give it some more thought and try to imagine the concrete process.
As francisgranada points out, other languages borrowed words like alcohol and alchemy with the article—but they can be "forgiven" for not recognizing "al-" as an article.
The question remains for the Castilian-speakers, who had "el" in their language.

I have an intriguing word for your list: the word for Eggplant, apparantly Albadinjan in Arabic. It goes into Spanish (Berenjena) and Portuguese (Berengela) this time without the Al, but in Catalan (Albergínia) and French (Aubergine) the articles was kept.

I think in general since Al needs to be used in most cases in Arabic a Spanish speaker just adopted the loan word in a given context unaware of the grammatical components, just like English speakers when they say "the alcohol". I don't think the type of people who took these words into the Spanish language and created its future were skilled Arabic scholars or even bilingual speakers. Reminds me of cowboys in contact with Mexicans in the US Far West. They went as masters, didn't speak Spanish or have any desire to learn it. They heard juzgado and adopted it as Hooscow jail. The Castilians picked up a few words while they were assimilating and/or expelling Arab-speaking people.

 

[Cenzontle]

Thank you, merquiades, for "eggplant" and for your thoughts about the borrowing process. I think Spanish "berenjena" would indeed qualify as an article-less borrowing (its date of first documentation in Corominas is "princ. S. XV"—beginning of the 15th century). You said...

I think in general [...] a Spanish speaker just adopted the loan word in a given context unaware of the grammatical components, just like English speakers when they say "the alcohol". I don't think the type of people who took these words into the Spanish language and created its future were skilled Arabic scholars or even bilingual speakers.

I can't agree that Spanish-speakers adopting Arabic nouns were "just like" English-speakers saying "the alcohol". My original question was based on the fact that Spanish-speakers had "el" in their language (both grammatically and phonetically similar to Arabic "al-"), while English-, French-, etc. -speakers did not. Not to mention the fact that the medieval Spanish-speakers lived closer to—if not in bilingualism with—the source language. I haven't come across any studies investigating, or even speculating about, the role of bilingualism in Arabic-to-Spanish borrowing. My own speculation has been that the Mozarabs needed to know Arabic in order to get along with the dominant society and its institutions, and that they may have borrowed Arabic into their Romance speech almost in the manner of code-switching. Look how Arabic words—including grammatical particles—are sprinkled in the otherwise-Romance Mozarabic kharjas: "Ya nuemne dolje" ("O sweet name"), "In non, si non keris" ("If not, if you don't wish to"), "Ke farey, yaummi?" ("What can I do, O Mother?"). This doesn't come from "skilled Arabic scholars", but rather from vernacular Romance-speakers with fairly intimate knowledge of Arabic. And yet, in one of the kharjas quoted by francisgranada, we get...

Meu al-habib

...whose inclusion of the article after the possessive determiner seems also to be ignoring the article function of "al-".
Where was the breakdown in understanding that "al-" = "el"?

(The other half of my speculation is that the reconquering Castilians, et al. found it easy to borrow vocabulary from their fellow-Romance-speakers, the Mozarabs, rather than from their "enemies" the Moors.)

For an analogy in English, we would need a loanword from a language whose article sounds like "the", such as Dutch "de". A famous gate in Maastricht has the name "Helpoort". (I'm getting far outside my expertise here.) Being a unique entity in the world, I imagine the Dutch would call it "de Helpoort". But I can't imagine English-speakers calling it "the de Helpoort".

 

[Hulalessar]

My own speculation has been that the Mozarabs needed to know Arabic in order to get along with the dominant society and its institutions, and that they may have borrowed Arabic into their Romance speech almost in the manner of code-switching.

Whilst there may have been some continuum between them, Castilian and Mozarabe were two different things. "Islamic Civilisation" was present in the Iberian for some seven centuries. It would be surprising if Arabic had not had at least some influence on the Romance languages of Iberia. Whatever the territorial limits of Al-Andalus may have been at any given moment (and there were not always clear cut boundaries because with shifting alliances there were periods when Muslim and Christian were aligned against Christian and Muslim) commerce and other interaction must have taken place between the two communities. Those engaged in commerce (especially on the Christian side - we are talking about a time when a woman who could read was thought a disciple of the Devil) were not necessarily literate. Universal literacy is only a quite recent phenomenon in Europe. As mentioned above, the illiterate or semi-literate (whether children or adults) sometimes confuse word boundaries in their own language. They are much more likely to do so when picking up words in the market place. A Castilian speaker selling his produce may have gone home and said to his wife: "It's surprising how many vegetables in Arabic begin with al-".

 

[Hulalessar]

By way of an imperfect analogy consider the difference between English and French. In English an article may be required if a noun is a count noun, but not if it is not a count noun. In French a non-count noun may require "de" to be used. So, an English speaker pointing respectively to a dog, cat, water and rice and asking a French speaker what they are will get the answers: "C'est un chien", "C'est un chat", "C'est de l'eau" and "C'est du riz". He may, not unreasonably and assuming French follow English, conclude that in French "cé" means "it's", "tun" means "a", "chien" means "dog", "chat" means "cat", "dleau" means "water" and "duriz" means" rice".

 

[Forero]

I assume zanahoria was deemed long enough without an al-.

 

[rbrunner]

What about Andalucía? The name of this region in Spanish is not Alandalucía, and the name of the Muslim state seems to have been written as Al-Ándalus, with a hyphen between article and proper name, which could show awareness about the article.

 

[francisgranada]

... ...whose inclusion of the article after the possessive determiner seems also to be ignoring the article function of "al-"

Not necessarily. If we suppose that the cited phrase "Meu al-habib est' ad yana" in medieval Spanish (Romance) required the article (*el meu amore est' ad yana), then the article "al" could be used consciousely. It's position here is given by the usage of "habib" instead of "amore", i.e. the adequate Arabic article precedes directly the Arabic noun. *Al meu habib should be indeed a bit strange (and also ambiguous), as the Arabic article was used only with Arabic nouns and not as a separate loanword in Spanish.

On the other hand, in the phrase "Tanto amare, tanto amare, habib, tant amare!" the noun "habib" seems to be in vocative function, thus without article.

What about Andalucía? The name of this region in Spanish is not Alandalucía, and the name of the Muslim state seems to have been written as Al-Ándalus, with a hyphen between article and proper name, which could show awareness about the article.

Andalucía seems to me a conscious derivation from the Arabic and not a "spontaneous" loanword.

 

[Hulalessar]

It is not necessarily the case that Al-Andalus is of Arabic origin. See here: http://en.wikipedia.org/wiki/Al-Andalus#Etymology

 

[fdb]

the name of the Muslim state seems to have been written as Al-Ándalus, with a hyphen between article and proper name, which could show awareness about the article.

There is no hyphen in Arabic. The article is always written as part of the following word. The hyphen comes with the European transcription of Arabic.

 

[rbrunner]

There is no hyphen in Arabic. The article is always written as part of the following word. The hyphen comes with the European transcription of Arabic.

Ok, understood. Anyway, whoever transcribed it was aware about the article. But I admit that for the purpose of this discussion that's not a strong argument because the broad mass of the "normal" Spanish-speaking people interests far more than the people who wrote and transcribed things.

 

[xari]

Is the "el" and "al" similarity actually relevant to the issue though?

Many (most?) of Spanish words of Arabic origin are also present in Portuguese and we have quite different articles (o and a).

 

[Forero]

Is the "el" and "al" similarity actually relevant to the issue though?

Many (most?) of Spanish words of Arabic origin are also present in Portuguese and we have quite different articles (o and a).

What were the Portuguese articles during the period in question? Were Spanish articles still elo and ela?

 

[إسكندراني]

You make the presumption that the Mozarabs, who were bilingual, added the “el” to definite Arabic nouns, but what makes you believe that? Isn’t it more plausible to think that the Castilian article was added after the speakers forgot Arabic?

This must surely be the case. It is unheard of that anyone with even the faintest exposure to Arabic grammar would double the article. This phenomenon must have appeared after the Reconquista, since Arabic was banned in that period. This also explains the absence of verbs. Broken nouns casually maintained as loanwords are relatively tolerable to the reconquistador - especially if their article and pronunciation were corrupted (this happened to place names as well, though I admit that I still find the variety of Spanish regional pronunciations rather bewildering).

 

[Outsider]

It seems to me that redundancy with proper nouns or set phrases is fairly common in many languages. An example in English: "ATM machine" -- the "M" already stands for "machine", but even native speakers are so used to the acronym that they often forget that. This is even more common with loanwords. Newspapers everywhere have been talking about "Al Qaeda" for the last ten years, with no regard for the fact that the "al" is actually an article, not strictly a part of the noun. And then there are all those cities whose name in foreign languages includes the article in the original language: Oporto ("o" is an article), The Hague, Los Angeles, and so on.
Another factor is that in Arabic the definite article often assimilates with the noun. That probably tends to make people reanalyse the article as an integral part of the noun.

 

[rbrunner]

Newspapers everywhere have been talking about "Al Qaeda" for the last ten years, with no regard for the fact that the "al" is actually an article, not strictly a part of the noun.

Well, yes, but I see a big difference to the situation in Spain: Western newspapers and their Western readers are nowhere as close to or even surrounded by Arabic like the Spanish were. Of course the more distant from a language you are the less you know about it and the more you tend to throw articles and words together, not knowing better.

This phenomenon must have appeared after the Reconquista, since Arabic was banned in that period.

May well be that the Reconquista is an important part of the puzzle: Maybe because of it Arabic went out of favor pretty quickly, but the loanwords were already established enough to survive, but without the supporting knowledge of Arabic also quickly assimilated their articles.

 

[xari]

Is it possible that the "al" was mistakenly perceived as a case marker instead of article?

 

[francisgranada]

... This phenomenon must have appeared after the Reconquista, since Arabic was banned in that period ...

I do not agree. It seems to me much more logical that the Arabic loanwords should have been borrowed before the Reconquista, when the co-existence of Arabic and Spanish (Hispanic Romance) was intense enough (borrowings do not "happen" during a night). It is hardly imaginable that the reconquistadores would have "re-attached" the Arabic article to already existing loanwords ... In other words, I think, the Arabic loanwords had to be adopted in Spanish (Hispanic Romance) together with the article al from "the very beggining", i.e. since they had been spontaneousely borrowed, and then they have survived until today.

It is unheard of that anyone with even the faintest exposure to Arabic grammar would double the article.

I agree, but where is it written that all the Spanish (Hispanic Romance) speaking people had "even the faintest exposure to Arabic grammar"? ...

In linguistically mixed regions it is absolutely not obvious that all the people know or speak both (all) the languanges spoken in the region. For example, in my country, there live Hungarian-speaking and Slovak-speaking persons side by side, in the same village/town/street (or even in the same building or family) and the majority of Slovaks have no idea about the Hungarian grammar, articles, etc ... (practically, about the Hungarian language) at all. At the same time, they use Hungarian loanwords in their everyday speech. (For correctness: the opposite may also happen, but it is not typical today, as the Slovak is the official language. But some decades ago it was not so rare to meet older Hungarian persons which didn’t speak Slovak at all). In other words, the co-existence of two languages at the same place/village/town/region does not automatically lead to bilinguism (especially not in case of different cultural traditions, religion, writing system, etc …)

 

[Ben Jamin]

I can imagine a following process: the bilingual inhabitants under Arab rule used the Arabic article in the loanwords correctly. The newcomers that came with the victorious Reconquista armies learned those words from them and began to use the words indiscriminately with the article. The bilinguals died out in a generation or two.

 

[Outsider]

Well, yes, but I see a big difference to the situation in Spain: Western newspapers and their Western readers are nowhere as close to or even surrounded by Arabic like the Spanish were. Of course the more distant from a language you are the less you know about it and the more you tend to throw articles and words together, not knowing better.

But my main point was that redundancies can crop up even in one's native language, let alone in a bilingual situation, no matter how closely the two languages may have been in contact.

I agree that most of these words must have been borrowed - with the article - prior to the Reconquista, and by people with at least some passive knowledge of Arabic.

Is it possible that the "al" was mistakenly perceived as a case marker instead of article?

I find it unlikely since neither Portuguese nor Spanish (nor Mozarabic) have cases, but I don't know enough about cases in Arabic to form an opinion. Can you explain your idea better?

 

[merquiades]

I can imagine a following process: the bilingual inhabitants under Arab rule used the Arabic article in the loanwords correctly. The newcomers that came with the victorious Reconquista armies learned those words from them and began to use the words indiscriminately with the article. The bilinguals died out in a generation or two.

I think that's the crux of the matter, Ben Jamin. This map I posted would point to that too.

 

[Quiviscumque]

"After the Reconquista"??? What date is that? Just at random, here is a vulgar Latin text from 922 (near Burgos):

deinde in villa que nuncupant Cavia, que est sita in alfoz de Munno[...]

Believe me, the Reconquista was not yet finished in 922 A.D.

The question posed by this thread has been recently discussed by scholars. You can read

http://www.uni-muenster.de/imperia/md/content/romanistik/noll/noll-art.pdf

I am a complete, absolute, total ignorant about these issues, but it seems that scholars consider some features of Andalusian Arabic as the source of these "al-" Hispanic words.