How would a Romance Language with Arabic as a parent evolve?

Say the Romans or a Latin speaking people conquered and culturally assimilated Arabia or an Arabic speaking country in much the same way as Iberia and so on, to the point where after the fall of Rome or this hypothetical polity, a Romance language was appearing with Arabic as its other parent.

Using other languages as an example, how would this new language sound like? What would the influences be? Would the [p] or the [q] phoneme be in it, or both?

If this isn't the best place for this, is there a more appropriate sub-forum to be discussing the topic? or is there an alternate languages forum?

I'm guessing it would be just like Maltese

Farfromhome said:

I'm guessing it would be just like Maltese

Click to expand...

Maltese is a Semitic language in the Siculo-Arabic group, its main parent is Arabic as opposed to latin/italian. I am proposing a Romance language, where the main parent is vulgar latin or church latin as opposed to Arabic.

Not quite, since Maltese is a Semitic language with a ton of Romance influence (much like English is to other Germanic languages, only even worse than that). The OP asks for an Arabic-influenced Romance language. Something like Mozarabic would be closer to what is asked for, I'd think.

EDIT: NINJAAAAA'D! But yeah, the Mozarabic dialects would be a good start (esp. since it retained the Latin "f" instead of Spanish "f>h" mutation), albeit written in Latin script instead of Arabic abjad.

Most interesting is question of pharyngealization in Arabic.

You mean heavily latinized Arabic language or arabized Romance ?

Chrzęścimir said:

Most interesting is question of pharyngealization in Arabic.

You mean heavily latinized Arabic language or arabized Romance ?

Click to expand...

I actually already made that clear in the OP, a Romance language with arabic as a parent.

Mozarabic from what I understand is actually more latin visigothic than arabic influenced, at least according to wikipedia.

Chrzęścimir said:

Most interesting is question of pharyngealization in Arabic.

You mean heavily latinized Arabic language or arabized Romance ?

Click to expand...

Pharyngealization is likely to be lost (as did in Maltese IIRC).
Note that, in general, the substrate languages did not influence Romance very much. There is relatively little Celtic in French, for example, and, AFAIK, little recognizable Dacian in Romanian (the same holds for English; very few words have a clear Britonic orgin, and no morphological process I know of). It is also worth noting that a similar situation is true of most (not all) spoken Arabic vernaculars: the recognizable influence of substrate languages varies, but is usually relatively limited (I know that important exceptions exist in some parts of Yemen and, probably, in the sedentary varieties of the Maghreb).

To the OP, it is also worth asking "what Arabic"? The Classical Arabic language emerged as a documented variety at a time when Romance had already likely undergone very significant internal differentiation that anticipates the more recent forms (altough this is murky because the written documents do not really bear much testimony of this).
The linguistic situation of Arabic in earlier times is only now beginning to be somewhat clarified, and there is a hell of a lot we do not know.
My expectation is that Arabic influence is likely to be primarily phonological. While, as I said, pharyngealization/velarization is likely to be lost (assuming it was the relevant articulation of the "emphatics" at the time; there's some reason to believe that an ejective emphasis was the norm into Roman times in some varieties at least) I think that the primary Arabic influence is likely to be phonological. A p>f shift is a distinct possibility; also you might see a reduction of the vowel inventory, perhaps even into a binary opposition (open/closed) if one considers the scarce phonematic distinction and frequent free variation between u and i in Classical Arabic (most spoken varieties reduced both to schwa in unstressed syllables).
A q phoneme is possible, but not necessarily likely (a realization as /g/, now common in Bedouin varieties, might have been more common in the past). The glottal stop is also up in the air - it appears to have been lost in at least some varieties (including Hijazi dialects) before Islam, and not marked in earlier ortography as such, but it re-emerges as a reflex of other sounds (primarily /q/) in many other dialects, esp. Syrian and Egyptian (sedentary, as Bedouin in those areas apparently lacked and lacks /q/).
I expect vocabulary to be largely Latin, with relatively few Arabic loanwords. The distinctive Semitic root system may survive as a apophonic system of some sort, but only as a residue - no way it remains a productive grammatical tool in a Romance context.
A interesting reflex of Arabic might be the loss a verb corresponding to "have"; Arabic lacks it, and in most Romance instead the equivalent it expanded its semantic role greatly, often becoming an auxiliary verb (the same happened in English and German, and I am fairly sure that this is strongly related with the parallel Romance development). A Romance that instead loses it (and develops an auxiliary system exclusively based on "be", or with other auxiliary forms - possibly "go" for the continuous/imperfective, like some forms of spoken Arabic?) would be interesting.
But here I am assuming a substrate Arabic that is very close to what would be known as Classical Arabic. There was plenty of related, different varieties in pre-Islamic Arabia, which are still somewhat poorly understood.

Isfendil said:

Say the Romans or a Latin speaking people conquered and culturally assimilated Arabia or an Arabic speaking country in much the same way as Iberia and so on, to the point where after the fall of Rome or this hypothetical polity, a Romance language was appearing with Arabic as its other parent.

Using other languages as an example, how would this new language sound like? What would the influences be? Would the [p] or the [q] phoneme be in it, or both?

If this isn't the best place for this, is there a more appropriate sub-forum to be discussing the topic? or is there an alternate languages forum?

Click to expand...

Well Portuguese is a Romance language, with origins on the Old Portuguese or Medieval Galician, as you prefer to call it, that originated from a Vulgar Latin dialect. And it as Arabic influences and many words have arabic origins (ex: Azulejo (az-zillij), Mesquita (masdjid), Romã (ramaan), Xarope (xarab), etc...).

Portuguese as much arabic influence having more than 800 words of arabic origin. But in the north the language as less influence, while in the southern parts the influence is stronger

A great example of an Arabic-Romance language is the Mozarabic. It descends from Latin but had huge arabic influence.

The other Iberian language also have great arabic influence, but if you are looking for a comparison language outside Iberia, the Sicilian language is also a great example having some 500 words of arabic origin.

Falecius said:

My expectation is that Arabic influence is likely to be primarily phonological. While, as I said, pharyngealization/velarization is likely to be lost (assuming it was the relevant articulation of the "emphatics" at the time; there's some reason to believe that an ejective emphasis was the norm into Roman times in some varieties at least) I think that the primary Arabic influence is likely to be phonological. A p>f shift is a distinct possibility; also you might see a reduction of the vowel inventory, perhaps even into a binary opposition (open/closed) if one considers the scarce phonematic distinction and frequent free variation between u and i in Classical Arabic (most spoken varieties reduced both to schwa in unstressed syllables).
A q phoneme is possible, but not necessarily likely (a realization as /g/, now common in Bedouin varieties, might have been more common in the past). The glottal stop is also up in the air - it appears to have been lost in at least some varieties (including Hijazi dialects) before Islam, and not marked in earlier ortography as such, but it re-emerges as a reflex of other sounds (primarily /q/) in many other dialects, esp. Syrian and Egyptian (sedentary, as Bedouin in those areas apparently lacked and lacks /q/).
I expect vocabulary to be largely Latin, with relatively few Arabic loanwords. The distinctive Semitic root system may survive as a apophonic system of some sort, but only as a residue - no way it remains a productive grammatical tool in a Romance context.
A interesting reflex of Arabic might be the loss a verb corresponding to "have"; Arabic lacks it, and in most Romance instead the equivalent it expanded its semantic role greatly, often becoming an auxiliary verb (the same happened in English and German, and I am fairly sure that this is strongly related with the parallel Romance development). A Romance that instead loses it (and develops an auxiliary system exclusively based on "be", or with other auxiliary forms - possibly "go" for the continuous/imperfective, like some forms of spoken Arabic?) would be interesting.

Click to expand...

I have bolded the items that I would like you to elaborate on, as I feel I did not understand them quite well.

Also, I think that I mean Classical Arabic in this case for the purposes of ease,

Furthermore, I would like to say that I am inexperienced, but judging from what you've said, I decided to do a really crude transformation of a latin phrase, and I would like you to tell me how I can improve it. Also, I did it in (again, very crude) IPA transcription, so if there are errors in that, please tell me.

Here is the phrase

entia non sunt multiplicanda praeter necessitatem
[IPA: ɛntiɑ non sunt multiplicɑndɑ prɑjtɛɾ nɛkɛsitɑtɛm]

Here is a really crude transformation that I'm guessing at with what you've told me. It is most certainly horribly wrong, because I am not at all well-versed in any arabic, but I'm trying to create a starting point.

[IPA: non sunt multəfləcɑndɑ alkjənɑtɑ frɑɛtur xtɑgɑ]

Would something like Sabir qualify?

Jonathan Edelstein said:

Would something like Sabir qualify?

Click to expand...

I actually found out about Sabir recently and find it quite interesting, although I'd think it would be more the parent of some kind of Neo-Roman common language, as it has way too many different parents.

I'll echo others in saying that Mozarabic fits the OP's question exactly; that is, it was a Romance language spoken in an Arab country. However, it was also extremely conservative, so maybe Arabic didn't influence it as much as it could have, had it evolved a bit faster.

I know of two Semito-Romance conlangs. Judajca was written for the Ill Bethisad world, and if I can remember correctly it uses Hebrew for the substrate level. All I can find online are brief sketches, though.

A much richer conlang is Syrunian, a Romance language written for Syria. Its main deep substrate is Aramaic/Syraic, with Arabic influence coming in to play later. It is written in Hebrew letters, for whatever reason. Syrunian has some major Semitic-like features:

• The default sentence pattern is VSO.
• The definite article is a prefix rather than a separate word (it's /h-/, from Latin hic).
• Direct objects are marked with a prefix as well (/la-/, taken from the source of the Spanish/French word "la" but behaving like a Syriac prefix).
• Three noun states (emphatic, absolute, and construct) rather than noun cases
• Two verb forms, imperfect and perfect (and the perfect forms actually descend from the Latin imperfects)

Jonathan Edelstein said:

Would something like Sabir qualify?

Click to expand...

Sabir was more of a pidgin than a real language. Something coming from it would have been basically a creole. My one real conlang is actually a creole based on Sabir, created within the Ill Bethisad world like Judajca. It develops in the Black Sea rather than the Mediterranean, so Turkic rather than Arabic becomes the main other parent. However, it does have the "two verb forms" feature that Syrunian uses, perfect and imperfect. Its perfect forms come from the perfect in Romance languages, which makes sense since the language evolves from medieval Romance rather than Latin. (You can see it via the link "Para veni sabu" in my signature.)

False Dmitri said:

I'll echo others in saying that Mozarabic fits the OP's question exactly; that is, it was a Romance language spoken in an Arab country. However, it was also extremely conservative, so maybe Arabic didn't influence it as much as it could have, had it evolved a bit faster.

I know of two Semito-Romance conlangs. Judajca was written for the Ill Bethisad world, and if I can remember correctly it uses Hebrew for the substrate level. All I can find online are brief sketches, though.

A much richer conlang is Syrunian, a Romance language written for Syria. Its main deep substrate is Aramaic/Syraic, with Arabic influence coming in to play later. It is written in Hebrew letters, for whatever reason. Syrunian has some major Semitic-like features:

• The default sentence pattern is VSO.
• The definite article is a prefix rather than a separate word (it's /h-/, from Latin hic).
• Direct objects are marked with a prefix as well (/la-/, taken from the source of the Spanish/French word "la" but behaving like a Syriac prefix).
• Three noun states (emphatic, absolute, and construct) rather than noun cases
• Two verb forms, imperfect and perfect (and the perfect forms actually descend from the Latin imperfects)

Click to expand...

This is nitpicking, but these features are not universally Semitic; most look actually like specifically Northwest-Semitic or just Aramaic (except the prefixed article, that is identical to Hebrew; note that, in most Romance languages and most phonosyntactical contexts, the article is indeed phonologically a prefix, although spelled separately. Romanian unites it to the word in spelling, but it is a suffix there - as in Aramaic and Sayhadic within Semitic -).

I doubt that the VSO order is likely to be mantained by a Romance with Arabic substrate, as the overwhelming majority of Arabic spoken varieties has SVO, like most Romance (but unlike Latin, which had a preferred SOV order).

Falecius said:

[/LIST]
This is nitpicking, but these features are not universally Semitic; most look actually like specifically Northwest-Semitic or just Aramaic (except the prefixed article, that is identical to Hebrew; note that, in most Romance languages and most phonosyntactical contexts, the article is indeed phonologically a prefix, although spelled separately. Romanian unites it to the word in spelling, but it is a suffix there - as in Aramaic and Sayhadic within Semitic -).

I doubt that the VSO order is likely to be mantained by a Romance with Arabic substrate, as the overwhelming majority of Arabic spoken varieties has SVO, like most Romance (but unlike Latin, which had a preferred SOV order).

Click to expand...

I bolded several items in the quote for post #9 because I did not understand them and would've liked some clarification. I understand that my attempt at construction is far too horrible to comment on but could you at least clarify those questions for me?

Isfendil said:

I bolded several items in the quote for post #9 because I did not understand them and would've liked some clarification. I understand that my attempt at construction is far too horrible to comment on but could you at least clarify those questions for me?

Click to expand...

Sure, wait a moment

Isfendil said:

I have bolded the items that I would like you to elaborate on, as I feel I did not understand them quite well.

Click to expand...

1) There seems to be reason to believe that in the past, the letter now written and spoken in standard Arabic as /q/ was commonly pronounced as /g/ (or /G/). I suppose you are perplexed by the fact that the letter now realised mainly as /d͡ʒ/ used to be realized likewise as /g/ at some point, as clearly shown by semitic cognates (and by the realization in Modern Cairene spoken Arabic, where it seems to be a retention).

It seems that there this is explained by dialect difference within pre-islamic Arabic; in some dialects /g/ became /d͡ʒ/ and in a subset of those at least, /q/ became a new /g/. This situation was widespread and probably considered the standard prnounciation at some point. But later (perhaps in the ninth century) the standard was defined as having both a /d͡ʒ/ and a /q/ sound, but no /g/ sound. This is possibly because the /q/ realization was seen a feature of "urban" speech while most Bedouin speech form had /g/ (as shown in Iraq by the oppostion between the so called "gilit" and "qelti" varieties - from the respective voices for the perfect of the verb "to say" ("qult-"/ in Standard Arabic).

2) I do not understand what you don't understand about lexicon. I said that if this language is Romance on Arabic substrate, the number of Arabic lexical items that enter it is likely to be relatively low.

3) Semitic languages have a distinctive (and very discussed) root-pattern system. Essentially, the basic meaning of the word is conveyed by a primarily consonantal root, normally composed of three letters. These are arranged in fixed patterns with infixes (usually vowels) prefixes and suffixes to articulate meaning and inflection. Textbook example: "darasa" = "he studied", "yadrusu"="he studies" "durisa" = "it was studied" (by unknown agents), "yudrasu"= "it is studied" (by unknown agents) "darrasa"="he taught" (made study), "mudarris"="teacher", "daaris"="he who is studying", "madrasa" = "school" (place to study), "dars"="lesson", "duruus"="lessons", etc.
This system is also sometimes extended, through reanalysis, to loanwords (for instance "film" has a plural "aflaam").

This is not a feature that could be passed into Romance by the substrate, but I was saying that there's a possibility that some residue of reanalysis on the root/pattern lines can show in alternances where the change of the vowel sound in a root signal a shift in meaning (this is something that occurs in Indo-European and was hugely productive in the pre-history of Latin, but not very productive in Classical Latin anymore).

4) Arabic does not a verb for "to have". Latin did. This verb expanded greatly its function in Romance, coming to be used as an auxiliary verb, in roughly the same way it is used in English and in German.
A Romance language with an Arabic substrate might not have the same development, and instead lose the verb entirely. In that case, its verbal system would differ from the other Romance system as it would be using the equivalent of "to be" as the primary, if not sole, auxiliary verb. The verb "to be" is the only auxiliary verb in Classical Arabic.

I also argued that other auxiliary verb may develop, especially to signal the future (or continuous, etc.). This is what happened in spoken Arabic, in some Romance languages (Italian uses "to come" as an auxiliary for the passive, Spanish and Portuguese developed "to stay" and "to hold/have" as auxiliaries too, etc., so this sounds like a probable outcome).

I am sorry if this sounds still unclear.

Isfendil said:

Also, I think that I mean Classical Arabic in this case for the purposes of ease,

Click to expand...

Technically somewhat inaccurate but certainly justifiable on the basis of ease, yeah.

Isfendil said:

Furthermore, I would like to say that I am inexperienced, but judging from what you've said, I decided to do a really crude transformation of a latin phrase, and I would like you to tell me how I can improve it. Also, I did it in (again, very crude) IPA transcription, so if there are errors in that, please tell me.

Here is the phrase

entia non sunt multiplicanda praeter necessitatem
[IPA: ɛntiɑ non sunt multiplicɑndɑ prɑjtɛɾ nɛkɛsitɑtɛm]

Here is a really crude transformation that I'm guessing at with what you've told me. It is most certainly horribly wrong, because I am not at all well-versed in any arabic, but I'm trying to create a starting point.

[IPA: non sunt multəfləcɑndɑ alkjənɑtɑ frɑɛtur xtɑgɑ]

Click to expand...

I do not understand why you chose to turn "entia" into "kjenata". Are you using the Arabic root for "to be"? It is unlikely to survive in a Romance version I think. While Arabic "xtaga" is a reasonable loanword. For the rest, not bad.
As I said, I don not believe the final language will preferred VSO order, but it is not impossible either.

Falecius said:

1)
- snip -

Click to expand...

This is much more clear now, thank you.

I do not understand why you chose to turn "entia" into "kjenata". Are you using the Arabic root for "to be"? It is unlikely to survive in a Romance version I think. While Arabic "xtaga" is a reasonable loanword. For the rest, not bad.
As I said, I don not believe the final language will preferred VSO order, but it is not impossible either.

Click to expand...

Kjenata was "to be" and in light of what you just re-explained, I understand that now, although I was trying to replace the noun. As for VSO, that was me personally trying to make the language more "unpredictable". I guess what I'm envisioning is a romance language where the substrate has more heavy influence, like in French, as opposed to Spanish.

I wonder, what a Romance language with the Arabic singular, double and plural forms would be like?

For those who don't know, Arabic has three forms rather than two. Singular, double and plural. So for instance, two people would never be plural, only thee and more can be plural and double is obviously only two.

John7755 يوحنا said:

I wonder, what a Romance language with the Arabic singular, double and plural forms would be like?

For those who don't know, Arabic has three forms rather than two. Singular, double and plural. So for instance, two people would never be plural, only thee and more can be plural and double is obviously only two.

Click to expand...

Something about the language that I have always found admirable. I'm not sure though whether it would go in, because when languages hybridize, some things simplify and some things become more complex.

Actually, if the Arabic cultures are so used to this separate conceptualization (and I am fond of it even though it's not a function of any language I speak), then I think it would go into the language, simply by dint of the people being too used to it.

Isfendil said:

Something about the language that I have always found admirable. I'm not sure though whether it would go in, because when languages hybridize, some things simplify and some things become more complex.

Actually, if the Arabic cultures are so used to this separate conceptualization (and I am fond of it even though it's not a function of any language I speak), then I think it would go into the language, simply by dint of the people being too used to it.

Click to expand...

Unlikely, although it's cool.
Modern spoken Arabic largely lost (although not entirely) functional, productive dual. Romance and Latin do not have it (except very residual fixed forms). This is the sort of thing that does not survive through substrate influence. Influence which, I repeat at the of cost of being boring, often tends to be relatively limited in contexts of language substitution; what you usually have is a restructuring of the new language due to imperfect learning environments. What it really works is normally adstratal/superstratal influence. For instance, French does not seem to have that much Gallic into it*; most of its significant deviation from the Latin norm is probably best explained by internal evolution and Frankish superstrate. This is because language substitution tends to be relatively rapid, while adstrate/superstrate contact has of more time to exert influence.

* Gallic is also poorly documented, which may make assessment of its influx over French complicated. Still, the vast majority of French lexicon has clear either Latin or Germanic etyma. You see roughly the same situation almost everywhere in Romance, and commonly in spoken Arabic.