Disputatio:Ǧamāhīrīyya

The name of this article

The "Y" here is very problematic. In various dialects of Arabic or historical periods of Arabic, the ج consonant can be pronounced either like English "soft" g (as in the word gem) or like English "hard" g (as in the word gift), or further slight variations of these two, but it is never standardly pronounced like y in the word "yes". Furthermore, the main purpose of the letter "y" in the Latin language is actually to transcribe Greek upsilon (which during the Classical period was a front rounded vowel like modern German u-umlaut) -- something which is even further remote from accepted pronunciations of Arabic ج. AnonMoos 07:26, 8 Iulii 2008 (UTC)

Wikipedia clarifies the situation: under en:Arabic, it says the consonant ج is regularly pronounced [dʒ]. It adds that it's "pronounced as [ɡ] by some speakers. This is especially characteristic of the Egyptian and southern Yemeni dialects. In many parts of North Africa and in the Levant, it is pronounced as [ʒ]." I suggest that, insofar as possible, Vicipaedia should accommodate the practice of the vast majority of Arabic-speakers, and not of Egyptian & southern Yemeni minorities. IacobusAmor 19:27, 10 Iulii 2008 (UTC)

Eheu! Nempe ducunt Iemeniam matrem fuisse gentis Arabicae, Aegyptum autem fontem litterarum pellicularum canticorum plerumque quibus gloriantur filii Orientis.--Ceylon 21:17, 10 Iulii 2008 (UTC)

It's true that in some words an English "J" (i.e. soft g or "dzh" sound) corresponds to a Latin "I", but those words have a different source and history than Arabic Jamahiriyya -- and the letter "Y" is not usually used in connection with them. AnonMoos 07:26, 8 Iulii 2008 (UTC)

Tell it to the Spanish at es:Yamahiriyya ! Which seems to be good romance-language evidence for rendering ج as "y." In much of the Spanish-speaking world, word-initial "y" can be pronounced like the "j" in jam. (For example, its sound in "yo" often approximates that of "J" in English "Joe.") In large parts of South America, even medial "y" varies between that and the "z"-sound of English "azure." If in Spanish, why not in Latin? We make far more wrenching changes in the language when we use newfangled words like televisificus and transcendentalismus. IacobusAmor 11:16, 8 Iulii 2008 (UTC)

Are you suggesting we should apply Spanish spelling conventions to Latin? No, seriously, I agree with AnonMoos. I you want parallels, see the g in Algebra, in Rigel (cf. en:Rigel), or Gibraltar. --Fabullus 11:46, 8 Iulii 2008 (UTC)

Those aren't parallels. The parallel would be with Giamahiria, not Gamahiria. The sound of gia would seem pretty close to what's wanted, but the sound of ga wouldn't, so maybe Giamahiria would be the best Latin spelling. You can't ordinarily get the "j"-in-"jam" sound out of "g" unless it's followed by an "e" or an "i." In English, the only exception that comes to mind is margarine, and the ga in its cousin, margarite (the other derivative from the same Greek word) is pronounced regularly. IacobusAmor 11:51, 8 Iulii 2008 (UTC)

And what about gaol? --Fabullus 18:46, 10 Iulii 2008 (UTC)

A bizarrerie of no account here! According to the Oxford English Dictionary, "The archaic spelling gaol . . . is obsolete in the spoken language, where the surviving word is jail" (s.v. jail), and "In British official use the forms with G are still current; in literary and journalistic use both the G and the J forms are now admitted as correct, but all recent Dictionaries give the preference to the latter" (s.v. gaol). IacobusAmor 19:19, 10 Iulii 2008 (UTC)

I seem to remember that Ceylon wrote something about this problem somewhere. Why not ask him for his opinion? --Fabullus 11:55, 8 Iulii 2008 (UTC)

Re vera illam ipsam tractavimus quaestionem, ubi disputavimus, quomodo Gihad recte scriberetur. Num dubium est quin eadem ratione et Gamahiriyya (sive lenius Gamahiria) pro جماهيرية ǧamāhirīyya scribere cogamur?--Ceylon 21:04, 10 Iulii 2008 (UTC)

Quomodo Arabice enunciatur ǧa? Haec littera ǧ non in en:Arabic phonology apparet.‡ Suspicor ǧa enuntiari /dʒa/, ergo Latine gia vel dia vel fortasse ia vel etiam ya, sed certe non ga.——‡"/dʒ/ is pronounced as [ɡ] by some speakers. This is especially characteristic of the Egyptian and southern Yemeni dialects. In many parts of North Africa and in the Levant, it is pronounced as [ʒ], and in certain regions of Oman it is pronounced as [j]. In classical Arabic, this was either [ɟ] or [gʲ]." IacobusAmor 21:29, 10 Iulii 2008 (UTC)

Certe melius nomen Latinum est Giamahiria (vel [Y/I]amahiria), non Gamahiria. Vide nexus: ar:جماهيرية, bg:Джамахирия, cs:Džamáhíríje, da:Jamahiriyya, de:Dschamahiriyya, en:Jamahiriya, es:Yamahiriyya, fr:Jamahiriyya, it:Jamahiriya, lt:Džamahirija, no:Jamahiriya, pl:Dżamahirijja, ru:Джамахирия, sk:Džamáhíríja, uk:Джамахірія. Nulla vicipaedia primis litteris Ga- vel sonis /ga/ utitur. IacobusAmor 11:26, 12 Iulii 2008 (UTC)

Certe non ignoras, Iacobe, legem sive regulam de hac re esse constitutam, cui obtemperantes certa utimur translitteratione. Quod ad Arabicam linguam attinet, haec est translitteratio ad normam ISO, scilicet ǧamāhirīyya, aut fastigiis abiectis Gamahiriyya. Si vis hanc regulam repudiare - quod absque omni dubio fas est - quaestio illuc disputanda mihi videtur, non hic.--Ceylon 12:16, 12 Iulii 2008 (UTC)

Amice, ubi invenis hanc litteram ǧ pro littera Latina g (quod enuntiamus /g/) valere? Secundum commentationes en:Arabic language et en:Arabic phonology, ج‎ enuntiatur (in Alphabeto Phonetico Internationali) /dʒ/, quod ante a Latine, ut videtur, optime scribitur gia vel dia vel fortasse etiam ia vel ya, sed haudquaquam ga. Litterae Latinae ga pro IPA /ga/ valent, sed sonus hic desideratus est /dʒ/. IacobusAmor 13:24, 12 Iulii 2008 (UTC)

Vide hic et illic et istic. Est haec translitteratio ubique terrarum maxime usitata in libris doctis qui de rebus orientalibus disserunt.--Ceylon 14:22, 12 Iulii 2008 (UTC)

Scio et scio et scio, sed ubi haec littera ǧ pro littera Latina g (quod enuntiamus /g/) valet? Nobis dicis litteram ǧ esse litteram g. Immo: dico litteram ǧ esse ǧ, et litteram g esse g. Secundum translitterationes citatas, nomen huius commentationis recte est ǧamahiria, non Gamahiria. Nihilominus, Iamahiria fortasse est melius. Vide infra. IacobusAmor 21:58, 12 Iulii 2008 (UTC)

---

Classical Latin just doesn't have any letter which closely corresponds in pronunciation to a phonetic "dzh" or "zh" sound. In medieval Latin "g" was most often used to correspond to Arabic ج. I don't necessarily have a strong objection to "Giamahiria" (though it's actually more of an Italian spelling than Latin), but the letter "y" is inappropriate for this word... AnonMoos 21:43, 12 Iulii 2008 (UTC)

My point all along has been that in Latin the letter G produces the sound /g/ (a "hard G"). If you want to pronounce Gamahiria that way, and that way is the standard way of pronouncing it, fine with me! You're going to pronounce its first three letters almost exactly like the English word gum. My worry is that the cited evidence, including the pages cited by Ceylon above, suggest that this word should not be pronounced with a "hard G," except in Egypt and South Yemen. IacobusAmor 21:58, 12 Iulii 2008 (UTC)

Sometimes when you only have about 24 letters in your alphabet and there are thousands of sounds in the languages of the world, then you have to make compromises. Latin G for Arabic ج was an established medieval convention, AnonMoos 22:07, 12 Iulii 2008 (UTC)

Was it? In all environments? or just when ǧ was followed by e or i? Fabullus's list (Nomina Latina Mediaevalia e lingua Arabica mutuata) has no examples in which Arabic ǧa became Latin ga. ¶ For a different contradiction, in which Arabic ǧ = Latin i, see below. IacobusAmor 00:22, 13 Iulii 2008 (UTC)

and "g" also represents the actual pronounciation used in an Arabic dialect which is fairly strongly known through the Arabic world (though not the dialect which is most often used in Libya). If one letter among the 24 has to be chosen, then "g" has many fewer problems than all of the other 23, and so is left standing as the only semi-plausible candidate...AnonMoos 22:07, 12 Iulii 2008 (UTC)

But it doesn't: two letters can be chosen! IacobusAmor 00:22, 13 Iulii 2008 (UTC)

...then "g" has many fewer problems than all of the other 23, and so is left standing as the only semi-plausible candidate... AnonMoos 22:07, 12 Iulii 2008 (UTC)

So then: (1) how can we explain the fact that, according to Fabullus's list, Arabic initial ǧu (in ǧulāb) became Latin iu (in iulapium), not Latin initial gu? (2) Also, Arabic al-ǧabr became Latin algebra. Therefore, perhaps we could rightly say that attested usage indicates that the title of this article should be something like (according to attestation 1) Iamahiria or (according to attestation 2) Gemahiria. IacobusAmor 00:22, 13 Iulii 2008 (UTC)

I have been wondering about iulapia myself. I guess that this word was not imported directly into Latin, but via English or French were initial J may have just the required value. By the way, I also found the alternative Latin form zulapium. --Fabullus 12:40, 13 Iulii 2008 (UTC)

What we know: ¶ The standard modern (ISO, but not IPA) way of transliterating the Arabic syllable in question is ǧa. ¶ This syllable typically has the sound /dʒa/. ¶ The typical (or at least a very, very common) way that Latin transcribes this sound is ia. ¶ Vicipaedia already has English Jamaica = Latin Iamaica, and Java = Iava, and Jordan = Iordania. ¶ Latin ia approximates /dʒa/ more closely than Latin ga.——What we may conclude: ¶ Unless we'd rather violate all sorts of attestations and regularize Vicipaedia according to the Egyptian & South Yemeni pronunciations of a particular Arabic sound, thereby respelling these names as Gamaica, Gava, and Gordania, we should spell the name of the present article Iamahiria. IacobusAmor 12:12, 13 Iulii 2008 (UTC)

What these examples are telling us is that at some period in time, and probably depending on the author's mother tongue, Arabic ǧ might be rendered in Latin as g (before e or i), as gi (before other vowels), as j (by speakers of French and English), and z (compare the vulgar Latin spelling Zabolus for Diabolus). Should we permit all those different spellings now as well? I don't think Latin i for English j supports your proposition either. The way I see it this is not a phonetical transciption, but a mechanical equation, not much different from g for ǧ. Iordania is not a case in point; it is named after the river that already in classical times bore the name Iordanes. Your parallels are not really parallels at all. Algebra has been Latinized, Gamahiria has not, and since we have all agreed not to invent new Latin words, Gamahiria will not be Latinized, it is a foreign word. Gamahiria or better still Gamahiriyya is nothing but an internationally approved transcription, with the diacritics omitted. To prevent further misunderstanding I suggest moving the page to Ǧamāhirīyya, and use Gamahiriyya (without diacritics) only as a redirect. --Fabullus 12:40, 13 Iulii 2008 (UTC)

That's a good idea. Andrew Dalby (disputatio) 12:43, 13 Iulii 2008 (UTC)

What I might then recommend would be a lemma that gives both forms: one that most readers (ignorant of what the character ǧ means) will have a chance of pronouncing fairly correctly, and then the visually finicky one, thus: Giamahiria, recte Ǧamāhirīyya, est. . . . IacobusAmor 21:24, 13 Iulii 2008 (UTC)

Giamahiria would be late Latin for the target pronunciation [dʒa-]. Why late Latin? Why not classical (Zamahiria)? But (still more) seriously, I agree with Fabullus' suggestion. --Neander 00:49, 14 Iulii 2008 (UTC)

I agree (with Fabullus and Adrew). If the exemptio diacriticorum principle works in å ä ö (e.g. Thomas Lovqvist for Thomas Lövkvist), why not elsewhere? But if we’re asking how the old Romans heard [dz] or [dž], they obviously heard more or less [z] and repressnted this sound as < s > and later < z >. Plautus represented ζώνη / ζώνα as sona, and Ζῆθος is inscriptionally represented as Setus. From the 1st c. B.C., < z > was used, as Fabullus says. --Neander 20:58, 13 Iulii 2008 (UTC)

So you're saying that the ζ in ζώνη was pronounced [dž] and is therefore relevant here? I don't have Allen's book on Greek pronunciation, so I can't check it myself, but I do recall reading in various places that ζ was [dz] and/or [zd]. IacobusAmor 21:18, 13 Iulii 2008 (UTC)

In the classical Greek that Allen's book covers, the zeta is [zd]. And of course by the time of classical Latin (several hundred years later) it had passed to [zz], as Allen shows in his book on Latin. —Mucius Tever 02:56, 14 Iulii 2008 (UTC)

Yes, there are etymological and morphological reasons for assuming both pronunciations for Greek ζ. However, from the Roman point of view, the problem is how to represent the sound which they obviously heard more or less as [z], whatever the objective sound was. Phonetically untrained people tend to hear sounds phonologically, through the functional sound system of their own language. In principle, then, had Cicero heard the objective sound [dz] (or [dž]) coming from the pronunciation of a Greek or an Arabic word, he’d probably have written it < z >. --Neander 22:15, 13 Iulii 2008 (UTC)

I see that Iustinus gave us the article Ziudo, so he'd presumably favor Ziamahiria, but we can't be sure without hearing from him. Say, where is he, anyway? IacobusAmor 22:01, 13 Iulii 2008 (UTC)

Careful, now. The spelling of ziudo is not based on the J sound, but on the actual Japanese: じゅう- is zyuu- in ISO 3602 and phonemically /zjɯː/.

I bet Cicero would've written Zudo.   :–)   --Neander 22:20, 13 Iulii 2008 (UTC)

... or Ziudo.   :–)   --Neander 00:28, 14 Iulii 2008 (UTC)

Nah, ziudo would probably have had too many syllables for him. (Of course Cicero would more likely have had a G in jamahiriya, as the proto-Arabic of the age hadn't turned the stop into an affricate...) —Mucius Tever 02:49, 14 Iulii 2008 (UTC)

Now that's the most interesting statement in this thread! IacobusAmor 10:57, 14 Iulii 2008 (UTC)

Moving the article to Ǧamāhirīyya would certainly be correct, but it would also run counter to the general observance of not using diacritics in lemmata, which is probably due to both practical (diacritics not being displayed in all browsers, and not familiar to everybody) and aesthetic reasons. Therefore I would still prefer Gamahiriyya (which is, as Fabullus says above, merely a mechanical translitteration, and does not aspire to be a phonetically correct transcription - which Latin would not allow, anyway). ¶ The rendering of ج as g in some mediaeval words borrowed from Arabic may have something to do with the fact that /g/ as a pronunciation is less exotic than Iacobus makes it appear: --Ceylon 06:29, 14 Iulii 2008 (UTC)

The ordinary Western European late medieval pronunciations of ge and gi were presumably [dže] & [dži], or maybe [že] & [ži], not [ge] & [gi]. What might be dispositive would be examples of ج+[a], or ج+[o], or ج+[u], which medieval Europeans transliterated as ga, or go, or gu, respectively. Fabullus's list does have an example of ج+[u], but Europeans transliterated it as iu, not gu. So whatever the present-day pronunciation that can be heard among Arab-speakers, it would appear that at the time that the principles of transliteration were being developed (and Latinists were thereby establishing forms that became our algebra and Gibraltar and such), Europeans were hearing ج as [dž] or [ž]. IacobusAmor 10:55, 14 Iulii 2008 (UTC)

I agree that the list Fabullus has compiled does not (yet) contain enough evidence to reconstruct a pattern of transcription. (This is precisely one of the reasons why I would prefer Vicipaedia to follow a system 'based on the eye' in a systematic fashion, rather than one 'based on the ear' - for which it is hard to come up with consensual and clear rules). Regarding the mediaeval pronunciation of Latin, you certainly have a point as far as Italy, probably the main line of contact with the Arab world, is concerned. But calling it "The ordinary Western European late medieval pronunciations of ge and gi" seems too sweeping for my taste: The Germanic and Slavic world universally pronounced Latin ge and gi as /g/ (and ce and ci as /ts/).--Ceylon 13:13, 14 Iulii 2008 (UTC)

Ah, excellent point. (Germans & Slavs don't seem to be Western Europeans to me; Central Europeans, maybe.) I was indeed thinking of Italian, Spanish, French, and British scribes, or should we say the people that Arabs might have called the Franks, whom I'd have expected to be more involved in Africa & the Levant than Germans & Slavs, and therefore likelier to have been the initial transcribers (who naturally were imposing their substrate pronunciations on what they were transcribing), but of course that thought could be erroneous. IacobusAmor 13:24, 14 Iulii 2008 (UTC)

Yea, but our guys fought in the crusades too :) --Ceylon 15:45, 14 Iulii 2008 (UTC)

It is a variant which due to the prominence of Egyptian film etc. is almost on a par in frequency with the standard /dʒ/ and the Levantine /ʒ/. ¶ Maybe an elaboration of the rules on Vicipaedia:De nominibus propriis is called for, including references to translitteration tables for specific languages?--Ceylon 06:29, 14 Iulii 2008 (UTC)

I agree! Go ahead (if everyone agrees, of course). --Fabullus 06:47, 14 Iulii 2008 (UTC)

I have moved the page to Gamahiriyya (without diacritics), but throughout the page it now says Ǧamāhīrīyya. (Da veniam, Iacobe) --Fabullus 05:44, 15 Iulii 2008 (UTC)