Talk:Romance languages/Archive 4

Vulgar Latin
The term "Vulgar Latin" is associated with vernacular speech forms postdating Classical Latin, beginning approximately in the 2nd or 3rd century and continuing well into Late Antiquity. It is more than problematic to equate it with Proto-Romance, as this article and other Wikipedia articles do. By about 300 AD at the latest, there is uncontested evidence of regional differences reflecting later divisions within Romance, and according to some authors, the divergence of Sardinian may have started even earlier than that, significantly so, as is pointed out in the article. Moreover, while Vulgar Latin is described as having lost the Latin case system almost completely, or featuring at most a two-case system, the use of articles and the loss of the neuter, the absence of articles, the neuter and three cases can be solidly reconstructed for Proto-Romance: nominative, accusative and genitive-dative. Harris/Vincent are rightly critical of the term "Vulgar Latin" – including for its vagueness. Certainly, by about 300 AD we should speak of an Early Romance dialect continuum that was spoken across the Roman Empire, although the vernacular was still identified as being the same language as literary Latin by its speakers. --Florian Blaschke (talk) 01:45, 6 March 2021 (UTC)
 * Are you saying that Vulgar Latin is in fact the dialect continuum of early Romance? 98.143.65.225 (talk) 17:08, 8 March 2021 (UTC)

Portuguese pronunciation of "meu".
The page says "Final -m was dropped in Vulgar Latin. Even in Classical Latin, final -am, -em, -um (inflectional suffixes of the accusative case) were often elided in poetic meter, suggesting the m was weakly pronounced, probably marking the nasalisation of the vowel before it. This nasal vowel lost its nasalization in the Romance languages except in monosyllables, where it became /n/ e.g. Spanish quien < quem "whom", French rien "anything" < rem "thing"; note especially French and Catalan mon < meum "my (m.sg.)" pronounced as one syllable (/meu̯m/ > */meu̯n/, /mun/) but Spanish mío and Portuguese and Catalan meu < meum pronounced as two (/ˈme.um/ > */ˈme.o/).".

I'm a native Portuguese speaker, and that info's wrong. "Meu" in Portuguese is just one syllable, although its pronunciation is either /meu̯/ and /mew/ depending on who you ask. Most sources (Wiktionary included) say it should be the latter, though. 187.0.109.249 (talk) 12:42, 9 March 2021 (UTC)
 * It's monosyllabic in Catalan too. I'll change that. --Jotamar (talk) 20:07, 9 March 2021 (UTC)
 * To clarify: the forms now have just one syllable, but they are descendants of a Latin pronunciation variant with two syllables. The Catalan mon descends from the Latin one-syllable pronunciation. Rua (mew) 20:27, 9 March 2021 (UTC)

Removed Mozarabic
I removed the Mozarabic sample, added by 187.140.8.87 on April 22nd, 2016, due to various issues. First, Mozarabic is quite tentatively reconstructed phonetically, and it is questionable whether to include such an old and scantily attested language... I don't think all the words are attested. Second, and more seriously, the sample text is wrong, and does not follow the conventions of the attested Mozarabic. To mention one problem, kharjas typically show final -e either written with an Arabic yaa' (ي), or not at all, e.g. an infinitive *leváre in kharja 24 on this site written lbry (لبري), and yet in kharja 23 one can see *nóhte as nxt (نخت), and *iréyme as yrym (يريم) (right above *gárme as ġrmy (غرمي), no less!), but the guest user used alif (ا) instead. I would rather not have to make up what the Mozarabic would have been.--Ser be etre shi (talk) 13:13, 10 March 2021 (UTC)
 * I completely agree, we know too little about Mozarabic to create new sentences in it. --Jotamar (talk) 18:15, 15 March 2021 (UTC)

Listing/counting Romance languages
Copied form User talk:Austronesier:

I'm refraining for the moment from reverting your reversion of my attempt to clarify that the number of Romance languages can't be established, in hopes that an edit war can be avoided. You object:

''This sounds as if we had some white spots in Romance linguistics, and that's certainly not the case. It's a matter of the cut-off line between language and dialect, and this is clearly stated here.''

I don't know what 'white spots' means in this context, so I can't address that. It is the case that there is no definitive count of Romance languages, and any attempt to establish one runs into -- prime among other difficulties -- the intractable language/dialect of dilemma. I don't see where that is clearly stated for the general reader, who can't be expected to grasp it without exemplification from Because it is difficult to assign rigid categories to phenomena such as languages, which exist on a continuum. No small % of general readers are operating with no awareness of the underlying Romance continuum, i.e. conceptually at the level of "In France they speak French, in Italy they speak Italian". No way for them to intuit that e.g. Bolognese and Ferrarese even exist, much less that there could be any principled debate as to whether they're different languages or "dialects of" Emilian-Romagnol, nor to understand what's at work if, for example, an ingenuous native of Llanes is surprised, even puzzled, to learn that the indigenous Romance of the town is Asturleonese.

Rather than the more extensive re-write that's actually called for, including -- if mention of Dalby's count is useful at all -- noting that closely cognate different languages can be highly mutually comprehensible, thus the criterion is of dubious value at best, I tweaked the text just slightly with the goal of offering a glimpse of reality: the list offered is (necessarily) incomplete and can be expanded. I'm quite open to different wording or to a complete re-write of the paragraph beginning Because... What shouldn't be left to stand is even a hint of implication that the list below the text is in any sense complete. Barefoot through the chollas (talk) 17:35, 27 March 2021 (UTC)
 * First to explain why I have talked about "white spots": the word "complete" in no count is definitively complete immediately evoked to me the picture of areas like the Amazon, which are still underresearched, and where the number of languages could still rise with increased knowledge and more detailed surveys. A list of Romance languages can be "complete" as long as the criteria are sharply defined. The choice of such criteria is of course relative/arbitrary.
 * But I fully share your worries that such a list in the lede gives in the impression to the casual reader that it is some kind of authoritative listing. I was blinded by the Dalby source, which of course becomes useless once one leaves a backdoor for "additional current, living languages". With such a backdoor, it is totally arbitrary. On second thought, your edit is useful in this respect, but comes out a bit like a piece of tape on a pot with countless cracks.
 * Why actually try to give a detailed list at all in the lede? We can do this in Kartvelian languages, but not here; the internal structure of Romance is just too complex to squeeze it into the lede. Ideally, the first section of the article should be "Languages", which gives a detailed outline and also explains the complexity of the matter, and the fuzziness of language/dialect divisions. An inventorizing "lingocrat" source like Dalby (which is just a catalogue/index) should not be the only reference for it. –Austronesier (talk) 11:41, 28 March 2021 (UTC)
 * With perhaps a couple of minor quibbles that shouldn't be difficult to resolve, it sounds as though we pretty much agree. I'll try to be brief at this point.


 * Dalby's list based infelicitously on presumed mutual intelligibility can have a doubly blinding effect: 1) it's the only reference, thus is inevitably granted some degree of authority in the eyes of at least some readers (and attempts to solicit a reference for the more expanded list that someone set up have fallen on deaf ears); 2) it can leave the impression that mutual intelligibility is an accepted and trustworthy criterion for determining same/different language. Pursuing your metaphors, I'd claim that Dalby's list leaves its own front door wide open for the observation that even if his MI list is accurate, the MI criterion in itself disqualifies his 23 as a credible count of languages. Any other criterion or set of criteria hits a similar wall of arbitrariness. The real-world fact is that the pot is inherently cracked.


 * The list presented now in the lede isn't really one of languages, but sort of cladistic typologies, with a few representatives mentioned for most types (but not all, i.e. "Sardinian" stands inexplicably alone with no mention of even just Logudorese vs. Campidanese), elaborated somewhat incoherently also in that those mentioned vary from very local (Romanesco) to broad subtypologies encompassing numerous differentiated varieties (e.g. Portuguese, French/Oïl languages, Venetian -- by which is meant not actual Venetian, but Venetan). I mostly, perhaps even totally, agree with your last paragraph above. But since an exhaustive list is not possible (at any point in the article), a cleaned up version of the typological list that's there now can serve as a useful orientation.


 * I said I'd be brief, so I'll stop rather than finish. My overarching concern re the paragraph in question and the list below it is that the information provided be as clear and accurate as possible in limited space, and that readers not be led astray. Some minor editing can achieve that for the nonce, a holding pattern until someone with the time and knowledge chooses to carry out a more ambitious re-write. Barefoot through the chollas (talk) 22:47, 28 March 2021 (UTC)

Removed Mozarabic
I removed the Mozarabic sample, added by 187.140.8.87 on April 22nd, 2016, due to various issues. First, Mozarabic is quite tentatively reconstructed phonetically, and it is questionable whether to include such an old and scantily attested language... I don't think all the words are attested. Second, and more seriously, the sample text is wrong, and does not follow the conventions of the attested Mozarabic. To mention one problem, kharjas typically show final -e either written with an Arabic yaa' (ي), or not at all, e.g. an infinitive *leváre in kharja 24 on this site written lbry (لبري), and yet in kharja 23 one can see *nóhte as nxt (نخت), and *iréyme as yrym (يريم) (right above *gárme as ġrmy (غرمي), no less!), but the guest user used alif (ا) instead. I would rather not have to make up what the Mozarabic would have been.--Ser be etre shi (talk) 13:13, 10 March 2021 (UTC)
 * I completely agree, we know too little about Mozarabic to create new sentences in it. --Jotamar (talk) 18:15, 15 March 2021 (UTC)

Listing/counting Romance languages
Copied form User talk:Austronesier:

I'm refraining for the moment from reverting your reversion of my attempt to clarify that the number of Romance languages can't be established, in hopes that an edit war can be avoided. You object:

''This sounds as if we had some white spots in Romance linguistics, and that's certainly not the case. It's a matter of the cut-off line between language and dialect, and this is clearly stated here.''

I don't know what 'white spots' means in this context, so I can't address that. It is the case that there is no definitive count of Romance languages, and any attempt to establish one runs into -- prime among other difficulties -- the intractable language/dialect of dilemma. I don't see where that is clearly stated for the general reader, who can't be expected to grasp it without exemplification from Because it is difficult to assign rigid categories to phenomena such as languages, which exist on a continuum. No small % of general readers are operating with no awareness of the underlying Romance continuum, i.e. conceptually at the level of "In France they speak French, in Italy they speak Italian". No way for them to intuit that e.g. Bolognese and Ferrarese even exist, much less that there could be any principled debate as to whether they're different languages or "dialects of" Emilian-Romagnol, nor to understand what's at work if, for example, an ingenuous native of Llanes is surprised, even puzzled, to learn that the indigenous Romance of the town is Asturleonese.

Rather than the more extensive re-write that's actually called for, including -- if mention of Dalby's count is useful at all -- noting that closely cognate different languages can be highly mutually comprehensible, thus the criterion is of dubious value at best, I tweaked the text just slightly with the goal of offering a glimpse of reality: the list offered is (necessarily) incomplete and can be expanded. I'm quite open to different wording or to a complete re-write of the paragraph beginning Because... What shouldn't be left to stand is even a hint of implication that the list below the text is in any sense complete. Barefoot through the chollas (talk) 17:35, 27 March 2021 (UTC)
 * First to explain why I have talked about "white spots": the word "complete" in no count is definitively complete immediately evoked to me the picture of areas like the Amazon, which are still underresearched, and where the number of languages could still rise with increased knowledge and more detailed surveys. A list of Romance languages can be "complete" as long as the criteria are sharply defined. The choice of such criteria is of course relative/arbitrary.
 * But I fully share your worries that such a list in the lede gives in the impression to the casual reader that it is some kind of authoritative listing. I was blinded by the Dalby source, which of course becomes useless once one leaves a backdoor for "additional current, living languages". With such a backdoor, it is totally arbitrary. On second thought, your edit is useful in this respect, but comes out a bit like a piece of tape on a pot with countless cracks.
 * Why actually try to give a detailed list at all in the lede? We can do this in Kartvelian languages, but not here; the internal structure of Romance is just too complex to squeeze it into the lede. Ideally, the first section of the article should be "Languages", which gives a detailed outline and also explains the complexity of the matter, and the fuzziness of language/dialect divisions. An inventorizing "lingocrat" source like Dalby (which is just a catalogue/index) should not be the only reference for it. –Austronesier (talk) 11:41, 28 March 2021 (UTC)
 * With perhaps a couple of minor quibbles that shouldn't be difficult to resolve, it sounds as though we pretty much agree. I'll try to be brief at this point.


 * Dalby's list based infelicitously on presumed mutual intelligibility can have a doubly blinding effect: 1) it's the only reference, thus is inevitably granted some degree of authority in the eyes of at least some readers (and attempts to solicit a reference for the more expanded list that someone set up have fallen on deaf ears); 2) it can leave the impression that mutual intelligibility is an accepted and trustworthy criterion for determining same/different language. Pursuing your metaphors, I'd claim that Dalby's list leaves its own front door wide open for the observation that even if his MI list is accurate, the MI criterion in itself disqualifies his 23 as a credible count of languages. Any other criterion or set of criteria hits a similar wall of arbitrariness. The real-world fact is that the pot is inherently cracked.


 * The list presented now in the lede isn't really one of languages, but sort of cladistic typologies, with a few representatives mentioned for most types (but not all, i.e. "Sardinian" stands inexplicably alone with no mention of even just Logudorese vs. Campidanese), elaborated somewhat incoherently also in that those mentioned vary from very local (Romanesco) to broad subtypologies encompassing numerous differentiated varieties (e.g. Portuguese, French/Oïl languages, Venetian -- by which is meant not actual Venetian, but Venetan). I mostly, perhaps even totally, agree with your last paragraph above. But since an exhaustive list is not possible (at any point in the article), a cleaned up version of the typological list that's there now can serve as a useful orientation.


 * I said I'd be brief, so I'll stop rather than finish. My overarching concern re the paragraph in question and the list below it is that the information provided be as clear and accurate as possible in limited space, and that readers not be led astray. Some minor editing can achieve that for the nonce, a holding pattern until someone with the time and knowledge chooses to carry out a more ambitious re-write. Barefoot through the chollas (talk) 22:47, 28 March 2021 (UTC)

Age of the divergence of the vowel systems
Just a quick OR observation: In, it is pointed out that in inscriptions, short close and long mid vowels are often confused, indicating that they were pronounced fairly similarly, like near-close and near-close vowels [ɪ] : [eː] and [ʊ] : [oː], preparing the Italo-Western Romance quantity collapse. Maybe we should go further, though: This frequent confusion could also mean that their relative vowel spaces were already encroaching on each other, and overlapping at least partly, even if their qualities might not have merged yet completely. Or, in fact, their qualities had already merged, and they were only distinguished by their relative quantities. Either way, they were approaching each other in quality so much that the quantity collapse would immediately cause their merger once it occurred.

Now consider that dialects with Sardinian-type vowel systems (once apparently spoken in several regions in the south of Italy – not only Sardinia and perhaps Corsica) show almost no trace of this merger. This indicates that short /i/ was pronounced [i ~ ɪ], not [ɪ ~ e] in these dialects before a quantity collapse occurred in them too. Even if this development may ultimately have been triggered by a Greek adstratum, or some other kind of adstratum, this conclusion seems inevitable. Hence, there must have been a genuine difference in pronunciation between (regional?) varieties of Latin already in the Classical Latin period, perhaps as early as Cicero's time. And further, this (regional?) difference foreshadowed the split into dialects that developed the Italo-Western Romance vowel system and into dialects that developed the Sardinian-type vowel system (as well as dialects that developed Romanian-like "compromise" vowel systems, which may however originally have been dialects with Sardinian-type vowel systems that were overlaid by dialects with the apparently more mainstream Italo-Western system. (It is very suggestive to propose that the first type of pronunciation was found in more northern regions of the Roman Empire, being more mainstream, and the second type in more southern regions, although I do not know if there is evidence along the lines of vowel confusion being less common in inscriptions found in southern Italy, Sicily, Sardinia, Corsica and Africa.)

If this conclusion is correct (and inscriptional evidence could bolster it), this would mean that the fact that a full merger of short /i/ and long /eː/ is not found before the late third or fourth century AD is immaterial: The first signs of intra-Romance divergence can then already be dated to the Classical Latin period, and "Proto-Romance" must then be effectively identical to spoken Latin around that time (the first century BC, perhaps slightly earlier or later, but in any case significantly earlier than the third century AD; I don't know the precise dating of the inscriptions that show confusion between short close and long mid vowels). Traits that must be reconstructed to "Proto-Romance" based on the evidence of the medieval and modern languages might then have started already in the Classical Latin period, even if they were not complete even in the spoken language; and for developments like [ae] > [ɛː], there is inscriptional evidence pointing into this direction at a quite early period, even though the merger probably became general only in the first century AD (judging by the evidence of the Pompejan inscriptions), and was still resisted by more learned registers until Late Antiquity.

However, the most basilectal registers of Latin at least may have shown certain developments foreshadowing the situation throughout Romance (such as the merger of most cases, with only the nominative, accusative and to some extent genitive remaining more stable, at least in early Romance, even though they tended to merge later too) quite early, and then it makes more sense to posit that interference from other languages – such as Koine Greek, Gaulish and even Oscan – in bilingual speakers (who spoke Latin only as a second language and significantly deviated from the literary standard and even the spoken Latin of lower-class speakers who were native to Latium) triggered these changes – such as confusing the genitive with the dative and ablative, conflating the quantities, avoiding many "short words" and particles and preferring longer ones, etc. It is to be expected that these developments would be far less visible in the written record as late as the second century AD, even if they were already occurring, due to the constant efforts by learned speakers, and the influence of first-language speakers in general, to suppress these "barbarisms". It is not surprising that with the decline of Roman culture in Late Antiquity these developments would become more apparent, and that the innovations could not be as thoroughly suppressed anymore. --Florian Blaschke (talk) 16:32, 31 August 2021 (UTC)


 * I suggest you learn more about Romance Linguistics, preferably through reading a modern reference work or two (from cover to cover), before you embark on fantastical Quixotic quests to overturn the entire field. — Preceding unsigned comment added by 2600:1700:1150:1920:20CD:D56C:3712:BAF4 (talk) 07:18, 26 December 2021 (UTC)

Wiki Education Foundation-supported course assignment
This article is or was the subject of a Wiki Education Foundation-supported course assignment. Further details are available on the course page. Student editor(s): Kateybeck.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 08:19, 17 January 2022 (UTC)

Resolving "This article possibly contains original research. (June 2019)"
The template has been here almost a year, and needs resolution. Whoever placed it there in the first place would be doing everyone a favor if s/he would point out a few examples of what s/he judges to be original research, so that they can be corrected. Barefoot through the chollas (talk) 16:58, 25 April 2020 (UTC)


 * Pretty much all of the article is OR because a lot of it isn't sourced.
 * Asparagusus (talk) 16:50, 29 March 2022 (UTC)
 * Probably due to much of it being common knowledge at sort of the undergraduate level, found in just about any basic intro. In any case, the OR template is pretty much useless without an indication of which bits of text were felt by someone to need references. Barefoot through the chollas (talk) 20:17, 29 March 2022 (UTC)

How many Catalan-speakers are there?
In the first paragraph, listing populations of speakers, the number for Catalan has been recently changed from 4.5 million to 14.5 million. A website called  says 9 million, "with more than 4 million speaking it as a native language, according to official census data". Who's right, and where can we get unbiased figures? Kotabatubara (talk) 16:10, 17 April 2022 (UTC)
 * Excellent questions, for which answers are hard to come by. What can be said for sure is that 14.5 million is absurdly large, given that the population of Catalunya is a bit over 7.7 million, and a large contingent of residents, especially in Barcelona, do not speak Catalan fluently. Speakers outside of Catalunya will add up to a substantial number, but not many millions. But wait! There's more! The count is meant to be "by number of native speakers", meaning that those who immigrated to Catalunya (or Andorra or Valencia) for work years after first-language acquisition but who over the years became fluent in Catalan are not to be counted -- which makes the figure of 14.5 million even more absurd. I haven't found figures that seem trustworthy, but this source might be sort of a start: https://www.idescat.cat/pub/?id=eulp&lang=en&utm_campaign=home&utm_medium=cercador&utm_source=estad Barefoot through the chollas (talk) 07:03, 18 April 2022 (UTC)
 * I can remember that there was a pretty serious study which estimated the number of native speakers at around 4.3 million. I haven't been able to locate that study so far, however I've changed the figure in the page because the previous one was completely ludicrous. --Jotamar (talk) 16:39, 23 April 2022 (UTC)

Most spoken Romance languages by number of native speakers - sources
The lead currently has an invisible comment saying:

followed by a listing of the 6 most spoken Romance languages by number of native speakers. Recently, an IP editor did in fact increase the number of speakers of Catalan to 10 million, even though that figure includes non-native speakers. I think I'll undo that edit, but I'm wondering where do the older figures come from? Ethnologue? Erinius (talk) 06:06, 15 August 2022 (UTC)

Italian is readily intelligible to speakers of other Romance languages ???
From the article: "standard Italian is a “central” language (i.e., it is quite close and often readily intelligible to all other Romance languages)"

There are a couple of difficulties with this assertion.

1) It needs to be clarified whether this statement refers to spoken or written texts.

2) In either case, this is too strong a statement. We can dispute the degree of intelligibility, but to say "readily intelligible" and "all" other Romance languages doesn't seem correct (even with the qualification of "often".) Omc (talk) 16:46, 11 January 2023 (UTC)


 * Agreed that the statement is much too strong. True enough that Italian has generally undergone less phonological erosion than other major Romance varieties, thus basic lexicon can very often be more recognizable than in types less conservative phonologically (extremes: compare Fr. and It. versions of the month 'August'). And it may be that lexical innovation and borrowing has been somewhat less prominent in Italian (e.g. 'dog' as cane, no gos or perro or ghjacaru). But that in no way implies, much less certifies, that Italian is readily intelligible to all. Also, the intended resource backup, Britannica, can veer off without warning from spot on to dilettante nonsense (the latter: "Venetian (a dialect of Italian)" (emphasis added). Barefoot through the chollas (talk) 17:55, 11 January 2023 (UTC)
 * The statement is also literally plagiarized from Encyblopedia Britannica, so it needs to be rewritten for that reason as well. Erinius (talk) 19:09, 11 January 2023 (UTC)
 * And I've rewritten it and added some more from that Encyclopedia Britannica link. Feel free to improve anything I wrote. Erinius (talk) 19:25, 11 January 2023 (UTC)
 * Much better, Erinius. Barefoot through the chollas (talk) 21:53, 11 January 2023 (UTC)

Re Vowel prosthesis: The association of /i/ ~ /j/ and /s/ also led to the vocalization of word-final -s in Italian, Romanian, certain Occitan dialects, and the Spanish dialect of Chocó in Colombia.
This claim should be deleted of it's not fleshed out. As it stands it is not relevant to the syllable structure basis of vowel prosthesis which is the topic of the paragraph, it misstates Sampson's p. 63, and there is no example from chocoano nor reference to it. It could, perhaps, if stated more carefully and illustrated, be weaved into discussion of the fate of final /s/ in the apocope section. Barefoot through the chollas (talk) 02:42, 31 July 2022 (UTC)
 * Almost six months on and it appeared that no one was interested in cleaning up that text, so I've removed it. Barefoot through the chollas (talk) 17:43, 13 January 2023 (UTC)

Information in the lead
On at least three occasions (the first as an IP editing on the 4th of November between 8:30 and 8:39), the user Venezia Friulano has tried to do two things: to remove information from the lead and to add new information to the effect that Spanish ranks third in closeness to Latin. I do not wish to engage in edit warring, so I am simply stating here the reasons why the arguments he provides for those changes are wrong. 1) Regarding similarity to Latin. This user has stated that the sources employed (sources 2, 3 and 4, although 3 and 4 are actually the same) support the idea that Spanish comes third in the similarity ranking. But that’s plainly false. In the first place, the Britannica source only mentions Italian and Sardinian. And, on the other hand, Friulano’s statement that Pei’s percentages concern "phonology, inflection, syntax, vocabulary, and intonation" is, quite simply, wrong. I am quoting literally from the paper, which is open access and can therefore be downloaded from here:
 * "Let us give a brief and very incomplete demonstration of this tentative methodology, which is all that the space at our disposal permits. For the purposes of this demonstration, we shall restrict ourselves to a single division of phonology, accented vocalism, and to seven of the principal Romance varieties: standard literary French, Spanish, Italian, Portuguese and Rumanian, standard literary Old Provençal, and Logudorese Sardinian. A fuller demonstration, including unstressed vowels, consonants, the divisions of morphology and those of syntax, and applied to a larger number of Romance varieties, is reserved for a lengthier work."

It is pretty obvious, by the way, that a twelve-pages paper is simply too short to contain any thorough comparison of "phonology, inflection, syntax, vocabulary, and intonation" in all Romance languages. The author is simply restricting himself to a very specific phenomenon, and he is not even comparing all Romance languages. You only need to check page 138, which is the place where the percentages representing the degree of similarity were extracted from, and you will see that they only concern accented vocalism (and only in a limited set of languages). I.e., you need not even read the full paper.

2) Regarding the removal of Catalan. The user asks "why six languages instead of five". But the question can be reversed: why five instead of six? Only because evolution gave us five fingers per hand? By the same token, it could be argued that the list of languages most similar to Latin should be reduced to just two instead of three (after all, we only have a pair of legs). All six languages are state-level official and I don't see why one of them should be removed (is Andorra any less than Italy or Romania?). As for Friulano's argument to the effect that "The difference in speakers between Romanian and Catalan is large enough", it is debatable at least. If understood in absolute numbers, the difference between both languages is 19.5 million (L1) speakers, which is far less than that between, say, Italian and Romanian (48 million). If understood percentually, the situation is a bit different, for it is true that the difference between the two is the largest in the list (Romanian has five times as many speakers as Catalan), However, I do not think that it is that far away from the second largest difference in the list, namely that between Portuguese and French (3.6 times). And on the other hand, if we considered speakers with a native level instead of just L1 speakers, the numbers would vary drastically (and many sources provide larger numbers, see this and this, for example.

3) Regarding the removal of the dates. I think it is pretty clear that the approximate dates in which Romance languages started to exist is important enough to be left in the introduction. That's how it is done in similar pages such as the one on Semitic languages, whose introduction states even the dates in which they started to be studied scientifically.

4) Regarding the change from "less commonly referred to as Latin languages" to "sometimes referred to as Latin languages". Again, it is quite obvious for anybody familiar with this field of study that the former is the most common name. A simple search on Google (or better, on Google Scholar) should confirm it.

2A0C:5A81:302:1D00:954A:5080:3B4A:5418 (talk) 18:06, 8 November 2022 (UTC)


 * 1)I have not looked into it, but if in fact Pei's work is misquoted and misused then we should intervene. Britannica states that by most measures, Sardinian and Italian are the least divergent while French and Romanian are the most differentiated. No ranking, no mention of Spanish on par with Italian and Sardinian, so i agree with you. That being said, I think User: Venezia Friulano is in good faith but a bit concerned (in general on Wikipedia) that Spain and Spanish are neglected in various articles. Venezia, i think you should look at the positive side here. No one is neglecting Spain and Spanish, no one reading these articles gets that idea.


 * 2)Should we mention Catalan on par with the 5? Honestly, i think the national Romance languages are universally considered to be Spanish, Portuguese, French, Italian and Romanian. I don't think Catalan is generally included in that category. Yes there is Andorra with Catalan, but it's a micro-state; Catalan is more of a regional language.
 * Barjimoa (talk) 11:00, 14 November 2022 (UTC)


 * 1) Well, lets see: "... analyzing the degree of difference from a language's parent (Latin, in the case of Romance languages) indicated the following percentages (the higher the percentage, the greater the distance from Latin): Sardinian 8%, Italian 12%, Spanish 20%, Romanian 23.5%, Occitan 25%, Portuguese 31%, and French 44%" - Source of the article.


 * It is evident that it makes it clear that Spanish is the third closest to Latin, the other sources only indicate that the closest are Sardinian and Italian, which doesnt exclude that the third closest is the Spanish. I don't understand the controversy. If you want, we write all the languages from least to greatest differentiation with Latin. I also see no problem with the source, they are all estimates, not accurate maths. All the studies on the subject are like this, including about Sardinian and Italian, they are methodologies that are difficult to calibrate and are never perfectly complete, they are estimates. Mario Andrew Pei was renowned linguist on the subject, their sources are and have been reliable and used on Wikipedia.


 * 2) I also believe that User:Barjimoa has good faith, but believes that Italy deserves more recognition. No one is neglecting Italy and Italian, no one reading these articles gets that idea. Im Italian, like you, so that point here is worthless, just we dont need Fratelli d'Italia in Wikipedia. I live in Spain and know Spanish, its not a problem if Im mainly interested in Spanish articles.


 * 3) Catalan language is not a main language by many measures, thats why its not necessary to include the 4M of its speakers.
 * Venezia Friulano (talk) 12:07, 14 November 2022 (UTC)
 * User: Venezia Friulano, did you read the discussion between me and 2A0C:5A81:302:1D00:954A:5080:3B4A:5418 (talk)? this quote


 * "...analyzing the degree of difference from a language's parent (Latin, in the case of Romance languages) indicated the following percentages (the higher the percentage, the greater the distance from Latin): Sardinian 8%, Italian 12%, Spanish 20%, Romanian 23.5%, Occitan 25%, Portuguese 31%, and French 44%...""


 * only refers to accented vocalism. The author of the paper is analysing the degree of difference from Latin specifically with regards to accented vocalism. It's not about grammar, not about vocabulary, etc etc. You cannot use it to to make a general statement. The only general statement here is that of Britannica: by most measures, Sardinian and Italian are the least differentiated and French the most divergent. No ranking provided. And regarding you being an Italian that moved to Spain, again I don't see how these articles are pro-italian and anti-spanish.


 * I have to remove that information. It's wrong, it's a mistake.Barjimoa (talk) 14:28, 14 November 2022 (UTC)
 * Well, by itself that source deleted by you, which is valid, even if you don't like it, should be a valid sign of the closeness of Spanish with Latin, there are also quite a few sources (some in the article itself) that indicate that the Spanish is intrinsically close to Latin, even by Romance language standards. The source of Britannica is not even the result of a study that can be seen, it is literally just a mention that Sardinian and Italian are the closest, nothing more, that could be improved, good options exist. I'm wickedly curious as to what you think is the third closest language to Latin. Surely you know that the third most similar is Spanish, but for some reason that fact is annoying. It doesn't like that other languages appear over something related to Latin, only the languages of the Italian peninsula should be highlighted in the lead, just to clumsily make believe indirectly that only Italy is related to Latin and ancient Rome. It is exactly the same case with Trajan and Hadrian, it bothers some Italians that these Roman emperors have Hispanic origins and were not born in the Italian Peninsula, and therefore their Italicity must be mentioned as much as possible in order to make them understand that they are like "Italians" or something exclusively from Italy and its history.
 * The mind in these cases works like this: "Its very annoying that another country that is not Italy can have or be proud of something Roman, and on top of that it is about Trajan! The Roman emperor who led the Roman Empire to its maximum historical extension, what a humiliation for us Italians! No, we can't allow it, that fact must be Italianized! Christopher Columbus? He was not a Genoese from the Republic of Genoa. Even though Italy and Italian were just geographical terms in that period, like "Iberia" and "Iberian", we must isolate they can read the fact that he was Genoese at all cost, and imply that he somewhat belongs to the country of Italy. He was Italian! Like the current ones!"
 * It is the classic paranoia and rhetoric of Italian fascist nationalism of the early and mid-20th century: Feverishly thinking that Italy, as a country, is millennia old and establishing an anachronistic continuity between ancient Rome and present-day Italy, a country born in the middle of the 19th century, claiming that Roman heritage, history and culture is an exclusively "Italian" phenomenon, and therefore only applicable to the current Italian population the right to be heirs of ancient Rome, repudiating other nationalities of Latin heritage as mere borrowers of "their culture".
 * That is the idiosyncrasy and politics of Italian fascism for much of the 20th century, and the seed of the problems and shame it inflicted on itself. Of course, I'm not saying that you are that, but there are traces and I recognize well an Italian nationalist. I am not at all surprised by the current political situation and inclination of Italy or that it has the only far-right government in Western Europe.
 * You can pretend I've made a movie out of this or that Im exaggerating, of course, but we both know the reality and how it works. Anyway, I'm not going to follow an Edit Warring or continue talking about the topic. I do not want to be fined or expelled, besides, it is a waste of time considering the panorama.
 * All the best. Venezia Friulano (talk) 16:17, 14 November 2022 (UTC)
 * I agree that Spanish is 3rd closest, but too far bro with Italy stuff. 173.163.240.165 (talk) 17:32, 28 March 2023 (UTC)
 * Since the question has been brought up, I'll take advantage to explain my point of view: I think that the closeness of Romance languages and Latin is a matter of a very secondary relevance. If anyone that knows a couple of Romance languages and also a little bit of Classical Latin (which is probably the case with most editors of this page) analyzes objectively those languages, I'm convinced that the person will arrive to the conclusion that whatever chosen Romance languages are clearly closer to each other than to Classical Latin. In fact I'm pretty sure I have read something like that from a well-known author, but unfortunately I can't remember which one. :( --Jotamar (talk) 22:10, 16 November 2022 (UTC)
 * Good points. First off, other than satisfying idle curiosity, what's the point of observing that Romance language x is closer to Latin than Romance language b? And even if curiosity itself licenses the comparison, by which criteria is the comparison established? The oft-cited study from decades ago is phonological. A meaningful comparison would include lexicon, morphology, and syntax, at the very least. Any one of those categories requires cherry picking and artful dodging, even at the level of basic lexicon, leading to the question "What do you mean by Latin?". Parler / parlare and falar / hablar reveal two sets of reflexes from two distinct Latin verbs meaning 'speak', and something like latine loquor complicates the picture further, a picture that blurs more with slightly more complex vīsne mēcum latīnē loquī. "I'm hungry" is about as basic and banal as it can get, yet calculating the distance between esurio and mi-e foame / ho fame / j'ai faim / tengo hambre [...] is more than a bit of a challenge (and Catalan tinc gana is not at all helpful). Etc. Barefoot through the chollas (talk) 21:46, 8 April 2023 (UTC)

Excessive examples
As suggested by @Barefoot through the chollas, I am taking up my issue with the samples section on the talk page (although there are other sections with the same problem, I'll restrict my arguments to the samples section for now). Currently, this section consists mostly of a very long list of examples followed by an unsourced bit of prose. The point of examples is to show the breadth and scope of something without having to exhaustively list every single item (in this example, languages). I quote from MOS:LONGSEQ: "Keep lists and tables as short as feasible for their purpose and scope: material within a list should relate to the article topic without going into unnecessary detail; and statistical data kept to a minimum per policy." I believe that the samples section does just that; rather than keeping it to a minimum it's seems to invite "maximums", which everyone translating their favorite Romance language and adding it to the list. I can't even fit all of them on one page, and I have a decently large monitor. Excessive examples is something that is a problem with a lot of language articles I have found.

Having made my point, I now ask other people on this page what should be done. As far as I see, these are the options:

A: Keep it as is and allow more examples to be added freely.

B: Make it a collapsible list.

C: Limit the examples as recommended by the policy (how to choose which ones to cut will be decided later)

I vote for C. I think that curating the examples will provide a much better reader experience rather than bombarding them with minute differences between closely related varieties. I have explained why I don't like option A, and I'm willing to discuss option B if there is enough interest.

Megaman en m (talk) 23:43, 11 September 2021 (UTC)
 * Not the central point now, but let me say that I agree with an opinion that a lot of the text in the Romance article can and should be condensed, and much of it left to e.g. Hispano-Romance, Gallo-Italian and the like.


 * To the point, I also agree entirely with the guiding principle "Keep lists and tables as short as feasible for their purpose and scope". The question at issue is the purpose of the list of Romance renditions of the sentence, and what the list's scope should be.


 * Cap'n Obvious here: One of the many great merits of Wikipedia is enabling a wide gamut of readers to find information of interest to them at the click of a mouse. Some are "just" curious, some are seriously searching for knowledge. Judging from experience, the latter especially, but also many of the former once they get a whiff of what's going on, want an answer to "How many Romance languages are there?" The article skirts that question with Dalby's scanty undercount, followed by a list of typological groupings with a few languages listed for each (but not for all: Sardinian tout court?). Not until Samples does the reader get a clear view of some of the vast variety of Romance.


 * I vote for C above, but not in a version that necessarily implies cutting examples (nor that totally precludes judicious additions). With that, some minor culling might be in order, given that sub-types are not treated uniformly. Just one version for Occitan, Picard, Sicilian, for example, but three Emilian types and two Lombard. I wouldn't be against reducing Emilian to one major type, perhaps Bolognese, and similar for Lombard, but with the location identified, again a major center if possible. Ladin should be reduced. Some relabeling is called for, such as Campidanese and Logudorese rather than the vague Southern and Northern Sardinian, and, if they can be established, locales for what's listed now as Northern and Southern Corsican.


 * The vote for C assumes a brief paragraph under Samples that introduces the list and explains that it is only a sampler, not by any means complete -- perhaps illustrating there the difficulty of achieving a full count of Romance languages by comparing Reggiano and Bolognese very briefly, mutually understandable and only 60 kilometers apart, yet distinct in grammatical detail. Barefoot through the chollas (talk) 03:57, 12 September 2021 (UTC)
 * Hi. For what it's worth, I copied this long list of examples in case it (or parts of it) disappear. Fascinating! As a place for such fascination I suggest (without careful deliberation) an option D, a Wikipedia Salon des Refusés, or "exhibition of rejects." I don't dispute the rule "Keep lists and tables as short as feasible for their purpose and scope," except to turn variations around in my mind: Keep knowledge as slight as feasible, or, keep fascination as slight as possible. Ugh, no. Like our biology, our ideas evolve (endlessly we hope), including our languages, our knowledge of languages, and the language with which we describe languages. Evolution needs abundant mutations that can at first strike us as a mess or ferment, and I for one redirect my suspicions to any kind of finality or order that fights rather than promotes that evolutionary process. Can't we in some way accommodate this, for the many Wikipedia readers who link from one article to another to another, as time allows, and come away knowing more than they knew to plan for? John Hicks (talk) 12:37, 29 November 2022 (UTC)
 * Couldn't the larger list simply move to another page? Even as a "list of" page? I do agree that some people will appreciate the details, but we should find ways to accomodate multiple audiences, perhaps. Jim Killock (talk) 06:52, 18 April 2023 (UTC)

Counting and listing Romance languages in the lead
This archived discussion is still relevant Erinius (talk) 23:00, 13 January 2023 (UTC)


 * Thought I'd revisit this now that the Dalby URL no longer works (at least it's available on the Internet Archive).
 * Something I think worth mentioning is that several reliable sources (The Cambridge History of the Romance Languages and The Oxford Guide to the Romance Languages) I've found just don't even attempt a comprehensive listing or counting of Romance languages. The Encyclopedia Britannica's article on the Romance languages doesn't attempt a comprehensive list either, although it does offer a rather unsatisfactory map of the Romance languages in Europe.
 * Points made by and  are still relevant - the current "list" and paragraph above still have their problems. They can still mislead readers into thinking they're a complete listing of all Romance languages, they don't go far enough in explaining the Romance dialect continuum for the unaware reader, and they can give the impression that mutual intelligibility is an unproblematic criterion for establishing different languages. Also, they can give the impression that the largely "geohistoric" language groups used (ie Ibero-Romance) are valid, well-established and uncontroversial clades with no overlap between them.
 * Also, the current list in the article's lead isn't even the same as Dalby's list.
 * A whole "Languages" section may be too ambitious for now, but it might be possible to fix up this vague list of Romance language areas in the lead and its associated paragraph and avoid misleading any uninformed readers. Erinius (talk) 23:40, 10 April 2023 (UTC)
 * Erinius, bullseye. The editors of the CUP and OUP works you mention are serious Romance linguists who know their stuff, and (thus) are not going to engage in the fool's errand of attempting to count Romance languages. Illustrated explanation of why not could fill a thick tome, but just clicking on Occitan in Dalby's typological list eventually leads to the crux of the impasse:
 * "...there is no single written standard language called "Occitan" [...] "Occitan is fundamentally defined by its dialects, rather than being a unitary language. That point is very conflictual in Southern France, as many people do not recognize Occitan as a real language and think that the next defined "dialects" are languages." [...] every settlement technically has its own dialect, with the whole of Occitania forming a classic dialect continuum that changes gradually along any path from one side to the other."
 * The same can be said of the subtypes listed that haven't been subjected to what someone once called "the dead hand of standardization" -- and note even with that, the odd classification French/Oïl languages, i.e. highly engineered national standard along with autochthonous local "dialects" (in the terminology adopted for Occitan).
 * As you say, the list is very misleading as it is, in more than one way. But it could be salvaged as layered typologies with some judicious editing and the appropriately informative text, making it clear that the list is not of "languages" but of types, all of which exist in varieties, such as Gallo-Romance > Oïl languages > Picard (> Rouchi, etc.). Oh, and drop, explaining why, the notion that Romance languages can be counted. Barefoot through the chollas (talk) 19:15, 11 April 2023 (UTC)
 * A "Languages" section is not that hard to create. As a first emergency step, just move the final paragraph of the lede down to a section of its own. The section "Samples" also needs to be moved down.
 * But: I've just intended to do that and scrolled down to find a proper place, down and down and down...the page is endlessly long! The section "Classification" has 40k of code (the corresponding "main article" only 30k!), "Sound changes" has 60k! The first is a classic case of content forking, and the second one is a highly notable topic of its own ("Phonological history of the Romance languages"). Let's put this into a more readable shape. –Austronesier (talk) 20:37, 11 April 2023 (UTC)
 * There are some big swaths of content that are just unsourced (including in the "Classification" section), possibly misleading and/or largely irrelevant - I'm hoping none of you would object to them being deleted. Likewise the "samples" section seems to be just a vague listing of as many Romance languages as possible, alongside translations of an example sentence. It's possible that one or two of the translations for a language with relatively few speakers is totally inaccurate, but since none of them are sourced, who would know?
 * I agree with you that a simple "Languages" section would be a good idea, and I think it would make the most sense to put it right above "Classification". Erinius (talk) 21:25, 11 April 2023 (UTC)
 * Oh and of course I agree that much of "Classification" should be transferred to the subject's main article. As for "Sound changes", there is already a main article "... to Proto-Romance" and the related article on Vulgar Latin, although I think a lot of these sound changes would take place a little after "proto-Romance". Erinius (talk) 21:30, 11 April 2023 (UTC)
 * I just did a preliminary rewrite of the paragraph in question - I'd appreciate further changes to phrasing, since I really wasn't sure how to explain things best. Also, IIRC, the Encyclopedia Britannica article, and the "Classifications" chapter of The Oxford Guide to the Romance Languages, explain why counting the Romance languages is impossible, a bit about how the whole language vs dialect thing is based on social factors as well as purely linguistic ones, and how a lot of somewhat medium-level classifications (ie Ibero-Romance) are largely geographical. Erinius (talk) 23:58, 25 April 2023 (UTC)
 * Erinius, pressed for time just now, but I've made one little tweak, from various languages to numerous languages, to open the way for what's to come. I think you've done an excellent job of editing. Barefoot through the chollas (talk) 01:01, 26 April 2023 (UTC)
 * "One, two, three, many". might be interested in this. 04:28, 12 April 2023 (UTC)

Content to remove
While a lot of the "Classification" section should be transferred to the corresponding main article, and most of "Sound Changes" should also go in a separate article, there's a lot of content on this page that I think should be removed entirely because it's either totally unsourced (and possibly misleading) or totally irrelevant to this article's scope. I don't want to start deleting things en masse, and I can see reasonable arguments being made in favor of keeping some of these things, so I'll just make a list here of everything I think should be purged from the article: My current mood is that we should delete first and, if citations show that certain content actually should be included, restore later. I'd appreciate any feedback. Erinius (talk) 23:42, 17 April 2023 (UTC)
 * the "Samples" section - Totally unsourced, might not be entirely relevant for this article, possible that one or two of the smaller languages have really unnatural or inaccurate translations (who would know since it isn't sourced), and the "Vulgar Latin" example seems really reconstructed-y
 * Within "Classification":
 * Within "Proposed Divisions"
 * The giant table showing verb conjugations in different languages, supposedly ordered from conservative to innovative - unsourced which leaves the door open to possible errors, not very relevant for a classification of the Romance langs, selection of languages may be questionable, the relative ordering of languages in relation to each other as more/less innovatory may also be questionable and seems OR-y. Selection of verb conjugations is arbitrary, and the table falsely presents each language's verb forms as largely mapping onto the others' one-to-one
 * The "conservative" vs "innovatory" bullet point - neither of its sources actually establish conservative vs innovatory as a widely-used way of classifying different Romance languages (sorry, Dante!)
 * Three consecutive paragraphs beginning with "The usual solution to these issues...", "Probably a more accurate description..." and "This would explain why some of the "northwest" features..." - unreferenced and misleading
 * " Outcome of stressed Classical Latin vowels in dialects of southern Italy, Sardinia and Corsica" - It's likely that all the info in that table is accurate and that it's compiled from one or more good sources - but the question marks on the reflexes for Latin /au/ and the parentheses and the lack of citations make me inclined to delete it
 * Gallo-Romance languages - not particularly relevant for a very high-level classification of Romance dialects, and uncited
 * Conlangs - Not sure how they're relevant to this page
 * "Writing systems" - "Romance orthographies" actually is a coherent topic (which surprised me - Encyclopedia Britannica article does include a section on orthographies, and the Oxford Guide to the Romance Languages has a chapter on writing systems as well) - but this section doesn't need to be as long as it currently is - it certainly doesn't need to include every consonant digraph and trigraph
 * Vocabulary comparison - the big table feels OR-y, while the lexical similarity table could be moved to the "classification" page


 * This seems reasonable; I've a few questions, as I feel the Latin pages are suffering from similar problems of focus / bloat.
 * does WP have a guide or a good model by example of how a language topic should be constructed?
 * or are there other examples; for instance, what topics does Britannica cover in its Romance languages article?
 * is it possible to spin sub-pages off for any of this where the content is decent, eg "writing systems"?
 * In general, these pages seem to fill with changes to linguistic patterns, and other quite technical information, which may interest someone with a deeper interest in linguistics, while the page contents tend to lack thematic discussions or other information which would engage someone whose interest was more general, I would say. Jim Killock (talk) 06:47, 18 April 2023 (UTC)
 * Super pressed for time just now, but I'm chiming in -- I hope it's clear in a collegial spirit -- to urge restraint in executing bold deletions and other sweeping changes. Issues in no specific order...
 * Samples. Such a list seems very useful in that most readers probably have no idea of the variegated richness of Romance, find themselves stuck at national languages with maybe Catalan and a couple of others tossed in. The sample list also makes it clear that the little list under Languages is the iceberg's tip only, not exhaustive and not intended to be (notwithstanding the unfortunate Dalby input). Agreed that the list is bloated in terms of variants in subgroups. Emilian, Ladin, Lombard, for example; one will do, readers can turn to specific articles for detail. The main issue in Samples is attribution. That does need resolution, which a concerted effort should be able to reach. ("Vulgar Latin" is pretty much by definition a reconstruction; there's no real need to have it in the list, though.)
 * Writing systems. This is pretty well done, though running text can be made a good bit more concise. Seems to me the question is whether to edit carefully or just present a summary paragraph of high points in this article, and move the detail to an article "Romance language writing systems".
 * Vocabulary comparison. Not just sound shifts, but also lexical selection, and, like Samples, surely informative for the neophyte. A bit of a mess as it is, though. Especially for that readership who can't correct as they read (no, Italian 'man' is not phonetically [ˈwɔmo], but [ˈwɔːmo]), it needs a lot of cleaning up for inconsistencies and inaccuracies in IPA -- though presented as phonetic, many entries look phonemic, or partially one, partially the other. And then there's French, all phonemic, yet said to represent pronunciation. OR doesn't seem to me an issue, as they're all basic dictionary entries easily available to one and all. Definitely check suspect entries for accuracy, though: e.g. is Sicilian 'water' really [ˈakːua], with [u]?
 * Classification. If the main article were a good bit fleshed out, I'd say sure, just a paragraph or two here, with a see... the main article. But the main article is still far from containing even the (à mon avis, useful) Extent of variation table and much else that's informative that is present in this Romance article.
 * Basta. Must run. Once again, I'm just urging let's take it easy. Babies and bathwater sort of thing. Barefoot through the chollas (talk) 17:34, 18 April 2023 (UTC)
 * Maybe the first issue is actually the content order. The sections run in an odd order. The most helpful parts are probably "modern status" and "history" which are in the second half of the running order; "name" and "Languages" are also introductory, while "samples" is to my mind background information. I would also give these sections more helpful titles, eg:
 * The Romance Language family [combine 'name' and 'languages']
 * Romance languages today
 * History
 * On 'classification of Romance languages', move this to a separate article and add a simmar to part 1 above
 * Then look at the rest of the page and work out if it needs slimming down / separating out Jim Killock (talk) 18:51, 18 April 2023 (UTC)
 * Thanks for the replies! I won't go ahead with any big deletions. Erinius (talk) 22:54, 19 April 2023 (UTC)
 * I completely agree. It is very odd to find the History chapter for example mid-way trough the article. Aristeus01 (talk) 12:39, 19 July 2023 (UTC)

Sound changes
Looking at the article flow a big chunk in the middle is taken by the "Sound changes" part. The topic is too complex to be compressed in a satisfactory manner. I suggest moving most of its content to "Phonological changes from Classical Latin to Proto-Romance". It would add to the main article's depth, and, at the same time, improve readability of the Romance languages article. Aristeus01 (talk) 13:07, 19 July 2023 (UTC)


 * Agree, and I'm not sure anyone would disagree. Most of that content should be in a separate article. A word of caution though, not all the changes in the "sound changes" part affected all of Romance, so they wouldn't strictly be sound changes to Proto-Romance. I'm not totally sure what should be done about those. Palatalization which affected all of Romance to some degree, but some varieties more extensively, is already briefly mentioned in Phonological changes... so I think transferring most of this article's Palatalization section there would be fine. I guess we should also see what some of the main contributors to the Proto-Romance articles (namely ) have to say and I'm also curious if other language families have their own sound changes articles. Erinius (talk) 16:15, 19 July 2023 (UTC)
 * You could certainly add a section labelled 'subsequent regional sound changes', or similar, to the article about Classical Latin > Proto-Romance changes. Nicodene (talk) 16:59, 19 July 2023 (UTC)
 * I ask only that whatever content transferred there be cited to the same standard as the rest of the existing article. Nicodene (talk) 17:00, 19 July 2023 (UTC)
 * A fair request. We should start by adding refences to our current text in question and when we are happy with the result we can look again at all the suggestions and decide on the best way to move forward. Aristeus01 (talk) 18:27, 19 July 2023 (UTC)

Agreed that the historical phonology in the Romance languages article is in some ways too detailed -- though in an article of its own, one could argue not nearly detailed enough, starting with too many languages not even mentioned. Before venturing into that quagmire, though, Phonological changes from Classical Latin to Proto-Romance as basis for a revised more complete version would have to be re-named ("moved"), but the present title (reduced to Latin to Romance) and content could provide an excellent first portion if the new article were to be articulated in two phases, either Latin to Romance, then local developments as two large sections, or each feature or set of features in two phases. The present text would need some massaging (e.g. glosses supplied, sources updated, statements such as this cleaned up: In cases where a long vowel precedes a geminate consonant, one of the elements often shortens unpredictably). The other task, deciding on how much historical phonology remains in Romance languages, is a bit thorny, but it can be done. Barefoot through the chollas (talk) 18:21, 19 July 2023 (UTC)

Length of the Roman rule and the Romance Languages Map
Proposing to remove the map due to not meeting quality standards of an academic material. General issues:
 * it arbitrarily and subjectively sets the time frames based on the length of rule of Dacia and Britannia, then on a 100 years basis, then on a generic "over 500" with no clear explanation of the relevance of such categorisation.
 * it sets the end of Roman rule on the end date for the Western Roman Empire, misrepresenting the rule in Moesia that lasted until the 6th century under Eastern Roman Empire ( so "over 500 years")
 * it ignores the difference between rule in most of Italian peninsula and the other areas presented as "over 500".
 * it misrepresents the gradual conquest of Iberian peninsula that took over 200 years, possibly misleading the reader to think it happened in two waves.
 * as above it makes no distinction between southern France and the territories added by Julius Caesar.
 * it completely misrepresents North African provinces and timeline, including the geographical extent of Roman border along Libyan coast, and the length of rule in Mauretania Caesariensis ( should be "over 500")

Language specific issues:
 * it mixes representation of national languages with regionally recognised languages, oddly showing only some of the latter.
 * it draws the Jireček Line in an amateurish manner, following neither Jireček's indications nor one of the latter academic proposals.
 * controversially assigns "Romanian" speaking label to Aromanian language (which is very poorly illustrated, by the way).
 * confusingly shows a single clade of languages ("Galloitalic") touching on the subject of clades then refuses to elaborate.

Overall, the map adds little to no value to Romance languages' history and grouping, and poorly illustrates the chronology of the Roman Empire and its limits. Aristeus01 (talk) 10:00, 27 August 2023 (UTC)


 * 1. It is a well known fact that most Romance languages are spoken in lands that were under Roman rule for more than 500 years, so the time frames are not arbitrary in the article's context.
 * 2. Roman rule came to an abrupt end along the Lower Danube with the arrival of the Huns in the first half of the 5th century, and the Romans reclaimed the territory only decades after the fall of the Hunnic Empire.
 * 3. And why is it important?
 * 4. According to the map, the conquest of the Iberian lasted for about two centuries.
 * 5. Why should if both territories were ruled by the Romans over 500 years?
 * 6. We should not ignore that Mauretania Caesarensis was lost to the Vandals in the early 5th century and reconquered by Justinian in the 530s.
 * 7. I think you refer to Gallego and Gallo-Italic when speaking of "regional languages". Actually, the list of Romance languages is still uncertain.
 * 8. New archeological finds continuously move the Jireček Line since it is a theoretical line based on inscriptions.
 * 9. In scholarly literature it is not unusual that Aromanian (or Macedo-Romanian) is presented together with Daco-Romanian, Megleno-Romanian, and Istro-Romanian (see, for instance ) Borsoka (talk) 11:13, 27 August 2023 (UTC)
 * I understand your concerns; however, I feel that the map should be improved, rather than deleted. The original author in Commons is Andras Bereznay (probably no longer active), and it's been modified by commons:User:Liviojavi. Why don't you try to contact them? --Jotamar (talk) 11:18, 27 August 2023 (UTC)
 * A re-do is definitely in order if the map is to be kept. In addition to sorting out the time questions, there's this concern: "It mixes representation of national languages with regionally recognised languages, oddly showing only some of the latter.", troublesome in multiple ways. If regionally recognised means the usual here, the designation is far more political than linguistic, thus misleading at best, and guaranteed to be less informative than might be helpful to non-expert readers. This is then compounded by the labeling of e.g. French through almost all of France and Italian for most of Italy, with nods to very few local languages or typologies (Galloitalic), not just misleading, but misinforming the non-expert. None of this is easy to deal with in a small map, but attempts at coherent language labeling can be made. Barefoot through the chollas (talk) 14:28, 27 August 2023 (UTC)
 * I've contacted both editors a few days ago bit so far no reply. It's possible that the map, with improvements, could be useful for the topic but I do not have high hopes the authors will make any changes. Aristeus01 (talk) 09:17, 3 September 2023 (UTC)

Historical stages of palatalization
I've been combing through the sources available to me and none seem to mention the 5 stage model presented in the Wikitable at the Palatalization paragraph (see for example: Palatalizations in the Romance Languages by Daniel Recasens - Oxford Research Encyclopedias site). Could someone with a better knowledge of the topic confirm this is not OR? Aristeus01 (talk) 16:23, 8 September 2023 (UTC)


 * It's not, or not necessarily, OR, but it could use some sourcing, even if as basic as Boyd-Bowman's From Latin to Romance in sound charts. Barefoot through the chollas (talk) 18:02, 8 September 2023 (UTC)
 * Decided to scan through From Latin to Romance in sound charts and I'll note here where the changes in question are mentioned - the palatalizations in that chart are mentioned in rules 5 (CA-, GA-, pp. 28-31), 6 (k-/g- before e/i/j, pp. 32-34), 18 (yod, gj-/dj-, gi/ge, these consonants all merging as yod, p. 65), 19 (intervocalic yod/-gj-/-dj-/-gi-/-ge-, merged as yod, p. 68), 22 (-lj-, p. 77), 26 (-gn- and nj-, p. 89), 32 (p/r/s/m/b/v before j, p. 106), and 40 (tj, p. 129). Also there are sections on the development of the Spanish and French sibilants.
 * I haven't read through the notes on each palatalizing environment, so I can't say for sure whether or not they support the model presented there. Erinius (talk) 22:32, 8 September 2023 (UTC)
 * Guess I should add - I accessed it through HathiTrust - url here: https://babel.hathitrust.org/cgi/pt?id=uiug.30112058502060&seq=7 The scan is NOT perfect, the left edges of some pages have been cut off Erinius (talk) 10:26, 9 September 2023 (UTC)
 * @Barefoot through the chollas and @Erinius, thank you for the source and indications! I added a few refs from it, will add more as time permits. I'm still of the opinion the 5 stage model might need to be part of a larger conversation on timeline and phases of palatalization, but that would probably be better to do after moving the paragraph. Aristeus01 (talk) 11:35, 9 September 2023 (UTC)
 * I also have been reading about the outcomes of palatalizations. I just corrected a blatant error (this article formerly indicated that Sardinian did not palatalize kj and gj, which is false) but the overall topic seems to have many complex details ("The Early History of Romance Palatalizations", Marcello Barbato 20 June 2022, is another helpful resource that can be accessed via the Wikipedia Library). I think it would be a good idea to split "Palatalization in the Romance languages" out to a separate article so it can get more room for coverage and so that this article can be shortened; does that sound like a good idea?--Urszag (talk) 17:01, 9 September 2023 (UTC)
 * Sure. It sounds to me like that absolutely has enough coverage and notability to be its own article. Erinius (talk) 00:33, 10 September 2023 (UTC)
 * I agree. At this stage the paragraph is perhaps too detailed for the article but in an article of its own could and should be expanded considerably. I suggest we keep the first 3 lines and the table as brief description while the body can be moved to the new article. Aristeus01 (talk) 12:18, 10 September 2023 (UTC)
 * Sounds like a good idea to me. At present it suffers from lack of examples in the table; perhaps those could be included. Also, a decision should be made as to what to do with -NN- and -LL- > /ɲ/ and /ʎ/ and initial PL-, CL-, FL- in Spanish (and a few other languages), and whether to treat historical restructuring and modern allophony together or separately. Barefoot through the chollas (talk) 20:44, 10 September 2023 (UTC)
 * Move completed. Please see Palatalization in the Romance languages. Aristeus01 (talk) 14:20, 11 September 2023 (UTC)

Portuguese: first, or second language for Africans?
The article says Portuguese "is spoken as a first language by perhaps 30 million residents of that continent, most of them second-language speakers." It looks like a contradiction to me. Kotabatubara (talk) 22:58, 12 August 2022 (UTC)
 * My bad. I replaced first language with primary language, which doesn't imply native. A quick search on the internet makes me think that the figures for Portuguese speakers in Africa are not reliable, and in any case they're growing with every year. --Jotamar (talk) 06:17, 13 August 2022 (UTC)
 * Hi,
 * Is appears now that they are native speakers now.
 * Native Dark Blue Circle in Luanda, Angola per sources:
 * Angola in particular - Portuguese is the native language of almost half the country.
 * Link 1:
 * https://www.worldatlas.com/articles/what-languages-are-spoken-in-angola.html#:~:text=The%20official%20language%20of%20Angola,most%20spoken%20language%20of%20Angola.
 * " According to a 2014 study, nearly 71% of the population of Angola speak Portuguese. The language is the mother tongue of 39% of the population of the country while many more speak it as a second language. Portuguese is also the most spoken and sometimes the only language that is spoken by younger Angolans. Portuguese speakers are more common in the urban areas of Angola where 85% of the population converse in Portuguese. "
 * Link 2:
 * Other sources put the number much higher:
 * " The 2016 CIA World Fact Book reports that 12.3 million, or 47% of the population, speaks Portuguese as their first language. However, many parents raise their children to speak only Portuguese"
 * Angolan Portuguese
 * https://www.britannica.com/place/Angola
 * "Portuguese is often the only language spoken in Luanda and in much of the interior extending beyond the city and in other parts of the country"
 * Basically the whole coast only uses Portuguese, with native languages being used in land. IntelloFR (talk) 01:06, 14 August 2023 (UTC)[ reply]
 * Have Kikongo and Umbundu really given way to Portuguese on the coast? I would think those two languages at least would be robust. — kwami (talk) 02:14, 14 August 2023 (UTC)
 * 2. Native dark blue circles in Libreville Gabon and Younde, Cameroon, and Lafayette LA.
 * Libreville, Gabon:
 * "It has been noted that French is increasingly be-coming the mother tongue and the initial language of the younger generations in urban Gabon (Pambou, 1998:147; Ndinga-Koumba-Binza, 2005a:72 & 2005b:141; Idiata, 2008:85; cf. Blanchon, 1994). In fact, studies by Ntong Amvame (1984), Bouché (1998), Mbondzi (1998), Ompoussa (1998), Itembo (1999) and Mouloungui Nguimbyt (2002) have shown that pupils of various ages and grades at schools learn French more efficiently than any other Gabonese language. Idiata (2008:200 & 2009:126) has also noted that some pupils do not speak any of the Gabonese native languages at all. One of the reasons for this phenomenon (i.e. French being the mother tongue of younger generations) is cross-ethnic marriages.
 * In fact, many couples of mixed ethnicity prefer French rather than Gabonese native languages as the code for better communication within the family. Children from a family of this kind have no choice but to acquire French as their first language. The children learn the language at home from the parents before they even get to school, therefore lessening the chances of learning any of the Gabonese native languages.
 * This urbanisation is also to be considered as a cause for French being the initial language of Gabonese younger generations. In fact, “in certain urban contexts there is a large degree of learning by contact at an early age” (Lafage, 1993:216)."
 * While this does not give a specific number, it states clearly that language transmission in the urban areas is French to French - making French natively spoken per evidence in Gabon.
 * 2. Younde, Cameroon:
 * In Cameroon, per research: "French usage twenty years ago and in 2004 show a loss of the LWDs, which goes along with a loss of Cameroonian languages in urban areas. Bitjaa Kody has further found that the national languages are disappearing even in endogamic households where family members speak the same national language. Francophone adults used French in 42 percent of the domestic communications which were studied, whereas the young (10–17 years old) used French in 70 percent of the communication. In addition 32 percent of the young between ten and seventeen years old interviewed in Yaounde did not know any national language and had French as their L1 (Bitjaa Kody 2001a). There is a clear change in language use from the parent generation to the generation of their children (Bitjaa Kody 2005:95). As a result, there will be even less national language speakers in future, since the future parent generation will not be able to transmit a Cameroonian language to their children."
 * Additionally here, this map shows Douala at 91% Francophone: https://translatorswithoutborders.org/wp-content/uploads/2021/11/Cameroon-Language-Map-option-2.png
 * This again does not give a specific number, but points to the fact that in urban Cameroon, language transmission is French to French - showing French is spoken natively.
 * 3. Lafayette, Louisiana :
 * The history of French in Louisiana has not allowed the vast majority of francophones the opportunity to learn to read and write French. CODOFIL states, In the 1990 census, approximately 250,000 Louisianians responded that French was the language they spoke at home. The 2000 census showed 198,784 Louisiana francophones over the age of 5, including 4,470 who speak Creole French. Programs such as the Centre de la francophonie des Amériques and CODOFIL continue to identify future avenues of research and encourage francophone to produce knowledge that will lay the foundations for the future.
 * "This generation of new speakers, often younger than 40, represents a generational shift that developed after home transmission of French faded in the mid to late 20th century. Their grandparents likely spoke French as a first language, "
 * https://www.theadvertiser.com/story/news/local/louisiana/2023/05/30/sustainability-and-economic-developments-of-the-french-language-in-louisiana/70244915007/
 * https://www.theadvocate.com/curious_louisiana/curious-louisiana-how-many-people-speak-cajun-french-in-our-state/article_947551b6-a982-11ed-b305-d7c4959e9102.html
 * https://www.krvs.org/2023-05-30/new-speakers-of-french-in-louisiana-continuing-a-legacy
 * @Kwamikagami IntelloFR (talk) 03:29, 15 August 2023 (UTC)
 * @Kotabatubara @Jotamar IntelloFR (talk) 03:30, 15 August 2023 (UTC)
 * @Salvabl @Kwamikagami Open to your thoughts here IntelloFR (talk) 16:04, 13 October 2023 (UTC)
 * Sorry, I'm out of towwn and haven't had an opportunity to look into this further. — kwami (talk) 23:47, 13 October 2023 (UTC)

Counting and listing Romance languages in the lead
This archived discussion is still relevant Erinius (talk) 23:00, 13 January 2023 (UTC)


 * Thought I'd revisit this now that the Dalby URL no longer works (at least it's available on the Internet Archive).
 * Something I think worth mentioning is that several reliable sources (The Cambridge History of the Romance Languages and The Oxford Guide to the Romance Languages) I've found just don't even attempt a comprehensive listing or counting of Romance languages. The Encyclopedia Britannica's article on the Romance languages doesn't attempt a comprehensive list either, although it does offer a rather unsatisfactory map of the Romance languages in Europe.
 * Points made by and  are still relevant - the current "list" and paragraph above still have their problems. They can still mislead readers into thinking they're a complete listing of all Romance languages, they don't go far enough in explaining the Romance dialect continuum for the unaware reader, and they can give the impression that mutual intelligibility is an unproblematic criterion for establishing different languages. Also, they can give the impression that the largely "geohistoric" language groups used (ie Ibero-Romance) are valid, well-established and uncontroversial clades with no overlap between them.
 * Also, the current list in the article's lead isn't even the same as Dalby's list.
 * A whole "Languages" section may be too ambitious for now, but it might be possible to fix up this vague list of Romance language areas in the lead and its associated paragraph and avoid misleading any uninformed readers. Erinius (talk) 23:40, 10 April 2023 (UTC)
 * Erinius, bullseye. The editors of the CUP and OUP works you mention are serious Romance linguists who know their stuff, and (thus) are not going to engage in the fool's errand of attempting to count Romance languages. Illustrated explanation of why not could fill a thick tome, but just clicking on Occitan in Dalby's typological list eventually leads to the crux of the impasse:
 * "...there is no single written standard language called "Occitan" [...] "Occitan is fundamentally defined by its dialects, rather than being a unitary language. That point is very conflictual in Southern France, as many people do not recognize Occitan as a real language and think that the next defined "dialects" are languages." [...] every settlement technically has its own dialect, with the whole of Occitania forming a classic dialect continuum that changes gradually along any path from one side to the other."
 * The same can be said of the subtypes listed that haven't been subjected to what someone once called "the dead hand of standardization" -- and note even with that, the odd classification French/Oïl languages, i.e. highly engineered national standard along with autochthonous local "dialects" (in the terminology adopted for Occitan).
 * As you say, the list is very misleading as it is, in more than one way. But it could be salvaged as layered typologies with some judicious editing and the appropriately informative text, making it clear that the list is not of "languages" but of types, all of which exist in varieties, such as Gallo-Romance > Oïl languages > Picard (> Rouchi, etc.). Oh, and drop, explaining why, the notion that Romance languages can be counted. Barefoot through the chollas (talk) 19:15, 11 April 2023 (UTC)
 * A "Languages" section is not that hard to create. As a first emergency step, just move the final paragraph of the lede down to a section of its own. The section "Samples" also needs to be moved down.
 * But: I've just intended to do that and scrolled down to find a proper place, down and down and down...the page is endlessly long! The section "Classification" has 40k of code (the corresponding "main article" only 30k!), "Sound changes" has 60k! The first is a classic case of content forking, and the second one is a highly notable topic of its own ("Phonological history of the Romance languages"). Let's put this into a more readable shape. –Austronesier (talk) 20:37, 11 April 2023 (UTC)
 * There are some big swaths of content that are just unsourced (including in the "Classification" section), possibly misleading and/or largely irrelevant - I'm hoping none of you would object to them being deleted. Likewise the "samples" section seems to be just a vague listing of as many Romance languages as possible, alongside translations of an example sentence. It's possible that one or two of the translations for a language with relatively few speakers is totally inaccurate, but since none of them are sourced, who would know?
 * I agree with you that a simple "Languages" section would be a good idea, and I think it would make the most sense to put it right above "Classification". Erinius (talk) 21:25, 11 April 2023 (UTC)
 * Oh and of course I agree that much of "Classification" should be transferred to the subject's main article. As for "Sound changes", there is already a main article "... to Proto-Romance" and the related article on Vulgar Latin, although I think a lot of these sound changes would take place a little after "proto-Romance". Erinius (talk) 21:30, 11 April 2023 (UTC)
 * I just did a preliminary rewrite of the paragraph in question - I'd appreciate further changes to phrasing, since I really wasn't sure how to explain things best. Also, IIRC, the Encyclopedia Britannica article, and the "Classifications" chapter of The Oxford Guide to the Romance Languages, explain why counting the Romance languages is impossible, a bit about how the whole language vs dialect thing is based on social factors as well as purely linguistic ones, and how a lot of somewhat medium-level classifications (ie Ibero-Romance) are largely geographical. Erinius (talk) 23:58, 25 April 2023 (UTC)
 * Erinius, pressed for time just now, but I've made one little tweak, from various languages to numerous languages, to open the way for what's to come. I think you've done an excellent job of editing. Barefoot through the chollas (talk) 01:01, 26 April 2023 (UTC)
 * "One, two, three, many". might be interested in this. 04:28, 12 April 2023 (UTC)

Sound changes
Looking at the article flow a big chunk in the middle is taken by the "Sound changes" part. The topic is too complex to be compressed in a satisfactory manner. I suggest moving most of its content to "Phonological changes from Classical Latin to Proto-Romance". It would add to the main article's depth, and, at the same time, improve readability of the Romance languages article. Aristeus01 (talk) 13:07, 19 July 2023 (UTC)


 * Agree, and I'm not sure anyone would disagree. Most of that content should be in a separate article. A word of caution though, not all the changes in the "sound changes" part affected all of Romance, so they wouldn't strictly be sound changes to Proto-Romance. I'm not totally sure what should be done about those. Palatalization which affected all of Romance to some degree, but some varieties more extensively, is already briefly mentioned in Phonological changes... so I think transferring most of this article's Palatalization section there would be fine. I guess we should also see what some of the main contributors to the Proto-Romance articles (namely ) have to say and I'm also curious if other language families have their own sound changes articles. Erinius (talk) 16:15, 19 July 2023 (UTC)
 * You could certainly add a section labelled 'subsequent regional sound changes', or similar, to the article about Classical Latin > Proto-Romance changes. Nicodene (talk) 16:59, 19 July 2023 (UTC)
 * I ask only that whatever content transferred there be cited to the same standard as the rest of the existing article. Nicodene (talk) 17:00, 19 July 2023 (UTC)
 * A fair request. We should start by adding refences to our current text in question and when we are happy with the result we can look again at all the suggestions and decide on the best way to move forward. Aristeus01 (talk) 18:27, 19 July 2023 (UTC)

Agreed that the historical phonology in the Romance languages article is in some ways too detailed -- though in an article of its own, one could argue not nearly detailed enough, starting with too many languages not even mentioned. Before venturing into that quagmire, though, Phonological changes from Classical Latin to Proto-Romance as basis for a revised more complete version would have to be re-named ("moved"), but the present title (reduced to Latin to Romance) and content could provide an excellent first portion if the new article were to be articulated in two phases, either Latin to Romance, then local developments as two large sections, or each feature or set of features in two phases. The present text would need some massaging (e.g. glosses supplied, sources updated, statements such as this cleaned up: In cases where a long vowel precedes a geminate consonant, one of the elements often shortens unpredictably). The other task, deciding on how much historical phonology remains in Romance languages, is a bit thorny, but it can be done. Barefoot through the chollas (talk) 18:21, 19 July 2023 (UTC)