Language desk
< June 8	<< May \| June \| Jul >>	June 10 >

Welcome to the Wikipedia Language Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

June 9

French vowels

Where did French get /y/, /ø/, /œ/, and nasal vowels, since nasal vowels are only in Portuguese and the other three vowels aren't in any other Romance languages at all? --108.227.31.161 (talk) 03:13, 9 June 2012 (UTC)[reply]

Why did they have to "get" them from anywhere? If all linguistic change occured because a language "got" something from somewhere else, we'd have never had more than one language. Some languages evolve or change arbitrarily: that is, without outside influence, and in random ways. Indeed, the fact that French phonology is so unlike any of its neighbors or antecedants, it is quite likely something that "just happened". --Jayron32 03:30, 9 June 2012 (UTC)[reply]

The rounded vowels are likely a Frankish language superstratum influence, since they are found in northern dialects and not southern. See Austrasia. -- 03:41, 9 June 2012‎ User:Medeis

Maybe -- I don't think it's very clear that there was phonemic umlaut in the relevant Germanic languages during the Merovingian period (as opposed to low-level conditioned allophones), and some scholars of past generations were too quick to ascribe anything the slightest bit out of the ordinary to hypothetical substrata or superstrata. I find it interesting that the French shift is parallel to Hellenistic Greek (in broad terms, u → ü, then o → u to fill the resulting gap)... AnonMoos (talk) 06:21, 9 June 2012 (UTC)[reply]

The point is that it is not clear when exactly the shift from /u/ to /y/ happened in French; conceivably it was not in the Merovingian period, but shortly before the shift from from /o/ to /u/, in which case it could conceivably have happened under the influence of neighbouring Germanic dialects. I am not sure however where exactly bilingual areas were found in the High Medieval period (when /y/ was certainly already a phoneme in German). That said, long /yː/ (from earlier /iu/ and its umlauted allophonic variant [iy]) seems to have been present as a distinct phoneme in late Old High German, judging by the spelling, and according to Merovingian dynasty#Language Old West Franconian (which may have included dialects resembling Old High German prior to its extinction) did not go extinct prior to the 10th century. Still, there is admittedly a remaining major problem: If Old West Franconian was (partly) like OHG, it would have had both /uː/ and /yː/, and the motivation for the shift from /u/ to /y/ in Old French remains obscure even if we grant that a significant (either in number or in status) portion of speakers of OF were bilingual in OWF. If OWF influenced OF phonologically, this would mean that bilingual speakers at the time used OWF more intensely than OF, otherwise one would expect the reverse direction of influence. I don't think loans from OWF into OF, even if they contained /yː/, could have caused the shift; if anything, they could have introduced the phoneme, but why should it be generalised even to native words? The only plausible explanation, to my mind, is a fad by which the speakers of OF shifted their /u/ phonemes to [y] as a sort of overgeneralisation, to sound more "Germanic", after getting used to the sound as a distinctive sound typical of Germanic languages. Considering the many other ways in which OF did adapt to Germanic not only lexically but also structurally, I find that believable. One might compare the (not only) phonological Turkification of Cappadocian Greek (which even affected the native lexicon) or perhaps also Armenian dialects which have rounded front vowels, or the way rounded front vowels have appeared in Slavic and specific Slavic dialects at various points, apparently under the influence of Turkic, Finnic, Hungarian and German. --Florian Blaschke (talk) 10:46, 9 June 2012 (UTC)[reply]

I think can't be a coincidence that among Romance languages front vowels like /y/ and /ø/ occur only in French and North Italian dialects − both of which are adjacent to Germanic-speaking areas. Whether it is a result of Germanic substrate in French or not, cerainly it is a shared areal feature of the languages of west-central and northern Europe (French, Northern Italian, German, Dutch, Scandinavian, Finnic). --BishkekRocks (talk) 11:23, 9 June 2012 (UTC)[reply]

It's no wonder that when in the 14-15th centuries diphthongs /eu/ and /ue/ had monophthongized they became just rounded /e~ɛ/ that is /ø~œ/. This is simply one of phonetic universals. The same is for nasals from combinations "vowel + n/m", they can't become anything but nasal vowels. As for /y/ it exists also in Occitan quite well, so German superstratum is very unlikely. I think the reason for the changes is a tendency of French to open syllables and simple (non-composite) phonemes.--Luboslov Yezykin (talk) 11:46, 9 June 2012 (UTC)[reply]

Occitan is not a compelling counter-argument as the fronting of /u/ is not necessarily very old there: it was not necessarily already present in Old Occitan (especially the oldest attested stage – we cannot date the change precisely just like in French), and the fronting can thus be ascribed to French influence. Also, I harbour little trust in universals as universals all have exceptions in my experience, except very trivial or definitory universals and perhaps some implicative universals. Same for tendencies: tendencies hold no explanatory value because they are merely a description of observed developments, and essentially you are saying that a language exhibits a tendency because it exhibits a tendency which is patently tautological (and sometimes even blatantly unscientific, like when certain tendencies are imputed to languages that happen to surface only centuries later). Insofar as tendencies are real phenomena, they are more plausibly explained as due to contact/areal influences. --Florian Blaschke (talk) 20:21, 9 June 2012 (UTC)[reply]

Labov, in Principles of Linguistic Change vol I, has an interesting and quite non-tautological discussion about why there may be a universal, internally-caused tendency for vowel changes (at least in long/tense vowels) to go in a fronting/raising direction. This would cover [u]>[y] changes, as in French, Dutch, Greek and others. Fut.Perf. ☼ 19:35, 11 June 2012 (UTC)[reply]

"an ethnic group which ..."?

While editing Utsul I was struck by the phrasing "The Utsuls [...] are a tiny ethnic group which ~~live~~ lives on the island of Hainan", as I would have expected who, since an ethnic group is a group of people and not an object. However, I can see why even native speakers might be confused in cases such as this one (I am not entirely sure in this case myself and did not realise the problem immediately, either; but then, I am not a native speaker), and in fact, a Google search for "ethnic group which" yields more results than "ethnic group who". However, from a prescriptive point of view based on logical reasoning, shouldn't who be the only correct option? Unfortunately, our article English relative clauses is not clear on this point. --Florian Blaschke (talk) 09:20, 9 June 2012 (UTC)[reply]

I just realised I made an error that seems to point to the heart of the matter here: a group who live sounds correct, but the other option is a group which lives, so the group conceived as many individual people is person-like while the group conceived as a collective is (or at least can be) considered an object, intuitively. Fascinating. --Florian Blaschke (talk) 09:26, 9 June 2012 (UTC)[reply]

Another possibility is "... a tiny ethnic group that lives on ...". In fact, I'd prefer "that" over "which" in such a construction. I never use "that" for humans, preferring "who", but "group" is not grammatically human so "that" is OK. -- ♬ Jack of Oz ♬ ^{[your turn]} 11:35, 9 June 2012 (UTC)[reply]

I am pretty sure you would use "that" over "which" in this case (as the original sentence is constructed). Although it is not absolute, and it is a point of contention. See English relative clauses#That or which. Thanks. Joseph A. Spadaro (talk) 16:58, 9 June 2012 (UTC)[reply]

In prescriptivist usage of English, "that" is the only correct form here, as a distinction is "supposed" to be made between:

Utsul, which is a Hainanese language, is very interesting."

and

Utsul is a Hainanese language that is very interesting.

Although in practice "which" is used in the second sentence as well. saɪm duʃan Talk|Contribs 17:20, 9 June 2012 (UTC)[reply]

I don't write American English so I don't observe this rule in any case, but thanks for reminding me of it. While that is certainly an option, there's another one I was contemplating: using a participle to dodge the issue. The Utsuls are a tiny ethnic group living on [...] – that would work, too, wouldn't it? (By the way, Utsul is only the name of the ethnic group, their language is Tsat.) --Florian Blaschke (talk) 20:18, 9 June 2012 (UTC)[reply]

Yes, that would work, but lucky for you they're still around. There's no past tense counterpart of the participle, so if the Utsuls were an an extinct people, you have to revert to "that lived" or, if you prefer, "which lived". You could also write "The Utsuls were an ethnic group living on ...", but that would probably be used for a group that had relatively recently ceased to live there, not for a group that had died out millennia ago. -- ♬ Jack of Oz ♬ ^{[your turn]} 00:34, 10 June 2012 (UTC)[reply]

Counterexample: "once living" or "formely living" or even "once found living". μηδείς (talk) 02:31, 10 June 2012 (UTC)[reply]

Indeed. -- ♬ Jack of Oz ♬ ^{[your turn]} 04:00, 10 June 2012 (UTC)[reply]

They wouldn't, however, necessarily be the best word choice. Russian has past active participles, although not in the article on Russian grammar, but here: Participle#Russian. μηδείς (talk) 04:16, 10 June 2012 (UTC)[reply]

This is also not, as Saimdan says, a matter of restrictive versus non-restrictive clauses. In his example there would have to be two Utsul languages for that: "The Utsul that is a Hainan language is very interesting" would be a restrictive clause--but it makes no sense if there is no non-Hainan Utsul. Unless we have already heard that all the other Hainan languages are uninteresting, what would be meant here is the equivalent of "Utsul, which is a Hainan language, is very interesting." That's non-restrictive. But the non-restrictive which is certainly not mandated here. The real issue is animacy, not restriction. Which sounds funny because it contrasts with who and implies inanimacy. In any case, I prefer The Utsuls are a tiny ethnic group that lives on Hainan or The Utsuls are a tiny ethnic group who live on Hainan.μηδείς (talk) 02:52, 10 June 2012 (UTC)[reply]

The creator of the D'ni language

Apparently prematurely, I assumed that the philosopher and speleologist Richard A. Watson and the creator of the D'ni language in the video game series Myst, Dr. Richard A. Watson, also known as RAWA, were the same person, and redirected Richard A. Watson to the philosopher. See Richard A. Watson the philosopher's website and Richard A. Watson the language creator's weblog. Now there are several links pointing from D'ni-related material to the philosopher, which would seem to be misleading. However, I'm still confused about the identity or non-identity of the two people. Web searches do not give any clear statements, at best indications. What do I do now? --Florian Blaschke (talk) 20:48, 9 June 2012 (UTC)[reply]

You can always email one or both of them and ask. But given the pictures at [1] and [2] I'm pretty sure they aren't the same person. -- BenRG (talk) 07:04, 10 June 2012 (UTC)[reply]