(Conlangs) Q&A Thread - Quick questions go here

A forum for all topics related to constructed languages
Ælfwine
roman
roman
Posts: 940
Joined: 21 Sep 2015 01:28
Location: New Jersey

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ælfwine »

What might be a clean way to develop a form of pitch accent in a Romance language?

I was thinking at first to derive a pitch accent from the loss of Latin final consonants, such as -s and -m, leading to a simple tonal system perhaps marked on the stressed syllable.

example:
trēs > trì (falling accent)
hic > í (rising accent)

I am wondering if this is a realistic way to go about it, or if this would be too "out there." A less conservative option would be instead to derive tone from a later loss of final vowels, like -u

Another side effect is that case would be preserved:

ex:
MULTUS > mùot
MULTUM > múot

The background of this language is it is a romance conlang spoken in Croatia, so the development of pitch accent and the preservation of cases at least are areal features. Though I am aware of the overdoneness of the whole case preserving romlang (and indeed it wasnt even my intention at first, just a side effect of my desire to develop a pitch accent!)
My Blog

A-posteriori, alternative history nerd
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

I generally have four go-to paths if I'm thinking about tone:

a) from the loss of codas
b) from neutralisation of onsets
c) from vowel length
d) from changes in word shape

I don't think neutralisation is a great answer for a romlang spoken in Europe, because voiced/voiceless contrasts (or equivalent) are virtually ubiquitous in the SAE area.

The other three options all seem plausible, however.

The most obvious route is from the loss of -m. This is 'obvious' because we know that -m was weird. It probably wasn't pronounced as a consonant even in the late Republic, and certainly wasn't in vulgar latin a century later, but we also know it was doing SOMETHING. Both because Latin speakers said it did, and because -um and -us have different outcomes in many daughter languages. [mostly, -um acts like it had a long vowel, or a half-long vowel, or something, so that at least temporarily it was't lowered, so it triggered metaphony]. It's likely that the consonant dropped with compensatory nasalisation; this might also explain the loss of nasals before fricatives in daughter languages (if we assume that nasalisation itself left no trace in non-final position). But nasalisation could very easily transform into a tone contrast.

That by itself wouldn't give a lot of tone in the daughter language, unfortunately. But it's a start, and once there's some tone other tone can join it. You're right that other coda consonants could - less probably but still not implausibly - yield tone. Given the widespread loss of final consonants, a change of, for instance, final stops > glottal stop > tone seems quite plausible to me.

Word shape (i.e. maintaining tonal contour in the face of syllabic loss) is always a good option, even in languages with no other tonal influence (eg Scandinavian, Limburgish). In a language that already has tone and that's surrounded by other languages with tone, it seems very plausible indeed.

Long vowels can easily turn into tone - either by loss of length, or by the creation of new long vowels distinguished from the old by tone (eg original a: vs ada > a:). This is less likely in Latin because of the universal loss of length without trace in daughter languages, but it's it's conceivable to imagine a distant elative that keeps length longer. Or, indeed, to say unfalsifiably "the loss of length in Southern Romance was actually just the replacement of length with tone, which was later neutralised in Sardinian").
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

I was thinking of the word shape route. Romance languages altered word shapes in various ways (prothesis, syncope, etc). If the pitch contour remained constant, this could yield pitch accent contrasts between derived and inherited word shapes. Of course this could be combined with any of the other options Sal mentioned.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
VaptuantaDoi
roman
roman
Posts: 1066
Joined: 18 Nov 2019 07:35

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by VaptuantaDoi »

I think loss of coda s is the most areally believable one. I could even see coda /s/ triggering falling pitch accent from a distance, e.g.

LŪNAM → /lúna/
LŪNĀS → /lûna/

Like how Andalusian Spanish spreads [–ATR] (or something accoustically similar) from final syllables throughout the word e.g. come /kome/ vs. comes /kɔmɛ/.

Alternatively, you could have metaphony (PORCUM PORCĪ > /pɔrk pwɔrk/) then allophonic prenuclear accent > rising pitch (as in Spanish again) followed by remonophthongisation (/pɔrk pwɔrk/ > [pɔrk˥ pwɔrk˩˥] > /pórk pǒrk/). You could also keep case that way with a bit of fiddling, although it might not be ideal as it only really gives a phonemic contrast on mid vowels.
User avatar
WeepingElf
greek
greek
Posts: 531
Joined: 23 Feb 2016 18:42
Location: Braunschweig, Germany
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by WeepingElf »

Now this is an interesting variation of the otherwise overdone Romance conlang theme, I must say.
... brought to you by the Weeping Elf
My conlang pages
User avatar
GoshDiggityDangit
greek
greek
Posts: 549
Joined: 18 Dec 2018 21:27
Location: Oakwood OH, USA
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by GoshDiggityDangit »

How does topic-prominence work? Does the topic marker work like a pronoun or a case?
“Like billowing clouds, Like the incessant gurgle of the brook,
The longing of the spirit can never be stilled.” ― St. Hildegard von Bingen
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Omzinesý »

Salmoneus wrote: 21 May 2023 17:05and because -um and -us have different outcomes in many daughter languages. [mostly, -um acts like it had a long vowel, or a half-long vowel, or something, so that at least temporarily it was't lowered, so it triggered metaphony].
Just interested, in which modern language the difference is visible and how?
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

GoshDiggityDangit wrote: 22 May 2023 19:34 How does topic-prominence work? Does the topic marker work like a pronoun or a case?
Topic-prominent languages do not need a topic marker. Being the topic is more like a syntactic position or role. In general, a topic marker is neither a case nor a pronoun, but if it helps you to picture it -- it is a bit like a case marker in the same way the notion of subject is case-like. Diachronically, it can stem from a pronoun in the same way that definite articles can derive from pronouns. Topics, definite nouns, and pronouns all have in common that they usually mark 'given' or 'old' information.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Omzinesý wrote: 22 May 2023 20:37
Salmoneus wrote: 21 May 2023 17:05and because -um and -us have different outcomes in many daughter languages. [mostly, -um acts like it had a long vowel, or a half-long vowel, or something, so that at least temporarily it was't lowered, so it triggered metaphony].
Just interested, in which modern language the difference is visible and how?
It's visible in central and southern Italian languages, Sardinian, Asturian, and Portuguese, although it's only sporadic in Portuguese.

The difference shows up either in raising or in diphthongisation of the preceding vowel.

Eg in Asturian pilu, 'hair', pelos, 'hairs' (and I gather from wikipedia also pelo, 'hair' (mass noun). In Western Asturian final -u and -o merge, so that would be pilu (count) vs pelu (mass).


Note that I'm not aware of this ever leading to case retention, and in some dialects it's also neutralised by analogy in plural forms, so there often aren't any direct -u/-o contrasts left. However, the effect of -u is still widely seen in, for instance, adjective gender contrasts (vs -a). The same processes also occur with final -i: (eg Portuguese fiz, I did, vs fez, he did, < feci: vs fecit).
Ælfwine
roman
roman
Posts: 940
Joined: 21 Sep 2015 01:28
Location: New Jersey

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ælfwine »

Thank you for the detailed responses, everyone!

My thinking now is that I develop tone as a byproduct of language contact with Serbo-Croatian, instead of a natural development from VL. And there seems to be precedent, Wikipedia states this under "Origin and Development" of its article about linguistic Tone: Tone is sometimes an areal rather than a phylogenetic feature. That is to say, a language may acquire tones through bilingualism if influential neighbouring languages are tonal or if speakers of a tonal language shift to the language in question and bring their tones with them. The process is referred to as contact-induced tonogenesis by linguists.[44] In other cases, tone may arise spontaneously and surprisingly fast: the dialect of Cherokee in Oklahoma has tone, but the dialect in North Carolina does not although they were separated only in 1838.

Perhaps tones could arise during the redevelopment of Early Romance vowel length. From Wikipedia's article on Romance Language, under "Loss of vowel length, reorientation:" The Proto-Romance allophonic vowel-length system was rephonemicized in the Gallo-Romance languages as a result of the loss of many final vowels. Some northern Italian languages (e.g. Friulian) still maintain this secondary phonemic length, but most languages dropped it by either diphthongizing or shortening the new long vowels.
My Blog

A-posteriori, alternative history nerd
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

One way to go for contact-induced tonogenesis is to look for other phonetic cues for pitch accent. IIRC, in many varieties of BCMS, pitch accent interacts in complex ways with both phonological length and phonetic duration. (I think there is work by Draga Zec on this.) So, this could be an argument for this route. Bilingual speakers of your conlang could transfer the phonetic pitch changes onto their own length distinction, at first merely as additional phonetic cues. Lateron, the pitch changes could become the main phonetic cue and be reinterpreted as part of the underlying phonological form.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Ælfwine
roman
roman
Posts: 940
Joined: 21 Sep 2015 01:28
Location: New Jersey

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ælfwine »

Yes this was my thinking. I was think that vowel length would be lost (the second time) like many romance languages but the pitch accent would remain as a phonetic quality. I have to refigure out what Serbocroatian does. I remember vaguely there was some relation in many dialects between tone and vowel length, where in dialects that lost the former still kept a distinction in the later.
My Blog

A-posteriori, alternative history nerd
User avatar
Omzinesý
mongolian
mongolian
Posts: 4079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Omzinesý »

Salmoneus wrote: 23 May 2023 01:19
Omzinesý wrote: 22 May 2023 20:37
Salmoneus wrote: 21 May 2023 17:05and because -um and -us have different outcomes in many daughter languages. [mostly, -um acts like it had a long vowel, or a half-long vowel, or something, so that at least temporarily it was't lowered, so it triggered metaphony].
Just interested, in which modern language the difference is visible and how?
It's visible in central and southern Italian languages, Sardinian, Asturian, and Portuguese, although it's only sporadic in Portuguese.

The difference shows up either in raising or in diphthongisation of the preceding vowel.

Eg in Asturian pilu, 'hair', pelos, 'hairs' (and I gather from wikipedia also pelo, 'hair' (mass noun). In Western Asturian final -u and -o merge, so that would be pilu (count) vs pelu (mass).


Note that I'm not aware of this ever leading to case retention, and in some dialects it's also neutralised by analogy in plural forms, so there often aren't any direct -u/-o contrasts left. However, the effect of -u is still widely seen in, for instance, adjective gender contrasts (vs -a). The same processes also occur with final -i: (eg Portuguese fiz, I did, vs fez, he did, < feci: vs fecit).
Interesting
I learned a new fact
Thanks
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
Arayaz
roman
roman
Posts: 1218
Joined: 07 Sep 2022 00:24
Location: Just south of the pin-pen merger
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Arayaz »

I'm working on fully fleshing out the second of my three Isanyá languages: Anopí, and the romanization is being messed up by the fact that the two registers (High and Low) each have their own allophonic quirks and, unfortunately, mergers.

Low Anopí, the one spoken in colloquial contexts (and thus by the vast majority of people), has reduced /k/ to /h/. In addition, it merged Early Modern Anopí's palatal series with the velars, with the exception of ʃ (which became ç). I've romanized it as though it were Low Anopí always, with two exceptions: 1) LA /ŋ/ can be spelled ng or , depending on whether its HA equivalent is /ŋ/ or /ɲ/, and 2) LA /j/ is y or depending on whether it is /j/ or /ʎ/ in HA.

However, the overdots are annoying, since Anopí is already peppered with acute and grave accents due to tone. As a (nonsense) example:
Ása hahya hyuṅá iríṅì nkihù nsaí nsàmya opí yau ẏò

So what should I do?
Proud member of the myopic-trans-southerner-Viossa-girl-with-two-cats-who-joined-on-September-6th-2022 gang

:con: Ruykkarraber languages, Izre, Ngama, Areyaxi languages, ???, 2c2ef0
my garbage

she/her
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by DesEsseintes »

Üdj wrote: 03 Jun 2023 03:59 Ása hahya hyuṅá iríṅì nkihù nsaí nsàmya opí yau ẏò

So what should I do?
I personally like the dots, though that’s probably not the answer you were looking for.
shimobaatar
korean
korean
Posts: 10371
Joined: 12 Jul 2013 23:09
Location: UTC-04:00

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by shimobaatar »

Üdj wrote: 03 Jun 2023 03:59 I'm working on fully fleshing out the second of my three Isanyá languages: Anopí, and the romanization is being messed up by the fact that the two registers (High and Low) each have their own allophonic quirks and, unfortunately, mergers.

Low Anopí, the one spoken in colloquial contexts (and thus by the vast majority of people), has reduced /k/ to /h/. In addition, it merged Early Modern Anopí's palatal series with the velars, with the exception of ʃ (which became ç). I've romanized it as though it were Low Anopí always, with two exceptions: 1) LA /ŋ/ can be spelled ng or , depending on whether its HA equivalent is /ŋ/ or /ɲ/, and 2) LA /j/ is y or depending on whether it is /j/ or /ʎ/ in HA.

However, the overdots are annoying, since Anopí is already peppered with acute and grave accents due to tone. As a (nonsense) example:
Ása hahya hyuṅá iríṅì nkihù nsaí nsàmya opí yau ẏò

So what should I do?
Perhaps <ng> LA & HA /ŋ/ vs. <gn> LA /ŋ/ but HA /ɲ/, and <y> LA & HA /j/ vs. <ly> LA /j/ but HA /ʎ/?

Ása hahya hyugná irígnì nkihù nsaí nsàmya opí yau lyò
User avatar
Arayaz
roman
roman
Posts: 1218
Joined: 07 Sep 2022 00:24
Location: Just south of the pin-pen merger
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Arayaz »

DesEsseintes wrote: 03 Jun 2023 09:46
Üdj wrote: 03 Jun 2023 03:59 Ása hahya hyuṅá iríṅì nkihù nsaí nsàmya opí yau ẏò

So what should I do?
I personally like the dots, though that’s probably not the answer you were looking for.
Well, there's no accounting for taste, as they say.
Proud member of the myopic-trans-southerner-Viossa-girl-with-two-cats-who-joined-on-September-6th-2022 gang

:con: Ruykkarraber languages, Izre, Ngama, Areyaxi languages, ???, 2c2ef0
my garbage

she/her
User avatar
Arayaz
roman
roman
Posts: 1218
Joined: 07 Sep 2022 00:24
Location: Just south of the pin-pen merger
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Arayaz »

shimobaatar wrote: 03 Jun 2023 12:54
Üdj wrote: 03 Jun 2023 03:59 I'm working on fully fleshing out the second of my three Isanyá languages: Anopí, and the romanization is being messed up by the fact that the two registers (High and Low) each have their own allophonic quirks and, unfortunately, mergers.

Low Anopí, the one spoken in colloquial contexts (and thus by the vast majority of people), has reduced /k/ to /h/. In addition, it merged Early Modern Anopí's palatal series with the velars, with the exception of ʃ (which became ç). I've romanized it as though it were Low Anopí always, with two exceptions: 1) LA /ŋ/ can be spelled ng or , depending on whether its HA equivalent is /ŋ/ or /ɲ/, and 2) LA /j/ is y or depending on whether it is /j/ or /ʎ/ in HA.

However, the overdots are annoying, since Anopí is already peppered with acute and grave accents due to tone. As a (nonsense) example:
Ása hahya hyuṅá iríṅì nkihù nsaí nsàmya opí yau ẏò

So what should I do?
Perhaps <ng> LA & HA /ŋ/ vs. <gn> LA /ŋ/ but HA /ɲ/, and <y> LA & HA /j/ vs. <ly> LA /j/ but HA /ʎ/?

Ása hahya hyugná irígnì nkihù nsaí nsàmya opí yau lyò
I do like this, though I've never understood <gn> for /ɲ/. It's from Italian, right?
Proud member of the myopic-trans-southerner-Viossa-girl-with-two-cats-who-joined-on-September-6th-2022 gang

:con: Ruykkarraber languages, Izre, Ngama, Areyaxi languages, ???, 2c2ef0
my garbage

she/her
shimobaatar
korean
korean
Posts: 10371
Joined: 12 Jul 2013 23:09
Location: UTC-04:00

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by shimobaatar »

Üdj wrote: 03 Jun 2023 15:02
shimobaatar wrote: 03 Jun 2023 12:54
Üdj wrote: 03 Jun 2023 03:59 I'm working on fully fleshing out the second of my three Isanyá languages: Anopí, and the romanization is being messed up by the fact that the two registers (High and Low) each have their own allophonic quirks and, unfortunately, mergers.

Low Anopí, the one spoken in colloquial contexts (and thus by the vast majority of people), has reduced /k/ to /h/. In addition, it merged Early Modern Anopí's palatal series with the velars, with the exception of ʃ (which became ç). I've romanized it as though it were Low Anopí always, with two exceptions: 1) LA /ŋ/ can be spelled ng or , depending on whether its HA equivalent is /ŋ/ or /ɲ/, and 2) LA /j/ is y or depending on whether it is /j/ or /ʎ/ in HA.

However, the overdots are annoying, since Anopí is already peppered with acute and grave accents due to tone. As a (nonsense) example:
Ása hahya hyuṅá iríṅì nkihù nsaí nsàmya opí yau ẏò

So what should I do?
Perhaps <ng> LA & HA /ŋ/ vs. <gn> LA /ŋ/ but HA /ɲ/, and <y> LA & HA /j/ vs. <ly> LA /j/ but HA /ʎ/?

Ása hahya hyugná irígnì nkihù nsaí nsàmya opí yau lyò
I do like this, though I've never understood <gn> for /ɲ/. It's from Italian, right?
Indeed, some Romance languages, including Italian and French, use <gn> for /ɲ/ because one possible source of /ɲ/ in these languages is Latin <gn> /gn/ [ŋn]. For instance, Latin agnellus > Italian agnello & French agneau.
User avatar
LinguoFranco
greek
greek
Posts: 613
Joined: 20 Jul 2016 17:49
Location: U.S.

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by LinguoFranco »

One of my conlangs has a phonemic distinction between short and long vowels. The thing that is unusual about it is that the language doesn't have any true long vowels at all. Rather, the "long vowels" are realized closer to half-long.

Thus, while native speakers can tell short and long vowels apart easily, foreigners have a much harder time.

What are your thoughts on this? Would this be something that could be phonemic, or would a short vs half long be more allophonic?
Post Reply