Where did Western Romance <ch> come from?

[Christian Weisgerber]

How did the spelling <ch> for /tʃ/ arise in the Western Romance
languages?

The Romans used <ch> to transcribe the aspirated stop of Greek χ /kʰ/.
Eventually the Greek pronunciation shifted to fricative /x/, which
presumably underlies the use of <ch> for /x/ in German and other
languages.

But where is <ch> for /tʃ/ from? Did this spread from Old French?

Phonologically, Western Romance /tʃ/ has various origins. In Old
French it was the result of a highly distinctive sound shift
/ka/ > /tʃa/. In Spanish it is the reflex of Latin -ct-. In
Old Portuguese it is the reflex of Latin cl-. (There may be
additional sources in Spanish and Portuguese, I don't know.)
French and Portuguese have since deaffricated /tʃ/ to /ʃ/.

None of the medieval languages had /x/, so <ch> could be repurposed,
but is it possible to pinpoint an origin?

--
Christian "naddy" Weisgerber na...@mips.inka.de

[Peter T. Daniels]

Or, why does <h> happen to be a convenient diacritic in English
for "sounds that sre similar," giving us ch gh ph sh th wh (kh is later,
rh simply transliterates the original Greek)?

Because the model existed in Latin, as you noted?

[Daud Deden]

On Saturday, April 29, 2023 at 9:30:08 AM UTC-4, Christian Weisgerber wrote:

I don't know.
My experience of finding out that Dutch uses tj instead of ch came through reading Malay Indonesian, where benci (angry) was earlier written bentji due to Dutch influence. At first it made no sense, as I expected the sounds tj and ch to be very different, but then it dawned on me that they are about the same sound. I had been saying "bent - ji" not "benchi" (which is close to "Benji").

[Ruud Harmsen]

Sat, 29 Apr 2023 13:07:58 -0000 (UTC): Christian Weisgerber
<na...@mips.inka.de> scribeva:

>How did the spelling <ch> for /t?/ arise in the Western Romance
>languages?
>
>The Romans used <ch> to transcribe the aspirated stop of Greek ? /k?/.

>Eventually the Greek pronunciation shifted to fricative /x/, which
>presumably underlies the use of <ch> for /x/ in German and other
>languages.
>

>But where is <ch> for /t?/ from? Did this spread from Old French?
>
>Phonologically, Western Romance /t?/ has various origins. In Old

>French it was the result of a highly distinctive sound shift

>/ka/ > /t?a/. In Spanish it is the reflex of Latin -ct-. In

>Old Portuguese it is the reflex of Latin cl-. (There may be
>additional sources in Spanish and Portuguese, I don't know.)

Also pl. https://rudhar.com/etymolog/prantoen.htm

>French and Portuguese have since deaffricated /t?/ to /?/.

Portuguese did, yes. Modern Galician still has the original affricate.

>None of the medieval languages had /x/, so <ch> could be repurposed,
>but is it possible to pinpoint an origin?

--

Ruud Harmsen, http://rudhar.com

[Ruud Harmsen]

Sun, 30 Apr 2023 04:03:58 -0700 (PDT): Daud Deden
<daud....@gmail.com> scribeva:

>On Saturday, April 29, 2023 at 9:30:08?AM UTC-4, Christian Weisgerber wrote:
>> How did the spelling <ch> for /t?/ arise in the Western Romance
>> languages?
>>
>> The Romans used <ch> to transcribe the aspirated stop of Greek ? /k?/.

>> Eventually the Greek pronunciation shifted to fricative /x/, which
>> presumably underlies the use of <ch> for /x/ in German and other
>> languages.
>>

>> But where is <ch> for /t?/ from? Did this spread from Old French?
>>
>> Phonologically, Western Romance /t?/ has various origins. In Old

>> French it was the result of a highly distinctive sound shift

>> /ka/ > /t?a/. In Spanish it is the reflex of Latin -ct-. In

>> Old Portuguese it is the reflex of Latin cl-. (There may be
>> additional sources in Spanish and Portuguese, I don't know.)

>> French and Portuguese have since deaffricated /t?/ to /?/.

>>
>> None of the medieval languages had /x/, so <ch> could be repurposed,
>> but is it possible to pinpoint an origin?
>>
>> --
>> Christian "naddy" Weisgerber na...@mips.inka.de
>
>I don't know.
>My experience of finding out that Dutch uses tj instead of ch came through reading Malay Indonesian, where benci (angry) was earlier written bentji due to Dutch influence. At first it made no sense, as I expected the sounds tj and ch to be very different, but then it dawned on me that they are about the same sound. I had been saying "bent - ji" not "benchi" (which is close to "Benji").

Or tsj. The Malay sounds are palatal, and not affricates, I think.
Java is Djava in modern Indonesian?

[Christian Weisgerber]

On 2023-04-29, Peter T. Daniels <gram...@verizon.net> wrote:

> Or, why does <h> happen to be a convenient diacritic in English
> for "sounds that sre similar," giving us ch gh ph sh th wh (kh is later,
> rh simply transliterates the original Greek)?
>
> Because the model existed in Latin, as you noted?

Yes. The Romans originally used ph, th, ch to transcribe the Ancient
Greek aspirated stops φ [pʰ], θ [tʰ], χ [kʰ], which makes a certain
amount of sense phonetically, as well as rh for ῥ, apparently
devoiced [r̥]. Then, sometime around the start of the Common Era,
the Greek stops shifted to fricatives [f], [θ], [x], but the
transcriptions stayed the same.

Once the pattern had been established, it could spread.

* The Strasbourg Oaths of 842 use dh and th for [ð] and [θ],
respectively.
* Early Old High German used th for [θ~ð], Tatian, c. 830.
* The digraph ch was reused for [tʃ] in Western Romance, see my
original question. The second-oldest document in Old French, the
Eulalia sequence, c. 880, uses ch inconsistently for [k] and [tʃ].
* Galician-Portuguese picked up lh for [ʎ] and nh for [ɲ] from Old
Occitan in the 11th century.

In conclusion, the practice was well established by the time it
reached Middle English.

[Dingbat]

On Sunday, April 30, 2023 at 8:54:20 PM UTC+5:30, Ruud Harmsen wrote:
> Sat, 29 Apr 2023 13:07:58 -0000 (UTC): Christian Weisgerber
> <na...@mips.inka.de> scribeva:
>
> >How did the spelling <ch> for /t?/ arise in the Western Romance
> >languages?
> >
> >The Romans used <ch> to transcribe the aspirated stop of Greek ? /k?/.
> >Eventually the Greek pronunciation shifted to fricative /x/, which
> >presumably underlies the use of <ch> for /x/ in German and other
> >languages.
> >
> >But where is <ch> for /t?/ from? Did this spread from Old French?
> >
> >Phonologically, Western Romance /t?/ has various origins. In Old
> >French it was the result of a highly distinctive sound shift
> >/ka/ > /t?a/. In Spanish it is the reflex of Latin -ct-. In
> >Old Portuguese it is the reflex of Latin cl-. (There may be
> >additional sources in Spanish and Portuguese, I don't know.)
> Also pl. https://rudhar.com/etymolog/prantoen.htm
>

It's a pity that Gallego/ Galician is not included, since I've seen
it described as retaining its original affricates. I'm not sure what
'thst description means though. If other Romance languages
didn't retain their original affricates, what did they do with them?
You too say 'original affricate'. Please explain what that means.

[Ruud Harmsen]

Tue, 2 May 2023 19:37:37 -0700 (PDT): Dingbat
<ranjit_...@yahoo.com> scribeva:

>On Sunday, April 30, 2023 at 8:54:20?PM UTC+5:30, Ruud Harmsen wrote:
>> Sat, 29 Apr 2023 13:07:58 -0000 (UTC): Christian Weisgerber
>> <na...@mips.inka.de> scribeva:
>>
>> >How did the spelling <ch> for /t?/ arise in the Western Romance
>> >languages?
>> >
>> >The Romans used <ch> to transcribe the aspirated stop of Greek ? /k?/.
>> >Eventually the Greek pronunciation shifted to fricative /x/, which
>> >presumably underlies the use of <ch> for /x/ in German and other
>> >languages.
>> >
>> >But where is <ch> for /t?/ from? Did this spread from Old French?
>> >
>> >Phonologically, Western Romance /t?/ has various origins. In Old
>> >French it was the result of a highly distinctive sound shift
>> >/ka/ > /t?a/. In Spanish it is the reflex of Latin -ct-. In
>> >Old Portuguese it is the reflex of Latin cl-. (There may be
>> >additional sources in Spanish and Portuguese, I don't know.)
>> Also pl. https://rudhar.com/etymolog/prantoen.htm
>>
>It's a pity that Gallego/ Galician is not included, since I've seen
>it described as retaining its original affricates. I'm not sure what
>'thst description means though.

In Standard Portuguese, chorar (to cry) sounds like English shoo-rahr.
In Northern dialect and in Galician, it sounds like English chew-rahr.
(Or maybe chaw-rahr.)

>If other Romance languages
>didn't retain their original affricates, what did they do with them?
>You too say 'original affricate'. Please explain what that means.

Used to be, but no longer.

[Ruud Harmsen]

Wed, 03 May 2023 07:29:06 +0200: Ruud Harmsen <r...@rudhar.com>
scribeva:

>In Standard Portuguese, chorar (to cry) sounds like English shoo-rahr.
>In Northern dialect and in Galician, it sounds like English chew-rahr.
>(Or maybe chaw-rahr.)

Or rather, t-show-rahr.

[Christian Weisgerber]

On 2023-05-03, Dingbat <ranjit_...@yahoo.com> wrote:

> It's a pity that Gallego/ Galician is not included, since I've seen
> it described as retaining its original affricates. I'm not sure what
> 'thst description means though. If other Romance languages
> didn't retain their original affricates, what did they do with them?
> You too say 'original affricate'. Please explain what that means.

Deaffrication of affricates is a common soundchange, e.g., [ts] > [s],
[dz] > [z], [tʃ] > [ʃ], [dʒ] > [ʒ]. Such changes have happened
independently in several Romance languages.

The following is based on
https://en.wikipedia.org/wiki/Comparison_of_Portuguese_and_Spanish#Sibilants

"The most marked phonetic divergence between Spanish and Portuguese
in their modern period concerned the evolution of the sibilants.
In the Middle Ages, both had a rich system of seven sibilants [...]
and spelled virtually the same in Spanish and Portuguese."

Medieval Spanish Modern Modern Modern
and Portuguese Portuguese Spanish Galician
/s/ /s/ /s/ /s/
/z/ /z/ /s/ /s/
/ts/ /s/ /θ/ or /s/ /θ/
/dz/ /z/ /θ/ or /s/ /θ/
/ʃ/ /ʃ/ /x/ /ʃ/
/dʒ ~ ʒ/ /ʒ/ /x/ /ʃ/
/tʃ/ /ʃ/ /tʃ/ /tʃ/

The column for Modern Galician is not from Wikipedia and reflects
my personal understanding.

Also, while both Medieval Galician-Portguese and Medieval Spanish
had a phoneme /tʃ/, it had arisen from different Latin sounds.

[Christian Weisgerber]

On 2023-05-03, Christian Weisgerber <na...@mips.inka.de> wrote:

> Medieval Spanish Modern Modern Modern
> and Portuguese Portuguese Spanish Galician
> /s/ /s/ /s/ /s/
> /z/ /z/ /s/ /s/
> /ts/ /s/ /θ/ or /s/ /θ/
> /dz/ /z/ /θ/ or /s/ /θ/
> /ʃ/ /ʃ/ /x/ /ʃ/
> /dʒ ~ ʒ/ /ʒ/ /x/ /ʃ/
> /tʃ/ /ʃ/ /tʃ/ /tʃ/

In other words, it looks like the development of the sibilants in
Galician mirrors that of Northern/Central Spanish and stands in
sharp contract to Portuguese. Both Galician and Spanish merged the
voiced sibilants into their voiceless counterparts; Galician only
lacks the Spanish retraction of /ʃ/ > /x/.

For a description of the Spanish development, see
https://en.wikipedia.org/wiki/Phonological_history_of_Spanish_coronal_fricatives#Historical_evolution

[Dingbat]

On Thursday, May 4, 2023 at 2:00:07 AM UTC+5:30, Christian Weisgerber wrote:
> On 2023-05-03, Christian Weisgerber <na...@mips.inka.de> wrote:
>
> > Medieval Spanish Modern Modern Modern
> > and Portuguese Portuguese Spanish Galician
> > /s/ /s/ /s/ /s/
> > /z/ /z/ /s/ /s/
> > /ts/ /s/ /θ/ or /s/ /θ/
> > /dz/ /z/ /θ/ or /s/ /θ/
> > /ʃ/ /ʃ/ /x/ /ʃ/
> > /dʒ ~ ʒ/ /ʒ/ /x/ /ʃ/
> > /tʃ/ /ʃ/ /tʃ/ /tʃ/
> In other words, it looks like the development of the sibilants in
> Galician mirrors that of Northern/Central Spanish and stands in
> sharp contract to Portuguese.
>

Where does Western Ladino stand in relation to Galician, Northern/
Central Spanish and Portuguese? If you know, that is. It was spoken
in NW Africa and known by various names including Haketia.

[Christian Weisgerber]

On 2023-05-05, Dingbat <ranjit_...@yahoo.com> wrote:

> Where does Western Ladino stand in relation to Galician, Northern/
> Central Spanish and Portuguese? If you know, that is.

Only what Wikipedia tells me. Ladino aka Judeo-Spanish branched off
from Old Spanish and didn't participate in most sound shifts that
set apart Modern Spanish. Here's the sibilant table again:

Medieval Spanish Modern Judeo- Modern Modern
and Portuguese Portuguese Spanish Galician Spanish
/s/ /s/ /s/ /s/ /s/
/z/ /z/ /z/ /s/ /s/
/ts/ /s/ /s/ /θ/ /θ/ or /s/
/dz/ /z/ /z/ /θ/ /θ/ or /s/
/ʃ/ /ʃ/ /ʃ/ /ʃ/ /x/
/dʒ ~ ʒ/ /ʒ/ /dʒ/, /ʒ/ /ʃ/ /x/
/tʃ/ /ʃ/ /tʃ/ /tʃ/ /tʃ/

Judeo-Spanish is more conservative than Modern Galician, which is
more conservative than Modern Spanish.