Latin ‘c’ with ‘k’ sound

< Previous | Next >

PersoLatin

Senior Member
UK
Persian - Iran
Why did Latin not use ‘k’ for cases where ‘c’ is pronounced with a ‘k’ sound, e.g. klass vs class or kolour vs colour?
 
Last edited:
  • Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    C was always pronounced with /k/ sound in Latin, the fricative or affricated pronunciations associated with that letter (/s/, /ʃ/ /tʃ/, /θ/, etc) in its descendants or in church Latin are the product of later sound changes.

    Latin inherited its alphabet from the Etruscans, who got theirs from western Greek people, who ultimately got theirs from Phoenician. Since each of those languages has its own consonant inventory that didn't match up with that of Phoenician, some letters got used for the same sound or reused for something completely different, and new letters had to be created.

    C comes from the Phoenician letter for /g/ (Arabic ج, Greek γ), but became used for /k/ in Etruscan that didn't have voiced stop consonants like /g/ or /b/.

    K comes from the Phoenician letter for /k/ (Arabic ك, Greek κ) and stayed that way

    Q comes from the Phoenician letter for (probably) /q/ -the uvular stop- (Arabic ق, Archaic Greek Ϙ, but also possibly Greek φ). Since Greek didn't have uvular stops, Ϙ was used to spell /k/ before back vowels like /u/ or /o/, possibly in imitation of the backing effect of the uvular/emphatic stop in its Semitic source. This was later abandoned, but not before the alphabet reached Etruscan, and thus Latin.

    Since Latin inherited 3 letters for its /k/ (and none for its /g/, leading to the invention of G, a modified C), it at first did something similar to the Greek treatment of Ϙ: They used all three to spell /k/, depending on what followed. C was used before I and E, K was used before A and Q was used before U and O (and as part of the digraph QU, that spelling thus representing both /ku/ and /kʷ/). Such a system was needlessly complicated however, and eventually abandoned, with the rarer K essentially falling out of use except in some abbreviations and the rare important word, like kalenda. Q was relegated to the digraph QU, now spelling only /kʷ/, while /ku/ would be written CU as in cum or locus.
     

    berndf

    Moderator
    German (Germany)
    ...that is, only when followed by another vowel, as in quantum, quæro, quid, quoniam, etc. :)
    I don't think there are any cases where <qu> isn't followed by a vowel, so we can't really tell. Theoretically, /kʷ/ could be part of a consonant cluster but I don't think this existed in Latin.
     

    berndf

    Moderator
    German (Germany)
    There aren't, that was exactly my point.
    Ok, let me phrase it differently:
    While being correct, your comment is unnecessary as the sound /kʷ/ did not occur except in front of a vowel. While you didn't explicitly say that, one might understand your comment as implying there are other cases of /kʷ/ where the sound was transcribed differently.
    I wanted to make sure you didn't mean to say that. We seem to be all in agreement. :)
     

    PersoLatin

    Senior Member
    UK
    Persian - Iran
    C was always pronounced with /k/ sound in Latin, the fricative or affricated pronunciations associated with that letter (/s/, /ʃ/ /tʃ/, /θ/, etc) in its descendants or in church Latin are the product of later sound changes.
    Thank you very much.

    So in Latin C was pronounced with /k/ sound even in cases where it was followed by /i/ & /e/.

    Did the start of sound changes you mentioned, marked the start of descendant languages appearing?

    Is there any information about which of the above changes appeared first or whether there was one or more intermediate sounds that have now disappeared?

    Also whether if one can predict the future sound changes based on what is known now?
     

    berndf

    Moderator
    German (Germany)
    First to /tʃ/ before the split up of proto-Romance. In Italian it remained like this. Reflexes in other languages like /s/ in French and /θ/ in Iberian Spanish are later developments.
     

    berndf

    Moderator
    German (Germany)
    Any ideas about when CE/CI started being pronounced with /tʃ/ sound? I've read contradictory hypotheses on this.
    So have I. Allen says there is no evidence for a /k/-palatalization "before the fifth century". That is probably the only thing we can say.
     

    berndf

    Moderator
    German (Germany)
    It is actually thought that -CE- and -CI- were affricated to /tse/ and /tsi/ in proto-Romance before ending up as /tʃ-/ in Italian, /θ-/ in northern Spain, and /s-/ elsewhere.
    That is difficult to say as neither Latin nor Greek nor Germanic distinguished /s/ and /ʃ/ at the time and the Latin /s/ was probably moving between [ s ] and [ʃ] like in modern Greek.

    The High German t-affricatization, which happened roughly at the same time, certainly produced [ts] rather than [tʃ] but if it was the same in Latin is difficult to ascertain.
     
    Last edited:

    pollohispanizado

    Senior Member
    Inglés canadiense
    Latin /s/ was probably moving between [ s ] and [ʃ] like in modern Greek
    Indeed, and an affricate would be different enough to distinguish. This is why in the North of Spain when CI /tsi/ was fricativized it was fronted as well to differentiate it from the apical S, whereas in the South SI and CI merged as predorsal. (I can't see /tʃ/ being fronted to /θ/, which is why I think the intermediate /ts/ makes the most sense.)

    (PD: On second thought, in Madrid /tʃ/ can be pronounced /c/ and almost /ts/, which then could theoretically get fronted further to /θ/... 🤔🤔🤔)

    I imagine for Italian, much like French, it's hard to draw a straight line back due to the fact that the modern national languages were not prestige or common languages/dialects in the past.
     
    Last edited:

    berndf

    Moderator
    German (Germany)
    In French, /tʃ/ and /ts/ probably separated because of the opposition between the palatalized c before a and before e/i, respectively.
     

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    In Picard, which didn't palatalise /ka/, patalised c became /ʃ/, the opposite outcome of Spanish.

    In Walloon and I think Lorrain, where Latin intervocalic /sk/ and /ss/ became /ʃ/ early and /ka/ palatalised, the deaffrication of /tʃ/ was blocked and patalised c merged into /s/ (Eastern dialects of Walloon show some traces of a late survival of the opposition, since word final /s/ from /ts/ is still pronounced (amicum > amice, bracchium > bresse), while word final /s/ from /s/ has disappeared).
     

    Catagrapha

    Member
    Malagasy
    Latin already has qu and ch for c, e.g., squilla for scilla, parochia for paroecia.
    Wiktionary says Italian macina is inherited from Latin machina, whereas macchina is borrowed.
     

    pollohispanizado

    Senior Member
    Inglés canadiense
    Do you know how French ended up with Q without U in words like coq,
    The problem with French orthography is that it's been artificially replenished with "etymological" and archaic spelling conventions. (Poids is spelled with a D because of the erroneous belief that it derived from pondus and not pensum.)

    Coq comes from coquus.
    Cinq comes from quinque.
     
    Last edited:

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    It is. Cf. catch-chase, a double loan of the same word, the former from northern French and the latter from standard French.
    The other evidence that English provides is that /s/ from /s/ sometimes ends up as /ʃ/ (cash, leash, etc.), while I can't think of any word with /s/ from /ts/ where this happened.

    I can't think of any French loanword in English that preserves the palatal affricate as /ts/, it's only /s/ or the very rare case like catch (where I have to wonder if the /tʃ/ isn't the result of the yod that was still present in most Oil varieties at the time: cacier /katsjer/ or /katʃjer/?).
     

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    Yes I'm aware. What I'm saying is that captiare would have first evolved into /tʃatsjer/ in OF and /katsjer/ in Old Normand and Picard (because stressed open syllable a diphthonged in palatal contexts, see carus > cher (OF chier) or canis > chien). This glide was later, with a few exceptions, deleted after palatal consonants (but not in Belgium, where Picard has /kaʃiː/ and Wallon /tʃasiː/, from a regular je > iː sound change), so depending when the loan happened, the input in English might have had /tsj/
     

    danielstan

    Senior Member
    Romanian - Romania
    In Aromanian the CE and CI are pronounced with /ts/ in Latin inherited words:
    arom. fatsiri - Wiktionary < lat. facere
    Aromanian has the sound /tʃ/, for example in Slavic loanwords:
    arom. ciocan < slav. čokan ("hammer")
    so this proves that the evolution of the Latin CE and CI stopped at the /ts/ phase.
    Although Aromanian does not have written sources earlier than 18th century, it seems to be one of the few (or the only) Romance language that preserved this /ts/ pronounciation from the pre-Romance time until today: Aromanian language - Phonology

    In Romanian the CE and CI groups are pronounced with /tʃ/ like in Italian,
    but some words still preserve the /ts/ pronounciation:
    rom. ghiață < lat. glaceus
    rom. față < lat. facia
    rom. soț ("husband")< lat. socius ("partner")
    rom. laț < lat. *laceus (laqueus)
    rom. desculț < lat. *disculceus (disculceatus)

    proving that a /ts/ phase existed in this language, too.

    As a personal opinion, the /ts/ sound was consistently preserved at the end of Romanian words
    because if it would have followed the phonetic evolution to /tʃ/
    it would have resulted in words like:
    *soci, *saci, *laci
    with an -i ending, a tipical plural marker in Romanian (as in Italian).

    In fact there is such a case in Romanian:
    rom. arici < lat. ericius
    which has identical forms at singular and plural.
     
    Last edited:

    Catagrapha

    Member
    Malagasy
    CH was a digraph used to represent the aspirated (and sometimes unaspirated) velar stops (χ, κ) in words borrowed from Greek. It's not a native spelling convention to my knowledge. The only word that is spelled with it that doesn't seem to come from Greek is pulcher, pulchra.
    Sepulchrum, less common than sepulcrum, perhaps influenced by pulchrum.
    Perhaps che chi and ce ci were conflated in medieval Latin, Romance arcipelago (from archipelagus), macina (from machina), and more seem to be the results thereof.
     

    pollohispanizado

    Senior Member
    Inglés canadiense
    As a personal opinion, the /ts/ sound was consistently preserved at the end of Romanian words
    because if it would have followed the phonetic evolution to /tʃ/
    it would have resulted in words like:
    *soci, *saci, *laci
    with an -i ending, a tipical plural marker in Romanian (as in Italian).
    What you say is very interesting. I know dreadfully little about the Eastern Romance languages. May I ask, what are the real plural forms of those words ending in ț?
     

    symposium

    Senior Member
    Italian - Italy
    In old Italian/Tuscan words that in modern standard Italian have a /tʃ/ sound used to have a /ts/ sound: one of the most egregious examples being the Uffizi (offices) gallery, where "zi" had and still has a /ts/ sound. The modern spelling of the word "offices" is "uffici", where "ci" has a /tʃ/ sound.
     

    danielstan

    Senior Member
    Romanian - Romania
    What you say is very interesting. I know dreadfully little about the Eastern Romance languages. May I ask, what are the real plural forms of those words ending in ț?
    Singular - plural forms:
    soț - soți (noun, masculin gender, meaning "husband")
    laț - lațuri (noun, neuter gender, meaning "noose")
    desculț - desculți (adjectiv, masculin gender, meaning "bare-footed, shoeless")

    Note that Romanian plurals of masculin words usually ends in -i, like in Italian, but the Romanian -i is a short sound that does not make a separate syllable (in phonetic notation: soți is pronounced [sotsĭ] in a single syllable)
    This evolution is thought to be influenced by the Slavic infinitiv which has a palatalized sound at the end,
    like Russian ходить - Wiktionary

    (I don't say that Russian had an influence on Romanian, but I give an example easier to understand).

    On another note, Romanian preserved the Latin neuter (with some deviations) with the plural ending -uri (from Latin ending -ora)
    for a big number of nouns which designate non animated notions (objects, thus not living beings).
    Example (singular - plural):
    Romanian timp - timpuri < Latin tempus - tempora
    This explains the unusual plural laț - lațuri.
     
    Last edited:

    danielstan

    Senior Member
    Romanian - Romania
    In old Italian/Tuscan words that in modern standard Italian have a /tʃ/ sound used to have a /ts/ sound: one of the most egregious examples being the Uffizi (offices) gallery, where "zi" had and still has a /ts/ sound. The modern spelling of the word "offices" is "uffici", where "ci" has a /tʃ/ sound.
    Another example in modern Italian:
    calza < lat. calcea (feminine of calceus) (Etimologia : calza;)

    Romanian has the verb încălța < lat. *incalciare (incalceare)
     

    pollohispanizado

    Senior Member
    Inglés canadiense
    Another example in modern Italian:
    calza < lat. calcea (feminine of calceus) (Etimologia : calza;)

    Romanian has the verb încălța < lat. *incalciare (incalceare)
    It's interesting that the derivatives in Romanian and in Italian maintain the /ts/, but in the West they continued the expected evolution (Sp. calza /kalθa/, Fr. chausse /ʃos/). I wonder if this is a coincident or if it had something to do with the dialect continuum between Italy and Romania.
     

    danielstan

    Senior Member
    Romanian - Romania
    Certainly it has to do with the dialectum continuum of Italian dialects South of La Spezia–Rimini Line and the Eastern Romance (Romanian, Aromanian etc.).

    Most of the Roman colonists in Dacia (and in Balkan Peninsula) were from Central and Southern Italian Peninsula.
    The usual road followed by the Roman army to the Balkans was:
    Rome (by Appian Way) - Brindisi - (crossing the Adriatic sea) - Dyrrachium (in today's Albania).
    As a confirmation, the closest Italian dialect to modern Romanian is Pugliese, located in the 'heel' of the Italian 'boot'.

    I don't know for sure from what territory were recruited the Roman colonists emigrated to Hispania and Gallia,
    but I guess they were from Central and Northern Italian Peninsula.
     
    Last edited:

    pollohispanizado

    Senior Member
    Inglés canadiense
    Rome (by Appian Way) - Brindisi - (crossing the Adriatic sea) - Dyrrachium (today Albania).
    As a confirmation, the closest Italian dialect to modern Romanian is Pugliese, located in the 'heel' of the Italian 'boot'.
    Thanks for the info. I knew about the Spezia-Rimini line, but I had never put together the sea crossing being the principal way east. Now it all makes much more sense.
     

    Sobakus

    Senior Member
    Regading what symposium mentions:
    In Italian and at least in the rest of Western Romance, there's been a lot of confusion between the reflexes of /tj/ as in puteum > pozzo, and /kj/ as in faciam > faccia (so regularly), but coriāceam > corazza, cuminitiāre > comenzare & cominciare. French has e.g. justice but tristesse, both from -itia; Old Spanish ŭncia > onza and onça. These make the most sense if we assume that the /ts/ phoneme had already existed by the time /kj/ started shifting to an affricate, and the initial outcome was that same existing phoneme, probably because it wasn't specified for being palatal or dental. The words that failed to affricate in the first round later underwent the same development as velars before front vowels (as opposed to /j/). In fact, by about the 4th century the standard pronunciation of /tiV/ (before vowels) was with an affricate, and [ti~j] was considered a barbarism by the grammarians, so we must conclude that the palatalisation of velar stops is a later development, which took a different course in different areas. Certainly the palatalisation of /kj/ must precede that before vowels. That and the fact that the former, but not the latter has happened in Sardinian.

    By the way, this is also the source of the -ci-/-ti- confusion rampant in Late and Medieval Latin.
     
    Last edited:

    pollohispanizado

    Senior Member
    Inglés canadiense
    I was looking up the Romanian word oţet (vinagre), figuring it came from acetum, however I wasn't expecting it to have come through "Proto-Slavic" (Wiktionary cites the PS root as *ocьtъ, and gives cognates in all the Slavic languages, including Bulgarian/Makedonian оцет). This isn't exactly evidence, but I think it does show that this word was borrowed from Latin with the /ts/ pronunciation. (According to my minimal research, the Proto-Slavic period ended by the 6th century CE.)
     
    Last edited:

    danielstan

    Senior Member
    Romanian - Romania
    About Romanian word oţet I found some etymologies online and all are deriving it from a Slavic (here there are variations, but no Proto-Slavic) word which on its turn comes from Latin acetum:
    oțet - dexonline
    oțet - Wikționar

    This oţet is not inherited directly from Latin because the sound evolution
    a > o is unusual in Romanian.
    See other Romanian words with similar beginings in Latin:
    rom. ață < lat. acia (ață - dexonline)
    rom. ac < lat. acus (ac - dexonline)
    rom. acru < lat. *acrus (acer) (acru - dexonline)

    On another hand see:
    rom. oltar < slav. олътарь < lat. altarus (oltar - dexonline)
    which suggest the transformation a > o did happen in Slavic for Latin loanwords.

    If you like to investigate by yourself the Romanian vocabulary in order to establish some ad-hoc phonetic rules,
    see this site:
    CuvinteCare.ro - Cuvinte care încep sau se termină cu anumite litere ("words starting or ending with certain letters")
    and build in it search criteria like:
    'cuvinte care încep cu ac' ("words starting with 'ac'")
    'cuvinte care conțin ac' ("words containing 'ac'") etc.

    Then use www.dexonline.ro to find etymologies compiled from many other Romanian dictionaries.
    (DEX = short for "Dicționar EXplicativ al limbii române")
     
    Last edited:
    < Previous | Next >
    Top