Romance languages: Did the ancient Romans pronounce the letter 'H'?

  • lazarus1907

    Senior Member
    Spanish, Spain
    No one has ever heard Romans speak and the pronunciation evolved in time, so people don't seem to agree completely.

    There are a few pronunciation "schools":

    Classic: H is pronounced at the beginning of a word almost like in English, but it is not pronounced everywhere else (pulcher, thesaurus). "Ph" sounds like "f".

    Ecclesiastic: It has no sound, except in "mihi", where it sounds like a "k"
     

    Jhorer Brishti

    Senior Member
    United States/Bangladesh English/Bengali
    From what I know(which doesn't mean that I'm necessarily correct!) the H was only pronounced by high officials and not by the common people or at least this is what the situation had fallen into towards the end of the Roman Empire.. I do know that the Ph should be pronounced as an aspirated P(normal english P) rather than an F in classical latin.
     

    Outsider

    Senior Member
    Portuguese (Portugal)
    rambler said:
    I've noticed that the letter 'H' is not pronounced in Italian, Spanish, and French.

    (1) Is there a theory about why this is so?
    I'm not sure that there is a specific theory about it, but the silencing of the aspirated h seems to be a common phonetic phenomenon. It happened twice in French, and it's even occurred in some words of Germanic languages, including English.

    rambler said:

    (2) Did the ancient Romans pronounce it?
    Yes, they did.
     

    diegodbs

    Senior Member
    Spain-Spanish
    Outsider said:
    I'm not sure that there is a specific theory about it, but the silencing of the aspirated h seems to be a common phonetic phenomenon. It happened twice in French, and it's even occurred in some words of Germanic languages, including English.

    Yes, they did.
    I once read that the aspirated H in Latin was something taken from the Greek "spirit" written on some vowels to make them sound aspirated, as in Greek. Perhaps that aspiration was something that only the most cultured or richer classes did. I don't know.
     

    Outsider

    Senior Member
    Portuguese (Portugal)
    The H is not only present in loan words from Greek. It's also found in genuinely Latin words, and you can find it in texts and inscriptions dating as far back as the archaic period. Perhaps you are thinking of the digraphs CH, PH, RH and TH, which were indeed introduced into Latin by hellenophile intellectuals.
     

    diegodbs

    Senior Member
    Spain-Spanish
    Outsider said:
    The H is not only present in loan words from Greek. It's also found in genuinely Latin words, and you can find it in texts and inscriptions dating as far back as the archaic period. Perhaps you are thinking of the digraphs CH, PH, RH and TH, which were indeed introduced into Latin by hellenophile intellectuals.
    You are right Outsider. On second thoughts, I wonder how I could believe what I was told. :)
     

    polaco

    Member
    Poland/polish
    Outsider said:
    The H is not only present in loan words from Greek. It's also found in genuinely Latin words, and you can find it in texts and inscriptions dating as far back as the archaic period. Perhaps you are thinking of the digraphs CH, PH, RH and TH, which were indeed introduced into Latin by hellenophile intellectuals.
    Of course for example heres - English heir

    greetings
     

    rambler

    Senior Member
    English Canada (blizzards!)
    I greatly appreciate all the interesting responses to my question.

    If I may, I'd like to summarize the responses, in case I have misunderstood anyone.
    (1) Is there a theory about why this is so?

    The silencing of the aspirated H is a common phonetic phenomenon.
    It happened in Italian, Spanish, and French, and even in some English words (e.g. ‘heir’).

    (2) Did the ancient Romans pronounce it?


    (a) Yes, they did.

    (b) No, ‘H’ was pronounced at the beginning of a word, but nowhere else.

    (c) No, the ‘H’ was only pronounced by high officials and not by the common people.

    (d) It was never pronounced, except in "mihi", where it sounded like a "k"
    Are there any other opinions?

    Again, thank you, everyone, for you responses!
     

    MarX

    Banned
    Indonesian, Indonesia
    the silencing of the aspirated h seems to be a common phonetic phenomenon. It happened twice in French
    You're right. In Indonesian there is also a tendency to drop the H.

    By the way, what do you mean it happened twice in French?
     

    Outsider

    Senior Member
    Portuguese (Portugal)
    By the Vulgar Latin period, all Latin and Greek aitches had become silent. But then the /h/ phoneme reappeared in a number of French loanwords from Old German (where /h/ still existed). There is evidence that the "h" was pronounced in these loanwords, because they behave differently from others in liaison. But in modern French even those words are no longer pronounced with an /h/ sound. They are called words with an ache aspiré.
     

    MarX

    Banned
    Indonesian, Indonesia
    By the Vulgar Latin period, all Latin and Greek aitches had become silent. But then the /h/ phoneme reappeared in a number of French loanwords from Old German (where /h/ still existed). There is evidence that the "h" was pronounced in these loanwords, because they behave differently from others in liaison. But in modern French even those words are no longer pronounced with an /h/ sound. They are called words with an ache aspiré.
    I get it. :)

    Merci!
     

    Forero

    Senior Member
    Ancient Romans certainly pronounced the h.

    The Romans knew next to nothing about Indoeuropean etymology, but they wrote hs in native Latin words in places that correspond to where other Indoeuropean languages had/have related consonants (Greek chi, English g or y, German g or k, Sanskrit gh, etc.).

    For example, etymologically speaking, vehiculum corresponds to wagon, habeo to give, trahere to drag, hortus to yard and garden, hospes to guest, etc. The Romans did not get these forms from us, nor we from them, but they go back to a common source, of which the Romans were presumably unaware.
     

    tashana

    New Member
    Serbia & serbian
    Here is an example : Una hirunda non facit ver ( translation:One martin doesn't make the spring) I think I spelled it OK. It's one of my favorite sayings.
     

    SerinusCanaria3075

    Senior Member
    México, D.F. (Spanish)
    Ancient Romans certainly pronounced the h.

    The Romans knew next to nothing about Indoeuropean etymology, but they wrote hs in native Latin words in places that correspond to where other Indoeuropean languages had/have related consonants (Greek chi, English g or y, German g or k, Sanskrit gh, etc.).

    For example, etymologically speaking, vehiculum corresponds to wagon, habeo to give, trahere to drag, hortus to yard and garden, hospes to guest, etc. The Romans did not get these forms from us, nor we from them, but they go back to a common source, of which the Romans were presumably unaware.
    I can see your comparison between Greek and Latin, but in no way can you compare English to Latin :thumbsdown:since English is a Germanic language that evolved in the middle ages.

    It is not pronounced in Portuguese too either.
    The "h" is pronounced in Romanian, which if I'm not mistaken is a transliteration of the Greek/Slavic [Cyrillic] "Χ" rather than a conservation of the h from Latin words:

    (Russian/Bulgarian) Дух.
    (Czech/Polish) Duch.
    (Romanian) Duh.

    (Greek) Χριστός.
    (Russian/Bulgarian) Христос.
    (Romanian) Hristos.

    Of course this has nothing to do with Latin, but it proves that Romanian pronounces the "h".
    I doubt any h's were kept from Latin.

    I've noticed that the letter 'H' is not pronounced in Italian, Spanish, and French.
    Actually, some French nouns have the h "aspiré" where one does pronounce it:
    La honte, le héros.
     

    Forero

    Senior Member
    I can see your comparison between Greek and Latin, but in no way can you compare English to Latin :thumbsdown:since English is a Germanic language that evolved in the middle ages.
    Spanish also evolved in the Middle Ages. But English and Spanish both have ancestors from before the Middle Ages, Anglo-Saxon and Latin, respectively. In fact, Anglo-Saxon was a Germanic language and Latin was an Italic language, but they had a common ancestor, called Proto-Indoeuropean. The h in native Latin words corresponds to an aspirated voiced gh sound in Proto-Indoeuropean. The Germanic version of PIE gh is g, voiced but not aspirated, and the Greek version is chi (looks like an x, sounded like kh), aspirated but not voiced. English changes some of the Germanic gs to ys, for example yard from Anglo-Saxon geardum, related to Gothic garda, Phrygian gordum, Hittite gurtas, Greek chortos, and Latin hortus.
    Actually, some French nouns have the h "aspiré" where one does pronounce it:
    La honte, le héros.
    The French don't pronounce any consonant at the beginning of honte or héros. The h here is called aspiré only because the vowels do not contract together (l'onde but la honte).
     

    Montesacro

    Senior Member
    Italiano
    Yes, ancient Romans pronounced the letter H.
    But it was kind of an upper class thing.
    Common people would normally drop their aitches in everyday speech.

    All in all the situation was pretty much similar to the present usage in a large part of England.


    Here’s Catullus’s carmen 84: it says it all…

    LXXXIV. ad Arrium
    CHOMMODA dicebat, si quando commoda uellet
    dicere, et insidias Arrius hinsidias,
    et tum mirifice sperabat se esse locutum,
    cum quantum poterat dixerat hinsidias.
    credo, sic mater, sic liber auunculus eius.
    sic maternus auus dixerat atque auia.
    hoc misso in Syriam requierant omnibus aures
    audibant eadem haec leniter et leuiter,
    nec sibi postilla metuebant talia uerba,
    cum subito affertur nuntius horribilis,
    Ionios fluctus, postquam illuc Arrius isset,
    iam non Ionios esse sed Hionios.



    The dangers of hypercorrection! :D
     

    Hulalessar

    Senior Member
    English - England
    When the Romans started to write Latin they adapted the Etruscan alphabet including a symbol for /h/. If they did not have the sound they would not have borrowed the symbol.
     

    Maroseika

    Moderator
    Russian
    The "h" is pronounced in Romanian, which if I'm not mistaken is a transliteration of the Greek/Slavic [Cyrillic] "Χ" rather than a conservation of the h from Latin words:

    (Russian/Bulgarian) Дух.
    (Czech/Polish) Duch.
    (Romanian) Duh.
    This sound in said Slavic word has nothing to do with the Greek "X". Slavic дух/duch corresponds to the Greek θεός < *θεσός.
     

    Forero

    Senior Member
    How "ancient" is Catullus’s carmen 84? It appears that the h was not pronounced in that particular place and time.

    There was a graffito found in Pompeii, written around the 1st century AD, that says "abiat Venere Bompeiiana iratam qui hoc læserit" ("May he who messes with this have the anger of Pompeii's Venus"). The first word is traditionally written "habeat" and goes back to a PIE root with aspirated gh that is also the source of our verb "give".

    The earliest Latin inscription on record is on a belt buckle from the 7th century B.C. that says "Manios med fhefhaked Numasioi" ("Manius made me for Nummerius"). This too has roots in PIE aspiration, in particular the aspirated dh sound (twice) in the PIE source of our word "did".

    Source: The Story of Latin and the Romance Languages by Mario Pei.

     

    sokol

    Senior Member
    Austrian (as opposed to Australian)
    Catull lived in the 1st century BC.

    Also Petron's Cena Trimalchionis (1st century AD) has, if memory serves right, hypercorrected 'h' in cases where there is no 'h' at all (you could compare Petron's comic use of this to hypercorrect 'h' in British English which would tell you that the speaker could be a Cockney native trying to mask this): which is the sign of an upstart trying to mask his roots in lower class where 'h' wasn't pronounced any more.

    But then it is possible that the loss of 'h' only was a Southern Italian feature at the time - I don't really have much knowledge of the history of Latin.


    Anyway, Cena Trimalchionis shows that in the 1st century AD it was still considered proper and correct use to pronounce 'h', which as well tells us that the loss of 'h' already was a marker for lower class Vulgar Latin speakers.

    Such processes as the loss of 'h' anyway do not occur in an instant and then spread trough the whole language community - they start in some region and/or in some social class and spread from there. It may take centuries before such a process is completed.
     

    berndf

    Moderator
    German (Germany)
    The earliest Latin inscription on record is on a belt buckle from the 7th century B.C. that says "Manios med fhefhaked Numasioi" ("Manius made me for Nummerius"). This too has roots in PIE aspiration, in particular the aspirated dh sound (twice) in the PIE source of our word "did".
    It seems to be consensus view today that the Praeneste fibula is a hoax.

    EDIT 9 years later: It now looks that it is genuine.
     
    Last edited:

    Miguel Antonio

    Senior Member
    Galego (Rías Baixas)
    except in "mihi", where it sounds like a "k"
    What about nihil? In Spanish the verb is spelled aniquilar, reflecting the purported original pronunciation of the letter H. The name of the letter itself in Galician and Portuguese betrays its pronunciation too: AHA>aka>agá.

    Egyptian Arabic tends to drop hard consonants for glottal stops: maqalaqah (teaspon) is pronounced "ma'ala'a". A similar phenomenon may have happened to the Latin H

    vale
     

    Outsider

    Senior Member
    Portuguese (Portugal)
    But I think sometimes an aspirated can become a [g] without an intervening [k]. For example, in Russian español becomes gishpanskii (from Hispanus?), and harmonía becomes garmoniya...
     

    berndf

    Moderator
    German (Germany)
    As far as I know the sound shift form to [g] is regularly employed in Slavic languages when assimilating foreign words. Another example is the name of the philosopher Hegel which is pronounced Gegel in Russian. This means just that [g] is the native consonant which sounds closest to the foreign to a Slavic ear.
     

    Toma

    Member
    Bulgarian
    In short:

    IE did not possess a separate H sound.
    The sound H did exist in Latin and its weakening is a later phenomenon.
    The dissappearance of the H sound in Romance languages is a later development.
    To slavic ears foreign H deos not sound as G. This is typical only for Russian and only at the beginnig of words.
     

    Forero

    Senior Member
    In short:

    IE did not possess a separate H sound.
    The sound H did exist in Latin and its weakening is a later phenomenon.
    The dissappearance of the H sound in Romance languages is a later development.
    To slavic ears foreign H deos not sound as G. This is typical only for Russian and only at the beginnig of words.
    Hi, Toma.

    Any idea when or where the h sound was consistently pronounced in Latin?
     

    sokol

    Senior Member
    Austrian (as opposed to Australian)
    But I think sometimes an aspirated can become a [g] without an intervening [k]. For example, in Russian español becomes gishpanskii (from Hispanus?), and harmonía becomes garmoniya...


    This is because in southern Russian and Ukrainian the northern Russian 'g' is pronounced 'h', so this is only motivated through script: these names seem to have been established in Russian either at a time when the 'h' pronunciation of 'g' was dominant [probably when Kiev was politically dominant, but I am not sure if this were an established fact or only a theory, or not even that one], or alternatively they took the route (geographically) over the south to the north.

    The transformation of 'h' to 'k' or 'g' (in Spain and Portuguese) of course is a possible one; if such a things happen then one can be sure that the previous 'h' in Latin (at least the Vulgar Latin as spoken in those regions) still was pronounced.
     

    Toma

    Member
    Bulgarian
    Hi Forero,

    I don't have with me the book with a very good article on Latin H, but thia is waht Sihler writes in his New Comparative Greek and Latin Grammar (a rework of Buck's comparative grammar, but a quesitonable one) ff.158.
    'the sound written h wa faintly sounded and probably absent in collaoquial speech from an early period. (There seem to be some discrepancies in spelling, but only a few are mentioned)' finally he adds 'the letter h was sometimes used as a sign of hiatus as in AHENUS beside AENUS (made of bronze) where the H stands for no consonant, only to distinguish aee (long e) from the diphthong ae in writing. Such a use porves that H was used sporadically at best, since otherwise the lwtter H would hvae been no more suitable than any other consonant...'

    Salvete
     

    surangamas

    New Member
    English
    I have to second with Forero above, to add my two cents into this discussion, I will bring up the following for people to think about:

    1. If we compare Latin "homini" with modern standard Italian "uomini", I would be inclined to assume that "h" in classical Latin might've sounded like either "gh" or a muddy "h" (IPA symbol: ɦ). The disappearance of the Latin h in modern Italian might have been weakened, or gone through a process called "assimilation." (I might not be correct on this so feel free to correct me if I am wrong). In French and Spanish however, the aspirated h was eliminated rather than being assimilated.

    2. In English and German, the retention of the aspirated h might've gone through a different process, that is, "gh" simply became voiceless rather than being assimilated into another sound.

    Hope this helps.
     

    danielstan

    Senior Member
    Romanian - Romania
    We know how ancient Latin pronounced or not the H from ancient grammarians who mentioned this in their writings.

    In Vox Latina, page 45, is quoted St. Augustine which complains on H not being pronounced in hominem
    (source: Vox Latina)

    ... ut qui illa sonorum uetera placita teneat aut doceat, si contra disciplinam grammaticam sine adspiratione primae syllabae ominem dixit, displiceat magis hominibus quam si contra tua praecepta hominem oderit
    Saint Augustine lived in between 354 - 430 AD, thus before the fall of Western Roman Empire (476 AD), so his testimony applies to Vulgar Latin spoken in late Antiquity, centuries before Romance languages emerged.
    No wonder that no Romance language has preserved the Latin sound H.

    ---------------------
    Romanian has not inherited the H sound from Latin (no Romanian word of Latin origin preserved this sound).
    The H entered in Romanian massively by Slavic loandwords and loanwords from other languages like modern Greek, Hungarian and so.
     

    berndf

    Moderator
    German (Germany)
    The disappearance of the Latin h in modern Italian might have been weakened, or gone through a process called "assimilation." (I might not be correct on this so feel free to correct me if I am wrong). In French and Spanish however, the aspirated h was eliminated rather than being assimilated.
    It seems very clear that the loss of the Latin h began during the classical period. There is no point in theorising about different processes in various Romance languages.

    The French aspirated h appears in Germanic loans and is unrelated the Latin h which was at that time already lost.

    2. In English and German, the retention of the aspirated h might've gone through a different process, that is, "gh" simply became voiceless rather than being assimilated into another sound.
    The Germanic h is a spirantisation of /k/ as per Grimm's law and evolved through a very different process than the Latin h. See English give and its Latin cognate habere and English have and its Latin cognate capere.
     

    Sardokan1.0

    Senior Member
    Sardu / Italianu
    I have to second with Forero above, to add my two cents into this discussion, I will bring up the following for people to think about:

    1. If we compare Latin "homini" with modern standard Italian "uomini", I would be inclined to assume that "h" in classical Latin might've sounded like either "gh" or a muddy "h" (IPA symbol: ɦ). The disappearance of the Latin h in modern Italian might have been weakened, or gone through a process called "assimilation." (I might not be correct on this so feel free to correct me if I am wrong). In French and Spanish however, the aspirated h was eliminated rather than being assimilated.

    2. In English and German, the retention of the aspirated h might've gone through a different process, that is, "gh" simply became voiceless rather than being assimilated into another sound.

    Hope this helps.

    I think that the Italian "uomini" it's due to grammatical errors happened during the evolution from vulgar Latin to ancient Italian

    homo -> illu homo -> lu omo -> l'uomo

    a similar thing but reversed can be seen in the name of the region Apulia -> illa Apulia -> la Apulia -> la Puglia

    While in Sardinian the most conservative language of the Romance family there is no trace of a particular pronounce of the H, it's always mute, and often also omitted in writing

    habere -> háere (to have)
    hominem -> homine (man)
    hibernum -> hiérru (winter)
    heri -> héris (yesterday)
    hoc annus -> hoc annu (this year)
    in hoce -> in hoche, in hoghe (here)
    in hice -> in hie (there)
     

    Dymn

    Senior Member
    I think that the Italian "uomini" it's due to grammatical errors happened during the evolution from vulgar Latin to ancient Italian

    homo -> illu homo -> lu omo -> l'uomo
    But the ò > wò diphthongization happened in open syllables, many inside a word (fuoco, suono, etc.). I don't think it has to do with the definite article here. Puglia is another case in my opinion.
     

    Sardokan1.0

    Senior Member
    Sardu / Italianu
    Same in Corsican "omu", but how the hell happened that in Italian the pronounce turned to "uomo, fuoco, suono, luogo, tuono" etc.etc. it's a recent innovation? since in the middle ages we still find "omo, foco" etc.etc.

    While in Corsican and Sardinian the things remained as they were in origin


    Italian - Corsican - Sardinian

    uomo - omu - homine
    fuoco - focu - fogu
    suono - sonu - sonu
    luogo - locu - logu
    tuono - tonu - tronu
     

    Nino83

    Senior Member
    Italian
    "uomo, fuoco, suono, luogo, tuono" etc.etc. it's a recent innovation?
    It seems that in Tuscany there were two forms, /wò/ (literary) and /ò/ (popular), according to Rholfs. The former entered the Italian language, the latter is still present today in the Tuscan dialects.
    On the other end, the Vulgar Latin /è/ in open syllables is a diphthong in the Tuscan dialects (piètra, Italian and Tuscan, pètra in other peninsular Romance languages like Sicilian, Romanesco and so on).
     
    Last edited:

    Dymn

    Senior Member
    And Spanish has fuego, sueno (archaic), luego ("later") and trueno. In Aragonese and most Asturian dialects there's also (ò > we), and in some Asturian dialects (ò > wo). Is Italian the only language in Italy to feature this diphthongization? How about Neapolitan, Lombard, Piedmontese, etc.?
     

    Nino83

    Senior Member
    Italian
    Is Italian the only language in Italy to feature this diphthongization? How about Neapolitan, Lombard, Piedmontese, etc.?
    Among the languages south of the La Spezia-Rimini line, Tuscan and Standard Italian are the only languages with these changes /è/ > /jè/, /ò/ > /wò/ (/ò/ in popular Tuscan) in open syllables.
    North of the La Spezia-Rimini line, the Gallo-Italian languages have or had diphthongs in open syllables, for both open and closed vowels, like in French.
    French: è > jè, ò > wò > ø, é > éi > ói > oè > wa, ó > óu > éu > ø
    Piedmontese: è > jè > é, ò > wò > ø, é > éi, ó > óu > u
    Lombard: è > jè > é, ò > wò > ø, é > éi > é, ó > óu > u
    Bolognese: è > jè > ii, ò > wò > uu, é > èi > ai, ó > òu > au
     
    Last edited:

    Sardokan1.0

    Senior Member
    Sardu / Italianu
    On the other end, the Vulgar Latin /è/ in open syllables is a diphthong in the Tuscan dialects (piètra, Italian and Tuscan, pètra in other peninsular Romance languages like Sicilian, Romanesco and so on).
    Same things on these shores

    Italian - Corsican - Sardinian

    pietra - petra - pedra
    fieno - fenu - fenu
    tieni - teni - tene
    vieni - veni - beni
    dieci - deci - deche, deghe
    ieri - eri - héris

    etc.etc.
    about the vowels, Sardinian doesn't make distinction between long or short vowels, they are pronounced all in the same way, that's why when a Sardinian speaks Italian is unable to differentiate them, like in pèsca or pésca, we pronounce both in the same way.

    from Wikipedia :

    The Proto-Romance loss of phonemic length originally produced a system with nine different quality distinctions in monophthongs, where only original /ă ā/ had merged. Soon, however, many of these vowels coalesced:

    • The simplest outcome was in Sardinian, where the former long and short vowels in Latin simply coalesced, e.g. /ĕ ē/ > /e/, /ĭ ī/ > /i/: This produced a simple five-vowel system /a e i o u/.
     
    < Previous | Next >
    Top