Evolution of consonant+rhotic clusters

Dymn

Senior Member
All Indo-European languages, or at least those in Europe, seem to include a great variety of consonant+rhotic clusters such as pr tr kr br dr gr fr vr, and I don't know any of them which has changed them into something else, even if the rhotic sound itself might have shifted recently in some of these languages...

Is there any example of a language simplifying those clusters? Not necessarily from the Indo-European family. The closest thing I can imagine is -str- > -s- in Galician-Portuguese (nosso, amossar...).

Thanks
 
  • Zec

    Senior Member
    Croatian
    There is commonly dissimilation or epenthesis in sibilant + rhotic clusters - which is why clusters like sr and zr are not as common as the ones you mentioned. Changes such as sr > str, sr > fr or sr > ʃ often happen. The last one happened in a bunch of Iranian languages (something similar is in the process of happening in many varieties of English - listen to how words with tr and dr can be pronounced with affricates instead of stops), and Middle Indo-Aryan assimilated all dental + r clusters into retroflexes, while Tibetan assimilated all stop + r clusters into retroflexes.
     

    Lusus Naturae

    Senior Member
    Cantonese
    Fredericus > Federico
    I'm interested in finding more examples like it.

    I have a slightly different question. It's about rhotic+consonant. I saw Ordessus & Odessus, but not Oddessus. Did Ordessus become Oddessus before becoming Odessus (kinda like ursus > osso > oso)?
     

    Dymn

    Senior Member
    Fredericus > Federico
    I'm interested in finding more examples like it.
    In Spanish:

    l and r switched: Argelia, milagro, palabra, peligro
    r dropped: Federico, fragancia, orquesta, propio
    r added: estrella
    r changed place: cocodrilo, costra, entregar, preguntar
    r to l: árbol, azul, bolsa, cárcel, Catalina, Cristóbal, escolta, miércoles, papel, quilate, roble, templar, tiniebla
     

    Lusus Naturae

    Senior Member
    Cantonese
    ^
    Great examples!

    How often do consonants other than l become r?

    *medidies > meridies
    Did medidies become melidies before becoming meridies (like cicada > cicala > cigarra)?
     

    danielstan

    Senior Member
    Romanian - Romania
    Romanian (all its dialects: Daco-Romanian, Aromanian, Megleno-Romanian and Istro-Romanian) has a phonetic rule that distinguish it in the family of Romance languages:
    - intervocalic single -L- from Latin > -R-
    lat. solem > rom. soare
    lat. caelum > rom. cer
    Note: the intervocalic double -LL- > -L-
    lat. pellem > rom. piele
    I suppose that the distinction -L- vs. -LL- was difficult to maintain and the speakers chose these replacement -L- > -R- (both L and R being liquids) in order to keep this distinction with simplified means.

    Another rhotacism in some Romanian dialects is:
    N > R
    In standard Romanian there are few examples:
    lat. fenestra > rom. fereastră
    lat. minutus > rom. runt
    But in Istro-Romanian and in the regional Romanian spoken in Maramures (North-West of Romania) this is almost regular:
    lat. bonum > istr. bur
    lat. bene > istr. bire
    lat. lumen > istr. lumira
    (Istro-Romanian language - Wikipedia)
     

    Sardokan1.0

    Senior Member
    Sardu / Italianu
    In many Sardinian dialects the Latin "Pl or Cl" has become "Pr and Cr"

    Latin - Sardinian (Nuorese - Logudorese - Campidanese)

    platea - pratta - piatta - pratza (square)
    ecclesia - cresia - creja, cheja - cresia (church)
    florem - frore - fiore - frori (flower)
    clavem - crave - jae - crai (key)
    cravus - cravu - jau - crau (nail)
    oculus - ocru - oju - ogu (eye)
    placere - pràghere - piàghere - pràxiri (to like)


    While in some northern Sardinian dialects and also in Gallurese (a Corsican dialect spoken in the north-eastern corner of Sardinia) there is a change from the original R to L.

    Sardigna -> Saldigna
    Perdere -> Peldere (to lose)
    Surdu -> Suldu (deaf)
    Màrtis -> Màltis (Tuesday)
    Mèrcuris -> Mèrculis, Mèlculis (Wednesday)
     

    danielstan

    Senior Member
    Romanian - Romania
    The difference between Romanian and other Romance languages is the intervocalic position of single -L- which become intervocalic -R-.
    Anyway, L and R are 2 sounds with similarities (they "flow") and there is no surprise for their interchange.
    By the way, it happened to me to hear small children pronouncing L instead of R everywhere.
    If I well remember, Japanese has a regular transformation L > R for loanwords.
     

    Määränpää

    Senior Member
    Finnish
    From Swedish (IE) into Finnish (non-IE)

    prost - rovasti (provost)
    trall - ralli (folk song)
    krabba - rapu (crab)
    brillor - rillit (eyeglasses)
    dräng - renki (farmhand)
    grammofon - rammari (gramophone)
    franska - ranska (French)
    vrak - raakki (wreck) (hey, the w is silent in the English equivalent!)
     

    Cossue

    Senior Member
    Galician & Spanish
    Also some Galician dialects change /z/ to [r] at the end of words before voiced consonants I think.
    Yep, that's rather common; in fact it has affected the current form of a number of place names:

    Cardalda, Vilanova de Arousa, Pontevedra < Casa d'Alde 1295
    Cardiguimbra, Bande, Ourense < *Casa de Guímara
    Cardanachama, Allariz, Ourense < *Casa de Dona Chamoa < *Casa de Domna Flammula
    ...
     

    Penyafort

    Senior Member
    Catalan (Catalonia), Spanish (Spain)
    Is there any example of a language simplifying those clusters? Not necessarily from the Indo-European family. The closest thing I can imagine is -str- > -s- in Galician-Portuguese (nosso, amossar...).
    Nuesu, vuesu in Eastern Asturian and nueso, vueso in Western Aragonese can also be found locally. I also recall it in some Old Spanish poems. Maybe -STR- > -SS- was not that uncommon in some Late Latin forms.
     

    Penyafort

    Senior Member
    Catalan (Catalonia), Spanish (Spain)
    It's not Old Spanish but maese Pérez for example?
    Clearly. According to Coromines, maestre is taken from Catalan or Occitan but maese would come from the vocative of MAGISTER. That'd explain why it is not maeso.
     

    apmoy70

    Senior Member
    Greek
    Ιn Greek, the rhotacism l > r before a consonant is a phenomenon regular in demotic MoGr, appearing in occasional form as early as the 2nd c. CE.
    The earliest known attestation is the commemorative plaque at the Delphi Oracle, on the occasion of the honorary citizenship bestowed upon the Roman emperor Marcus Aurelius (reigned from 161-180 CE):
    «Μ. Αὐρ. Δερφοί πολείτην εποίησαν ἐτείμησαν δὲ καὶ ἀνδριάντος ἀναστάσει»
    "Derphi [Delphi] made M. Aur. a citizen and honoured him with his statue".

    Besides the rhotacism of l > r in Δερφοί, one can also see the misspellings of ἐτείμησαν (instead of ἐτίμησαν) & πολείτην (for πολίτην) which proves the completion of the rising of the diphthong «-ει-» to /i:/ in the Koine Greek of 2nd c. CE.

    In demotic MoGr the usage of «αδερφός» [aðelˈfos] --> brother for «αδελφός» [aðelˈfos], and «κόρφος» [ˈkorfos] instead of «κόλπος» [ˈkolpos] is common.
    The Thessalian «Βούργαρος» [ˈvurɣaɾos] for «Βούλγαρος» [ˈvulɣaɾos] --> Bulgarian, is also interesting (possibly influenced by the Aromanian change l > r).

    One very interesting and uncommon (at least in Greek) rhotacism, is the shift of the intervocalic d > r in the Koine of Pamphylia (Pamphylian Greek).
    On an epitaphic plaque found in Aspendos (2nd c. CE) the name of the dead is spelt «Ἐπιτῑμίρᾱς» Ĕpĭtīmírās instead of «Ἐπιτῑμίδᾱς» Ĕpĭtīmídās, while the Classical v. «ἀείδω» ăeí̯dō (<*ἀϝείδω *ăweí̯dō, Attic «ᾄδω» ắ̩dō) is «ἀβήρω» ăbḗrō.
     

    danielstan

    Senior Member
    Romanian - Romania
    The Thessalian «Βούργαρος» [ˈvurɣaɾos] for «Βούλγαρος» [ˈvulɣaɾos] --> Bulgarian, is also interesting (possibly influenced by the Aromanian change l > r).
    I guess you are talking about the transformation:
    intervocalic single -L- > -R-
    which is present in Daco-Romanian and Aromanian, but is not applicable in your Greek example.
    Examples:
    lat. solem > rom. soare, arom. soari

    For the sake of discussion, in Latin the single intervocalic -L- was pronounced differently than the double intervocalic -LL-, i.e. the syllable boundaries were different:
    lat. mola [mo-la] > rom. moară [moa-ră]
    lat. pellem [pel-lem] > rom. piele [pie-le]
    So, in my opinion, the initial difference in pronunciation was kept in Romanian, but simplified to the difference between intervocalic -l- and -r-.

    Because this phonetic rule is present in all Romanian dialects, we may suppose it was produced before the separation of Aromanian from Daco-Romanian, which is estimated in 10th century (when Byzantine sources mention the arrival of first Vlachs in Salonica and other cities).
    By the way, the Aromanian name of Salonica is Săruna ( < lat. Salonae).

    The text Peri Ktismaton/De Aedificiis (written by Procop of Caesarea around 560 AD) mentions some Vulgar Latin toponyms in the Balkans which do not have this phonetic change applied:
    Dorostolon (in Dobrudja), later named Durostor, Dristor, Dârstor in the Middle Ages.

    Also, the vast majority of Slavic loanwords in Romanian do not have the invervocalic -l- > -r- transformation:
    slav. bole > rom. boală
    slav. milo > rom. milă
    slav. sila > rom. silă

    So, this phonetic change was produced probably between 7th - 10th centuries, but we may not be more precise because the Slavic loanwords could have been imported anytime in these centuries.
    At the time the already existing Proto-Romanian language (named also Common Romanian, before the dialectal separation) had a social status less important than the Greek, the official language of Byzantine Empire.
    In these conditions is less probable that Greek imported some phonetic change from Aromanian.

    In fact many phonetic changes in Aromanian could be traced as originating from Greek.
     
    Last edited:

    apmoy70

    Senior Member
    Greek
    ^^Thank you for this, very interesting, but I wonder, if «Βούργαρος» is not influenced by the Aromanian l > r, then how could one explain the Aromanian Virgaru (for Bulgarian) > Thessalian «Βούργαρος»?
     

    danielstan

    Senior Member
    Romanian - Romania
    I repeat myself:
    the phonetic context of Aromanian and Daco-Romanian L > R change is:
    vowel + L + vowel > vowel + R + vowel.
    The phonetic context of«Βούργαρος» is different:
    vowel + L + consonant > vowel + R + consonant
    In Greek there is also a similar change (I did some googleing, I may have spelling errors):
    Albanoi (Ἀλβανοί) > Arvanite (Αρβανίτε)

    In order to assess an Aromanian influence of Virgaru > Βούργαρος
    we need to find a phonetic rule inside Aromanian that would explain Bulgaru > Virgaru (I don't know other Aromanian examples with such change).
    I don't know much Aromanian, I use few online dictionaries when I give examples, but I read Romanian linguistic studies comparing all Romanian dialects and explaining the most important phonetic changes from Vulgar Latin to current days.
    I never heard of such change.

    In order to assess such an influence we need more examples of Greek loans from Aromanian, otherwise is more probable that Greek gave this word to Aromanian.
    The Greek influence on Aromanian (especially its southern dialect) is big in vocabulary and some phonetics.
     
    Last edited:

    Sardokan1.0

    Senior Member
    Sardu / Italianu
    Bulgaru > Virgaru
    In Sardinian language there are words where the original U of Latin has become I, like in the example above, and also the phenomenon of betacism B>V or V>B is a very common thing.

    Latin - Sardinian

    manducare - mandicare, mandigare (to eat)
    umbilicus - ìmbilicu, ìmbiligu (navel)
     

    OBrasilo

    Senior Member
    Brazil, Brazilian Portuguese
    killerbee256 said:
    Portuguese does something similar, if in a slightly different situation:
    V lat. blancus > branco
    lat. obligatus > obrigado
    There is also praça for town square (Spanish plaza), praia for beach (Spanish playa), and prazer for enjoyment (Spanish placer).

    danielstan said:
    If I well remember, Japanese has a regular transformation L > R for loanwords.
    The Japanese sound is somewhere between L and R.
     
    Last edited:

    OED Loves Me Not

    Senior Member
    Japanese - Osaka
    The Japanese sound is somewhere between L and R.
    You're right, I guess. In fact, most Japanese never get to learn to tell L's from R's however hard
    they may work to throughout their lives.

    In that connection, I would like to mention an interesting phonetic phenomenon in the Korean language.
    That is, in standard Korean, they always pronounce an L at the end of a word while they always utter
    an R in the middle.

    For example, Seoul the capital of South Korea is pronounced somewhat
    like English "soul," with the L sound pronounced not much unlike the L in English. But when a suffix
    "-eso" comes at the end of the word Seoul, then it becomes SeouR-eso, not SeouL-eso.
    Why? because the sound which used to be at the end of the word Seoul is now in the middle
    (*Seoul-eso), so the L turns into an R, hence "SeouR-eso."

    At the same time, standard speakers of Korean never utter an R sound at the end of a word. If they have to
    choose between an L and an R at the end of a word, they always utter an L.

    I'm not much versed in Korean, so I hope a Korean native speaker will be kind enough to explain the rest.
     

    OED Loves Me Not

    Senior Member
    Japanese - Osaka
    In various variants of Chinese, I believe that rhotacism has frequently occurred
    throughout their history.

    Where many Southern variants of Chinese utter a Z sound, mandarin Chinese
    utter an R. For example, the set of Chinese characters meaning "Japan" (日本) is pronounced
    somewhat like ziben in southern variants of Chinese as opposed to something like
    riben (meaning the same thing) in mandarin Chinese. This R-Z correspondence is seen
    everywhere between some southern variants of Chinese and mandarin Chinese.

    By the way, the southern variant pronunciation ziben quite resembles Cipangu,
    which word was mentioned to mean Japan in Marco Polo's famous book and is
    again transliterated as Zipangu elsewhere.
     
    Last edited:

    OBrasilo

    Senior Member
    Brazil, Brazilian Portuguese
    Both ziben and riben are ultimately from *ndjit-pon, from where also Japanese Nihon and Nippon (and Korean Ilbon), as well as the southeast asian Jepang and Jepun from where all the western names for Japan, including "Japan" itself.
     
    < Previous | Next >
    Top