Vulpe

Rainbowlight

Senior Member
Spanish
Hello,

I am writing this message because of an interest in how consonants shift from one language to another. Case in point is the Italian word volpe and Latin vulpes. Interestingly, the first consonant in these two words turns into the phoneme g in several European countries. Take French goupil or Galician golpe or golpiño. It seems to clear to me that the v has turned into a g. Is there a systematic rule that might explain this shift?

Thank you very much.
 
  • Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    Whenever you see a Latin /w/ evolve into /g(w)/ or /w/, it's usually the result of the influence of a Germanic cognate with in /w/ or /χw/. See vespam + Frankish waspa > fr. guêpe (not *vêspe), Friulan gjespe (not viespe), Walloon wesse (not vesse); vadum + Germanic wadą > fr. gué, it. guado, wal wé; vastare + germanic *wōstjan > fr. gâter, it. guastare, wal. waster, etc.

    In the case of vulpe > golp-, the culprit is generally thought to be wulf (wolf), but von Wartburg also mentions hwelp (whelp) as a possibility.
     

    Catagrapha

    Member
    Malagasy
    v > g is not often, but not rare, e.g.

    ripa > riva > rigagnolo
    sebum > sevo > sego
    uvula > ugola

    levis > leviarius > legier, leggero
    niv- > nevicare, negier
    serviens > sergent
     

    Rainbowlight

    Senior Member
    Spanish
    Whenever you see a Latin /w/ evolve into /g(w)/ or /w/, it's usually the result of the influence of a Germanic cognate with in /w/ or /χw/. See vespam + Frankish waspa > fr. guêpe (not *vêspe), Friulan gjespe (not viespe), Walloon wesse (not vesse); vadum + Germanic wadą > fr. gué, it. guado, wal wé; vastare + germanic *wōstjan > fr. gâter, it. guastare, wal. waster, etc.

    In the case of vulpe > golp-, the culprit is generally thought to be wulf (wolf), but von Wartburg also mentions hwelp (whelp) as a possibility.
    Thank you!
     

    Welsh_Sion

    Senior Member
    Welsh - Northern
    Initial Latin /w-/ written as <u> or <v>, regularly becomes /g(w)-/ in Cymraeg/Welsh. Not all /gw-/ are of course Latin based.

    Examples:
    uinum > gwin 'wine'
    Uenerem > Gwener 'Friday', 'Venus'
    uesper > gosper, gosber 'vespers'
    vagina > gwain 'sheath'
     
    Last edited:

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    levis > leviarius > legier, leggero
    niv- > nevicare, negier
    serviens > sergent
    Those orthographic "g" are /(d)ʒ/ instead, and don't come from the /w/ but from the following /i/ > /j/. Servientem > VL /sɛr.'βjɛn.te/ > proto-French */sɛrβ.'jɛn.te/ > */sɛrβ.'dʒɛn.te/ > OF /sɛr.'dʒɛnt/

    It's the same mechanism that created sache out of sapiam or rage out of rabiam, V.CjV resyllabifying to VC.dʒV.
     

    Rainbowlight

    Senior Member
    Spanish
    Initial Latin /w-/ written as <u> or <v>, regularly becomes /g(w)-/ in Cymraeg/Welsh. Not all /gw-/ are of course Latin based.

    Examples:
    uinum > gwin 'wine'
    Uenerem > Gwener 'Friday', 'Venus'
    uesper > gosper 'vespers
    vagina > gwain 'sheath'
    Thank you very much.
     

    Cenzontle

    Senior Member
    English, U.S.
    It is odd to think of [v] turning into [g] — what could they have in common?
    But it is not so odd to think of the glide [w] turning into either of them.
    The [w] sound, like the vowel [ u ], entails a narrowing of the speech channel at two places:
    the raising of the back of the tongue and the rounding of the lips.
    Take either of these two gestures to its extreme (making it a consonant), while eliminating the other,
    and you get [ɣ] or [g] on one hand, [ᵬ]* or [v] on the other.
    ---------------
    *Beta.
     

    Welsh_Sion

    Senior Member
    Welsh - Northern
    [ɣ] fell out of use in early in Middle Welsh - it was only weakly pronounced anyway.

    [g], and [v] (<f>) of course still exist.

    Initially <g> disappears in soft mutation, often leaving a glide, [w], or a lateral, [l] behind. Or in some cases <w + l>.

    gwan 'weak' > yn wan 'weakly'
    gwlad 'world' > dy wlad 'thy world'

    Initial <f> is rare and is often the result of soft mutation of .

    dy fys 'thy finger' < bys 'finger'
     

    Rainbowlight

    Senior Member
    Spanish
    It is odd to think of [v] turning into [g] — what could they have in common?
    But it is not so odd to think of the glide [w] turning into either of them.
    The [w] sound, like the vowel [ u ], entails a narrowing of the speech channel at two places:
    the raising of the back of the tongue and the rounding of the lips.
    Take either of these two gestures to its extreme (making it a consonant), while eliminating the other,
    and you get [ɣ] or [g] on one hand, [ᵬ]* or [v] on the other.
    ---------------
    *Beta.
    I myself think that we have taken for granted the very pronunciations of the characters in the Latin alphabet. That is, I do not know how these words were pronounced in, say, classical Latin. Nor have I encountered a systematic approach to Latin pronunciation that was written back then. Then again, maybe there is a phonetic guide written during those times and I have never heard of it. Suffice it to say, vulpe is indeed still pronounced as goupil in French and golpe in Galician and that is enough reason for me to ponder some things that we seem to take for granted. I certainly can't find a natural relationship between the letters [v] and [g]. By the way, and [v] are not differentiated in current Spanish: these two different characters share the same sound. Thoughts, anyone?
     

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    Think of vulpes in classical Latin as pronounced something like "huolpes" would be in Spanish (not quite that, because the stressed vowel was intermediate between Spanish /u/ and /o/ and the e was much longer, but it's a good approximation of the v). We write it "v" because that's what the sound became in most of Latin's descendants.

    This initial /w/ sound, be it in Latin vulpes /ˈwul.peːs/, in Spanish huevo /webo/, in English world /wɜːld/ or in French oui /wi/, was "labio-velar". That is, it's pronounced with a constriction at two points in the mouth: at the velum (the soft palate), like k, g or the jota and at the lips, like m, p or b (that's the labio- part).

    In the languages where it became /v/ (or b, as in Spanish), the labial part won out. In those where it went to /g/, it's the velar part who won.

    French is a part of the first group, where /w/ became labial (well labio-dental, but close enough) /v/. But because speakers of Gaulish Romance still had /gʷ/ (another labio-velar sound!) in words like lingua or agua (< aqua), they could use that sound when borrowing Germanic words that started with /w/, like werra > guerra > guerre, or wardian > guarder. They also extended that pronunciation to words like volpe (< vulpem) that kinda sounded like the Germanic words wulf and hwelp and referred to similar animals. Thus volpe became *gwolpe (and vulpeculum became *volpeclo then *gwolpeilo then eventually goupil)

    This whole w > g thing really shouldn't surprise you, since there's American Spanish varieties that pronounces words like huevo as güevo, which is the first part of such a shift.
     

    Rainbowlight

    Senior Member
    Spanish
    Think of vulpes in classical Latin as pronounced something like "huolpes" would be in Spanish (not quite that, because the stressed vowel was intermediate between Spanish /u/ and /o/ and the e was much longer, but it's a good approximation of the v). We write it "v" because that's what the sound became in most of Latin's descendants.

    This initial /w/ sound, be it in Latin vulpes /ˈwul.peːs/, in Spanish huevo /webo/, in English world /wɜːld/ or in French oui /wi/, was "labio-velar". That is, it's pronounced with a constriction at two points in the mouth: at the velum (the soft palate), like k, g or the jota and at the lips, like m, p or b (that's the labio- part).

    In the languages where it became /v/ (or b, as in Spanish), the labial part won out. In those where it went to /g/, it's the velar part who won.

    French is a part of the first group, where /w/ became labial (well labio-dental, but close enough) /v/. But because speakers of Gaulish Romance still had /gʷ/ (another labio-velar sound!) in words like lingua or agua (< aqua), they could use that sound when borrowing Germanic words that started with /w/, like werra > guerra > guerre, or wardian > guarder. They also extended that pronunciation to words like volpe (< vulpem) that kinda sounded like the Germanic words wulf and hwelp and referred to similar animals. Thus volpe became *gwolpe (and vulpeculum became *volpeclo then *gwolpeilo then eventually goupil)

    This whole w > g thing really shouldn't surprise you, since there's American Spanish varieties that pronounces words like huevo as güevo, which is the first part of such a shift.
    Thank you very much for the very useful and detailed explanation. I actually studied English Philology for some time and had a very hard time understanding these matters. I am still trying to catch up, though. :)

    What I was trying to convey in my earlier message is that there is not an actual record of how these letters were actually voiced in Latin, as we do not have a physical, audible embodiment of that language, its varieties and its sounds. As for the vulpe case, some Italian dialects actually retain that /g/ sound and pronounce the word as golpa, gulp and even gurpi, a fact that makes me think about the original sound of these characters and how they were originally voiced. Indeed, huevo is sometimes pronounced as güevo in Spanish, but what really piques my interest is the inner logic of the rule that establishes a link between two clearly differentiated sounds such as the r > l shift (think of Spanish amor versus amol in some Latin American countries) or the t > l shift (think of Spanish lengua, Italian lingua or French langue versus English tongue), a phenomenon that has always puzzled me.
     

    danielstan

    Senior Member
    Romanian - Romania
    but what really piques my interest is the inner logic of the rule that establishes a link between two clearly differentiated sounds such as the r > l shift ... or the t > l shift ..., a phenomenon that has always puzzled me.
    I guess you assume such sound shifts are happening inside a generation of speakers. A possible scenario:
    - an individual in his adult age starts one day to pronounce 'amol' instead of 'amor', then his neighbours are imitating his speech and after 100 years an entire region in a country does the same

    In fact most of the sound shifts are happening from one generation to another and the "initiators" are the children in their early years when they are reproducing the parent speech with imperfections.
    For the r > l shift there are examples of children who cannot pronounce the "r" up to the age of 10, replacing it by "l".
    In a forced analogy, the process of passing a language from one generation to another is similar to copying a photo in a Xerox machine, one copy from the one before, and comparing at the end the original with the 100th copy.
     

    Rainbowlight

    Senior Member
    Spanish
    I guess you assume such sound shifts are happening inside a generation of speakers. A possible scenario:
    - an individual in his adult age starts one day to pronounce 'amol' instead of 'amor', then his neighbours are imitating his speech and after 100 years an entire region in a country does the same

    In fact most of the sound shifts are happening from one generation to another and the "initiators" are the children in their early years when they are reproducing the parent speech with imperfections.
    For the r > l shift there are examples of children who cannot pronounce the "r" up to the age of 10, replacing it by "l".
    In a forced analogy, the process of passing a language from one generation to another is similar to copying a photo in a Xerox machine, one copy from the one before, and comparing at the end the original with the 100th copy.
    Thank you very much your insightful commentary. I wholeheartedly agree with some of your views, but my point is that some of these shift patterns sometimes bear no logic at all. I can understand the p sound flowing into a b sound, as they are both bilabial. Or take /d/ and /t/, which again are extremely similar sounds.

    What puzzles me are the shifts between sounds that sometimes are radically different. Could road, say, may give way to load? Did French article le turn into English the for no apparent reason? (A piece of interesting trivia: French pronounce approximately the article the as "tse" or "ze")Why did English stick with the and not the l-based articles used in many European countries, such as French le or even Spanish el or Italian il? Was this article in a very distant actually pronounced as le in the UK?
     

    Penyafort

    Senior Member
    Catalan (Catalonia), Spanish (Spain)
    In Aragonese, confusion between b and g is common in some words (IUGU > chugo/chubo 'yoke', FAGU > fago, fabo, fau 'birch', MUGA > buaga/güaga 'border', etc.)

    Rabosa ended up being the word for 'fox', but golpella (< VULPECULA) also existed in Old Aragonese.
     

    Swatters

    Senior Member
    French - Belgium, some Wallo-Picard
    What puzzles me are the shifts between sounds that sometimes are radically different. Could road, say, may give way to load? Did French article le turn into English the for no apparent reason? (A piece of interesting trivia: French pronounce approximately the article the as "tse" or "ze")Why did English stick with the and not the l-based articles used in many European countries, such as French le or even Spanish el or Italian il? Was this article in a very distant actually pronounced as le in the UK?
    The Germanic and Romance definite articles aren't directly related! In both cases, they come from demonstratives, but not the same one. The Germanic the/de/etc comes from the same demonstrative root that gave Spanish isto, while the Romance article (bar Sardinian's) comes from a root that designated far away or foreign things and eventually became a demonstrative in Latin (but also alter > otro and ultra). It doesn't have a real counterpart in Germanic, but the same root eventually became English all.
    what really piques my interest is the inner logic of the rule that establishes a link between two clearly differentiated sounds such as the r > l shift (think of Spanish amor versus amol in some Latin American countries) or the t > l shift (think of Spanish lengua, Italian lingua or French langue versus English tongue), a phenomenon that has always puzzled me.
    /r/ and /l/ are close enough to each other that many languages don't distinguish them at all.

    But if you want a biomechanical explanation, notice first that /t/, /d/, /n/, both Spanish r (/r/ and /ɾ/), /l/, /s/ and /z/ are all pronounced at the same point in the mouth: with the tip of the tongue touching the flat area behind the teeth (depending on language, it can be right behind the teeth or a bit more back toward the alveolae, but roughly speaking all of those sounds can be articulated with the tongue in the exact same position)

    So just like /b/, /m/ and /v/ can rather easily shift from one to another, so can all those (denti-)alveolar consonants.

    Let's start from /d/. It's produced by blocking your airway with the tongue, letting pressure build up then releasing the air flow, which creates a burst. Now make the same tongue gesture that produces /d/ but flick your tongue back down immediately once it touches your palate without letting any pressure build up and thus no release burst. You've pronounced /ɾ/, the single Spanish r, as in caro. This sound change is happening right now in American English, where pudding can sound like pu/ɾ/ing (that's why Japanese borrowed that word as /puriN/).

    Starting from /ɾ/, if you get your tongue in that same position but hold it for longer while letting air stream around the sides of the tongue, you get a /l/. If you let air stream between the tongue and the palate, it's going to make a lot of noise as it goes through that thin space then hits your teeth, which we hear as /s/ or /z/. Do the same thing with just the right amount of air pressure and tension in your tongue muscle, so that instead the tip of your tongue starts closing the airway then opening it again in a rhythmic pattern, and you get /r/ (it's a complicated gesture, that's why kids have trouble with it).

    /n/ is like a /d/ where instead of letting air build up in your mouth, you open the passage to your nose so that the air exits through your nostrils.

    And /t/ is just a /d/ pronounced with the vocal folds not vibrating until you release the occlusion and produce the airburst (or even later, that depends on the language).

    In other words, any of those sounds can shift to some of the other with just one or two changes in the way you produce them, and so they do, quite often.

    As for lingua and tongue, you're right that they're related and that the Latin word should have been dingua. The going theory is that because the verb lingare (meaning lick, and in fact sharing a root with that English verb) existed, and had a related meaning to dingua, it caused some confusion in the mind of the speakers of Latin, who ended up shifting dingua to lingua to match the verb.
     

    Rainbowlight

    Senior Member
    Spanish
    Whenever you see a Latin /w/ evolve into /g(w)/ or /w/, it's usually the result of the influence of a Germanic cognate with in /w/ or /χw/. See vespam + Frankish waspa > fr. guêpe (not *vêspe), Friulan gjespe (not viespe), Walloon wesse (not vesse); vadum + Germanic wadą > fr. gué, it. guado, wal wé; vastare + germanic *wōstjan > fr. gâter, it. guastare, wal. waster, etc.

    In the case of vulpe > golp-, the culprit is generally thought to be wulf (wolf), but von Wartburg also mentions hwelp (whelp) as a possibility.
    Thank you very much!
     

    Rainbowlight

    Senior Member
    Spanish
    It is odd to think of [v] turning into [g] — what could they have in common?
    But it is not so odd to think of the glide [w] turning into either of them.
    The [w] sound, like the vowel [ u ], entails a narrowing of the speech channel at two places:
    the raising of the back of the tongue and the rounding of the lips.
    Take either of these two gestures to its extreme (making it a consonant), while eliminating the other,
    and you get [ɣ] or [g] on one hand, [ᵬ]* or [v] on the other.
    ---------------
    *Beta.
    Thank you very much.
     

    Rainbowlight

    Senior Member
    Spanish
    [ɣ] fell out of use in early in Middle Welsh - it was only weakly pronounced anyway.

    [g], and [v] (<f>) of course still exist.

    Initially <g> disappears in soft mutation, often leaving a glide, [w], or a lateral, [l] behind. Or in some cases <w + l>.

    gwan 'weak' > yn wan 'weakly'
    gwlad 'world' > dy wlad 'thy world'

    Initial <f> is rare and is often the result of soft mutation of .

    dy fys 'thy finger' < bys 'finger'
    Thank you very much.
     

    Rainbowlight

    Senior Member
    Spanish
    Those orthographic "g" are /(d)ʒ/ instead, and don't come from the /w/ but from the following /i/ > /j/. Servientem > VL /sɛr.'βjɛn.te/ > proto-French */sɛrβ.'jɛn.te/ > */sɛrβ.'dʒɛn.te/ > OF /sɛr.'dʒɛnt/

    It's the same mechanism that created sache out of sapiam or rage out of rabiam, V.CjV resyllabifying to VC.dʒV.
    Thank you very much!
     

    Rainbowlight

    Senior Member
    Spanish
    Think of vulpes in classical Latin as pronounced something like "huolpes" would be in Spanish (not quite that, because the stressed vowel was intermediate between Spanish /u/ and /o/ and the e was much longer, but it's a good approximation of the v). We write it "v" because that's what the sound became in most of Latin's descendants.

    This initial /w/ sound, be it in Latin vulpes /ˈwul.peːs/, in Spanish huevo /webo/, in English world /wɜːld/ or in French oui /wi/, was "labio-velar". That is, it's pronounced with a constriction at two points in the mouth: at the velum (the soft palate), like k, g or the jota and at the lips, like m, p or b (that's the labio- part).

    In the languages where it became /v/ (or b, as in Spanish), the labial part won out. In those where it went to /g/, it's the velar part who won.

    French is a part of the first group, where /w/ became labial (well labio-dental, but close enough) /v/. But because speakers of Gaulish Romance still had /gʷ/ (another labio-velar sound!) in words like lingua or agua (< aqua), they could use that sound when borrowing Germanic words that started with /w/, like werra > guerra > guerre, or wardian > guarder. They also extended that pronunciation to words like volpe (< vulpem) that kinda sounded like the Germanic words wulf and hwelp and referred to similar animals. Thus volpe became *gwolpe (and vulpeculum became *volpeclo then *gwolpeilo then eventually goupil)

    This whole w > g thing really shouldn't surprise you, since there's American Spanish varieties that pronounces words like huevo as güevo, which is the first part of such a shift.
    Thank you very much.
     

    Rainbowlight

    Senior Member
    Spanish
    In Aragonese, confusion between b and g is common in some words (IUGU > chugo/chubo 'yoke', FAGU > fago, fabo, fau 'birch', MUGA > buaga/güaga 'border', etc.)

    Rabosa ended up being the word for 'fox', but golpella (< VULPECULA) also existed in Old Aragonese.
    Thank you very much!
     

    Rainbowlight

    Senior Member
    Spanish
    The Germanic and Romance definite articles aren't directly related! In both cases, they come from demonstratives, but not the same one. The Germanic the/de/etc comes from the same demonstrative root that gave Spanish isto, while the Romance article (bar Sardinian's) comes from a root that designated far away or foreign things and eventually became a demonstrative in Latin (but also alter > otro and ultra). It doesn't have a real counterpart in Germanic, but the same root eventually became English all.

    /r/ and /l/ are close enough to each other that many languages don't distinguish them at all.

    But if you want a biomechanical explanation, notice first that /t/, /d/, /n/, both Spanish r (/r/ and /ɾ/), /l/, /s/ and /z/ are all pronounced at the same point in the mouth: with the tip of the tongue touching the flat area behind the teeth (depending on language, it can be right behind the teeth or a bit more back toward the alveolae, but roughly speaking all of those sounds can be articulated with the tongue in the exact same position)

    So just like /b/, /m/ and /v/ can rather easily shift from one to another, so can all those (denti-)alveolar consonants.

    Let's start from /d/. It's produced by blocking your airway with the tongue, letting pressure build up then releasing the air flow, which creates a burst. Now make the same tongue gesture that produces /d/ but flick your tongue back down immediately once it touches your palate without letting any pressure build up and thus no release burst. You've pronounced /ɾ/, the single Spanish r, as in caro. This sound change is happening right now in American English, where pudding can sound like pu/ɾ/ing (that's why Japanese borrowed that word as /puriN/).

    Starting from /ɾ/, if you get your tongue in that same position but hold it for longer while letting air stream around the sides of the tongue, you get a /l/. If you let air stream between the tongue and the palate, it's going to make a lot of noise as it goes through that thin space then hits your teeth, which we hear as /s/ or /z/. Do the same thing with just the right amount of air pressure and tension in your tongue muscle, so that instead the tip of your tongue starts closing the airway then opening it again in a rhythmic pattern, and you get /r/ (it's a complicated gesture, that's why kids have trouble with it).

    /n/ is like a /d/ where instead of letting air build up in your mouth, you open the passage to your nose so that the air exits through your nostrils.

    And /t/ is just a /d/ pronounced with the vocal folds not vibrating until you release the occlusion and produce the airburst (or even later, that depends on the language).

    In other words, any of those sounds can shift to some of the other with just one or two changes in the way you produce them, and so they do, quite often.

    As for lingua and tongue, you're right that they're related and that the Latin word should have been dingua. The going theory is that because the verb lingare (meaning lick, and in fact sharing a root with that English verb) existed, and had a related meaning to dingua, it caused some confusion in the mind of the speakers of Latin, who ended up shifting dingua to lingua to match the verb.
    Thanks for the very useful and detailed explanation.
     

    Delvo

    Senior Member
    American English
    What puzzles me are the shifts between sounds that sometimes are radically different. Could road, say, may give way to load?
    L and R (at least the English glide R, not necessarily the Spanish tap R) only seem radically different because you're used to languages in which that difference matters a lot. But physically, they're made in very similar ways. Both are voiced, neither completely blocks the flow of air through the mouth, and both involve moving the sides of the tongue up closer to, or touching, the gums. (It is also possible to make another version of the R-sound by curling the front of the tongue back instead, but most Englishers don't, and it would sound different if they did, more like the retroflex sounds of Indic languages.) The only difference is whether the tongue is slightly higher & flatter (L) or slightly lower & more deeply curled up at the sides (R).

    For all kinds of sound shifts in general, it's helpful to remember that people sometimes mispronounce things, or different dialects/accents in one language can have different versions which are already part of the way to becoming something that would eventually seem completely distinct. And a sound shift is simply a mispronunciation that became so common that it became normal. So pay attention to how people mispronounce things, or how people speaking your language but from another country pronounce things differently from what you're used to, and you'll get a bunch of examples of how sound shifts could begin. Even the mispronunciations or dialectical differences that haven't become standard sound shifts yet are still examples of how sound shifts can begin.

    For example, when I listen to Scottish people speaking English, I often hear their Rs after vowels as either Ls or something somewhere between an R and an L, or maybe a sequence consisting of first R then L. Sometimes I get that with people from some parts of England, too; Benedict Cumberbatch has tricked me into hearing his "force" as "false" more than once. It's a very easy pair of sounds to either mix up or merge or circumstantially switch. I've never met an American who would do it except rarely by accident, but accidents do happen. For example, I've also heard /k/ come out as /either /k͡x/ or just /x/ in English, which reflects a sound shift that has really happened in some languages, but English isn't usually thought of as even having /x/ at all. That's the beginning of a fairly common type of sound shift, which might eventually catch on and become standard, but hasn't yet and might not ever do so at all.

    The relationship between "w" and "ɡ" (when they aren't both descendants of an earlier /ɡʷ/ or such) isn't quite so simple because it probably involves a coarticulation of both "w" and some kind of velar. But once you recognize coarticulations (including accidental unintended ones) as a real thing, it's fairly straight forward, with only two steps in it. And again, accidental mispronunciations I've heard in English demonstrate that this kind of coarticulation can sprout up spontaneously. For example, I've heard the name "Chris Wallace" pronounced as not /krɪs-wαlɜs/* but /krɪs-ɣʷαlɜs/. Why would that happen? The sound of /w/ in English comes from what the speaker does with his/her lips, so it doesn't matter whether the middle/back of the tongue is slightly raised or not, so, at least sometimes for some people, it is, whether they think about it or not. It's not standard or universal, but it's also essentially inaudible when it does happen, so there's no restriction against doing it either. And then it only takes a slight slip, just putting the tongue another tiny bit higher than intended, for it to get close enough to start adding a /ɣ/ (or possibly /ɡ/) to the original /w/*. So if enough people did that often enough, /ɣʷ/ or /ɡʷ/ would become the new standard. From there, it's only one more step to /ɡ/: just dropping the /w/* part.

    *In fact, if you look up "/w/" on Wikipedia, at least in English, you even get a strange dual definition: that it represents a labiovelar (coarticulation) in languages that have such a thing, or just the bilabial alone in languages without a distinct labiovelar. But...
    • having two definitions for one symbol violates the IPA's principle of one symbol per sound and one sound per symbol, and
    • using different definitions in different languages violates the IPA's principle of each symbol's sound being defined universally for all languages rather than language-dependent,
    ...so only one of these two definitions for /w/ can really be valid as far as I'm concerned. And it clearly has to be the just-bilabial one, because...
    • the labiovelar one also violates the IPA's principle of one symbol per sound and one sound per symbol with coarticulations being represented with superscripts & other such modifiers rather than having any single symbol represent both parts of a coarticulation, and
    • otherwise every use of the symbol "/w/" in languages like English would be claiming there's an extra articulation in there as standard pronunciation, which there isn't, and
    • with /w/ meaning "labiovelar", accurately representing the plain bilabial glide would then be impossible because the IPA has no way to indicate the subtraction of one part of a coarticulation represented by a single symbol (because the IPA doesn't have single symbols for coarticulations in the first place and this problem is one of the reasons why).
    Still, the fact that the two definitions for "/w/" have been asserted at all does demonstrate the point that that labiovelar coarticulation is a real thing and can easily get mixed up with its just-bilabial counterpart, which is all it takes for a sound shift to occur.
     
    Last edited:

    pollohispanizado

    Senior Member
    Inglés canadiense
    in Spanish huevo /webo/
    And huevo ends up getting pronounced güevo /ɣʷebo/ in certain dialects or circumstances. If emphasis is put on the word (like it is often, for example, in the expresión "hacer algo alguien por sus huevos") the initial /ɣʷ/ can actually drag on quite long, becoming the most salient sound in a word in which it is not even a "real" phoneme. Also, some words have two spellings: one with hue- and one with güe-.
    In Aragonese, confusion between b and g is common in some words (IUGU > chugo/chubo 'yoke', FAGU > fago, fabo, fau 'birch', MUGA > buaga/güaga 'border', etc.)
    Sort of to this point: güeña (a type of chorizo) comes from the older boheña. This is the word used in Aragon for chorizo de sábado.
     
    Last edited:

    pollohispanizado

    Senior Member
    Inglés canadiense
    Se me había olvidado... There's also the case of the extremely used Mexican vocative güey/wey, which is pronounced /ɣʷej/~/ˠwej/~/wej/ in free variation, depending on many things including stress, intonation and personal choice. (It is said that güey derives from buey, showing the b > g shift, although apparently it could have come from a Náhuatl word huey that means "honorable/venerable". Of course, maybe both are true...)
     
    Last edited:
    Top