Linguistic Miscellany Thread

[Richard W]

Should those be "fricative, affricate, affricate", then?

Affricates are stops.

[Travis B.]

Should those be "fricative, affricate, affricate", then?

Affricates are stops.

Affricates are plosives, as are stops, but affricates and stops are not one and the same.

[bradrn]

Should those be "fricative, affricate, affricate", then?

Affricates are stops.

Affricates are plosives, as are stops, but affricates and stops are not one and the same.

I’d say it’s the other way around: affricates and plosives are both stops, with affricates and plosives being different things.

[Travis B.]

Affricates are stops.

Affricates are plosives, as are stops, but affricates and stops are not one and the same.

I’d say it’s the other way around: affricates and plosives are both stops, with affricates and plosives being different things.

I checked the Wiki and it's very confusing. It seems to indicate that affricates are not plosives, and may or may not be stops depending on whom you ask, but if one really wants a term to cover plosives and affricates, there is the term occlusive, which is unfortunate because that term also includes nasals...

[bradrn]

Affricates are plosives, as are stops, but affricates and stops are not one and the same.

I’d say it’s the other way around: affricates and plosives are both stops, with affricates and plosives being different things.

I checked the Wiki and it's very confusing. It seems to indicate that affricates are not plosives, and may or may not be stops depending on whom you ask, but if one really wants a term to cover plosives and affricates, there is the term occlusive, which is unfortunate because that term also includes nasals...

The Wiki is generally confusing. I’d say that affricates and plosives are both stops (as are ejectives and implosives), and stops and nasals are both occlusives — but it really does depend on who you ask. Ladefoged and Maddieson (in The Sounds of the World’s Languages, 1996) seem to agree with me, but it’s not so uncommon to see ‘stops’ and ‘affricates’ distinguished in reference grammars.

[Rounin Ryuuji]

I had always thought affricates were coarticulations (is that the right word?) involving a stop followed by a fricative.

[Travis B.]

I had always thought affricates were coarticulations (is that the right word?) involving a stop followed by a fricative.

I always understood coarticulations as phones with multiple simultaneous articulations at different POA, such as the [kp] and [gb] found in many Niger-Congo languages.

[bradrn]

I had always thought affricates were coarticulations (is that the right word?) involving a stop followed by a fricative.

I always understood coarticulations as phones with multiple simultaneous articulations at different POA, such as the [kp] and [gb] found in many Niger-Congo languages.

This is my understanding too. An affricate is not coarticulated; neither is it a plosive followed by a fricative. An affricate is a plosive with a fricated release.

[Richard W]

Another, slightly different, example of the of the affricate with the later voicing onset time (VOT) becoming a fricative can be found in Lao after the great consonant shift, which has /cʰ/ > /s/, while /c/ remains. The written evidence is that the reflexes of Proto-SWT *z and *ʝ are now written the same, with back-up from Pali syllable-initial <s>, <j> and <jh> differinɡ only in tone. The same shift seems to have happened in Northern Thai and Tai Lue, but their only native input was the slightly suspect Proto-SW Tai *cʰ.

It's a bit confusing, because the the principal source of this /cʰ/ in Lao was Proto-SWT *ʝ. The Siamese-Lao area had a VOT flip-flop on non-aspirates, believed to be a chain of feature transfers:

1. initial voicing > breathy voice on vowel
2. tone split in tone languages
3. breathy voice on vowel > aspiration on initial stop

This wrinkle on the Great Consonant Shift also affected at least one Northern Tai language (Saek), at least one Central Tai language, and at least one, non-tonal, Monic language (Nyah Kur).

The old and new aspirates have merged non-segmentally, but a distinction is carried by tone (except where a tone merger has occurred) in the Tai languages and register (Mon-Khmer language(s)).

[anteallach]

I had always thought affricates were coarticulations (is that the right word?) involving a stop followed by a fricative.

I always understood coarticulations as phones with multiple simultaneous articulations at different POA, such as the [kp] and [gb] found in many Niger-Congo languages.

This is my understanding too. An affricate is not coarticulated; neither is it a plosive followed by a fricative. An affricate is a plosive with a fricated release.

I think "stop with fricative release" is closest to my understanding. But there are some grey areas, like the Standard German [pf], where the closure is apparently normally bilabial but the fricative is labiodental. (Ladefoged and Maddieson describe what happens in some detail.)

For many languages it's reasonable to think of there being a single "stop" class which contains both affricates and ordinary plosives, and Sanskrit is presumably like this (if the sounds in question were actually affricates at all).

[Ephraim]

I always understood coarticulations as phones with multiple simultaneous articulations at different POA, such as the [kp] and [gb] found in many Niger-Congo languages.

I think the term coarticulation is used in multiple, but related, ways. It can refer to the doubly articulated consonants you mentioned, or to secondary articulation (labialization, palatalization etc.), or as a cover term for both.
https://en.wikipedia.org/wiki/Co-articulated_consonant

Coarticulation can also refer to overlapping articulation in continuous speech, i.e. essentially allophonic secondary articulation such as consonants being articulated with rounded lips next to rounded vowels. I think this may actually be the most common use of the term.

Wikipedia also suggests that it can refer to assimilation in a broader sense.
https://en.wikipedia.org/wiki/Coarticulation

I had always thought affricates were coarticulations (is that the right word?) involving a stop followed by a fricative.

Maybe the term you were thinking of was countour segments (or just contours). This includes things like diphthongs and prenasalized plosives, i.e. single phonemic segments that include some type of transition.
https://en.wikipedia.org/wiki/Contour_(linguistics)

Affricates are contour segments but they are typically not thought of as coarticulated segments. Labialized and palatalized segments may involve both coarticulation and a contour, since they typically have a noticeable on- or offglide.

[zompist]

Fun fact: in Kumzari, a SW Iranian language, دُمب [dumb] means 'stupid person'. Not cognate with English 'dumb', but cognate with the Farsi word for 'tail'.

(From a linguist on Twitter.)

[Travis B.]

Does anyone know how Spanish, Portuguese, and Galician gato, Asturian gatu, Catalan and Occitan gat (but also Occitan cat), ended up with initial /g/?

[Estav]

Does anyone know how Spanish, Portuguese, and Galician gato, Asturian gatu, Catalan and Occitan gat (but also Occitan cat), ended up with initial /g/?

Voicing of initial /k/ to /g/ is fairly common in Iberian Romance (or in some cases Western Romance more generally): compare golpe and graso. I don’t know why /k/ in particular and not /p/ and /t/ was affected by this sporadic process.

[Linguoboy]

Does anyone know how Spanish, Portuguese, and Galician gato, Asturian gatu, Catalan and Occitan gat (but also Occitan cat), ended up with initial /g/?

Sporadic /k/ > /g/ in initial position just seems to be an area feature. See also (all examples from Catalan):

• crassus > gras (though Italian grasso points to a general Vulgar Latin *grassus)
• cumulus > *culumus > gom
• cavea > gàbia
• *caveola > Old Catalan cajola > garjola (perhaps influenced by gàbia)
• calix > calze/galze
• *carulia > garolla/garota

[Richard W]

Voicing of initial /k/ to /g/ is fairly common in Iberian Romance (or in some cases Western Romance more generally): compare golpe and graso. I don’t know why /k/ in particular and not /p/ and /t/ was affected by this sporadic process.

The reason is that velar contact is further back in the mouth. The pressure difference due to voicing is less, and I think It's also easier for timing errors to change the perceived voicing. It's apparently also quite common in Oceanic. Unstable voicing also noticeably shows up in English surnames, such as Tyson and Dyson.

[Travis B.]

Here is a recording of two pronunciations of final /r/, both in the word four, in my idiolect. The first one to me is an ordinary clear uvular approximant with a bit of pharyngealization. A "bunched /r/" as they call it. However, I'm not sure what the second is. Unlike the first, it is almost entirely realized back in the throat, without uvular realization or any involvement of the tongue (even in the form of pharyngealization). It seems to be a trill of some sort, but what kind of trill would be pronounced without the tongue? A voiced epiglottal trill? This is not me being a snowflake, either, because I've heard my mother pronounce final /r/ the very same way on multiple occasions, and it is very hard to mistake this particular realization of /r/.

[Travis B.]

I remember a bit of discussion between myself and someone else here about yod-coalescence in English and how do you IMD is resistant to it (i.e. I very often have /dju/ [tjy(ː)] or /djə/ [tjə(ː)] for it). I just remembered another case of yod-coalescence resistance IMD, which is in just so you, which I commonly realize as /ˈdʒʌssjə/ [ˈtʃʌsːjə(ː)].

[anteallach]

Here is a recording of two pronunciations of final /r/, both in the word four, in my idiolect. The first one to me is an ordinary clear uvular approximant with a bit of pharyngealization. A "bunched /r/" as they call it. However, I'm not sure what the second is. Unlike the first, it is almost entirely realized back in the throat, without uvular realization or any involvement of the tongue (even in the form of pharyngealization). It seems to be a trill of some sort, but what kind of trill would be pronounced without the tongue? A voiced epiglottal trill? This is not me being a snowflake, either, because I've heard my mother pronounce final /r/ the very same way on multiple occasions, and it is very hard to mistake this particular realization of /r/.

It's hard to tell from a recording, but attempting to mimic the second one does feel like I'm doing something in the pharyngeal area.

I don't believe that the typical English "bunched r" is accurately described as an uvular approximant, even pharyngealised, but we've been through this before.

[bradrn]

Here is a recording of two pronunciations of final /r/, both in the word four, in my idiolect. The first one to me is an ordinary clear uvular approximant with a bit of pharyngealization. A "bunched /r/" as they call it. However, I'm not sure what the second is. Unlike the first, it is almost entirely realized back in the throat, without uvular realization or any involvement of the tongue (even in the form of pharyngealization). It seems to be a trill of some sort, but what kind of trill would be pronounced without the tongue? A voiced epiglottal trill? This is not me being a snowflake, either, because I've heard my mother pronounce final /r/ the very same way on multiple occasions, and it is very hard to mistake this particular realization of /r/.

Not quite sure, but it sounds like it could well just be plain [ʕ̞]. It doesn’t sound like any kind of trill to me (especially since a trill involves very considerable movement of the active articulator, which is hard to mistake even when epiglottal).