Linguistic Miscellany Thread

[Travis B.]

Okay, to break it down, while I have the same standard set of English sibilant phonemes as everyone else (i.e. /s z ʃ ʒ tʃ dʒ/), as I have mentioned on here they are complicated by that each of them has front and back allophones based on adjacent consonants and vowels in a rather complex fashion that I am not going to attempt to detail here.

Also, from a superficial analysis one might come to the conclusion that these front and back realizations are actually separate phonemes (as a consequence of that /st/ outside of final position can be realized as [sʲː] and /ʃtʃ/ can be realized as [ɕː], and in quick speech these are sometimes shortened to [sʲ] and [ɕ]), but I have come to the conclusion that such an analysis is mistaken.

As I mentioned above, /s/ has a front realization that is laminal dentialveolar and a back realization that is palatalized laminal alveolar.

/z/ has a front realization that is laminal alveolar and a back realization that is palatalized laminal alveolar.

/ʃ ʒ tʃ dʒ/ all have front realizations that are laminal palatoalveolar and back realizations that are laminal alveolopalatal.

Note that all of the lenis sibilants have voiceless allophones, and /dʒ/ is always voiceless except when deliberately overpronounced (and yes, minimal pairs with /tʃ/ based on aspiration or preceding vowel length alone can be formed).

I don't mean to belittle you by saying this, but do you think you could give audio recordings of your speech? Particularly longer passages where you are less likely to be able to speak carefully. I suppose some of the details you normally give are common to phonetic transcriptions of (Great Lakes) American English, but a lot of the time i'm confused and wonder what is going on.

I'll record my speech later after I get home from work. About my transcriptions, many of the features in them are frequently found in Inland North dialects in general from my experience but are not familiar to many who are not intimately acquainted with the details of NAE dialectal phonology and phonetics. My speech really is less odd than one may think. (Most people only really know NAE phonology and phonetics from the perspective of an idealized GA and are liable to attempt to make their own transcriptions fit this.)

[Starbeam]

Living in Chicago for almost three years now, i've gotten familiar with the Great Lakes (Inland North) accent to the point i catch myself imitating some features without realizing it. That said, i do not know much about transcriptions beyond what Wikipedia and Labov have given.

[Travis B.]

Living in Chicago for almost three years now, i've gotten familiar with the Great Lakes (Inland North) accent to the point i catch myself imitating some features without realizing it. That said, i do not know much about transcriptions beyond what Wikipedia and Labov have given.

Key differences between conventional GA-type transcriptions and my transcriptions are:

• I mark the NCVS.
• I mark allophonic vowel length.
• I mark phonemic consonant length and reduction of consonant clusters to geminates.
• I mark allophonic fortis plosive aspiration (note that conventional GA-type transcriptions sometimes do this but sometimes don't).
• I mark coda fortis plosive (pre)glottalization.
• I mark allophonic obstruent devoicing.
• I mark /l/-vocalization and dorsal realizations of /l/.
• I mark dorsal realizations of /r/.
• I mark American/Canadian raising (my own speech has elements of both).
• I mark front/back allophony of GOOSE, FOOT, and GOAT.
• I mark the weak vowel merger and the resulting distribution of allophones.
• I mark vowel nasality (but note that while I actually have both weakly and strongly nasalized vowels, I normally do not distinguish these in my transcriptions).
• I mark post-tonic intervocalic elision of /t d n nt/, often /nd/, and inconsistently /b v ð/.
• I mark resolution of many intervocalic elisions in resulting diphthongs.
• I mark homorganic coda nasal elision before fortis plosive phonemes in the same coda.
• I mark other frequent consonant elisions such as initial elision of /t/ in to.
• I mark allophonic consonant labialization before GOOSE, FOOT, and /w/.
• I mark front/back allophones of sibilants and other palatalization of coronals.
• I mark assimilation of initial /ð/ across word boundaries.
• I mark affrication of initial /θ/ and /ð/ and stopping of initial /ð/.

While there are definite differences between GA and my own speech, in many cases differences between my transcriptions and others' is more a matter of notation than a matter of underlying phonetic realities.

[rotting bones]

I still feel self-conscious if I aspirate the initial stop consonants in English.

[Travis B.]

I still feel self-conscious if I aspirate the initial stop consonants in English.

You do? In Standard English initial fortis plosives are pretty uniformly aspirated aside from some exceptions such as often /t/ in to, today, tomorrow, and tonight. (Note, however, that they are not typically as strongly aspirated as aspirated voiceless consonants in Indic languages.) Conversely, in Standard English initial lenis plosives are often entirely devoiced, such that in isolation the contrast lies solely on aspiration.

[rotting bones]

You do? In Standard English initial fortis plosives are pretty uniformly aspirated aside from some exceptions such as often /t/ in to, today, tomorrow, and tonight.

Are they in Indian English? I feel like I'm faking a Western accent if I do it.

[Travis B.]

You do? In Standard English initial fortis plosives are pretty uniformly aspirated aside from some exceptions such as often /t/ in to, today, tomorrow, and tonight.

Are they in Indian English? I feel like I'm faking a Western accent if I do it.

Well, lack of aspiration is a key feature of Indian English, and probably is in reaction to that native Indic aspirated consonants are 'too' aspirated by Standard English, well, standards.

[Travis B.]

I should note that the lack of aspiration is something that I found the hardest to get used to with my Indian coworkers, because I would hear their initial /p t tʃ k/ as /b d dʒ ɡ/ (since then I have acclimatized to this, though).

[rotting bones]

Well, lack of aspiration is a key feature of Indian English, and probably is in reaction to that native Indic aspirated consonants are 'too' aspirated by Standard English, well, standards.

The aspirated and unaspirated consonants are very strongly distinguished in Indo-Aryan languages. Treating them as phonemic equivalents seems like it can't possibly be right. In Bengali, [tʃɔl] is 'go', and [tʃʰɔl] is 'style'.

[Travis B.]

Well, lack of aspiration is a key feature of Indian English, and probably is in reaction to that native Indic aspirated consonants are 'too' aspirated by Standard English, well, standards.

The aspirated and unaspirated consonants are very strongly distinguished in Indo-Aryan languages. Treating them as phonemic equivalents seems like it can't possibly be right. In Bengali, [tʃɔl] is 'go', and [tʃʰɔl] is 'style'.

Just for the record (I assume you already know much of this):

Initially the plosive fortis/lenis contrast in NAE and EngE is first an aspiration contrast and only second an optional voicing contrast. (In my own dialect it also involves vowel length allophony for vowels in final syllables of preceding words that do not have obstruents in their codas.) The same applies to word-internal stressed syllables' onsets.

(In other positions plosives are not typically aspirated, even though ejectivization may be perceived as aspiration finally.)

[Travis B.]

As requested, here is an off-the-cuff recording of my speech; it's rather halting because I did not really plan what I was going to say ahead of time.

[Starbeam]

Key differences between conventional GA-type transcriptions and my transcriptions are:

• I mark the NCVS.
• I mark allophonic vowel length.
• I mark phonemic consonant length and reduction of consonant clusters to geminates.
• I mark allophonic fortis plosive aspiration (note that conventional GA-type transcriptions sometimes do this but sometimes don't).
• I mark coda fortis plosive (pre)glottalization.
• I mark allophonic obstruent devoicing.
• I mark /l/-vocalization and dorsal realizations of /l/.
• I mark dorsal realizations of /r/.
• I mark American/Canadian raising (my own speech has elements of both).
• I mark front/back allophony of GOOSE, FOOT, and GOAT.
• I mark the weak vowel merger and the resulting distribution of allophones.
• I mark vowel nasality (but note that while I actually have both weakly and strongly nasalized vowels, I normally do not distinguish these in my transcriptions).
• I mark post-tonic intervocalic elision of /t d n nt/, often /nd/, and inconsistently /b v ð/.
• I mark resolution of many intervocalic elisions in resulting diphthongs.
• I mark homorganic coda nasal elision before fortis plosive phonemes in the same coda.
• I mark other frequent consonant elisions such as initial elision of /t/ in to.
• I mark allophonic consonant labialization before GOOSE, FOOT, and /w/.
• I mark front/back allophones of sibilants and other palatalization of coronals.
• I mark assimilation of initial /ð/ across word boundaries.
• I mark affrication of initial /θ/ and /ð/ and stopping of initial /ð/.

I want to keep track of the conversation, so i'm going to bold what examples i think are the result of your speech actually being different from GA and leave alone what's merely notation. Let me know if i got any wrong. Thanks for the recording btw

[Travis B.]

Key differences between conventional GA-type transcriptions and my transcriptions are:

• I mark the NCVS.
• I mark allophonic vowel length.
• I mark phonemic consonant length and reduction of consonant clusters to geminates.
• I mark allophonic fortis plosive aspiration (note that conventional GA-type transcriptions sometimes do this but sometimes don't).
• I mark coda fortis plosive (pre)glottalization.
• I mark allophonic obstruent devoicing.
• I mark /l/-vocalization and dorsal realizations of /l/.
• I mark dorsal realizations of /r/.
• I mark American/Canadian raising (my own speech has elements of both).
• I mark front/back allophony of GOOSE, FOOT, and GOAT.
• I mark the weak vowel merger and the resulting distribution of allophones.
• I mark vowel nasality (but note that while I actually have both weakly and strongly nasalized vowels, I normally do not distinguish these in my transcriptions).
• I mark post-tonic intervocalic elision of /t d n nt/, often /nd/, and inconsistently /b v ð/.
• I mark resolution of many intervocalic elisions in resulting diphthongs.
• I mark homorganic coda nasal elision before fortis plosive phonemes in the same coda.
• I mark other frequent consonant elisions such as initial elision of /t/ in to.
• I mark allophonic consonant labialization before GOOSE, FOOT, and /w/.
• I mark front/back allophones of sibilants and other palatalization of coronals.
• I mark assimilation of initial /ð/ across word boundaries.
• I mark affrication of initial /θ/ and /ð/ and stopping of initial /ð/.

I want to keep track of the conversation, so i'm going to bold what examples i think are the result of your speech actually being different from GA and leave alone what's merely notation. Let me know if i got any wrong. Thanks for the recording btw

Many of these features are actually not all that uncommon for Inland North and northern para-GA varieties here in the US:

The dialect here takes vocalization of /l/ and dorsal realizations of /l/ to an extent that goes beyond most NAE varieties, even though /l/ vocalization in codas and when syllabic is common in modern NAE varieties.

Similarly, the dialect here has significantly more frequent dorsal realizations of /r/ than most NAE varieties, even though it is common for modern NAE varieties to have pharyngealization of /r/.

American raising is very common in modern para-GA NAE varieties; my own idiolect differs in particular because I also have allophonic raising of MOUTH.

Allophonic centralization of GOOSE and FOOT is not uncommon in para-GA NAE varieties; the dialect here though takes this to its final conclusion and also allophonically fronts GOAT.

Intervocalic flap elision is actually not uncommon in para-GA NAE varieties, but it is not nearly as aggressive in most cases as it is in the dialect here.

Similarly, stopping of initial /ð/ is not uncommon in urban Inland North dialects; this may be due to substratum influence from other languages originally spoken by immigrants from Europe.

[zompist]

As requested, here is an off-the-cuff recording of my speech; it's rather halting because I did not really plan what I was going to say ahead of time.

Thanks for taking the time to do this. It seems very close to the speech here in Chicago, with a few noticeable differences:

* lost n's: anyway [ɛ̃we]
* lost l's in -ly words: carefully [kerfʊj], probably [praji]

If I'm not mistaken there are two instances of that's, one with [d] and one with [ð].

[Travis B.]

As requested, here is an off-the-cuff recording of my speech; it's rather halting because I did not really plan what I was going to say ahead of time.

Thanks for taking the time to do this. It seems very close to the speech here in Chicago, with a few noticeable differences:

* lost n's: anyway [ɛ̃we]
* lost l's in -ly words: carefully [kerfʊj], probably [praji]

If I'm not mistaken there are two instances of that's, one with [d] and one with [ð].

To me the biggest phonetic differences between Milwaukee dialect and Chicago dialect are the realizations of TRAP and THOUGHT; Chicago dialect is more likely to diphthongize TRAP, and when both do it Chicago dialect tends to do it more strongly, and Chicago dialect tends to unround THOUGHT where Milwaukee dialect tends to preserve its rounding. However, a friend of mine says that people she knows from Chicago say that Milwaukeeans have really "long" vowels compared to their own.

[jal]

As requested, here is an off-the-cuff recording of my speech; it's rather halting because I did not really plan what I was going to say ahead of time.

It's of course impossible to me to actually hear what you're saying here (given my bad hearing), but it otherwise sounds like "American" to me. The only other observation is that you sound like you're having a pretty bad cold.


JAL

[Travis B.]

As requested, here is an off-the-cuff recording of my speech; it's rather halting because I did not really plan what I was going to say ahead of time.

It's of course impossible to me to actually hear what you're saying here (given my bad hearing), but it otherwise sounds like "American" to me. The only other observation is that you sound like you're having a pretty bad cold.

About the cold, you probably just are hearing the nasal vowels, which for me are more prominent than for many English-speakers.

[jal]

About the cold, you probably just are hearing the nasal vowels, which for me are more prominent than for many English-speakers.

That might indeed be it :).


JAL

[Starbeam]

Many of these features are actually not all that uncommon for Inland North and northern para-GA varieties here in the US:

The dialect here takes vocalization of /l/ and dorsal realizations of /l/ to an extent that goes beyond most NAE varieties, even though /l/ vocalization in codas and when syllabic is common in modern NAE varieties.

Similarly, the dialect here has significantly more frequent dorsal realizations of /r/ than most NAE varieties, even though it is common for modern NAE varieties to have pharyngealization of /r/.

American raising is very common in modern para-GA NAE varieties; my own idiolect differs in particular because I also have allophonic raising of MOUTH.

Allophonic centralization of GOOSE and FOOT is not uncommon in para-GA NAE varieties; the dialect here though takes this to its final conclusion and also allophonically fronts GOAT.

Intervocalic flap elision is actually not uncommon in para-GA NAE varieties, but it is not nearly as aggressive in most cases as it is in the dialect here.

Similarly, stopping of initial /ð/ is not uncommon in urban Inland North dialects; this may be due to substratum influence from other languages originally spoken by immigrants from Europe.

I did intentionally leave off things common to multiple subdialects of American English, but generally considered substandard or regional. Two things that stood out to me were the intervocalic flap elision (i'm guessing that makes 'patting' sound like 'pa'ing'?) and pharyngealizing r. I've been more broadly curious about the exact nature of prevocaling r in rhotic accents (some dialects are alveolar, others a retroflex) honestly.

[Travis B.]

Many of these features are actually not all that uncommon for Inland North and northern para-GA varieties here in the US:

The dialect here takes vocalization of /l/ and dorsal realizations of /l/ to an extent that goes beyond most NAE varieties, even though /l/ vocalization in codas and when syllabic is common in modern NAE varieties.

Similarly, the dialect here has significantly more frequent dorsal realizations of /r/ than most NAE varieties, even though it is common for modern NAE varieties to have pharyngealization of /r/.

American raising is very common in modern para-GA NAE varieties; my own idiolect differs in particular because I also have allophonic raising of MOUTH.

Allophonic centralization of GOOSE and FOOT is not uncommon in para-GA NAE varieties; the dialect here though takes this to its final conclusion and also allophonically fronts GOAT.

Intervocalic flap elision is actually not uncommon in para-GA NAE varieties, but it is not nearly as aggressive in most cases as it is in the dialect here.

Similarly, stopping of initial /ð/ is not uncommon in urban Inland North dialects; this may be due to substratum influence from other languages originally spoken by immigrants from Europe.

I did intentionally leave off things common to multiple subdialects of American English, but generally considered substandard or regional. Two things that stood out to me were the intervocalic flap elision (i'm guessing that makes 'patting' sound like 'pa'ing'?) and pharyngealizing r. I've been more broadly curious about the exact nature of prevocaling r in rhotic accents (some dialects are alveolar, others a retroflex) honestly.

Intervocalic flap elision is actually pretty common in NAE, even though the dialect here takes it to 11. The general rule is that flaps resulting from /t d n nt/ and sometimes /nd/ are often elided starting from the start of the syllable after a stressed syllable, but never in two syllables in a row. This can also apply finally if the next word begins with a vowel, but is typically suppressed if such elision has occurred in the penultimate syllable. The distinctions between these are largely preserved by the fact that preceding vowel allophony (i.e. vowel length allophony, vowel nasality allophony, and American/Canadian raising) stays as-is. Also, in many cases the new hiatuses are resolved by turning them into diphthongs or, if the following vowel phoneme was a schwa before /r/ or /l/, a lengthened vowel; however, hiatuses may be preserved if the preceding vowel is tense (but oftentimes following /i/ becomes [j] even if the preceding vowel is tense). When diphthongs or lengthened vowels are produced, the resulting vowel is one unit of length longer than the preceding vowel's original length; short vowels becomes long vowels and long vowels become overlong vowels.

About pharyngealization of /r/, that just means that in many cases /r/ takes on a pharyngeal coarticulation. In my dialect coda /r/'s are often primarily pharyngeal, but onset /r/'s may also have pharyngeal coarticulation except if they follow a coronal, where then they lack any pharyngeal coarticulation. (My /r/'s also commonly have uvular articulation, even though the strength of this may vary, e.g. it may be weak in codas; also, my /r/'s have postalveolar articulation after coronals. And, of course, as is typical of English varieties my /r/'s are typically labialized in initial position.)