Pronunciation of Standard English in America (1919)

[Emily]

(haven't forgotten about this thread btw, my computer broke so the next update will happen once i'm able to make it to the library since i'm not about to try to deal with bbcode and ipa characters on my cell phone lol)

[Emily]

aː
this is described as basically a compromise between [ɑ] and [æ], "chiefly in somewhat conscious and academic speech", basically in those words that have /æ/ in GA but /ɑ/ in RP: grass, half, laugh, path, paths, halves, aunt, branch, aunt (which, then as now, can also be [ænt]); the author laments that [aː] tends to be given priority in dictionary pronunciations over [æ] in these words, blaming it in part on british speech but also on the fact "that New Enland has exerted, and to some extent continues to exert, a strong influence upon formal instruction and upon notions of cultivation and refinement throughout the country". the author also claimes to have "tested" the sound on many speakers and that everyone he has ever heard use it "did not do so without a certain degree of self-consciousness"

æ, æː

• words like carry, carriage, Clara, caret, claret, parent, Paris, parish, marry, tarry (the verb, not the adjective form of tar) are usually pronounced [æ]; some speakers pronounce them [ɛ], but the [æ] pronunciation "is to be preferred"
• the word radish is normally [ˈrædɪʃ], but "the popular dialects often have [ˈrɛdɪʃ]"
• the first vowel in apricot can vary between [æ] and [eː], as today, but the final syllable can be either [ˌkɑt] or [kət]
• apparently the past tense of bid is bade which is pronounced [bæd]
• halibut can be either [ˈhæləbət] or [ˈhɑləbət]
• as someone who used to have family in spokane, it's nice to see him clarify to his readers that it is in fact pronounced [spoˈkæn]

e, eˑ, eː
the line between the different lengths can vary; he lists the relevant syllables in chaotic, archaic, Baconian as short and in fatally, bakery, bay-berry, payroll as half-long; vacation is [veˈkeˑʃən]

• as described in a different post, the sound is long [eː] and often becomes becomes [eɪ] when stressed in monosyllables, especially when word-final (day, they, whey) or before a voiced consonant (fade, grave, haze)
• this vowel is rare before [r], but is sometimes found in formal speech in vary, Mary, chary, parent, vagary, wary, creating a distinction between these and marry, very, cherry, etc.
• Danish is standardly [ˈdeːnɪʃ], but in popular speech often pronounced [ˈdænɪʃ]
• alongside [ˈɔlˌweːz] ot [ˈɔlˌweˑz], always is often pronounced [ˈɔlwəz], [ˈɔlwɪz], or even [ˈɔləz]
• the "more general" pronunciation of Isaiah is [aɪˈzeːə], but [ɪˈaɪə] is also in current use

ɛ, ɛː

• words like very, perish, terrible, ferry, merit are commonly pronounced with [ɛ]
• "A variant form yelk [jɛlk] exists by the side of yolk [joːk]
• epoch still wavered between [ˈɛpək] and [ˈiːpɑk] just like today, with the latter labeled "very formal"
• [diːf] for deaf is a frequent dialectal pronunciation alongside standard [dɛf]
• words where [ɛ] is commonly pronounced as [ɪ] include get, chest, yet, instead
• "in the popular dialects", [ɛg] is often pronounced [eɪg]: egg, beg, leg, nutmeg — i'm surprised this pronunciation is over a hundred years old! it still seems to be a pronunciation in flux: i always have it in egg, leg; never in beg, peg; and either way in keg, nutmeg. in high school i had a teacher from wisconsin whose bag sounded to us like "bayg", i wish i could remember how she pronounced "egg". the pronunciation of [ɛ] instead of [eɪ], combined with consonant devoicing, is what made me take fully a minute to figure out that a german exchange student saying [ɛk] meant "egg"
• in apparently a three-way contrast, the words chary, fairy, hairy, Mary, vary, wary are pronounced [-ɛːrɪ]
• in non-rhotic dialects, the final glide in fair, hair, etc. [-ɛːɚɹ] becomes [-ɛːə] or even [-ɛːɑ], though this last pronunciation is "nowhere general in America, though sometimes cultivated in imitation of what is taken to be Eastern American or British usage"
• though today's pronunciation [ˈɛːrə-] or [ˈɛːro-] is "quite generally used", the "standard formal" pronunciation prefix aero- is [ˈeːərə-] or [ˈɛːərə-]; [ˈɛrɪə-] is "popular and dialectal"

[foxcatdog]

I'm very suprised at the pronounciation of danish but i am australian.

[bradrn]

I'm very suprised at the pronounciation of danish but i am australian.

‘Standard’ [deːnish] would be the equivalent of Australian [dæ͡ɪnɪʃ]. The whole point is how odd the ‘popular’ pronunciation is.

(Also, I didn’t know you were Australian too, though I did have a feeling that might be the case.)

[Nortaneous]

the line between the different lengths can vary; he lists the relevant syllables in chaotic, archaic, Baconian as short and in fatally, bakery, bay-berry, payroll as half-long; vacation is [veˈkeˑʃən]

Could this be related to the distinction between <ā ē ī ō ū> and <ā̇ ē̇ ī̇ ō̇ ū̇> in this 1892 dictionary?

"In several regions of the Atlantic seaboard a glide vowel is introduced between a preceding [k], [g] and [ɑː], as in the Virginia pronunciation of carter [kɪˈɑːtə], garden [gɪˈɑːdən], but this pronunciation is distinctly local or dialectal" — what the fuck??

Yeah, I find that pretty baffling. I even just now listened to some recordings of Tangier and Smith Islanders to see if this was some highly regressive Tidewater feature but there's not a trace of it in their accents (which are rhotic anyway).

Among accents of European English, I strongly associate this sort of palatalisation with Northern Ireland. Maybe there are conservative accents among the Scotch-Irish in Appalachia with this feature?

Not in Ocracoke either.

Interestingly, the paper I found mentions the practice of making tea from yaupon, which is listed as a distinctive word of the dialect - and backing of START to [ɔr], which is widespread and still alive - the Inland North [a̟ɻ] sounds wrong to me.

This paper says rounding was stigmatized, which I'm surprised by. I've also never heard of the START-NORTH merger that "may have been a majority variant" on the Delmarva Peninsula, but I'm from the wrong side of the Chesapeake for it.

That paper also says:

During the 19th century, insertion of [j] in such words as car [cʰjɑː]~[cʰjɑɚ], garden, and Carter was widespread in coastal plain and Piedmont sections of the South, though perhaps less so in the Appalachians. This variation probably began to decline in the late 19th century and has now entirely disappeared.

(Right before a paragraph on the South as a yod-retention holdout, which tracks. I think some people still retain yod today, but it's subtle.)

And:

Plantation areas typically showed certain dialectal features, particularly non-rhoticity and intrusive [j] in car [cʰjɑː], garden, etc. Plantations occupied the better farmland, such as the “Black Belt” of central Alabama and the Mississippi valley, while poor white farmers predominated in less arable regions, such as the rugged terrain of northern Alabama and the sandy “Piney Woods” region that stretched from southern Georgia and northern Florida to southern Mississippi, with a disjunct area in western Louisiana and eastern Texas.

edit: "One other variant, the tap [ɾ], may have occurred in some older Southern speech after [θ], as in three, but the evidence is unclear." - I don't have this but it's common where I grew up

[axolotl]

I think cot/caught is regional, not temporal. I will try to test this on my nieces at Easter.

It's becoming less regional and more temporal by the day.

[Travis B.]

I think cot/caught is regional, not temporal. I will try to test this on my nieces at Easter.

It's becoming less regional and more temporal by the day.

This is, there are dialects like that here in southeastern Wisconsin which firmly resist the cot-caught merger to this day regardless of age.

[Nortaneous]

The cot-caught merger is generally absent from the Mid-Atlantic and South as well. From New Jersey to Maryland (or maybe northern Virginia) THOUGHT is realized with a schwa offglide in various circumstances, ranging from unconditionally to optionally under complicated circumstances to only in the word "on", pronounced [uən]. This is paralleled by the TRAP-BATH split producing a phonemic contrast between [æ] and [eə], but that hasn't traveled as far south.

(I used to think there was a marginal contrast between /æ/ and /eə/ in DC GenAm, but now I think it's about syllabification, with æ-breaking applying before nasal place assimilation across a syllable boundary, but after nasal place assimilation within a coda - so banker /bænk.ər/ [ˈbæjŋkɚ], but panko /pæn.kʌw/ [ˈpeəŋkəw]. The surname Bernanke can be syllabified either /bər.nænk.ɨj/ [bɚnæjŋki] or /bər.næn.kɨj/ [bɚneəŋki] - I've heard both.)

[alynnidalar]

I think cot/caught is regional, not temporal. I will try to test this on my nieces at Easter.

It's becoming less regional and more temporal by the day.

This is, there are dialects like that here in southeastern Wisconsin which firmly resist the cot-caught merger to this day regardless of age.

Yes and no. Inland North historically avoided the cot-caught merger because in the Northern Cities Vowel Shift, we fronted LOT to [a] before THOUGHT lowered to [ɒ], which means for us to merge the two, we'd either have to shift THOUGHT forward (not impossible, but it'd leave us with pretty weird back vowels) or shift LOT back again (also kind of weird).

Buuuut that reversal of LOT actually may be happening among younger speakers. I've seen papers from at least upstate NY and Michigan's Lower Peninsula with evidence of this. So who knows. Maybe in a few years I'll--gulp--start hearing cot-caught merged in Michigan. I can only pray we avoid this terrible fate.

(I'm joking, but I find NCVS delightful, especially the atrocity we perpetuate by making /æ/ a diphthong. I'd be sad to lose parts of it.)

[Travis B.]

It's becoming less regional and more temporal by the day.

This is, there are dialects like that here in southeastern Wisconsin which firmly resist the cot-caught merger to this day regardless of age.

Yes and no. Inland North historically avoided the cot-caught merger because in the Northern Cities Vowel Shift, we fronted LOT to [a] before THOUGHT lowered to [ɒ], which means for us to merge the two, we'd either have to shift THOUGHT forward (not impossible, but it'd leave us with pretty weird back vowels) or shift LOT back again (also kind of weird).

Buuuut that reversal of LOT actually may be happening among younger speakers. I've seen papers from at least upstate NY and Michigan's Lower Peninsula with evidence of this. So who knows. Maybe in a few years I'll--gulp--start hearing cot-caught merged in Michigan. I can only pray we avoid this terrible fate.

(I'm joking, but I find NCVS delightful, especially the atrocity we perpetuate by making /æ/ a diphthong. I'd be sad to lose parts of it.)

My daughter at least does not back her LOT (except adjacent to /r w h kw gw/, like myself), and THOUGHT is firmly a rounded vowel in the dialect here - but my mother, who is from Kenosha, not that far south of Milwaukee, has an unrounded THOUGHT and has only a central-versus-back contrast between LOT and THOUGHT. I would expect the dialect in Kenosha to cot-caught merge before that here, because that dialect only needs to lose the central-versus-back contrast whereas that here also has to lose the rounding contrast as well.

[Travis B.]

Interestingly enough, though, the dialect here does not uniformly diphthongize TRAP - very many people here only diphthongize it before nasals, like in GA, and some people like myself don't natively diphthongize it at all (I do now, but I didn't when I was growing up - I learned to diphthongize it before nasals as an adult). However, it is practically universal here to raise TRAP - my native TRAP is a bit higher than my native DRESS, which is centralized except when speaking carefully, where then my TRAP and DRESS are really close together.

[alynnidalar]

I hadn't been taking into account rounding--that's an interesting point to consider. Perhaps the youths will be saved after all.

[axolotl]

I think cot/caught is regional, not temporal. I will try to test this on my nieces at Easter.

It's becoming less regional and more temporal by the day.

This is, there are dialects like that here in southeastern Wisconsin which firmly resist the cot-caught merger to this day regardless of age.

There are places that resist, yes. I suspect that "General American" (to the extent it exists) either will be fully merged soon or arguably already is, with the lack of merger ultimately persisting only in speakers who have "marked regional accents" of the Great Lakes, South, and Urban Northeast regions. And sometimes not even then. I have relatives in central Virginia, and those close in age to me (I'm 22) sound distinctly southern (although not strongly so) yet have the merger.

[Travis B.]

There are places that resist, yes. I suspect that "General American" (to the extent it exists) either will be fully merged soon or arguably already is, with the lack of merger ultimately persisting only in speakers who have "marked regional accents" of the Great Lakes, South, and Urban Northeast regions. And sometimes not even then. I have relatives in central Virginia, and those close in age to me (I'm 22) sound distinctly southern (although not strongly so) yet have the merger.

I would only say this is true because "General American" today is closely aligned with Midland and Western dialects in the first place, which tend to be cot-caught merged or close to, by exclusion of dialects in the Great Lakes region, the Northeast, and the South.

[Emily]

all right, got a new keyboard so i can type again (except it has the insert/home/pgup/del/end/pgdn block shifted down one row! i hate this! never ever do this!!!! aaaaaaaaahhhh)

ə

• appears in dialect speech in final unstressed syllables where standard speech has a clear vowel: yellow, potato, tomato, piano, window, fellow, thorough (as well as the aforementioned always [ˈɔlwəz]) — i had a teacher in high school who was from indiana and who would pronounce "window" [ˈwɪndə]
• he lists words where "careless and rapid speech" tends to delete schwas; this includes words with vowel hiatus such as poem (standard [ˈpoːɛm], [ˈpoːəm], [ˈpoːɪm], nonstandard [poːm]) and violet (standard [ˈvaɪəlɪt], nonstandard [ˈvaɪlɪt] or [ˈvɑːlɪt]), but also words such as moral and towel, where he contrasts "popular" [mɔɹl], [tɑʊl] with "standard" [ˈmɔrəl], [ˈtɑʊəl]—this is sort of mindblowing to me, since i can't even figure out how to pronounce these words as one clear syllable unless i pronounce the /l/ as a clear, "light" /l/ like in german
• he also lists the standard pronunciation of diamond as three syllables
• apparently deal, seal, peal are standardly [-iːl] but real is standardly [ˈriːəl], which he chalks up to the different etymologies of the words
• word final unstressed [-ə] is frequently [-ɪ] in popular speech, with the examples opera, era, extra, America, Noah, Martha, and sometimes Iowa; this is obviously the source of the word "Op'ry" today, and you'll sometimes hear newsboys say "extry" in cartoons, but i think even in a cartoon if i heard someone say "ameriky" i'd fall over backwards
• a popular pronunciation of cupola transposes the final syllables: [ˈkjupəlˌo] for standard [ˈkjupələ]; i can't say as i've ever said this word out loud
• a [ə] is often inserted into [lm] in popular pronunciations of elm, film, realm, etc, as well as into [θl] in athlete, athletic; another pronunciation i've only ever heard on tv
• he notes that british speakers tend to insert [ə] before [r] when it follows [ɪ], [ɛ], [aɪ] (examples he gives are period, peeress, parent, miry, Byron), "but this [ə] is scarcely ever heard in America"; british posters can tell me if these words are still pronounced this way
• an exception to the above is that in fiery, the pronunciation can be either [ˈfaɪrɪ] or [ˈfaɪərɪ], and the unstated implication seems to be that the latter is standard; to me, pronouncing this without the glide has the same artificial sound as when the gym teacher in Carrie is dressing down the other girls for bullying carrie, and she says "carrie hwite has fee-lings" (instead of "feel-ings"), which i presume is mostly because the actress was probably trained in classical stage diction

ɚ
again, the symbol he uses is not one i've been able to find in unicode, he prints it as a schwa but with the end (the northwest point) curling back in to point at the curve

• can be found in stressed or unstressed syllables" bird, burr, sir, fir, etc. [bɚɹd], [bɚɹ], [sɚɹ], [fɚɹ]; typically sound is short, "but may be prolonged in exceptional circumstances" (e.g. the "somewhat exotic word" myrrh)
• non-rhotic accents tend to substitute [ʌː] for the stressed sound, but when word-final (e.g. burr, fur often [ʌːə]; in unstressed words the sound becomes just [ə]
• the standard rhotic and non-rhotic pronunciations for girl are [gɚɹl] and [gʌːl], "but [gɛːɹl], [gæɹl], [gɪːɹl] are sometimes heard and are often cultivated as refined pronunciations" (!!!)
• courteous, courtezan are usually pronounced with ['kɚɹt-] but courtier with [ˈkɔrtʃ-]; i only really know "courtesan" from baz luhrmann's moulin rouge!, where all the british-accented french people pronounce it with [ɔ]
• words with standard [ɚɹ] followed by a consonant are instead often pronounced [əɪ] in new york and its vicinity, "commonly represented in dialect stories by the spelling oi . . . This pronunciation has not made its way into cultivated usage" — hey, we had a thread about this!
• in stressed words with [aɪ] or [aʊ] followed by [ɹ], standard pronunciation regularly inserts [ɚ], rendering hire and higher homophones: [ˈhaɪɚɹ]; he describes "a slovenly kind of pronunciation" where the [ɚ] is deleted and the diphthong becomes [ɑː]: flower, flour [flɑːɹ]; fire [fɑːɹ]; our, hour [ɑːɹ]. (it's unclear from how it's worded whether this phenomenon also takes place in words that are "properly" two syllables such as "higher".) he does note that this pronunciation of our is so common "that perhaps one cannot characterize it as slovenly"

[foxcatdog]

I can confirm in australian english liquids after diphthongs are pronounced with an epenthetic schwa but i don't see it after lax vowels.

[Darren]

• apparently deal, seal, peal are standardly [-iːl] but real is standardly [ˈriːəl], which he chalks up to the different etymologies of the words

As another australian I have this. Almost all -eal words are /iːl/ (realised as [ɪːɫʷ]) except for real, reel, eel, seal (the animal) and wheel which are usually bisyllabic /iːəl/ ([əi̯ɫ̩ʷ]), but sometimes drift to /iːl/ in "slovenly" speech. I guess it could be etymological; Old English -eō̆l(h) gives /iːəl/ in reel, seal and wheel, and real comes from Middle English /-eːal/ which seems to be unique. eel is the odd one out, as OEng -ǣ̆l gives /iːl/ otherwise.

[foxcatdog]

• apparently deal, seal, peal are standardly [-iːl] but real is standardly [ˈriːəl], which he chalks up to the different etymologies of the words

As another australian I have this. Almost all -eal words are /iːl/ (realised as [ɪːɫʷ]) except for real, reel, eel, seal (the animal) and wheel which are usually bisyllabic /iːəl/ ([əi̯ɫ̩ʷ]), but sometimes drift to /iːl/ in "slovenly" speech. I guess it could be etymological; Old English -eō̆l(h) gives /iːəl/ in reel, seal and wheel, and real comes from Middle English /-eːal/ which seems to be unique. eel is the odd one out, as OEng -ǣ̆l gives /iːl/ otherwise.

In my case i struggle to not pronounce the bisyllabically

[Travis B.]

• apparently deal, seal, peal are standardly [-iːl] but real is standardly [ˈriːəl], which he chalks up to the different etymologies of the words

As another australian I have this. Almost all -eal words are /iːl/ (realised as [ɪːɫʷ]) except for real, reel, eel, seal (the animal) and wheel which are usually bisyllabic /iːəl/ ([əi̯ɫ̩ʷ]), but sometimes drift to /iːl/ in "slovenly" speech. I guess it could be etymological; Old English -eō̆l(h) gives /iːəl/ in reel, seal and wheel, and real comes from Middle English /-eːal/ which seems to be unique. eel is the odd one out, as OEng -ǣ̆l gives /iːl/ otherwise.

To me real, reel, eel, seal, and wheel are all monosyllabic, as -[i(ː)ɯ̯].

[Nortaneous]

words such as moral and towel, where he contrasts "popular" [mɔɹl], [tɑʊl] with "standard" [ˈmɔrəl], [ˈtɑʊəl]—this is sort of mindblowing to me, since i can't even figure out how to pronounce these words as one clear syllable unless i pronounce the /l/ as a clear, "light" /l/ like in german

How many syllables does "Carl" have?