North American English Dialects, Based
on Pronunciation Patterns
Web-Based Survey A
group of linguists are gathering data on North American English dialects
using a web-based survey. They currently have about 1000 responses but would
like to double or triple that if possible. This survey is distinct from the
information on this page, but they have asked for our help. If you would like to take this survey, it can be
found at: pantheon.yale.edu/~clb3/NorthAmericanDialects. 6-May-2010 |
This is just a little hobby of mine, that I thought might be interesting to a lot of people. Some people collect stamps. Others collect coins. I collect dialects. Please let me know what you think of this page. - Rick Aschmann (Last updated: August 28, 2010.)
There are 8 major
English dialect areas in North America, listed below the map at left. These
are shown in blue, each with its number, on the map and in the Dialect Description Chart below, and are
also outlined with blue lines on the map. The first 6 of these begin at the eastern seaboard and proceed west,
reflecting western settlement patterns. Adj.
24-Aug.-2010 The many subdialects
are shown in red on the map and in the chart, and are outlined with red lines
on the map. All of these are listed in the margins of the map as well. Adj. 28-Aug.-2010
In the Dialect Description Chart additional
features not shown on the map are provided for distinguishing the dialects.
What’s New? All additions or changes within the last two or three months are marked with “ New! ” and the date, or with “Adj.” (for “Adjusted”.) To see this new information, simply search for these words. Adj. 24-Aug.-2010 Recent additions: The Cajun English dialect, which exists alongside Cajun French in Louisiana. Suggested by Philip Batton and Martin Ball. Thanks! I still don’t have many samples, so send me some, please! New! 23-June-2010 Web Forums: There are several web forums or blogs that refer to my map. The most recent one that I know of, set up on June 7, is phonetic-blog.blogspot.com/2010/06/american-dialect-areas.html. An older one is forum.skyscraperpage.com/showthread.php?t=175333, from November, 2009. I have made a few adjustments based on comments made there, and am adding some sound samples. However, the main complaint, that the map is too complicated and confusing, I can’t really fix: the subject is complicated, and I am well aware that I have tried to include too many features. However, if people have ideas on how to make the map or web page less confusing, I am all ears! Adj. 20-June-2010 Humble pie: Oops! I was dead wrong about
Montreal and southwestern Quebec being a part of the Northwestern New England
dialect, as contributor Chris Harvey and my recent research have shown. See The U.S. - Canada Border and the “Badge of
Identity”. Adj. 24-Aug.-2010 |
|
|||||||
The small map below is the same as the Full-Scale Dialect Map that follows, but shows the entire width of the map (on most monitors). Adj. 24-Aug.-2010
Click on any part of this map to move to the equivalent part of the Full-Scale Dialect Map. (For now this only moves to the far left or the far right of the Full-Scale Dialect Map, so unfortunately it doesn’t work well for the middle portions, and you will just have to scroll over.) Adj. 24-Aug.-2010
Instructions
For many of the cities or towns on this map, you can listen to an audio or video sample of speech of a native (more specifically, someone who was raised there, though not necessarily born there, and whose dialect clearly represents that place). All of the cities or towns with a green center have such an audio or video sample that can be listened to (and a few of the ones with pink centers do also). I will continue adding new audio and video samples, so check back from time to time. So far there are over 500 samples listed, a number of which are from contributors! Thanks! New! 24-Aug.-2010
Use the scroll bars to move around on this map, or, even simpler, start at the tiny map above and click the country (U.S. or Canada) that you want to look at. This will take you to the Small-Scale Dialect Map. Click again to zoom in further on your location. (For now this only moves to the far left or the far right of the Full-Scale Dialect Map, so unfortunately it doesn’t work well for the middle portions, and you will just have to scroll over.) Adj. 24-Aug.-2010
So far about 3/4 of the states and
provinces on this map are clickable, taking you to the list of samples for that
state or province. Place the mouse over a particular state or province to see
if it is currently clickable. If it is not, you will have to go to the Audio
Samples of Local Dialects section and find the samples for a particular
state. Only those locations with green centers,
and a few with pink centers, have a sample
so far. If there aren’t any of these samples yet for a particular state or
province, then it will not be clickable, since there is no data to go to! (The map guides, showing the meaning of all
the colors, are on the top right and bottom left of the map.) Adj. 24-Aug.-2010
Help! For many places I haven’t found an audio sample yet. If you know of an audio or video sample on the Internet that features a speaker who was raised in a particular place, and whose dialect clearly represents that place, please let me know, whether that place is currently listed or not! Although many of the people in these samples are prominent people, I actually prefer ordinary local people, but anyone at all will do, as long as their pronunciation represents the local dialect. (The ones I especially need, and cannot find, are those with an orange-yellow center.) Also, if you think that one of the audio examples does not truly represent the local dialect, please let me know in the same way.
Numbered Locations: Thanks to
enthusiastic contributor Eli K., much of Kentucky and
Tennessee and neighboring areas are about as thoroughly mapped as they
possibly can be. That’s the kind of help needed to really fill out this map! In
fact, I was forced to go to a numbering system
for such areas, since the scale of map is already big enough! I put the key to
the numbers off the east coast of Canada. New! 24-Aug.-2010
|
|
|
|
|
Data from the Atlas of North American English (ANAE)
I am grateful to the Atlas of North American English (ANAE) by William Labov, Sharon Ash, and Charles Boberg, for a good part of the data on which this map was based. Specifically, much of the information on the map above and in the Dialect Description Chart below was obtained from: ling.upenn.edu/phono_atlas/Atlas_chapters/Ch11_2nd.rev.pdf, which is chapter 11 of a draft version of the atlas, as well as from many other chapters of the same work, with a few ideas from a much older version of the same: ling.upenn.edu/phono_atlas/home.html. (The Table of Contents of the draft version of the atlas can be seen at: www.ling.upenn.edu/phonoatlas/ANAE_ToC.pdf, but this does not link directly to the chapters.)
However, the names of a number of the dialects are my own, and I have made many adjustments to their borders (especially Inland and Lowland South, West Midland, and Allegheny Midland). Also, a lot of the data is from my own research and analysis. - Rick Aschmann
The following notes refer to numbers on the map, and show the corresponding section of the ANAE:
1: Pin-pen merger: See Map 9.5 in www.ling.upenn.edu/phono_atlas/Atlas_chapters/Ch09_2nd.rev.pdf and www.ling.upenn.edu/phono_atlas/maps/Map3.html. This is the only feature in which I find myself in significant disagreement with the ANAE: I have found that the pin-pen merger area is much larger than he shows, especially in the west. (See The Pin-Pen Merger, The Kansas-Nebraska Act, and the Texas Cattle Drives below.)
2: Long /ō/ [oʊ] fronting: See Map 20.2 in www.ling.upenn.edu/phono_atlas/Atlas_chapters/Ch20_2nd.rev.pdf. 2: The boundary between central-back and central-front (the yellow dots) was used by the ANAE to define the boundary between North and Midland, but this line then extends into the West. The deep dip that it takes southwards in Utah and Nevada would seem to indicate settlement of these areas by Northerners, probably represented by the Mormon settlement. Thus this dip corresponds to a large degree to the “Mormon Corridor”. Many of these settlers were originally from the Palmyra, New York, area and from Kirtland, Ohio. Another northern contribution may have been the early northeastern organized crime influence in Las Vegas. Now I’m not saying that people in these areas sound like northeasterners: they don’t, they sound like westerners, with this one feature being dragged south because of this origin. 20-Jan.-2010
3: R-dropping: See Map 7.1 in www.ling.upenn.edu/phono_atlas/Atlas_chapters/Ch07_2nd.rev.pdf. R-droppers are also called non-rhotic English speakers. There are two types of r-droppers, which I call Systematic R-droppers and Simple R-droppers.
Systematic R-droppers are found in the northeastern U. S., in much of England, and in Australia, New Zealand, South Africa, India, much of the Caribbean, and other places. Systematic R-droppers have linking and intrusive r’s. John F. Kennedy is an excellent example of a Systematic r-dropper. In a speech he gave prior to being elected, he says “The hungry children I sawr in West Vaginia.” This quote has one intrusive r, and one dropped r. In another speech during the Cuban missile crisis he says: “Theah has been a great deal of talk on the situation in Cubar in recent days...”, which again has one intrusive r, and one dropped r.
Simple R-droppers are found in parts of the Lowland South. As a general rule, they do not have linking and intrusive r’s. All of the areas in the South marked as r-droppers on my map are Simple R-dropper areas.
Numerous examples of both kinds of r-droppers are given in the audio samples below. This pattern is receding, so occasionally only much older speakers retain the r-dropping in a given location. These are surrounded on the map by a dotted green circle, and in the chart below will be indicated with the phrase “Older r-dropper:”. 3-Apr.-2010
4: Canadian and Tidewater raising: See Map 15.5 in www.ling.upenn.edu/phono_atlas/Atlas_chapters/Ch15_2nd.rev.pdf, noting only the info for the /ou/ [aʊ] vowel (which he writes /aw/), not the /ī/ [aɪ] vowel (which he writes /ay/). For Tidewater I have gleaned the info from various sources, including stray comments in ANAE. Adj. 3-July-2010
5: Bite-bout line: See Map 14.1 in www.ling.upenn.edu/phono_atlas/Atlas_chapters/Ch14_2nd.rev.pdf.
6: “On” line: See Map 14.2 in www.ling.upenn.edu/phono_atlas/Atlas_chapters/Ch14_2nd.rev.pdf. This is the only lexical item included in this analysis, but it seems to correlate with the phonological data. Why it turns north at its western end in the Dakotas and does some contortions is unclear. (The ANAE did not have Mitchell, SD or Ashley, ND, which are the surprises. Actually, Mitchell is not a surprise, or even borderline, but one speaker seems to be anomalous.) Miles City, Montana, an outlier of Western North, is also below the line. 24-Mar.-2010
|
|
vowel of “lot” fronted almost as much as vowel of “let” |
vowel of “cot” more fronted than vowel of “cut” |
vowel of “too” much more fronted than vowel of “toe” |
Vowel of “far” fronted |
Vowel of “caught” strongly raised |
“hoarse” = “horse”, “mourning” = “morning”, “four” = “for” |
Unique Features |
Chapter and map in ANAE |
|
Chapter and map in ANAE |
||||||||
|
7. North Central |
|
yes |
Like Western North, but “cot” = “caught” |
|||||
|
Mat-Su Valley, Alaska |
yes |
|
yes |
Strongly like North Central, but with some admixture from the main Alaska dialect. (See Sarah Palin.) |
||||
|
8. West |
yes |
|
yes |
Vowel of “too” significantly more fronted than vowel of “toe”, “cot” = “caught” |
||||
|
Alaska |
yes |
|
yes |
Same as West (ANAE says there are significant differences, but does not make clear what they are.) |
||||
|
Silver City, NM |
yes |
|
yes |
Same as West, but “cot” <> “caught” |
||||
|
1. Canada (main area) |
yes |
very little |
|
yes |
Same as West, plus Canadian vowel shift, vowel of “cat” central, raising of “bite”, “bout” |
|||
|
Atlantic Provinces |
mixed? |
yes |
|
yes |
vowel of “far” fronted (but not “father”) |
|||
|
2. Northern New England |
yes |
|
no |
“far” & “father” fronted, systematic r-dropping, “cot” = “caught”, “father” & “bother” don’t rhyme |
||||
|
NW New England |
very little |
very little |
yes |
|
yes |
vowel of “far” fronted (but not “father”) , “cot” = “caught” (Changed name Aug. 1, 2009) |
||
|
3. The North |
mixed |
mixed |
mostly |
|
almost all |
Back vowels strongly backed |
||
|
Western North |
mixed |
mixed |
mostly |
|
yes |
Least distinctive dialect of the North, some sections are “General American” Changed name 27-Nov.-2009 |
||
|
Inland North |
Yes |
yes |
mostly |
|
almost all |
Northern Cities Shift: “bat” strongly raised, most short vowels shifted |
||
|
St. Louis Corridor |
Yes |
yes |
mixed |
|
mixed |
Northern Cities Shift: “bat” strongly raised, most short vowels shifted, but many other vowels like Midland |
||
|
Indiana Intrusion[1] |
no |
no |
yes |
Only “North” because of the pronunciation of long /ō/ [oʊ], the raising of “bat”, and the bite-bout line[2]. Pin=pen. (Added this dialect 16-Feb.-2010) |
||||
|
Eastern North |
yes |
yes |
very little |
mixed |
yes |
Mostly like Western North, but some similarities to New York City (Changed name July 21, 2009) |
||
|
Albany |
yes |
yes |
very little |
yes |
yes |
Many vowels like New York City, but no r-dropping |
||
|
Providence |
yes |
|
no |
vowel of “cat” central, systematic r-dropping, “cart” = “cot”, which is not seen anywhere else in the world! 3-Apr.-2010 |
||||
|
4. Greater New York City |
yes |
Yes |
Various unusual vowels, systematic r-dropping, “bad” & “had” don’t rhyme, “father” & “bother” don’t rhyme (30-Mar.-2010) |
|||||
|
The Hamptons |
|
Yes |
A lot like New York City, but more research needed! 16-Mar.-2010 |
|||||
|
Downtown New Orleans |
|
yes |
More like New York City than anything else, although “bad” & “had” probably don’t rhyme |
|||||
|
5. The Midland |
|
almost all |
In many ways is intermediate between Northern and Southern[3] Changed name 27-Nov.-2009 |
|||||
|
Central Midland |
|
almost all |
Least distinctive dialect in the U.S., many sections are “General American” |
|||||
|
Canton, Ohio |
|
Yes |
“bat” strongly raised, “on” rhymes with “don”, not “dawn” |
|||||
|
Cincinnati, Ohio |
|
Yes |
Many vowels are pronounced like New York City |
|||||
|
West Midland (Not in ANAE) |
|
Yes |
“cot” = “caught” |
(19) |
||||
|
Allegheny Midland[4] |
|
Yes |
“cot” = “caught” (Changed name June 2, 2009) |
|||||
|
Pittsburgh |
|
Yes |
Pittsburgh vowel shift: “out” is pronounced with no diphthong, the way a Bostonian says “art” |
|||||
|
Oklahoma City (Not in ANAE) [5] |
|
Yes |
Like the parts of the Central Midland south of the pin-pen line (Added this dialect July 11, 2009) |
|||||
|
East Midland (Not in ANAE) |
yes |
Yes |
Like the Central Midland, with influences from Atlantic Midland (Added June 3, 2009) |
|||||
|
Atlantic Midland[6] |
yes |
Yes |
“bad” does not rhyme with “had”, like New York City (20-Jan.-2010) |
|||||
|
North Florida |
|
Yes |
Like the Central Midland, “pin” = “pen” |
|||||
|
South Florida |
|
Yes |
Like the Central Midland, “pin” <> “pen” |
|||||
|
El Paso |
|
Yes |
“cot” <> “caught”, “pin” = “pen” |
|||||
|
Corpus Christi |
|
Yes |
Mixture of Southern and Northern features (but no Southern shift), “pin” = “pen” |
|||||
|
Galveston (Not in ANAE) |
|
Yes |
Very similar to East Midland, or even to Atlantic Midland, except that “bad” rhymes with “had” |
|||||
|
San Francisco Bay |
yes |
yes |
Very similar to East Midland, or even to Atlantic Midland, except that “bad” rhymes with “had” |
(11) |
||||
|
6. The South |
|
mixed |
Partial to complete Southern shift: vowels of “ride” and “buy” have no diphthong |
|||||
|
Lowland South |
|
mixed |
Partial Southern shift: vowels of “ride” and “buy” have no diphthong, but “right” does, “pin” = “pen” |
|||||
|
Savannah |
|
yes |
R-dropping, “pin” <> “pen” |
|||||
|
|
yes |
East is R-dropping, west apparently not, “pin” = “pen”, French influence, th > t,d New! 23-June-2010 |
||||||
|
New Orleans, Mid City |
|
yes |
See New Orleans inset on map and the New Orleans section below |
|||||
|
New Orleans, Irish Channel |
|
yes |
See New Orleans inset on map and the New Orleans section below |
|||||
|
Inland South |
|
almost all |
Full Southern shift: vowels of “ride”, “buy”, and “right” all have no diphthong, “pin” = “pen” |
|||||
|
Charleston |
|
yes |
No Southern shift, vowels of “bait” and “boat” are not diphthongs |
|||||
|
Outer Banks (Not in ANAE) |
|
|
|
|
|
yes |
No Southern shift, long /ī/ [aɪ] vowel often almost like /oi/ [ɔɪ], “pin” <> “pen” |
|
Colors: |
|
|
Transitional areas within
main dialects |
Distinctive or innovative features of a given dialect |
|
Transitional areas
outside main dialects |
Intermediate or partial features |
I have added and adjusted a lot of the information on the
map based on the following audio and non-audio data. - Rick Aschmann
Regional non-audio data
|
Location |
Source |
|
English, French, and indigenous mother-tongue areas of
Canada |
atlas.nrcan.gc.ca/site/english/maps/peopleandsociety/lang/languages2001/mt |
|
French mother-tongue areas of Maine |
|
|
Indigenous languages |
|
|
the eastern boundary of Inland North |
|
|
Greater New York City |
|
|
Eastern boundary between Eastern New England and Providence |
ling.upenn.edu/~johnson4/pwpl_draft.pdf 30-Mar.-2010 |
Multiple-region audio samples found on the Internet
|
Location |
Source |
Comment |
|
Samples from almost all U. S. states |
Few indications are given as to whether the speakers are
natives of the area in which they were interviewed, or if they represent well
the local dialect. Also, locations are often limited, and often only urban
locations are given. Even so, in many cases the data is useful, and I have
used it (see the next chart), especially when clearer indications are given
of “nativeness”. |
|
|
Samples only from north-central U. S. |
Again, not always clear if the speakers are natives of the
area in which they were interviewed, or if they represent well the local
dialect. However, includes rural speakers, which can help fill in holes. Used
occasionally. |
Guide to the Sounds of North American English
In many places on this web page the pronunciation of a name or other word will be given after it. These pronunciation guides will have two forms: a phonemic guide between slashes / /, based on the Traditional Dictionary Pronunciation System (TDPS) that is found in many dictionaries, and a phonetic guide (providing the phonetic details) between square brackets [ ], based on the International Phonetic Alphabet (IPA). (Thanks, Maria Mikkonen![7])
The advantage of the phonemic guide is that it allows different dialects to use the same pronunciation key and get the right result for each dialect. For this guide I have used the system used in the American Heritage Dictionary, rather than the one used by Merriam-Webster, since it is more complete and applies to more dialects.
However, I have included the phonetic guide as well, partly to give a more precise phonetic pronunciation of local variants, and partly because many have complained that they prefer the IPA or are familiar only with it. Adj. 3-July-2010
In the phonemic guide I have followed the American Heritage Dictionary system to the letter, except for a few minor adjustments in the vowel system, and the following differences:
1. I write the syllable with primary stress using bold and underline, and syllables with secondary stress with just bold, rather than using an apostrophe after it like the AHD. In other words, I show the pronunciation of “underneath” as /ŭndərnēth/, whereas the AHD does it as /ŭn'dər-nēth'/.
2. I do not separate syllables with a hyphen except when absolutely necessary, as in “cartridge” /kärtrij/ versus “cartwright” /kärt-rīt/, or “mission” /mĭshən/ versus “mishap” /mĭs-hăp/; although technically in these two cases the underlining of the primary-stressed syllable gives enough information, even so the hyphens help to clarify.
3. I
show the pronunciation of words like “needle” and “sudden” as /nēdəl/ and /sǔdən/, rather than
treating them as having syllabic /l/ or /n/. Adj.
3-July-2010
The ANAE does not use
either the TDPS or
the IPA, but instead uses a completely different
transcription system, described in ling.upenn.edu/phono_atlas/Atlas_chapters/Ch02_2nd.rev.pdf. This
system is phonemic, like the TDPS.[8] Adj. 3-July-2010
How Many Vowels are there in American English?
No, the answer is not: “Five: a, e, i, o u.” Granted, in traditional English spelling those are the vowel letters, yes, but I’m talking about our spoken language: How many significant vowel sounds are there? Well, if you consult any popular American English dictionary, and study the Pronunciation Key, there will be a long list of vowels. In the Pronunciation Key to the American Heritage Dictionary, 19 different vowel symbols are listed (not counting the ones only used in foreign words)! However, some of these are special vowels that only occur before the /r/ sound, which are “colored” by the /r/, so these can be separated out as special cases. One of these vowels, /ə/, only occurs in completely unstressed syllables, never in stressed syllables (whether primary stressed or secondary stressed), so it also can be separated out as a special case. This leaves us with 15 vowels that can occur in stressed syllables. Not all North American English speakers have all of these vowels: Some have 14, some have only 13. New York City has 16, including one that is not usually listed in dictionary pronunciation guides, and which I have chosen to spell /ăə/! These 16 vowels are listed below in the second column, with sample words shown in the first column. Adj. 17-July-2010
The remaining columns show what happens to vowels before final r, showing the “R-colored” vowels used in most of North America, and showing the Southern System in the final column, representing the system used in much of the South, which does not have “R-colored” vowels.
(I have included the IPA equivalents of these vowels in brackets [ ] as well. However, keep in mind that the actual pronunciation of a given phonemic vowel may vary greatly from region to region. For example, the /ŏ/ vowel is pronounced as [a], an open front unrounded vowel, in much of the Inland North, but is pronounced as [ɔ], an open-mid back rounded vowel, in England. A whole gamut of vowel sounds in between these two occurs somewhere in North America: in much of Canada and in some other cot = caught areas the pronunciation is [ɒ], whereas most others use [ɑ] or [a] or something in between. Many other vowels have similar variants. The most distinctive Southern pronunciation is shown in a separate column. However, keep in mind that I have not listed all possible variants for any region.) Adj. 3-July-2010
|
Ordinary Stressed Vowels |
|
|
|
“R-colored” Stressed Vowels |
||||||||
|
phonemic |
IPA |
IPA South |
|
|
keepers |
droppers |
Southern System |
|||||
|
beat, bee, fleece |
ē |
[i] |
[ɪi] |
|
|
fear, pier, peer, near |
îr |
[ɪr] |
[ɪə] |
intemediate between ē and ĭ |
/ēər/ [ɪiə(r)], rhymes with “skier” |
|
|
bit, fill, kit |
ĭ |
[ɪ] |
[iə] |
|
|
|||||||
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
bait, bay, face |
ā |
[eɪ/e] |
[æɪ] |
|
|
fair, bear, care, square, |
âr |
[er] |
[eə] |
For most speakers, intemediate between ā and ě, but ăər [eə(r)] in areas with ăə. |
/ăr/ [ær/æə], “hairy” = “Harry” [hærɪ], but “merry” is different. |
|
|
bet, help, dress |
ě |
[ɛ] |
[e] |
|
|
|||||||
|
bat, had, trap |
ă |
[æ] |
[æiə] |
┐ |
|
|||||||
|
bad, man |
ăə |
[eə] |
|
┘ |
(NYC, Atlantic Midland) |
|
||||||
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
father, Bach’s, palm |
ä |
[a/ɑə] |
|
┐ |
(E. New England, NYC) |
|
far, farther, heart, start |
är |
[ar/ɑr/ɒr] |
[a/ɑ/ɒə] |
Everyone has this![9] |
/är/ [ɒ(r)/ɔr] |
|
cot, lot, bother, box, |
ŏ |
[a/ɑ/ɒ] |
[ɒ] |
┤ |
|
|
|
|
|
|
|
|
|
caught, thought,
balks, |
ô |
[ɒ/ɔ/oə] |
[ɒʊ] |
┘ |
(Eastern U.S. See map.) |
|
for, horse, morning, |
ôr |
[ɔr/or] |
[ɔə/oə] |
Varies a lot! |
[ɒ(r)/ɔr/ɒʊ]; /är/ for many speakers, /ôr/ for others |
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
boat, goat, toe |
ō |
[oʊ/o] |
[ɘʊ] |
|
|
four, hoarse, mourning, |
(ōr) |
[or] |
[oə] |
/ôər/ [ɒʊə(r)], rhymes with “rawer” |
||