This article is about the cultivated accent blending American and British English. For the native dialect of the Mid-Atlantic region of the United States, see
Philadelphia English.
This article includes inline links to audio files. If you have trouble playing the files, see
Wikipedia Media help.
The Mid-Atlantic accent, or Transatlantic accent,[1][2][3] is a consciously learned
accent of English, fashionably used by the
American upper class and
entertainment industry of the late 19th century to mid-20th century, that blended elements from both
American and
British English. Specifically, it blended features from both prestigious coastal
Northeastern American English and from
Received Pronunciation, the standard speech of England. The accent was embraced in private independent American
preparatory schools, especially by members of the Northeastern upper class, as well as in schools for film, radio, and stage acting,[4] with its overall use sharply declining after the
Second World War.[5] The Mid-Atlantic accent is not a native or regional accent; rather, according to voice and drama professor
Dudley Knight, "its earliest advocates bragged that its chief quality was that no Americans actually spoke it unless educated to do so".[6]
A similar accent that resulted from different historical processes,
Canadian dainty, was also known in Canada, existing for a century before waning in the 1950s.[7] More broadly, the term "mid-Atlantic accent" can also refer to any accent with a perceived mixture of American and British characteristics.[8][9][10]
Elite use
History
In the 19th century and into the early 20th century, formal
public speaking in the United States focused primarily on song-like intonation, lengthily and tremulously uttered vowels (including overly articulated
weak vowels), and a booming resonance.[11] Moreover, since at least the mid-19th century, upper-class communities on the
East Coast of the United States increasingly adopted many of the phonetic qualities of
Received Pronunciation—the standard accent of the British upper class—as evidenced in recorded public speeches of the time, with some of these qualities, like
non-rhoticity (sometimes called "r-lessness"), also shared by the regional dialects of
Eastern New England and
New York City.
SociolinguistWilliam Labov et al. describe that such "r-less pronunciation, following Received Pronunciation, was taught as a model of correct, international English by schools of speech, acting, and elocution in the United States up to the end of World War II".[5]
Early recordings of prominent Americans born in the middle of the 19th century provide some insight into their adoption (or not) of a carefully employed non-rhotic Mid-Atlantic speaking style.
PresidentWilliam Howard Taft, who attended public school in Ohio, and inventor
Thomas Edison, who grew up in Ohio and Michigan in a family of modest means, both used natural rhotic accents. Yet presidents
William McKinley of Ohio and
Grover Cleveland of
Central New York, who attended private schools, clearly employed a non-rhotic, upper-class, Mid-Atlantic quality in their
public speeches that does not align with the rhotic accents normally documented in Ohio and Central New York State at the time; both men even use the distinctive and especially archaic affectation of a "
tapped r" at times when r is pronounced, often when between vowels.[12] This tapped articulation is additionally sometimes heard in recordings of
Theodore Roosevelt, McKinley's successor from an affluent district of New York City, who used a cultivated non-rhotic accent but with the addition of the
coil-curl merger once notably associated with
New York accents.[12] His distant cousin
Franklin D. Roosevelt also employed a non-rhotic Mid-Atlantic accent,[13] though without the tapped r.
In and around
Boston, Massachusetts, a similar accent, in the late 19th century and early 20th century, was associated with the local urban elite, the
Boston Brahmins. In the
New York metropolitan area, particularly including its affluent
Westchester County suburbs and the
North Shore of
Long Island, other terms for the local Transatlantic pronunciation and accompanying facial behavior include "
Locust Valley lockjaw" or "
Larchmont lockjaw", named for the stereotypical clenching of the speaker's jaw muscles to achieve an exaggerated enunciation quality.[14] The related term "boarding-school lockjaw" has also been used to describe the accent once considered a characteristic of elite New England boarding-school culture.[14]
Vocal coach and scholar
Dudley Knight describes how the Australian phonetician William Tilly (
né Tilley), teaching at
Columbia University from 1918 to around the time of his death in 1935, introduced a version of the Mid-Atlantic accent that, for the first time, was standardized with an extreme and conscious level of phonetic consistency.
Linguistic prescriptivists, Tilly and his adherents emphatically promoted their new Mid-Atlantic speech standard, which they called "World English". World English would eventually define the pronunciation of American classical actors for decades, though Tilly himself actually had no special interest in acting. Mostly attracting a following of
English-language learners and New York City public-school teachers,[15] he was interested in popularizing his standard of a "proper" American pronunciation for teaching in public schools and using in one's public life:[16]
World English was a speech pattern that very specifically did not derive from any regional dialect pattern in England or America, although it clearly bears some resemblance to the speech patterns that were spoken in a few areas of New England, and a very considerable resemblance ... to the pattern in England which was becoming defined in the 1920s as "RP" or "Received Pronunciation". World English, then, was a creation of speech teachers, and boldly labeled as a class-based accent: the speech of persons variously described as "educated," "cultivated," or "cultured"; the speech of persons who moved in rarefied social or intellectual circles; and the speech of those who might aspire to do so.[17]
As a phonetically consistent version of Mid-Atlantic pronunciation, World English (known at the time by a
variety of names) was advocated most strongly from the 1920s to the mid-1940s and was particularly embraced during this period in the Northeastern independent
preparatory schools accessible to and supported by wealthy American families. However, the prestige of Mid-Atlantic accents had largely ended by 1950, presumably as a result of cultural and demographic changes in the United States in the aftermath of the Second World War.[18]
U.S. President
Franklin D. Roosevelt, who came from a privileged New York City family, has a
non-rhotic accent, though it is not an ordinary
New York accent; one of Roosevelt's most frequently heard speeches has a falling
diphthong in the word fear, which distinguishes it from other forms of surviving non-rhotic speech in the United States.[46] "
Linking r" appears in Roosevelt's
delivery of the words "The only thing we have to fear is fear itself"; this pronunciation of r is also famously recorded in his
Pearl Harbor speech, for example, in the phrase "naval and air forces of the Empire of Japan".[47]
Decline
After the accent's decline following the end of World War II, this American version of a "posh" accent has all but disappeared even among the American upper classes, as Americans have increasingly dissociated from the speaking styles of the East Coast elite;[13] if anything, the accent is now subject to ridicule in American popular culture.[48] The clipped, non-rhotic English accents of
George Plimpton and
William F. Buckley Jr. were vestigial examples.[49]Marianne Williamson, a self-help author and a
2020 and
2024 Democratic presidential candidate, has a unique accent that, following her participation in the first
2020 presidential debate in June 2019,[50][51][52] was widely discussed and sometimes described as a Mid-Atlantic accent.[53] An article from The Guardian, for example, stated that Williamson "speaks in a beguiling mid-Atlantic accent that makes her sound as if she has walked straight off the set of a
Cary Grant movie."[54]
Theatrical and cinematic use
When the 20th century began, classical training for actors in the United States explicitly focused on imitating upper-class British accents onstage.[18] From the 1920s to 1940s, the "World English" of William Tilly, and his followers' slight variations of it taught in classes of theatre and oratory, became popular affectations onstage and in other forms of high culture in North America. The codification of a Mid-Atlantic accent in writing, particularly for theatrical training, is often credited to
Edith Warman Skinner in the 1930s,[6][55] a student of Tilly best known for her 1942 instructional text on the accent: Speak with Distinction.[3][56] Skinner, who referred to this accent as Good (American) Speech or Eastern (American) Standard, described it as the appropriate American pronunciation for "classics and elevated texts".[57] She vigorously drilled her students in learning the accent at the
Carnegie Institute of Technology and, later, the
Juilliard School.[6]
It is also possible that a clipped, nasal, "all-
treble" acoustic quality sometimes associated with the Mid-Atlantic accent arose out of technological necessity in the earliest days of
radio and
sound film, which ineffectively reproduced natural human bass tones.[58] As used by actors, the Mid-Atlantic accent is also known by various other names, including American Theatre Standard or American stage speech.[55]
American cinema began in the early 1900s in
New York City and
Philadelphia before becoming largely transplanted to
Los Angeles beginning in the mid-1910s, with
talkies beginning in the late 1920s. Hollywood studios encouraged actors to learn this accent into the 1940s.[49] For instance, in the 1952 movie Singin' in the Rain, the Skinner-like elocution coach who entreats Lina Lamont to use "round tones" is attempting to teach her American stage speech.
Although it has disappeared as a standard of high society and high culture, the Transatlantic accent has still been heard in some media in the second half of the 20th century, or even more recently, for the sake of historical, humorous, or other stylistic reasons. Actors working this period who used the accent included
Edward Herrmann,[68]Kelsey Grammer, and
David Hyde Pierce:[69]
In the film
Auntie Mame (1958), Gloria Upson's accent identifies her as a “lockjawed prep princess” from Connecticut's
WASP elite.[70]
Elizabeth Banks uses the Mid-Atlantic accent in playing the flamboyant, fussy, upper-class character
Effie Trinket in the
Hunger Games film series, which depicts enormous class divisions in a futuristic North America.[2]
Mr. Burns,
Sideshow Bob, and
Cecil Terwilliger from The Simpsons all speak with a Mid-Atlantic accent, with the latter two characters voiced by the aforementioned Kelsey Grammer and David Hyde Pierce, respectively.[73]
In the animated television series The Critic,
Franklin Sherman (an affluent former governor of New York) and his wife
Eleanor Sherman both speak with pronounced Locust Valley Lockjaw accents.
Mark Hamill's vocal portrayal of Batman villain
the Joker adopts a highly theatrical Mid-Atlantic accent throughout the character's many animation and video game appearances.[74]
Alexander Scourby was an American stage, film, and voice actor who continues to be well-known for his recording of the entire
King James Bible completed in 1953. Scourby was often employed as a voice actor and narrator in advertisements and in media put out by the
National Geographic Society. His well-refined mid-Atlantic accent was considered desirable for such roles.[76]
Phonology
The Mid-Atlantic accent was carefully taught as a model of "correct" English in American elocution classes,[5] and it was also taught for use in American theatre into the 1960s, after which it fell out of vogue.[77] It is still taught to actors for use in playing historical characters.[78]
A codified version of the Mid-Atlantic accent, American Theatre Standard, advocated by voice coaches like
Edith Skinner ("Good Speech" as she called it) and Margaret Prendergast McLean, was once widely taught in acting schools of the early-mid-20th century.[79]
Trap–bath split: The Mid-Atlantic accent commonly exhibits the TRAP-BATH split of RP. However, unlike in
RP, the BATH vowel does not retract and merge with the back vowel of PALM. It is only lowered from the near-open vowel [æ] to the fully open vowel [a].
No
/æ/ tensing: While most dialects of
American English have the TRAP vowel
tensed before nasals, the vowel is not particularly tensed in this environment in Mid-Atlantic accents.[56]
Father–bother variability: The "a" in father is unrounded, while the "bother" vowel may be rounded, like RP. Therefore, the father-bother distinction exists for some speakers, particularly those following the 20th-century American Theatre Standard in the vein of Skinner, but not necessarily in aristocratic speakers trained before that time or outside of the entertainment industry, like Franklin Roosevelt, who indeed shows a merger.[82] The bother vowel is also used in words like "watch" and "quad".[84]
No
cot–caught merger: The vowels in cot and caught (the LOT vowel and THOUGHT vowel, respectively) are distinguished, with the latter being pronounced
higher and longer than the former, like RP.
Lot–cloth variability: Like contemporary RP, but unlike conservative RP and
General American, Theatre Standard promoted that the words in the CLOTHlexical set use the LOT vowel rather than the THOUGHT vowel.[85][86][nb 1] However, speakers trained before the Theatre Standard, like Franklin Roosevelt, indeed show a LOT-CLOTH split, with the latter aligning to the THOUGHT vowel.[82] The THOUGHT vowel is also used before /l/ in words such as "all", "salt", and "malt".
Lack of
happy tensing: Like conservative RP, the vowel /i/ at the end of words such as "happy" [ˈhæpɪ] (listenⓘ), "Charlie", "sherry", "coffee" is not tensed and is thus pronounced with the SIT vowel [ɪ], rather than the SEAT vowel [iː].[56] This also extends to "i", "y", and sometimes "e", "ie", and "ee" in other positions in words. For example, the SIT vowel is used in "cities", "remark", "because", "serious", "variable".
No
Canadian raising: Like RP, the diphthongs /aɪ/ and /aʊ/ do not undergo Canadian raising and are pronounced as [aɪ] and [ɑʊ], respectively, in all environments.
Back /oʊ/, /uː/, /aʊ/: The vowels /oʊ/, /uː/, /aʊ/ do not undergo advancing, being pronounced farther back as [oʊ], [uː] and [ɑʊ], respectively,[88] like in conservative and Northern varieties of American English; the latter two are also similar to conservative RP.
No weak vowel merger: The vowels in "Rosas" and "roses" are distinguished, with the former being pronounced as [ə] and the latter as either [ɪ] or [ɨ]. This is done in General American, as well,[89] but in the Mid-Atlantic accent, the same distinction means the retention of historic [ɪ] in weak preconsonantal positions (as in RP), so "rabbit" does not rhyme with "abbot".
Lack of
mergers before /l/: Mergers before /l/, which are typical of several accents, both British and North American,[90][91][92] do not occur. For example, the vowels in "hull" and "bull" are kept distinct, the former as [ʌ] and the latter as [ʊ].
American, British and Mid-Atlantic low vowels comparison
In a Mid-Atlantic accent, the postvocalic /r/ is typically either
dropped or vocalized.[93] The vowels /ə/ or /ɜː/ do not undergo
R-coloring.
Linking R is used, but Skinner openly disapproved of
intrusive R.[93][94] In Mid-Atlantic accents, intervocalic /r/'s and
linking r's undergo
liaison.
When preceded by a long vowel, the /r/ is vocalized to [ə], commonly known as schwa, while the long vowel itself is laxed. However, when preceded by a short vowel, the /ə/ is elided. Therefore, tense and lax vowels before /r/ are typically only distinguished by the presence/absence of /ə/. The following distinctions are examples of this concept:
Mirror–nearer distinction: Hence mirror is [mɪɹə], but nearer is [nɪəɹə].
Mary–merry distinction:[56] Hence merry is [mɛɹɪ], but Mary is [mɛəɹɪ]. Mary also has an opener variant of [ɛ] than merry.
"Marry" is pronounced with a different vowel altogether. See further in the bullet list below.
Other distinctions before /r/ include the following:
Cure–force–north distinction: The vowels in cure and force–north are distinguished, the former being realized as [ʊə] and the latter as [ɔə], like conservative RP.
Thought–force distinction: The vowels in thought and force–north are distinguished, the former being realized as [ɔː] and the latter as [ɔə]. Hence saw[sɔː], sauce[sɔːs] but sore/sour[sɔə], source[sɔəs].[95] This does not agree with /ɔː/horse and /ɔə/ for hoarse in traditional Received Pronunciation, but it keeps the distinction observed in rhotic accents like General American.
Hurry–furry distinction: The vowels in hurry and furry are distinguished, with the former pronounced as /ʌ/ and the latter pronounced as /ɜː/. (listenⓘ)
Palm–start distinction: The vowels in palm and start are distinguished, the former being realized as [ɑː] and the latter as [ɑə]. Hence spa[spɑː], alms[ɑːmz] but spar[spɑə], arms[ɑəmz].[96] This keeps the distinction observed in rhotic accents like General American, but not made in RP.
Wine-whine distinction: The Mid-Atlantic accent resists the modern
wine–whine merger: The consonants spelled w and wh are pronounced differently; words spelled with wh are pronounced as "hw" (/ʍ/). The distinction is a feature found in conservative
RP and
New England English, as well as in some Canadian and Southern United States accents, and sporadically across the Mid-West and the West. However, it is rarely heard in contemporary
RP.[97]
Pronunciation of /t/: the alveolar stop /t/ can be pronounced as a glottal stop, [ʔ], only if it is followed by a consonant in either the same word or the following word. Thus grateful can be pronounced [ˈɡɹeɪʔfɫ̩]ⓘ. However, Skinner recommended avoiding the glottal stop altogether; she also recommended a "lightly aspirated" /t/ in place of the
flapped /t/ typical of American speakers whenever /t/ appears between vowels.[98] Likewise, winter[ˈwɪntə] is not pronounced similarly or identically to winner[ˈwɪnə], as it is by some Americans.[nb 2] Generally, Skinner advocated for articulating /t/ with some degree of aspiration in most contexts.
Resistance to
yod-dropping: Dropping of /j/ only occurs after /r/, and optionally after /s/ and /l/.[100][101] Mid-Atlantic also lacks palatalization, so duke is pronounced ([djuːk]ⓘ) rather than ([dʒuːk]ⓘ).[102] All of this mirrors (conservative) RP.
A "dark L" sound, [ɫ], may be heard for /l/ in all contexts, more like General American than RP. However, Skinner explicitly discouraged darker articulations.[103]
A
tapped articulation of post-consonantal or inter-vocalic /r/ is heard in many of the earliest recordings of Mid-Atlantic accents, likely for dramatic effect in
public speaking. Skinner, however, disapproved of its usage.[104]
Other pronunciation patterns
Skinner approved of the -day suffix (e.g. Monday; yesterday) being pronounced as [deɪ] or as [dɪ] ("i" as in "did"), without any particular preference.[105]
Instead of the unrounded STRUT vowel, the rounded LOT vowel (listenⓘ) is used in everybody, nobody, somebody, and anybody; and when stressed, was, of, from, what. This is more like RP than General American. At times, the vowels in the latter words can be reduced to a schwa.[106] However, "because" uses the THOUGHT vowel.
Polysyllabic words ending in -ary, -ery, -ory, -mony, -ative, -bury, -berry: The first vowel in the endings -ary, -ery, -ory, -mony, -ative, -bury, and -berry are all pronounced as [ə], commonly known as a schwa. Thus inventory is pronounced [ˈɪnvɪntəɹɪ], rather than General American [ˈɪnvɨntɔɹi] or rapidly-spoken RP [ˈɪnvəntɹi].[107]
^A similar but unrelated feature occurred in RP. As one attempt of middle-class RP speakers to make themselves sound polished, words in the CLOTH set were shifted from the THOUGHT vowel back to the lot vowel.[87] Also see
U and non-U English for details.
^"The t after n is often silent in [regional] American pronunciation. Instead of saying internet [some] Americans will frequently say 'innernet.' This is fairly standard speech and is not considered overly casual or sloppy speech."[99]
^
abcKnight, Dudley. "Standard Speech". In: Hampton, Marian E. & Barbara Acker (eds.) (1997). The Vocal Vision: Views on Voice.Hal Leonard Corporation. pp. 174–77.
^Riedel, Michael (10 December 2010).
"You don't know Jack (yet)". New York Post. Retrieved 17 May 2022. My dad was born in Queens but affected this mid-Atlantic accent. The old neighborhood accent only came out when he got mad at us.