All articles · Published 2026-10-23

What is the Transatlantic accent? Old Hollywood's voice was taught to Americans by an Australian and a Canadian

Watch a Hollywood film from the 1930s or '40s and the actors sound strange: no American r, a whiff of England, yet unmistakably American. This is the Transatlantic accent, also called the Mid-Atlantic accent — "the accent from the middle of the ocean" — and nobody ever grew up speaking it naturally. It was taught in classrooms: an Australian phonetician designed it, and a Canadian wrote the textbook. Here is where it came from, how to spot it, why it nearly vanished after World War II, and the most useful lesson it holds for TOEIC learners: there is no such thing as a neutral accent.

Why do actors in old movies talk so strangely?

Put on The Philadelphia Story (1940). The moment Katharine Hepburn opens her mouth, something sounds off. Car comes out as "cah" — the r at the end is gone. The vowel in can't is broader than an American's, closer to a Brit's. Every consonant lands crisply, as if she were talking to the back row of a theater.

No need to imagine it. Here is Hepburn herself:

(English subtitles are on by default; if the embed won't play, watch it on YouTube.)

Yet she was American: born in Connecticut, educated at Bryn Mawr College in Pennsylvania. Cary Grant, in the same film, speaks in much the same way — and he was English, born Archibald Leach in Bristol in 1904, who came to America as a teenager. An American and an Englishman, speaking one accent on screen, and the accent belonged to neither country.

That is the Transatlantic accent. It is also called the Mid-Atlantic accent — "mid-Atlantic" as in halfway between Britain and America, as though it had been picked up on an ocean liner crossing between the two.

But ocean liners don't grow accents. This one was designed and taught.

Who invented the Transatlantic accent?

An Australian: William Tilly

The story starts with William Tilly, an Australian-born phonetician. From 1918 he taught phonetics in the extension program at Columbia University in New York, arguing that the English-speaking world needed one standard pronunciation that rose above local accents. He called it "World English." To his mind, a strong regional accent was a mark of provincialism.

His standard took Received Pronunciation — RP, the upper-class English accent of old-school BBC announcers — as its base, with some American adjustments. The key point is that every vowel and every consonant in it was precisely specified. It could be written down as rules and drilled.

A Canadian: Edith Skinner

Tilly's most important student was Edith Skinner. She was born Edith Warman in 1902 in Moncton, New Brunswick, Canada, and in 1926 she became Tilly's assistant.

In 1942 Skinner published Speak with Distinction, turning the system into a textbook for actors. She called it "Good Speech," or "Good American Speech." Her own definition, roughly: a North American English free of regional features — recognizably North American, yet suited to classical texts, effortless to articulate, and easily understood in the last row of a theater.

She spent decades teaching stage speech at the Carnegie Institute of Technology (today's Carnegie Mellon University), and class after class of her students went on to Broadway and Hollywood. The book is still in print; a revised edition came out in 2007.

So here is the lineage: an Australian invented it, a Canadian wrote the textbook, and Americans were taught to sound a little bit English. At the source, there isn't a single American.

How do you recognize a Transatlantic accent?

Taking Skinner's Good Speech as the reference, four features are the easiest to hear:

Feature Transatlantic General American today Example words
r after a vowel silent (as in England) pronounced car, park, mother
the a in "ask" words broader, between American cat and British father same vowel as cat ask, dance, can't, half
t between vowels a clean t often a soft d better, water, city
words starting with wh a puff of breath first; which and witch differ the two sound the same which, what, when

The first feature is the most important and the best documented. In the Atlas of North American English (2006), the linguists Labov, Ash and Boberg write that dropping the r, as a characteristic of British RP, "was also 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.

The second feature shows the artificiality plainly. British RP says ask and dance with the broad a of father; Americans use the a of cat. Skinner chose neither. She prescribed a vowel in between, and supplied a complete "Ask-List" telling you which words take it and which don't. An accent you need a lookup table to speak did not grow up on its own.

Why did old movie actors all talk like that? Was it the microphones?

A popular explanation online goes like this: early microphones and radios were so poor that performers had to enunciate unusually clearly, and the accent grew out of that.

It sounds plausible, but we could not find any linguistic research supporting it. And it has an obvious hole: the same accent was taught in theaters and classrooms too. A stage has no microphone, and Tilly's phonetics classes were not designed for recording.

A social explanation fits better. In early-twentieth-century America, an upper-class English accent signaled education and culture; the classical repertoire was Shakespeare, and stage actors already imitated English speech. Actor training, elocution schools and radio stations all wanted a "standard," and Tilly and Skinner offered one that was clearly written down and could be practiced. When Hollywood switched to sound in 1927, it suddenly needed a great many actors who could "speak properly" — and this accent was ready and waiting.

Besides the taught version, the upper class of the American Northeast had a similar inherited way of speaking, handed down through New England boarding schools. President Franklin D. Roosevelt went to Groton School in Massachusetts, and his "fireside chats" on the radio were delivered in that r-less elite accent. The actors paid to learn it; Roosevelt simply grew up talking that way.

Why did the Transatlantic accent disappear?

After the war it retreated fast. As the Atlas of North American English puts it, the accent was taught up to the end of World War II — after that, speech schools stopped treating it as the standard, and the next generation of students simply didn't learn it.

Film tastes changed too. In the early 1950s Marlon Brando played A Streetcar Named Desire with "the Method": mumbled lines, a street-level accent, a voice that sounded like a real person. Next to him, the crisp Transatlantic accent suddenly sounded affected, old-fashioned and distant.

Skinner's method was later criticized from within her own profession. In his 1997 essay "Standard Speech: The Ongoing Debate," the drama professor Dudley Knight argued that "Good American Speech" was elitist at its core: calling one class's accent "correct" amounts to calling everyone else's incorrect. The essay stirred up the voice-teaching world for some time.

Katharine Hepburn, for her part, never changed. In 2004 Cate Blanchett played her in The Aviator, accent and all — and won the Academy Award for Best Supporting Actress. Today you mostly hear the accent when someone is imitating old movies or deliberately playing an upper-crust character.

Careful: "Mid-Atlantic" has another meaning

If you look up "Mid-Atlantic accent," you may find something completely different.

In the United States, "the Mid-Atlantic" is also a geographical term for the middle stretch of the East Coast — New York, New Jersey, Pennsylvania, Delaware, Maryland. When linguists talk about Mid-Atlantic English, they often mean the regional accent of Philadelphia and Baltimore — a real, home-grown accent, and almost the opposite of the old-movie voice.

One "mid-Atlantic" means the middle of the ocean; the other means the middle of the coast. That is why some people prefer "Transatlantic accent" for the Hollywood sound, to avoid the mix-up.

What does this have to do with the TOEIC? There is no neutral accent

Tilly dreamed of a pronunciation of English with no local features, usable all over the world. The way that dream failed is instructive for anyone learning English today.

First, "accentless" English is always somebody's accent. The Transatlantic accent claimed to be neutral, but underneath it was upper-class English with an American seasoning; the pronunciation now treated as "standard American" is likewise the accent of particular regions. Everyone who speaks English has an accent. The only questions are whose accent it is, and whose accent counts as "correct" at the time.

Second, the TOEIC doesn't aim for a single accent at all. The TOEIC listening section is recorded in four real accents: American, British, Australian and Canadian. ETS's public official sample test labels every line of its listening script with the speaker's accent — W-Am, W-Br, M-Au, M-Cn. The test doesn't check you against a "standard accent" that doesn't exist; it checks whether you can understand different people in the real world.

Third, when you speak, the goal is to be understood, not to pass as someone else. Hollywood actors spent years drilling Skinner's Ask-List because it was their job. You don't need to say ask like an Englishman, and you don't need to pretend you're from California. Clear, consistent, and understood by the listener — that is good English.

As usual, a few caveats

The sources are thinner than we'd like. The history of the Transatlantic accent lives mostly in books and articles on actor training; rigorous linguistic research on it is scarce. Tilly's and Skinner's biographies, the line from teacher to student, and the publication date of Speak with Distinction are where accounts agree. Claims about who taught whom, or which actors learned the accent and which came by it naturally, are often just lore; we have only included what we could trace to a source.

We have not proven the microphone theory wrong. All we can say is that we found no research supporting it, and that the accent was taught in places with no microphones. Early recording technology may well have rewarded crisp diction — it just isn't where this accent came from.

"Artificial" does not mean "fake." For educated Northeasterners of Hepburn's or Roosevelt's generation, this was simply how they talked. An accent that was taught is not so different from any other accent that was learned — everyone picks up their accent from the people around them.

The TOEIC won't ask you to identify accents. This post is about the history of an accent, not about handling any particular test question — and you won't hear the Transatlantic accent on the TOEIC.


Accent Compare in Handy 990 plays the same sentence in American, British and Australian English, and Accent ID asks you to name the accent from a single clip — the point isn't which accent is correct, but understanding all of them. For the words that differ between Britain and America, read British vs. American English vocabulary; for spelling, read British vs. American spelling.

← All articles