How to Lose Your Accent in (American) English

by Margaret Magnus

copyright by Margaret Magnus
all rights reserved

My Web Site
Margo's Magical Letter Page


Before I begin, I must warn you that I'm about the least competent person to speak on this subject, because by now there's not a single language I speak without an accent... including the ones I have spoken since childhood. But in periods when I tried really hard, I've pretty much managed to lose my accent in a few of them intermittently off and on... until I got lazy and the whole thing became mush.

Be further warned that this little treatise won't help you at all with your grammar. This is for the legions of you naturalized Americans who actually speak English better than I do, but who have this irritating accent you'd like to be free of. And it's for all you others too... who soon will speak English better than I do. It's a good idea to get the accent right from the start, if that matters to you.

First some general points:

1. Were you over 40 when you first learned to say 'hello' in English? Everybody has told you all your life that it's in principle impossible to lose or improve your accent? Wrong. Losing your accent does not require some kind of genius or magic. You do it like you do anything else... by understanding what is required and by practicing. It takes more practice after you're 40, but it's not impossible, and it's fairly easy in most cases to radically improve.

2. The basic principle is this. Slow down. Everyone can speak American English without any accent at all... Everyone! Right now! If they talk slowly enough.... by 'slowly enough', I mean in most cases saying just one word at a time - each word carefully considered and carefully pronounced.

3. Most people find that when it really gets down to pronouncing that one word just like we Americans do, it's not that they can't pronounce it that way... it's that they don't want to. It's just too embarrassing to actually say that word just the way we really say it. I openly confess that it's pretty ridiculous that a whole nation in all seriousness actually talks like this, but that's the way it is. The point of this whole Web site is that the sounds you pronounce in part define your personality. In order to talk like an American, you have to be American. There's no way around this. So before you go on, ask yourself honestly whether you are truly willing to incorporate the American psyche into yourself on the cellular level. If the answer is no, then sorry. Can't help you. If you feel more comfortable being British, then you could perhaps take that up instead. If you are willing to actually be a native English speaker of some continent, then imagine that you are American or Australian or whatever. It's reasonably safe to do this. You're not lying to anybody... just imagining. It's like acting in a play. After you've pretended for a few minutes, you can always go back and be yourself again... no permanent damage is done.

4. Now listen very carefully to that one word and say it just as you hear it, not as you think it should be pronounced. There are three barriers that adults must overcome in learning a language. One is that the neurons controlling their mouth movements have been programmed to speak another language. This barrier is overcome by repetition. Another barrier is the one just mentioned, that they aren't aware that what's required is to become American, and once they realize this, they don't want to do it... it threatens their self-definition too much. And a third barrier is that they already have notions about how things are pronounced, and they rely on their notions rather than on what they actually hear. So at this point, slow down to one word at a time, and say it over and over again without an accent. Every time you say the word the wrong way, you are programming your neurons the wrong way... so say it as often as you can the right (all-American, apple pie) way. Don't speed up until it's easy to do so without losing your accent. You'll notice that whatever you can't do fast, you can't do slowly either. You'll also notice if you pay attention that you're often simply mistaken about which vowel is actually used... you're thinking it's rounded at the lips like 'o', when it's actually a nice wide 'ah', as in 'hot'.

American Specific:

All the above advice is equally applicable to learning any language. However, in addition to this, there are a few handy tricks they teach you in linguistics school about the differences between English pronunciation and those of other languages.

1a. The vowels are usually the hardest for people. English has a lot of them compared to most languages, so the differences between them is likely to be more subtle than what you're used to. Like the other Germanic languages, English has a series of long vowels and another series of short vowels. Foreigners usually have the most trouble with the distinction between short 'e' (as in bed) and short 'a' (as in bad) and between long 'i' as in 'sheet' and short 'i' as in 'fit'. : - ) The long vowels are mostly diphthongized in English... this is not usually true of other Germanic languages (in case you're German or Swedish). This means that many of our long vowels are a combination of two vowels ­ the last one is 'y' or 'w'. Our long 'a' as in 'fate' has a generous 'y' at the end.

* You can draw the vowel in 'feeeeeeyyyyyyyt' out for a long time (2 seconds) without gaining an *accent*... An accent happens when you pronounce the wrong vowels, not when you pronounce them for too long. When you pronounce them for too long, you just sound stupid, not foreign. And make sure that initial 'e' starts out low enough in the mouth... most languages let it get a little higher than we do, because they don't have to fit both short 'e' and short 'a' into such a small space.
* Draw out also 'shuuuuuuuuuuut' (shoot) - that 'u' is not powerfully velarized in most dialects of American (unlike most languages), which means you don't lift the back of your tongue so high toward the velum... pronounce it much farther toward the front of your mouth - even between your lips and your front teeth. I don't perceive a shift in color for the duration of this vowel (unlike 'fate', it stays the same throughout)
* 'looooowwwwwd' (load) - again the 'o' is not powerfully velarized. Pronounce it in a little space the size of a golf ball behind your front teeth. Most people have to work on the American 'w' as well. It's pronounced lightly... don't purse your lips.
* 'laaaaaaawwwwwwd' (loud) - This 'a' isn't velarized either

1b. Nearly 1/3 the vowels in my dialect of spoken American English are 'uhhhhhh' as in 'duuuuhhhh' or 'ummmmm'. Practically all the unstressed vowels go to 'U'. Before an 'r' it become 'u' like in 'took'. Many dialects (New England) also let unstressed vowels go to 'i' in some positions. attentive -> UtentUv or utentiv, depending on the dialect. combination -> kambUneyshUn, German -> JurmUn..., practically -> praektUkliy, color -> kUlur, unbelievable -> UnbUliyvUbUl. This is coincidentally also the stupidest, dullest and least vibrant of the vowels... speaking phonosemantically... in case you're wondering why we are as we are.

1c. A following 'r' strongly colors our vowels. For me, the vowel in 'for', 'fir' and 'fur' is all the same - basically the same vowel as in 'book'. 'Four' and 'far' each retain their vowels, except in the midwest where 'card' and 'cord' are pronounced the same and 'horse' rhymes with 'farce'. In my (Colorado) dialect 'Mary', 'merry' and 'marry' are all pronounced the same, but in New England, they're all different

1d. In New England and other regions, they have a more rounded 'a', which doesn't exist in the West. For me 'cot' and 'caught' are pronounced the same. Not so out here in the East.

2. Now some words about consonants... Linguists distinguish different kinds:

labial (pronounced at lips): b,p,v,f,m,r(initially),w
dental (pronounced just behind the teeth): th (thing) and th (that)
alveolar (pronounced in English behind the alveolar ridge): d,t,j,ch,z,s,n,r
palatal (flatten the tongue in the middle of the mouth): sh,l,y
velar (pronounced at the velum): g,k,h,ng

stop (fully stops the airflow in the mouth): b,d,g,p,t,k
fricative (air flows through a narrow opening in the mouth): v,th,z,f,th,s,sh,h
affricate (stop and fricative at the same time): j (d+zh), ch (t+sh)
nasal (air flows through the nose): m,n,ng
liquid (historically always mixed up): l,r
glides (vowels used as consonants): w,y

voiced (pronounced while using the voice): b,d,g,v,th(this),z,j,m,n,ng,r,l,w,y
unvoiced (pronounced while voice is still): p,t,k,f,th(thing),s,sh,h,ch

3. What I have called alveolars here are dentals in Slavic and Romance and many other language families. That is, they are pronounced in Slavic by touching the tongue to the back of the teeth. When you pronounce them in English, the tongue does not touch the teeth, but rather hits just behind the alveolar ridge (that bump in the roof of your mouth just behind the teeth).

4. The unvoiced stops in American are aspirated at the beginning of a syllable. This is not true in Slavic or Romance. So there's a slight puff of air after the 'p' in 'pot' and 'prize', the 't' in 'top' and 'train', and the 'c' in 'cat' and 'climb'. This is not true of the unvoiced stops in 'stop', 'spill', 'ship', 'skip',... After 's' and at the end of the syllable, the sound remains unaspirated. The aspiration in English is much lighter than in the Indian languges, at it's limited to unvoiced stops.

5. That brings us to the 't' at the end of 'fate'. The 't' doesn't even have to be fully pronounced. If you want, you can just let the tongue hit the top of your mouth... or you don't even have to do that. 't' at the end of a word like that can in many dialects even be pronounced with a glottal stop way at the back... like the first consonant in Hebrew 'Aleph. The glottal stop precedes all words in English that start with a vowel too... we just don't perceive it as an actual consonant like the Semites do because its distribution in our speech is completely predictable.

6. When you have two stops in a row in American English, you don't let go of the first one before you pronounce the second. In French or Norwegian, for example, you pronounce a full 'c/k', then let it go and then start the 't' in 'ct'. But in the American word 'act', you don't let go of the closure of the 'c' before you start the closure for the 't'. Your tongue is actually touching both the velum and the alveolar ridge at the same time.

7. Dentals, labials and palatals in most languages is usually velarized more than in American English. Don't lift your tongue up in back so much. This makes the sounds 'lighter', higher pitched. 'l' is an exception. 'l' for me is quite heavily velarized. You Slavs have a distinction between 'hard', velarized consonants and 'soft' palatalized consonants. We don't have that distinction. All our consonants are sort of in the middle between these two.

8. In a lot of languages, particularly Slavic, voiced sounds are pronounced with stronger voicing than in American. In Slavic, you start moving your vocal chords before you even start articulating the 'z' or 'v'. In American, you start articulating the 'v' or 'z' before the vocal chords start moving. You Russians... have your American friends pronounce the Russian word 'Zdravstvyujte' for you. See if their lame 'z' in the beginning doesn't sound almost like an 's' to you.

9. In a great many languages, there can be no voiced consonants at the end of the syllable. This is not true in English. 'Log' is different from 'lock', 'bid' is different from 'bit', 'raze' is different from 'race'. Be especially careful of those plurals and 3rd person singular verbal endings. Most of them are spelled 's' but actually pronounced 'z'... anything after a vowel or another voiced consonant is pronounced 'z', not 's' - 'has', 'watches', 'bugs', 'news', 'pans', 'initiations'.

10. The American 'r'. When you guys make fun of our pronunciation, you overdo the 'r'. It is pronounced with rounded lips in initial position, but not final position. And don't velarize it so much. Our 'r's and our vowels vary the most dialectally, so you have to decide what version of American you want to speak.

11. 'th' is pronounced behind the teeth, where most languages pronounce 'd' and 't'. It's not achived by sticking your tongue between your teeth.

12. 'd' and 't' between vowels in American are pronounced the same. It's what we call a 'flap', in some ways similar to one strike against the roof of the mouth of a rolled 'r'. It's a voiced sound. 'Bitter' and 'bidder' are pronounced the same. It also happens in speeach between words "What did you do?" becomes 'wUDijUdu?' The final 'd' there is not pronounced as a flap, because it comes at the beginning of a stressed syllable.

13. Once you've got individual words down cold, start listening to the intonation. American rises and falls less dramatically in normal speech than most languages... except when we get emotional.

That's all I can think of for now.