I'm not a native speaker. However, I have tried a lot during last 10 years to learn English at a high level of proficiency and to become fluent in conversation.

However, when I talk to some of my friends in US over Skype (found via my profession), they tell me that I talk like a foreigner. But they don't know why is that.

What is/are the reasons that can make a not-native English speaker sound like a foreigner? Is it about pronunciation, or about syntax, or is it maybe just a false assumption coming from a psychological background due to the fact that the other party knows that you are not an English man/woman.

Update: I asked this question specifically about English, since English is the most widespread language on the earth and tens of different dialects and millions of different idiolects seem to alleviate the problem of "sounding-foreign".

  • 6
    I find it is pronunciation that makes some one sound foreign, and in particular the accent and prosody. One can have impeccable grammar, extensive vocabulary, fluid execution, even the 'th's and dental flaps right, but the intonation and stress and 'character of vowels just slightly off and the latter says immediately non-native (or from Boston). – Mitch Sep 24 '11 at 13:13
  • 3
    If you add a recording of yourself, we can pinpoint the details. Most likely your pronunciation has just a few more differences than the expected differences of dialects. – Paul Dexter Sep 24 '11 at 13:21
  • being pedantic on the update: in what was is English the most widespread language on earth? It definitely isn't by native speakers, although I guess it could be depending on how you count foreign speakers. – Artem Kaznatcheev Sep 24 '11 at 18:35
  • 1
    @ArtemKaznatcheev. I suspect that English is indeed the most "widespread", in having large communities of native speakers all over the place. – TRiG Sep 24 '11 at 21:36
  • I would think English is most widespread in that it's the language with the most speakers in general, of varying levels of fluency/proficiency. I mean, if you asked every person on earth which languages they spoke, English would be mentioned the most. Of course, no-one can actually do this, so I can't prove it :D – Jürgen A. Erhard Sep 25 '11 at 10:16

10 Answers 10


There are so many ways that a person's speech can indicate that he is not a native speaker of English (or that he is a speaker of English from certain region). Basically, accents happen when you are using non-English rules (can be phonetic, syntax, idiomatic, etc.) when speaking English. Many speakers do not realize that they are using rules from their own language that are not compatible with the rules of English, since they assume the rules to be true. Because non native speakers often don't notice this, it is very hard to fix unless they practice with a native speaker and have him/her correct them.

Some examples

  • Phonemes. One's native language has a different set of phonemes than English, and some English phonemes are not available in one's language. For example, Japanese doesn't have the sound 'l' (they use something like an 'r' instead), French doesn't have 'th' (use 'z' or 's' instead) and a lot of English vowels are not available in French (so they merge them with the available vowels)

  • Syllable pattern. For example, in Japanese a syllable can begin with at most one consonant and must end in a vowel, so they pronounce girlfriend as "garufurendo", etc.

  • Pronunciation rules. For example, French don't pronounce the letter 'h', etc. Plus English is notorious for very unpredictable pronunciation rules, it is often hard for non native speakers to use the right pronunciation naturally.

  • Intonation pattern. This is also hard to get right because it is unpredictable for non native speakers, other languages' pattern are often really different from English, and people often don't realize that they are speaking with a non native-like intonation pattern

  • Syntax. For example, speakers of isolating languages tend to forget using the right inflection (plural, past tense, etc). After you learn for a while, you will not make such simple mistakes, but non native speakers often don't get the finer point of the syntax, like what hippietrail noticed

  • Idioms

  • Choice of words

There are many other ways you can do it "wrong", so the best suggestion is to speak with a native speaker who knows about these things and have him/her correct you.

  • 1
    Minor correction: Japanese "r" is a (possibly lateral) apical postalveolar flap [ɽ], which I wouldn't say is any more like an alveolar approximant [ɹ] than a lateral one [l]. Also it has a syllabic [n], so it'd be garufurendo. – Jon Purdy Sep 25 '11 at 0:38
  • @jonpurdy please go ahead and edit it, if you don't mind.. I'm on the go right now – Louis Rhys Sep 25 '11 at 7:15
  • Yes the Japanese sound is somewhere between the English l and r but to English speakers it sounds like a kind of r. In Chinese it's the other way around but hey English speakers always confuse Japanese and Chinese accents despite how utterly different they sound \-: But yes garufurenudo should be garufurendo. I'll fix it... – hippietrail Sep 25 '11 at 8:38
  • @hippietrail: Chinese has both [l] and [ɹ], but [l] only occurs syllable-initially and [ɹ] syllable-finally. But yeah, Japanese and Chinese accents sound completely different; I'd expect people to confuse Japanese and Korean more. – Jon Purdy Sep 25 '11 at 17:10
  • 1
    @JoePineda: Yes the Japanese "r", the Spanish "r", and the inervocalic "t" and "d" are all the same sound (though some slight differences might be amplified in some analyses?) but Japanese speakers will use the same sound when they need to pronounce English "l", which Spanish speakers will not do. – hippietrail Feb 28 '14 at 9:28

Here are a few things in your English from your question that signal you as non-native for me:

  1. *during last 10 years -> during the last 10 years
  2. *some of my friends in US -> some of my friends in the US
  3. *But they don't know why is that -> But they don't know why that is
  4. *not-native English speaker -> non-native English speaker.

Many native English speakers might also do some of these at least as typos or thinking faster than they write, but not 3. and probably not 4. Taken together I would expect you were a foreigner. My first impression from 3. is that you might be a French speaker.

So even without pronunciation your syntax would give you away, as would any and all the many points where two languages can differ. (Vocabulary, pronunciation, syntax, exceptions to rules, intonation, contractions, idioms, etc, etc)

(You seem to have excellent English by the way)

  • Looking at his name and location, I don't think he's a French speaker. Arabic seems more likely. – Peter Olson Sep 25 '11 at 2:54
  • 3
    I saw the Arabic name but wanted to include my instinct whether it be right or wrong just as an interesting illustration. But many Arabic/French bilingualism is very common especially in Northern Africa and France and sometimes when speaking your third language you get the accent of your second language rather than your first! Which might be a good idea for another question (-: – hippietrail Sep 25 '11 at 8:35
  • 3
    Initially this question had other problems that I have edited. See the revision english.stackexchange.com/posts/40198/revisions – Theta30 Sep 26 '11 at 8:03
  • 1
    @BogdanLataianu Ironically, in this case editing his question made it less obvious what his actual issue was, I fear! :) – Mark Beadles Jan 22 '12 at 1:05
  • 1
    I am a native Russian speaker, and I also several times had made this mistake (3rd one). This is not because it is a Russian pattern but rather a hypercorrection. Because in English classes the teacher said in all questions in English the verb goes first. I only recently learned it is not always the case and constructions like "why that is" are possible: ell.stackexchange.com/questions/36533/… In Russian I definitely would put the verb at the end in this sentence. – Anixx Oct 16 '15 at 6:18

I've also heard about some research by Chilin Shih (but can't find it on her website) that suggests that highly proficient speakers of a second language don't do as much phonetic reduction as native speakers do. So to some native speakers, highly proficient L2 speakers sound like they're hyper-articulating a little.

  • Since sandhi rules are never a part of English courses, we never know which reductions are even possible :( – Joe Pineda Feb 28 '14 at 6:22

I'm afraid I cannot cite sources, but I was once told that one key to sounding fluent is to make sure you practice getting the "fillers" correct - the umms and errs in English are quite different from French or Spanish fillers. Fillers are sometimes neglected and it is easy to relapse into a native filler when speaking a foreign language.

Some of the information contained in this post requires additional references. Please edit to add citations to reliable sources that support the assertions made here. Unsourced material may be disputed or deleted.

  • Agree! And not just fillers: In Mexican Spanish we say 2 "m" sounds in succession both for "yes" and "no", depending if tone rises or falls - very similar to English "mm-hmm". Brazilian Portuguese speakers use a short, medium pitch "m" to indicate "what?" (we'd use a medium, high-falling pitch for that) whereas Japanese would use a medium pitch, short "m" to signal an informal "yes!" No course or book teaches you that, must deduce meaning by context with plenty of potential for misunderstanding. – Joe Pineda Feb 28 '14 at 6:20

There are some phenomena at play here.

First, yourself. As others have mentioned, there are some markers that will undeniably identify a person as a foreign speaker, such as pronunciation. As children, we have the possibility of learning how to pronounce most consonants and vowels from most languages. Some sounds (the English th for example) are only mastered later. As we grow older, we lose that ability and remain "stuck" with a limited range of sounds we can reproduce, even when we learn a new language.

Secondly, your interlocutors. While all the words you use can be English, and your structure can be the correct grammar, there are levels of formality in every language which second language speakers do not always master. This is where sociolinguistics step in. There is a "known social norm" for speech in certain situations, and native speakers are keen to those, albeit unconsciously. It is not bound by fixed lexical or grammatical rules, it's hardly quantifiable, but it exists, native speakers have a "culture of speech" that is part of language and yet cannot be taught directly.

I get it too. I'm a foreign speaker myself, and people tell me I talk too formally. Even when I try and play it down it's still there.

  • Do you have any references for this "being stuck"? Because I doubt it, and I think a subset (majority/minority, I don't know) of linguists disagree. NB: I don't say this because I like demanding references, I'm just curious what the scientific community thinks. – Jürgen A. Erhard Sep 25 '11 at 10:10
  • 1
    @JürgenA.Erhard I've got one part, that is, that children start with the ability to learn sounds from all languages: R. Jakobson, Kindersprache, Aphasie, und Allgemeine, Uppsala, Sweden: Almqvist and Wiksell, 1941. (English translation by A. Keiler. 1968. Child language, Aphasia, and Phonological Universals. The Hague: Mouton). That itself found in V. Fromkin, R. Rodman, An Introduction to Langage - Fifth Edition, Orlando, FL, USA: Harcourt Brace & company, 1993. – MPelletier Sep 25 '11 at 15:39
  • 1
    @JürgenA.Erhard, Patricia Kuhl has done a lot of work on perceptual narrowing, some of which can be found on her website: bit.ly/qrHemw, and some of which she covered in her TED talk: on.ted.com/9iZR – JoFrhwld Sep 26 '11 at 4:43
  • 1
    I've seen dozens of Spanish learners suffering with thrilled r's to no avail. Its usually the last or one of the last sounds children master, so it's hard for everyone. – Joe Pineda Feb 28 '14 at 6:13
  • 2
    I've seen some Spanish learners thrilled when they could finally produce their first rolled r (-; – hippietrail Mar 1 '14 at 7:01

I work with a large number of non-native English speakers, and probably the biggest indicator of fluency for me is the use of articles (a, an, the), which is what Hippie Trail's answer covers.

This is entirely observational, I do not have research to support my findings.

  1. Definite articles ("the") are missing. I think this is because in many languages, definite articles communicate an extra level of meaning, or maybe have more defined rules for usage. In English, It's easy to know that you must use "the" with a superlative ("The fastest"), but it's hard to figure when to use it with a proper noun ("The United States"). Especially when other languages have more rigid rules about article usage with proper nouns.
  2. Indefinite articles ("a", "an") are missing. I suspect this is because in many languages the indefinite article can be assumed.
  3. Articles instead of determiners on mass nouns . I recognize this a great deal in American vs. British English: "a cup of coffee" vs."a coffee". American English doesn't allow you to say, "I need a coffee". We say, "I need a cup of coffee" or "I need some coffee".

Here's my response again, but, with the articles modified:

I work with large number of non-native English speakers, and probably biggest indicator of fluency for me is use of articles (a, an, the), which is what Hippie Trail's answer covers.

This is entirely observational, I do not have some research to support my findings:

  1. Definite articles ("the") are missing. I think this is because in many languages, definite articles communicate extra level of meaning, or maybe have more defined rules for usage: In English, It's easy to know that you must use "the" with superlative ("The fastest"), but it's hard to figure when to use it with proper nouns ("The United States"). Especially when other languages have more rigid rules about article usage with some proper nouns.
  2. Indefinite articles ("a", "an") are missing. I suspect this is because in many languages, indefinite article can be assumed.
  3. Articles instead of determiners on mass nouns . I recognize this great deal in American vs. British English: "a cup of coffee" vs."a coffee". American English doesn't allow you to say, "I need a coffee". We say, "I need a cup of coffee" or "I need some coffee".

See? Sounds more...non-native.

Everyone else here has offered a ton of great answers, and I agree with them whole-heartedly. I think a fascinating area of research would be determining native language by misuse of articles in English.

  • "I need a coffee" is fine in American English. – hunter Oct 12 '15 at 19:53
  • What region of the US are you from? In the midwest and the south, "I need a coffee" is not common. Native speakers would say, "I need a cup of coffee". – paceaux Oct 12 '15 at 20:56
  • I am a native speaker and from the south. "I need a coffee" is something I say all the time, or "let's get a coffee," or ... – hunter Oct 12 '15 at 21:17
  • dicussed these with some friends just now. we agreed that nobody would say "i need a wine" or "i need a tea" but for coffee and beer these are acceptable. at any rate +1 on your response – hunter Oct 12 '15 at 21:43

I think the biggest hurdle on the way to getting taken for a native speaker is intonation. That's the hardest part. You can have a big vocabulary; you can learn that from books easily. You pronounce each and every single word correctly (in isolation). Also easily to be learned from books (every dictionary has the proper pronounciation listed... at least for the difficult or unusual words).

And still, if your intonation is wrong, it's a dead giveaway. And most native speakers would probably not be able to say what made you sound foreign, since all you pronounce all the words correctly. (Okay, I'm assuming this from your description)

How can you learn intonation? Not from books. Only from listening (a lot), and then practicing what you heard.


This touches me on a personal level for I've tried hard my whole life to get rid of my accent, only to be told time and again I still have a strong Hispanic accent. I've learned a few interesting things on the road, though:

  • Rhythm. Syllables of Spanish are all more or less same length and at a semi-constant tempo. English tends to accelerate the tempo for unstressed syllables and slow it down a bit for stressed ones. Other languages have their own rhythmic and mellodic-line patterns which differ from English ones.

  • Sounds. Consonants have allophonic variations depending on environment. And most of the 10+ vowels of English lack equivalents in most other European languages. Nothing really prepares you to utter them save tons of practice. And then, there's a lot of variation in the realizations of some vowels in different dialects. Plus, in most dialects they get blurred in unstressed positions, which takes me to...

  • Sandhi. All languages have it, but English has a moderately complicated one that's usually not represented in its written form (in comics or informal texts it is sometimes, but not consistently). Sounds, both written and not, interact with their neighbours not just within a word but also at the boundaries of words, mutating into others.

  • Über-naïf native speakers. Most of them are not aware of the different pronunciations they use - some even are unable to tell 2 sounds appart even though they clearly make the distinction when speaking. Of course they do can detect when you're sounding "weird" or "better" - but can't state why :(

  • 1
    I've spoken Spanish for about half of my life; working with Spanish speakers has taught me a few things about English fluency with regard to rhythm. Romance languages (especially Spanish) have both a rhythm and a melody. Spanish has strong rules for where the emphatic syllable is in a word, and the written language indicates when the word breaks the rule. English doesn't do that at at all. I've never met a non-native speaker who could figure out how to pronounce "hippopotamus" on their own. – paceaux Dec 12 '14 at 17:59
  • 1
    Spanish is a syllable-timed language; every syllable takes about the same amount of time to pronounce, producing a steady rhythm of syllables. English, by contrast, is a stress-timed language, where the rhythmic unit is not the syllable, but the time between two successive stressed syllables. So, HE's a DOCtor, HE's a good DOCtor, HE's a very good DOCtor, and HE's an extremely good DOCtor all take about the same amount of time to say. The result is that unstressed English syllables are reduced almost to nothing at normal speech rates. – jlawler Nov 16 '15 at 15:49

The vast majority of people who learn English, even to very high levels, still sound 'foreign' because they do not employ the rules of connected speech.

See the following link:


  • The way I see it, shouldn't these "connected speech" rules be considered special forms of sandhi? – Joe Pineda Feb 28 '14 at 6:06
  • Yes, and that's what linguists often call it. But they studied Sanskrit so they already know the term. "Fast speech rules" or "rules of connected speech" are more descriptive. – jlawler Nov 16 '15 at 15:43

OP has "they don't know why is that" which is not a natural word order to me as UK Southern-received. More expected, if not 'natural', would be "they don't know why that is.". This would suggest to me that the OP has some gaps in syntatical fluency which may make native speakers aware that they are dealing with a non-native, thus increasing the likelihood that accent and pronunciation generally would be picked up as variant. All in all, though, I think the OP should stick with it and get exposed to as much native speech and writing as possible to clear away these few remaining issues.

  • 2
    That point was already noted in @hippietrail's answer, so this answer does not add anything new to the existing answers. – jknappen Sep 23 '18 at 20:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.