"Amarillo by Morning" sung by a Mongolian

« previous post | next post »

Mongolian gets 97 points for singing "Amarillo by Morning" on US TV show but didn't understand a word he was singing. His pronunciation was perfect.

[VHM:  The YouTube video linked to here is currently unavailable, but our resourceful Language Log readers have elsewhere found this song sung by Enkh Erdene and others by him as well, some of them captioned.  See the comments below.]

Title on YouTube:  "Enkh Erdene: Monglian Singer With ELVIS Voice! Meet The Monglian Cowboy | World's Best 2019"

"Mongolian Singer Nails A Classic Country Song – Stunning", by Dale Mussen, WYRK (2/4/19):

Last night we were introduced to a singer from Mongolia who doesn't speak a word of English.  He had an interpreter on stage with him as he spoke with show host James Corden.  Then it was time for him to sing and I couldn't believe what I was hearing.  And we found out he doesn't really know the meaning of the words he's singing, but considering where he's from his performance is stunning.

Brought lots of tears to my eyes.  Incredible performance.

[h.t. Bathrobe]

[Update:  See now also a side thread on this in the comments to "Massive borrowing" (2/18/19).]


  1. Rube said,

    February 4, 2019 @ 12:46 pm

    Back when the song was popular, I had a friend, a native English speaker, who thought it was "Amaryllis" by Morning.

    So, a Mongolian not getting it doesn't astonish me.

  2. Martha said,

    February 4, 2019 @ 12:55 pm

    He never got curious and looked up a translation or asked someone?

    Even if I didn't understand it word for word, I'd at least want to know the gist of the song.

    It makes me wonder if he (or the interpreter) misunderstood Corden's question, when he asked if he knew what he was singing about, which I took to be a question about whether he understood the gist, rather than understanding the words, as the South African panelist asked.

  3. Laura Morland said,

    February 4, 2019 @ 12:56 pm

    So… how did he achieve his "perfect" accent? From a linguistic point of view, that's what I'd like to know.

  4. J.W. Brewer said,

    February 4, 2019 @ 1:24 pm

    I assume a fair number of opera singers can and do perform phonetically-memorized arias in languages in which they can't successfully order lunch or check in to a hotel. Is this qualitatively different than that, or does it just seem novel because of the genre of music and/or the fact that there's a whole accent-coaching infrastructure built into the professional training of opera singers whereas this guy may be more of an autodidact? I suppose it might be the case that this C&W style of singing is less abstracted from an ordinary conversational tone in West Texas English than the operatic style is, and thus perhaps requiring more nuance for a non-native-speaker to get right? I guess one practical question might be how many new or different phonemes would someone starting with native fluency in Mongolian need to master from scratch in order to sing this particular text with a plausibly American accent?

  5. Miles Archer said,

    February 4, 2019 @ 1:38 pm

    His accent isn't perfect at near the end. He tends to pronounce it amareeyo almost like it's a Spanish name or something.

    The singing, however, is spot on.

    Previously, I saw a video of a chinese guy singing Take Me Home Country roads. His singing was great, accent was spot on, but didn't know any English. The video has been taken down. It's funny though, I went to Shanghai with a co-worker from West Virginia and the first thing anyone said when they heard where she was from was Take Me Home, Country Roads.

  6. Thaomas said,

    February 4, 2019 @ 2:28 pm

    I found "Country and Eastern" :) music quite popular i Indonesia a few years ago.

  7. AntC said,

    February 4, 2019 @ 2:59 pm

    I assume a fair number of opera singers can and do perform phonetically-memorized arias in languages in which they can't successfully order lunch or check in to a hotel.

    Er, you assume wrong. If you see interviews with opera singers, they'll say how they always "learn their languages". Now maybe they only learn enough for the words in the opera, a few polite phrases, and how to check in to a hotel; but they want enough to cover the cadence and sound-pattern.

    There I'm talking about opera singers. Professionals.

  8. David Morris said,

    February 4, 2019 @ 3:17 pm

    If he doesn't understand a word, then how does he know that the song is anywhere appropriate for him to sing on a tv show?

  9. DCBob said,

    February 4, 2019 @ 4:36 pm

    His pronunciation is by no means perfect, but it's remarkably good. I think it's pretty clear that he doesn't know what all the words mean, but I have the sense that he understands the gist of it. Makes sense that a young man from Mongolia would like a great rodeo song.

  10. Ian said,

    February 4, 2019 @ 4:44 pm

    From my time in Mongolia I realized that basically all Mongols are really great at singing. I heard tons of people singing from the karaoke bars of UB to horseback in the countryside and everybody had amazing voices. I think it's genetic. Doesn't explain the pronunciation, but interesting point nonetheless.

  11. Jerry Friedman said,

    February 4, 2019 @ 6:59 pm

    David Morris: If he doesn't understand a word, then how does he know that the song is anywhere appropriate for him to sing on a tv show?

    Maybe he's seen it on TV shows. And I don't know anything about the show in question, but I wouldn't be surprised if the performers have to clear their songs with the producers. I assume he can pronounce and probably write the title.

  12. Bill Benzon said,

    February 4, 2019 @ 7:12 pm

    Alas, that video seems to have disappeared, at least for me. But I was able to find his performance of "Don't be cruel". I was excellent, but the pronunciation was a bit off here and there. I also found a song that he sang, as winner of "Mongolia's Got Talent" on "Got Talent Global" for 2016 and, I assume, he was singing in his native language. But the idiom was thoroughly Western, big band derived with a very heavy back beat.


  13. Victor Mair said,

    February 4, 2019 @ 8:41 pm

    The video appears to have been removed everywhere, even from the WYRK article where I first saw it. His agents are thorough, if nothing else.

  14. John Swindle said,

    February 4, 2019 @ 9:21 pm

    It's available on YouTube at
    as I write. YouTube is however trying to feed me some orange-haired jerk's anti-immigration ads as a price for watching. I hope their inscrutable process will feed you better ads.

  15. david said,

    February 4, 2019 @ 9:21 pm

    Amarillo by morning


  16. david said,

    February 4, 2019 @ 9:22 pm

    Amarillo by morning via youtube


  17. Bathrobe said,

    February 4, 2019 @ 10:25 pm

    Apart from a Mongolian singing C&W, I found it interesting that a country-music station based in Buffalo would have what appears to be a Pommy (English) guy compering the show and a South African on the panel…

  18. Chas Belov said,

    February 5, 2019 @ 12:22 am

    Original performance on Mongolia's Got Talent, with English closed captions. https://www.youtube.com/watch?v=ddBLFdiylFg

    @Miles Archer: It was news to me that Amarillo isn't pronounced like in Spanish; that's how I would have said it.

  19. AntC said,

    February 5, 2019 @ 4:51 am

    His pronunciation was perfect.

    But what language are all those fake heads in the bucket seats talking? Are they bots or holograms of some sort? Just what is that oily, mincing argot?

    what appears to be a Pommy (English) guy compering the show and a South African on the panel…

    Yes James Corden (Pom, comedian, actor) is a late night talk show host on the West Coast. There's quite a few non-American late night talk show hosts: John Oliver, another Pom; Trevor Noah, South African.

    And quite a few Poms as 'judges' on talent shows: most conspicuously Simon Cowell; also the ubiquitous whinging Pom Piers Morgan.

    I can only assume it's a major challenge to find 'celebritys' to appear on these things, to up the ratings.

    If it seems like I'm some sort of TV addict:
    Hours spent in the past 5 years watching talent shows: zero.
    Hours spent in the past 2 years watching late shows: substantial — they have the best ever scriptwriter-in-chief.

  20. Bathrobe said,

    February 5, 2019 @ 7:56 am

    I can only assume it's a major challenge to find 'celebritys' to appear on these things, to up the ratings.

    That's strange. I would have thought it was because non-Americans had something different to offer, perhaps a different sense of humour or a slightly different slant on things. I thought it was rather refreshing to see non-Americans on such a show.

  21. Victor Mair said,

    February 5, 2019 @ 8:08 am

    Also, Americans are addicted to their lovely accents!

  22. Ralph Hickok said,

    February 5, 2019 @ 9:34 am

    When I was in college back in the late 1950s, there was a Japanese singer who had her 15 minutes of fame for doing the same thing. Actually, it was somewhat more than 15 minutes; IIRC, she had a hit album containing her covers of some pop songs. Then she disappeared. There was nothing particularly good about her singing and I'm sure she owed her brief popularity entirely to all the publicity about the fact that she didn't know English and simply memorized the sounds.

    On a somewhat related note, my granddaughter took part in those statewide (Illinois) music competitions when she was in junior and senior high school. One year the competitors had to sing an Italian song. In the finals, Kendra forgot a couple of lines but filled in with with some nonsense syllables that sounded like Italian, a la Sid Caesar.

    Four of the five judges gave her a perfect score of 5.0; the fifth judge gave her a 4.5 with the comment that he had to knock off half a point because she forgot some of the lyrics. Evidently the other judges didn't notice or were so impressed with her ability to improvise that they didn't penalize her.

  23. Andrew Usher said,

    February 5, 2019 @ 9:58 am

    I would not call his pronunciation anywhere near perfect; it's understandable – no worse than a British accent would be – and the rendition is OK. And maybe it's the best one can get when singing in a language one doesn't know, but it certainly doesn't require any highly unusual talent to explain.

    Also, he seems to pronounce 'Amarillo' the normal way as well as the 'Spanish' one, as well as other deviations. This suggests that he _is_ more familiar with English than we might pretend.

    I'd never heard this song before but am pretty sure that I can not understand a literal meaning to the lyrics either …

    k_over_hbarc at yahoo.com

  24. J.W. Brewer said,

    February 5, 2019 @ 10:24 am

    I'm assuming he learned the lyrics phonetically not by looking at printed words on a page plus some rough sense of English orthographic conventions but instead largely by listening to recordings of it sung by actual Americans over and over again (possibly only the bit-hit version by George Strait, possibly some other version or multiple versions). A few of the lines do require some background knowledge of the conventions of rodeo competition to make sense of and thus might be as puzzling for some Anglophones (probably including some percentage of Americans) as a song full of cricket-and/or-rugby references might be to most Americans.

  25. Philip Taylor said,

    February 5, 2019 @ 10:50 am

    It is not my sort of music, but I was extremely impressed. I have no way of knowing (or even judging) whether the accent that he assumed for the purposes of this song was authentic, but to my (British) ear there was nothing to indicate to me that he was not a native speaker, and a good singer to boot.

  26. BZ said,

    February 5, 2019 @ 11:39 am

    It's not unlike many religious Jews who often don't understand the Hebrew words they are saying during prayer, and some of whom read a phonetic transliteration. Due to the daily repetition of the ritual, many of them know the prayers by heart as well. This is of course not officially condoned as a preferred course of action, but given the holiness of the Holy Language, it may be preferred to reading a translation.

  27. Victor Mair said,

    February 5, 2019 @ 12:25 pm

    Here's a good version of his "Amarillo by Morning":


  28. J.W. Brewer said,

    February 5, 2019 @ 12:28 pm

    Actually, upon re-listening it appears that Enkh Erdene cut out the middle verse, which has the most "technical" and thus potentially opaque lyrics. Here's a version that's annotated for the benefit of those who would otherwise have no idea what "I'll be looking for eight when they pull that gate" might possibly need. https://genius.com/George-strait-amarillo-by-morning-lyrics

  29. J.W. Brewer said,

    February 5, 2019 @ 12:34 pm

    Last word of my prior comment should have been "mean" rather than the damnyouautocorrected "need." Separately, the 2016 back-in-Mongolia clip VHM just posted includes the missing verse left out in the 2019-in-America version. I think his pronunciation has gotten notably closer to American in the intervening years.

  30. Veronica said,

    February 5, 2019 @ 12:46 pm

    As a George Strait fan from way back, I know this song extremely well. For the first few words of the song, ("Amarillo by morning / Up from San Antone") I would have thought I was listening to Strait himself singing it. He had what I would think of as a "country sound," throughout, which was really kind of amazing . . . but his accent is not really perfect. His pronunciation of "got" and "county" don't sound like anything in American English, and I wonder if his native language lacks the "th" sound, since he tended to gloss over that sound altogether. Also, English-speaking Americans universally say "Santa Fe" with the same vowel sound that they use for "Santa Claus," not the "Sahnta Fe" of Spanish.

  31. Victor Mair said,

    February 5, 2019 @ 12:50 pm


    I noticed that about his "Sahnta Fe" too.

  32. Chris Button said,

    February 5, 2019 @ 1:40 pm

    You can actually get away with quite a lot pronunciation-wise when singing. To shamelessly quote myself from a comment I made a while ago on a separate LLog thread…

    Singing often neutralizes allophonic variation to cause a degree of convergence which goes well beyond obvious intonational/tonal effects. What's interesting is that it is not just restricted to synchronic phenomena like Americans and Brits no longer replacing "t" with flaps and glottals, but goes all the way through to things like /æ/ being lengthened in accordance with the melody to encroach upon /ɑ:/ which correlates with the diachronic explanation for "grass" as Southern British /grɑ:s/ versus General American /græs/ due to lengthening before fricatives etc. While Adele might indeed be easier to understand for Americans when singing than when speaking naturally, this has little to do with any notion of her intentionally putting on an American accent when singing (whether she chooses to do so or not). If you listen closely there are always giveaway clues as to a singer's origin. For example, while rhotic dialects often lose some of the prominence of their syllable final "r" sounds when extended out across a melody there is often still some vocalic coloring at least; the fact that Boy George has none of this coloring in "Karma Chameleon" leaves little doubt as to his origin.

  33. Victor Mair said,

    February 5, 2019 @ 3:58 pm

    Have had a Mongolian-English-Texas cowboy earworm in my head for the last two days. Don't know how to get rid of it.

  34. Rube said,

    February 5, 2019 @ 4:40 pm

    @Victor Mair "Have had a Mongolian-English-Texas cowboy earworm in my head for the last two days."

    From a linguistics point of view, isn't THAT a sentence.

  35. Bill Benzon said,

    February 5, 2019 @ 6:00 pm

    @Veronica: You said, He had what I would think of as a "country sound," throughout, which was really kind of amazing ….

    Yes. While I"m not particularly knowledgeable about country & western music, I'm certainly familiar with it, and you're right about the sound. And yes, it is amazing.

    * * * * *

    Above I'd inserted a link to a different performance:


    I've read through some of the comments to that and learned that he took the lyrics to a Mongolian song, Nart tengeriin door, and set them to Michael Bublé's version of "Feeling Good." That's why it sounded familiar.

  36. Anthony said,

    February 5, 2019 @ 7:49 pm

    His first "got" sounds like German "gut" more than like any English accent, even though the very next "got" is perfect. He repeats that when the first verse repeats.

    There's a "leg" pronounced "league"

    The first vowel in the first "Santa Fe" is wrong in Spanish, too, almost an 'o' sound.

    There are a few other blips here and there, but if he really learned it just by listening and doesn't know any English, it's pretty good.

  37. Jerry Friedman said,

    February 5, 2019 @ 11:08 pm

    Veronica: Also, English-speaking Americans universally say "Santa Fe" with the same vowel sound that they use for "Santa Claus," not the "Sahnta Fe" of Spanish.

    Maybe 99.999%, but here in northern New Mexico I do occasionally hear native English speakers, including both Hispanics and Anglos, say "Sahnta Fe".

  38. Jerry Friedman said,

    February 5, 2019 @ 11:32 pm

    Now what I want to know is whether O.Enkh Erdene can sing "El Rey" in George Strait's Texan-accented Spanish.

  39. Andrew Usher said,

    February 6, 2019 @ 12:07 am

    Despite his imperfect pronunciation, as others have noted, he does hit the target of sounding acceptably 'country'. This 'sound' in singing is separate from accent, I suppose it corresponds to what we call prosody or intonation in speech – I just encountered this in a different context when reading (and noticing) how close 'Weird Al' comes to sounding like the artists he parodies – even though he practically does not change his GA accent.

    Re: 'Sahnta Fe' I've never actually heard it. It really is so back in the song that I could not say whether a Spanish speaker would map it to 'a' or to 'o'.

  40. Philip Taylor said,

    February 6, 2019 @ 3:49 am

    Would it be possible to update the URL of the embedded video at the start of the article so that those reading it for the first time are not put off by a clearly broken link ? Or is that thought to be liable to provoke even more instances of the video being withdrawn ?

  41. Victor Mair said,

    February 6, 2019 @ 5:13 am

    Better not mess with the situation regarding the inactive video at the beginning. My note there makes it very clear what has happened and the remedy. It would be a real tragedy if we put another one up there and the singer's overzealous agents (or somebody else who took down the first one), would proceed to take down the others too.

  42. Anthony said,

    February 6, 2019 @ 10:08 am

    Andrew is right about this guy sounding "country" even if his pronunciation is imperfect.

    Another good example is him singing "Sixteen Tons" – https://www.youtube.com/watch?v=RyNTER91e-g (Compare the sound to Paul Robeson – https://www.youtube.com/watch?v=niClWAnP_14 – who also mispronounces or mumbles some of the words.)

    Incidentally, Mongolia is now one of the top 20 coal-mining nations in the world, and produces more coal per person than any country besides Australia. A translation of "Sixteen Tons" could have a social impact.

  43. maidhc said,

    February 7, 2019 @ 5:40 am

    Here in the Bay Area there are theatrical performers who sing on stage in professional productions in Irish and Scottish Gaelic with creditable pronunciation, but have not a clue about what the song is about. I have heard some of them go on the radio to plug their performances, and the host might say, what a beautiful song, what does it mean. Who knows? Is it a song about two people who are so much in love? Is it a song about a woman who has been abandoned pregnant by a married man? Is it a song about a woman whose husband's head has been cut off and stuck on a pole? Who can tell, but doesn't it have a lovely melody?

    This attitude is more common with Celtic languages (and quite possibly other minority languages, but I don't have any first-hand knowledge) than it is with languages like Italian or German.

    My wife when she was a teenager used to learn foreign-language songs by people like Miriam Makeba, Theodore Bikel and Nana Mouskouri, but at least she took the time to read the translations. She can still floor people by singing a song in Japanese or Hindi, admittedly the most common ones in those cultures. And she knows what it means.

    I was surprised a couple of years ago when I went to the Obon festival in our local Japantown. This is a festival that happens in the summer when the spirits of people who have died come back, and there is dancing and singing to welcome them. There are particular songs and dances that are done at this festival.

    I was amazed at the number of people who knew the songs and dances, but didn't look very Japanese. Are these Japanophiles who spend their time studying Japanese language and culture, and in the summer they travel round to all the Obon festivals on the West Coast?

    I know Scottish or Scottish-American people who go around to all the Highland Games, so it would be quite believable. It's just not something I have encountered before.

  44. Chris Button said,

    February 7, 2019 @ 11:29 am

    Here in the Bay Area there are theatrical performers who sing on stage in professional productions in Irish and Scottish Gaelic with creditable pronunciation, but have not a clue about what the song is about.

    This makes me think of the Welsh song Suo Gân that is prominently featured in "Empire of the Sun" with Christian Bale (who incidentally was apparently born in Wales) being voiced over by an English singer James Rainbird (who I'm guessing was not a Welsh speaker?):


  45. Trogluddite said,

    February 7, 2019 @ 11:56 am

    I can remember wondering for many New Year's Eves quite what an "ine" was and why "Old Lang's" one was particularly special. Likewise, it wasn't until I moved to Yorkshire that I learned why it was advisable not to wander on Ilkley Moor "bar tat".

  46. Victor Mair said,

    February 7, 2019 @ 7:21 pm

    From Brendan O'Kane:

    I've heard a few people in the Philadelphia area singing Irish-language songs they learned phonetically. I can understand the impulse, I guess — as maidhc says, they're lovely tunes, and it's a less obnoxious way of getting back in touch with one's roots (which is the usual motivation) than, say, joining the Ancient Order of Hibernians and wearing a kilt.
    A bit more objectionable to me is the practice of singing English-language songs with a cod-Irish accent, though even there one can sometimes identify extenuating circumstances: I can think of at least one song in which the rhymes ("town," "time," "Boyne") don't really work outside Northern Ireland. (This is one of the many things that make me glad not to be dealing with reconstructed pronunciation: imagine dedicating a lifetime to Old Chinese phonology and then finding out that the authors and compilers of the Shi jing had the equivalent of a Belfast accent.)

  47. John Swindle said,

    February 7, 2019 @ 9:52 pm

    @maidhc: I don't know about Obon in North America, but in Hawaii there certainly are people who make the circuit (pun not intended) and go to a different Bon dance each weekend of the season if they can. Distances of course are smaller. The dances here are mostly organized around Japanese Buddhist temples, and the temple members attend each other's festivities. Temples offer practice, but anybody joins in with or without prior practice. Some but by no means all of the non-Japanese faces you see will be those of temple members, and some but not all of those will have Japanese family members. Enjoying the Bon dance is enough, although there's usually also a religious service available. I gather that the pattern in Japan is different, more a matter of civic religion practiced at shopping centers, like Santa Claus.

  48. Chris Button said,

    February 8, 2019 @ 9:32 am

    imagine dedicating a lifetime to Old Chinese phonology and then finding out that the authors and compilers of the Shi jing had the equivalent of a Belfast accent

    That quite neatly sums up the problem with many of the current models of OC reconstruction being peddled – an unfortunate blurring of the distinction between underlying phonology and surface phonetics.

  49. Jack said,

    February 9, 2019 @ 6:29 am

    For those of you who are wondering how Mongolian language sounds like, here is your answer. https://www.youtube.com/watch?v=6WFFUso5v9Y
    I have to say that I am really impressed with his accent. Because I have been trying to fix my accent for a long time and It is a quite struggle. By the way. I am also Mongolian.

  50. IA said,

    February 9, 2019 @ 8:02 pm

    @ Chris Button:

    Have I understood correctly that you have not yet presented your own OC reconstruction, but will be doing so, and that it will in some way be (relative to ZZSF/PWY, Baxtert-Baxter …) Pulleyblank-ish?

    When can this be looked forward to?

  51. IA said,

    February 9, 2019 @ 8:08 pm

    ( 'Baxter–Sagart', that should have been in the post just now. )

  52. Chris Button said,

    February 10, 2019 @ 8:11 am

    @ IA

    Yes. To give a simple example, a word like 肘 *trə̀wʔ < *kʷrə̀ɣʔ "elbow" is clearly related to a word like 弧 gʷáɣ "arc" (via a sense of "bow") via the pervasive ə/a ablaut underlying the OC lexicon and originally identified by Pulleyblank. A Baxter-Sagart style reconstruction like 肘 *t-[k]uʔ and 弧 *[g]ʷˤa does nothing to elucidate this.

    As for when my "Derivational Dictionary of Chinese and Japanese Characters" will be appearing, unfortunately I am financially unable to work on it full-time so it is taking an inordinately long time and looks to be several years out I'm afraid.

  53. Chris Button said,

    February 10, 2019 @ 9:17 am

    The medial -r- component in the B&S form for 肘 did not come out for some reason. It should be *t-[k]uʔ

  54. Chris Button said,

    February 10, 2019 @ 9:20 am

    Ok let's go with *t-[k]ruʔ as clearly the "" symbols are messing the formatting up.

  55. Chris Button said,

    February 10, 2019 @ 9:21 am

    I give up…. Sigh…

  56. IA said,

    February 10, 2019 @ 10:36 pm

    @Chris Button

    It's all good (as the expression goes). Easily enough looked up at http://ocbaxtersagart.lsait.lsa.umich.edu … missing are the angle brackets around the 'r'.

    As for 'Derivational Dictionary of Chinese and Japanese Characters', I ardently look forward to it. 'Speed the Day!' (As the old-fashioned optative has it.)

  57. IA said,

    February 12, 2019 @ 11:10 pm

    @ Chris Button:

    'That quite neatly sums up the problem with many of the current models of OC reconstruction being peddled – an unfortunate blurring of the distinction between underlying phonology and surface phonetics.'

    It would be instructive if you could give an example or two. Thanks.

  58. Chris Button said,

    February 13, 2019 @ 3:21 pm

    @ IA

    To give a comparable example from English, the current situation is not dissimilar to someone a few thousand years from now reconstructing American English and definitively stating that "cot" and "caught" are either homophones or not. That is not to say we should be reconstructing separate dialects for Old Chinese (incidentally I have big issues with how this is currently being proposed in other ways for OC particularly in terms of onsets), but rather that we should be reconstructing an underlying phonological system that can flex to accommodate different phonetic realizations on the surface.

RSS feed for comments on this post