Understanding across varieties of English

« previous post | next post »

Yesterday, I posted on Speakout/Truthout about Rachel Jeantel's African American Vernacular English use in the Zimmerman trial/verdict: "Race, Credibility, Communication and Evidence in the Zimmerman Trial, and Beyond", 7/30/2013.

Readers of my Language Log post of July 10 — written before the verdict was announced — may recall that I felt that despite its vernacular character, Jeantel's testimony would be understood by the jury, but that they might not relate to her. Turns out I was both right and wrong. They didn't relate to her, didn't even find her credible, but they (at least Juror B37) also found her difficult to understand.

This case raises an interesting research question about the extent to which African American Vernacular English (and other social/regional dialects of English) are understood by speakers of Mainstream American English or Standard English. We seem to have very little research evidence about this, apart from studies focused on individual grammatical or lexical features, like the study of stressed BIN that I mentioned in my blog, and studies of individual lexical items like cut-eye and suck-teeth.

Here's another bit of anecdotal evidence that makes me question the widespread assumption that people from other dialect backgrounds understand AAVE quite well. Last weekend, my wife and I went to see Fruitvale Station –a powerful new film about the fatal shooting of handcuffed Oscar Grant, a young Black man who was "just trying to get home" on the train after New Year's Day midnight festivities in SFO (). As the credits rolled at the end of this powerful movie, I heard an older white man across the aisle comment that "The acting was superb," to which his wife responded, "But I couldn't understand a word of the dialogue!" To which I thought, "Not a word?!" And I wondered to myself again how much of African American vernacular (especially if fluently spoken) Whites really understand. We really don't know. But think about the implications, if teachers, jurors, job interviewers and so on miss some or all of what students, witnesses, defendants, job seekers say! This is an area crying out for research.

Please let me know if you're aware of any good relevant research on this specific point (intelligibility of AAVE to non-AAVE users), or on intelligibility among speakers of other ethnic, social, and regional varieties, especially of English, but of other languages/dialects as well. I'm as interested in the various methods people use to assess intelligibility reliably, as in their substantive results. I'm aware of Labov's relevant stuff, e.g. the Gating Experiments from the Project on Cross-Dialectal Comprehension reported on in his 2010 Principles of Linguistic Change, vol. 3, as well as his older (1973) paper on "The Boundaries of Words and their Meanings." And of Gary Simons' 1979 monograph on Language Variation and Limits to Communication. What about other studies, especially recent ones, and ones that also take into account attitudes and ideologies and how those affect cross-dialect comprehension?



  1. Duncan said,

    August 1, 2013 @ 4:57 am

    Not directly AAVE related, but given that I spent six years as a kid in East Africa, plus the traval (4 one-way-trips) via Europe to/from there from the US Northwest, for years I had no idea just how difficult it was for many to understand much of the multi-accented and multi-venaculared English I've always simply understood with little trouble at all.

    Then my high school had a public showing of Chariots of Fire, which I thoroughly enjoyed, but which many of my classmates had serious trouble with, because they simply couldn't understand large parts of it either verbally or culturally. Ever since then I've had a much keener appreciation for just how much I /can/ understand, that many of the less travelled USians around me since my family returned from East Africa simply can't. (Never-the-less, back when it was current I read Salmon Rushdie's Satanic Verses and found it one of the most difficult books to get thru I've ever read, simply because I lacked exposure to many of of the cultural cues references it made and that I guess he took for granted that his audience would know. But in the process it revealed an entire region and multiple cultures to me that I had hardly glimpsed to that point. So I guess I understand some of the pain my classmates were going thru with Chariots of Fire.)

    But you're right, that AAVE and the jury's lack of understanding thereof may well have changed the verdict in that trial. The implications are indeed disturbing.

  2. Verena Minow said,

    August 1, 2013 @ 5:25 am

    I came across a couple of intelligibility studies when I was researching on varieties of South African English. The following article came to my mind immediately:

    van der Walt, Christa. 2000. The international comprehensibility of varieties of South African English. In: World Englishes 19(2): 139-153.

    I think the focus is mainly on non-native varieties of English but it does give some insight into methods used to assess intelligibility.

  3. David C said,

    August 1, 2013 @ 5:40 am

    One of the many quirks surrounding dialects in German-speaking Switzerland is the extreme pride that the Swiss have in their ability to understand dialects from different regions. Except, of course, the Kanton of Wallis, whose dialect is particularly opaque.

  4. Levantine said,

    August 1, 2013 @ 5:43 am

    I'm a British English speaker who went to the States for his graduate education, and I didn't have any trouble understanding Rachel's testimony. Maybe this was because it was more audible when watched on television. I assumed it was her quietness more than anything that was making it difficult for those in the courtroom to understand her. The only feature of her accent that may have thrown me had I not previously encountered it in speakers of AAVE was her dropping of possessive S.

  5. R said,

    August 1, 2013 @ 6:10 am

    Great idea! I would love to see a study that combined both objective and subjective measures of intelligibility. I have a strong feeling that people are likely to report understanding less than they due to avoid identifying with "those people". My only evidence comes from some personal experience as a white person from a border state. I have relatives who claim to speak/understand no Spanish, even though I know for a fact that this is not the case.

  6. Street-Smart Language Learning said,

    August 1, 2013 @ 6:29 am

    This post reminded me of my roommate in college. I went to college in DC and many of the workers in the dining hall were African American and spoke AAVE. I grew up in a mostly white suburb right outside of Philly but regularly ran into African Americans that entire time, and because of that I presume I never had any trouble understanding what they were saying. My roommate grew up in an even whiter suburb of Philly about an hour farther away than mine, and he was always frustrated by how he couldn't understand a word of what the dining hall workers were saying. I always chalked it up to him simply having had less exposure to AAVE due to where he grew up, but that never seemed like a wholly satisfying answer due to how close it was to where I grew up.

  7. Michael Newman said,

    August 1, 2013 @ 6:40 am

    Anecdote and comment.
    Anecdote: One year I had a White vernacular NYCE speaking freshman doing some transcriptions. When he was transcribing a AAE speaking gang member, he kept asking for clarifications. This is a kid who went to a high school with at least some kids with similar backgrounds, and he was pretty vernacular himself, but he had difficulty.

    Comment: Isn't there some research showing that even fluent Asians (I'm not sure if Asian Americans) sometimes provoke claims of 'difficult to understand' that at least plausibly involves self-fullfilling expectations? Wouldn't it be almost impossible to disentangle phonological and affective sources of incomprehension? I'm not saying the research is worth doing, and there might not be ways of at least starting to differentiate where the causes lie, but it is something to keep in mind.

  8. Dick Margulis said,

    August 1, 2013 @ 7:20 am

    Couple of factors that any such research ought to take into account (in addition to the points noted above that some cultural exposure outside one's own immediate neighborhood and a bit of curiosity about the outside world are of immense value): Age of the listener and length of exposure. I find that when I go to a movie set in England and employing any sort of regional or working-class dialect, it takes me several minutes before I can understand a word of it, but then it all seems quite clear. So if you were to expose me to brief snippets of some test dialect, I'd likely fail on every one of them, but if you allowed several minutes of acclimation (again, perhaps depending on the age of the test subject), I'd do fine. Clarity, loudness, lack of background noise, and the subject's hearing acuity would also be factors, but presumably they would be controlled for anyway.

  9. möngke said,

    August 1, 2013 @ 7:25 am

    Another issue is, quite possibly, the ideology of comprehensibility – i.e., speakers may claim to understand / not understand a language even though they in fact have little trouble doing so (or make that claim on part of others). A lot of Slovenes, for instance, would say that younger generations in Slovenia can't understand Serbian or Croatian speakers very well, that the languages aren't mutually intelligible to any great degree (at least not more than any other two Slavic languages), etc. – which are claims that don't really reflect linguistic reality fully. So what people want to believe about another person's speech variant's comprehensibility may be highly relevant for the claims regarding whether or not they had understood them. (Regardless of whether they did in fact understand them or not.)

    Sorry for the off-kilter linguistic anthropology plug, but it was the first thing I thought of when I read the article! Unfortunately I don't know very much (or anything at all) about studies of actual understanding/intelligibility, but I would be very interested to break out of the "constructivist ghetto" in this respect and learn more about it. (Primarily because I specialize in Arabic, where this is a huge topic.)

  10. Ray Girvan said,

    August 1, 2013 @ 7:56 am

    @R: I have a strong feeling that people are likely to report understanding less than they due to avoid identifying with "those people"

    Agreed. I've probably said this before, but in the UK, I've run into quite a lot of people who claim to genuinely misunderstand, or to not understand at all, British working-class sociolect forms such as "you wasn't" or multiple negations for emphasis. They claim to really believe a speaker who says "I don't want none of that" means "I want some of that".

    I find this hard to believe, given the wide occurrence of such forms in real life and the media – I think they're just making some kind of snarky point. But I've no idea how to prove it, except to expose them to such forms in a safety-critical situation like the one in the book You Only Live Twice where James Bond (pretending to be a Japanese miner who can"t understand English) is put on a seat that'll squirt molten lava in a few minutes, and is told in English what'll happen.

  11. Tyler Morkin said,

    August 1, 2013 @ 8:22 am

    @Michael Newman: "Isn't there some research showing that even fluent Asians (I'm not sure if Asian Americans) sometimes provoke claims of 'difficult to understand' that at least plausibly involves self-fullfilling expectations?"

    There's the Rubin 1992 study. This found that undergrad students comprehended less when they believed their speaker was Asian, even though the speaker in question was a native speaker of SAE/mainstream English. In other words, they projected an accent on someone who didn't have an accent just because the speaker was of a different ethnic origin.

    I believe there's a lot more socio stuff at play here than people really want to admit. A study that measured comprehensibility of non-mainstream varieties would certainly need to take this into account.

  12. Nic Subtirelu said,

    August 1, 2013 @ 8:42 am

    I'm not aware of any research on intelligibility of AAVE for White English speakers. It does seem like a really important question. I hope someone will take this up.

    Of course, as others have mentioned, language attitudes are a well-documented confounding factor (and are extremely likely to play a role here). At the risk of shameless self-promotion, Stephanie Lindemann and I just published a paper looking at some of the research on intelligibility (from more of a second language acquisition perspective) and how the listener plays a role in these perceptions. We cite a number of studies I think would be relevant to this question.

    Here's the link to that paper:

    And here's a few of the more relevant references:

    Gooskens, C., & van Bezooijen, R. (2006). Mutual comprehensibility of written Afrikaans and Dutch: Symmetrical or asymmetrical? Literary and Linguistic Computing, 21, 543–557.

    Gooskens, C., van Heuven, V. J., van Bezooijen, R., & Pacilly, J. J. A. (2010). Is spoken Danish less intelligible than Swedish? Speech Communication, 52, 1022–1037.

    Hilton, N. H., Schüppert, A., & Gooskens, C. (2011). Syllable reduction and articulation rates in Danish, Norwegian and Swedish. Nordic Journal of Linguistics, 34, 215–237.

    Jensen, J. B. (1989). On the mutual intelligibility of Spanish and Portuguese. Hispania, 72, 848–852.

    Kang, O., & Rubin, D. L. (2009). Reverse linguistic stereotyping: Measuring the effect of listener expectations on speech evaluation. Journal of Language and Social Psychology, 28, 441–456.

    Lindemann, S. (2002). Listening with an attitude: A model of native-speaker comprehension of non-native speakers in the United States. Language in Society, 31, 419–441.

    Rajadurai, J. (2007). Intelligibility studies: A consideration of empirical and ideological issues. World Englishes, 26, 87–98.

    Rajagopalan, K. (2010). The soft ideological underbelly of the notion of intelligibility in discussions about “World Englishes.” Applied Linguistics, 31, 465–470.

    Rubin, D. L. (1992). Nonlanguage factors affecting undergraduates’ judgments of nonnative English-speaking teaching assistants. Research in Higher Education, 33, 511–531.

    Tang, C., & van Heuven, V. J. (2009). Mutual intelligibility of Chinese dialects experimentally tested. Lingua, 119, 709–732.

    Wolff, H. (1959). Intelligibility and inter-ethnic attitudes. Anthropological Linguistics, 1, 34–41.

  13. Jay said,

    August 1, 2013 @ 8:49 am

    This puts me in mind of the time I was watching TV coverage of Hurricane Katrina with an Irish friend of mine, and at times he was having a great deal of difficulty understanding the AAVE of some of the interviewees. He was frustrated and surprised because, he said, he always thought he could understand AAVE speakers perfectly well, since he never had trouble when watching American TV dramas and movies that featured them.

    Several reasons present themselves. For one, spontaneous speech is probably harder to understand in general than scripted speech. Also, people being interviewed after a hurricane's struck their city are likely to be upset and speaking quickly or indistinctly. I suspect, too, that the AAVE you hear in scripted TV/film is rather homogenized, lacking some regional markers that spontaneous speech would have – much of it may even be written by non-speakers of the dialect, considering how many Hollywood writers are white. So, when you ask whether someone can understand AAVE… what type of AAVE are they hearing, and what is the context?

    I agree that some people exaggerate their lack of understanding out of prejedice against the speakers of a given dialect, and I think the reverse also happens – some people may be reluctant to admit they don't understand a speaker, for fear of being thought prejudiced against their group.

  14. Mr Fnortner said,

    August 1, 2013 @ 9:12 am

    I reminded of the jive scene in Airplane with Al White, Norman Gibbs, and of course, Barbara Billingsley? http://www.youtube.com/watch?v=7fkZdz4Vz10

  15. Dan Lufkin said,

    August 1, 2013 @ 9:27 am

    This is a perennial topic in Scandinavian, too, where intelligibility is intransitive. Danes can usually understand mainstream Swedes but not vice-versa. Norwegians sometimes can't understand someone from 50 miles away but still in Norway. Young Danes have trouble understanding Danish movies from the 1950s while mature Danes, having gone through the transition gradually, can understand youngsters.

    On a recent trip to Ireland, I found that my ability to understand our guide, a cheerfully voluble chap with a deep West-country accent, went from zero to 90% in a couple of hours. That couldn't happen with a genuinely different language.

  16. Marc Leavitt said,

    August 1, 2013 @ 9:53 am

    I always thought I was fairly good at understanding other dialects ontil I watched the BBC series, "Monarch of the Glen" on Netflix; I had to pay very close attention, aND. It was nae easy ta kenbtogf English, NDING

  17. Marc Leavitt said,

    August 1, 2013 @ 9:55 am

    Sorry. I hit the wrong key.

  18. Mark P said,

    August 1, 2013 @ 10:00 am

    It seems to me that there are at least three factors that come into play, familiarity, speaking face-to-face, and a willingness to try to understand. When I was in graduate school, especially when taking undergrad classes, I heard a lot of complaints from students about not being able to understand foreign-born instructors. I found that after a few classes it was much easier to follow the instructor's words.

  19. wally said,

    August 1, 2013 @ 10:04 am

    On the point Dick Margulis made on length of exposure, my teenage daughter and I saw the movie of Shakespears Much Ado About Nothing recently, and we both found that we had trouble understanding the speakers at first, but we understood the speakers with little trouble by the end of the movie.

    As for Slovenes, I talked to one young man 25 years ago who said that they all studied Serbo Croatian for one year in school, and then they had no problem with it.

  20. J.W. Brewer said,

    August 1, 2013 @ 10:30 am

    I have not followed the post-trial stuff closely, but I believe a number of other members of the jury have as a general matter distanced themselves from B37's public statements. Witnesses at criminal trials come from all walks of life, all ethnic/socio-economic backgrounds, all dialectical backgrounds (and some may speak English as a second language, but well enough that having them testify in their L1 via translator might cause as many problems as it solves). It is the lawyer's job to maximize the effectiveness of the presentation of the witnesses they call. Juries certainly do not convict only or primarily in cases where the prosecution's witnesses all speak the prestige variety of standard AmEng (or the native dialect/variety of the majority of the jurors), or our prisons would not be nearly as crowded as they are. Trial testimony is not necessarily "scripted" but it's not spontaneous, either, or at least, the lawyer conducted the examination has lots of opportunities to ask for clarification/rephrasing if he/she doesn't think (for whatever reason, including failures of comprehension across dialect boundaries) that the testimony is being understood by the jury. Now, it is *possible* that these prosecutors did not correctly assess how well the jury was following the testimony (and this is all happening in real time in a high-stress environment and there are lots and lots of tactical decisions that need to be made quickly with incomplete information) and thus failed to avail themselves of those techniques, but I wouldn't bet on it.

  21. Barbara said,

    August 1, 2013 @ 10:47 am

    "And I wondered to myself again how much of African American vernacular (especially if fluently spoken) Whites really understand." Whites? Do you mean non-mother tongue speakers of AAVE? And those can be of any race?

  22. Graham Strong said,

    August 1, 2013 @ 10:48 am

    I don't have any research for you about intelligibility, but I do have a somewhat related news note — a manslaughter conviction in Northwestern Ontario was recently overturned due to under-representation of First Nations on the jury.


    There are many ways in which cross-cultural misunderstandings — including, but not limited to, language — lead to dire consequences. It happens in the courts, it happens in healthcare, it happens in day-to-day life. Bridging the cultures, or in this case, ensuring appropriate cultural representation, are key to reducing these misunderstandings.

    I think everyone reading this blog realizes that language is an imperfect method of communication, especially cross-culturally. Bottom line in this case — the reason there needs to be a "jury of peers" is because there is a better understanding of motivations and thinking (and, in the case you describe above, language) so that jurors can get the "whole truth". It also reduces the risk of prejudices getting in the way of justice.


  23. Jeff DeMarco said,

    August 1, 2013 @ 11:12 am

    It seems to me that the attorneys involved should have arranged for an interpreter. It a situation where shades of meaning are very important, it could make a significant difference.

  24. bianca steele said,

    August 1, 2013 @ 11:14 am

    On implausible claims not to understand: I once worked with someone who'd comment on just about every e-mail from certain sectors of Corporate, the ones that were filled with smiley-face nonsense jargon kinds of things (written in English but some, but not all, written by non-Americans and presumably not native English speakers) with "I didn't understand a word of it!" It's difficult to know what to reply to that, when the overt meaning was pretty self-evident, even if the assumptions were difficult to believe.

    I was struck by the juror who said she felt sorry for Rachel because she clearly wasn't able to make herself understood in "normal" society and probably felt deep inferiority because of it.

  25. Levantine said,

    August 1, 2013 @ 11:45 am

    Jeff DeMarco, really?! She was speaking English, after all. She was made to repeat everything enough times that everyone knew what she was saying by the end.

  26. J.W. Brewer said,

    August 1, 2013 @ 11:46 am

    Avoiding inappropriate exclusion of members of racial/ethnic minorities from juries is something the US court system has spent a lot of time wrestling with. Language competence is actually a different and more complicated issue – there have been arguments about whether in cases where some witnesses will testify in Spanish via an English translator it is legitimate to exclude Spanish-speakers from the jury on the theory that the jury's decision-making might be distorted if everyone else is going to be going by the what the translator said the testimony was but two or three out of twelve jurors would be able to form their own view as to what the testimony really was and/or argue to their fellow jurors that the translator had gotten it wrong. At least some courts have said trying to remove jurors on that basis is not the same as trying to remove them based on Hispanic ethnic identity despite the positive statistical correlation. Relatedly, counsel will sometimes question prospective viewers about their attitudes toward a defendant or key witness who may testify through an interpreter, in order to weed out those who may have negative views about someone who has lived in the U.S. for X years (and perhaps even gotten through the naturalization process, where the rigor of the English-proficiency requirement may not be that demanding in practice) but hasn't mastered the language.

  27. Rodger C said,

    August 1, 2013 @ 12:00 pm

    I'm a white West Virginian from a working-class background, and when I went to graduate school in Indiana I was constantly annoyed by how often I had to repeat what I said. I was speaking what passed for middle-class small-town Upland Southern in the 1970s, and to my own ear I sounded exactly like my peers around me. It probably didn't help that when I was trying to be friendly I no doubt automatically lowered my register a notch. But for years I just thought, "It's true about these Yankees, they don't respect us–they never listen the first time you say something." Eventually I caught on and learned to sound like Harry Von Zell.

  28. Andrew (not the same one) said,

    August 1, 2013 @ 12:23 pm

    I remember, as a child, genuinely not understanding the line 'I had nothing; no, I didn't have nothing' from A. A. Milne's 'Market Square'. Did he have nothing, or didn't he? But I was about six at the time. Milne was writing for an educated middle-class readership, and clearly expected them to understand it.

    But while this kind of failure to understand is implausible, being unable to pick out the words someone is saying is another matter. I have certainly had the experience of being initially unable to follow someone speaking in an unfamiliar dialect, although I'm sure that if I were given a transcript of what they said I could read it without difficulty.

  29. Chris C. said,

    August 1, 2013 @ 2:33 pm

    Off-topic because I have nothing really intelligent to add, but referring to San Francisco by its airport abbreviation is almost as bad as calling it Frisco. The airport isn't even actually in the city.

  30. J.W. Brewer said,

    August 1, 2013 @ 2:47 pm

    One source of data on perceptions of comprehensibility is to look for movies where dialogue in some variety of English, broadly construed, is subtitled because it is assumed that a mainstream Anglophone audience won't be able to follow it otherwise. Well-known examples of such subtitling include The Harder They Come (Jamaican Patois/Creole) and Trainspotting (the young-Edinburgh-heroin-addict register of Scots). Would subtitling some of the AAVE dialogue in Fruitvale Station have genuinely increased mainstream audience comprehension, or would it have seemed like a patronizing/condescending gesture that treats AAVE as more alien/incomprehensible than it in fact is?

  31. Mark Mandel said,

    August 1, 2013 @ 4:59 pm

    J.W. Brewer asks: "Would subtitling some of the AAVE dialogue in Fruitvale Station have genuinely increased mainstream audience comprehension, or would it have seemed like a patronizing/condescending gesture that treats AAVE as more alien/incomprehensible than it in fact is?"

    I would put this as "would many viewers have taken it as a patronizing/condescending gesture…?", because the interpretation is very much in the mind of the individual viewer, with lots of multidimensional variation that the cover term "mainstream" ignores.

  32. J.W. Brewer said,

    August 1, 2013 @ 6:12 pm

    Sure. One challenge for any sort of empirical research in this area is that it's not as if we just have one idealized AAVE speaker (who never code-switches regardless of social context) and one idealized white or "mainstream" (or just "non-AAVE-proficient")* listener (who always invests the same degree of attention and good will in trying to understand what is being said regardless of social context). So multidimensional variation is going to be ubiquitous.

    *Even leaving aside the rapidly increasing percentage of the U.S. population that cannot be adequately described as either "white" or "black," if like me you live in the NYC area you are reminded on a daily basis of the existence of a large and increasing population of black Americans who do not speak AAVE at all (rather, their pronunciation etc. tells you that they grew up in Jamaica or Barbados or Haiti or Senegal etc.) and have no greater intrinsic ability to understand AAVE than anyone else of any skin color.

  33. Adrian said,

    August 1, 2013 @ 8:10 pm

    It's been my belief for a long time that British people have less trouble with accents than American people do, so it'd be good if someone could research whether or not it's true!

  34. John said,

    August 1, 2013 @ 8:22 pm

    @Rodger C: My wife, fluent in five languages at the time, was absolutely flummoxed when we went on a trip to WV. The White, rural accent was opaque to her; she couldn't hear individual words. She had no problem in the towns, but in the countryside, it was simply a different language that sounded like English, much the way Hebrew sounds like Arabic.

  35. Grep Agni said,

    August 1, 2013 @ 9:17 pm

    I suspect fatigue matters quite a bit here. The testimony lasted hours and the jury had to pay attention the entire time. If it took even a little more effort to understand her than other witnesses after a few hours the extra work could be significant. I've had lectures which I thought I understood fairly well but left me feeling drained from needing to concentrate extra hard.

    Also, If a juror couldn't make out (say) half a dozen phrases over the two days that might be enough to claim the witness was hard to understand.

  36. Levantine said,

    August 1, 2013 @ 9:21 pm

    Adrian, I certainly feel that I have less trouble as a British person understanding Americans than they do understanding me, but I've always put that down to the fact that I grew up watching American films and TV programmes. I don't think Americans have the same sort of exposure to other kinds of English in their media, notwithstanding several very successful imports like Downton Abbey.

    Also, I feel that there is more of an emphasis in the States on 'standard' pronunciation than there is in the UK, where regional accents are a matter of pride for many people. As a graduate student at an American university, I was surprised to discover that Americans meeting for the first time very often couldn't determine each other's regional backgrounds from their respective accents. When I asked why this was, I received answers like, 'He's speaking without an accent', which puzzled me even more, until I finally realised that educated Americans often speak a sort of standardised version of the language that can't be geographically placed with any ease. I suppose this was once the situation in the UK with RP, though no longer. But the idea of accentless English was/is completely foreign to my British mind — even the Queen's English constitutes an accent, after all.

    To sum up the point I'm trying to make, I do get the sense that we Brits are typically exposed to a greater variety of accents than most Americans are.

  37. Jeff P said,

    August 1, 2013 @ 10:27 pm

    I believe that I heard a paper given about this very subject at LSA in 2009 (at least I think it was 2009–it was for sure in San Francisco). The author(s?) specifically studied comprehension of doctors with AAVE-speaking patients, I believe somewhere in the American South. It would probably have been under the aegis of the American Dialect Society. Sorry I can't be more specific.

  38. julie lee said,

    August 1, 2013 @ 10:46 pm

    On a visit to London I was sitting in a bus and the dark-skinned young couple sitting behind me was talking to each other in a foreign language. They looked like they were Africans. Since I am interested in languages, after about fifteen minutes I turned around and asked them: "Excuse me, may I ask what African language are you speaking?" The woman replied, with a giggle: "We're speaking English." After that they resumed talking and, listening again, I realized they were indeed speaking English, but with a heavy accent.

  39. Jenny Chu said,

    August 1, 2013 @ 11:16 pm

    Wouldn't it be interesting to do an experiment like this (or has it already been done?)…

    People of different backgrounds are asked to watch different versions of a video: the audio is the same, but with different actors on screen (who seem to be) saying the words.

    It could be the same text, once in AAVE perhaps and once in each of several other varieties of English, but with the video being different people with different appearances (dark skinned, light skinned, dressed in stereotypical clothes representing one or another group, and maybe one "ambiguous"). I am sure this is easily with readily-available technology. (I'll ask an 11-year-old I know; that should do it.)

    Then, we (a) find out how much the participants *actually* understand, eg through a little content quiz ; (b) ask the participants how much they feel they have understood

    This might shed some light on the impact of prejudices on comprehension.

  40. ohwilleke said,

    August 1, 2013 @ 11:25 pm

    Subtitling AAVE for Standard English speakers is sometimes done on television (typically in nightly news interviews of someone with a particular difficult to understand dialect for Standard English speakers). Even outright translating AAVE is not unprecedented and now and then someone is hired to do those translations in real life (less seriously recall the humorous volunteer translator of "Jive" in the movie "Airplane").

    The context where translation often comes up in court cases is in efforts to assist juries in understanding wiretaps or hidden microphone recordings of conversations of African African gang members, especially in drug cases. I've never heard of it being done in cases involving live witness testimony, however.

  41. Mark said,

    August 2, 2013 @ 12:04 am

    I don't think there are relevant studies of the comprehension of AAE by SAE speakers and vice versa, but I do know some are underway. At this point, the person to ask might be John McWhorter, who said in his article, "Wasting time on an illusion", The Black Scholar, vol. 27, 1997, "It is a fact that Black English is not different enough from standard English to pose any significant obstacle to speaking, reading, or writing it." and later "[Black English] is mutually intelligible with standard English both on the page and spoken…." He doesn't cite any empirical studies bearing on these claims, but takes the points to be obvious once one understands the linguistic issues properly.

  42. Eorrfu said,

    August 2, 2013 @ 12:49 am

    Part of it was her accent that sat atop her AAVE was much different than the AAVE I have heard in the past. I assume it was the Hatian influence.

  43. AlexB said,

    August 2, 2013 @ 3:21 am

    Strange that nobody has brought this up yet.


  44. dainichi said,

    August 2, 2013 @ 3:47 am


    "I have relatives who claim to speak/understand no Spanish, even though I know for a fact that this is not the case."

    This kind of reminds me of something which might or might not be related:

    I hear many stories of kids with one immigrant parent who do not speak (or pretend not to speak) the language of said immigrant parent at all, although their parents want them to learn it. I get the impression that this is often because the kid doesn't want to stand out from their friends by speaking a "strange language".

    I have a similar mixed background, but have been speaking both of my parents' languages since birth, and I've often wondered why the above phenomenon never happened to me. I guess either I was surrounded by very open-minded kids who didn't exert monolingual peer pressure, or I was so much of a weirdo that I didn't really care what social effects being bilingual would have on me.

  45. Jason said,

    August 2, 2013 @ 5:16 am


    John McWhorter decided to exchange linguistics for a sinecure at the Manhatten Institute, where he has generated one work of political hackery after another. It's precisely because he's provided no empirical evidence that "[Black English] is mutually intelligible with standard English both on the page and spoken…" that he should be assumed to be arguing in bad faith on the matter.

    There are still some real researchers in the world. John McWhorter can continue to act as a spin doctor for the institution that hired Jason Richwine, he's obviously happier in that role anyway.

  46. Erica said,

    August 2, 2013 @ 8:21 am

    @R wrote:
    "I have a strong feeling that people are likely to report understanding less than they due to avoid identifying with "those people". My only evidence comes from some personal experience as a white person from a border state. I have relatives who claim to speak/understand no Spanish, even though I know for a fact that this is not the case."

    Though I agree some people will deny knowing any Spanish out of prejudice, or to emphasize that they don't want Spanish spoken in the US, I think sometimes that reaction ("I don't understand a word of it") is not conscious. I'm an English speaker from a border state, and I don't consciously feel dislike towards Spanish, but I usually underestimate my ability to understand it. I used to work in retail, and often had the experience of customers approaching me and asking a question in Spanish, just in case I could speak it — and I was often surprised to realize that I actually did understand, even though just as often I wasn't sure how to formulate a grammatical sentence to answer them, and had to answer in English or get a co-worker who spoke Spanish properly.

    I think many English speakers border states acquire quite a bit of passive comprehension of Spanish without actually being aware of it. I would guess the reverse happens too.

  47. J.W. Brewer said,

    August 2, 2013 @ 10:21 am

    Maybe it would be useful to consider a parallel question. How intelligible is AustEng to Americans? If you go by the Australian McDonald's ad linked a few posts back, the answer would be "not very intelligible at all." But that's an extreme example, because the fast pace of the speech makes the phonological differences more challenging for AmEng hearers than they might be at a slower pace and the whole (at least semi-humorously intended) point of the script was to load up on very distinctively opaque-to-outsiders Australian lexical items to the maximum extent possible. If by chance an AustEng speaker happened to be passing through Florida on vacation and saw something that made him a key eyewitness at a high-profile criminal trial, I suspect the testimony would not resemble either in delivery or in word choice the script of that ad. Rather, one would expect the witness to try extra-hard to be intelligible to a non-AustEng audience, the listeners to try extra-hard to comprehend the testimony, and the lawyers and perhaps even judge to actively intervene to maximize the odds of understanding by (politely) asking the witness to slow down, repeat something, or explain an unfamiliar word or phrase he had just used. The difference of course, is that there's not the same historical background, so that both speaker and listeners would be aware of a potential comprehension gap due to dialect differences without it being "anyone's fault" and without it implying that speaker and listeners were not social equals – an American tourist could end up by chance having to testify in an Australian court and you'd have the same situation in reverse. But it seems useful to try to separate out the question of "how much of an intelligibility gap exists" from the related (and equally important) question of "how will the social/cultural/historical baggage associated with the difference between the speakers of the two language varieties affect attempts to overcome any such gap."

  48. Mark Dowson said,

    August 2, 2013 @ 11:35 am

    I recall reading that, during at least the first quarter of the 20th century, there were 35 mutually incomprehensible dialects of British English, and that the later homogenization was down to the BBC, where only RP (standard) English was heard up to the early 1960's (the "35" could be eskimo-words-for-snow inflation, but may not be that far out). Even in the 1960's dialects survived in the speech of older people. For example, in 1965 at an Exeter (south west England) bus stop, as a packed bus approached, I heard one old lady say to another " Us won't get on ee"

  49. Linda Seebach said,

    August 2, 2013 @ 12:04 pm

    I was at a conference once where I was interviewing members of a high school gospel choir from Oakland, Calif. The (African American) choir members were entirely intelligible when they were speaking to me, or to other visitors to the green room, even though their speech was heavily influenced by AAVE. When they were talking among themselves, though, I couldn't understand them at all.

  50. Jenny said,

    August 2, 2013 @ 12:28 pm

    I think it is impossible to say whether Jeantel was intelligible to any given person. It will vary by person. I listened to some of her testimony and found her easy enough to understand, but I believe those who say they didn't.

    I am a well-educated white speaker of Standard American English. I grew up with a lot of friends from Vietnam, and I find those accents nearly unnoticeable. I rarely have to ask someone with such an accent to repeat anything. On the other hand, if someone has a German accent, it is rare that I will understand them the first time they say something.

    My husband, with a similar background, is the opposite. German accents are perfectly clear to him, while Vietnamese accents are nearly unintelligible.

    Those accents obviously aren't variants of English, but they are variants of pronunciation and word choice, so I think it is a similar situation.

  51. Rodger C said,

    August 2, 2013 @ 12:29 pm

    @J. W. Brewer, John: Some years ago I was watching a BBC show of the "weird sh*t" variety, and the snake-handling preacher of Jolo, WV was subtitled. I was somewhat bemused (and amused), since I could clearly understand every word he was saying.

  52. Jeff P said,

    August 2, 2013 @ 12:46 pm

    Further googling has identified the presentation I referred to above as "The Misinterpretation of African-American English in Medical Examinations" by Susan Tamasi of Emory University at the American Dialect Society's 2009 annual meeting in San Francisco.

  53. Ray Dillinger said,

    August 2, 2013 @ 1:12 pm

    One thing I'm noticing here is that most of you are talking about AAVE as though it were a single dialect.

    It isn't.

    Oakland AAVE ("ebonics") is different from east coast AAVE ("jive") is different from Kansas City AAVE("black talk") is different from New Orleans AAVE(which you may remember from coverage of Katrina(and the waves)).

    When we talk about the accents white people have, we talk about regionalisms and local markers, and we expect Appalachian Kentuckians to speak differently from Minnesotans. Having lived a few different places, I find the accents and vocabularies that African-Americans have seem to me to vary just as much by region. I don't recall having difficulty learning to understand them, but I do remember that a new learning experience (at least a few hours) was required each time before I could rely on getting it right.

    You are very correct in thinking that 'AAVE' appearing in movies, etc, is mostly written by white guys who don't necessarily speak it very well themselves, and besides are writing for white audiences to understand. The Los Angeles variety is highly overrepresented in Hollywood movies and the New York variety in TV content, mostly because of where the actors live and where many of them come from.

    Finally, actors from elsewhere who speak some other variety of English often are making a specific effort to learn "the" AAVE dialect, and learn the variety prevalent near places where movies (or TV shows) are produced.

    A woman hearing a new variety for the first time in a theater, without directly interacting with the speakers, should not necessarily be expected to pick it up by the middle of one movie.

  54. J.W. Brewer said,

    August 2, 2013 @ 1:23 pm

    Ray: that's interesting because I seem to specifically recall some of the early scholarship (maybe stuff by John Baugh written in the '70's which I would have read as an undergrad in the '80's?) making the contrary claim, namely that AAVE was remarkably uniform nationwide, all things considered. I have myself been somewhat skeptical of that uniformity claim (even setting aside the point I made earlier about the obvious counterexample of the language varieties spoken by the increasingly numerous population of black immigrants from the West Indies, Africa, and many other places), but that's been a pretty vague skepticism uninformed by much empirical evidence. (I think there has been in the traditional scholarship a carveout for Geechee/Gullah, but that's a pretty specific exception spoken by a fairly small and geographically localized population.)

  55. Xmun said,

    August 2, 2013 @ 2:27 pm

    @julie lee
    I had a similar experience in New Zealand, when I asked some Indians (people from India) what language they were speaking. They replied: "English".

  56. julie lee said,

    August 2, 2013 @ 3:38 pm

    Glad to hear you had a similar experience. I chuckle when I think about it.

  57. Chris C. said,

    August 2, 2013 @ 5:36 pm

    Mark Dowson has already adumbrated what I was going to say in response to Levantine, that pride in regional accents in the UK is largely a recent phenomenon. In previous decades it was far and away the norm for educated people to have their native accents trained out of them. I can't claim to have read more than one book specifically on the subject, but the author took the dwindling of regional UK accents as more or less a given, and was both surprised and pleased when he encountered a university professor who was not ashamed to speak like the Brummie he was.

    That a similar trend exists in the US is undeniable, but I think it's a sign that the US is catching up to the former UK habits. I have an older co-worker from Arkansas who doesn't mind speaking as if he was, and he's an Aggie to boot. On the other hand, I follow a YouTube series made by a group of former film students from Alabama. They speak with only a hint of a drawl, and have been pretty clear they consider the typical regional dialect to be "redneck". If not for that, I might have put it down to the influence of a California-based media.

  58. Levantine said,

    August 2, 2013 @ 6:04 pm

    Chris C., I didn't mean to imply that British pride in regional accents has always been there. In fact, I compared the current situation in the States to what once obtained in the UK, when RP was the standard regardless of where you came from. Since I'm only 31, I have no personal experience of that period. I hope you're right about regional accents becoming less stigmatised in the US, though I can't say I saw any evidence of it at graduate school.

  59. Chris C. said,

    August 2, 2013 @ 8:03 pm

    I meant it the other way around: that regional accents were becoming more stigmatized in the US. When a group of Alabama natives consider what was once a typical regional accent that could be heard from all socioeconomic classes to be "redneck", there's a clear stigma.

  60. Levantine said,

    August 2, 2013 @ 8:41 pm

    Ah, sorry for misunderstanding. That really would be a shame, and I hope it's not the case.

  61. David said,

    August 2, 2013 @ 9:10 pm

    @ Mark Dowson: We won't get on it?

  62. Pete B. said,

    August 2, 2013 @ 9:15 pm

    @ Rodger C: Yeah, I've noticed that they often subtitle people who aren't really that difficult to understand. I kind of feel a bit offended (for the person) when that happens.

  63. Alex Bollinger said,

    August 3, 2013 @ 8:16 am

    Maybe it's because I study a social science where we don't believe anything anyone says, but I wouldn't say that B37 found Jeantel hard to understand, but that she said that she found Jeantel hard to understand. B37 attempted to speak for the jury and the others have distanced themselves from her, so maybe she isn't a reliable source.

    Other commenters have pointed out that people often see an accent and then complain that they can't understand. It should at least be controlled for in experimental work. I just hopped over to youtube a listened to a few minutes of Jeantel's testimony and I understood her just fine (grew up in the suburbs in Indiana, went to college, then moved to France), which makes me wonder how a juror who lives in the same region would have so much trouble understanding her.

    As someone else pointed out, though, 5+ hours is a long time.

  64. Mark Dowson said,

    August 3, 2013 @ 12:34 pm

    @David: Exactly so. "Us won't get on ee" = "We won't get on it" where "ee" = "he" = "it"

  65. Mark Dowson said,

    August 3, 2013 @ 1:33 pm

    In the UK, the shift from the BBC only allowing RP English to be heard probably started with pop music programs in the early 60's, and slightly later with cop soaps. The recent BBC series "The Hour" about a 1950's TV news program gives a fairly accurate portrayal of what the BBC sounded like then. Much more recently (1990's) some British politicians (like Tony Blair) started to adopt "Estuarine English" a light version of the (already light) dialect spoken around the Thames estuary to the east of London, presumably in an attempt to sound closer to "the people" than their natural upper middle class Oxbridge educated speech would suggest. Somewhere, a John Le Carre character describes Estuarine English as "The last stages of Linguistic decay".

  66. Levantine said,

    August 3, 2013 @ 2:48 pm

    His curious unstressed vowels notwithstanding, Tony Blair still sounds pretty posh to me — not really what I think of as Estuary English (or is 'Estuarine English' something different still?). But yes, he certainly sounds closer to 'the people' than someone speaking the RP of old (or even the RP still spoken by Boris Johnson).

  67. Ken Brown said,

    August 3, 2013 @ 4:11 pm

    Totally anecdotal. I used to work for an American company, and often spoke on the phone with our office in Houston, Texas. I had difficulty understanding one person. So. I asked someone where he came from. Bombay. As soon as I knew that I could understand him. It was a normal Asian accent but I had expected him to be some kind of American and so misheard what he said. Once I knew where he was from I could follow his speech.

    FWIW I live in London and was brought up in Brighton and have a south-east English accent AKA "Estuary".

  68. Mark Dowson said,

    August 3, 2013 @ 6:17 pm

    "Estuary English" seems more common than "Estuarine" but I meant the same thing (and there is a Guardian reference to "Estuarian".

    @Ken Brown
    As a Londoner who also lived in Brighton for a while I didn't hear the local accent (what there was of it) as close to Estuary, but maybe I was associating with too many academics.

    There is a nice article on the phonetics of estuary English (by John Wells from UC London Linguistics Department) at http://www.phon.ucl.ac.uk/home/estuary/whatis.htm

  69. Levantine said,

    August 3, 2013 @ 8:54 pm

    I thought of Estuary English as the accent represented (albeit in exaggerated form) by Catherine Tate's Paul and Sam characters: http://www.youtube.com/watch?v=-YKrpzKELJU. I first became aware of this accent in secondary school when my maths teacher said the word 'scale'; I honestly thought he was saying 'scow', and was very confused until I realised my mistake.

    At the risk of veering completely off topic, the accent used by Catherine Tate's foul-mouthed grandmother character sounds to me like a wonderful study of traditional Cockney, which, with its trilled Rs and disyllabic pronunciation of words like 'poor', is no less moribund than conservative RP: http://www.youtube.com/watch?v=zLxSa-XOOKA.

  70. Anna said,

    August 4, 2013 @ 1:21 am

    I know some folks have made reference to the jive talk scene from "Airplane," but the discussion here also made me think of the exchanges between Mira and Marjorie from "Little Britain": http://www.youtube.com/watch?v=YpiAhxaE_oI.

  71. Suburbanbanshee said,

    August 4, 2013 @ 11:00 am

    Expectations make things hard to understand and hear correctly, particularly when it comes to the prosody of speech. Expectations of content do the same thing, as do expectations of word formation and name formation.

    But yes, any hearing problems, neural problems, etc. that are normally masked by the redundancies in normal-to-the-hearer speech patterns will be revealed by unfamiliar-to-the-hearer accents.

    My particular problem is choral music. I love songs, but have difficulty making out words in most pop songs. I love listening to choral music, but I routinely have trouble figuring out the language that choirs sing in, if I expect one and get the other. I don't normally have this trouble with single singers, especially if the instrumentation arrangement is spare rather than busy. Yet my hearing is supposedly normal.

  72. Mar Rojo said,

    August 4, 2013 @ 4:51 pm

    British English speaker from NW England. Never been to the States. I also had no trouble understanding Rachel`s testimony.

  73. John R. Rickford said,

    August 5, 2013 @ 12:54 pm

    Wow! 72 responses! Thanks to everyone for their feedback. Anecdotes, references, ideas–all are helping me get a grasp on this (for me) relatively new area. Here's my take on some of the points that came up:
    1. Objective/Subjective or Phonological/Affective~Attitudinal sources of misunderstanding. A recurrent theme. The Lindemann and Subtirelu 2013 reference seems like the most recent and comprehensive discussion although it's so recent (July 2013) that it's not yet available online at Stanford, so I'll have to head on over to the reading room to get it.
    2. Like the Lindemann/Subtirelu study, non-native or second-language speech seems to be the focus of a lot of relevant intelligibility studies. Especially interesting for my purposes is a study Lev-Ari and Keysar 2010 on the influence of accent on credibility (J of Experimental Social Psychology 46:1093-1096) that Stanford grad student Annette D'Onofrio told me about. Basically, non-native English speakers were heard as less credible than native speakers, and heavily accented speakers even less so than mildly accented speakers.
    3. Length of exposure point raised by Margolis and others (e.g.understanding more of a foreign accent in a movie by the end than at the beginning) is certainly relevant, as is the effect of experience with a variety of accents, raised by Duncan and many commentators.
    4. Translation/Interpretation of AAVE in court and movies: Raised by DeMarco, Levantine, Brewer, others. A friend of mine suggested I start a petition to have AAVE translated in courts in California, but the issues ARE complex, and as many of you point out, it could/would be found demeaning, as though questioning the speaker's status as "American" or "English speaker." Of course, as several of you point out, an attorney can and should anticipate and help to offset issues of unintelligibility. But I think the prosecutors did an especially bad job of this in the case of Jeantel's testimony in the Zimmerman trial. They did not even help her understand that she would get questions from prosecution and defense attorneys, with the latter more combative, as noted by Rod Eudell, the lawyer who Jeantel's family brought in to help her. Jeantel was better understood and came across more positively on the Piers Morgan show than she did in court. The latter was a more stressful environment, to be sure, but the difference is also that Morgan did a far better job of calming and preparing her (he talked about this explicitly) than prosecutor Bernie De la Rionda did.
    5. The references were especially useful. I'll write Susan Tamasi for her 2009 ADS paper. Thanks, Jeff P.
    6. Regional Variation in AAVE (Dillinger). Yes, this has been a topic of increasing attention in AAVE studies over the past few years. But the variability is most striking precisely where Dillinger notes–at the level of phonetics/phonology and lexicon. Incidentally, we know very little about Florida AAVE, and there are some unusual features of Jeantel's AAVE that may be due to region or to Haitian French Creole and Dominican Spanish influence (her parent's lgs, in which she has some degree of competence herself). We're JUST beginning to analyze her speech in the testimony and the Morgan interview, and hope to present details at forthcoming NWAV and ADS meetings…

  74. Lane said,

    August 6, 2013 @ 7:39 am

    One of the takeaway points I remember from Thinking, Fast and Slow is that we believe things that are easier to process, giving them a greater credibility than things that are harder to understand and remember. Experiments have shown this for low-contrast type (dark-gray on light-gray text was judged less reliable than crisp black-and-white) and for ryhme ("woes unite foes" strikes people as more true than "common problems unite enemies".) So yes, racial animus or just distance might have made Jeantel less credible. But this effect would have been strengthened by the mere lack of comprehension. Our lazy brains (Kahneman's "System 1") strongly prefer evidence that is easy to take in, even if it's a worse fit with the facts.

  75. Nancy Friedman said,

    August 6, 2013 @ 10:45 am

    Sometimes, to fuel my orneriness, I read the comments at WSJ.com. I usually end up in despair, but today I was rewarded with this unexpectedly germane nugget, posted in response to a feature article about the lives of middle managers in America:

    > My VP at Charles Schwab, a white man named Chris Nichols, told me that none of my top direct reports in my IT group (all Asian-Americans) could be promoted because of their "foreign accents". Thus ended my career as a middle-level manager. And I couldn't be happier. <

    (Link: http://on.wsj.com/16xhozQ )

  76. John R. Rickford said,

    August 7, 2013 @ 12:08 am

    Re Lane: Yes, processing difficulty is the intermediary variable in the Lev-Ari and Keysar 2010 study on the influence of accent on credibility (J of Experimental Social Psychology 46:1093-1096). From their abstract: "Non-native speech is harder to understand than native speech. We demonstrate that this “processing difficulty” causes non-native speakers to sound less credible." Whether linguists would reach the same conclusion is another matter.

    Re Nancy Friedman: Sounds like the Charles Schwab VP should read Lippi-Green's English With An Accent, 2011 2nd ed., esp. chap. 9, which deals with discrimination-in-the-workplace cases that were brought before the courts (11 of the 25 cases surveyed involved plaintiffs from Asia and the Pacific Rim), and chap. 15, "The unassimilable races: What it means to be Asian."

  77. Rhoda said,

    August 10, 2013 @ 2:43 am

    Julie Lee, that's what we're apparently calling MLE now, http://en.wikipedia.org/wiki/Multicultural_London_English, and it's a genuine London takeover. Mark Dowson, my Exeter mother-in-law constantly interchanges pronouns – 'her's got a new job, working with he!' And 'am yous alright?'

  78. Found while foraging (September 23, 2013) | Inspiring Science said,

    September 23, 2013 @ 8:04 am

    [...] nearly any accent of English, but I know it can be challenging for others.  Language Log raises the question of how intelligible African American Vernacular English is to speakers of Mainstream English and [...]

RSS feed for comments on this post · TrackBack URI

Leave a Comment