WEBVTT - Accents 0:00:00.040 --> 0:00:04.600 Hey, this podcast includes an unbleeped swear word that's important 0:00:04.680 --> 0:00:08.119 to a story we tell, so if you've got kids 0:00:08.280 --> 0:00:10.800 or sensitive years around, you might want to save this 0:00:10.840 --> 0:00:18.840 one for later. Olivia Coleman, the impossibly talented British actress, 0:00:19.320 --> 0:00:21.599 worked with an accent coach to prepare to play the 0:00:21.680 --> 0:00:24.960 Queen of England, and the coach explained that to pronounce 0:00:25.000 --> 0:00:28.800 the word yes, as her Majesty would, Olivia should say 0:00:28.840 --> 0:00:33.800 the word that's spelled E A R S, like the 0:00:33.880 --> 0:00:36.479 things on the side of your head. Okay, this might 0:00:36.520 --> 0:00:39.199 work slightly better if you have a British accent, but 0:00:39.240 --> 0:00:45.320 I'm gonna give it a go. Yes, yes, I love that, 0:00:45.880 --> 0:00:50.519 I love it. Okay, here's another language parlor trick may 0:00:50.520 --> 0:00:55.120 be suited better for a new world accent. Say the 0:00:55.120 --> 0:01:00.720 words rise up lights in quick succession, rise of lights, 0:01:00.920 --> 0:01:04.520 rise of lights, and boom you're an Australian sand razor blades. 0:01:08.760 --> 0:01:12.360 I am fascinated by accents and regional dialects, and I 0:01:12.360 --> 0:01:14.560 think a lot of us are. We like knowing how 0:01:14.600 --> 0:01:17.320 other people swear, the kind of slang they fling around 0:01:17.680 --> 0:01:21.720 that the American zucchini is the British Courgette that in India, 0:01:21.800 --> 0:01:25.360 the opposite of postponing something is to prepon it, to 0:01:25.480 --> 0:01:29.040 hear our own familiar language. The substrate of our thoughts 0:01:29.440 --> 0:01:32.880 made new and novel is just a little electric thrill. 0:01:35.520 --> 0:01:39.400 So how exactly our accents forged? Can they change? And 0:01:39.440 --> 0:01:42.480 how might they affect the trajectory of our lives? You 0:01:42.520 --> 0:01:45.840 are listening to deeply human and I am Dessa hereby 0:01:45.880 --> 0:01:53.840 asking the question why do you talk like that? My 0:01:53.920 --> 0:01:57.880 name is John Ball. It rhymes with however you pronounced 0:01:57.960 --> 0:02:02.760 the words l A w or or saw. That takes 0:02:02.760 --> 0:02:07.200 into account than any different regional accents. So if someone 0:02:07.320 --> 0:02:10.239 is from the South, is John Bower, you know? Or 0:02:10.280 --> 0:02:13.440 if it's a New Yorkist, John Blow? All right? So 0:02:13.520 --> 0:02:17.120 it rhymes with law and so. John ba is the 0:02:17.120 --> 0:02:20.359 president of the Linguistic Society of America and a professor 0:02:20.440 --> 0:02:24.520 at Washington University in St. Louis. Lawyers sometimes called John 0:02:24.600 --> 0:02:27.880 to ask for his help in what's called linguistic forensics, 0:02:28.240 --> 0:02:31.320 legal analysis of speech to help shed light on a crime. 0:02:32.120 --> 0:02:34.200 He's been asked to serve as an expert witness for 0:02:34.280 --> 0:02:37.520 more than one trial involving homicide, and, as he puts it, 0:02:38.000 --> 0:02:41.360 a person's accent could be the difference between life and death. 0:02:42.040 --> 0:02:44.440 Let's back up to a pivotal moment in his career. 0:02:44.840 --> 0:02:48.440 In the nineties, he became particularly interested in an insidious 0:02:48.440 --> 0:02:55.320 phenomenon now called linguistic profiling. So linguistic profiling occurs when 0:02:55.840 --> 0:03:02.119 someone calls for goods or services over the telephone and 0:03:02.400 --> 0:03:07.840 site unseen, the person who receives the call denies those 0:03:07.880 --> 0:03:12.880 goods or services based on stereotypes about the person's speech. 0:03:13.240 --> 0:03:17.160 And this interest wasn't entirely academic. John himself was making 0:03:17.200 --> 0:03:20.560 calls looking for a new apartment, and in a few instances, 0:03:20.600 --> 0:03:23.800 when I showed up in person, I was told that 0:03:23.880 --> 0:03:27.000 nothing is available. You know, and I'm African American. A 0:03:27.080 --> 0:03:29.639 lot of people can't tell that from my professional voice. 0:03:30.280 --> 0:03:34.880 And no one said, oh, I am so sorry. Had 0:03:34.960 --> 0:03:38.640 you sounded African American over the telephone, I would have 0:03:38.720 --> 0:03:41.480 never given you an appointment, right, I mean that that's 0:03:41.520 --> 0:03:45.040 a slam dunk lawsuit I win. So they would always 0:03:45.040 --> 0:03:48.200 come up with some other excuse. But you know, I 0:03:48.280 --> 0:03:54.920 was suspicious, So he decided to run a formal experiment. Yeah, 0:03:54.960 --> 0:03:57.080 I can modify my speech. So I grew up in 0:03:57.280 --> 0:04:00.040 l A you understanding the in the city, and in 0:04:00.480 --> 0:04:04.320 Los Angeles I had a lot of Mexican American friends, 0:04:04.520 --> 0:04:09.720 So in the experiment I modified my own speech and used, Hello, 0:04:09.960 --> 0:04:12.960 I'm calling about the apartment you have advertised in the paper. 0:04:13.920 --> 0:04:15.880 And I'd say Hello, I'm calling about the apartment you 0:04:15.920 --> 0:04:19.359 have advertised in the paper, or along coming about the 0:04:19.360 --> 0:04:22.440 apartment you have advertised in the paper. The sentence was 0:04:22.440 --> 0:04:25.159 always the same, and John was always the person calling, 0:04:25.279 --> 0:04:27.760 so the voice was the same. The only difference was 0:04:27.800 --> 0:04:32.039 the accent. As you might guess, the accent made a 0:04:32.080 --> 0:04:36.159 big difference. Landlords were screening out potential renters whose speech 0:04:36.200 --> 0:04:40.320 implied they were from minority communities, and as a general rule, 0:04:40.480 --> 0:04:44.360 the discrimination was worse in affluent neighborhoods. And it's international. 0:04:44.400 --> 0:04:48.680 We've done these experiments in South Africa and Brazil, in France, 0:04:49.000 --> 0:04:53.919 and you know, every place that's an advanced industrialized society 0:04:54.320 --> 0:04:59.200 has its own version of this. Linguistic profiling is also 0:04:59.240 --> 0:05:03.640 bigger than ray. Other research revealed that female voices calling 0:05:03.640 --> 0:05:06.320 to ask about an open executive position in a bank 0:05:06.600 --> 0:05:09.680 were informed that it had already been filled, whereas mail 0:05:09.720 --> 0:05:12.240 callers were more likely to be told it was still available. 0:05:12.920 --> 0:05:16.320 Gay men have also reported being denied goods or services 0:05:16.360 --> 0:05:19.760 because of the way that they speak. Okay, quick self 0:05:19.800 --> 0:05:22.400 interruption here to note that none of us speak in 0:05:22.440 --> 0:05:25.520 exactly the same way at all times. The way that 0:05:25.560 --> 0:05:28.440 you'd voice frustration at having been cut off in traffic 0:05:28.520 --> 0:05:31.320 by some ding dong in a miyata with flame decals 0:05:31.560 --> 0:05:33.880 is not the same way that you'd voice frustration in 0:05:33.920 --> 0:05:37.839 an all staff meeting. What's commonly called code switching, though 0:05:37.920 --> 0:05:41.200 john Ball might more accurately call it style shifting, is 0:05:41.240 --> 0:05:44.880 the way we change our speech based on context or company. 0:05:47.000 --> 0:05:50.800 The ability to identify even really subtle details of dialect 0:05:51.040 --> 0:05:54.400 has made John Ball a powerful expert witness and criminal court. 0:05:54.839 --> 0:05:58.160 In one particular case, a murder case, he was called 0:05:58.200 --> 0:06:01.840 in to analyze the recording of a phone call. If 0:06:01.880 --> 0:06:04.719 you're arrested and you go to jail in the United States, 0:06:05.120 --> 0:06:08.880 any phone call that you make can be recorded by 0:06:09.000 --> 0:06:12.520 law enforcement, and anything you say during that phone call 0:06:12.600 --> 0:06:14.880 can be used against you in a court of law. 0:06:15.440 --> 0:06:18.520 A crucial bit of the prosecution's case centered on the 0:06:18.560 --> 0:06:21.240 content of a call in which the defendant spoke to 0:06:21.320 --> 0:06:24.880 his cousin on a jailhouse phone. Both speakers were young 0:06:24.960 --> 0:06:29.479 black men, and so his cousin thought that it would 0:06:29.480 --> 0:06:33.159 be to his advantage to have a speedy trial, and 0:06:33.240 --> 0:06:34.920 he asked him, He said, well, why don't you do 0:06:34.960 --> 0:06:38.920 a speedy trial. The defendant's answer to this question, according 0:06:38.960 --> 0:06:43.600 to the prosecutor, was a smoking gun. The prosecutor produced 0:06:43.640 --> 0:06:47.839 a transcript that said, why would I want a speedy 0:06:47.880 --> 0:06:51.760 trial when I know I committed this ship And that 0:06:51.880 --> 0:06:56.200 for the prosecutor was the admission of guilt that the 0:06:56.240 --> 0:07:00.880 prosecutor needed in order to pursue the convict sction aggressively. 0:07:03.600 --> 0:07:06.240 But when John heard the recording, he wasn't sure that's 0:07:06.279 --> 0:07:09.320 what the defendants said at all. He heard a totally 0:07:09.360 --> 0:07:12.320 different sentence, which will explain in a moment, because a 0:07:12.360 --> 0:07:16.160 particular vowel caught his ear. Both speakers were using what 0:07:16.280 --> 0:07:20.320 scholars referred to as African American vernacular English, and John 0:07:20.360 --> 0:07:23.000 wondered if the prosecutor just didn't know how to interpret it. 0:07:23.400 --> 0:07:28.120 Maybe the confession wasn't a confession at all. The prosecutor 0:07:28.280 --> 0:07:31.440 likely wasn't very well acquainted with a dialect used by 0:07:31.440 --> 0:07:34.560 the young man on trial. Our accents and patterns of 0:07:34.600 --> 0:07:38.360 speech can vary dramatically, even for people who live very 0:07:38.400 --> 0:07:42.760 near one another, and often those differences are perpetuated on purpose. 0:07:44.840 --> 0:07:50.080 People from Boston don't want to sound like people for 0:07:50.240 --> 0:07:53.400 New York. Right, If you're a Red Sox fan, you 0:07:53.400 --> 0:07:55.960 don't want to sound like a Yankees fan. That is 0:07:56.000 --> 0:08:00.680 Aaron Dinkin. He teaches socio linguistics at San Diego State University, 0:08:00.960 --> 0:08:04.360 and he focuses on variations in American English dialects and 0:08:04.400 --> 0:08:08.920 how they change over time, features of linguistic variation, just 0:08:09.200 --> 0:08:14.800 get all of this subconscious social freight assigned to them, 0:08:14.840 --> 0:08:17.800 not just in terms of regional dialects, but also in 0:08:17.960 --> 0:08:24.200 terms of gender and social class and ethnicity. That word 0:08:24.400 --> 0:08:27.800 dialect describes all the features of a manner of speaking 0:08:27.800 --> 0:08:33.080 a language, our pronunciation, rhythms, melodies, and the particularities of 0:08:33.120 --> 0:08:36.760 our vocabularies. So think of the British lift versus the 0:08:36.760 --> 0:08:42.000 American elevator, or the British adjective hinch versus the American honky, 0:08:42.280 --> 0:08:46.600 the British fly tipper versus the American. Okay, admittedly, I 0:08:46.760 --> 0:08:49.360 just learned that word this morning, and I think it's 0:08:49.400 --> 0:08:52.240 a person who dumps a lot of trash without paying 0:08:52.240 --> 0:08:55.560 for proper disposal. I'm not sure we have those in America. Well, 0:08:55.559 --> 0:08:57.400 I'm sure we have people who do that, but I 0:08:57.480 --> 0:08:59.880 don't think we've given them a proper name, So the 0:09:00.200 --> 0:09:04.000 just call them garbage scoundrels. Anyway, your accent is shaped 0:09:04.120 --> 0:09:06.600 not only by where you're from, but by all sorts 0:09:06.600 --> 0:09:10.440 of super fine tuned markers. Aaron told me about this 0:09:10.480 --> 0:09:14.319 one study by a researcher named Suzanne Wagner that illustrates 0:09:14.320 --> 0:09:16.920 how a single vowel sound can serve as a hook 0:09:17.080 --> 0:09:19.559 on which all sorts of identities are home. So we're 0:09:19.600 --> 0:09:23.800 talking about the long eye vowel in words like fight 0:09:24.080 --> 0:09:29.800 and price and rice. It has been found that statistically 0:09:30.360 --> 0:09:34.280 in Philadelphia, there is a change in the pronunciation of 0:09:34.280 --> 0:09:38.800 that taking place where what was originally I as in 0:09:38.960 --> 0:09:43.760 fight and rice is changing more towards that I as 0:09:43.800 --> 0:09:48.920 in fight and rice. And it has been found that 0:09:49.559 --> 0:09:53.400 again statistically, not like a hard and fast distinction, but 0:09:53.440 --> 0:09:57.280 just you know, on average, men are more likely to 0:09:57.280 --> 0:10:00.200 pronounce the vowel a little bit closer to A as 0:10:00.240 --> 0:10:03.000 in fight and rice, and women are more likely to 0:10:03.080 --> 0:10:06.600 pronounce it a little bit less far along that curve 0:10:06.960 --> 0:10:10.520 and it sounds so brey right, so it got that 0:10:10.559 --> 0:10:15.800 associated with it. But she found that that I pronunciation 0:10:16.640 --> 0:10:21.320 was a little bit more common among the girls who 0:10:21.400 --> 0:10:29.240 were less invested in conforming to conventional femininity and who 0:10:29.280 --> 0:10:34.720 were more invested in like appearing tough, or the ones 0:10:34.720 --> 0:10:40.520 who were more likely to get into fights, which is 0:10:40.559 --> 0:10:44.080 perceived as a masculine thing. And there was an ethnic 0:10:44.200 --> 0:10:48.480 correlation as well, because for the most part, the more 0:10:48.559 --> 0:10:52.880 conventionally feminine girls in this neighborhood in South Philly were 0:10:53.120 --> 0:10:59.240 Italian American and the Irish American girls were more interested 0:10:59.480 --> 0:11:07.320 in seeming tough. Word broy freaking. This one vowel, even 0:11:07.360 --> 0:11:10.880 just a little, really meant something. It meant something about 0:11:10.960 --> 0:11:14.120 Irish roots versus Italian ones. It meant something about the 0:11:14.160 --> 0:11:16.400 relative merit of trying to be fem versus trying to 0:11:16.440 --> 0:11:18.880 be tough, and the girls knew it. I read some 0:11:18.920 --> 0:11:22.480 of their research interviews. It was like two worldviews competing 0:11:22.480 --> 0:11:25.520 for dominance in that little dot above a lower case I. 0:11:26.760 --> 0:11:29.640 The vowel change in South Philadelphia is a subtle one, 0:11:30.160 --> 0:11:33.600 but accents can diverge enough to cause confusion between speakers 0:11:33.600 --> 0:11:36.280 of the same language. If I were watching a movie 0:11:36.320 --> 0:11:39.680 whose character spoken like Scottish brogue or something, I'd probably 0:11:39.679 --> 0:11:41.800 have to turn on the subtitles to track the action. 0:11:42.600 --> 0:11:45.600 Which brings us back to John Baw, the expert witness 0:11:45.600 --> 0:11:49.240 in the murder case and the defendant speaking African American 0:11:49.360 --> 0:11:55.280 Vernacular English on a call with his cousin. To study 0:11:55.320 --> 0:11:58.160 the roots of that dialect is to understand the extreme 0:11:58.200 --> 0:12:05.720 pressures that forged it. The linguistic heritage of slaves in 0:12:05.760 --> 0:12:09.800 the United States is unique in comparison to every other group. 0:12:10.559 --> 0:12:16.160 Slave traders separated slaves whenever they could based on language. 0:12:16.360 --> 0:12:19.280 Beginning in the slave factories on the West coast of Africa, 0:12:19.640 --> 0:12:22.960 people who spoke the same language were forcibly kept apart. 0:12:24.240 --> 0:12:27.680 If you don't share the same language, then you're less 0:12:27.720 --> 0:12:31.400 likely to be able to, you know, foment an insurrection. 0:12:32.400 --> 0:12:39.480 So that explains why no African language survived the Atlantic crossing. Intact, 0:12:39.640 --> 0:12:42.400 We've got lots of communities where Polish is spoken, or 0:12:42.480 --> 0:12:45.760 German is spoken, or Italian is spoken, but there's no 0:12:46.120 --> 0:12:50.160 community where TWEE survived. And then it was illegal to 0:12:50.200 --> 0:12:54.640 teach slaves to read and write African American vernacular English 0:12:54.760 --> 0:12:56.960 can differ from the kind of English that's taught in 0:12:57.040 --> 0:13:01.359 textbooks in many ways, in part because of this interrupted heritage, 0:13:01.720 --> 0:13:03.800 and it's also the product of a lot of really 0:13:03.920 --> 0:13:07.240 varied linguistic influences. If I say i'll be i'll be done, 0:13:07.280 --> 0:13:10.439 told this story a thousand times, Okay, That use of 0:13:10.559 --> 0:13:14.839 be done can be traced back to Scott's Irish dialects 0:13:14.880 --> 0:13:18.000 and the indentured servants that introduced that to the slaves. 0:13:18.559 --> 0:13:22.880 The dialect that's considered standard or correct, or the Queen's English, 0:13:23.000 --> 0:13:25.040 or whatever you want to call it, that kind of 0:13:25.080 --> 0:13:28.200 status is really usually just the product of the fact 0:13:28.240 --> 0:13:30.760 that the classes in power consider their own way of 0:13:30.800 --> 0:13:36.079 speaking as correct. But linguists recognize that, at least from 0:13:36.120 --> 0:13:39.760 a purely theoretical point of view, all languages and dialects 0:13:39.760 --> 0:13:43.840 are equal, but they do vary. And the dialect spoken 0:13:43.840 --> 0:13:45.960 by the defendant on trial for murder was not the 0:13:46.000 --> 0:13:49.680 same as the one spoken by the prosecutor. John took 0:13:49.720 --> 0:13:52.960 the taped call, which included the so called confession, to 0:13:53.080 --> 0:13:56.319 his language lab and he slowed it down to analyze 0:13:56.320 --> 0:13:59.440 every sound, and he began some research with speakers of 0:13:59.480 --> 0:14:04.760 the same dialect. I conducted some experiments with African American 0:14:04.880 --> 0:14:09.400 men similar background, same age, who were listening to the 0:14:09.440 --> 0:14:13.240 recording under headphones, and I had them produce at the 0:14:13.280 --> 0:14:17.000 same pace. You know, I know I committed this ship 0:14:17.240 --> 0:14:21.000 versus I know I ain't committed this ship. Okay, let's 0:14:21.000 --> 0:14:24.280 break that down. That one syllable from the defendant is 0:14:24.320 --> 0:14:28.680 the crucial detail on which John's testimony hinges and in actuality, 0:14:28.760 --> 0:14:31.320 what he said in black dialect was whim a dual 0:14:31.360 --> 0:14:34.520 speedy trial, want to know and committed this ship now 0:14:34.680 --> 0:14:40.040 when I know I ain't committed is a nasalized dip 0:14:40.160 --> 0:14:43.800 thong without the tea of the word ain't. What the 0:14:43.840 --> 0:14:47.440 prosecutor had taken for a confession, John understood is a 0:14:47.480 --> 0:14:50.760 statement of innocence. He took the stand and shared that 0:14:50.800 --> 0:14:53.560 analysis with the jury, and he presented them with an 0:14:53.600 --> 0:14:56.440 image called a voice print, a graphic that showed the 0:14:56.480 --> 0:14:59.960 sound waves of both sentences. You can see the different 0:15:00.000 --> 0:15:03.280 it's more clearly than you can hear the difference. After 0:15:03.320 --> 0:15:06.320 all the evidence was presented, the jurors went to deliberate, 0:15:06.680 --> 0:15:09.960 and the stakes of the verdict were really, really high. 0:15:10.120 --> 0:15:14.000 The crime that was committed was one where the defendant 0:15:14.560 --> 0:15:21.520 was eligible for execution, so essentially the single nasalized vowel 0:15:21.640 --> 0:15:25.160 sound potentially could be the difference between life and death. 0:15:26.040 --> 0:15:28.480 The jury was asked not only for a verdict, but 0:15:28.560 --> 0:15:31.960 to decide whether to impose capital punishment a death sentence 0:15:32.080 --> 0:15:35.800 if they found him guilty. In the end, the jurors 0:15:35.840 --> 0:15:39.640 didn't vote to equit the defendant. They did find him guilty, 0:15:39.680 --> 0:15:44.680 but the jury decided not to invoke the death penalty. 0:15:44.880 --> 0:15:48.440 So it could well be that John's testimony, undermining the 0:15:48.520 --> 0:15:52.600 otherwise damning confession, stopped them from issuing a death sentence. 0:16:00.040 --> 0:16:03.480 Your particular dialect, which can shape your life and reflect 0:16:03.520 --> 0:16:06.920 all sorts of facets of your identity, probably started to 0:16:07.000 --> 0:16:13.520 gel before you learned your first word. Okay, we get 0:16:13.560 --> 0:16:15.560 the idea. You can really only take so much of 0:16:15.600 --> 0:16:20.080 that anyway. Researchers led by Kathleen Vermke at the University 0:16:20.080 --> 0:16:22.960 of Wurtzburg in Germany have found that infants cry and 0:16:23.040 --> 0:16:25.520 the melody of the language they've heard while in the womb, 0:16:26.200 --> 0:16:29.480 so French babies tend to produce cries with the rising 0:16:29.520 --> 0:16:33.640 melody contour, whereas German babies produce more falling ones. Both 0:16:33.680 --> 0:16:37.720 patterns distinguishable from babies crying in Mandarin. But as soon 0:16:37.760 --> 0:16:40.000 as we're old enough to spend time with our peers, 0:16:40.040 --> 0:16:43.720 the parental influence on our language quickly dims in comparison 0:16:43.760 --> 0:16:47.040 to that of our own social groups. Back to socio 0:16:47.120 --> 0:16:52.480 linguist erindiccken to a kindergartener. If there is nobody cooler 0:16:52.520 --> 0:16:56.120 than a first grader, right, So that's who becomes your 0:16:56.160 --> 0:17:01.440 new lingualistic models. But the profits of language teenagers, the 0:17:01.480 --> 0:17:05.000 way that like fifteen, sixteen, seventeen year old speak, that 0:17:05.240 --> 0:17:07.879 is what's in the crystal ball, the nascent trends that 0:17:07.920 --> 0:17:11.240 will shape the speech of the future, and about of 0:17:11.320 --> 0:17:17.280 language innovations are first used by women or girls. So 0:17:18.040 --> 0:17:22.120 what changes are currently underway? What sort of language forecast 0:17:22.160 --> 0:17:25.119 are we in for? Well, in the North American corner 0:17:25.160 --> 0:17:27.960 of the map, there is a front coming in. It's 0:17:28.000 --> 0:17:31.280 been called the California shift, it's been called the third 0:17:31.320 --> 0:17:34.720 dialect shift, it's been called the Canadian shift, the short 0:17:34.840 --> 0:17:38.480 front vowel shift, or the low back Merger shift. So 0:17:38.760 --> 0:17:42.000 I'm talking about the short O vowel as in c 0:17:42.240 --> 0:17:46.840 ot cot and then this other vowel as in c 0:17:47.160 --> 0:17:53.720 A U g h T cont and since I'm from Massachusetts, 0:17:53.960 --> 0:17:56.639 I pronounced those two the same already. I say caught 0:17:56.720 --> 0:18:00.800 and caught, but somebody from New York City, for example, 0:18:01.080 --> 0:18:04.119 might say c ot cot and c a U g 0:18:04.359 --> 0:18:09.199 h T court. Essentially, two distinct vowel sounds are collapsing 0:18:09.280 --> 0:18:12.320 into one single sound, and this is creating a chain 0:18:12.400 --> 0:18:16.960 reaction all over Vowelville. So other vowels are essentially shifting 0:18:17.240 --> 0:18:20.320 to fill in the space left by this convergence. Think 0:18:20.359 --> 0:18:23.919 of how tectonic plates shift and collide and push one 0:18:23.920 --> 0:18:28.280 another around. So the third dialect shift involves the short 0:18:28.359 --> 0:18:32.400 a as in the cat and the hat, moving backwards 0:18:32.920 --> 0:18:35.960 towards like the cat and the hot the e and 0:18:36.080 --> 0:18:39.240 dress and the eye and fit those are morphing too, 0:18:39.680 --> 0:18:44.080 So that's what the third dialect shift is. The short A, 0:18:44.320 --> 0:18:48.120 short E, and short eye vowels all moving kind of 0:18:48.560 --> 0:18:54.240 down and back in the geography of the mouth. So 0:18:54.640 --> 0:18:57.720 maybe when we hit the ten thousandth episode of Deeply 0:18:57.800 --> 0:19:01.960 Human will be calling it a podcast as about human behavior, 0:19:02.640 --> 0:19:06.879 I don't talk about the vowel sounds in English seem 0:19:06.960 --> 0:19:10.600 particularly susceptible to this kind of evolution. This may be 0:19:10.600 --> 0:19:13.679 because English, as compared to other languages, has a bunch 0:19:13.760 --> 0:19:18.280 of vowel sounds somewhere between eleven and sixteen, which means 0:19:18.280 --> 0:19:20.160 that each of them has sort of a narrow lane, 0:19:20.680 --> 0:19:22.440 like if you move your tongue just a little bit 0:19:22.480 --> 0:19:26.639 one way or the other, you wander into the neighboring sound. Spanish, 0:19:26.680 --> 0:19:30.040 by comparison, has only five vowel sounds, and Aaron says 0:19:30.080 --> 0:19:32.400 that some of those have been stable for as long 0:19:32.440 --> 0:19:36.040 as linguists have been able to trace their history. Okay, 0:19:36.440 --> 0:19:45.480 speaking of foreign languages, James, that's the signal, James, my guy, 0:19:46.160 --> 0:19:50.240 play the next clip. My name's Katie Harris, thank you, 0:19:50.600 --> 0:19:54.200 and I run a website called Drew of Languages, which 0:19:54.560 --> 0:19:59.160 is where we help adults learn a foreign language by 0:19:59.200 --> 0:20:02.879 making it far and making the grammar explanations really really simple. 0:20:03.400 --> 0:20:05.760 Katie lives in Italy, and I'm told that her Italian 0:20:05.800 --> 0:20:09.919 accent is pretty think good, like almost native. But it 0:20:10.119 --> 0:20:14.480 was not always. So I've been living in Italy for 0:20:14.520 --> 0:20:17.200 a year, nobody had ever told me about the difference 0:20:17.240 --> 0:20:23.119 between single and double sounds. So in Italian there's this 0:20:23.200 --> 0:20:29.160 difference between how long you hold a sound. So, for example, 0:20:29.359 --> 0:20:32.760 if you say I'm not with two ends, then that 0:20:32.800 --> 0:20:36.280 means year. But if you say are not with one 0:20:36.520 --> 0:20:41.639 end a shorter end, then that means anus. So it 0:20:41.680 --> 0:20:45.119 can be quite dangerous, especially because in Italian, so to 0:20:45.160 --> 0:20:47.919 say your age, it's like Spanish, where you say I 0:20:48.000 --> 0:20:51.080 have thirty five years. So I thought I'd been walking 0:20:51.080 --> 0:20:53.399 around telling people that. Well at the time, you know, 0:20:53.520 --> 0:21:01.640 I had twenty three anuses. It's unfortunate. Yeah, Katie, who 0:21:01.680 --> 0:21:05.280 I will note also speak Spanish and French and some 0:21:05.480 --> 0:21:08.680 German and some Mandarin, had to really focus on getting 0:21:08.680 --> 0:21:12.200 her Italian accent down, whereas it comes more easily to 0:21:12.480 --> 0:21:17.199 very young humans. Babies when they're born are able to fully, 0:21:18.040 --> 0:21:21.120 of course, perceive the difference between any sound in any 0:21:21.119 --> 0:21:24.119 of the world's languages. It's almost as if their brain 0:21:25.240 --> 0:21:28.240 starts to lock on to the first language that they learned, 0:21:28.400 --> 0:21:31.879 so that then it becomes more difficult to hear the 0:21:31.920 --> 0:21:35.240 difference between other sounds and therefore pronounce them as well. 0:21:35.960 --> 0:21:39.440 If you focus on learning the sounds of a language 0:21:39.440 --> 0:21:41.440 in the same way that you focus on learning the words, 0:21:42.160 --> 0:21:45.680 it's quite possible to train yourself to pronounce new sounds. 0:21:46.280 --> 0:21:48.040 You have to be patient because it's a bit like 0:21:48.080 --> 0:21:51.719 going to the gym. You have to change the muscles 0:21:51.720 --> 0:21:53.560 in your mouth, so that takes a little bit of time, 0:21:53.600 --> 0:21:58.560 but it's absolutely possible to learn them. I spoke to 0:21:58.600 --> 0:22:01.320 an accent coach, Jack Alas, who said that part of 0:22:01.320 --> 0:22:03.919 the challenge is just to make people aware of all 0:22:03.920 --> 0:22:07.359 their muscles. So sometimes teachers will ask students to roll 0:22:07.440 --> 0:22:09.800 a blueberry on their tongue just to get a sense 0:22:09.840 --> 0:22:12.560 of what's really going on in there. Being a language 0:22:12.560 --> 0:22:15.800 teacher and learner herself, Katie is dialed into all the 0:22:15.840 --> 0:22:20.959 tiny variations of speech sounds around her. For example, my 0:22:21.000 --> 0:22:25.679 partner speaks wonderful English. He's Italian, but he still has 0:22:25.680 --> 0:22:28.960 an Italian accent. He asked me about the pronunciation of things, 0:22:29.040 --> 0:22:33.480 for example, saying, the Italians when they speak English, tend 0:22:33.520 --> 0:22:37.160 to say because they don't have the sound, and if 0:22:37.200 --> 0:22:40.159 we just sit and I explain how to make it, 0:22:40.280 --> 0:22:44.199 he can make it perfectly. It sounds native. And then 0:22:44.240 --> 0:22:47.240 I asked him why he doesn't actually do it when 0:22:47.240 --> 0:22:50.080 he's speaking English, and he said, it feels silly. I 0:22:50.119 --> 0:22:52.040 feel like I'm trying to pretend to be somebody that 0:22:52.080 --> 0:22:54.520 I'm not. I don't want to pretend to be an 0:22:54.520 --> 0:22:58.080 English person. I'm Italian. I'm happy to sound like an Italian. 0:22:58.119 --> 0:23:01.760 He's speaking English. Okay. By this point you've probably picked 0:23:01.800 --> 0:23:04.639 up on the fact that identity is a big theme here. So, 0:23:04.680 --> 0:23:08.119 for example, if I think about chefs who might have 0:23:08.119 --> 0:23:12.400 been living in the country for twenty thirty years, say 0:23:12.400 --> 0:23:15.680 an Italian chef living in the US. For some reason, 0:23:15.680 --> 0:23:18.239 I've always noticed that chefs always really tend to keep 0:23:18.280 --> 0:23:20.760 their accents, even if they've been living there for decades. 0:23:21.440 --> 0:23:24.480 I think it's probably because it's useful to them. They 0:23:24.520 --> 0:23:28.119 have no reason to adopt a new local identity, because 0:23:28.520 --> 0:23:30.879 the stronger their accent is, the more that people can 0:23:30.920 --> 0:23:33.399 associate them with where they come from. So you know, 0:23:33.520 --> 0:23:37.680 I personally would trust a pizza made by an Italian 0:23:38.160 --> 0:23:41.080 chef that has a stronger Italian accent than they had 0:23:41.080 --> 0:23:45.359 an American accent. Why you talk the way you do, 0:23:45.680 --> 0:23:48.919 there's a product of a lifetime of layered factors. The 0:23:49.000 --> 0:23:51.120 melodies of speech that make their way to you while 0:23:51.160 --> 0:23:54.479 you're still an incubation, the words and phrases learned from 0:23:54.480 --> 0:23:57.240 your parents, the rough and tumbled talk of the first 0:23:57.240 --> 0:24:01.199 greater jet set, and then in humorable, fine layers of 0:24:01.240 --> 0:24:04.119 identity informed by the slang at the girl's table, in 0:24:04.160 --> 0:24:06.840 the lunch room, the talk in the brake room, maybe 0:24:06.840 --> 0:24:09.640 the language of the Sunday sermon. You're like a walking 0:24:09.760 --> 0:24:13.560 archaeological dig of language, and of course they are your 0:24:13.560 --> 0:24:16.720 own aspirations to speak like the person you'd like to be. 0:24:17.160 --> 0:24:21.199 My job as a scientist is to try to have 0:24:21.440 --> 0:24:28.000 people appreciate and respect that someone who's linguistic background is 0:24:28.040 --> 0:24:33.879 substantially different from your own is equally worthy of your respect, 0:24:34.680 --> 0:24:39.800 your kindness, and the assumption that they may be just 0:24:39.960 --> 0:24:44.560 as intelligent and just as capable as those people that 0:24:44.640 --> 0:24:49.359 you prefer in your own linguistic comfort zone. Snaps Wow, 0:24:49.640 --> 0:24:53.320 landed Man, Well, let me give you your shoutout girlfriend, 0:24:53.320 --> 0:24:58.320 because you are good. What kind of maniacal, self celebratory 0:24:58.400 --> 0:25:02.320 podcast host would include that complimentary bit of audio. If 0:25:02.359 --> 0:25:12.280 you guessed this guy, yes you are correct, Watch out Olivia. Also, 0:25:12.440 --> 0:25:15.359 let me share a quote from author Amy Chua, but 0:25:15.440 --> 0:25:17.760 struck a chord with me. Do you know what a 0:25:17.800 --> 0:25:23.320 foreign accent is? It's a sign of bravery. Okay, let 0:25:23.359 --> 0:25:26.679 the record show what a highbrow podcast you are listening to, 0:25:27.000 --> 0:25:28.920 and that we got almost all the way to the 0:25:29.000 --> 0:25:31.879 end of this thing before even mentioning the fact that 0:25:32.080 --> 0:25:39.800 some accents are just really sexy, like unwholesomely charged with 0:25:39.960 --> 0:25:44.040 pure animal Hang on, we'll come back to that in 0:25:44.080 --> 0:25:48.520 a second. Got into the credits first. Deeply Human is 0:25:48.560 --> 0:25:52.359 a BBC World Service and American Public Media co production 0:25:52.480 --> 0:25:56.440 with I Heart Media, and it's hosted by Tessa. Find 0:25:56.480 --> 0:26:05.760 her online on Instagram, Darling on Twitter. My French friend 0:26:05.800 --> 0:26:10.399 who lives in Milatin. When she was in the States 0:26:10.440 --> 0:26:14.040 for a while, she found the perfect pickup line was 0:26:14.119 --> 0:26:19.600 to say, I'm French, but they live in Milan. Is 0:26:19.640 --> 0:26:26.320 she single? Yes? But she wasn't when she was in 0:26:26.359 --> 0:26:29.040 the States because it was so easy. Everyone would melt 0:26:29.240 --> 0:26:35.919