WEBVTT - The Whistling, Part 2 0:00:00.160 --> 0:00:05.280 My Welcome to Stuff to Blow Your Mind production of 0:00:05.320 --> 0:00:14.319 My Heart Radio. Hey you welcome to Stuff to Blow 0:00:14.360 --> 0:00:17.560 your Mind. My name is Robert Lamb, and I'm Joe McCormick, 0:00:17.640 --> 0:00:21.680 and we're back with part two of our series on whistling. 0:00:22.400 --> 0:00:25.640 In the last episode, we talked a bit about how 0:00:25.680 --> 0:00:29.080 whistling works physically, what happens when you're creating a sort 0:00:29.080 --> 0:00:33.479 of resonator cavity within the mouth. We also talked about 0:00:33.720 --> 0:00:38.480 the whistle speech of the Maze Taco languages in Wahaka 0:00:38.560 --> 0:00:42.120 in Mexico, and I wanted to start off today's episode 0:00:42.240 --> 0:00:46.680 by talking about some other examples of whistled languages and 0:00:46.760 --> 0:00:49.520 some of the common characteristics between them, because, of course, 0:00:50.240 --> 0:00:53.240 the Maze Teco whistle speech is not the only example 0:00:53.720 --> 0:00:58.160 of a whistled language that carries information. In fact, I 0:00:58.200 --> 0:01:00.880 was looking at a paper by an author named Julian 0:01:01.000 --> 0:01:06.200 Meyer called Environmental and Linguistic Typology of Whistled Languages in 0:01:06.240 --> 0:01:10.199 the Annual Review of Linguistics. It's a very recent paper, 0:01:10.880 --> 0:01:14.640 uh And according to Meyer, there are reports of more 0:01:14.680 --> 0:01:19.720 than eighty languages around the world that contain a whistled lexicon, 0:01:20.360 --> 0:01:23.600 and about half of those have been confirmed by formal 0:01:23.640 --> 0:01:27.800 studies and published recordings, so really solid documentation of at 0:01:27.880 --> 0:01:32.080 least forty or so whistled languages around the world, And 0:01:32.120 --> 0:01:34.720 so I think it's worth mentioning a few more examples 0:01:34.760 --> 0:01:37.160 of these and describing how they work and seeing what 0:01:37.200 --> 0:01:41.759 we can compare and contrast with them. So one example 0:01:41.800 --> 0:01:45.400 I was reading about was in a really interesting article 0:01:45.640 --> 0:01:50.960 in BBC Travel by Elliott Stein, and the story here 0:01:51.000 --> 0:01:55.280 goes like this. In Greece, there is a remote mountain 0:01:55.400 --> 0:02:00.200 village called Antia, which is found on the southern an 0:02:00.200 --> 0:02:03.800 eastern coast of the Greek island of Evia in the 0:02:03.880 --> 0:02:08.360 Aegean c And within this village there has long been 0:02:08.600 --> 0:02:13.920 a whistle based language called Sphyria, which allows speakers to 0:02:14.120 --> 0:02:17.560 communicate across great distances, and it seems to have been 0:02:17.600 --> 0:02:21.080 passed down from parents to children among the shepherds and 0:02:21.120 --> 0:02:24.720 the farmers of the village for literally thousands of years, 0:02:24.760 --> 0:02:27.639 for more than two thousand years to read from Stein 0:02:27.720 --> 0:02:31.399 here quote. But in the last few decades, antias population 0:02:31.480 --> 0:02:34.840 has dwindled from two hundred and fifty to thirty seven, 0:02:35.240 --> 0:02:38.840 and as older whistlers lose their teeth, many can no 0:02:38.919 --> 0:02:43.000 longer sound sphery as Sharp notes today there are only 0:02:43.120 --> 0:02:46.160 six people left on the planet who can still speak 0:02:46.400 --> 0:02:50.040 this unspoken language. Now this was five years ago as 0:02:50.080 --> 0:02:52.400 of this recording, so I don't know how that number 0:02:52.440 --> 0:02:55.440 six has changed since then. There there are descriptions of 0:02:55.480 --> 0:02:58.840 some efforts to try to teach it to more people. Um, 0:02:59.280 --> 0:03:02.160 but of whenever you're talking about a language with that 0:03:02.320 --> 0:03:06.920 few speakers, it it's certainly extremely endangered. In fact, this 0:03:07.000 --> 0:03:10.919 is considered one of the most endangered languages in the world. Now. 0:03:10.960 --> 0:03:14.760 Apparently the existence of this language sphery A here was 0:03:15.000 --> 0:03:19.400 not documented anywhere in the outside world until the year 0:03:19.480 --> 0:03:23.040 nineteen sixty nine, when a plane crashed in the mountains 0:03:23.080 --> 0:03:26.359 nearby and there was a rescue team that was attempting 0:03:26.360 --> 0:03:30.799 to locate the pilot and they reported hearing strange, melodious 0:03:30.800 --> 0:03:35.440 whistling echoing through the hillsides. And this led to investigation 0:03:35.560 --> 0:03:38.240 and brought the language to the attention of the media 0:03:38.440 --> 0:03:42.120 and to academics. Uh So, a big question here is 0:03:42.200 --> 0:03:45.560 where does a language like this come from. Linguists do 0:03:45.760 --> 0:03:49.160 seem to agree that it dates back to ancient times. 0:03:49.200 --> 0:03:51.520 It's been around for a long time, but exactly how 0:03:51.560 --> 0:03:55.720 it was created is less certain, uh and apparently local 0:03:55.880 --> 0:04:00.120 legends abound. So one story I came across this was 0:04:00.160 --> 0:04:04.080 described in in some detail in a documentary piece on 0:04:04.120 --> 0:04:08.440 PBS News Hour that was about Sphyria, and it claimed 0:04:08.520 --> 0:04:12.920 that the language was invented about two thousand five years ago, 0:04:13.400 --> 0:04:17.880 not by Greeks, but by Persians after they were defeated 0:04:18.080 --> 0:04:21.920 at the Battle of Salamis. So Salamus was a battle 0:04:22.160 --> 0:04:25.839 in four a d b c. During the Persian invasion 0:04:25.880 --> 0:04:29.560 of Greece under zerk Sees the Great, and so Salamus 0:04:29.600 --> 0:04:32.080 was a It was a naval battle where the coalition 0:04:32.120 --> 0:04:34.839 of Greek city states was able to fight off and 0:04:34.880 --> 0:04:38.839 defeat the larger Persian allied fleet. And I think this 0:04:38.960 --> 0:04:41.719 is widely considered the battle, or one of the battles 0:04:41.760 --> 0:04:44.560 that turned the tide of the war in favor of 0:04:44.560 --> 0:04:48.560 the Greek defenders and pushed back the Persian invasion. But anyway, 0:04:48.680 --> 0:04:51.960 the legend about the whistle speech goes that the Persian 0:04:52.120 --> 0:04:55.680 survivors of the battle I guess they were, you know, 0:04:55.720 --> 0:04:58.279 their their ship was sank or defeated in some way 0:04:58.279 --> 0:05:00.599 and they managed to swim to shore on the island 0:05:00.640 --> 0:05:03.800 of Evia, where they had to survive hiding in the 0:05:03.839 --> 0:05:08.960 mountains inhabited by hostile native Greeks, and one way they 0:05:09.000 --> 0:05:12.560 avoided detection was by coming up with a way of 0:05:12.600 --> 0:05:15.600 speaking in whistles that would sound just like the birds, 0:05:15.640 --> 0:05:18.200 so they could they could communicate with each other, but 0:05:18.360 --> 0:05:22.080 their speech would not be intelligible and in many ways 0:05:22.120 --> 0:05:25.839 would probably not even be detected. But that that story 0:05:25.839 --> 0:05:27.840 has a kind of legendary quality. I'm not sure how 0:05:27.920 --> 0:05:30.279 much there is behind that so, but but it's a 0:05:30.279 --> 0:05:34.160 great story nonetheless, and and steinsite some other local legends 0:05:34.160 --> 0:05:37.520 as well. Some residents believe it was invented during the 0:05:37.560 --> 0:05:40.640 Byzantine Empire by locals who wanted a secret way to 0:05:40.680 --> 0:05:44.080 communicate that would elude the understanding of pirates and people 0:05:44.120 --> 0:05:47.719 from hostile nearby villages. And so a common theme here 0:05:47.760 --> 0:05:50.120 seems to be the idea that somehow this language was 0:05:50.240 --> 0:05:53.279 created to be a secret way of communicating, to allow 0:05:53.320 --> 0:05:57.159 the locals to communicate across distance and understand each other 0:05:57.760 --> 0:06:01.920 without other people detecting or understanding what they were saying. Now, 0:06:01.960 --> 0:06:05.320 the author of this BBC Travel piece describes visiting the 0:06:05.400 --> 0:06:08.400 village and spending time with with the handful of people 0:06:08.440 --> 0:06:12.560 there who still use the whistle language, and they apparently 0:06:12.640 --> 0:06:15.800 use it in many of the same scenarios described in 0:06:15.839 --> 0:06:18.359 that paper on the Massa Teco whistle speech that I 0:06:18.440 --> 0:06:22.000 talked about in the last episode, A big scenario of 0:06:22.080 --> 0:06:26.880 you seems to be communicating across great distance on the 0:06:26.920 --> 0:06:30.760 mountain side and sort of greeting or summoning people from 0:06:30.839 --> 0:06:36.080 far away. And Stein cites a Greek linguist named Demitra 0:06:36.200 --> 0:06:40.560 Hanggen who studies Spheria, and she says that Sphoria is 0:06:40.600 --> 0:06:45.200 in some sense a whistled version of spoken Greek, where 0:06:45.480 --> 0:06:51.320 specific whistled tones correspond to specific phonetic syllables or letters, 0:06:51.680 --> 0:06:55.240 and you can build words out of them. Now, again, 0:06:55.480 --> 0:06:57.520 this is another way that it's similar to the Massa 0:06:57.560 --> 0:07:00.880 Teco example, because in both cases the wistled language is 0:07:00.920 --> 0:07:05.719 not like an uh totally independent, unique language. Instead, it 0:07:05.880 --> 0:07:10.280 is in some way adapting an existing spoken language two whistles, 0:07:10.680 --> 0:07:14.160 and of course that would lean us more towards the 0:07:14.520 --> 0:07:17.400 Greek origin story as opposed to the Persian one. Yeah, 0:07:17.440 --> 0:07:19.560 I thought about that, but I don't know if it 0:07:19.600 --> 0:07:22.800 actually informs that one way or another. But yeah, I 0:07:23.040 --> 0:07:26.040 had the same intuition at least. So one of the 0:07:26.080 --> 0:07:29.640 most remarkable things about Sphyria, again similar to the mesa 0:07:29.680 --> 0:07:34.000 Teco whistle speech example, is that it is intelligible at 0:07:34.000 --> 0:07:37.920 a great distance. You can understand messages in Sphyria up 0:07:37.960 --> 0:07:43.320 to about four kilometers away on this mountainous terrain, which 0:07:43.560 --> 0:07:47.480 Hangen says is about ten times farther than you can 0:07:47.560 --> 0:07:51.440 usually understand speech, loud speech or shouting. And I saw 0:07:51.480 --> 0:07:54.600 that number of the ten times distance multiplier mentioned in 0:07:54.640 --> 0:07:58.280 other sources, such as a Cambridge University press paper that 0:07:58.360 --> 0:08:00.960 I looked at. But there's a great part in this 0:08:01.080 --> 0:08:06.240 article where Stein quotes a local shop owner named Maria Kafalas, 0:08:06.720 --> 0:08:09.640 who tells a story about some of the like the 0:08:09.720 --> 0:08:13.480 social opportunities offered by the whistle speech. And so her 0:08:13.520 --> 0:08:16.760 story goes like this quote. One night a man was 0:08:16.800 --> 0:08:19.840 in the mountains with his sheep when it started snowing. 0:08:20.360 --> 0:08:22.880 He knew that somewhere deep in the mountains there was 0:08:22.920 --> 0:08:26.400 a beautiful girl from Antia with her goats. So he 0:08:26.400 --> 0:08:29.280 found a cave, built a fire and whistled to her 0:08:29.320 --> 0:08:32.720 to come keep warm. She did, and that's how my 0:08:32.760 --> 0:08:35.880 parents fell in love. Well, that is a better ending 0:08:36.520 --> 0:08:40.160 than something like and then the descendants of of of 0:08:40.160 --> 0:08:44.400 Persian soldiers slaughtered him in the woods, so I will 0:08:44.440 --> 0:08:47.600 say that much. Yeah, and then the Greeks finally came 0:08:47.679 --> 0:08:50.480 for revenge. Yeah. I continue to just have a lot 0:08:50.520 --> 0:08:54.120 of questions about the Persian origin theory. It just seems 0:08:54.240 --> 0:08:57.120 like it seems and I could I could be missing 0:08:57.160 --> 0:08:59.520 something major here, but it seems like it begs more 0:08:59.600 --> 0:09:02.280 questions than it answers. Well. Yeah, as as I said, 0:09:02.920 --> 0:09:05.680 it sounds more like legend to me than like an 0:09:06.280 --> 0:09:10.480 strongly evidence based explanation. I agree, away from being a 0:09:10.520 --> 0:09:14.320 full blown ghost story. But but so I've described two 0:09:14.320 --> 0:09:18.400 examples here of whistled language in detail, and as I 0:09:18.440 --> 0:09:21.160 mentioned before, there are many others around the globe. There 0:09:21.160 --> 0:09:25.280 are something like eighty ish that have been reported somewhere 0:09:25.320 --> 0:09:28.800 around forty of them are very well documented. So an 0:09:28.840 --> 0:09:32.240 obvious question to ask is what do these languages have 0:09:32.320 --> 0:09:37.120 in common? What causes a whistled language to arise? So 0:09:37.160 --> 0:09:39.160 I was looking at a few sources here. One of 0:09:39.200 --> 0:09:42.600 them is that paper in the Annual Review of Linguistics 0:09:42.600 --> 0:09:47.280 by Julian Meyer already mentioned. Another is a an article 0:09:47.360 --> 0:09:51.560 in Smithsonian magazine from one by Bob Holmes, which cites 0:09:51.720 --> 0:09:55.080 that paper and summarizes some other research in this area, 0:09:55.400 --> 0:10:00.200 for example, focusing on a whistled variant of Spanish that 0:10:00.360 --> 0:10:03.760 is used in the Canary Islands, on the mountainous islands 0:10:03.840 --> 0:10:08.440 of Lagomera and Elierro in uh they're both in the 0:10:08.440 --> 0:10:13.320 Canary Islands. And that paper by Julian Meyer tries to 0:10:13.440 --> 0:10:16.440 gather together all of these languages and say, okay, are 0:10:16.440 --> 0:10:21.440 there common topic, graphical or or sort of geographical features 0:10:21.960 --> 0:10:25.000 that these languages tend to have in common? And he 0:10:25.080 --> 0:10:29.160 finds yes, indeed there are almost all of the whistled 0:10:29.240 --> 0:10:33.360 languages occur in two different types of environments, either in 0:10:33.760 --> 0:10:39.560 mountainous terrains rugged mountains, or in dense vegetation like dense 0:10:39.640 --> 0:10:42.680 forest or dense savannah. So why would it be those 0:10:42.679 --> 0:10:46.680 two places, mountains or in dense vegetation. Well, to focus 0:10:46.679 --> 0:10:51.200 on mountains first, Myer writes that in mountainous terrain, settlements 0:10:51.200 --> 0:10:53.640 and the people living in the mountainous terrain tend to 0:10:53.679 --> 0:10:58.760 be much more scattered across larger distances that are more 0:10:58.800 --> 0:11:03.040 difficult to travel verse quickly than people in other types 0:11:03.080 --> 0:11:08.200 of topographical settings. So uh here, so Meyer writes quote 0:11:08.400 --> 0:11:11.720 in Aliero and Lagomera, in the Canary Islands, in the 0:11:11.760 --> 0:11:15.679 region around kush Koy in Turkey, in the High Altars, 0:11:15.840 --> 0:11:19.120 or in the Pyrenees near the village of Os two points, 0:11:19.160 --> 0:11:22.880 only five dred meters apart can easily represent an hour 0:11:23.120 --> 0:11:27.640 in walking time. Thus, whistled forms of languages serve as 0:11:27.679 --> 0:11:31.640 soon as the spoken forms become ineffective between forty and 0:11:31.679 --> 0:11:35.520 a hundred meters. Depending on terrain, whistles can be heard 0:11:35.600 --> 0:11:39.719 up to seven kilometers away in some valleys. Okay, So 0:11:39.800 --> 0:11:42.960 the idea here is that in mountainous terrain you have 0:11:43.440 --> 0:11:47.080 the problem of people are often situated farther apart from 0:11:47.080 --> 0:11:51.079 each other, and those distances to cross are difficult to cross. 0:11:51.120 --> 0:11:53.480 They take a long time. So if you need to 0:11:53.520 --> 0:11:56.720 communicate actually coming to be close enough together that you 0:11:56.760 --> 0:12:01.319 could understand each other by shouting, would that's along? That's 0:12:01.320 --> 0:12:04.360 a big time investment. So it's actually worth your time 0:12:04.520 --> 0:12:07.800 to learn a whistle speech that will carry better across 0:12:07.880 --> 0:12:11.120 longer distances and save you all of that climbing and 0:12:11.160 --> 0:12:15.360 walking time. Now, what about the forests or the dense vegetation. Well, 0:12:15.400 --> 0:12:19.720 here Myer writes quote the vegetation in dense tropical forests 0:12:19.760 --> 0:12:25.160 and savannah's restricts visual contact and limits the propagation of sound. 0:12:25.920 --> 0:12:30.280 In such contexts, Whistled speech frequencies are also well shielded 0:12:30.320 --> 0:12:35.199 against acoustic energy loss due to reverberation, which is particularly 0:12:35.240 --> 0:12:40.079 important in densely vegetated environments because the whistled frequencies belong 0:12:40.160 --> 0:12:44.280 to the most favorable frequency window, ranging from one to 0:12:44.440 --> 0:12:50.000 three kiloherts, within which reverberation in forests varies less with distance. 0:12:50.640 --> 0:12:55.239 In dense vegetation, whistled language facilitates the coordination of individuals 0:12:55.320 --> 0:12:59.920 during group movements, especially during hunting and fishing. Whistling all 0:13:00.040 --> 0:13:04.040 so allows human dialogue to go undetected by animals, blending 0:13:04.080 --> 0:13:08.160 in with natural sounds, since many animal species also use whistling. 0:13:08.920 --> 0:13:12.040 Other advantages are that whistles are easy to locate and 0:13:12.080 --> 0:13:16.760 difficult for strangers to recognize, especially other tribes, even those 0:13:16.840 --> 0:13:21.040 that speak different dialects of the same language. Whistled communications 0:13:21.080 --> 0:13:24.400 are used for distances from about ten ms up to 0:13:24.480 --> 0:13:29.400 five hundred meters, depending on the density of vegetation. Okay, 0:13:29.400 --> 0:13:31.959 so there are a lot of advantages in the forest 0:13:32.080 --> 0:13:35.600 or thick savannah. So the idea is that, of course 0:13:35.600 --> 0:13:39.319 whistling speech allows you to communicate without being able to 0:13:39.360 --> 0:13:42.760 see each other. Sidelines are limited by the by the 0:13:42.840 --> 0:13:47.840 vegetation itself. But also whistling just carries better in the forest. 0:13:48.000 --> 0:13:51.600 It it propagates better through the forest without being drowned 0:13:51.640 --> 0:13:54.960 out by the sort of the the reverberation effects of 0:13:55.000 --> 0:13:58.440 having all that foliage there. Uh. And it also seems 0:13:58.480 --> 0:14:02.480 to pierce through and and sound much better. And Holmes 0:14:02.520 --> 0:14:06.400 also summarizes uh some of these advantages of whistling in 0:14:06.400 --> 0:14:11.280 in the Smithsonian paper, saying that if you're good at whistling, 0:14:11.360 --> 0:14:14.079 and you've been practicing this all your life, sometimes you 0:14:14.120 --> 0:14:16.559 can reach a hundred and twenty decibels with a whistle, 0:14:16.640 --> 0:14:20.080 which is loud. That's like uh. He compares it to 0:14:20.320 --> 0:14:22.520 a car horn and says it's actually louder than a 0:14:22.560 --> 0:14:26.480 car horn. Um, and that whistles pack almost all of 0:14:26.520 --> 0:14:31.600 that energy into the perfect frequency range, the most piercing 0:14:31.640 --> 0:14:34.760 frequency range, which Holmes says is between one to four 0:14:34.880 --> 0:14:37.880 killer hurts. Meyer said between one to three killer hurts. 0:14:37.880 --> 0:14:41.400 But it's roughly the same space um, which Holmes says 0:14:41.520 --> 0:14:45.600 is above the pitch of most ambient noise. And this 0:14:45.680 --> 0:14:49.360 is interesting because I was thinking about, um, why do 0:14:49.440 --> 0:14:53.760 we keep noticing that whistling sounds, the ones made by 0:14:53.880 --> 0:14:58.080 humans are similar to birds song. Well, one thing that 0:14:58.080 --> 0:15:01.600 occurs to me here is that Earth's song is probably 0:15:01.760 --> 0:15:05.960 shaped by natural selection to propagate through vegetation and to 0:15:06.040 --> 0:15:09.720 cut through ambient noise from the environment so so as 0:15:09.720 --> 0:15:11.600 to be clear, you know, to be clear and audible 0:15:11.600 --> 0:15:14.360 at a distance where maybe a potential mate could hear it. 0:15:14.960 --> 0:15:18.960 So whistle speech probably sounds like birds song, having similar 0:15:19.000 --> 0:15:22.880 frequency ranges because similar forces are shaping them. In the 0:15:22.920 --> 0:15:25.600 case of birds, it would be evolution, and in the 0:15:25.640 --> 0:15:29.520 case of humans, it would be people intentionally selecting whatever 0:15:29.600 --> 0:15:32.600 noise they are able to make with their bodies that 0:15:32.840 --> 0:15:37.080 is the clearest at the longest distance, cutting through ambient noise, 0:15:37.160 --> 0:15:41.680 and and losing the least energy to reverberation in the forest. 0:15:41.760 --> 0:15:43.960 And that just happens to be the whistle that sounds 0:15:43.960 --> 0:15:47.000 like a bird. Yeah, yeah, this this is interesting to 0:15:47.040 --> 0:15:51.240 think think about it. On one hand, I'm a quiet whistler. 0:15:51.280 --> 0:15:54.160 My whistler, my whistle was not very loud, and therefore 0:15:54.520 --> 0:15:57.040 it can be a little surprising when I encounter someone 0:15:57.040 --> 0:15:59.680 who has a very loud whistle, and you reminded just 0:15:59.720 --> 0:16:03.120 how loud and whistle can be. So that's important to 0:16:03.280 --> 0:16:06.680 factor into all of this. And uh. And another connection 0:16:06.800 --> 0:16:08.560 that came up in some of the research I was 0:16:08.600 --> 0:16:12.520 doing was that you you end up encountering this whole 0:16:12.600 --> 0:16:18.200 realm of of non linguistic sounds that humans can make 0:16:19.000 --> 0:16:21.760 that can be used to communicate ideas or to to 0:16:22.280 --> 0:16:25.280 gain attention, etcetera. And you you also see things like 0:16:25.360 --> 0:16:31.040 yodeling thrown in there. Yodeling um also an art form 0:16:31.080 --> 0:16:34.440 if you will, or a performance are to sound that 0:16:34.680 --> 0:16:37.920 developed that also had to do with communicating or calling 0:16:37.960 --> 0:16:41.800 animals or or communicating with other herdsmen across across long 0:16:41.840 --> 0:16:44.880 distances in the wild. When you're trying to speak normal 0:16:44.960 --> 0:16:48.600 phonemes like we're using in words here, I think a 0:16:48.640 --> 0:16:51.720 lot of that information probably easily gets lost to the distance. 0:16:51.800 --> 0:16:54.440 Like you might be able to hear that somebody is shouting, 0:16:54.760 --> 0:16:57.400 but you can't hear the difference between consonants. Are they 0:16:57.520 --> 0:17:00.080 making a T sound or a case sound like I 0:17:00.280 --> 0:17:03.200 don't know at a distance that that kind of all disappears. 0:17:03.640 --> 0:17:07.440 But if you're if you're judging more on sequences of pitches, 0:17:08.480 --> 0:17:13.600 then suddenly the confusion created by distance is reduced. Yeah. Yeah, 0:17:13.840 --> 0:17:17.199 just just yelling doesn't necessarily cut it, right, because if 0:17:17.200 --> 0:17:19.920 you can't, if you're particular words are not going to 0:17:19.960 --> 0:17:22.119 be overheard, then you might end up having to do 0:17:22.240 --> 0:17:24.800 something like just some sort of rhythmic barking. And if 0:17:24.800 --> 0:17:27.240 you're doing some sort of rhythmic barking, well why not 0:17:27.520 --> 0:17:31.200 further develop that and get somewhere and get to somewhere 0:17:31.240 --> 0:17:34.199 where it is yodeling or you shift over into a 0:17:34.240 --> 0:17:37.040 whistling uh, and that develops into some sort of a 0:17:37.040 --> 0:17:40.199 whistling language. So yeah, it's just the more you look 0:17:40.240 --> 0:17:41.880 at it, the more since it makes for this kind 0:17:41.920 --> 0:17:51.040 of purpose than Now, another thing going on with whistled 0:17:51.080 --> 0:17:54.840 languages is that, um, most of them, perhaps all of them, 0:17:54.840 --> 0:17:56.600 but I'm not sure about that. So I'm going to 0:17:56.640 --> 0:18:00.520 say at least the vast majority of whistled languages appear 0:18:00.640 --> 0:18:04.919 to be not wholly independent languages of their own, but 0:18:05.119 --> 0:18:09.639 whistled versions of spoken languages. So this was true of 0:18:09.960 --> 0:18:12.639 all the examples I've talked about before. You know, the 0:18:12.680 --> 0:18:15.720 maze Teco whistle speech was a whistled variant of the 0:18:15.760 --> 0:18:19.800 tonal meze Teco language. Spherria appears to be a whistled 0:18:19.840 --> 0:18:23.440 system for encoding spoken Greek. The whistle speech system of 0:18:23.480 --> 0:18:26.840 the Canary Islands, called sill Bo, is a whistled version 0:18:26.880 --> 0:18:31.160 of Spanish, and so forth, and for this reason, one 0:18:31.160 --> 0:18:34.800 of the main differences in whistled languages appears to be 0:18:34.920 --> 0:18:38.560 whether they are encoding a tonal language or a non 0:18:38.640 --> 0:18:43.240 tonal language, and based on that distinction, the encoding process 0:18:43.359 --> 0:18:46.480 is different. Tonal languages tend to be whistled in a 0:18:46.520 --> 0:18:49.720 way that preserves the tones of the spoken words, and 0:18:49.920 --> 0:18:53.240 in the last episode we talked about tonal languages. Tonal 0:18:53.320 --> 0:18:56.560 languages where you know the syllables of the words, also 0:18:57.119 --> 0:19:00.880 carry information based on the tone you use when speaking them. 0:19:01.240 --> 0:19:04.480 So say like a high, high pitched version of the 0:19:04.520 --> 0:19:08.200 syllable MA means something different than a lower pitched version 0:19:08.240 --> 0:19:11.040 of the syllable MA, or an up gliding tone on 0:19:11.119 --> 0:19:13.320 that syllable, and so forth, like the tone of the 0:19:13.320 --> 0:19:16.719 syllable actually makes a difference. Uh, non tonal languages are 0:19:16.760 --> 0:19:19.840 not like so in in English, we we don't encode 0:19:19.880 --> 0:19:23.080 much information into the tones of syllables. It's just like, 0:19:23.160 --> 0:19:27.840 what are the vowels and consonants. Yeah, the tone can 0:19:28.680 --> 0:19:31.520 contain some information, but not nearly to the to the 0:19:31.600 --> 0:19:34.560 extent that you find in true tonal languages. Right. No, 0:19:34.720 --> 0:19:38.520 it's not not lexical information, more like, uh, maybe sort 0:19:38.560 --> 0:19:44.399 of contextual mood information or inflection. Like the difference between 0:19:44.440 --> 0:19:47.280 saying I would like you to walk the dog and 0:19:47.520 --> 0:19:49.879 I would like you to walk the dog. Well, that 0:19:49.960 --> 0:19:52.679 implies that maybe something the last time the dog was 0:19:52.680 --> 0:19:56.119 walked it was not it was not good enough, or 0:19:56.119 --> 0:19:58.719 it was or maybe you ran the dog, you know, 0:19:58.800 --> 0:20:01.240 like that, that sort of thing. But it doesn't change 0:20:01.280 --> 0:20:05.840 the actual yeah, uh information contained in the word. No, 0:20:06.000 --> 0:20:09.400 it's more like about the implied information about the attitude 0:20:09.440 --> 0:20:12.440 of the speakers, or I need you to walk the dog. 0:20:13.760 --> 0:20:15.800 Maybe you didn't walk the actual dog. Maybe you took 0:20:15.920 --> 0:20:18.560 some other creature or item from the house with you 0:20:18.640 --> 0:20:21.480 on the walk instead. Okay, so you've got a tonal 0:20:21.600 --> 0:20:23.879 language and you want to make a whistled version of 0:20:23.920 --> 0:20:27.480 that you In most cases, it seems like you preserve 0:20:27.600 --> 0:20:31.560 the tones of the spoken words. Non tonal languages that 0:20:31.720 --> 0:20:35.960 have whistled speech tend to involve a sort of approximation 0:20:36.000 --> 0:20:40.399 of consonants and vowels, and the Homes article I mentioned 0:20:40.480 --> 0:20:44.520 quotes the scholar Julian Meyer explaining that we already use 0:20:44.840 --> 0:20:50.719 subtle differences in frequencies to distinguish between spoken phonemes, like 0:20:51.480 --> 0:20:55.560 the differences between certain vowels and consonants. Uh So, think 0:20:55.560 --> 0:20:59.639 about the vowels E and O. A long E vowel 0:20:59.680 --> 0:21:03.280 sound has a higher pitch than along O vowel sound, 0:21:03.359 --> 0:21:04.920 and if you say them back to back, you can 0:21:04.960 --> 0:21:09.400 listen to the descending melody of those vowel sounds EO 0:21:09.440 --> 0:21:12.520 E O, And in fact, though it's harder to hear 0:21:12.560 --> 0:21:15.320 at first, the same is sort of true of consonants, 0:21:15.359 --> 0:21:19.320 like a T sound contains more high frequencies than a 0:21:19.480 --> 0:21:23.280 K sound, and these differences can to some extent be 0:21:23.480 --> 0:21:28.439 reproduced in whistles and uh So, the discussion of this 0:21:28.480 --> 0:21:32.680 in the article got me thinking about, even without having 0:21:32.760 --> 0:21:35.520 an established version of a language like this, and without 0:21:35.600 --> 0:21:40.080 any training, can you sort of attempt to whistle English 0:21:40.119 --> 0:21:43.960 phrases and have people understand what you're saying. In some 0:21:44.000 --> 0:21:46.320 cases you can. And I actually tried this out with 0:21:46.359 --> 0:21:49.440 my wife Rachel before we recorded here, I, Uh, this 0:21:49.480 --> 0:21:52.160 was a kind of weird exercise, but I was like, hey, 0:21:52.200 --> 0:21:54.479 can you tell what I'm saying here? And so I 0:21:54.520 --> 0:22:00.800 tried things like um, which she took a minute on 0:22:00.920 --> 0:22:03.440 but decided I was saying hello, nice to meet you, 0:22:03.560 --> 0:22:06.880 which is what I was trying to say. Uh, So 0:22:07.040 --> 0:22:10.240 that one worked a few other phrases I tried did 0:22:10.280 --> 0:22:13.760 not work as well, But the ones that really seemed 0:22:13.800 --> 0:22:18.000 to work immediately were the ones where it was phrases 0:22:18.080 --> 0:22:21.720 she had heard me say before, especially when I tried 0:22:21.760 --> 0:22:25.639 to whistle, common phrases that we use with our dog, 0:22:26.200 --> 0:22:33.520 so um immediately she heard as all buddy um. And 0:22:33.720 --> 0:22:35.800 I think this ties into something we've talked about on 0:22:35.840 --> 0:22:40.439 the show before, the exaggerated musicality that humans tend to 0:22:40.560 --> 0:22:44.480 use when speaking to babies and pets for some reason. Uh, 0:22:44.560 --> 0:22:48.280 there may be evolutionary reasons for this, that we when 0:22:48.280 --> 0:22:51.840 we speak to cute things that need our care and attention, 0:22:52.080 --> 0:22:55.840 you know, babies or pets as kind of it might 0:22:55.840 --> 0:22:57.719 be creepy to think about them this way, but in 0:22:57.800 --> 0:23:01.840 to some extent kind of psychologically sir at babies, um, 0:23:01.880 --> 0:23:06.320 that we we speak with an exaggerated musicality or tonal 0:23:06.440 --> 0:23:09.399 variation that we don't use when speaking to adults, and 0:23:09.440 --> 0:23:13.880 that stereotyped phrases within this kind of highly musical speech 0:23:14.480 --> 0:23:17.280 are much easier to recognize when you try to whistle 0:23:17.359 --> 0:23:20.800 them instead of say them phonetically. So are you gonna 0:23:20.840 --> 0:23:23.160 keep whistling? Have you? Was it? Was it a big 0:23:23.240 --> 0:23:25.480 enough success? Oh? No, I think that would be a 0:23:25.480 --> 0:23:30.560 horrible idea. Also, strangely, the dog did not seem to 0:23:30.560 --> 0:23:33.800 get it. So when when I whistled all buddy Rachel 0:23:33.800 --> 0:23:36.160 could tell what I was saying, but but Charlie did 0:23:36.160 --> 0:23:41.280 not seem affected. I tried whistling to my cat whilst 0:23:41.320 --> 0:23:45.080 researching information for for these episodes, and yes, you didn't care. 0:23:45.640 --> 0:23:48.080 And my wife was like, like you, you can't speak 0:23:48.080 --> 0:23:50.480 to a cat and whistles. You have to use the 0:23:50.520 --> 0:23:53.440 kissy sound. That's what they understand. That's what that's the 0:23:53.520 --> 0:23:57.960 language they speak. It is known. But the kissy sound clicks, 0:23:58.080 --> 0:23:59.959 I mean, these are all these are not too far 0:24:00.000 --> 0:24:02.760 are removed from whistling. Some of these these types of 0:24:02.760 --> 0:24:06.080 sounds will come up again later on. Indeed. Now, one 0:24:06.119 --> 0:24:08.600 of the most interesting lines of thought emerging from all 0:24:08.640 --> 0:24:13.919 this is that some experts think that studying whistled languages 0:24:14.119 --> 0:24:17.760 might help us understand the origin of human language as 0:24:17.760 --> 0:24:23.000 a whole, because again, some linguists think that these whistled 0:24:23.040 --> 0:24:27.160 languages could be similar to the first languages that probably 0:24:27.240 --> 0:24:31.000 emerged in human evolution. Now why on earth would that be? Well, 0:24:31.040 --> 0:24:32.960 a couple of thoughts here. One. I just want to 0:24:33.480 --> 0:24:37.199 rate a passage from the Holmes article in Smithsonian quote. 0:24:37.840 --> 0:24:40.720 One of the big challenges for of language is the 0:24:40.840 --> 0:24:44.399 need to control the vocal chords to make the full 0:24:44.520 --> 0:24:48.399 range of speech sounds. None of our closest relatives, the 0:24:48.400 --> 0:24:52.840 great apes, have developed such control, but whistling maybe an 0:24:52.840 --> 0:24:57.480 easier first step. Indeed, a few orangutans and zoos have 0:24:57.600 --> 0:25:01.840 been observed to imitate zoo employee is whistling as they work. 0:25:02.520 --> 0:25:06.000