1 00:00:04,720 --> 00:00:08,079 Speaker 1: Imagine that one of these centuries we make contact with 2 00:00:08,240 --> 00:00:09,760 Speaker 1: an alien civilization. 3 00:00:10,039 --> 00:00:11,600 Speaker 2: It's a big. 4 00:00:11,400 --> 00:00:15,880 Speaker 1: Cosmos with quintillions of planets, so it's bound to happen 5 00:00:15,920 --> 00:00:18,279 Speaker 1: at some point. But how the heck are we going 6 00:00:18,320 --> 00:00:22,320 Speaker 1: to understand what they're saying? How are we going to 7 00:00:22,400 --> 00:00:26,960 Speaker 1: decode their language? After all, they might not communicate with 8 00:00:27,240 --> 00:00:30,280 Speaker 1: air compression waves. Maybe they do something visual, but in 9 00:00:30,360 --> 00:00:32,640 Speaker 1: ranges of light we can't even pick up with our eyes. 10 00:00:33,320 --> 00:00:37,000 Speaker 1: We won't have a Rosetta stone, So how are we 11 00:00:37,080 --> 00:00:40,199 Speaker 1: going to decipher what they are trying to say to us? 12 00:00:40,680 --> 00:00:43,120 Speaker 1: And this might seem speculative, but what I want to 13 00:00:43,240 --> 00:00:46,600 Speaker 1: draw our attention to is that we currently are in 14 00:00:46,640 --> 00:00:50,480 Speaker 1: the same position right now, right here at home, which 15 00:00:50,520 --> 00:00:54,400 Speaker 1: is that we can't tell what a single one of 16 00:00:54,520 --> 00:00:58,800 Speaker 1: the two million species on our planet are saying, not 17 00:00:58,920 --> 00:01:03,080 Speaker 1: even the six six hundred species of mammals, which are 18 00:01:03,160 --> 00:01:07,319 Speaker 1: presumably kind of like us. We're not having conversations with 19 00:01:07,640 --> 00:01:12,360 Speaker 1: anyone but ourselves. With all these species, I'm willing to 20 00:01:12,400 --> 00:01:15,000 Speaker 1: bet you don't listen to a single podcast not. 21 00:01:15,120 --> 00:01:16,479 Speaker 2: Made by a human. 22 00:01:17,360 --> 00:01:21,640 Speaker 1: But today we're going to see some hope, some pathways 23 00:01:21,680 --> 00:01:27,080 Speaker 1: for how we might get to animal translation and relatively soon. 24 00:01:31,120 --> 00:01:34,480 Speaker 1: Welcome to Inner Cosmos with me David Eagleman. I'm a 25 00:01:34,520 --> 00:01:38,600 Speaker 1: neuroscientist and an author at Stanford and in these episodes 26 00:01:39,040 --> 00:01:44,120 Speaker 1: I examined the intersection of science and our lives, and 27 00:01:44,200 --> 00:01:54,360 Speaker 1: today we're going to talk about understanding animals. When I 28 00:01:54,400 --> 00:01:57,440 Speaker 1: was a kid, I saw some episodes of Star Trek, 29 00:01:57,520 --> 00:02:00,920 Speaker 1: the original one with Kirk and Spock, and the thing 30 00:02:00,960 --> 00:02:04,200 Speaker 1: that always struck me was how every week on schedule 31 00:02:04,680 --> 00:02:10,040 Speaker 1: they discovered new alien civilizations, which is not so crazy 32 00:02:10,160 --> 00:02:14,839 Speaker 1: given that the universe is presumably teeming with life. There 33 00:02:14,880 --> 00:02:18,399 Speaker 1: are about one hundred billion galaxies, and each of these 34 00:02:18,440 --> 00:02:22,960 Speaker 1: has about one hundred billion stars, and most stars have 35 00:02:23,080 --> 00:02:28,200 Speaker 1: some planets rolling around them, so it's extraordinarily unlikely that 36 00:02:28,280 --> 00:02:31,040 Speaker 1: we are the only planet with life on it. But 37 00:02:31,160 --> 00:02:35,440 Speaker 1: before I get into alien communication, let's quickly address something first. 38 00:02:35,480 --> 00:02:39,800 Speaker 1: You've probably heard of the Fermi paradox, and if you haven't, 39 00:02:39,840 --> 00:02:43,519 Speaker 1: it's a very important question. It's the question of why, 40 00:02:43,680 --> 00:02:46,160 Speaker 1: if there's all this life in the cosmos, why have 41 00:02:46,280 --> 00:02:50,480 Speaker 1: we not heard a peep from anyone. This paradox is 42 00:02:50,560 --> 00:02:54,680 Speaker 1: named after the physicist Enrico Fermi, who raised this question 43 00:02:54,840 --> 00:02:59,800 Speaker 1: if there are so many potential alien civilizations, why haven't 44 00:02:59,840 --> 00:03:03,400 Speaker 1: we detected any signals or encountered any of them. Yet 45 00:03:03,880 --> 00:03:06,160 Speaker 1: we're living in a moment of history where there seems 46 00:03:06,160 --> 00:03:10,919 Speaker 1: to be this very strange contradiction between the high probability 47 00:03:11,040 --> 00:03:16,160 Speaker 1: of extraterrestrial civilizations and the lack of any shred of 48 00:03:16,240 --> 00:03:19,920 Speaker 1: evidence for them. So over the decades, people have proposed 49 00:03:19,960 --> 00:03:25,320 Speaker 1: all kinds of possible explanations for the Fermi paradox. The 50 00:03:25,360 --> 00:03:28,920 Speaker 1: first is that maybe aliens don't exist, which is the 51 00:03:28,960 --> 00:03:34,080 Speaker 1: simplest explanation but presumably not terribly likely given the size 52 00:03:34,120 --> 00:03:37,280 Speaker 1: of the cosmos. So some people point out that maybe 53 00:03:37,320 --> 00:03:41,440 Speaker 1: the problem just has to do with the enormous distances, 54 00:03:41,840 --> 00:03:46,360 Speaker 1: the vastness of space, and the limitations of our current 55 00:03:46,360 --> 00:03:50,360 Speaker 1: technology that might make it hard to detect other civilizations 56 00:03:50,440 --> 00:03:54,119 Speaker 1: even if they exist, because the distances between stars are 57 00:03:54,800 --> 00:03:59,080 Speaker 1: enormous and signals may take thousands or millions of years 58 00:03:59,120 --> 00:04:02,240 Speaker 1: to reach us. Okay, so that's a possibility. Or A 59 00:04:02,360 --> 00:04:06,640 Speaker 1: related idea is what's called the rare Earth hypothesis, which 60 00:04:06,680 --> 00:04:11,400 Speaker 1: is that Earth like planets, which are capable of supporting 61 00:04:11,560 --> 00:04:16,400 Speaker 1: complex life, are exceedingly uncommon in the universe, and that 62 00:04:16,440 --> 00:04:22,760 Speaker 1: makes the emergence of intelligent civilizations a rare event. But again, 63 00:04:23,240 --> 00:04:27,840 Speaker 1: given that there are something like seventy quintillion planets, that's 64 00:04:27,880 --> 00:04:33,039 Speaker 1: a seven followed by nineteen zeros Earth like planets can't 65 00:04:33,080 --> 00:04:39,120 Speaker 1: be too rare. So another idea is the technological singularity idea. 66 00:04:39,720 --> 00:04:44,640 Speaker 1: Some thinkers have proposed that advanced civilizations might always end 67 00:04:44,720 --> 00:04:51,840 Speaker 1: up reaching a technological singularity when their technology suddenly accelerates rapidly, 68 00:04:52,400 --> 00:04:56,080 Speaker 1: and one possible outcome of this is that these civilizations 69 00:04:56,480 --> 00:05:00,640 Speaker 1: tend to self destruct, and a related hypothesis is that 70 00:05:00,960 --> 00:05:04,919 Speaker 1: when other civilizations hit this singularity, this leads them to 71 00:05:05,080 --> 00:05:10,640 Speaker 1: a post biological existence. They're no longer products of nature, 72 00:05:10,680 --> 00:05:14,320 Speaker 1: but instead they build themselves into other sorts of devices, 73 00:05:14,800 --> 00:05:17,800 Speaker 1: which would be hard for us to detect given the 74 00:05:17,800 --> 00:05:22,120 Speaker 1: ways that we're searching. And other people suggest that advanced 75 00:05:22,160 --> 00:05:28,279 Speaker 1: civilizations intentionally avoid broadcasting their signals or their presence for 76 00:05:28,400 --> 00:05:33,560 Speaker 1: fear of attracting unwanted attention or causing conflicts with less 77 00:05:33,600 --> 00:05:38,440 Speaker 1: advanced civilizations. Or maybe they're just not interested in contacting us. 78 00:05:38,440 --> 00:05:40,760 Speaker 1: They might be too busy with their own problems, or 79 00:05:40,800 --> 00:05:44,600 Speaker 1: they simply don't see us as an interesting threat or ally, 80 00:05:44,680 --> 00:05:46,920 Speaker 1: so there's no reason to pick up the phone. And 81 00:05:46,960 --> 00:05:52,080 Speaker 1: then there's the possibility that they use extremely different communication 82 00:05:52,279 --> 00:05:56,200 Speaker 1: methods than we do, so different that we can't currently 83 00:05:56,600 --> 00:05:57,560 Speaker 1: understand them. 84 00:05:57,520 --> 00:05:59,719 Speaker 2: Or even know what we should be looking for. 85 00:06:00,480 --> 00:06:03,800 Speaker 1: So ultimately we don't know why we haven't heard from 86 00:06:03,880 --> 00:06:07,760 Speaker 1: anyone yet. There may be other reasons, or maybe multiple 87 00:06:07,800 --> 00:06:10,680 Speaker 1: of the reasons I mentioned, or all at play, but 88 00:06:10,720 --> 00:06:12,719 Speaker 1: for now we just have to live with the fact 89 00:06:12,800 --> 00:06:16,920 Speaker 1: that we haven't yet heard from anyone. So this was 90 00:06:17,000 --> 00:06:19,960 Speaker 1: part of the appeal of Star Trek. Every week there 91 00:06:20,480 --> 00:06:22,960 Speaker 1: spacewarping off to some new coordinates, and. 92 00:06:22,920 --> 00:06:26,320 Speaker 2: Everywhere they go they meet new civilizations. 93 00:06:26,960 --> 00:06:29,560 Speaker 1: Now, the thing that always struck me as funny and 94 00:06:29,720 --> 00:06:33,360 Speaker 1: the point of this episode is that each week they 95 00:06:33,480 --> 00:06:37,320 Speaker 1: end up meeting these new aliens. And often these aliens 96 00:06:37,360 --> 00:06:40,840 Speaker 1: look like a female movie star in a cool jumpsuit, 97 00:06:40,960 --> 00:06:44,400 Speaker 1: but with subtle differences like pointy ears and green skin. 98 00:06:44,760 --> 00:06:48,800 Speaker 1: But the key thing is that all these aliens speak English. 99 00:06:49,360 --> 00:06:52,719 Speaker 1: Usually it's a slightly broken English with a difficult to 100 00:06:52,800 --> 00:06:57,920 Speaker 1: discern accent, but nonetheless pretty easily understandable, which is of 101 00:06:57,920 --> 00:07:01,479 Speaker 1: course very lucky for these Star warsvoyagers who happen in 102 00:07:01,560 --> 00:07:05,119 Speaker 1: several hundred years from now to speak English themselves. Now, 103 00:07:05,920 --> 00:07:08,279 Speaker 1: why did the writers of Star Trek choose to make 104 00:07:08,360 --> 00:07:09,960 Speaker 1: everyone speak just like we do? 105 00:07:10,680 --> 00:07:10,920 Speaker 2: Well? 106 00:07:11,000 --> 00:07:14,920 Speaker 1: This is a basic constraint of storytelling. It's the only 107 00:07:14,960 --> 00:07:18,760 Speaker 1: thing that will work for telling a narrative that people 108 00:07:18,800 --> 00:07:22,440 Speaker 1: will tune into. It's hard to tell a story if 109 00:07:22,520 --> 00:07:26,080 Speaker 1: the alien is some kind of weird fungus thing that 110 00:07:26,160 --> 00:07:30,280 Speaker 1: doesn't speak or lives at a different timescale than we do, 111 00:07:30,480 --> 00:07:34,120 Speaker 1: like a tree. If we land on a planet of mute, 112 00:07:34,160 --> 00:07:37,680 Speaker 1: slow fungus creatures, it's not going to make good television. 113 00:07:38,360 --> 00:07:41,920 Speaker 1: So the stories we tell will always have aliens that 114 00:07:41,960 --> 00:07:44,840 Speaker 1: we can talk with and that serve as not so 115 00:07:45,040 --> 00:07:49,560 Speaker 1: distant reflections of ourselves. Okay, so no problem, that's what 116 00:07:49,920 --> 00:07:54,920 Speaker 1: storytelling requires. But in real life, it's much more likely 117 00:07:55,320 --> 00:07:58,400 Speaker 1: that we're going to have a very very difficult time 118 00:07:58,920 --> 00:08:03,760 Speaker 1: communicating much of anything to aliens when we find them. 119 00:08:04,160 --> 00:08:06,840 Speaker 1: You might think that we can get by with something 120 00:08:06,960 --> 00:08:09,480 Speaker 1: like take me to your leader or some sort of 121 00:08:09,520 --> 00:08:13,240 Speaker 1: hand signals, but in fact none of that's going to work. Now, 122 00:08:13,280 --> 00:08:17,040 Speaker 1: what's the reason that I say this? Why should we 123 00:08:17,080 --> 00:08:20,400 Speaker 1: think that communication is going to be so difficult. Well, 124 00:08:20,440 --> 00:08:23,320 Speaker 1: the aliens we find on other planets are going to 125 00:08:23,360 --> 00:08:27,440 Speaker 1: have a totally different evolutionary history. They may not be 126 00:08:27,520 --> 00:08:31,640 Speaker 1: based on DNA like all earthly creatures are, but instead 127 00:08:31,720 --> 00:08:35,959 Speaker 1: may have found a completely different way of encoding information 128 00:08:36,320 --> 00:08:41,240 Speaker 1: and managing replication and building societies. And it's possible they 129 00:08:41,280 --> 00:08:44,719 Speaker 1: won't even be carbon based like all creatures on Earth are, 130 00:08:45,040 --> 00:08:48,560 Speaker 1: but instead based on something like the element silicon, which 131 00:08:48,600 --> 00:08:51,640 Speaker 1: also has a tetrahedral structure and can make lots of 132 00:08:51,760 --> 00:08:54,959 Speaker 1: useful elements and so on. So there's a whole field 133 00:08:55,000 --> 00:08:59,800 Speaker 1: of this called astrobiology, where astro refers to star and 134 00:09:00,080 --> 00:09:05,040 Speaker 1: often known as exobiology, where exo means outside, and the 135 00:09:05,080 --> 00:09:08,480 Speaker 1: idea with this area of study is to search for 136 00:09:08,760 --> 00:09:13,679 Speaker 1: naturally evolved life in the universe, mostly on other planets 137 00:09:13,720 --> 00:09:17,800 Speaker 1: in the habitable zone, that's the Goldilocks zone of planets 138 00:09:17,800 --> 00:09:21,080 Speaker 1: who rotated just the right distance from their star so 139 00:09:21,120 --> 00:09:24,000 Speaker 1: it's not too hot not too cold. I'll also mention 140 00:09:24,080 --> 00:09:29,080 Speaker 1: there's a closely related field called zenobiology, which means alien 141 00:09:29,200 --> 00:09:33,320 Speaker 1: or foreign biology, and that term is usually reserved to 142 00:09:33,440 --> 00:09:37,640 Speaker 1: refer to biology that is synthetic, not found in nature 143 00:09:38,200 --> 00:09:41,760 Speaker 1: that science has no clue about yet. The general idea 144 00:09:41,920 --> 00:09:47,520 Speaker 1: is that astrobiologists try to detect and eventually analyze life 145 00:09:47,559 --> 00:09:52,439 Speaker 1: elsewhere in the universe, while xenobiologists will attempt to design 146 00:09:52,880 --> 00:09:56,560 Speaker 1: forms of life with a totally different biochemistry or different 147 00:09:56,559 --> 00:10:00,360 Speaker 1: genetic code than on planet Earth. Now, when we search 148 00:10:00,440 --> 00:10:03,280 Speaker 1: for life in the universe, I don't really see any 149 00:10:03,480 --> 00:10:08,000 Speaker 1: reason to imagine a distinction between naturally evolved life on 150 00:10:08,040 --> 00:10:12,400 Speaker 1: other planets and synthetic forms of life, because a planet 151 00:10:12,440 --> 00:10:15,000 Speaker 1: ten thousand years ahead of us would be well on 152 00:10:15,080 --> 00:10:17,840 Speaker 1: its way to building other species in the same way 153 00:10:17,840 --> 00:10:22,480 Speaker 1: that our species existence might simply serve as the spark 154 00:10:22,800 --> 00:10:26,559 Speaker 1: to build an artificial species that colonizes the Solar System. 155 00:10:27,200 --> 00:10:29,320 Speaker 1: And so for this reason we should always be thinking 156 00:10:29,360 --> 00:10:34,000 Speaker 1: about new biologies, new animals that could use a completely 157 00:10:34,000 --> 00:10:38,319 Speaker 1: different kind of biochemistry. On this planet, all we've ever 158 00:10:38,440 --> 00:10:43,640 Speaker 1: seen for coding information is DNA and RNA. We have 159 00:10:43,800 --> 00:10:47,200 Speaker 1: twenty amino acids from which we build all our proteins, 160 00:10:47,240 --> 00:10:50,559 Speaker 1: which are like little molecular machines. But instead of DNA, 161 00:10:51,400 --> 00:10:56,480 Speaker 1: we might find elsewhere what we'll call XNA zeno nucleic acid. 162 00:10:56,520 --> 00:11:00,560 Speaker 1: You might find a massively expanded genetic coat that uses 163 00:11:00,640 --> 00:11:03,880 Speaker 1: other amino acids to build totally new kinds of proteins, 164 00:11:04,240 --> 00:11:07,160 Speaker 1: or perhaps most likely, you could have something that's not 165 00:11:07,320 --> 00:11:10,760 Speaker 1: like our genetic code at all. Something to considers that 166 00:11:10,800 --> 00:11:13,959 Speaker 1: we only discovered the genetic code in nineteen fifty three, 167 00:11:14,000 --> 00:11:16,400 Speaker 1: which is not that long ago. I actually worked with 168 00:11:16,480 --> 00:11:20,280 Speaker 1: Francis Criek, the code discoverer of the structure of DNA, 169 00:11:20,400 --> 00:11:24,600 Speaker 1: So this is massively recent and we don't know what 170 00:11:24,720 --> 00:11:29,000 Speaker 1: we haven't thought of yet. So for this reason, I 171 00:11:29,000 --> 00:11:34,160 Speaker 1: think of the whole endeavor of space biology as zenobiology, 172 00:11:34,240 --> 00:11:37,240 Speaker 1: thinking about and looking for ways that life could be 173 00:11:37,320 --> 00:11:41,200 Speaker 1: built that we haven't yet imagined yet. And by the way, 174 00:11:41,240 --> 00:11:44,280 Speaker 1: it's thought exercises like this that make me wish I 175 00:11:44,320 --> 00:11:48,520 Speaker 1: could see the text books that our descendants will read 176 00:11:49,000 --> 00:11:51,440 Speaker 1: five hundred years from now or one thousand years from now. 177 00:11:51,960 --> 00:11:55,959 Speaker 1: There's going to be so much known that we currently 178 00:11:56,040 --> 00:11:59,720 Speaker 1: can't imagine. That's totally in our dark zone. Things we 179 00:11:59,840 --> 00:12:03,600 Speaker 1: have I haven't even realized that we don't know. Okay, 180 00:12:03,840 --> 00:12:06,920 Speaker 1: So all of this is to say life might be 181 00:12:07,040 --> 00:12:10,760 Speaker 1: massively different than what we have here on Earth, and 182 00:12:10,880 --> 00:12:14,480 Speaker 1: the question is will we figure out how to be 183 00:12:14,800 --> 00:12:18,040 Speaker 1: like Captain Kirk and beam onto their planet and have 184 00:12:18,160 --> 00:12:19,800 Speaker 1: a conversation with them. 185 00:12:20,600 --> 00:12:22,760 Speaker 2: Now, maybe all this talk. 186 00:12:22,600 --> 00:12:27,400 Speaker 1: About extraterrestrials seems abstract because we haven't discovered any yet. 187 00:12:28,000 --> 00:12:31,160 Speaker 1: But I want to point out that we are surrounded 188 00:12:31,440 --> 00:12:34,679 Speaker 1: by aliens. They are all around us, and they're making 189 00:12:34,880 --> 00:12:40,880 Speaker 1: constant sounds. We have measured this alien language with our recorders, 190 00:12:41,280 --> 00:12:45,440 Speaker 1: and at the moment we have no idea what they're saying. 191 00:12:52,120 --> 00:12:53,320 Speaker 2: Now, that's the. 192 00:12:53,360 --> 00:12:58,760 Speaker 1: Sound of whales on our planet, and lots of animals 193 00:12:59,160 --> 00:13:03,080 Speaker 1: make noise, And the question is, are these animals that 194 00:13:03,120 --> 00:13:04,720 Speaker 1: we're surrounded. 195 00:13:04,080 --> 00:13:06,480 Speaker 2: With speaking a language. 196 00:13:07,320 --> 00:13:09,920 Speaker 1: These are the aliens that we don't have to travel 197 00:13:10,080 --> 00:13:14,120 Speaker 1: very far to listen to, And the question is can 198 00:13:14,160 --> 00:13:15,200 Speaker 1: we understand them? 199 00:13:15,640 --> 00:13:16,920 Speaker 2: So maybe so, maybe not? 200 00:13:17,080 --> 00:13:21,920 Speaker 1: After all, these sounds could be like belches or like grunting, 201 00:13:22,000 --> 00:13:24,240 Speaker 1: the way you do when you're home alone and you 202 00:13:24,320 --> 00:13:26,480 Speaker 1: bang your knee the night before and you're going up 203 00:13:26,480 --> 00:13:29,320 Speaker 1: the stairs and you make a sound, but it's not 204 00:13:29,400 --> 00:13:31,240 Speaker 1: really meant for anyone in particular. 205 00:13:31,280 --> 00:13:32,720 Speaker 2: It's just a noise that you're making. 206 00:13:33,360 --> 00:13:37,559 Speaker 1: So how would we know if animals are actually implementing 207 00:13:37,920 --> 00:13:43,720 Speaker 1: language and communicating meaning to one another. Well, this is 208 00:13:43,760 --> 00:13:47,360 Speaker 1: an unanswered question right now, and it probably differs species 209 00:13:47,400 --> 00:13:50,640 Speaker 1: by species. One thing biologists look at to try to 210 00:13:50,679 --> 00:13:54,480 Speaker 1: address this is things like turn taking. Does one animal 211 00:13:54,520 --> 00:13:57,120 Speaker 1: make some sound and then the other animal goes, and 212 00:13:57,120 --> 00:13:59,480 Speaker 1: then the first one again, and then the second. That 213 00:13:59,559 --> 00:14:03,640 Speaker 1: feels more like language, right something had said, there's some response. 214 00:14:04,040 --> 00:14:07,200 Speaker 1: It feels like there's at least the possibility for some 215 00:14:07,280 --> 00:14:10,880 Speaker 1: real meaningful conversations that way. But there are a lot 216 00:14:10,920 --> 00:14:14,160 Speaker 1: of questions here. Even if we found that some species 217 00:14:14,200 --> 00:14:19,520 Speaker 1: were speaking language, would we be able to understand the meaning? 218 00:14:20,320 --> 00:14:22,840 Speaker 1: And I don't mean this in terms of this call 219 00:14:23,000 --> 00:14:26,200 Speaker 1: means this thing, but in terms of what does that 220 00:14:26,280 --> 00:14:31,080 Speaker 1: thing mean for a human? Imagine if a bee is 221 00:14:31,160 --> 00:14:33,760 Speaker 1: talking about some experience that you really have to see 222 00:14:33,800 --> 00:14:38,160 Speaker 1: in ultraviolet to understand, or your dog is experiencing something 223 00:14:38,200 --> 00:14:42,080 Speaker 1: about smell that we couldn't possibly get from our experience, 224 00:14:42,920 --> 00:14:45,840 Speaker 1: or a dolphin is talking about the joy of that 225 00:14:45,960 --> 00:14:49,000 Speaker 1: moment where there's no more little fish, and so the 226 00:14:49,040 --> 00:14:52,760 Speaker 1: whole pod suddenly turns upward and rockets through their world 227 00:14:53,120 --> 00:14:57,320 Speaker 1: and breaks through some surface where everything is different. Might 228 00:14:57,360 --> 00:15:00,320 Speaker 1: it be the case that there's simply no way we 229 00:15:00,360 --> 00:15:03,480 Speaker 1: could totally understand what they mean. There may be things 230 00:15:03,480 --> 00:15:06,440 Speaker 1: we can identify that they are talking about, but I 231 00:15:06,440 --> 00:15:10,720 Speaker 1: think there's a spectrum of how close we would actually 232 00:15:10,840 --> 00:15:14,120 Speaker 1: be in our interpretation. And by the way, I'll just 233 00:15:14,200 --> 00:15:17,080 Speaker 1: note this is true with our fellow humans as well. 234 00:15:17,440 --> 00:15:21,480 Speaker 1: Someone might tell you about an experience with hang gliding 235 00:15:21,760 --> 00:15:26,040 Speaker 1: or stamp collecting, or psychedelic drugs or whatever, and even 236 00:15:26,080 --> 00:15:28,680 Speaker 1: though you gnawd and you say, oh, I gotcha, I 237 00:15:28,680 --> 00:15:33,000 Speaker 1: can relate that to my own experience. There's a spectrum 238 00:15:33,080 --> 00:15:37,480 Speaker 1: of how closely you are actually capturing what they are describing. 239 00:15:37,920 --> 00:15:41,920 Speaker 1: Sometimes you might have an analogous experience that puts you close, 240 00:15:41,960 --> 00:15:45,920 Speaker 1: and sometimes your assumptions may be pretty distant. So back 241 00:15:45,960 --> 00:15:50,119 Speaker 1: to animals, how much could we even understand in a translation, 242 00:15:50,320 --> 00:15:55,240 Speaker 1: given that animals have such different sensory windows on the world, 243 00:15:55,600 --> 00:16:00,160 Speaker 1: and so their concepts might be very different from ours. Now, 244 00:16:00,200 --> 00:16:03,720 Speaker 1: even with all these caveats, how amazing would it be 245 00:16:03,760 --> 00:16:08,440 Speaker 1: if we could get even a low dimensional, blurry glimpse 246 00:16:08,680 --> 00:16:12,640 Speaker 1: of what they were talking about. People have always wanted 247 00:16:12,680 --> 00:16:15,800 Speaker 1: to understand animals, and they've tried in the past. But 248 00:16:15,920 --> 00:16:19,720 Speaker 1: we are at an amazing moment in history where I 249 00:16:19,760 --> 00:16:22,840 Speaker 1: think the time is right around the corner. And I'm 250 00:16:22,840 --> 00:16:24,840 Speaker 1: not saying this is a general claim. I'm saying this 251 00:16:24,920 --> 00:16:29,880 Speaker 1: for two specific reasons. First, we have incredible technology now 252 00:16:30,280 --> 00:16:34,600 Speaker 1: which makes it possible to do biologging. What is biologging. 253 00:16:34,840 --> 00:16:39,160 Speaker 1: This is about collecting data from an animal with a small, 254 00:16:39,280 --> 00:16:40,160 Speaker 1: lightweight device. 255 00:16:40,320 --> 00:16:41,680 Speaker 2: These are called biologgers. 256 00:16:42,240 --> 00:16:44,320 Speaker 1: You hook these up to an animal and that way 257 00:16:44,360 --> 00:16:48,360 Speaker 1: you can collect data for long windows of time without 258 00:16:48,520 --> 00:16:53,160 Speaker 1: humans being around. You can record sounds, and measure physiology 259 00:16:53,240 --> 00:16:55,960 Speaker 1: and track movements, and this is how you get a 260 00:16:56,200 --> 00:17:00,640 Speaker 1: secret window into an animal's world. Any thing is that 261 00:17:00,680 --> 00:17:04,320 Speaker 1: it gives new, rich data that you just can't get 262 00:17:04,359 --> 00:17:08,639 Speaker 1: otherwise about animals in their natural environment. Now, it's not 263 00:17:08,680 --> 00:17:12,280 Speaker 1: always easy to attach to the biologgers to animals, especially 264 00:17:12,280 --> 00:17:15,560 Speaker 1: if they're small and fast. But the main problem is 265 00:17:15,600 --> 00:17:19,560 Speaker 1: that the data collected by the biologgers can be really 266 00:17:19,680 --> 00:17:24,360 Speaker 1: complicated and difficult to analyze. So we humans have amassed 267 00:17:24,440 --> 00:17:27,760 Speaker 1: a ton of rich data that we're sitting on. And 268 00:17:27,800 --> 00:17:31,000 Speaker 1: that leads to the second specific reason why we're just 269 00:17:31,200 --> 00:17:35,680 Speaker 1: on the verge of something amazing, and that's artificial intelligence. 270 00:17:51,880 --> 00:17:56,040 Speaker 1: AI has been used for years to translate and decode 271 00:17:56,119 --> 00:18:00,000 Speaker 1: human languages, and now we have this incredible opportunity to 272 00:18:00,320 --> 00:18:05,080 Speaker 1: leverage it for understanding animal communication. And I'm lucky enough 273 00:18:05,080 --> 00:18:07,080 Speaker 1: to be friends with one of the people at the 274 00:18:07,240 --> 00:18:10,679 Speaker 1: lead of the effort to decrypt these alien languages that 275 00:18:10,720 --> 00:18:14,720 Speaker 1: were surrounded with So how does our modern technology give 276 00:18:14,840 --> 00:18:18,080 Speaker 1: us hope for decoding animal language. 277 00:18:19,119 --> 00:18:23,080 Speaker 3: So the insight originally came in twenty thirteen when I 278 00:18:23,119 --> 00:18:27,560 Speaker 3: was listening actually to NPR and there was a researcher 279 00:18:27,800 --> 00:18:29,440 Speaker 3: describing to a lot of monkeys. 280 00:18:29,920 --> 00:18:33,400 Speaker 1: That's Asa Raskin. He's a writer and entrepreneur and inventor, 281 00:18:33,840 --> 00:18:36,879 Speaker 1: and for the purposes of today, he founded the Earth 282 00:18:36,960 --> 00:18:42,280 Speaker 1: Species Project ESP, which is a nonprofit focused on using 283 00:18:42,440 --> 00:18:45,879 Speaker 1: AI to decode non human communication. 284 00:18:50,040 --> 00:18:53,119 Speaker 3: And these animals are They're credible. They live in the 285 00:18:53,160 --> 00:18:57,639 Speaker 3: Ethiopian highlands. They have like huge maines, red patches on 286 00:18:57,680 --> 00:19:01,879 Speaker 3: their chests. And what I did realize is that, according 287 00:19:01,920 --> 00:19:04,520 Speaker 3: to the researcher, they had one of the largest vocabularies 288 00:19:04,600 --> 00:19:08,119 Speaker 3: of any primate except for US humans. In fact, the 289 00:19:08,119 --> 00:19:12,480 Speaker 3: researchers swear that these animals talk about them behind their back. 290 00:19:13,280 --> 00:19:15,520 Speaker 3: And so the thought back then was like, well, if 291 00:19:15,560 --> 00:19:17,840 Speaker 3: people are just out there. The researchers are just out 292 00:19:17,840 --> 00:19:22,879 Speaker 3: there trying to understand what these beings are saying using 293 00:19:23,080 --> 00:19:25,800 Speaker 3: like a hand recorder in hand transcribing. Couldn't there be 294 00:19:25,840 --> 00:19:28,520 Speaker 3: a better way? Couldn't we use like machine learning AI 295 00:19:28,960 --> 00:19:32,159 Speaker 3: large scale microphone arrays. But of course, in twenty seventeen, 296 00:19:33,000 --> 00:19:36,560 Speaker 3: that wasn't yet possible because machine learning couldn't do something 297 00:19:36,880 --> 00:19:40,040 Speaker 3: that human beings couldn't already do. They couldn't translate a 298 00:19:40,119 --> 00:19:45,199 Speaker 3: language without a Rosetta stone, without any examples. And that 299 00:19:45,280 --> 00:19:49,480 Speaker 3: really changed in twenty seventeen when the machine learning community 300 00:19:49,480 --> 00:19:51,160 Speaker 3: there were two papers that came up back to back 301 00:19:51,200 --> 00:19:53,480 Speaker 3: and the Way this Thing Happens, that showed that you 302 00:19:53,520 --> 00:19:57,920 Speaker 3: could translate between any two human languages without the need 303 00:19:58,080 --> 00:20:01,680 Speaker 3: for examples or Rosetta stones. And I can dive into 304 00:20:01,840 --> 00:20:03,800 Speaker 3: how that works, but that was the moment that we 305 00:20:03,880 --> 00:20:07,080 Speaker 3: said we should get going and along the journey, I 306 00:20:07,119 --> 00:20:09,520 Speaker 3: just should say, like, all right, is there even a 307 00:20:09,600 --> 00:20:13,080 Speaker 3: there there? Like we say animal language, What does that mean? 308 00:20:13,160 --> 00:20:15,680 Speaker 3: What is a rich complex communication structure? What would that 309 00:20:15,760 --> 00:20:18,000 Speaker 3: look like? I just want to give a couple examples 310 00:20:18,040 --> 00:20:21,880 Speaker 3: for your listeners. So off the coast of Norway, every 311 00:20:21,960 --> 00:20:26,280 Speaker 3: year there's a group of false killer whales that all 312 00:20:26,560 --> 00:20:29,919 Speaker 3: phenomenologically speak one way, and a group of dolphins that 313 00:20:29,960 --> 00:20:33,359 Speaker 3: all speak another way, and they come together and they 314 00:20:33,440 --> 00:20:37,920 Speaker 3: hunt in a superpod. And when they do this, they 315 00:20:37,960 --> 00:20:43,200 Speaker 3: speak a third different way, which is just sort of crazy, 316 00:20:43,280 --> 00:20:45,560 Speaker 3: right like. And it turns out that whales, you know, 317 00:20:45,640 --> 00:20:48,679 Speaker 3: have a culture extending back thirty four million years. They 318 00:20:48,720 --> 00:20:51,399 Speaker 3: have dialects that sort of split off, which they can 319 00:20:51,520 --> 00:20:53,720 Speaker 3: understand each other with, and that can split all the 320 00:20:53,720 --> 00:21:00,560 Speaker 3: way into mutually unintelligible languages. Another example is I learned 321 00:21:00,560 --> 00:21:03,159 Speaker 3: this in twenty fourteen that for campell monkeys, hawk for 322 00:21:03,200 --> 00:21:06,520 Speaker 3: them means eagle, crack means leopard, and hawk ooh means 323 00:21:06,520 --> 00:21:09,359 Speaker 3: predator that's up, crack ooh means predator that's down. So 324 00:21:09,400 --> 00:21:12,119 Speaker 3: now we have a simple syntax. And then one of 325 00:21:12,119 --> 00:21:16,119 Speaker 3: my favorite studies is from the University of Hawaii in 326 00:21:16,280 --> 00:21:19,480 Speaker 3: nineteen ninety four, and here they taught dolphins two gestures, 327 00:21:19,800 --> 00:21:23,199 Speaker 3: and the first gesture was do something you've never done before, 328 00:21:24,000 --> 00:21:25,480 Speaker 3: which is sort of a crazy thing to be able 329 00:21:25,480 --> 00:21:27,680 Speaker 3: to communicate, but the dolphins will do it. And to 330 00:21:27,680 --> 00:21:29,960 Speaker 3: remember to do that, that means they have to remember 331 00:21:30,040 --> 00:21:33,399 Speaker 3: every single thing they've done before that session, understand the 332 00:21:33,400 --> 00:21:36,440 Speaker 3: concept negation not one of those things. Then invent whole cloth, 333 00:21:36,560 --> 00:21:39,479 Speaker 3: some new thing that they've never done before, but they 334 00:21:39,480 --> 00:21:41,440 Speaker 3: can do it. And then they'll teach the dolphins a 335 00:21:41,480 --> 00:21:45,399 Speaker 3: second gesture, do something you haven't done before. Together, and 336 00:21:45,440 --> 00:21:47,760 Speaker 3: they'll say, at the same time, do something you haven't 337 00:21:47,800 --> 00:21:51,280 Speaker 3: done before. Together. The dolphins go down, exchange sawnic information, 338 00:21:51,359 --> 00:21:53,320 Speaker 3: come up and do the same trick they've never done 339 00:21:53,320 --> 00:21:57,240 Speaker 3: before at the same time. And while that doesn't prove 340 00:21:57,320 --> 00:22:01,560 Speaker 3: representational language, it certainly places I think Auckham's razor. On 341 00:22:01,640 --> 00:22:04,159 Speaker 3: the other foot, it certainly seems that way. How do 342 00:22:04,240 --> 00:22:08,520 Speaker 3: you know, though, when you're approaching these that there exists languages? 343 00:22:08,600 --> 00:22:11,600 Speaker 3: For example, you said that whale culture or civilization is 344 00:22:11,600 --> 00:22:14,120 Speaker 3: thirty four million years old, But how do we know 345 00:22:14,200 --> 00:22:16,840 Speaker 3: that they're speaking a language that has the kind of 346 00:22:16,960 --> 00:22:20,560 Speaker 3: structure that we have that is capable of passing on 347 00:22:20,600 --> 00:22:25,680 Speaker 3: a culture or civilization. Yeah, great question, And of course 348 00:22:25,720 --> 00:22:29,439 Speaker 3: it's hard if you can't listen in and understand. But 349 00:22:29,520 --> 00:22:33,119 Speaker 3: you know, there are two hallmarks of language that human 350 00:22:33,160 --> 00:22:36,439 Speaker 3: beings have, one of which is to be able to 351 00:22:36,480 --> 00:22:40,600 Speaker 3: talk about something that isn't here and something that isn't now, 352 00:22:40,760 --> 00:22:42,720 Speaker 3: can you refer to things that are in a different time, 353 00:22:42,720 --> 00:22:46,080 Speaker 3: in a different place. And we can actually see already 354 00:22:46,240 --> 00:22:49,760 Speaker 3: from the research that the answer is. It appears to 355 00:22:49,800 --> 00:22:52,040 Speaker 3: be the case that at least some animals can do this. 356 00:22:52,160 --> 00:22:56,679 Speaker 3: So Adrian Lumero is a researcher on great apes, and 357 00:22:56,720 --> 00:22:59,800 Speaker 3: he's discovered in the last year or so that are 358 00:23:00,240 --> 00:23:03,000 Speaker 3: tangs do have a version of a past tense. They 359 00:23:03,040 --> 00:23:06,000 Speaker 3: can talk about things that are not now. And then 360 00:23:06,200 --> 00:23:10,800 Speaker 3: dolphins have names that they called each other by, and 361 00:23:11,240 --> 00:23:14,280 Speaker 3: Ian Yannick in twenty sixteen discovered that they will use 362 00:23:14,320 --> 00:23:17,120 Speaker 3: those names in the third person. They can talk about 363 00:23:17,880 --> 00:23:20,480 Speaker 3: one of their own that is not here. So now 364 00:23:20,480 --> 00:23:25,680 Speaker 3: we have two of the hallmarks not here not now. Now, 365 00:23:25,920 --> 00:23:28,560 Speaker 3: when we say language, we only have one example of 366 00:23:28,600 --> 00:23:31,800 Speaker 3: a species that speaks what we call language humans and 367 00:23:31,880 --> 00:23:33,960 Speaker 3: almost always like if you only know one color and 368 00:23:34,000 --> 00:23:36,480 Speaker 3: then you learn a second color, you discover an entire 369 00:23:36,560 --> 00:23:40,280 Speaker 3: rainbow in between. Like when we say language, you know 370 00:23:40,320 --> 00:23:43,120 Speaker 3: we're using that to be a catch all for rich 371 00:23:43,440 --> 00:23:47,240 Speaker 3: communication systems that can pass on cultural information. 372 00:23:47,720 --> 00:23:48,520 Speaker 2: Yeah, agreed. 373 00:23:49,800 --> 00:23:51,720 Speaker 1: Let me jump back to the orangutans for one second. 374 00:23:51,840 --> 00:23:54,399 Speaker 1: Is there any evidence that they have future tents. 375 00:23:54,880 --> 00:24:00,480 Speaker 3: That's a great question. We do not know. That's what's 376 00:24:00,480 --> 00:24:04,400 Speaker 3: so exciting about this field right now, is I think 377 00:24:04,480 --> 00:24:06,920 Speaker 3: of this it's sort of like the invention of the 378 00:24:06,960 --> 00:24:10,119 Speaker 3: Hubble telescope, right, It's like, and if you remember, I 379 00:24:10,119 --> 00:24:14,199 Speaker 3: think it was back in nineteen ninety five they pointed 380 00:24:14,200 --> 00:24:16,720 Speaker 3: the Hubble telescope and an empty patch in sky and 381 00:24:16,760 --> 00:24:19,200 Speaker 3: what they discovered was the most galaxies that have ever 382 00:24:19,240 --> 00:24:23,840 Speaker 3: existed in one spot. That's essentially what we're discovering here 383 00:24:23,960 --> 00:24:26,920 Speaker 3: is that we just haven't had the tools to look 384 00:24:27,480 --> 00:24:31,200 Speaker 3: and when we do, what we're discovering is much more 385 00:24:31,200 --> 00:24:33,159 Speaker 3: than everything we discovered. What we're discovering is everything. 386 00:24:33,280 --> 00:24:36,879 Speaker 1: Yeah, exactly right, the deep field experiment. So how do 387 00:24:36,960 --> 00:24:40,080 Speaker 1: we actually do it? How do we apply all the 388 00:24:40,119 --> 00:24:43,080 Speaker 1: modern tools of science to see if we can decode 389 00:24:43,119 --> 00:24:43,879 Speaker 1: a language? 390 00:24:44,440 --> 00:24:44,720 Speaker 2: Yeah? 391 00:24:44,800 --> 00:24:47,560 Speaker 3: Great question. So I'm going to start with this twenty 392 00:24:47,600 --> 00:24:50,600 Speaker 3: seventeen technology, and I just want your audience to remember 393 00:24:50,720 --> 00:24:54,000 Speaker 3: that twenty seventeen is essentially the Stone Age in AI. 394 00:24:55,080 --> 00:24:57,760 Speaker 3: But I think it's a really useful conceptual tool to 395 00:24:57,840 --> 00:25:00,720 Speaker 3: understand how it might work. So, how do you translate 396 00:25:00,800 --> 00:25:04,960 Speaker 3: between languages that don't have Rosetta stones. And it turns 397 00:25:04,960 --> 00:25:06,960 Speaker 3: out what you can ask AI to do is build 398 00:25:07,040 --> 00:25:09,920 Speaker 3: a shape that represents a language. So you say, feed 399 00:25:09,960 --> 00:25:12,520 Speaker 3: in all of Wikipedia, a whole bunch of text, and 400 00:25:12,560 --> 00:25:16,560 Speaker 3: the AI generates a shape that represents a language. Imagine 401 00:25:16,600 --> 00:25:20,800 Speaker 3: a galaxy where every star is a word, and words 402 00:25:20,840 --> 00:25:23,719 Speaker 3: that mean similar things are placed near each other, and 403 00:25:23,760 --> 00:25:28,520 Speaker 3: then words that share a sort of conceptual relationship get 404 00:25:28,560 --> 00:25:31,360 Speaker 3: turned into sharing a geometric relationship. What does that mean? 405 00:25:31,560 --> 00:25:35,000 Speaker 3: That means if you imagine king is to man as 406 00:25:35,040 --> 00:25:37,800 Speaker 3: woman is to queen, then in this shape, king is 407 00:25:37,840 --> 00:25:42,119 Speaker 3: the same distance direction to man as woman is to queen. 408 00:25:42,160 --> 00:25:44,119 Speaker 3: And so you actually just subtract king minus man. That 409 00:25:44,200 --> 00:25:47,080 Speaker 3: gives you a distance of direction. You add that to 410 00:25:47,320 --> 00:25:49,600 Speaker 3: boy and that'll equal prince. You add that to girl 411 00:25:49,640 --> 00:25:52,320 Speaker 3: eqal princess. You add that to woman and equal queen. 412 00:25:53,800 --> 00:25:56,520 Speaker 3: And so if you think about all of the relationships, 413 00:25:56,560 --> 00:25:58,840 Speaker 3: the internal relationships of a language, they think about the 414 00:25:58,840 --> 00:26:01,520 Speaker 3: word dog. Dog has relationship to man and to howl 415 00:26:01,680 --> 00:26:04,920 Speaker 3: and to wolf and to fer. If you it sort 416 00:26:04,920 --> 00:26:06,920 Speaker 3: of fixes in a point in space, and if you 417 00:26:07,640 --> 00:26:11,440 Speaker 3: solve this massive multi dimensional Sudoku puzzle of how every 418 00:26:11,560 --> 00:26:14,080 Speaker 3: concept relates to every other concept that gets turned into 419 00:26:14,119 --> 00:26:18,200 Speaker 3: a geometry, and out pops a rigid structure that represents 420 00:26:18,200 --> 00:26:20,439 Speaker 3: a language. Now the computer doesn't know what anything it means. 421 00:26:20,760 --> 00:26:22,600 Speaker 3: It just knows how they all relate to each other. 422 00:26:22,680 --> 00:26:26,840 Speaker 3: The shape represents all of the internal relationships of a language, 423 00:26:26,840 --> 00:26:29,359 Speaker 3: which is of course just a model of the world. 424 00:26:29,600 --> 00:26:32,000 Speaker 3: All right, So you have this shape for English, and 425 00:26:32,119 --> 00:26:34,360 Speaker 3: this is what the machine learners asked in twenty seventeen. 426 00:26:35,480 --> 00:26:38,760 Speaker 3: Is it possible? They said that the shape which is 427 00:26:38,800 --> 00:26:41,399 Speaker 3: English might be similar to or the same as, the 428 00:26:41,440 --> 00:26:45,000 Speaker 3: shape which is German. And if you ask anthropologists, they'd 429 00:26:45,000 --> 00:26:47,719 Speaker 3: be like, no, that's a silly thing to think, like. 430 00:26:47,840 --> 00:26:50,680 Speaker 3: They have different ways of viewing the world, different cosmologies. 431 00:26:50,880 --> 00:26:52,800 Speaker 3: But the machine is like, whatever, let's give it a try. 432 00:26:53,359 --> 00:26:55,400 Speaker 3: And it turns out that it works. You can take 433 00:26:55,440 --> 00:26:58,359 Speaker 3: the shape which is English and the shape which is German, 434 00:26:58,520 --> 00:27:01,320 Speaker 3: and literally rotate one shape on top of the other. 435 00:27:01,480 --> 00:27:03,119 Speaker 3: And even though there are words in one language that 436 00:27:03,119 --> 00:27:04,919 Speaker 3: don't appear in the other, if you blew your eyes, 437 00:27:05,760 --> 00:27:09,440 Speaker 3: the shapes are roughly the same, and the point, which 438 00:27:09,480 --> 00:27:12,120 Speaker 3: is dog ends up in the same in both. Now 439 00:27:12,160 --> 00:27:14,000 Speaker 3: you might be saying okay, but that's because English and 440 00:27:14,040 --> 00:27:18,159 Speaker 3: German are very similar languages. But it turns out this 441 00:27:18,200 --> 00:27:22,240 Speaker 3: works for Finnish, which is a really weird language, Turkish, Aramaic, Urdu. 442 00:27:22,920 --> 00:27:26,320 Speaker 3: Pretty much every human language fits in a kind of 443 00:27:26,760 --> 00:27:31,240 Speaker 3: universal human meaning shape, and the point, which is dog 444 00:27:31,440 --> 00:27:33,480 Speaker 3: ends up in the same spot in all of them, 445 00:27:34,400 --> 00:27:37,439 Speaker 3: and this lets you do translation without the need for 446 00:27:37,480 --> 00:27:41,600 Speaker 3: any examples. And this is I think, such a beautiful, 447 00:27:42,119 --> 00:27:48,080 Speaker 3: profound realization that there is a hidden structure underlying all 448 00:27:48,119 --> 00:27:52,400 Speaker 3: of us that unites our way of seeing. So that 449 00:27:52,520 --> 00:27:55,920 Speaker 3: was the sort of the core insight that said, well, 450 00:27:56,000 --> 00:27:59,480 Speaker 3: maybe now it's time to start building that shape for 451 00:27:59,840 --> 00:28:02,640 Speaker 3: animal communication, which by the way, is very hard because 452 00:28:02,680 --> 00:28:05,879 Speaker 3: it takes denoising and working with many partners to collect 453 00:28:05,920 --> 00:28:09,400 Speaker 3: like the years with the data that's required. But that's 454 00:28:09,400 --> 00:28:12,440 Speaker 3: sort of what we started to do. Now I'll pausit 455 00:28:12,440 --> 00:28:14,040 Speaker 3: for a second, but there are a couple other techniques 456 00:28:14,080 --> 00:28:15,160 Speaker 3: that can add to the top of this. 457 00:28:15,560 --> 00:28:17,280 Speaker 1: Great so let me jump in for one second. So 458 00:28:17,320 --> 00:28:20,200 Speaker 1: the fact that all the human languages have a similar 459 00:28:20,320 --> 00:28:25,240 Speaker 1: structure to them is in part because humans radiated out 460 00:28:25,280 --> 00:28:29,679 Speaker 1: of Africa sort of yesterday and as a result, you know, 461 00:28:29,720 --> 00:28:33,600 Speaker 1: we all have the same brain and it's not so surprising. 462 00:28:33,640 --> 00:28:37,240 Speaker 1: And the question is what do we expect when we're 463 00:28:37,240 --> 00:28:39,960 Speaker 1: looking at animal languages, which I'll come back to some 464 00:28:40,000 --> 00:28:41,680 Speaker 1: more questions on that a second, But what do we 465 00:28:41,760 --> 00:28:45,560 Speaker 1: expect in terms of the similarity there given that animals 466 00:28:45,560 --> 00:28:48,280 Speaker 1: are picking up on different signals from the world they're 467 00:28:48,440 --> 00:28:51,960 Speaker 1: umvelt is different the signals they can get and their 468 00:28:52,000 --> 00:28:53,320 Speaker 1: concepts might be very different. 469 00:28:53,640 --> 00:28:55,120 Speaker 2: How do you think about that? 470 00:28:55,200 --> 00:28:57,720 Speaker 3: This is a great question. It's just to repeat what 471 00:28:57,760 --> 00:29:01,400 Speaker 3: you're saying, is that the censorium, the way that animals 472 00:29:01,560 --> 00:29:03,600 Speaker 3: perceive the world, like what it is like to be 473 00:29:03,640 --> 00:29:06,360 Speaker 3: a bat, may be so completely different than what it 474 00:29:06,440 --> 00:29:08,400 Speaker 3: is like to be a human because they're seeing in 475 00:29:08,520 --> 00:29:11,880 Speaker 3: three D sound that we can never translate anything. And 476 00:29:11,880 --> 00:29:16,720 Speaker 3: that may turn out to be the case, but you know, 477 00:29:16,760 --> 00:29:19,400 Speaker 3: I think there's reason to believe that there may be 478 00:29:19,480 --> 00:29:22,880 Speaker 3: some kind of overlap with our experience. And to just 479 00:29:22,920 --> 00:29:29,000 Speaker 3: give a couple examples. You know, lemurs, for example, are 480 00:29:29,040 --> 00:29:33,320 Speaker 3: known to bite down on centipedes, literally to take a 481 00:29:33,400 --> 00:29:36,479 Speaker 3: hit off of centipedes to get high. They enter this 482 00:29:36,600 --> 00:29:39,960 Speaker 3: very trance like state, they get super cuddly. It looks 483 00:29:40,000 --> 00:29:43,400 Speaker 3: sort of like a scene from Burning Man. Dolphins too, 484 00:29:43,640 --> 00:29:48,280 Speaker 3: are known to intentionally inflate pufferfish to get high after 485 00:29:48,320 --> 00:29:51,400 Speaker 3: their venom and then pass them around literally puff pass. 486 00:29:52,320 --> 00:29:54,640 Speaker 3: Great apes are known to like hang off of vines 487 00:29:54,680 --> 00:29:58,280 Speaker 3: and spin to get dizzy. There is something about a 488 00:29:58,400 --> 00:30:03,200 Speaker 3: transcendent state of conscious altering our state that is at 489 00:30:03,240 --> 00:30:07,320 Speaker 3: least shared amongst the mammals, and so if they're communicating, 490 00:30:07,320 --> 00:30:10,640 Speaker 3: they may well communicate about that, and that's something we'd share. 491 00:30:10,720 --> 00:30:14,320 Speaker 3: Another example is something known as the mirror test. This 492 00:30:14,480 --> 00:30:18,400 Speaker 3: is a test where you take an animal, you paint 493 00:30:18,400 --> 00:30:20,320 Speaker 3: a dot on them where they can't see it. You 494 00:30:20,360 --> 00:30:22,800 Speaker 3: give them a mirror. They look in the mirror, they 495 00:30:23,200 --> 00:30:25,520 Speaker 3: see the dot, and they turn to the dot and 496 00:30:25,560 --> 00:30:28,320 Speaker 3: they try to brush it off of themselves or investigate it. 497 00:30:28,840 --> 00:30:30,719 Speaker 3: And in order for an animal to do that, they 498 00:30:30,760 --> 00:30:33,880 Speaker 3: have to associate the image that's in the mirror with themselves. 499 00:30:33,920 --> 00:30:35,120 Speaker 3: They have to look in the mirror and say like 500 00:30:35,200 --> 00:30:39,200 Speaker 3: that's me. So that means there's a rich sense of interiority, 501 00:30:39,640 --> 00:30:44,560 Speaker 3: like a self awareness. Dolphins past this test, elephants past 502 00:30:44,640 --> 00:30:47,080 Speaker 3: the tests. A number of other species pass this test. 503 00:30:47,280 --> 00:30:51,640 Speaker 3: So even the concept so profound as me self awareness 504 00:30:51,760 --> 00:30:55,479 Speaker 3: that seems to be shared. You know, examples of people 505 00:30:56,400 --> 00:31:01,640 Speaker 3: showing orangutangs magic tricks and they go crazy. It's worth 506 00:31:01,720 --> 00:31:03,760 Speaker 3: just looking them up on YouTube to see these kinds 507 00:31:03,760 --> 00:31:08,400 Speaker 3: of videos. Pilot whales carry they're dead young for three 508 00:31:08,560 --> 00:31:12,680 Speaker 3: four weeks, like grief is a shared part of the experience. 509 00:31:12,720 --> 00:31:15,600 Speaker 3: So if you imagine these shapes, where one of the 510 00:31:15,640 --> 00:31:17,680 Speaker 3: shapes is like human language, is one of these is 511 00:31:17,760 --> 00:31:20,600 Speaker 3: animal communication, I think we should expect to see some 512 00:31:20,720 --> 00:31:23,440 Speaker 3: part of those shapes overlap, and that should be the 513 00:31:23,480 --> 00:31:27,000 Speaker 3: part we should do direct translation. But then there's going 514 00:31:27,040 --> 00:31:28,960 Speaker 3: to be a huge portion of the shape that can 515 00:31:29,000 --> 00:31:31,640 Speaker 3: never be directly translated to human experience, and you'd sort 516 00:31:31,640 --> 00:31:33,760 Speaker 3: of expect that to be sticking out, like where we 517 00:31:33,760 --> 00:31:36,360 Speaker 3: can see complexity, but we don't know how to translate it. 518 00:31:36,760 --> 00:31:38,200 Speaker 3: And I still don't know which one of these two 519 00:31:38,280 --> 00:31:39,920 Speaker 3: is going to be more fascinating, the part where we 520 00:31:39,960 --> 00:31:43,280 Speaker 3: can directly translate the part we don't, because, as I'm saying, 521 00:31:43,360 --> 00:31:46,800 Speaker 3: human beings have been communicating vocally for one hundred thousand 522 00:31:46,840 --> 00:31:49,720 Speaker 3: to three hundred thousand years, passing up culture. Whales and 523 00:31:49,760 --> 00:31:53,440 Speaker 3: dolphins have been doing this for thirty four million years, 524 00:31:53,480 --> 00:31:56,320 Speaker 3: and that which is oldest correlates with that which is wisest. 525 00:31:56,760 --> 00:32:00,400 Speaker 3: So for something to survive thirty four million years, there 526 00:32:00,440 --> 00:32:03,480 Speaker 3: has to be some deep kernel of adaptive truth in there. 527 00:32:04,040 --> 00:32:07,920 Speaker 3: And whatever it is that is the solution to humanities problems, like, 528 00:32:07,960 --> 00:32:10,280 Speaker 3: it's not in our imagination, because if it is, we'd 529 00:32:10,280 --> 00:32:12,640 Speaker 3: probably be trying to do it. So this is a 530 00:32:12,680 --> 00:32:15,880 Speaker 3: way of starting to get the first polaroid sort of 531 00:32:15,920 --> 00:32:19,840 Speaker 3: blurry image pictures of that which is beyond our imagination. 532 00:32:20,920 --> 00:32:22,960 Speaker 2: Now, let me ask you this. If we're just looking 533 00:32:23,000 --> 00:32:25,280 Speaker 2: at the auditory. 534 00:32:24,760 --> 00:32:27,320 Speaker 1: Information that we get from animals, we can do this 535 00:32:27,400 --> 00:32:30,520 Speaker 1: kind of technique where we're looking to match one galaxy 536 00:32:30,560 --> 00:32:32,920 Speaker 1: of stars to the other galaxy of stars and see 537 00:32:32,920 --> 00:32:35,360 Speaker 1: what parts are sticking out and so on. But we 538 00:32:35,400 --> 00:32:38,600 Speaker 1: may well need much more than just the audio right 539 00:32:39,600 --> 00:32:42,800 Speaker 1: to understand the context of what the animal is saying 540 00:32:42,840 --> 00:32:47,560 Speaker 1: in a particular situation. So how are people pursuing that? 541 00:32:47,560 --> 00:32:51,440 Speaker 3: That's a great question. Even for humans, we know that 542 00:32:51,600 --> 00:32:53,840 Speaker 3: so much of the information that we convey. If you've 543 00:32:53,840 --> 00:32:56,720 Speaker 3: ever had to try to order food in a country 544 00:32:56,760 --> 00:32:58,840 Speaker 3: where you don't speak the language and somehow you can 545 00:32:58,880 --> 00:33:03,560 Speaker 3: do it, you can have communication without words the same 546 00:33:03,560 --> 00:33:08,080 Speaker 3: thing may be true for animals. So in fact, chimpanzees 547 00:33:08,120 --> 00:33:10,640 Speaker 3: are known to have sixty plus hand and feet gestures, 548 00:33:10,640 --> 00:33:12,560 Speaker 3: which seems to be at least as far as we know, 549 00:33:12,600 --> 00:33:16,120 Speaker 3: their predominant form of more symbolic communication. 550 00:33:16,560 --> 00:33:19,320 Speaker 1: Plus, we have indefinite references to things all the time, right, 551 00:33:19,400 --> 00:33:22,520 Speaker 1: So when I say she, I might be talking about 552 00:33:22,560 --> 00:33:25,200 Speaker 1: Marie Curie, or I might be talking about Michelle Obama 553 00:33:25,680 --> 00:33:28,000 Speaker 1: or something. But once I've introduced too I'm talking about 554 00:33:28,000 --> 00:33:30,240 Speaker 1: I can just use the word she. But somebody trying 555 00:33:30,280 --> 00:33:34,200 Speaker 1: to decode when I'm saying who doesn't speak English might 556 00:33:34,240 --> 00:33:36,320 Speaker 1: have a hard time understanding what the references to. 557 00:33:37,080 --> 00:33:40,920 Speaker 3: Yeah, that's exactly right, and what you're speaking to is 558 00:33:41,080 --> 00:33:45,280 Speaker 3: the importance of context. In order to understand what someone 559 00:33:45,560 --> 00:33:48,120 Speaker 3: is saying or what an animal is meaning, we have 560 00:33:48,160 --> 00:33:51,120 Speaker 3: to understand the context. Otherwise, like the same grunt may 561 00:33:51,160 --> 00:33:53,840 Speaker 3: mean like that monkey, or it may mean like I'm upset, 562 00:33:54,000 --> 00:33:57,200 Speaker 3: and it all sort of depends on social context. So 563 00:33:57,400 --> 00:34:00,000 Speaker 3: a lot of what we do now is multi mode. 564 00:34:00,360 --> 00:34:05,880 Speaker 3: That is to say, we work with biologists that have 565 00:34:06,640 --> 00:34:12,600 Speaker 3: tags on animals, that record often video, audio, and motion, 566 00:34:13,400 --> 00:34:16,319 Speaker 3: and that lets us begin to translate between all of 567 00:34:16,360 --> 00:34:19,000 Speaker 3: these different modalities, and in fact, with some of the 568 00:34:19,040 --> 00:34:22,600 Speaker 3: species to work with, these tags are on multiple animals 569 00:34:22,800 --> 00:34:26,440 Speaker 3: in the same group, so we can get social context. 570 00:34:26,760 --> 00:34:30,000 Speaker 3: And I want to return for a second to this 571 00:34:30,040 --> 00:34:33,080 Speaker 3: really interesting question you pose. You're like, well, maybe all 572 00:34:33,200 --> 00:34:36,000 Speaker 3: human languages fit in the same shape because we share 573 00:34:36,000 --> 00:34:39,840 Speaker 3: the same physical substrate, the same brains, in the same ears, 574 00:34:39,840 --> 00:34:43,399 Speaker 3: and the same eyes. But there's something deeper going on 575 00:34:43,520 --> 00:34:47,400 Speaker 3: in machine learning than just the ability to match the 576 00:34:47,440 --> 00:34:51,880 Speaker 3: shapes of languages. Maybe your audience has heard of or 577 00:34:51,960 --> 00:34:56,279 Speaker 3: seeing Dolly or mid journey or image diffusion where you 578 00:34:56,600 --> 00:35:00,359 Speaker 3: type in text and outcomes an image that is never 579 00:35:00,400 --> 00:35:05,160 Speaker 3: been seen before. How does that work? Well, these shapes 580 00:35:05,160 --> 00:35:06,960 Speaker 3: are actually really helpful to have in your mind to 581 00:35:07,000 --> 00:35:10,120 Speaker 3: understand how it works. So let's build now a shape 582 00:35:10,640 --> 00:35:13,640 Speaker 3: on human faces, and once again you end up with 583 00:35:13,719 --> 00:35:16,319 Speaker 3: a galaxy where every star now isn't a word, but 584 00:35:16,440 --> 00:35:21,399 Speaker 3: is a human face. Faces that share similar relationships share 585 00:35:21,480 --> 00:35:24,840 Speaker 3: geometric relationships. So if I take a picture of your face, David, 586 00:35:25,719 --> 00:35:27,680 Speaker 3: and then I take a picture of your face that's smiling, 587 00:35:28,040 --> 00:35:30,640 Speaker 3: there's this distance in direction that takes me between your 588 00:35:30,680 --> 00:35:32,839 Speaker 3: face and your faces just smiling. I subtract those two 589 00:35:32,840 --> 00:35:35,719 Speaker 3: as I get smilingness as a relationship. I can now 590 00:35:35,840 --> 00:35:39,160 Speaker 3: add that to any other face in the shape and 591 00:35:39,160 --> 00:35:40,840 Speaker 3: I'll get the smiling version of that face. 592 00:35:41,200 --> 00:35:41,359 Speaker 2: Right. 593 00:35:41,400 --> 00:35:43,719 Speaker 3: So now there's a direction that represents smiling, there's a 594 00:35:43,719 --> 00:35:48,280 Speaker 3: direction that represents frowning, that represents age, that represents gender, 595 00:35:48,360 --> 00:35:51,600 Speaker 3: more male, more female. You end up with a map 596 00:35:51,640 --> 00:35:55,120 Speaker 3: of all the semantic relationships, and you can now do 597 00:35:55,160 --> 00:35:57,200 Speaker 3: that not just for faces, you can do that for 598 00:35:57,320 --> 00:35:59,759 Speaker 3: all of images. And now you have a shape that 599 00:35:59,800 --> 00:36:02,520 Speaker 3: are presents images, a shape that represents languages. You look 600 00:36:02,520 --> 00:36:05,120 Speaker 3: at image caption pairs on the Internet and you can 601 00:36:05,480 --> 00:36:09,239 Speaker 3: align these two shapes. So now you have a way 602 00:36:09,239 --> 00:36:14,080 Speaker 3: of translating between text, language and images. So now you 603 00:36:14,160 --> 00:36:17,120 Speaker 3: just type in something like image I don't know, like 604 00:36:17,160 --> 00:36:22,080 Speaker 3: portrait of the country Chile as a woman. It goes 605 00:36:22,120 --> 00:36:25,359 Speaker 3: into the language space, gets translated to the image space, 606 00:36:25,440 --> 00:36:27,719 Speaker 3: the computer generates the image that's there, and you get that. 607 00:36:27,760 --> 00:36:31,720 Speaker 3: So that's how that technology works. There's something really deep 608 00:36:31,800 --> 00:36:34,960 Speaker 3: actually happening because it's not just working on language. It 609 00:36:34,960 --> 00:36:39,239 Speaker 3: seems to work on almost any modality out there. And 610 00:36:39,280 --> 00:36:42,960 Speaker 3: I think just like there is the unreasonable effectiveness of 611 00:36:43,239 --> 00:36:46,880 Speaker 3: mathematics where it's seems very strange. You go out on 612 00:36:46,960 --> 00:36:49,640 Speaker 3: some branch of abstract mathematics, it seems like it has 613 00:36:49,719 --> 00:36:51,319 Speaker 3: nothing to do with the world, and then it has 614 00:36:51,719 --> 00:36:54,240 Speaker 3: something profound to say about the world. You invent complex 615 00:36:54,280 --> 00:36:57,080 Speaker 3: numbers somehow that describes everything you're going to need to 616 00:36:57,480 --> 00:37:02,120 Speaker 3: deal with electricity. Who knew that's going on? In deep learning, 617 00:37:02,160 --> 00:37:04,680 Speaker 3: where there's an unreasonable effectiveness of deep learning, where the 618 00:37:04,800 --> 00:37:10,840 Speaker 3: same techniques are working across every modality from DNA to 619 00:37:11,040 --> 00:37:16,000 Speaker 3: fMRIs to language to audio to video to images to 620 00:37:16,840 --> 00:37:20,200 Speaker 3: computer code. There's nothing that says that had to have worked, 621 00:37:20,239 --> 00:37:23,360 Speaker 3: and yet it seems to be working. So we're learning 622 00:37:23,400 --> 00:37:26,880 Speaker 3: something I think profound about the structure of our universe. 623 00:37:27,320 --> 00:37:30,160 Speaker 3: But what that means for us specifically is that that 624 00:37:30,200 --> 00:37:32,800 Speaker 3: means we can build these kinds of shapes and embed 625 00:37:32,960 --> 00:37:37,319 Speaker 3: and translate between how an animal behaves and how it 626 00:37:37,480 --> 00:37:39,560 Speaker 3: sounds and what its body post is. So we can 627 00:37:39,600 --> 00:37:42,520 Speaker 3: say we're not quite there yet, but we're moving towards it. 628 00:37:42,640 --> 00:37:45,799 Speaker 3: Generate me the audio of two elephants coming together, and 629 00:37:46,239 --> 00:37:48,680 Speaker 3: that's going to view distribution of calls, some of which 630 00:37:48,760 --> 00:37:50,839 Speaker 3: might mean like hello, some of which might mean, this 631 00:37:50,920 --> 00:37:52,799 Speaker 3: is my name, some of which might mean I've missed you, 632 00:37:52,800 --> 00:37:54,400 Speaker 3: you don't know, but has something to do with affiliation. 633 00:37:54,480 --> 00:37:56,360 Speaker 3: Then you say, okay, now generate me the audio of 634 00:37:56,400 --> 00:37:58,480 Speaker 3: two elephants coming together, but where one of them's flapping 635 00:37:58,480 --> 00:38:00,839 Speaker 3: its ears and the other one's running quickly. What kind 636 00:38:00,840 --> 00:38:02,759 Speaker 3: of sounds does that make? And you can see that 637 00:38:02,840 --> 00:38:07,480 Speaker 3: this becomes a laboratory that lets you very quickly iterate 638 00:38:07,560 --> 00:38:10,719 Speaker 3: to understand what animals are saying. When you get to 639 00:38:10,960 --> 00:38:14,000 Speaker 3: do this in combination with the incredible biologists that are 640 00:38:14,000 --> 00:38:15,719 Speaker 3: out in the field and have already built up a 641 00:38:15,760 --> 00:38:20,040 Speaker 3: lot of that context from time, blood, sweat, and tears. 642 00:38:20,160 --> 00:38:23,960 Speaker 1: Excellent, and I think the biologging is becoming more sophisticated, 643 00:38:24,000 --> 00:38:28,120 Speaker 1: even right where they're looking at temperature and weather patterns 644 00:38:28,160 --> 00:38:31,160 Speaker 1: and gyroscopes and accelerometers and so on, where you get 645 00:38:31,239 --> 00:38:35,000 Speaker 1: all of this data from the animals, and in theory, 646 00:38:35,000 --> 00:38:37,560 Speaker 1: there's no limit to how much we can biologue, as 647 00:38:37,600 --> 00:38:39,640 Speaker 1: long as we can make it small and portable and 648 00:38:39,719 --> 00:38:43,560 Speaker 1: gets it on the animals, and then we're able to discover, hey, 649 00:38:44,239 --> 00:38:47,520 Speaker 1: these are the contextual cues that the animal is responding 650 00:38:47,560 --> 00:38:50,160 Speaker 1: to with the language. 651 00:38:50,840 --> 00:38:53,760 Speaker 3: Yeah, that's exactly right. And in fact, a big shift 652 00:38:53,800 --> 00:38:58,560 Speaker 3: that's happened in biology and conservation and ethology in the 653 00:38:58,640 --> 00:39:04,799 Speaker 3: last five years is because of cell phones driving the 654 00:39:04,840 --> 00:39:08,439 Speaker 3: cost of sensors lower. Biologists have gone from a world 655 00:39:08,480 --> 00:39:11,760 Speaker 3: where they're often data starved to where they're data drowned, 656 00:39:11,880 --> 00:39:15,880 Speaker 3: where they have access to terabytes of data, but they 657 00:39:15,880 --> 00:39:19,799 Speaker 3: don't yet have the tools to understand them. And so 658 00:39:20,960 --> 00:39:25,840 Speaker 3: our goal is to decode non human communication, translate animal language, 659 00:39:26,040 --> 00:39:29,399 Speaker 3: and use that to transform our relationship with the rest 660 00:39:29,440 --> 00:39:31,560 Speaker 3: of nature. But it's sort of like you're trying to 661 00:39:31,560 --> 00:39:33,320 Speaker 3: go to the moon along the way you invent velcro. 662 00:39:33,880 --> 00:39:38,640 Speaker 3: We're building the foundational tools that every biologist needs to 663 00:39:39,440 --> 00:39:42,919 Speaker 3: understand the data that they have now. And our hope 664 00:39:43,040 --> 00:39:47,080 Speaker 3: is that by building those foundational tools or nonprofit or 665 00:39:47,160 --> 00:39:49,600 Speaker 3: open source, we try to give back as much as 666 00:39:49,600 --> 00:39:53,200 Speaker 3: we can that that can broad scale accelerate all of 667 00:39:53,320 --> 00:39:57,360 Speaker 3: like conservation science, which we hope can also accelerate conservation itself. 668 00:39:57,719 --> 00:40:00,799 Speaker 1: Now, I have a technical question, which is if you 669 00:40:00,880 --> 00:40:04,919 Speaker 1: are looking at human languages and making this high dimensional 670 00:40:05,280 --> 00:40:09,120 Speaker 1: space of all the words when you're listening, when you're 671 00:40:09,160 --> 00:40:13,759 Speaker 1: eavesdropping on whales or lemurs or whatever species. How do 672 00:40:13,800 --> 00:40:16,239 Speaker 1: you know what a word is, what a unit of 673 00:40:16,400 --> 00:40:17,160 Speaker 1: meaning is. 674 00:40:18,280 --> 00:40:22,960 Speaker 3: Yeah, a hard problem and there's no one easy solution. 675 00:40:23,960 --> 00:40:25,960 Speaker 3: But one of the things that we can ask the 676 00:40:26,000 --> 00:40:32,520 Speaker 3: AI to do is try chopping up the audio in many, many, 677 00:40:32,560 --> 00:40:36,400 Speaker 3: many different ways and then see which one of those 678 00:40:36,719 --> 00:40:40,759 Speaker 3: ends up making good predictions for what comes next. And 679 00:40:40,800 --> 00:40:42,560 Speaker 3: so you can see if you're trying and varying and 680 00:40:43,600 --> 00:40:46,040 Speaker 3: you're not saying, well, which thing contains meaning, but which 681 00:40:46,040 --> 00:40:49,640 Speaker 3: things help make good predictions. When you try this on humans, 682 00:40:49,680 --> 00:40:53,960 Speaker 3: you end up with phonemes that you get out and 683 00:40:54,000 --> 00:40:57,560 Speaker 3: then those are then combinatorily built into words. So we're 684 00:40:57,560 --> 00:41:00,719 Speaker 3: playing with those kinds of techniques, but we don't have 685 00:41:00,840 --> 00:41:02,480 Speaker 3: like one surefire away yet. 686 00:41:02,600 --> 00:41:05,400 Speaker 1: And when you're thinking about predictions, one of the ways 687 00:41:05,480 --> 00:41:08,839 Speaker 1: that you could test a prediction is with playback, right, 688 00:41:08,880 --> 00:41:09,759 Speaker 1: So tell us about that. 689 00:41:11,239 --> 00:41:15,000 Speaker 3: Yeah, So this is the classic way that you test 690 00:41:15,040 --> 00:41:18,839 Speaker 3: your predictions in the field, where I'll just go out 691 00:41:18,880 --> 00:41:22,320 Speaker 3: and they will play a sound, often from the animal, 692 00:41:22,360 --> 00:41:24,840 Speaker 3: and they'll see whether the animal looks and for how long. 693 00:41:26,080 --> 00:41:28,839 Speaker 3: What we are starting to be able to do is 694 00:41:29,280 --> 00:41:32,800 Speaker 3: just like you can build a chat bot in text 695 00:41:33,160 --> 00:41:37,279 Speaker 3: that speaks Chinese without needing to speak Chinese. We are 696 00:41:37,320 --> 00:41:40,840 Speaker 3: on the cusp of being able to build these kinds 697 00:41:40,880 --> 00:41:44,319 Speaker 3: of chat bots, but that just directly speak in the 698 00:41:44,480 --> 00:41:47,040 Speaker 3: language of animals. So it's sort of like, imagine you 699 00:41:47,080 --> 00:41:49,920 Speaker 3: had a superpower, and your superpower was to go out 700 00:41:50,680 --> 00:41:52,880 Speaker 3: meet someone whose language you don't understand. You sort of 701 00:41:52,880 --> 00:41:54,680 Speaker 3: cock your head to the side and you listen for 702 00:41:54,680 --> 00:41:56,560 Speaker 3: a little bit and you're like, I don't know what 703 00:41:56,600 --> 00:41:58,880 Speaker 3: anything means, but I see that this sound pattern follows 704 00:41:58,920 --> 00:42:01,600 Speaker 3: this sound platted and this context. You just start to 705 00:42:01,640 --> 00:42:04,440 Speaker 3: babble and you have no idea what you're saying, but 706 00:42:04,520 --> 00:42:07,440 Speaker 3: the other person's like crosses the arm, like yeah, wow, 707 00:42:07,520 --> 00:42:10,680 Speaker 3: that's so meaningful. And at the end the person walks away. 708 00:42:10,680 --> 00:42:12,440 Speaker 3: I think they've had a great conversation. You're like, I 709 00:42:12,480 --> 00:42:14,280 Speaker 3: have no idea what I just said. I was just babbling. 710 00:42:15,280 --> 00:42:18,640 Speaker 3: But that's what that's literally what Chatchipdi does, and that's 711 00:42:18,680 --> 00:42:20,320 Speaker 3: the kind of thing that we are going to be 712 00:42:20,360 --> 00:42:23,320 Speaker 3: able to build in the next you know, like twelve months. 713 00:42:24,400 --> 00:42:25,239 Speaker 3: And what does that mean? 714 00:42:25,680 --> 00:42:27,719 Speaker 1: Just so it's clear, can you give an example of 715 00:42:28,400 --> 00:42:30,560 Speaker 1: playback and the kind of things that people are doing 716 00:42:30,640 --> 00:42:33,040 Speaker 1: right now with that yeah. 717 00:42:33,160 --> 00:42:36,520 Speaker 3: So an example of a playback might be one of 718 00:42:36,520 --> 00:42:40,200 Speaker 3: our partners, Michelle Fournet, and you can, actually your listeners 719 00:42:40,239 --> 00:42:45,200 Speaker 3: can go watch her incredible documentary Fathom. She was trying 720 00:42:45,239 --> 00:42:49,040 Speaker 3: to determine how do you say hello to humpback whale 721 00:42:49,600 --> 00:42:53,480 Speaker 3: and possibly include their name, So to say hello in 722 00:42:53,560 --> 00:42:57,040 Speaker 3: hump back, it turns out, is something like poop And 723 00:42:57,880 --> 00:43:02,280 Speaker 3: to test this, she recorded many different they're called whoop calls, 724 00:43:02,360 --> 00:43:06,359 Speaker 3: but they're hellos, many different whoop calls, and then went 725 00:43:06,400 --> 00:43:10,520 Speaker 3: out to Alaska, set up speakers underwater and would play 726 00:43:11,080 --> 00:43:14,520 Speaker 3: the hellos in a very controlled condition and would see 727 00:43:14,640 --> 00:43:19,759 Speaker 3: do the humpbacks respond? And the answer is yes, yes 728 00:43:19,880 --> 00:43:23,120 Speaker 3: they do. When she said hello, they would respond in 729 00:43:23,200 --> 00:43:26,840 Speaker 3: greater number of saying hello back. So that's an example 730 00:43:27,080 --> 00:43:28,760 Speaker 3: of a playback experiment. 731 00:43:44,960 --> 00:43:48,120 Speaker 1: I know you've thought a lot about the ethics involved 732 00:43:48,120 --> 00:43:51,320 Speaker 1: in this so far what we've been talking about sounds amazing, 733 00:43:51,360 --> 00:43:53,400 Speaker 1: and the question is what are the ethical things that 734 00:43:53,440 --> 00:43:54,360 Speaker 1: we need to keep an eye on. 735 00:43:55,560 --> 00:44:00,839 Speaker 3: Yeah, that is a great question because you know we 736 00:44:00,880 --> 00:44:05,080 Speaker 3: are going to be crossing this barrier very very soon, 737 00:44:05,320 --> 00:44:08,399 Speaker 3: which is I mean, this is the plot twist that 738 00:44:08,560 --> 00:44:12,440 Speaker 3: we will be able to communicate before we fully understand 739 00:44:12,480 --> 00:44:17,799 Speaker 3: what we're saying. That's again very surprising. I would have 740 00:44:17,840 --> 00:44:20,080 Speaker 3: not have guessed this if we rewound the clock three 741 00:44:20,239 --> 00:44:26,560 Speaker 3: or four years. What does this mean. This means that 742 00:44:26,640 --> 00:44:31,080 Speaker 3: if you're working with a species which has vocal learning, 743 00:44:31,840 --> 00:44:35,040 Speaker 3: well you might inject something that they say that then 744 00:44:35,440 --> 00:44:38,440 Speaker 3: changes their culture. So, to give an example, humpback whales 745 00:44:39,080 --> 00:44:43,160 Speaker 3: off the coast of Australia. For whatever reason, the Australian 746 00:44:43,239 --> 00:44:45,960 Speaker 3: humpbacks seem to be like the K pop singers, and 747 00:44:46,880 --> 00:44:49,400 Speaker 3: because they can sing halfway across an ocean basin and 748 00:44:49,400 --> 00:44:54,040 Speaker 3: they migrate across the world, often the songs that are 749 00:44:54,080 --> 00:44:57,600 Speaker 3: sung off the coasts of Australia will catch on and 750 00:44:57,640 --> 00:44:59,920 Speaker 3: be sung by much of the world population within a 751 00:45:00,000 --> 00:45:04,239 Speaker 3: couple of seasons. So it's, you know, the ultimate pop tune. Now, 752 00:45:04,480 --> 00:45:08,600 Speaker 3: we don't know as humans what truly the function of 753 00:45:09,000 --> 00:45:12,520 Speaker 3: humpback whale song is and how that culture works. So 754 00:45:12,600 --> 00:45:16,960 Speaker 3: if we just create a synthetic whale that sings, we 755 00:45:17,040 --> 00:45:20,200 Speaker 3: may infect a thirty four million year old wisdom tradition, 756 00:45:20,520 --> 00:45:24,800 Speaker 3: you know, create some kind of viral meme a whale QAnon. 757 00:45:25,200 --> 00:45:30,000 Speaker 3: We just don't know. So we have to be very 758 00:45:30,160 --> 00:45:35,560 Speaker 3: careful as we approach this new responsibility of what does 759 00:45:35,560 --> 00:45:39,880 Speaker 3: it mean to truly communicate with the other cultures of Earth? 760 00:45:40,200 --> 00:45:43,760 Speaker 3: And that means we should not go out and start 761 00:45:43,800 --> 00:45:47,719 Speaker 3: playing like two way communication real time. We should not 762 00:45:47,760 --> 00:45:51,879 Speaker 3: do those kinds of experiments with wild populations that vocally learn. 763 00:45:51,920 --> 00:45:53,719 Speaker 3: We have to think about what is it to have 764 00:45:53,840 --> 00:45:57,240 Speaker 3: like a prime direct of a Geneva convention for cross 765 00:45:57,239 --> 00:46:00,759 Speaker 3: species communication. And this is of course terrifying, and I 766 00:46:00,760 --> 00:46:03,759 Speaker 3: should say, everything that our species does we do with 767 00:46:03,800 --> 00:46:08,480 Speaker 3: biology partners and institutions. We are starting to work on 768 00:46:08,800 --> 00:46:11,759 Speaker 3: what are the ground rules before we even have the 769 00:46:11,800 --> 00:46:16,440 Speaker 3: technology for knowing when and how it is okay to 770 00:46:17,080 --> 00:46:21,760 Speaker 3: have these prime directive first contact moments because first contexts 771 00:46:21,760 --> 00:46:25,279 Speaker 3: have often not gone well for the beings being first contacted. 772 00:46:25,920 --> 00:46:30,319 Speaker 3: So I think the change in the relationship for how 773 00:46:30,360 --> 00:46:33,800 Speaker 3: we relate to nature is the point of our species 774 00:46:33,960 --> 00:46:35,920 Speaker 3: and it's exciting that we're getting to the place where 775 00:46:36,760 --> 00:46:38,000 Speaker 3: that becomes a necessity. 776 00:46:38,840 --> 00:46:42,120 Speaker 1: And what's your prediction for how long it'll be, what 777 00:46:42,280 --> 00:46:46,719 Speaker 1: year will have a meaningful conversation back and forth with 778 00:46:46,800 --> 00:46:50,239 Speaker 1: the species? And which do you think will be the 779 00:46:50,280 --> 00:46:51,000 Speaker 1: first species? 780 00:46:52,200 --> 00:46:52,480 Speaker 2: Yeah? 781 00:46:52,520 --> 00:46:54,920 Speaker 3: I mean This is science, so it's always very hard 782 00:46:54,960 --> 00:46:58,880 Speaker 3: to make predictions like this, and different people on my 783 00:46:58,880 --> 00:47:01,480 Speaker 3: team have different predictions, so I can just say mine, 784 00:47:01,920 --> 00:47:06,680 Speaker 3: but know that answers very I think certainly by twenty 785 00:47:06,800 --> 00:47:09,560 Speaker 3: thirty we will have had two way back and forth 786 00:47:09,640 --> 00:47:14,400 Speaker 3: to what degree we understand unknown, But I think we 787 00:47:14,440 --> 00:47:17,560 Speaker 3: will have a really good handle on it by then. 788 00:47:19,200 --> 00:47:24,680 Speaker 3: It's just so exciting. My personal favorite is Belugas, and 789 00:47:24,719 --> 00:47:27,359 Speaker 3: again everyone has their own personal favorite. But when you 790 00:47:27,400 --> 00:47:31,120 Speaker 3: listen to Belugas communicate, it sounds like an alien modem. 791 00:47:31,400 --> 00:47:35,000 Speaker 3: It sounds digital. There are lots of whistles in there. 792 00:47:35,360 --> 00:47:37,440 Speaker 3: It turns out that you know dolphins, they say their name, 793 00:47:37,480 --> 00:47:41,160 Speaker 3: their signature, whistle in a whistle. It's like a single band. 794 00:47:41,440 --> 00:47:44,359 Speaker 3: This is like full modem pack. It encodes their name, 795 00:47:44,480 --> 00:47:49,520 Speaker 3: it encodes their clan identity. And doctor Valeria Vergaro, with 796 00:47:49,520 --> 00:47:52,279 Speaker 3: whom we work on Beluga communication, she's sort of like 797 00:47:52,320 --> 00:47:55,000 Speaker 3: one of the preeminent scholars. It was her work that 798 00:47:55,080 --> 00:47:56,960 Speaker 3: showed that they have names that they call each other by. 799 00:47:57,719 --> 00:48:00,360 Speaker 3: And what blew my mind is that when she talks 800 00:48:00,360 --> 00:48:03,800 Speaker 3: about her data, she's like she had to throw away 801 00:48:04,680 --> 00:48:08,120 Speaker 3: ninety seven percent of her data in those studies because 802 00:48:08,120 --> 00:48:11,960 Speaker 3: she couldn't tell which beluga was speaking or disentangle them. 803 00:48:12,280 --> 00:48:14,520 Speaker 3: And that's because they are like forty belugas in a 804 00:48:14,560 --> 00:48:17,799 Speaker 3: tight mass that are moving around super fast. It's very 805 00:48:17,840 --> 00:48:22,120 Speaker 3: hard from a computational perspective. But that's where your listener's 806 00:48:22,200 --> 00:48:24,319 Speaker 3: ears should perk up, because here we have the most 807 00:48:24,400 --> 00:48:28,880 Speaker 3: vocal underwater species with the largest vocabulary that we know of, 808 00:48:29,520 --> 00:48:32,400 Speaker 3: and the super majority of data, like ninety seven percent 809 00:48:32,480 --> 00:48:35,080 Speaker 3: is unknown the ocean is what five percent explored Bluga 810 00:48:35,080 --> 00:48:37,840 Speaker 3: communication or at least this data sets are three percent explored. 811 00:48:38,200 --> 00:48:38,920 Speaker 2: Like this is. 812 00:48:38,840 --> 00:48:41,640 Speaker 3: Where you get like brand new discoveries. This is the 813 00:48:41,680 --> 00:48:42,400 Speaker 3: next frontier. 814 00:48:42,840 --> 00:48:45,319 Speaker 1: And do Belugas do turn taking, by the way, which 815 00:48:45,360 --> 00:48:47,880 Speaker 1: is one of the signatures of an actual language as 816 00:48:47,880 --> 00:48:49,560 Speaker 1: opposed to just broadcasting noise. 817 00:48:50,640 --> 00:48:53,719 Speaker 3: Yeah, a number of species to do turn taking, from 818 00:48:53,800 --> 00:48:58,040 Speaker 3: parrots to gelatas to many of the whale species. 819 00:48:58,719 --> 00:48:59,160 Speaker 2: Yeah, I know. 820 00:48:59,239 --> 00:49:02,439 Speaker 1: This is one of the signs that people look at 821 00:49:02,440 --> 00:49:04,120 Speaker 1: to try to figure out, how would we know if 822 00:49:04,120 --> 00:49:07,480 Speaker 1: this is actually a language versus they're just singing songs 823 00:49:07,560 --> 00:49:10,280 Speaker 1: or they're doing whatever, but they're not listening back and forth, 824 00:49:10,480 --> 00:49:14,799 Speaker 1: which leads to this question about if we find alien species, 825 00:49:15,000 --> 00:49:18,920 Speaker 1: eventually we find life on other planets. The question is 826 00:49:19,320 --> 00:49:22,440 Speaker 1: how much do we have to share with another species 827 00:49:22,440 --> 00:49:27,080 Speaker 1: for us to have some meaningful interpretation of the language, 828 00:49:27,440 --> 00:49:31,640 Speaker 1: Because fundamentally we're trapped in our internal model and a 829 00:49:31,719 --> 00:49:36,000 Speaker 1: species that's so different, we will impose an interpretation on 830 00:49:36,080 --> 00:49:40,719 Speaker 1: what they must mean by it. But I wonder when 831 00:49:40,800 --> 00:49:44,000 Speaker 1: we find an alien species, how we will ever be 832 00:49:44,080 --> 00:49:47,960 Speaker 1: able to know whether we understand enough of their language 833 00:49:48,000 --> 00:49:51,120 Speaker 1: to say that we have a meaningful interpretation of it. 834 00:49:52,840 --> 00:49:54,640 Speaker 3: I mean, when you say that, it just makes me wonder, 835 00:49:54,680 --> 00:49:57,279 Speaker 3: how do we ever know that when we're communicating with 836 00:49:57,320 --> 00:49:59,719 Speaker 3: each other's as humans, that we truly understand each other. 837 00:50:00,000 --> 00:50:05,920 Speaker 3: There's almost this undeniably huge and yet invisible gulf, like 838 00:50:05,960 --> 00:50:08,200 Speaker 3: the myth of communication is that it ever happened in 839 00:50:08,239 --> 00:50:11,359 Speaker 3: the first place. We never truly know. We can only 840 00:50:11,400 --> 00:50:15,400 Speaker 3: have clues that we are getting closer, that we're approaching knowing. 841 00:50:17,120 --> 00:50:20,359 Speaker 3: I think it's really important to call out how much 842 00:50:20,400 --> 00:50:23,320 Speaker 3: of our language is built on the metaphor of bodies. 843 00:50:23,960 --> 00:50:28,000 Speaker 3: Almost all of it is body and space, right Like, 844 00:50:28,120 --> 00:50:30,600 Speaker 3: even the things that we might think are really abstract, 845 00:50:30,840 --> 00:50:33,719 Speaker 3: like cursor on your computer. What is the root of 846 00:50:33,719 --> 00:50:36,520 Speaker 3: cursor in Latin, It's cursor the one who runs. It's 847 00:50:36,880 --> 00:50:41,400 Speaker 3: the man who runs impeded, impeded against foot. It's like 848 00:50:42,280 --> 00:50:46,160 Speaker 3: the deeper you look into language, the more you realize. 849 00:50:46,320 --> 00:50:48,879 Speaker 3: And George Lakeoff does an incredible job in a book 850 00:50:48,920 --> 00:50:53,360 Speaker 3: called Metaphors. We live by really deconstructing all of the 851 00:50:53,400 --> 00:50:55,960 Speaker 3: ways that what we think of as our most abstract 852 00:50:56,000 --> 00:50:59,040 Speaker 3: ideas can be traced back to a root of us 853 00:50:59,120 --> 00:51:01,840 Speaker 3: having bodies and talking about our bodies in a physical world. 854 00:51:01,719 --> 00:51:05,520 Speaker 2: And particular bodies, particular bodies. That that is true. 855 00:51:05,560 --> 00:51:07,280 Speaker 1: What I mean is when we find an alien species, 856 00:51:07,320 --> 00:51:09,400 Speaker 1: let's say they're more like slime, mold or something that 857 00:51:09,480 --> 00:51:14,080 Speaker 1: might make it very difficult for us to understand their metaphors. 858 00:51:14,239 --> 00:51:17,399 Speaker 3: That is exactly right. And I think the hope here 859 00:51:18,239 --> 00:51:22,239 Speaker 3: is that because we are conditioned on and live in 860 00:51:22,280 --> 00:51:24,880 Speaker 3: a physical world, that to the extent that there is 861 00:51:24,920 --> 00:51:29,399 Speaker 3: an outside world in which we share, and that that 862 00:51:29,440 --> 00:51:32,759 Speaker 3: will give the kind of grounding that's needed to do 863 00:51:33,000 --> 00:51:36,520 Speaker 3: some kind of translation. But I think it would be 864 00:51:36,560 --> 00:51:39,239 Speaker 3: wrong to say that the translations are going to look 865 00:51:39,280 --> 00:51:41,120 Speaker 3: like Google Translate. You're going to get word forward to 866 00:51:41,160 --> 00:51:43,880 Speaker 3: English it might end up looking like a translation is 867 00:51:43,920 --> 00:51:47,520 Speaker 3: more like a piece of art or poetry, where the 868 00:51:47,600 --> 00:51:51,000 Speaker 3: translation is very ambiguous, but if you spend enough time 869 00:51:51,040 --> 00:51:53,000 Speaker 3: with it, you start to get a felt sense of 870 00:51:53,040 --> 00:51:55,680 Speaker 3: what it's like. Or maybe you're right, maybe it'll be 871 00:51:55,760 --> 00:51:57,759 Speaker 3: so different that we'll never There's just some things we 872 00:51:57,840 --> 00:51:58,760 Speaker 3: will never be able. 873 00:51:58,560 --> 00:52:01,600 Speaker 1: To work right in between, will and pose an interpretation 874 00:52:01,840 --> 00:52:03,480 Speaker 1: even though it will be incorrect. 875 00:52:04,400 --> 00:52:07,760 Speaker 3: Yeah, that is true, And at least here on Earth. 876 00:52:08,120 --> 00:52:11,600 Speaker 3: There are sort of two failure modes. One is anthropomorphizing, 877 00:52:11,640 --> 00:52:15,160 Speaker 3: which is what you're talking about, is assuming that we can, well, 878 00:52:15,239 --> 00:52:17,560 Speaker 3: we can only relate to the experience of others through 879 00:52:17,600 --> 00:52:19,680 Speaker 3: our own experience. That's the only way to ever happened. 880 00:52:19,719 --> 00:52:21,640 Speaker 3: It's very simple to say, but it's actually profound when 881 00:52:21,640 --> 00:52:25,440 Speaker 3: you think about it. So there's a over projection of 882 00:52:25,480 --> 00:52:29,040 Speaker 3: ourselves onto others. And then the other side is human exceptionalism, 883 00:52:29,320 --> 00:52:32,320 Speaker 3: where we assume that our experiences are completely unique to 884 00:52:32,360 --> 00:52:35,200 Speaker 3: ushen we share nothing with other animals. And obviously the 885 00:52:35,440 --> 00:52:38,840 Speaker 3: answer that the truth is in between the two. And 886 00:52:38,880 --> 00:52:42,120 Speaker 3: then we have to have the self honesty to understand 887 00:52:42,160 --> 00:52:45,520 Speaker 3: and the way of asking questions that lets us determine 888 00:52:45,560 --> 00:52:47,320 Speaker 3: when we are over projecting. 889 00:52:47,160 --> 00:52:50,400 Speaker 1: Yes, exactly, And I'm really interested when and this might 890 00:52:50,440 --> 00:52:54,600 Speaker 1: not happen in our lifetimes, but when we discover completely 891 00:52:54,640 --> 00:52:59,239 Speaker 1: alien life, I mean as in living on other planets 892 00:52:59,280 --> 00:53:00,520 Speaker 1: that might be so different. 893 00:53:00,520 --> 00:53:02,480 Speaker 2: Maybe they don't have DNA, maybe they. 894 00:53:02,320 --> 00:53:06,280 Speaker 1: Have a different coding system, maybe they have very different bodies. 895 00:53:07,320 --> 00:53:13,799 Speaker 1: The question is how much is Earth exceptionalism true? You know, 896 00:53:13,880 --> 00:53:16,680 Speaker 1: in Star Trek they go around and they communicate with 897 00:53:16,719 --> 00:53:18,480 Speaker 1: all these aliens and they have a good time, and 898 00:53:18,640 --> 00:53:21,839 Speaker 1: you know, really understand each other to some degree. And 899 00:53:22,160 --> 00:53:25,000 Speaker 1: the question is whether that will be the case or not. 900 00:53:26,760 --> 00:53:30,239 Speaker 3: Yeah, I mean if you start thinking about I think 901 00:53:30,280 --> 00:53:32,799 Speaker 3: in Star Trek they have the crystalline entity, which is 902 00:53:33,280 --> 00:53:36,560 Speaker 3: a giant being the size of a whole planet. And 903 00:53:36,760 --> 00:53:39,080 Speaker 3: at that point, I think I'd come a little more 904 00:53:39,440 --> 00:53:41,960 Speaker 3: along your lines that the scale of which that being 905 00:53:42,160 --> 00:53:45,239 Speaker 3: is feeling and sensing is so broad. We probably share 906 00:53:45,400 --> 00:53:48,920 Speaker 3: very little but anything of roughly our size. And if 907 00:53:48,960 --> 00:53:51,800 Speaker 3: they have family structures, like then there is like hunger, 908 00:53:53,000 --> 00:54:01,400 Speaker 3: there's being tired, there's like safety, there's like familiar relationships, 909 00:54:01,640 --> 00:54:05,400 Speaker 3: there is gossip, and those things are probably conserved across many, 910 00:54:05,560 --> 00:54:09,520 Speaker 3: many different types of beings. 911 00:54:11,880 --> 00:54:15,719 Speaker 1: So we're entering a really exciting time, but the challenges 912 00:54:15,800 --> 00:54:18,400 Speaker 1: are real and there are still a lot of question marks. 913 00:54:18,960 --> 00:54:22,560 Speaker 1: For example, in the scientific literature, there's an ongoing debate 914 00:54:22,640 --> 00:54:27,719 Speaker 1: about which species might have languages. Some researchers listen to 915 00:54:27,760 --> 00:54:30,600 Speaker 1: a particular species and say that seems like that could 916 00:54:30,640 --> 00:54:33,399 Speaker 1: be language, and others listen and they say, no, way, 917 00:54:33,480 --> 00:54:37,200 Speaker 1: that's not language because there's no turn taking, and also 918 00:54:37,280 --> 00:54:40,560 Speaker 1: because the order of the sounds doesn't seem to make 919 00:54:40,600 --> 00:54:43,719 Speaker 1: any difference. And these are all valid debates because we 920 00:54:43,800 --> 00:54:47,920 Speaker 1: don't actually know what qualifies as a language and what doesn't. 921 00:54:48,719 --> 00:54:53,520 Speaker 1: Some species, for example, some songbirds do what's called dueting, 922 00:54:53,960 --> 00:54:56,560 Speaker 1: where they're singing at the same time. Does this mean 923 00:54:56,600 --> 00:55:00,520 Speaker 1: they're not doing language or is it possible? Are very 924 00:55:00,640 --> 00:55:04,600 Speaker 1: different ways of doing language. I'll give you a concrete 925 00:55:04,600 --> 00:55:07,799 Speaker 1: example of a different way of doing language, which is 926 00:55:08,080 --> 00:55:12,000 Speaker 1: sign language. It turns out that the temporal order doesn't 927 00:55:12,040 --> 00:55:14,600 Speaker 1: matter very much. In sign language. You can switch up 928 00:55:14,640 --> 00:55:17,080 Speaker 1: the order of the words and it can still mean 929 00:55:17,120 --> 00:55:20,200 Speaker 1: the same thing. And there are aspects of it that 930 00:55:20,239 --> 00:55:23,960 Speaker 1: are spatial. So, for example, an American sign language, you 931 00:55:24,000 --> 00:55:27,600 Speaker 1: can indicate that something happened in the past by doing 932 00:55:27,640 --> 00:55:30,600 Speaker 1: the signs slightly to your left, and if you make 933 00:55:30,640 --> 00:55:33,520 Speaker 1: the same signs over on your right side, that means 934 00:55:33,520 --> 00:55:37,520 Speaker 1: you're talking about the future. So it's the same signs 935 00:55:37,920 --> 00:55:41,359 Speaker 1: with this subtly different spatial position, and it can mean 936 00:55:41,400 --> 00:55:44,640 Speaker 1: different things. And I call it subtle because if someone 937 00:55:44,719 --> 00:55:48,759 Speaker 1: didn't know to watch for a slight spatial change, they 938 00:55:48,800 --> 00:55:51,839 Speaker 1: wouldn't even notice it. And even a language that only 939 00:55:51,960 --> 00:55:56,200 Speaker 1: uses sounds can be very difficult to decode because so 940 00:55:56,320 --> 00:56:00,400 Speaker 1: much of it depends on shared assumptions about meaning. So 941 00:56:00,640 --> 00:56:04,200 Speaker 1: just as an example, if I'm talking about someone named Aviva, 942 00:56:04,560 --> 00:56:07,280 Speaker 1: I use her name once, and in the next sentence 943 00:56:07,320 --> 00:56:10,600 Speaker 1: I just say her, and you know who I'm talking about. 944 00:56:10,640 --> 00:56:14,240 Speaker 1: I'm referencing Aviva. But if you are an alien working 945 00:56:14,280 --> 00:56:17,839 Speaker 1: to decode my language, you might be confused because one 946 00:56:17,840 --> 00:56:21,120 Speaker 1: minute later you hear me use the same utterance her, 947 00:56:21,520 --> 00:56:23,600 Speaker 1: but now I'm referring to someone else entirely. 948 00:56:23,640 --> 00:56:25,520 Speaker 2: I'm now talking about Sarah, but I. 949 00:56:25,480 --> 00:56:29,319 Speaker 1: Still use the word her, So the same word can 950 00:56:29,400 --> 00:56:32,480 Speaker 1: refer to totally different things. And the alien would be 951 00:56:32,600 --> 00:56:37,200 Speaker 1: very confused if it had concluded that her was the 952 00:56:37,280 --> 00:56:40,000 Speaker 1: word for Aviva, and in the same way when we 953 00:56:40,120 --> 00:56:43,680 Speaker 1: hear a whale make the same sound that we always hear, 954 00:56:43,760 --> 00:56:46,440 Speaker 1: it might be talking about something totally different than the 955 00:56:46,560 --> 00:56:51,239 Speaker 1: last time that used that sound. The context matters, and 956 00:56:51,440 --> 00:56:54,400 Speaker 1: this issue of context, in other words, what's going on 957 00:56:54,520 --> 00:56:58,759 Speaker 1: around the animal. This is why biologuers are interested in 958 00:56:58,800 --> 00:57:03,239 Speaker 1: collecting things beyond and just the audio data. Good biologuing 959 00:57:03,320 --> 00:57:08,120 Speaker 1: now uses video and gyroscope and altimeter and GPS and 960 00:57:08,480 --> 00:57:11,239 Speaker 1: any other measure they can get their hands on. And 961 00:57:11,280 --> 00:57:16,040 Speaker 1: this matters because so much of communication is about context, 962 00:57:16,280 --> 00:57:18,640 Speaker 1: and by the way, a lot of it is nonverbal. 963 00:57:19,320 --> 00:57:22,320 Speaker 1: Consider how you pick up stress from someone else even 964 00:57:22,400 --> 00:57:27,000 Speaker 1: without words, body language, the tightness of their facial muscles, 965 00:57:27,040 --> 00:57:31,160 Speaker 1: the way they're walking, and so on, and animals presumably 966 00:57:31,240 --> 00:57:36,560 Speaker 1: have many equivalents to this. Just think about smells and pheromones. 967 00:57:37,120 --> 00:57:39,439 Speaker 1: Take a close look at your dog the next time 968 00:57:39,440 --> 00:57:42,480 Speaker 1: you're on a walk. It's obvious that a lot of 969 00:57:42,520 --> 00:57:47,560 Speaker 1: your dog's language is happening silently. So all this is 970 00:57:47,600 --> 00:57:51,120 Speaker 1: to say that language can be complicated, and much of 971 00:57:51,160 --> 00:57:55,400 Speaker 1: it can be nonverbal, and this is why the challenge 972 00:57:55,440 --> 00:57:59,600 Speaker 1: of decoding animal language is a big one. And you know, 973 00:57:59,640 --> 00:58:02,080 Speaker 1: one of the things that I'm always on the lookout 974 00:58:02,160 --> 00:58:05,320 Speaker 1: for is whether we can see any evidence that animals 975 00:58:05,360 --> 00:58:09,600 Speaker 1: engage in something like storytelling. One of the classes that 976 00:58:09,640 --> 00:58:13,240 Speaker 1: I teach at Stanford is the Brain and Literature, and 977 00:58:13,280 --> 00:58:15,920 Speaker 1: I teach how weird it is that we go to 978 00:58:15,960 --> 00:58:18,600 Speaker 1: the theater or the movies or a lecture and someone 979 00:58:18,800 --> 00:58:23,160 Speaker 1: speaks and whosh, we get immediately transported into a different 980 00:58:23,520 --> 00:58:24,480 Speaker 1: space and time. 981 00:58:24,560 --> 00:58:26,760 Speaker 2: It's like a guided dream. 982 00:58:27,280 --> 00:58:29,280 Speaker 1: I'm going to do an episode on this issue soon, 983 00:58:29,320 --> 00:58:31,479 Speaker 1: but for now, I just want to point out that 984 00:58:31,920 --> 00:58:35,920 Speaker 1: we don't see bears congregating like hundreds of them on 985 00:58:35,960 --> 00:58:40,320 Speaker 1: a Saturday night listening to one bear grunt along. And 986 00:58:40,400 --> 00:58:42,480 Speaker 1: I'm not certain that we see that in any species, 987 00:58:42,560 --> 00:58:45,040 Speaker 1: but I don't know. But these are the kinds of 988 00:58:45,080 --> 00:58:48,600 Speaker 1: clues we would look for as we move forward. These 989 00:58:48,600 --> 00:58:51,000 Speaker 1: are the questions of not just do they have some 990 00:58:51,120 --> 00:58:54,720 Speaker 1: simple language, but what they can do with their language. 991 00:58:54,880 --> 00:58:57,160 Speaker 1: This is a tougher problem and one that we need 992 00:58:57,200 --> 00:59:01,440 Speaker 1: to keep our eye on. So plenty of remaining question 993 00:59:01,520 --> 00:59:05,320 Speaker 1: marks all around us, but what's clear is that technology 994 00:59:05,600 --> 00:59:10,160 Speaker 1: like biologuers and artificial intelligence are leveling us up into 995 00:59:10,200 --> 00:59:14,120 Speaker 1: a very exciting time. Not all species are going to 996 00:59:14,120 --> 00:59:18,640 Speaker 1: have something interesting to say, but many might. And if 997 00:59:18,680 --> 00:59:22,080 Speaker 1: we find we can decode animal language due to the 998 00:59:22,200 --> 00:59:26,040 Speaker 1: labors of Aseraskin and his co founders Katie and brit 999 00:59:26,200 --> 00:59:28,720 Speaker 1: and dozens of other people in this exciting field of 1000 00:59:28,720 --> 00:59:33,040 Speaker 1: animal communication, that will give us a very different view 1001 00:59:33,240 --> 00:59:38,520 Speaker 1: of ourselves and our species on this planet. Our grandchildren 1002 00:59:38,560 --> 00:59:42,080 Speaker 1: will grow up and they'll feel amazed that we considered 1003 00:59:42,120 --> 00:59:46,680 Speaker 1: ourselves the only ones and it wasn't even necessarily because 1004 00:59:46,720 --> 00:59:50,960 Speaker 1: of species chauvinism, but instead because we can only hear 1005 00:59:51,000 --> 00:59:53,920 Speaker 1: our own voices, and therefore we thought we were the 1006 00:59:53,920 --> 00:59:57,840 Speaker 1: only ones in the room. And with enough time, maybe 1007 00:59:57,920 --> 01:00:02,520 Speaker 1: we'll have enough technology and practice at decoding animal languages 1008 01:00:02,880 --> 01:00:06,880 Speaker 1: that eventually, in the more distant future, we can tackle 1009 01:00:07,440 --> 01:00:12,320 Speaker 1: extra planetary communication, and our great great grandkids will be 1010 01:00:12,360 --> 01:00:14,920 Speaker 1: amazed that there was a time when we thought we 1011 01:00:15,000 --> 01:00:17,000 Speaker 1: were the only ones in the galaxy. 1012 01:00:18,040 --> 01:00:19,880 Speaker 2: We maybe look back upon. 1013 01:00:19,960 --> 01:00:25,080 Speaker 1: As the era of loneliness, surrounded by voices of all 1014 01:00:25,160 --> 01:00:29,200 Speaker 1: types that we just didn't know how to hear. 1015 01:00:34,720 --> 01:00:36,360 Speaker 2: Please join me at eagleman. 1016 01:00:36,080 --> 01:00:40,160 Speaker 1: Dot com, slash podcasts more information and links to various 1017 01:00:40,160 --> 01:00:44,280 Speaker 1: animal communication projects and further reading. Send me an email 1018 01:00:44,320 --> 01:00:47,520 Speaker 1: at podcasts at eagleman dot com with questions or discussion, 1019 01:00:47,880 --> 01:00:50,360 Speaker 1: and I'll be making an episode soon in which I 1020 01:00:50,400 --> 01:00:54,439 Speaker 1: address those. Until next time, I'm David Eagleman, and this 1021 01:00:54,760 --> 01:01:03,840 Speaker 1: is Inner Cosmos.