Where does language come from?

Forget English or Spanish or Mandarin. Lots of people haven't even heard of what might be the most important language in human history, proto-Indo-European. Its last native speaker probably died thousands of years ago before it could ever be captured by any system of writing, which would seem to make studying it just slightly less challenging than describing the details of an ice sculpture based on a puddle of water. From KERA in Dallas, this is Think. I'm Chris Boyd. The thing is, there are clues. We know this early language arose somewhere between Europe and Asia and spawned hundreds of linguistic descendants that are still with us today, from Hindi to Russian to Portuguese. And scholars looking to trace those evolutions are finding ways to apply new science, like DNA sequencing, to reconstruct this language that only seems to have vanished. My guest, science writer Laura Spinney, covers all of this in her new book, Proto, How One Ancient Language Went Global. Laura, welcome to think. Thank you. It's a pleasure to be here. You write that the Big Bang of the Indo-European languages is easily the most important event of the last five millennia in the old world. That is such a wonderfully bold claim, so tell us a little bit about how you arrived at it. Well, because those languages, the Indo-European languages, of which there are about 400 living today, are now spoken by nearly half of humanity. So they're spoken on every inhabited continent today, even in the ancient world, they were spoken throughout Eurasia. And with the words, with the sounds of those languages, travelled ideas to some extent, stories to some extent. And so that radiation of the early Indoopian languages, you know, shaped the linguistic landscape for sure, but also in some ways the way that people think in those spaces. And after 1492, when the age of exploration got going and the invention of the ocean-going ship and those words, sounds, stories, ideas travel to other continents, the influence expanded again. At the very start of the book, you share some examples of shared or related words, starting with the Sanskrit word for the most powerful god in the ancient Indian pantheon. Will you take us through that and give us some examples of words that are surprisingly similar across languages? Yeah, it's really important to start with that, you're right, because that's what people noticed hundreds of years ago and what started the sort of quest for understanding this family and where it came from. So the name for the supreme deity in ancient India was Diao Peter in Sanskrit, in the Sanskrit language. Diao meaning sky and Peter meaning father. literally translated as Sky Father, we might say Father Sky. And the same name, the same deity in many respects, in the Roman pantheon was called Jupiter, which we rendered as English speakers into Jupiter. And the same deity again for the ancient Greeks was Seus Pater, same meanings of the two words, Seus Pater, Sky, Father. We shortened it obviously to...

Zeus or Zeus. In fact, they already shortened it, the ancient Greeks, sometimes called him Zeus simply. And so you can see over a vast landmass from Italy to Greece to the Indian subcontinent, the same idea, the same name is being repeated just with subtle sound changes. How can we know those common words or concepts derive from a shared ancestral language rather than being like loan words barred from nearby or transient populations? Yeah, really important because language is shaped by both processes, loans and sort of inheritance, vertical and horizontal processes. And in fact, you know, a large part of the work of historical linguists is trying to distinguish those two processes, especially looking backwards into a time when, you know, we have few written texts. But. One of the ways they can do it is that there are so-called sound laws. If you think about the vocal tract, the kind of larynx and the mouth and the lips and the tongue, all of that restricts in a certain way the sounds that we can make as human beings and the sounds that we can string together to make a language. And because of those restrictions, there are only certain ways in which sounds and words can... diverge from their common ancestors. And that generates a set of laws and linguists can sort of trace those laws back, sort of backpedal them if you like, reverse engineer them to understand what might have been the ancestral sounds of the ones that we speak today. That helps them also to distinguish loans from inheritances because the loans don't necessarily obey those laws, although once they come into the language, they are forced to. So we have clues as to what kind of influences we're looking at in each case. And how long do scientists think human beings have had the capacity for language of any kind? I would say that the consensus today is that it goes back at least to the beginning of homoseconds. But of course, it depends. So that would be 300,000 years, sorry. But that depends a little bit on how you define language. And it would certainly have been something very different to how we think about language today. Probably in many ways we might think if it is more primitive. But it is very much a... guess in game. And also, of course, we can look at our closest relatives in the primates. That's another source of information. But you can understand, I think, why it's an open debate and why there's plenty of room for divergent opinions in that discussion. So this very ancient grandmother language that we're talking about today is called proto-Indo-European. And I gather the name tells us something about where scientists think it originated. I mean, so, you know, strictly speaking, it's a hypothetical language because we've got literally no evidence that it existed. The evidence that we have is indirect, which is that we have this family of languages, not only the living Indo-European languages of today, but also historic ones that are now dead, but that we have preserved in text. So, for example, things like Sanskrit, ancient Greek and Latin. And those are similar enough. You know, we gave some examples earlier that linguists recognize a family of languages, and they...

define a root of that family, which they call proto-Indo-European, proto-simply meaning first, like parental. Obviously, the speakers of that language didn't call it that, and it may have looked much more like a chain of dialects than anything we might call a language today. Because if you think about just one simple reason for that, In the interim, in the thousands of years that separate us from the speakers of that language, we have invented amongst many other things, nation states. So we have today national languages. It doesn't take much thought also to see that those national languages rest on spectra of dialects. So there are plenty of different ways of speaking American, for example. And those cohabit quite comfortably most of the time. But in the ancient world, there were no national languages. So the most prominent feature of language would have been these extended dialect chains where, say, from one central settlement to the next, people understood each other, but the settlements at either end of the chain did not. So this language is theoretical, but lots of experts think it existed. And among those who do, some think there might have been only a few dozen living speakers at first. It's extraordinary. Well, yeah, but if you think about somewhere like Papua New Guinea today, which is one of the most linguistically diverse places on earth in the modern world, you can get whole languages which only have 2,000 or 3,000 speakers there. And that... linguists think was probably the norm of the ancient world. So, you know, we're talking about languages that had far smaller speaker populations than in Arbonne world, a very different linguistic landscape. But yes, you're right. There is this idea that the first speakers of the probably dialect rather than language, like one dialect in that chain that we might call proto-Indo-European, they... you know, they spoke a dialect that because of what happened to them in the subsequent stages of their existence, the fact that they were very successful in an economic way, they were very well adapted to their niche, which meant that their numbers increased, which meant that they radiated, and then took that dialect, which then became a language and probably fragmented into its own dialects, they carried that far and wide, and that was the origins of this family, which we still speak today. So it took maybe a thousand years for the descendants of that language to be spoken across an area that spread from Ireland to India and parts in between. If there were presumably other languages around during that time that were unrelated, do we have any idea why Proto-Indo-European was the one that sort of took over? Like, did it have features that positioned it to spread and evolve, or was this just the matter of the movement of people? Well, first of all, it was an incredibly fast spread at the beginning. I mean, we think that it had reached Ireland to the far west of Europe, but within a thousand years of the first speakers of Proto-Indo-European arriving at the eastern end of Europe. So we usually think of the starting point of the European expansion as Hungary because that's the eastern extremity of the Eurasian steppe. So the first speakers of Proto-Indo-European

were, the consensus again, is nomads, nomadic herders from the step. And they themselves probably did not go further than the step. They didn't leave the step, the environment to which they were so well adapted. But their descendants, who might have been the mixed offspring of European farmers and these herding people from the step, did go further. And they carried both genes and ideas and languages further west. And a similar kind of... process was unraveling in the east, going towards India, the subcontinent, and even as far as the front is in modern China. So that initial spread was incredibly fast, and it's a massive question as to why it happens so quickly. The explanation now is much more nuanced, much more multifactorial. Violence plays a much smaller part in it, although surely wasn't absent. But the other important aspect of this story is that the languages would have come into a pool of other languages so probably in the european case languages that came out of anatolia and the fertile crescent several thousand years earlier with the first farmers after the neolithic revolution unrelated languages at least unrelated in any reasonable time depth. And so there would have been some kind of hybridization. Now, the reason that we say that most Europeans speak Indo-European today is because Indo-European was definitely the dominant influence. But you can detect, and linguists do detect, a sort of substrate that comes from those other languages, and that on a human level would have would have... come into them, if you think about it, the nomads go west, they probably were mainly male to start with, they probably met women locally, farming women, who they married with, had children with. And you can understand how the languages might have become mixed through that process, through marriage, through reproduction, through new generations, growing up and speaking languages that were predominantly Indo-European, but had, you know, many influences from the languages that were spoken there earlier. What do we know about why some colonized places hold on to their language of origin while others very quickly convert to speaking the language of their conquerors? I mean, each... each meeting of languages is in a way an individual event. There probably are rules about how to explain what language comes out of that, but they're incredibly complex. In fact, at the end of my book, I talk about how AI might help us to sort out some of those influences effects and see if there are, you know, rules of language change that we can begin to detect in vast bodies of data. But for the moment, it's not really possible. So, you know, you can imagine some... factors that might shape it. Do the people who are coming in, are they more numerous than the locals or are the locals more numerous? Do they come in peace? Do they have technologies that give them some sort of superior edge? You know, everyone wants to be to join their gang, so to speak, and do like them and be wealthy and prosperous and have lots of prestige. That would be one reason why people switch languages. But there are all sorts of things. I mean, biologists have begun to detect.

diseases that were circulating at that time. And there's an idea which is gathering evidence that there might have been some kind of epidemic in Neolithic Europe, either that emerged from the farmers themselves or that was brought in by those step nomads and that decimated the local population. And you can imagine in that situation when sort of large stretches of land were... literally vacated because the people were dying, that the income is, especially if they had some kind of immune protection, which is also looking like a possibility, might have quite easily imposed themselves and their languages in those places. We should note here, Laura, there are a handful of very powerful language families somewhat on par with proto-Indo-European. How does the modern influence of... Indo-European compare with other very big original languages like Sino-Tibetan. Yeah, it's a really good point. So depending on how you define language and define a language family, there are between sort of 150 and 300 language families on Earth. And of those, there's a great imbalance in how many speakers each family has. So there's a top five which is spoken by most of us. And in those top five, there's some African families. There's the... Asian, Afroasiatic languages, which include things like Arabic. But the top two are Indo-European and Sino-Tibetan. The major representative of Sino-Tibetan is Mandarin Chinese. The major representative of Indo-European is the one we're speaking in now English. Mandarin Chinese has more native speakers than English, but Indo-European as a family has more native speakers than Sino-Tibetan. So they are really kind of vying there for top... top slot. And you know, you can imagine that some tilt in geopolitical power scape could change the order. It's certainly possible. But, you know, because there is, it's the thing we haven't discussed yet, but there is this hugely political dimension to language always has been its, it's inherited in its definition. You can't get away from it really. But yes, so Indo-European is not on its own at the top there. Yeah, what can we know about how the spread of language might also tell us something about the spread of ancient cultures and even cultural identities?

So, I mean, languages are highly dynamic. I think that, you know, that was something that a thousand years ago people didn't acknowledge. They thought their language was sort of static things and they could preserve them. Today, we don't think that, although I think there's some resistance amongst people to thinking their own languages might be changing and even dying, which is nevertheless a clear fact. We're losing languages quite fast in this world. So languages are dynamic. Languages don't necessarily travel. hand in hand with ideas or even with genes. These three things shape each other, genes, ideas and words, but they don't necessarily travel together. So they can do. I mean, for example, there's been some really interesting research lately to show that some of the oldest stories that we tell in Europe might have arrived with those Indo-European speakers who first came in about 5,000 years ago. there are also stories that are much younger and probably stories that are much older too, and stories probably change faster than languages. So there are the different dynamics to these things. But you can imagine that if you have a word for a concept and that word is handed down, then you might, you know, there are all ways in which the language that you speak shape the... way that you look at the world. That's not a concrete thing. I mean, there was an idea in the past that the language that you spoke sort of blinked you in some way as to how you saw the world. And we now know that you can think about things that you don't necessarily have words for. Nevertheless, different languages do push you to think about the world in different ways. They emphasize different aspects of the world. And so, you know, native speakers of English might... perceive the world or understand the world quite slightly differently from the speakers of Chinese, which I think is a fascinating idea. I mean, language obviously had benefits for humans who needed to work together to coordinate activities and keep one another alive. Why might storytelling specifically have provided a survival advantage? There are several theories, and it might serve several functions. We are inveterate storytellers as human beings. We've always done it and we can't help ourselves. And there is one idea that it's a way of handing on. essential information for survival. So if you look at some of the oldest stories, the story is told, for example, by Indigenous Australians, you'll find that they're very, or Native Americans indeed, you'll find that they're very tied to the landscape. And they tell you the person information about where you are and how to orientate yourself in that landscape and plants and animals that are found there and that might be dangerous or good to eat or toxic and so on.

And so there is this idea that it was a vehicle for survival information. But there's another kind of class of theories that hold that stories are about gluing us together. They're about social cohesion. If you think about the way that stories were recounted traditionally, especially in prehistory before writing, it was bards, it was poets who sang the tales to a group of people often assembled for a feast, especially in Indo-European traditions. And so it was a way of knitting people together, giving them a sort of shared identity. You know, there's even evidence to show that really good storytellers can have the effect on their audience of synchronizing heartbeats and pulses and even brainwaves. So, you know, there's even research to show that there's a biological basis to this idea of knitting people together and giving them a cohesive identity. Does language diversity track it all with biodiversity around the planet? Yeah, there is evidence for that that the richest parts of the world in terms of biodiversity, which would be generally speaking that the tropics also have the highest linguistic diversity. But there's some kind of... nuance to that argument again I come back to the example of Papua New Guinea which is so interesting because it is sort of this glimpse of how diverse the world might once have been linguistically and it was once thought that it was the kind of linguistic diversity of PNG mapped quite neatly onto its biological diversity, onto its very variable topography. You know, it has highlands, it has coastal areas, it has every kind of biome basically present on that island within a tropical context. And it turns out that there is no such mapping and that much of that diversity can be explained better by... by human behavior, human psychology, and groups essentially wanting to belong to other groups or distinguish themselves from other groups. And there are ethnographers and linguists who have captured, who have observed that kind of behavior. So it looks as if there are these social processes which were also very powerful in forming our linguistic landscape. I imagine this varies, Laura, but how long might it have taken before, assuming that there was this one? language that we now call proto-Indo-European. How long might it have taken before it evolved into different pockets of speakers so that it was impossible for people who spoke one of these derivatives to understand people who spoke another?

Yeah, it's a fascinating idea. So just to be clear, because maybe I was a bit ambiguous earlier, I mean, linguists do generally think that this was a real language spoken in a real place by real people. But maybe those were not, you know, like genetically homogeneous people might not even have been. people who saw themselves as one society, there might have been sort of related cultures spread out over a certain amount of space, especially since these people who we think spoke that language were very mobile. They were nomadic, and they moved about the landscape all the time. So we're probably, you know, thinking about a dialect chain again. And as I said, they were very successfully adapted to their niche. They were the first to become properly nomadic in that grass, stepland space. that enabled them with their herds to sort of unlock the huge energy reserves of the step for the first time, because their ancestors have been much more tethered to river valleys, essentially. And so what did they do with this energy? They plowed it into more of themselves. They were also very tall and lived to quite a long age, by the way. they radiated because when you have more people, you need more space, especially when you live off herds, the herds need grazing. So they moved out with quite an exceptionally fast rate, in fact. There's a whole sort of side debate about whether they had already domesticated the horse. If they were riding, that would explain how they managed to radio it so fast. They certainly had wagons and wheels and oxen to pull their wagons. So, you know, it was very fast that first radiation. Consequently, the fragmentation was probably quite fast because there was no such thing as the internet, no such thing as national borders or writing to sort of slow the change. But just to say that, you know, it can be variable. And when we're looking at these particular people, these speakers of proto-European, radiating out of their initial homeland cradle, it probably went quite fast. Just as an example of that rule of thumb, we of course speak modern English. If we're concentrating really hard, most of us can understand a fair amount of Shakespeare-era English, which came at kind of the tail end of the era of Middle English. Old English, on the other hand, feels mostly like a foreign language. And even in Shakespearean English, there are things that strike us today as not rhyming that would, in fact, have rhymed during Shakespeare's time. Yeah, as you say, we can understand it with concentration. If we go back to Chaucer, it gets a little bit harder again, Middle English. And if we go back to Bear Wolf, old English, it's really literally another language. So yeah, and, you know, we're talking about a couple of thousands of years there of linguistic evolution. And that's English. I mean, English is several languages in one. And, you know, We also have the future to look forward to will English, you know, evolve again into a language that will be not understandable to those of us who speak today is quite possible. Is there a pattern to the kinds of words within a language that are most likely to change over time and the ones or the ways of speaking that have the best chance of sticking around through many centuries? There are, yes, there is a kind of...

core vocabulary that historical linguists talk about, which is the sort of core that doesn't change very much over time and space or changes less than the rest. And these tend to be very important sort of universal concepts, things that, you know, are the bread and butter of our daily... conversation, the nuclear family, father, mother, son, daughter, etc. In the ancient world, the animals with which you lived perhaps, and of which, you know, on which your survival depended, basic concepts like, water, wind, night, star, so necessary for navigation in the old world. And body parts and maybe, you know, numbers are counting. So these things which, you know, misunderstandings in them are very difficult to tolerate because it might literally be a matter of life and death mean that they've been quite stable over time. Other more sort of peripheral... vocabulary might have changed more. And I think you could probably say that the words that change the most are the ones that are the taboo words. We might think of curse words today because they're unsayable. And so they're often being replaced, circumulacuted, euphemized. And you can think of any ready examples from our modern. lexicon of bad language, you know, we're often sort of changing, I can think of frick or fudge words that, that are less sensitive but are replacing an earlier taboo word. We should step back for a moment here and talk some more about how scientists can know anything about this language that emerged before anybody had created a system of writing to capture it. I think the oldest alphabet is like 3,500 years old. Tell us about this process of hunting down the ghost of a language that has not existed in its original form for thousands of years. Yeah. So I suppose the concept of an Indo-European family, though the shape of that family has changed over time. But the first kind of inkling that there might have been related families of this line is about 200 years old. Well, no, it's older than that because you can start with, for example, Dante in the Middle Ages, who recognized similarities across the descendants of Latin, you know, French, Spanish, Italian and so on. And he had quite a challenge on his hands and persuading his...

contemporaries that these languages had diverged from Latin, that there had been linguistic evolution, that even if you couldn't see that evolution happening in your lifetime, you know, just like you couldn't notice a person growing up, it was still happening. In fact, people laughed them pretty much out of town. But gradually, that idea took hold and people began to notice resemblances wider than... what is now an individual branch of the Indo-European family, so the Italic or Latin branch, the Romance branch. They began to see resemblances between, for example, the Germanic branch and the Italic branch, the Celtic branch, and the Germanic branch. People got very excited about resemblances, echoes, if you like, between Latin, Greek and Sanskrit, because, of course, they cover such a huge geographical space. So things like Pater in Latin meaning father and pita, the Sanskrit, Deus, and Deva, Raj. Raja and Rex, these similarities which seemed to them could not be coincidental. So gradually there was born this concept of of a family, that this family had 12 main branches. I think we've listed most of them, but Baltic, Slavic, Germanic, Greek, Albanian, Armenian, Iranian, Ironic, Indic, what have I left out? I definitely left out some because of the Nuff fingers. But there are 12 main branches. Yes, of course I've missed out, Anatolian and Tokarian, both of which have long been dead, but are nevertheless very important members of this family, Anatolian of which the main representative was Hittite. Hittite was the first Indo-European language to be written down. We had traces of Hittite words going back to 2000 BC, 4,000 years. You have to wait another roughly... 500 years before you see the next oldest Indo-European languages to be written down, which are ancient Greek and Sanskrit, and then Latin first sort of hoves into view in text around 500 BC. So those are the first written traces of Indo-European. But there is a big gap between the date when we think that Proto-Indo-European was spoken about 5,000 years ago, about 3,000 BC. and those first written traces in the daughter languages a thousand years later. So the evidence is sparse and there's a lot of extrapolation that goes on. Luckily we have, of course, all the living Indo-European languages as well as those historic ones. And so, you know, historical languages have enough to work on to start to trace out of the contours of that family and even to start to reconstruct the common ancestor, the last common ancestor. Laura, what is the difference between a language that truly goes extinct over time and one that evolves into a different form?

Yeah, it's a good question. It's a grey line. I mean, you might say, for example, that the languages that predated the Indo-European languages in Europe, the languages of the first farmers to come into the continent, are extinct. And yet, you know, this substrate is detectable in the Indo-European languages of words that came into them from those much older languages. You could look at Basque, which is a relic. of those older languages in Europe, protected a little bit by its mountain domain there in the Pyrenees, and now heavily affected by the Indo-European languages around it, but still recognisably Basque, recognisably different, a language isolate in the modern world. And so, yeah, are they extinct? There are relics. There are traces in... in the languages that we speak, you could even, you know, there are probably traces of some of the oldest human language has ever spoken in the languages we speak today. We just don't necessarily have the tools to detect them. So when, you know, when babies are growing up and listening to the sounds around them, the speech sounds around them, they are unwitting me filtering in a way and taking the most... you know, useful aspects of those languages. And they're always infinitesimally changing those languages. That's one of the main drivers of linguistic change is simply new generations growing up and learning them. And that would have been a great motor of linguistic evolution over time. And so it's, you know, it's an unceasing process. And where you place the line for one language ending in a new one beginning is to some extent arbitrary. Is there any way to know how common it is that languages that die die because they're spoken by smaller geographically isolated populations as opposed to they die because of the encroachment of like more commonly used languages that younger people start speaking in addition to their mother tongue? I mean, I think that there's many, many different ways that language change can come back. And you could look at something quite clear cut, like Vanuato, another Pacific archipelago, also like Papua New Guinea, very, very diverse. Some of the smaller languages spoken there, and they're all relatively small, were wiped out. I know for a fact, during the 1918 flu pandemic, which is the subject of my previous book, because those... Islanders were very heavily affected, badly affected by that infection. When it came in, they had relatively little immunity against it. So it was a complete catastrophe in those islands, and some of the languages were simply snuffed out with the population, sadly. And if you think about, you know, go to the other extreme in a language like English, I mean, English is very clearly fragmenting and diverging. quite fast at the moment to the extent that, you know, you can sometimes see the different variants of English subtitling each other on TV because, you know, they don't understand each other. But this is different, again, from the sort of explosion of Latin into the romance languages that happened, you know, hundreds of years ago, because the average citizen or subject of the Roman Empire would have been illiterate, whereas today we have a standard written language of English, which is shared by...

all the speakers of the different variants. And that is kind of holding it together into one language to some extent. So different processes are at work today. So talk more about how advances in DNA sequencing are telling us about the movement of language around the planet. Are scientists typically relying on people who live in a certain place and give samples because their ancestors have lived there a long time? Or are they taking DNA samples from the bones of ancient bodies that are found? Both, but the latter has really come into its own in the last decade or a couple of decades. So, I mean, to go back a step, obviously, when we're talking about a language as in the case of pro-tint European that was spoken before writing, we have no written texts to get at it. So we have to use other tools to try and understand where it was, who it was spoken by, and how it spread. archaeologists have long collaborated with linguists in this task in that you can see through archaeology the spread of ideas and culture and you can perhaps infer quite a lot about you know because language is often spread with culture but you can't necessarily tell about the movement of people so genetics is a really powerful third tool in this quest because it can shed light on you know, for example, how this person in this ancient cemetery is related to that person in that ancient cemetery, and therefore you can detect that there was some migration between the two. And migration turns out to be been really common in the ancient world. It was probably, you know, the main motor of language spread and dispersal. So, you know, and in, you can then kind of sort of triangulate with the archaeology, with the linguistics, and very often you find that the, um, the spread of people, the spread of ancient migrants, maps quite well to the diaspora of ancient languages. That's also the case with Indo-European. So, you know, it's not perfect, it's indirect, but in this triangulation, we have a... quite a powerful tool to, you know, a torch to shine on those ancient languages. And to, you know, get back to your question, we can also see traces of those movements in modern populations, of course, but the thing about looking in modern populations is that later movements sort of dilute earlier ones. So before you get very far back, you kind of lose the signal. You can't really see what was happening because, you know, it's crowded out by... by more recent movements. So the ability to extract DNA from ancient remains and, you know, differentiate it from modern contamination and analyze it, which has only really become widely possible in the last decade, has really transformed this field, this field of understanding ancient languages and is indeed the reason that I wrote my book. Can language offer any clues about the spread of technologies like smelting or the wheel or whatever the case may be that artifacts alone can't answer?

I mean, when you use a tool, for example, or a technology, you need a way to talk about it. And that holds for farming, for the animals you're domesticating as much as for the furnace and the metal and the specific artisanal tools that you're using to, you know, hammer a piece of copper into an all or a dagger or a piece of jewelry. But the thing is that, so yes, there is some, you know, parallel development of vocabulary and culture and technology. But we're quite good also as human beings at recycling old words. So the very simple example I give in my book is that mouse in English, which comes from the Latin moose, can mean both a small furry rodent, which is the only thing it meant to the Romans, or it can mean the cursor that you use to move, you know, the device that you use to move the cursor around your computer screen. So we have hijacked that word and given it a second meaning because we needed, we had this new concept of that device, which we needed to give a name in order to be able to talk about it. So one of the sort of corollaries of this is that when historical linguists are looking back and trying to understand what was the word, what was the world inhabited by those speakers of the, of an ancient reconstructed language, they have to be very careful that they don't map the wrong meaning onto a word and, you know, attribute technologies to those people that they might not have had or indeed deprive them of technologies that they did have, because all of that will go towards where we place them in space and in time and we might, you know, make errors in those. in those assumptions and in those conclusions. Laura, given the many different ways we use vowels and consonants in all the languages descended from proto-Indo-European, is there any way to guess what it might have sounded like? Yes. I mean, when historical linguists reconstructed, they're reconstructing. every aspect of it, not just the words, the vocabulary, but also the sound and the grammar. And so there have been many attempts over time to, you know, speak in that language. There are recordings you can find online. There's a lot of disagreement over it. I mean, it's hugely controversial because if you think that we can't even know what Oxytan, a language that was widely spoken in France in the Middle Ages, what it sounded like. and how it changed from village to village. You know, we're going back thousands of years earlier and we have no written texts. So it's that much more difficult. But there have been attempts and it follows these same rules, this comparative method where you're looking at the sounds across different but related languages and trying to understand how following the sound laws those might have descended from a word spoken in a certain way. I can easily understand how pronunciations and even words would morph over time as different populations evolve different versions of a language. What do we know, though, about why grammatical rules are so varied among languages with a common ancestor? Because they can be in a word.

you know, there are choices that are made over time, and often those choices are the result of historical events. So one of the messages that I try to get home through my book is that languages are sort of time capsules, they reflect their own journeys through the world. And if you think about Hittite, one of the Anatolian languages that died out more than a thousand years before. before the common ear and before the birth of Christ. That was a language that came into the Anatolian Peninsula, or its ancestor did, from outside, and that absorbed much of the vocabulary of the languages that it met there, unrelated languages, but it imposed its grammar on them. the elements that they absorbed then forced other changes in their language so that it could work as a structural whole. And so these are, you know, some of the ways in which a grammar must adapt to influences from outside. And that has happened since the beginning and is continuing to happen today. So do linguists think they can estimate the timing of the arrival of different languages that grew out of proto-Indo-European? Like did ancient Greek exist long before Old English? They can date these things by cross-referencing. So they have this idea of linguistic evolution of when different languages sounded.

had different sounds and then they can cross-reference that with historical text. They can cross-reference that with the archaeological record to see, you know, when people might have been splitting, when people might have thought of their languages as different and so on. And so, yes, they can start to put dates on those, on that linguistic evolution and that's how they start to construct their family trees. But again, it's not a... You know, it's a little bit of an art, a little bit of a science, and there's a lot of controversy in the construction of those trees. And anyway, the trees are, you know, in some ways, a simplication of reality. We spoke right at the beginning about how languages are shaped by borrowing by loans and horizontal processes as well as by vertical inheritance processes. And a tree only captures the vertical part of that, the inheritance part of that. When it comes to modern technologies, I mean, we have the internet and a lot is written down, but we can also hear one another speak languages with an ease that humans have never had before. Are these records we're creating of how people speak slowing down or speeding up the way languages evolve? Probably both. I mean, there are... There are some very slowing processes at work now. Writing, you know, I mean, we've only had widespread literacy for a couple of centuries. If you think about that in relation to the fact that the Indo-European languages are, what, 5,000, 6,000 years old, and language itself 300,000 years old, at least, probably. though that is controversial, then, you know, literacy is a very new invention, as our nation states, just a couple of centuries old. And these are all things that are sort of forcing language to standardize and to not change very much. But at the same time, you know, you can see how the internet is giving rise to new sort of nooks and crannies and linguistics. little, you know, archipelagos where people speak their own code languages, that's possible. And also, you know, popular culture, take for example, the huge popularity of Afro beats music or even just rap music, you know, how that spreads through the internet. So, and takes a certain form of language with it. So popular culture can be a huge driver for... what might be originally quite marginalized variants of a language and can, you know, blast them around the world. And there are always these two, you know, the kind of very macro level. There are two big forces which are intention, which are always shaping language. And they're very human, and language is a human tool, of course. It's our oldest tool. The first one is our desire to belong and, you know, to communicate with each other. And that's what, you know...

drives us to find a language, a common language, a common way of communicating. And then there's our desire to belong to our own group to differentiate our group from the out group. And that is the force that pushes us to distinguish our language from other people's languages. So these two forces are always in tension, and the relative power of one or other depends on the historical context and in the context of which speakers of different languages meet. But they are both always present to some extent. Science writer Laura Spinney is author of the book, Proto, How One Ancient Language Went Global. Laura, thank you so much for making time to talk about this. It's been a pleasure. Great conversation. Thank you. Think is distributed by PRX, the public radio exchange. Again, I'm Chris Boyd. Thanks for listening. Have a great day.