Cracking the Indus code
24:27 Share

On September 28th, the Global Citizen Festival will gather thousands of people who took action to end extreme poverty. Join Post Malone, Doja Cat, Lisa, Jelly Roll, and Raul Alejandro as they take the stage with world leaders and activists to defeat poverty, defend the planet, and demand equity. Download the Global Citizen app today and earn your spot at the festival. Learn more at globalcitizen.org.com.

Ryan Reynolds here from Mint Mobile. With the price of just about everything going up during inflation, we thought we'd bring our prices down. So to help us, we brought in a reverse auctioneer, which is apparently a thing. Mint Mobile Unlimited Premium Wireless. Give it a try at mintmobile.com slash switch.

About 100 years ago, Indian and British archaeologists were on a research expedition. They were traveling to a site along the Indus River in what's now eastern Pakistan.

When they got there, all they saw were just giant mounds of mud. But they started digging. So basically when they started excavating, they found a city. Rajesh Rao has spent the last decade writing about the impact of this discovery. They found urban planning that probably rivals modern standards in some ways because they found grid-like streets with houses on either side. And then they found amazing sanitation systems in terms of, you know, houses having toilets and sewage systems.

Archaeologists dated these systems to 2500 BCE, and they were way ahead of their time. There wouldn't be extensive urban sanitation systems like this for over a thousand years. It would probably come down to the more recent Roman and Greek civilizations, right? Even there, I think the kind of sophistication I would say is not there. Archaeologists named this city Mohenjo-daro, or Mound of the Dead Men.

And it was a lot more than just an impressive sanitation system. It was a full square mile of immaculately constructed roads lined with brick walls, multiple-story homes, courtyards, even a public bath. 4,500 years ago, it was likely one of the biggest cities in the world. And archaeologists eventually found more huge cities like this, and over a thousand settlements, all thought to be built by the same people, who they called the Indus Valley Civilization.

This society spanned at least 800,000 square kilometers through modern-day India, Pakistan, and Afghanistan,

A size on par with other major ancient empires like Egypt and Mesopotamia. And then in terms of the number of people, it's one of the biggest, about a million across all the different cities and settlements. And Indus people seem to have traveled even further. They're probably the first examples of globalization, right? So there were Indus merchants who traveled in ships all the way to, it appears, the Middle East. But as archaeologists kept excavating more and more sites...

they started to get kind of confused. Ancient civilizations this big and this influential pretty much always had huge, gaudy markers of royalty, of hierarchy, of conspicuous wealth.

But amazingly, there were no palaces, no huge tombs or any kind of monumental architecture, which was the surprising part for many people. Archaeologists also didn't find depictions of warfare or slavery, which are pretty much always found in major ancient societies like these.

So they started to wonder what these people might have really been like. There have been speculations about, you know, was this like an egalitarian society? You know, is it some kind of a society where people are more equal than unequal? We still don't know.

But with other ancient civilizations like Egypt or Mesopotamia, we can answer questions like these because they actually told us. They wrote down their beliefs, their stories, and then thousands of years later, we're able to read them.

With the Indus people, though, it's a lot trickier. On the one hand, archaeologists have found some tantalizing clues. These small carved shapes that look like writing. So we found about 4,000 artifacts with inscriptions. But to this day, no one's been able to read them. And it's really a pity that their script is not deciphered because we don't quite get them, right, in terms of who they were.

I'm Noam Hassenfeld, and this week on Unexplainable, the Indus Valley civilization was one of the largest, most technologically advanced civilizations in the ancient world. We barely know anything about them. And thousands of examples of what looks like writing have just been sitting there, unread for 4,500 years. So who were they? And what could they have been trying to say? ♪

So Rajesh first got hooked on the Indus mystery when he was a kid. I remember reading about it in a middle school textbook, and it was really quite fascinating because it helps me understand my roots, right, as somebody who was born in India. Rajesh didn't get right to deciphering, though. He studied computer science, and he eventually became a professor at the University of Washington. That's actually still his day job, where he works on AI and neuroscience.

But then about 15 years ago, he took a well-earned break. I decided that, you know, I needed something that was very different from what I do, which is what we do for sabbaticals, right? Is to essentially try to explore a question that you haven't explored, you know, as part of your sort of bread and butter research. So he thought back to his middle school days. I thought that this was a great opportunity to learn something about society.

this field I'm very fascinated about. And at the same time, to ask the question, can we answer at least some questions using the kinds of computer algorithms we have?

Rajesh started his research by looking at inscriptions that have been found across the Indus Valley. Most of the inscriptions are on these square seals. So these are about an inch, inch and a half squared. And these seals kind of look like rubber stamps. But instead of being etched into red rubber, they're made of a white or brownish stone, which could be stamped into wet clay.

And what you'd see are these small etched symbols, which are along the top of the seal. Some of them look like, you know, what people have called a fish symbol. Imagine sort of a vertical Jesus fish with like fins sticking out. There's also, you know, another symbol that has whiskers, you know, for like catfish have these little things that poke out from the front of them. There are symbols that look like little stick figures, there are jars, there are interlocking rings, etc.

When you see them all together, these little symbols look a whole lot like other ancient forms of writing. At the same time, though, the inscriptions on these seals are super short. These are only on average about five symbols in length, unlike language-based inscriptions in other parts of the world. Writing is just usually way longer than that, mainly because you can't say all that much with just five symbols.

which has led some scholars to pump the brakes on this whole Indus script thing. They've suggested that these fish, these stick figures, these jars, the rings, they might just be pictures. So things like European coat of arms, you know, traffic signs, totem poles, right? You have many examples of non-linguistic symbols. These kind of symbols do communicate things, but they can't come together to form complex, abstract ideas like a full language can.

So before trying to decipher the Indus script, Rajesh had to figure out whether it was a language at all. He started by asking himself a pretty basic question. How do you know if something is writing or if it's just pictures? There are no hard and fast rules here, but Rajesh had a few clues he was looking for.

qualities that tend to be found across all kinds of written languages. So just to list a few, you know, first of all, you see linearity. Written languages tend to be organized in a clear, straight line, unlike, say, a coat of arms, which could have symbols all over the place. Secondly, we know something about the directionality.

Languages tend to be written in a certain direction, so maybe right to left or top to bottom. And then finally, whether there is any kind of structure that resembles the kind of structures you find in a linguistic script. Take English, for example. Whenever you see a Q, you're usually going to see a U right after it. That's a clue that English is a written language and not just some random grouping of symbols. And Rajesh wanted to see if the Indus script checked these three boxes of written language.

First up, linearity. In the script, you always find them very nicely linearly written. There's an order to the symbol system. Next up, directionality. That's one of the things that everybody agrees on is it was mainly written from right to left.

There's a bunch of ways researchers have figured this out, but my favorite is that they noticed symbols would often end up getting smushed together as they got closer to the left side. They would try to squeeze in three symbols instead of just one symbol, right? Because they started out writing nice and spaced out on the right side, but then as they got towards the left side, they're like, oops, I don't have space. So the Indescript passes the first two tests. It's written in a line. It's written from right to left.

But the third sign that something's a written language, whether it has the kind of structure you'd find in a linguistic script, that's way harder to figure out. Especially when you have no idea what the language is actually saying.

So Rajesh started looking for patterns. And this is where his computer science expertise came in handy. So one thing we could do is we can calculate something called entropy, which is basically a measure of how much flexibility there is. So if every symbol can be followed by any other symbol, you have very high entropy. If every symbol can be followed by only one symbol, it's very rigid. Rajesh knew that languages are in this sort of sweet spot.

Linguistic scripts occupy a middle regime where it's not too flexible, it's not too rigid. It's somewhere in the middle. Too flexible and you could get random jumbles of letters. Too rigid and you wouldn't be able to express all kinds of ideas. So Rajesh took all 400 individual symbols of the Indus script and he cataloged them in a computer and

Then he took the 4,000 stone seals that archaeologists have found, so basically all the short words or phrases made up of these symbols, and he cataloged those too. And then we ran some statistical analysis on it. So think of it as an AI model, right? So we said, okay, learn the patterns in terms of which symbol follows which other symbol. So remember that English example? Qs are almost always followed by Us. Pretty rigid.

But then those U's are usually followed by vowels, like an A for quake or an I for quilt. But you're probably not going to find any kind of consonant next. It's still a pattern, but it's a lot more flexible. And Rajesh and his team showed that the indescript does a similar kind of thing. Our result was to show that the indescript falls right among all the languages.

In the Indus script, after a diamond symbol, you're probably going to find two parallel lines. After that, there's a good chance you'll see some kind of fish symbol, but you're never going to see things like stick figures or arrows. So with this analysis, Rajesh and his team showed that the Indus script does have these three markers of language.

Not just linearity and directionality, but linguistic structure. The bottom line was, you know, there's a confluence of evidence suggesting that it's a linguistic script. So we even, I think, kind of joke saying, you know, if it walks like a duck and quacks like a duck, it probably is a duck, right? So it probably is a linguistic script. But this is all just getting us to the starting line. That the end of script is likely writing. That it's saying something.

So how do you go about trying to decipher a language this unknown when you basically don't know anything about it or the society that spoke it? That's next. Support for Unexplainable comes from Greenlight. People with kids tell me time moves a lot faster. Before you know it, your kid is all grown up. They've got their own credit card.

And they have no idea how to use it. But you can help. If you want your kids to get some financial literacy early on, you might want to try Greenlight. Greenlight is a debit card and money app that's made for families. Parents can send money to their kids. They can keep an eye on kids' spending and saving. And kids and teens can build money confidence and lifelong financial literacy skills.

Oda Sham is my colleague here at Vox, and she got a chance to try out Greenline. There are videos you can watch on how to invest money. So we took a portion of his savings to put into investing where I told him, watch the videos so that he can start learning how to invest money as well.

Millions of parents and kids are learning about money on Greenlight. You can sign up for Greenlight today and get your first month free trial when you go to greenlight.com slash unexplainable. That's greenlight.com slash unexplainable to try Greenlight for free. greenlight.com slash unexplainable.

Isn't the point of traveling to get away from it all, to feel the best you've ever felt? Then maybe you should check out Aruba. You'll spend your time relaxing on cool white sandy beaches and floating in healing blue water. You'll meet locals brimming with gratitude for an island that redefines what a paradise can be. When your trip comes to an end, you won't need another vacation because you just had the vacation. That's the Aruba effect. Plan your trip at aruba.com.

Rajesh says the easiest way to decipher the indescript would be to just find an answer key. It might sound too good to be true, but this is actually what's led to a whole bunch of ancient languages being deciphered, like ancient Egyptian hieroglyphs. So in that case, the big breakthrough, of course, came with the discovery of the Rosetta Stone. The Rosetta Stone was discovered in 1799 during Napoleon's invasion of Egypt.

It was this enormous slab of rock, about four feet high by two feet across, and it was covered in ancient writing. It was a priestly decree written in three different scripts. The first was Ancient Hieroglyphs, which was the formal script of Ancient Egypt, and no one could read it. The second was Demotic, which was the common script of Ancient Egypt, and no one could read that one either.

But the third script was ancient Greek. People already knew how to read the Greek script, and so they were able to relate parts of the hieroglyphic and demotic scripts to the Greek script. So finding something like this is the hope for researchers studying the Indus script.

But at the same time, even if an Indus Rosetta Stone were discovered, it might not be as easy to decode as it was for hieroglyphs. A big reason researchers were able to decipher the Rosetta Stone is because they already knew so much about Egypt.

For one thing, they already knew the names of famous Egyptians. You know, they were able to essentially map these very specially marked parts of the hieroglyphic inscription to names of rulers like Ptolemy and Cleopatra. Researchers could sound out these words in the hieroglyphs. So they could figure out, okay, this is a P sound. This is a T sound.

But with the Indus Valley, we don't have this advantage. We don't know any proper names. There's no historical names of kings or rulers or people or gods. And Egyptologists also knew a lot about the spoken language.

Ancient Egyptian was still being spoken or used in the Coptic church. Coptic isn't the same exact language as spoken hieroglyphs, but it's similar. So researchers had a general sense of the grammar, the structure, and most importantly, the sound. How to actually pronounce the words. So when they were deciphering hieroglyphs, they had a way to know if their work sounded right. Something we just don't have with the Indus script. We don't even know if the script was representing the same language or multiple languages.

And there's one last advantage Egyptologists had that Indus researchers just don't. Finally, of course, the Egyptian hieroglyphics, there were very, very long inscriptions on walls and texts in papyri and so on, right? So there's lots of texts that you could use to essentially debug their decoding, right, of the script. To be clear, finding an Indus Rosetta Stone would be a huge breakthrough. But because the Indus Valley civilization is just still so unknown, there's no quick fix here.

But there is one potential clue. Scholars do have a pretty good guess about what kind of language they spoke.

The current consensus is that they were probably, you know, speaking the Dravidian language, some kind of proto-Dravidian. Dravidian is a family of languages that are mostly found in South India. But there's reason to believe these languages may have started in the Indus Valley and then migrated south later on. And there are a couple spots where people still speak Dravidian languages in Pakistan, right near where the Indus Valley civilization thrived.

So if you really believe in the fact that it may be some kind of proto-Dravidian or early form of Dravidian, then you can start to guess the meanings of particular words. Rajesh says that scholars have used this kind of guesswork to show that some of these seals might actually be names. And he walked me through how this works using one particular seal.

It takes a few steps to get there, so just bear with me for a sec. So you can see that there's these seven lines stacked in front of a fish symbol. Okay, so we've got seven lines and a fish. No idea what it means. But researchers started thinking about some ancient Dravidian words for seven and fish.

One of the words for seven is ilu, and a word for fish is mean. But then mean also happens to be another word for stars. So it turns out that ilu mean is like also a name for, you know, the seven stars, right? Or the Ursa Major. So ilu mean, this fish with seven lines, it could be referring to the Big Dipper.

But even more intriguingly, Illumine could be a name. So it turns out that Illumine is like a name that has been found in old Tamil literature, right? So the claim is that these are all names that are there on these seals.

So yeah, this is kind of a roundabout way to decipher things like this, and it's far from certain. But if this seal and others like it really are names, it could be huge. Because once you know a name, you can look for it in other texts, just like researchers did with the Rosetta Stone. Still, all of this is just a clue. And clues like this are basically all we've got at this point. So when it comes to really understanding the full language of

we're kind of at an impasse, even for computer scientists like Rajesh. Unfortunately, I think just throwing more computer power at it is not going to, I think, change the state of affairs. It's not a lack of, you know, computer analysis or tools or power because we don't have, you know, very long texts, as I mentioned, and we don't have enough of them either. So at this point, our best hope might not be in computer science or looking at related languages like with the Big Dipper example.

It might be in archaeology, just going out there and finding more texts, hopefully longer texts.

And there's reason to believe that those texts might really be out there. Because despite the thousands of artifacts we've found, we've barely scratched the surface of the Indus Valley. Only 10% of the archaeological sites have been excavated. So the hope is, you know, we may get lucky. And if we start excavating some of those other sites, including some very big sites, perhaps we can find artifacts that have longer, you know, inscriptions. But it's also possible, like some of the naysayers are saying, they may not even exist.

This possibility is something we need to consider. These short inscriptions might just be all there is. If we deciphered them, we might find records of some names or maybe business transactions. But would that really tell us who they were? What they believed? We might never really know. On the other hand, these kind of short, seemingly boring inscriptions are often just the start.

Take ancient Mesopotamia. So much of their earliest writing we've found is just short, simple stuff, like names or records of how much beer different workers got for their daily rations. But archaeologists ultimately found longer texts, like Hammurabi's Code, things that really did tell us what ancient people valued, what they were thinking when they laid down the foundation of our modern legal system. So given how large and advanced the Indus Valley civilization was,

just think about what we might be able to find, what we might be able to learn from deciphering their script. We basically, um,

not doing ourselves any justice by not knowing about these people, because basically it was the largest civilization of its kind in the ancient world at that time. But they're silent. And hopefully, if archaeologists can uncover longer texts, then we really start to learn about who these people were, what were their values, did they really have an egalitarian society? I think we would learn a lot about them in that case.

The Indus people were one of the most significant ancient civilizations, and they were essentially forgotten until the modern era. Imagine how differently we would think of the world if we were only now reading Hammurabi's code or Confucius' Analects or the Socratic dialogues. If we were only now tracing the formation of ideas like math, democracy, even monotheism. These ancient societies shaped how we understand the world.

And without reckoning with the Indus Valley civilization, a fundamental building block of the modern world is still essentially a black box. So I think the Indus civilization has lessons to teach us, you know, as modern humans. Just to say, hey, look, you know, we did all these amazing things, like practiced amazing sanitation, built these great cities. We did not try to glorify ourselves. And as far as we know, we haven't found these huge palaces. So I

So I think there are some examples that they can set for us. And giving them a voice through decipherment of the script is really a worthwhile endeavor. This episode was produced by me, Noam Hassenfeld. We had editing from Brian Resnick, mixing and sound design from Christian Ayala, music from me, and fact checking from Serena Solon. Manding Nguyen is a lone wolf, Meredith Hodnot is a rogue wave, and Bird Pinkerton waited for her eyes to adjust to the sudden darkness.

Suddenly, a deep voice rang out. If you have thoughts about this episode or ideas for the show, please email us. We're at unexplainable at vox.com. And we'd also love it if you left us a review or a rating. Stuff like that really helps new listeners find us. Unexplainable is part of the Vox Media Podcast Network, and we'll be back next week.