Was Indus Valley script like emojis? Read on

A wheel. A fish propped up on its tail fin. A stick-figure man. The symbols, contours of a lost world condemned to taunt prying imaginations, must surely mean something. But what?

The strange etchings on ancient Indus Valley artefacts like seals, those delicate "masterpieces of controlled realism" that jolted you awake in history class, remain one of world's last great unbroken codes. A mountain of research has yielded few points of scholarly concord, such the direction of reading - right to left for most inscriptions - and the idea that some symbols represent numbers. But bigger questions remain. Is the script similar to the sound-based alphabet you're reading? What language, or languages, did it encode? One group of researchers, in fact, has claimed rather controversially that the Indus Valley people, also known as Harappans, weren't literate. And they offered to pay $10,000 to anyone who could disprove their theory.

But most experts think that view is hogwash. One of them is Kolkata's Bahata Ansumali Mukhopadhyay, a software engineer now based in Bengaluru. Her independent research on the Indus script's structure - she calls it "innovative cryptography" - was published in the Nature-branded journal Palgrave Communications in July (It can be freely accessed here.). In two forthcoming papers, she deals with semantics (or meaning) and language. In an e-mail and in phone interviews with IndiaToday.in, Mukhopadhyay, 38, described her own personal quest to solve the Indus riddle and explained how her findings contribute to the existing corpus of research. Read on.

OF EMOJIS AND ANCIENT SCRIBES

Seals from the Indus Valley civilisation. (Photos: Sepia Times/Universal Images Group via Getty Images | Photo montage by ITGD Design Team)

Just how hard can it be to figure out an ancient writing system? If it's the Indus script you're interested in, here are your odds. To start with, you only get short strings of symbols to work with - the longest Indus inscription ever found is only 26 characters long. Next: you have no idea what language these engravings might encode, and no cheat-sheet like the Rosetta Stone, a slab of rock bearing the trilingual text that helped unlock the secrets of Egyptian hieroglyphs in the early 1800s. You don't even have references like names of public figures, like kings.

Faced with such restrictions, researchers have shown remarkable ingenuity, looking for clues in such features as the cramming of signs - the key to understanding the direction of writing - and the predictability of Indus symbol patterns, which revealed a similarity to other language scripts. It's this kind of clever observation that led Bahata Mukhopadhyay to make one of her paper's main claims. She has "surely proved", she says, that the Indus script wasn't phonetically written - in other words, its symbols don't represent speech sounds like the English alphabet.

Mukhopadhyay grouped the Indus symbols into classes based on their function - this in itself was an important contribution - and then looked for patterns that determined which signs couldn't appear together. These "co-occurence restrictions", she found, couldn't be based on sound, but only on meaning. In her own words: she used differences between "phonetic co-occurrence restrictions" (emerging from limitations of human speech production and auditory systems) and "meaning-driven" restrictions, to make her case.

So if the characters don't depict speech sounds, then what are they? Bahata Mukhopadhyay believes they are largely "logographic", meaning individual symbols represent things and concepts - not unlike the pitiless rows of laughing-so-hard-I'm-crying emojis you might WhatsApp in response to a particularly acerbic meme.

But there's an important caveat here. While Mukhopadhyay says the script instances found so far in seals and tablets aren't phonetic, she believes - given the Indus people's well-documented advancement - that "parallel phonetic scripts must have existed, the way the Egyptian Hieroglyphic script co-existed with the Egyptian Hieratic script, a more cursive script used for keeping records and writing letters."

"It is possible that we just haven't been able to get our hands on samples, as most of them have been possibly written in more perishable media (Indians traditionally wrote on leaves)," she said in an e-mail.

'COMMERCIAL SUB-LANGUAGE'

A diagram illustrating the "document-specific" syntax of "formalised data carriers" like stamps and -- Mukhopadhyay says -- Indus valley seals. (Photo courtesy: Bahata Ansumali Mukhopadhyay). CLICK TO ZOOM IN.

Bahata Mukhopadhyay's research also points to a number of features of Indus script samples like seals and tablets, such as the uniform positions of inscriptions and accompanying icons, to argue that these artefacts are not unlike modern coins or stamps. The technical term for such objects is "formalised data carriers" - documents whose text and icons obey rules that are not necessarily linguistic. (For example, think of the Sarnath lion emblem on your five-rupee coin, which is always in the centre with the text flanking it, never any other way.)

Mukhopadhyay says Indus artifacts have used many examples of this "document-specific syntax", both in inscriptions and accompanying icons (see image above). But what does all this tell us about the Harappans? Mukhopadhyay thinks it suggests they were "extremely civilised" - something that is already supported by archaeological evidence - and that the Indus script was used by what was likely a multi-lingual population as a "commercial-sub language": a clever, practical invention that made trade easier.

Since we're on the subject of language: one of Mukhopadhyay's forthcoming papers deals with precisely that thorny subject. And while she believes the region inhabited by the Harappans was linguistically diverse then as it is now, she has a "hunch" that one particular language group "was surely there". (She's quick to add this doesn't mean others weren't.)

She isn't giving away much right now, but she's sure of one thing.

"I think the proofs that I have are quite substantial."

'I CAN'T SLEEP FOR LONG'

A Cambridge academic introduced Bahata Mukhopadhyay (in photo) to the field that is her current passion. (Photo courtesy: Bahata Ansumali Mukhopadhyay)

It isn't easy being an independent researcher. Bahata Mukhopadhyay juggles the responsibilities of a mother and a software MNC employee with her great Harappan adventure, which began in 2014, when she was initiated into the field by the Cambridge academic Ronojoy Adhikari.

Though "deeply indebted" to Adhikari, Mukhopadhyay soon set out on her own path, partly due to a lack of time, towards developing her "multidisciplinary approach" to the Indus puzzle. She doesn't think plunging into full-time research work is currently a possibility, but is open to leaving software temporarily if she gets a scholarship. She's done it before: she quit her day job for 10 months in 2015 to devote herself completely to research.

How does she manage?

"I can't sleep for a long time," she says. "I can't go to social circles and I can't enjoy. I can't normally watch a movie at a stretch."

One wonders how the Harappans might have drawn that sentiment.