The Indus Valley script, or what five signs can hide
Sometime around 2600 BCE, in the brick-lined warehouse district of Mohenjo-daro — a city of at least 40,000 people with better sewage infrastructure than most of medieval Europe — a merchant pressed a small square seal into wet clay and fastened it to a bundle of goods. The seal was barely two centimetres on a side, carved from pale steatite, and bore five signs above the image of a one-horned bull. Those signs traveled with the bundle through the city gates, down the Indus River, perhaps all the way to a dock at Ur in Mesopotamia. What they said, no one today can tell you.
The Indus Valley Civilization at its height, roughly 2600–1900 BCE, stretched across what is now Pakistan and northwestern India — a territory larger than Mesopotamia and Egypt combined, with perhaps a million people in its orbit. Mohenjo-daro and Harappa were its great cities: laid out on near-perfect grids, supplied with municipal wells and public baths, administered by some bureaucratic apparatus that clearly required record-keeping. Somewhere in that apparatus, a writing system emerged.
We know it now as the Indus script. More than 5,000 inscribed objects have been recovered since Alexander Cunningham published the first seal drawing in 1875, having been handed a carved steatite tablet at Harappa without fully grasping which civilization had produced it. The corpus runs to roughly 400 distinct signs, appearing on square stamp seals, pottery sherds, bronze tools, and ivory tablets. It reads right to left. Its average inscription length is five signs — occasionally stretching to twenty-six, never more.
That brevity is one of three walls the script hides behind. The second is the absence of any bilingual text: there is no Indus equivalent of the Rosetta Stone, no parallel column in a known language to give scholars a handhold. The third wall is that the underlying language itself is unknown. Dravidian, Indo-Aryan, and Austroasiatic have all been proposed, along with language families that no longer exist at all. More than a hundred mutually exclusive decipherment proposals have been published since the 1920s, none accepted by the field as a whole.
In the early 1960s, the Finnish scholar Asko Parpola began what became a decades-long effort to crack the script using early computer analysis to find statistical patterns in sign sequences. His 1994 book Deciphering the Indus Script runs to nearly 400 pages of careful argument in favor of a Dravidian connection. The field remains politely unconvinced.
What makes the silence loud is the scale of the civilization that produced it. Harappan merchants sailed regularly to Mesopotamia — Indus seals have been found at Ur and Susa, dated to 2400–2100 BCE — and the civilization was sophisticated enough to standardize weights and measures across hundreds of kilometres of territory. Then, around 1900 BCE, likely due to climate shift and river rerouting, it contracted. The cities were abandoned, and the script with them.
The gap between the last Indus sign and the first Brahmi letter on the subcontinent is roughly fifteen hundred years — the longest documented interruption in any writing tradition we know of. Every other ancient script left a trail. This one left a question mark, pressed very small, in stone.
Sources
- Indus script — Wikipedia — sign count, inscription count, average inscription length, Mesopotamian finds at Ur and Susa, Cunningham’s 1875 publication.
- Harappan script — Britannica — 400 distinct signs, brevity of inscriptions, the decipherment problem, 100+ failed proposals.
- Indus Script — World History Encyclopedia — characteristics of the writing system, logo-syllabic debate, Parpola’s Dravidian hypothesis.