Skip to content
Stela
ENFR

Halls

We built IWe discovered IIWe imagined IIIWe changed the world IVWhat we lost VWhat we leave behind VI

Explore

TimelineMapJourneysSearch the archive

Language

Français

Settings

RH-0138LanguagesAsia

A seal from the Indus, and the readers we lost

Dated
2600 to 1900 BCE
Place
Mohenjo-daro, Larkana District, Sindh, Pakistan · The Indus (Harappan) civilisation, Mature Harappan period
Coordinates
27° 20′ N · 68° 08′ E (approximate)
Remains
Intact
Two small square objects side by side on a grey background. On the left, a seal of pale, whitish stone, chipped at the corners, its picture carved in reverse; on the right, its modern impression in buff clay, where the picture reads the right way round: a bull-like animal with a single long horn, in profile, a harness pattern on its shoulder, its head lowered towards a tall tiered object on a stand. Above its back runs a short row of signs, among them a diamond, a pointed arch filled with lines like a little tree, and a single upright stroke.The image could not be loaded.Image not saved on this deviceSee it on the original record
ZoomLoad zoom (≈ 97 KB)Zoom needs a connection

The Metropolitan Museum of Art, Dodge Fund, 1949, 49.40.1, CC0, via Wikimedia CommonsCC0-1.0Original record

Data saver is on. Images load only when you ask. You can change this in Settings at any time.

From the archivist

Someone pressed this square of stone into soft clay, and the clay kept a picture: an animal with a single horn, its head lowered before a tall object on a stand, and above its back a short line of signs.

In the cities of the Indus, at Mohenjo-daro and at Harappa, people did this thousands of times. They carved seals small enough to close in a fist, fired them hard and pale, and pressed them into clay, perhaps to mark a bundle of goods or a door. Whoever received the clay read the signs at a glance. A name, perhaps. A title, an office, a town, a god. It was obvious to them, the way a signature or a label on a shelf is obvious to us.

Then the cities emptied. The people who knew moved away, or stopped teaching their children, and the signs went on lying in the earth with no one left who could hear them.

Archaeologists have gathered some four thousand of these short inscriptions. They have counted the signs, measured their order, argued over whether they write a language at all. No one has been able to say a single word of them aloud.

What was lost was not the stone. It was the reader.

Established

  • Stamp seal of burnt steatite, 3.8 by 3.8 cm, with a modern impression; Indus civilisation, Mature Harappan period, about 2600 to 1900 BCE; Metropolitan Museum of Art, 49.40.1, bought in 1949 (Metropolitan Museum of Art).
  • The impression shows a one-horned animal facing a tall object that the museum identifies, with a question mark, as an incense burner, under a short inscription (Metropolitan Museum of Art).
  • Some 4,000 inscribed objects are known, on seals, tablets, pottery and other materials; an inscription averages about five signs, the longest on a single surface has 17, and some 400 distinct signs have been counted (Rao et al. 2010).
  • The script has not been deciphered, and no reading proposed for it has been confirmed (Rao et al. 2010; Sproat 2014).
  • Farmer, Sproat and Witzel argued in 2004 that the signs did not record speech; Rao and colleagues answered in 2009 that the order of the signs has a statistical structure close to that of languages (Farmer et al. 2004; Rao et al. 2010).

Interpretation

Whether the Indus signs write a language at all is debated. Farmer, Sproat and Witzel (2004) held that the inscriptions are too short, and the signs too unchanging over some six centuries, to be writing, and read them as a non-linguistic system of emblems with religious, political and social uses. Rao and colleagues (2009) measured the order of the signs and found it closer to languages than to non-linguistic systems such as DNA or a computer program; Sproat (2014) contested that such measures can tell the two apart, and Rao and colleagues maintain that their result adds weight to the other language-like features of the script rather than proving them alone. Proposals that the signs record an early Dravidian language, or another, remain unproved. The animal on this seal is called a unicorn by convention, and the object before it may be a standard, an altar or an incense burner.

Looking ahead

A few thousand years from now, someone may turn one of our bank cards over in their hand, or a door key, and ask the question we ask of these seals: was this writing, and what did it say?

Sources (5)
  1. The Metropolitan Museum of Art. Stamp seal and modern impression: unicorn and incense burner (?), Indus, Mature Harappan, ca. 2600 to 1900 BCE, burnt steatite, 49.40.1, Dodge Fund, 1949 (collection record, open access). · www.metmuseum.org
  2. Farmer, S., Sproat, R., Witzel, M. The collapse of the Indus-script thesis: the myth of a literate Harappan civilization. Electronic Journal of Vedic Studies 11 (2), 2004, p. 19-57.
  3. Rao, R. P. N., Yadav, N., Vahia, M. N., Joglekar, H., Adhikari, R., Mahadevan, I. Entropy, the Indus script, and language: a reply to R. Sproat. Computational Linguistics 36 (4), 2010, p. 795-805 (defending their paper in Science 324, 2009, p. 1165). · aclanthology.org
  4. Sproat, R. A statistical comparison of written language and nonlinguistic symbol systems. Language 90 (2), 2014, p. 457-481. · doi.org
  5. Connecticut College, Digital Asian art collection. Stamp seal (Metropolitan Museum of Art 49.40.1), provenance note. · diluo.digital.conncoll.edu

Related fragments

An old black and white photograph of a wooden tablet with rounded corners, one face covered from edge to edge with tight horizontal rows of small carved signs that stand out pale against the dark wood: little human figures, crescents, loops and zigzags.Image not loaded · data saverThe image could not be loaded.Image not saved on this deviceRH-0116

before 1864

Rongorongo, the tablets no one can read

Rapa Nui (Easter Island), Chile · Oceania

A small, thick tablet of pale clay, almost square, photographed against a dark grey background. Its face is divided into boxes by incised lines; inside them, signs pressed and drawn into the clay: ears of grain, round holes and crescent notches for numbers, a vessel, a triangle, a rectangular grid. On the lower left, a rolled seal has left a faint scene with an animal whose tail curls into a spiral.Image not loaded · data saverThe image could not be loaded.Image not saved on this deviceRH-0070

3300 to 2900 BCE

The first writing: proto-cuneiform tablets of Uruk

Warka, Al-Muthanna Governorate, Iraq · Asia

The Rosetta Stone lit against a black background: a dark grey granodiorite slab with a broken, sloping top edge and a band of pinkish stone at its upper left corner, its face covered from edge to edge in rows of small incised writing, in three bands that are hard to tell apart at this distance.Image not loaded · data saverThe image could not be loaded.Image not saved on this deviceRH-0018

196 BCE

The Rosetta Stone

Rashid, Beheira, Egypt; now in London, United Kingdom · Africa

A khipu laid flat behind glass: from a curved main cord hang dozens of long cords in dark brown, tan and cream, some tied with small knots at different heights, a few carrying shorter cords attached to them; a single thin cord trails away in a wide loop around them.Image not loaded · data saverThe image could not be loaded.Image not saved on this deviceRH-0052

1400 to 1800

The khipu

Peru · Americas

Enlarged image · RH-0138

Two small square objects side by side on a grey background. On the left, a seal of pale, whitish stone, chipped at the corners, its picture carved in reverse; on the right, its modern impression in buff clay, where the picture reads the right way round: a bull-like animal with a single long horn, in profile, a harness pattern on its shoulder, its head lowered towards a tall tiered object on a stand. Above its back runs a short row of signs, among them a diamond, a pointed arch filled with lines like a little tree, and a single upright stroke.

The Metropolitan Museum of Art, Dodge Fund, 1949, 49.40.1, CC0, via Wikimedia CommonsCC0-1.0Original record