Sahiwal District, Punjab, by an old bed of the Ravi, Pakistan

Harappa

Mature city about 2600 to 1900 BC (the mounds are the argument)

Beside a dry course of the Ravi, a chain of mounds is what remains of the city that gave the Indus civilisation its name. The mature town covered more than 100 hectares between about 2600 and 1900 BC. The sister city is Mohenjo-daro, downriver in Sindh.

Why this is extraordinary

Harappa sits by an old bed of the Ravi, in Punjab. The city is a set of mounds. The tentative listing gives the excavators’ sequence: a small farming settlement about 3300 to 2800 BC, an early phase about 2800 to 2600 BC, the Harappan phase about 2600 to 1900 BC when the place covered more than 100 hectares, then a transition and a late phase down to about 1300 BC. In those mature centuries there were at least four walled centres, each on its own mound. Mound AB is the western one. Mound E is to the east and south-east. The rubbish of occupation is over 13 metres deep.

What you walk on is baked brick, the same material as at Mohenjo-daro. On Mound F, long walls and platforms are still called a granary and a great hall. At Mohenjo-daro the platform beside the bath got the same name, and no store of grain was there to match it. The nickname can be kept as a label. It should not be kept as a finding. In the 1850s thousands of bricks were taken for railway ballast, so the latest levels are the ones most often missing.

Streets of this civilisation are often close to the cardinal points. A post’s shortest shadow at noon lies north-south. Equinox sunrise lies due east. Either is a sky method, and copying a neighbour’s wall is a third method that looks identical afterward. The sources used here print Harappa’s mounds and phases. They do not print a sunrise azimuth. What this changes: the type site of the Indus cities is a measured sequence of mounds, paired with Mohenjo-daro, not a single granary and not yet a surveyed solar plan. The seals carry the same short, unread signs.

Why it matters

Harappa is the type site. If its walled mounds are several centres rather than one citadel and one lower town, the Indus plan is more varied than a single diagram. The script is the same unread system as at Mohenjo-daro. A street near the cardinal points can be set from the sun, and this room does not have an azimuth that proves it was.

The tentative listing describes low mounds and cemeteries south of an old Ravi bed. The sequence proposed by the excavators runs from a small agricultural settlement around 3300 to 2800 BC, through an early phase around 2800 to 2600 BC, to the Harappan phase around 2600 to 1900 BC, then a short transition and a late phase to about 1300 BC. The deposit is over 13 metres deep. In the mature centuries the city held at least four walled centres on their own mounds. Mound AB is the western height. Mound E lies to the east and south-east. Thousands of bricks were taken in the 1850s for railway ballast, which cut away late levels. Dholavira and Lothal are other brick cities of the same civilisation.

Photo tour

3 photographs, unedited. On a phone, swipe the row. Choose one to look closely.

What is still unexplained

These sit beside the consensus account, with the same visual weight. None of them is marked solved.

Open Walls called a granary On Mound F the exposed brick includes long parallel walls and rows of platforms in the area long labelled a granary and a great hall. The same nickname was used for a platform at Mohenjo-daro, and there the grain never turned up. A working floor for grain is a fair reading of circular and rectangular platforms. A hall, a store for something else, or a foundation whose upper floor is gone fits the same brick. The tentative list dates the exposed structures on mounds AB and F to the third millennium. It does not say what was kept in them.

Open A grid and the sun Mature towns of this civilisation are often described with streets close to the cardinal directions. At Mohenjo-daro the lower town meets in right angles. Harappa’s plan is several walled mounds rather than one grid, and the sources used here do not print a sunrise azimuth for these walls. A right angle can be set in three ways you can still try. The shadow of a post is shortest at noon and lies north-south. The equinox sun rises due east and sets due west, so a sighting along that morning gives the same cross. Or a builder copies a line from a street already laid. The first two are sky readings. The third is a habit. They leave the same corner. What a cardinal street may have meant is practical, a city that shares one orientation, and possibly a year marked when the sun rose on the axis. Nothing written that we can read says which. Cemetery H, at the end of the sequence, is the listing’s picture of the urban, literate city breaking into a looser pattern.

Location and dating

Location: Sahiwal District, Punjab, Pakistan, south of a dry bed of the Ravi. The pin is the mounds. Part of the old city lies under fields and under the modern village.
Period label: Current consensus — open to revision The excavators’ Harappan phase is about 2600 to 1900 BC, inside a longer occupation from the fourth millennium to about 1300 BC. Current consensus on a mature Indus city of baked brick. Open on the granary label and on whether any wall was aimed at the sun.
How this dating was done: The phases are the archaeological sequence in the tentative listing, from pottery, brick, and the depth of the mounds. Radiocarbon has been used across the civilisation. This room does not replace the listing’s phases with one laboratory number. A cardinal sighting would be archaeoastronomy, and it has not been published here as a measured azimuth.
What could change it: A sealed date for one wall on Mound F, and a surveyed bearing of that wall against equinox sunrise. Or a text, if the script is ever read, that names the building.

Published ranges, drawn as bars. A bar is a claim about a span, not a point.

Harappan phase

Whole sequence at the site

3300 BCE2500 BCE1300 BCE

Harappan phase. The mature city, more than 100 hectares. Not the construction date of one wall.
Whole sequence at the site. From the Ravi/Hakra phase through the late Harappan phase, as the tentative list prints the excavators’ chronology.

What is actually there

More than 100 hectares in the mature phase, at least four walled mounds.
The deposit is over 13 metres deep.
Bricks were taken for railway ballast in the 1850s.
The script is undeciphered, as at Mohenjo-daro.

Open questions

Question	Status
Was Mound F a granary, a hall, or a foundation whose job has been named too soon?	Open
Were any of the walls aimed at equinox sunrise, or only built square?	Open. No azimuth is published in the sources used here.

The current narrative

Where publication stands today. It is the best-supported reading, and it is open to revision. It is not a verdict.

The current narrative — open to revision

A mature Indus city of several walled mounds

Harappa is a major city of the Harappan phase, about 2600 to 1900 BC, with at least four walled centres, baked brick, and the still-unread script shared with Mohenjo-daro. The “granary” is a nickname. Cardinal streets, where they exist, can be set without a recorded intention to track the sun.

What supports this

The tentative list gives the phase, the area over 100 hectares, and the walled mounds.
The same brick, seals, and script run across the Indus cities.
Railway robbing explains missing late levels without requiring a second, unknown city.

What pushes back

The granary name got ahead of the finds, here as at Mohenjo-daro.
A cardinal habit does not tell you whether noon, equinox, or copying set the first line.

What would settle it

It is the centre because the sequence and the sister cities agree. The open labels sit inside it, not in place of it.

Emerging narratives

Other published readings. Each may be right. None is adopted here unless the evidence carries it.

An emerging narrative

A plan set from the sun

The right angles of Indus towns were taken from the sky, either from the noon shadow or from equinox sunrise, so the city itself was a modest observatory. At Harappa the mounds would share that orientation even if a single street plan is harder to see than at Mohenjo-daro.

What supports this

A noon shadow and an equinox sight both produce a north-south or east-west line with no instrument but a post.
Shared orientation across cities is a real pattern in descriptions of the civilisation.

What pushes back

This room’s sources do not print a measured azimuth for a Harappa wall.
Copying a street produces the same corner, and a mound plan is not a grid on a drawing.

What would settle it

It is a way the corners could have been made. It is not a surveyed sightline. It waits on a bearing.

Should the current view be reevaluated?

Consensus holds well

The mature city, the dates, and the kinship with Mohenjo-daro hold. The granary label and any solar aim of the walls stay open, and neither moves the phase.

What drives this

A sequence from about 3300 to 1300 BC, with the mature city at about 2600 to 1900 BC.
Four walled mounds and a depth of over 13 metres.
An unread script shared with Mohenjo-daro.

This is the museum’s own weighing of the evidence. It is not a visitor vote, and it is separate from the slider.

Weigh it yourself

Anonymous. Nothing here is stored except on this browser. Your lean does not change the assessment beside it.

A mature Indus city of several walled mounds Unsure A plan set from the sun

If the contested reading is right

What would follow if a reading that is not the consensus turned out to be true, and what could settle it. This explains the stakes. It does not adopt the reading.

The walls were aimed at equinox sunrise

Indus planning would include a solar sight, and Harappa’s mounds would be checked for the same bearing as the streets at Mohenjo-daro. The script would still be unread.

Why it is still worth a test

A published azimuth of an original wall on Mound F or Mound AB, taken against equinox sunrise.

Sources

Archaeological Site of Harappa UNESCO World Heritage Centre, tentative list
Archaeological Ruins at Moenjodaro UNESCO World Heritage Centre

Image credits

Exposed brick at Harappa. The tentative listing says the structures you can see on mounds AB and F belong to the third millennium BC. . Hassan Nasir, CC BY-SA 3.0. Shown unedited.
Mound F, in the area called the granary and the great hall. Parallel brick walls and platforms. The label is older than any store of grain. . Muhammad Bin Naveed, CC BY-SA 3.0. Shown unedited.
A brick rectangle on the east of Mound F, with the mound rising behind it. Baked brick is the shared material with Mohenjo-daro. . Muhammad Bin Naveed, CC BY-SA 3.0. Shown unedited.

← Pakistan World map