IVC Religion & Script: Pashupati, Mother Goddess, and the Undeciphered Seal

Overview: What We Know and Don't Know

The religious life of the Harappans presents a paradox: a civilisation that left behind thousands of seals, hundreds of figurines, and elaborate drainage systems has told us almost nothing about its gods, priests, or rituals in any language we can read.

Unlike Mesopotamia or Egypt, which left extensive written religious texts, the IVC left behind a script that remains undeciphered. Everything we know about Harappan religion is inferred from material evidence — terracotta figurines, soapstone seals, architectural remains, and burial sites. As covered in the article on IVC discovery and extent, the civilisation spans over a million square kilometres and nearly seven hundred years of mature occupation, which means any religious practices identified are generalisations across a vast, likely heterogeneous culture.

What the archaeological evidence does suggest is: the Harappans venerated a mother/fertility goddess figure, may have worshipped a proto-Shiva deity, conducted fire rituals, revered certain trees and animals, and practised burial of the dead with grave goods. There is no confirmed temple, no identifiable priesthood, and no royal tomb — all of which distinguishes the IVC sharply from its Bronze Age contemporaries.

Mother Goddess and Fertility Figurines

The most numerous religious artefacts in the IVC are terracotta female figurines, found in large quantities at almost every Harappan site. These figurines are typically standing, wear elaborate headdresses and jewellery (necklaces, bangles, girdles), and are sometimes depicted with a child or a cup in hand. Most scholars interpret them as Mother Goddess or fertility deity representations — a female principle associated with creation, fecundity, and the earth.

The figurines vary considerably in style and sophistication across sites, which may indicate regional variations in worship rather than a single uniform cult. At Mohenjo-daro, the figurines are relatively elaborate. At smaller peripheral sites, they can be more schematic. Some figurines show a smoke-blackened cup near the head, interpreted as evidence of a lamp or incense being burned in front of them — suggesting active ritual use rather than mere decorative function.

Male figurines are also found, though far less frequently. Some are shown wearing a trefoil or pipal-leaf pattern robe (the so-called "priest-king" figure from Mohenjo-daro — a stone sculpture of a bearded man wearing a robe decorated with trefoils — is sometimes interpreted as a priest or ruler figure). The Priest-King sculpture from Mohenjo-daro (National Museum, New Delhi) is a notable example: it shows a bearded man with lowered eyes, typically interpreted as showing a meditative or ritual pose. While "priest-king" is the conventional label, it remains speculative.

The Pashupati Seal: Proto-Shiva Debate

Among the 3,500+ seals found in the IVC, none is more famous or more debated than the Pashupati seal — a small soapstone seal (3.56 × 3.53 cm) found at Mohenjo-daro, first published by John Marshall in 1931. The seal depicts a central figure seated in a yogic posture on a low throne, with feet touching heel-to-heel (a posture resembling the yoga asana called mulabandhasana), wearing a horned headdress. Around the figure are four animals: an elephant and tiger on the left, a rhinoceros and buffalo on the right, and two deer below the seat. The seal also bears an inscription in Indus script above the figure, which has not been conclusively read.

Marshall interpreted the figure as a prototype of the later Hindu deity Shiva, for several reasons: (1) the figure is seated in a yogic posture — Shiva is the Mahayogi; (2) it has three faces (or appears to), echoing Shiva's later three-faced (Trimurti) iconography; (3) it is surrounded by animals, matching Shiva's title Pashupati (lord of animals); (4) the horned headdress resembles the Shiva-crescent moon motif; (5) the figure may be ithyphallic, relating to Shiva's Linga worship. This identification has profoundly influenced Indian religious history — if correct, it pushes Shiva worship back to at least 2500 BCE.

However, the identification is contested. Some scholars (notably Alf Hiltebeitel, Wendy Doniger, and others) argue: (a) the "three faces" may be artistic perspective, not a three-faced deity; (b) the Pashupati epithet appears only in later Vedic texts (Rigveda, Atharvaveda) and the link is retrospective; (c) the figure could be a female goddess, a tree spirit, a buffalo-demon, or a water deity — the evidence does not uniquely support Shiva. The debate remains unresolved. For UPSC purposes: Pashupati seal = seated figure, horned headdress, surrounded by animals, found at Mohenjo-daro, interpreted as proto-Shiva.

For the broader context of IVC material culture — the economy that sustained these religious activities, including the craft production of soapstone seals themselves — see the companion article on IVC economy and trade.

Fire Altars at Kalibangan and Lothal

The most direct evidence of ritual fire use in the IVC comes from Kalibangan (Rajasthan), excavated by B.B. Lal and B.K. Thapar from 1961. On the citadel mound at Kalibangan, excavators found a row of seven fire altars — rectangular brick platforms with a central pit containing ash, charred animal bones, and terracotta cakes. The fire altars are aligned in a row, suggesting they were used simultaneously or in sequence for ritual purposes. A pit containing animal bones (possibly sacrificial) was found nearby.

Lothal (Gujarat) also has a fire altar. At both sites, the fire altars are located on raised civic platforms (citadels) rather than in private houses, suggesting they served a communal or state ritual function. This is significant: it is the closest thing to a public religious institution (comparable to a temple) that has been found in the IVC.

Some scholars have drawn a line from these Harappan fire altars to the later Vedic yajna tradition — the sacrificial fire ritual that is central to Rigvedic religion. This would suggest a cultural continuity between IVC and Vedic civilisation, a contentious claim in the larger debate about IVC identity. The more conservative scholarly position is that fire rituals were common across many ancient cultures, and the resemblance to Vedic yajna is not proof of direct lineage. Nevertheless, Kalibangan fire altars remain a key UPSC data point.

Tree, Animal, and Nature Worship

Harappan seals provide rich iconographic evidence for the veneration of trees and animals. The pipal tree (Ficus religiosa — the sacred fig, also known as the Bodhi tree in Buddhism) appears on multiple seals with a worshipper or figure standing before it, sometimes with a kneeling supplicant and a deity emerging from within the tree. This is among the earliest evidence of pipal worship in the subcontinent — a practice that continues in contemporary South Asian religious tradition.

The unicorn — a one-horned animal resembling a bull, depicted in profile with a standard object in front of it — is the single most common motif on Harappan seals, appearing on roughly 60% of all inscribed seals. The standard (a device in front of the animal that may be a filter, incense burner, or offering stand) is always present with the unicorn, suggesting it was not merely a decorative motif but part of a standardised ritual iconography. The underlying animal may be a stylised representation of a one-horned rhinoceros, a bull seen in profile, or a purely mythical creature.

Other animals with clear iconographic prominence include the humped bull (zebu/Bos indicus), the elephant, the tiger, the rhinoceros, the crocodile, and the cobra. The serpent motif is associated with water and fertility in many later South Asian traditions. The composite animal (part tiger, part elephant, part bull) appears on some seals, suggesting a mythological tradition of hybrid beings — common in Bronze Age religious iconography across West and South Asia.

Burial Practices and the Afterlife

Harappan burial practices offer indirect evidence about beliefs regarding the afterlife. The dominant mode is extended inhumation — the body laid out flat on its back in a rectangular pit, with the head oriented generally northward. The grave was often lined with mud bricks. Grave goods — pottery vessels (possibly containing food and water), ornaments (beads, bangles, necklaces), copper mirrors, and occasionally tools — were placed in the grave, strongly suggesting a belief in some form of existence after death that required material provisions.

Key burial sites include: Cemetery R37 at Harappa (over 200 graves excavated, richest mortuary assemblage), and cemeteries at Mohenjo-daro and Lothal. At Lothal, a remarkable double burial was found — two skeletons in the same grave, one male and one female, interpreted as possible simultaneous interment of spouses. At Kalibangan, the H-cemetery contains unusual oval graves and some fractional burials — secondary burials where the bones were placed in a pot or container after decomposition of the body elsewhere, suggesting a multi-stage mortuary ritual.

Crucially, there are no monumental royal tombs in the IVC — no pyramids, no elaborate death-pits like those at Ur (Mesopotamia), no royal hypogeum. This is consistent with the broader inference that the IVC did not have a divine monarchy or a god-king tradition. Grave goods show some differentiation by wealth (some graves have more ornaments than others) but the differences are modest, not the massive inequality visible in Egyptian or Mesopotamian burials.

This mortuary evidence ties directly into the broader picture of IVC town planning discussed in the article on IVC town planning and urban features — the absence of a palace or temple confirms that the IVC had a distinct socio-political structure from other Bronze Age civilisations.

The Indus Script: 4,000 Inscriptions, Zero Translations

The Indus script is the most consequential unsolved problem in the study of the IVC. Found on approximately 4,000 inscribed objects — primarily soapstone seals, copper tablets, terracotta sealings, and pottery — the script consists of roughly 400–600 distinct signs (most estimates converge on about 400 core signs, with variations across sites). The signs are small, abstract, and range from pictographic (recognisable as animals, plants, geometric forms) to more abstract linear marks.

Several key characteristics of the Indus script are well-established. The script was written right to left in most cases — this is inferred from the observation that signs are crowded on the right side of longer inscriptions (suggesting the scribe ran out of room after starting from the right). Some inscriptions show boustrophedon writing — alternating right-to-left and left-to-right lines. The texts are remarkably short: the average inscription is only 5 signs long. The longest known Indus inscription is about 26 signs. This brevity is the most significant obstacle to decipherment — there is simply not enough text to work with statistically.

No bilingual inscription — the Indus equivalent of the Rosetta Stone (which allowed decipherment of Egyptian hieroglyphs by providing the same text in Greek, Demotic, and hieroglyphic) — has ever been found. The nearest thing to a bilingual text is a handful of seals found at Mesopotamian sites that appear to have both Indus and cuneiform script; but these have not yielded a translation.

Theories of Decipherment

The most systematic effort at decipherment has been led by the Finnish scholar Asko Parpola, who has worked on the problem since the late 1960s. Parpola argues that the script represents an early Dravidian language — specifically Proto-Dravidian — and has proposed readings for many signs using a "rebus principle" (where a sign for one word is used to represent a phonetically similar word). His most famous proposed reading is that the fish sign (one of the most common Indus signs) represents the Proto-Dravidian word min (fish), which is homophonous with the word for "star" in Tamil — leading to the hypothesis that fish signs on seals may represent stellar or divine names.

A competing and controversial position has been advanced by Steve Farmer, Richard Sproat, and Michael Witzel (2004), who argued that the Indus script is not a true writing system — i.e., it does not encode language — but is rather a system of political, religious, or clan symbols comparable to heraldry. Their arguments include: the extreme brevity of texts, the apparent lack of long texts, the high sign frequency of a few signs (consistent with logo systems rather than phonetic scripts), and the absence of the script from private or domestic contexts. This view has been sharply contested by Parpola and others, and most scholars in the field continue to treat the Indus script as a genuine undeciphered writing system.

A third group of researchers (various Indian scholars) has proposed a Sanskrit/Vedic connection — suggesting the Indus script encodes an early Indo-Aryan language related to Sanskrit. This would support the idea that the IVC was the cradle of Vedic civilisation. However, this view has minimal support in mainstream academia and is generally considered to be driven more by nationalist historiography than by linguistic evidence.

Quick-Recall Table: IVC Religion & Script

The religion and script of the IVC tie directly into the question of the civilisation's identity and legacy — a debate that is examined in detail, alongside the major excavated sites, in the companion article on important IVC sites including Lothal, Kalibangan, Dholavira, and Rakhigarhi.