Indus Valley

A god that may not be a god

Sometime around 1928, in the ruins of Mohenjo-daro on the plain of the Indus, excavators turned up a small steatite seal, barely an inch and a half square. It shows a figure seated with legs folded beneath it, heels together, in what looks unmistakably like a yogic posture. The figure wears a tall horned headdress and is surrounded by animals — a tiger, an elephant, a rhinoceros, a water buffalo, and, below the seat, two deer. The great archaeologist Sir John Marshall, who was directing the excavation of the newly discovered Indus civilization, looked at this seal and saw something breathtaking: a proto-Shiva. Here, he argued, three thousand years before the Hindu god was worshipped by name, was Shiva in an early form — Pashupati, “Lord of Animals,” the horned yogi, master of the beasts, perhaps even three-faced.

It is one of the most famous objects in the archaeology of religion, and nearly a century later scholars are still arguing about almost every word of Marshall’s description. Is the figure male? Is it a god at all, or a chieftain, a shaman, a mythic hero? Are those three faces, or an artifact of a wide horned mask? Is it seated in yoga, or simply in the cross-legged way people sat? Is it Shiva’s ancestor, or a version of the “Master (or Mistress) of Animals” motif found from Mesopotamia to the Aegean in the same era? We cannot settle it — and the reason we cannot is the single most important fact about Indus religion, the fact that governs this entire chapter: their writing has never been read.

This is a civilization we can see in extraordinary material detail and cannot hear in its own voice at all. What follows is therefore a chapter with more honest uncertainty in it than any so far, and that is exactly as it should be.

The silent civilization

The scale is easy to underestimate. The Indus (or Harappan) civilization of its mature phase, roughly 2600 to 1900 BCE, was the largest of the Bronze Age — larger in area than Egypt and Mesopotamia combined, spread across more than a thousand known sites in what is now Pakistan and northwest India. Its great cities — Harappa, Mohenjo-daro, Dholavira, Rakhigarhi, Lothal — were laid out on grids, built of standardized kiln-fired brick, and served by the most advanced water and drainage engineering anywhere in the ancient world: covered sewers, private bathrooms, wells, and public reservoirs. Their system of weights was so precisely standardized that it hardly varied across the entire territory for centuries.

And then the strangeness. For a civilization this large and sophisticated, the Indus cities are almost eerily quiet about the things other Bronze Age powers shouted. There are no securely identified temples. No great palaces. No royal tombs stuffed with gold like Ur’s, no pyramids, no monumental inscriptions boasting of kings, no reliefs of rulers smiting enemies, no obvious images of war at all. Where Egypt and Mesopotamia broadcast a hierarchy of god-kings and priesthoods in stone, the Indus is, by comparison, faceless — egalitarian in appearance, decentralized, and reticent. Whatever their religion was, it did not express itself the way their contemporaries’ did, and that absence is itself one of the most intriguing pieces of evidence we have.

The wall we cannot read

Everything would be different if we could read the Indus script. We have it in abundance — some four thousand inscribed objects, most of them the little seals and their impressions — carrying a sign-system of roughly four hundred distinct characters. And it is, after a century of effort and dozens of confident “decipherments,” still completely undeciphered. No bilingual key like the Rosetta Stone has ever been found; we do not know what language, if any, lies behind the signs.

The longest single inscription only sharpens the frustration. At Dholavira, in Gujarat, excavators found the remains of a great “signboard” that had stood over the north gate of the citadel: ten large Indus signs, each about 37 centimeters high, once inlaid in white gypsum on a wooden board perhaps three meters long, which fell and rotted but left its letters legible in the ground. It is the closest thing the Indus world has to a monumental public text — a sign meant to be read by everyone entering the city — and we cannot read a character of it.

There is not even agreement that the signs encode language at all. The dominant view, argued by scholars such as Asko Parpola and Iravatham Mahadevan, holds that the script is true writing, most likely recording an early Dravidian language. But in a provocative 2004 paper, Steve Farmer, Richard Sproat, and Michael Witzel argued the opposite: that the inscriptions are too short (averaging around five signs), contain too many rare signs used only once, and lack the repetition real languages produce — and that the “script” may therefore be a system of non-linguistic emblems (for clans, deities, or goods) rather than writing. Parpola and others have pushed back hard, and computational studies have gone both ways. The debate is unresolved. Until it is — if it ever is — the Indus people keep their theology to themselves, and every claim about their gods is an inference from things, not texts.

What the objects suggest — held loosely

With that warning fixed firmly in place, here is what the material remains have been taken to suggest. Every item on this list is an interpretation, and the honest ones are offered as such.

The seated horned figure. Marshall’s “proto-Shiva” (the seal is catalogued as M-420) remains the flashpoint. The identification with Shiva rests on later features read backward: Shiva as Pashupati (lord of animals), as the archetypal yogi, as horned/bull-associated. Critics — among them Doris Srinivasan — have argued the figure may be a divine bull-man or buffalo deity, that the “three faces” are questionable, and that the whole “Master of Animals” composition belongs to a widespread Bronze Age motif rather than to a specifically Indian god. The scholar Shubhangana Atre has proposed the central deity of the Indus was in fact a goddess, a “Lady of Beasts.” No reading commands consensus.

The “Priest-King.” A small steatite bust from Mohenjo-daro, bearded and serene, its robe carved with a trefoil pattern once filled with red pigment, is among the most reproduced of all Indus objects — and its very name is a lesson in reading later categories backward. Marshall and his contemporaries called it the “Priest-King” on the analogy of the priest-kings of Mesopotamia, but the Indus has produced no textual, architectural, or funerary evidence of either priests or kings. We do not know whether the bust portrays a ruler, a deity, an ancestor, or an abstraction; the confident title is a modern projection onto a civilization that left no rulers’ names at all.

A goddess, or goddesses. Excavators have recovered large numbers of terracotta female figurines, some with elaborate headdresses, some bearing soot marks that may mean lamps or offerings were burned before them. These are widely read as evidence of a fertility or mother-goddess cult — a plausible and popular interpretation. But figurines are slippery: without texts we cannot be sure which were cult objects, which were votive, and which were toys, and recent scholarship urges caution against assuming a single unified goddess.

Water and purity. The most striking single structure is the Great Bath of Mohenjo-daro, a large, watertight sunken tank of fitted brick sealed with bitumen, reached by two staircases. It looks purpose-built for communal immersion, and it is very often read as a place of ritual bathing — a possible ancestor of the ritual-purity concerns that run through later Indian religion. That reading is attractive and unprovable; the tank could have served other public functions.

Trees, animals, and stones. Seals show a horned figure standing in a pipal (sacred fig) tree, and the pipal remains sacred in later Indian tradition; a deity or spirit in a tree is a reasonable inference. Cattle imagery is everywhere, especially the enigmatic single-horned “unicorn” bull that dominates the seals, hinting at a special status for the bull. Cones and ring-stones found at the sites have been interpreted — again first by Marshall — as lingam and yoni forms prefiguring Shaiva worship, though many archaeologists read them as ordinary architectural or utilitarian objects. At Kalibangan, rows of pits with ash and terracotta stelae have been claimed as fire altars suggesting a fire-ritual, but whether these are ritual or simply hearths is disputed.

The quiet dead. Harappan burials, where found, are generally modest: extended inhumations with some pottery and personal ornaments, and no lavish royal tombs. Compared with the death-pits of Ur or the tombs of Egypt, the Indus dead were sent off with striking restraint — which may say something real about how little this culture invested in an elaborate afterlife, or may simply reflect burial customs we have not fully recovered.

The believer's lens

Here, uniquely in this book, the believer's lens must be left almost blank — and honestly so. The Indus Valley people left a great civilization of planned cities, standardized weights, ritual bathing-tanks, and a script we cannot read, but not one word of what they believed. We cannot say what the "Priest-King" bust meant to them, whether the horned figure on the seals was a god, what the Great Bath at Mohenjo-daro was for, or whether they had gods at all in a sense we would recognize. To write a confident "believer's lens" for the Indus would be to invent one — which is exactly what a century of enthusiasts have done, reading back Hindu gods, yoga, and Vedic religion onto a silent people. The most faithful thing this book can say from the believer's side is that the Indus clearly had a rich ritual life — the bathing tanks, the fire-altars some claim, the careful figurines — and that its content is, for now, sealed behind an undeciphered script.

The skeptical lens

The skeptical lens here is really the whole discipline of the chapter: read the categories forward, not backward. Every claim that the Indus practiced proto-Hinduism, worshipped a "proto-Shiva," or invented yoga rests on interpreting mute images through the lens of religions that arose a thousand or more years later — a textbook case of reading the known onto the unknown. The famous seals, the "Priest-King," the Great Bath, and the "signboard" of Dholavira are genuine and fascinating, but their religious meaning is inferred, not established, and the Indus script remains undeciphered despite many confident "decipherments," none accepted. Even whether the script encodes a language at all has been questioned. The honest position is severe restraint: we have the material culture of Indus ritual in abundance and its meaning almost not at all, and the chapter's discipline is to resist the powerful temptation to fill that silence with a borrowed theology.

Symbology and sacred encoding

For a civilization defined by an unread script, the theme of sacred encoding is not a side-topic — it is the whole predicament in miniature.

The great cipher. The Indus script is the largest undeciphered corpus of the ancient world, and it sits mostly on seals — small carved stamps whose practical job was almost certainly administrative, pressed into clay to mark goods and ownership. Yet the same seals carry the civilization’s richest religious imagery: the horned figure, the tree spirit, composite animals, ritual scenes. So the encoded and the sacred are literally combined on a single object, and both halves resist us — we can neither read the signs nor confirm the meaning of the pictures. It is the purest case in this whole archive of a symbol system that is demonstrably there, plainly systematic, and still locked.

The unicorn and the altar-object. The most common seal image is the so-called unicorn — a bull-like animal in profile shown with a single horn — standing before a mysterious object usually called the “standard” or “manger,” possibly a ritual filter or incense-burner. What it meant is unknown; that it meant something, repeated on thousands of seals, is hard to doubt.

The swastika. Among Indus motifs is the swastika, an equilateral hooked cross that was, for thousands of years across Eurasia, a benign symbol of auspicious fortune — and remained sacred in Hindu, Jain, and Buddhist use long before its twentieth-century appropriation by Nazism inverted its associations in the West. Its presence here is part of the honest, and uncomfortable, long history of the sign.

The order in the weights. There is one Indus “code” we can read, and it is not religious: the standardized system of cubical stone weights, in a regular binary-then-decimal ratio, uniform across the whole territory. It is a reminder that not every encoded system is sacred — and a caution that, precisely because we cannot read the script, the temptation to pour mystical meaning into Indus symbols must be resisted. The safest honest statement is that the Indus people built elaborate symbolic and administrative systems whose religious content we can see the shape of and cannot yet read.

Connections

The Indus was not isolated. Sumerian texts speak of trade with a distant land called Meluhha, widely identified with the Indus region, and the archaeology confirms the link: Indus-style seals and distinctive etched carnelian beads have been found at Ur and elsewhere in Mesopotamia, and a class of “Persian Gulf” seals marks the trade route between them. The Indus and the Mesopotamia of Chapter 3 were contemporaries and commercial partners — two utterly different religious worlds, one loud with gods and kings, the other silent, exchanging goods across the Arabian Sea.

The forward connection is more fraught, and it is where this chapter hands off to the next. Around 1900 BCE the Indus cities declined — not by a single catastrophe but through a drawn-out combination of climate change, shifting and drying rivers, and the fading of the trade and urban systems that held them together. In the centuries after, the archaeological and linguistic record of northwest India is increasingly that of the Vedic culture and its Indo-Aryan language. How the two relate is one of the most contested questions in the field — and one heavily charged with modern political and national identity. Recent ancient-DNA work, including a genome from a Harappan individual at Rakhigarhi, indicates that the steppe-pastoralist ancestry associated with Indo-European speakers was absent in the Harappan population and arrived in South Asia largely after the Indus decline, around 2000–1500 BCE — as a substantial admixture layered onto the large existing population, not a wholesale replacement. Whether, and how much, Indus religion survived that transition to shape later Hinduism — the yogi, the goddess, the sacred bull and pipal, ritual bathing — is the great open question the next chapter, on the early Vedic world, must take up.

Walk this chapter’s room in the Museum (beta) →

From The Divine Archives · https://getconexto.com/chapters/ch04.html · a comparative library of the sacred.