Skip to content

Hindu, Hinduism and Sanātana Dharma

Lesson 2 of 30. Three names, three different histories. Only one of them comes from inside the tradition, and even that one has changed jobs.

10 min read

Depth: Seeker

Three words get used for the same subject, and they do not mean the same thing. Two of them were handed to the tradition from outside. One grew inside it, and even that one is doing a job today that it was not doing two thousand years ago.

This is not a quarrel about vocabulary. Names carry assumptions, and the assumptions travel silently. Ask what Hinduism's book is, and the question already contains a claim about what kind of thing it is. So it is worth knowing where each of the three names came from, who was holding the pen, and what each one was originally for.

A river, and a word for the people beyond it

Start with a river. The Sanskrit name for the great river of the north-west is Sindhu.

Iranian languages went through a sound change in which an initial s became an h. So in Old Persian the river and its region were Hindu, and the province in its lower basin was Hidūš. That province appears by name in the inscriptions of Darius I — Hiduš, listed among his lands on his tomb at Naqsh-e Rustam. The Greeks, borrowing westward again, dropped the h altogether and gave us Indos, and from it India.

Hold on to what that means. The oldest ancestor of the word Hindu is an administrative label in a Persian emperor's list of territories. It says where a place is. It says nothing whatever about what anyone there thought was holy.

And it stayed that kind of word for a very long time — for well over a thousand years in the usage of outsiders. When the Central Asian scholar al-Bīrūnī wrote his great study of the subcontinent around 1030, the title named the country: an inquiry into what is said of al-Hind, India. He was describing the people of a place, at length and with real curiosity, and the word in his title is a geographical one.

There is a second thing to notice, and it is easy to miss. A word like that does not create a group. Being described as "the people beyond the river" gives those people no reason at all to think of themselves as one thing. The label sorts from outside; it does not organise from inside.

When "Hindu" began to mean a religion

At some point the word crossed over. It stopped meaning from there and started meaning of this practice. When?

The honest answer is: gradually, and later than most people assume. In the usage of outsiders it remains largely geographical for most of its life. Recent scholarship tracing the term across its full span finds religious uses appearing only sporadically between about 1400 and 1600 — and finds them appearing, characteristically, in pairs.

That pairing is the key to it. The word turns religious in contexts where somebody is drawing a contrast: Hindu as against Turk, Hindu as against Muslim. In a poem in an early-fifteenth-century Indian vernacular, the Kīrtilatā of Vidyāpati, we get the dharmas of Hindus and of Turks set side by side. In verses attributed to Kabīr from the early sixteenth century, Hindus and Turks are contrasted constantly, and unmistakably as religious communities rather than as populations of places.

That is worth pausing on, because it tells you something about how the category formed. A great deal of what became "Hindu" as a religious identity took shape in the act of being distinguished from something else. Not invented in that act — the practices are far older — but bounded by it. A group that has never been asked to say what it is not has little reason to draw its own edge.

This is genuinely contested ground among historians, and you should know that rather than be handed one side of it. One position emphasises that the religious sense of the word is recent and that agreement about its contours arrives only with colonial rule. Another, argued at length by David Lorenzen, holds that a Hindu religious self-identification is clearly visible in Indian-language sources centuries before the British arrived, and that the pre-colonial evidence has been undercounted. What both sides accept is narrower and firmer than either headline: the word took on its religious sense late, and the practices it came to name did not.

Hinduism: an English word, with an English shape

Hinduism is English, and its shape gives it away.

The suffix -ism is what English does to convert a subject into a system: a body of doctrine, with positions, that a person can be said to hold or reject. It works tolerably for traditions built that way. Applied here it quietly promises a creed, and then the reader goes looking for one.

The word is also recent. The earliest use so far identified in print or letter is Hindooism, in a letter written from Calcutta in 1787 by Charles Grant, an evangelical connected with the East India Company. Rammohun Roy, the Bengali reformer, is among the first Hindus known to have used the English word of his own tradition, around 1816. So the abstract noun is roughly as old as the steam engine — younger than most of the practices it names by several thousand years.

Now, a caution, because this fact gets over-driven. You will hear it said that "Hinduism was invented by the British". That is a slogan, and it smuggles in a claim the evidence does not support. What was made around 1800 was a word, and with it a particular way of packaging a subject: one label, one entry in a survey of world religions, one column in a census. Temples, festivals, philosophical schools, sacred rivers, household rites and centuries of argument were not made in 1800. Neither, as we have just seen, was every trace of Hindus thinking of themselves as Hindus.

The accurate statement is narrower and more interesting: an old and vastly various set of traditions acquired a new English name, and the name brought expectations from elsewhere that the traditions had never undertaken to meet.

: old words doing a new job

The third name is the only one that is both Sanskrit and internal, and it is where the most care is needed — precisely because it is genuinely ancient.

Sanātana means eternal, perpetual, from of old. is one of the largest words in the language, and this course gives it a lesson of its own. Put together, the phrase is old, and it appears in revered texts.

But look at what it is doing there. In the older literature, sanātana overwhelmingly works as an adjective attached to something specific. It qualifies the self, or the divine, or an order of things, or a particular rule. It is not functioning as the proper name of a religion.

Take a famous line from the Manusmṛti: speak the truth, speak what is pleasant, do not speak an unpleasant truth, do not speak a pleasant untruth — eṣa dharmaḥ sanātanaḥ, this is the eternal norm. The phrase is doing its ordinary work. It is stamping a rule about speech as timeless. It is not announcing the name of anything.

And here is the fact that settles the matter most cleanly. The same phrase, in the same sense, appears in a Buddhist text. Dhammapada verse 5 closes: hatred is never appeased by hatred; by non-hatred alone is it appeased — esa dhammo sanantano, that is the eternal law. That is Pali, not Sanskrit, and it is scripture of a different tradition altogether. A phrase that both traditions use in the same way, to mean "this principle is timeless", is not the private name of either.

So when did it become a name? In the modern period, and largely in argument. During the late nineteenth and early twentieth centuries, as Hindus debated among themselves what reform should mean, sanātana dharma came into use as a banner for the position that the inherited tradition needed defending rather than remaking. Sanātana Dharma Sabhās were founded; schools and colleges took the name. In Benares, the Central Hindu College put the phrase on the cover of its school text-books, elementary and advanced. A phrase used for millennia to mean "this is timeless" was, in those decades, promoted to the title of the whole.

None of that makes the modern usage improper. Communities name themselves; that is one of the things communities do, and a name taken up from within is a different act from a label applied from outside. Millions of people use sanātana dharma today as the tradition's true name and mean it wholeheartedly. The only claim the history rules out is a narrow one: that this was always the tradition's name and the old texts prove it. The old texts, read as they stand, say something else.

Three names, three jobs

It helps to stop asking which name is correct and ask instead what each one is for.

Hindu began as a term of location and became a term of community. It is the word that draws a boundary, which is why it is the word that shows up in censuses, in law, and in arguments about who is inside.

Hinduism is a word for outsiders and for comparison. It exists to put an enormous subject on a shelf beside other subjects. That is a real use — you cannot teach a comparative course without some such word — but it is a filing word, and filing words flatten.

is a word of self-description and of value. It does not say where anyone lives or how they should be catalogued. It says: what we are holding on to is not a fashion, it is timeless. That is a claim about worth, and it is made from the inside.

Three different jobs. It is not surprising that they do not translate neatly into one another.

Who counts? The question the names cannot settle

If you want to see how unfinished this is, look at what happens when a modern state has to make the word operational.

Indian law needs to know who a Hindu is, because whole bodies of family law turn on it. The Hindu Marriage Act of 1955 answers in an unusual way. It applies to Hindus in any of their forms, and expressly to Buddhists, Jains and Sikhs. Then it adds a residual clause: it also applies to any other person domiciled in India who is not a Muslim, Christian, Parsi or Jew.

Read that again. For a large class of people, the statutory definition of Hindu is arrived at by subtraction — by not being four other things. The Constitution does something structurally similar in one clause about access to religious institutions, where a reference to Hindus is to be read as including Sikhs, Jains and Buddhists. Members of those communities have objected to being included in that way, and their objection is exactly the one this lesson has been circling: a name applied from outside does a different thing from a name chosen.

You do not have to resolve any of that to take the point. Two centuries after the English word was coined, the boundary it draws still has to be patched by hand.

What to carry out of this

Three things.

The names are younger than the thing. Every one of these words postdates by a long way the practices, texts and questions it now covers. Whenever a sentence begins "Hindus have always believed", it is worth asking which of the three words is doing the work, and when that word acquired the meaning the sentence needs.

Two names came from outside and one from inside, and adoption changes things. Hindu is now used by hundreds of millions of people about themselves, whatever its origin as somebody else's geography. A borrowed name that has been taken up and lived in is no longer merely borrowed.

And a habit, which this course will keep returning to: when a claim rests on a word, go and look at what that word was doing in the passage. It is remarkable how often a sentence that sounds like ancient authority turns out to rest on a modern reading of an old adjective.

Sources & context

These notes point at the texts and traditions a lesson draws on. They are references, not proof.

Bhagavadgītā

sanātana at 1.40, 2.24, 4.31, 7.10, 8.20, 11.18, 15.7reference being checked

Epic / classical

The text itself, read directly.

Parmeshwari's readingThe seven Gītā occurrences of sanātana, scanned mechanically across all eighteen chapters and listed verse by verse in the lesson.

Inscription of Darius I at Naqsh-e Rustam (DNa)

DNa §3, list of lands

Achaemenid royal inscription

The text itself, read directly.

Parmeshwari's readingThe lesson's date for the oldest attestation of the word behind Hindu rests on this land-list — an administrative record that says where a place is and nothing about what anyone there believed.

Taḥqīq mā li-l-Hind (al-Bīrūnī's account of India)

title

Eleventh-century Islamicate scholarship

The text itself, read directly.

Parmeshwari's readingEvidence that the word was still geographical in outsider usage around 1030. The claim is about the title, not about the contents.

Hindu: A History

Comparative Studies in Society and History 65:2 (2023), 246–271reference being checked

Modern scholarship on how this developed, rather than on what it means.

Parmeshwari's readingThe late-and-contrastive account of when Hindu turns religious. Given as one position in a disagreement the lesson names on the page.

Who Invented Hinduism?

Comparative Studies in Society and History 41:4 (October 1999), 630–659reference being checked

Modern scholarship arguing for a reading. One voice in a live conversation.

Parmeshwari's readingThe counter-position and its vernacular evidence. The lesson sets the two side by side and does not pick a winner.

Manusmṛti

4.138reference being checked

Dharmaśāstra

The text itself, read directly.

Parmeshwari's readingOne of the two non-Gītā attestations the lesson leans on for what sanātana dharma was doing in older texts.

Dhammapada (Yamakavagga)

verse 5

Theravāda Buddhist canon

A second text, held beside the first because they do not say the same thing.

Parmeshwari's readingHeld beside the Sanskrit attestations precisely because it is Buddhist and uses the same formula in the same sense. That is the whole argument.

Bhagavadgītā

11.18reference being checked

Epic / classical

The text itself, read directly.

Parmeshwari's readingRead closely because it is the verse most often quoted for the modern sense, and the one that shows most clearly what the modern sense adds.

A Brief History of the Eternal Religion: A Genealogy of Sanātana Dharma

International Journal of Hindu Studies 29:1 (2024), 1–32reference being checked

Modern scholarship on how this developed, rather than on what it means.

Parmeshwari's readingThe genealogy of the phrase's promotion to a name, and the negative finding the lesson quotes: no evidence that it was the original Sanskrit name.

Sanātana Dharma: An Elementary Text Book of Hindu Religion and Ethics, and Sanātana Dharma: An Advanced Text Book of Hindu Religion and Ethics

Foreword, pp. vii–viii (elementary)

The text itself, read directly.

Parmeshwari's readingCited for their titles and dates — a phrase of scripture serving as the title of a syllabus. Catalogue records of the volumes, not the volumes read through.

Hindu Marriage Act, 1955 (India), section 2

s. 2(1)(a)–(c)

Indian statute law

The text itself, read directly.

Parmeshwari's readingThe statutory definition of Hindu, quoted for its shape. The lesson uses it to show that the boundary still has to be patched by hand.

Constitution of India, Article 25, Explanation II

Article 25(2)(b), Explanation II

Indian constitutional law

The text itself, read directly.

Parmeshwari's readingThe parallel constitutional inclusion, and the objection to it, reported and not adjudicated.

Hindu Religious Identity with Special Reference to the Origin and Significance of the Term “Hinduism,” c. 1787–1947

in Rethinking Religion in India (Routledge, 2010), 41–55, at 45

Modern scholarship on how this developed, rather than on what it means.

Parmeshwari's readingWhere the 1787 attestation comes from originally. The lesson reaches it through Truschke and says “earliest so far identified” rather than “first ever used”.