What Hinduism Is

Most descriptions of a religion begin with a founder and a book. Somebody stood up and said something; it was written down; what follows is the history of people trying to live by it. That shape is so familiar that we reach for it without noticing, and it works well enough for several of the world's largest traditions.

It does not work here. Hinduism has no founder and no single book that every Hindu must accept. Those two absences are usually presented as difficulties, as though the description had been interrupted before it could get started. They are not difficulties. They are the description. Whatever else this lesson does, it should leave you holding that thought: the thing that makes Hinduism hard to summarise is not a missing piece of information. It is the shape of the object.

Nobody started it, and there is no one book

There is no first day. You cannot point at a year, a place or a person and say that Hinduism began there, in the way you can for traditions that begin with a life.

What we now group under the name is the long outcome of many communities across the Indian subcontinent asking, over several thousand years, how a person ought to live and what is finally real. They asked in different languages, in different centuries, under very different conditions. Some were priests attached to a fire ritual. Some were forest teachers with a handful of students. Some were poets whose songs got carried between villages before anybody wrote them down. Some were philosophers arguing in an idiom so technical that reading them today requires training.

They did not converge. That is the important part, and it is easy to say too quickly. In most accounts of a religion, disagreement is a stage: there is an original teaching, then a period of dispute, then a settlement, and the settlement is what the religion turns out to be. Nothing like that happened here. The different answers were kept. Not tolerated in the weak sense of being allowed to survive at the margins, but preserved, taught, commented on and handed forward by people who considered each other to be inside the same tradition.

A religion with a founder has a centre of gravity. Even where it splits, the pieces define themselves by their relation to the founder, and you can describe them in those terms. Take the founder away and something structural changes. There is no single point that everything is a distance from.

There are revered texts. There are a great many of them, and the next several lessons are about them, so this chapter will not describe their contents.

What matters now is the arrangement. There is no one book that functions the way a single scripture functions elsewhere: the text you are handed, the text you are expected to accept, the text a dispute is settled by. Instead there is a large inherited library whose parts are ranked differently by different communities. A text that sits at the centre of one household's religious life may be barely present in another's, and both households are Hindu without either of them being in error about it.

This is not the same as saying that the texts are unimportant, or that anything can be treated as scripture. Some texts carry enormous authority, and there are long-running arguments about exactly how much and of what kind. It means that authority is distributed rather than concentrated, and that where you stand in the tradition determines which parts of the library you stand closest to.

Ask a Hindu what the holy book is and you may get the Bhagavad Gītā, or the Vedas, or a text in a regional language that a family has read aloud for generations. None of those answers is a mistake. The question assumed a structure that is not there.

What holds it together instead

If not a founder and not a book, then what? The honest answer is that the unity is of a different kind, and describing it takes longer than naming a person would.

Think of it as a family of related ways rather than a single system. Members of a family may share no one feature that all of them have and no outsider has. What they share is a web of overlapping resemblances: this one has the nose, that one has the voice, a third has neither but has the walk. You recognise the family without being able to name the single thing that constitutes it.

Four kinds of thread run through this particular family.

A shared library, differently ordered. Communities that disagree about almost everything still draw on an overlapping set of texts. They rank them differently and read them differently, and the disagreements are often precisely about ranking and reading. But they are arguing over largely the same shelf.

A shared vocabulary, used to opposite ends. A cluster of words recurs across the whole tradition. They are not slogans; they are the terms in which the arguments are conducted. Two schools can use the same word and reach conclusions that contradict each other, which is only possible because both regard the word as worth fighting over. Month 1 spends several lessons on these words, one at a time, for this reason.

A shared ritual grammar. Underneath enormous local variation there are recognisable patterns: offering, hospitality, the treatment of certain things as set apart. Two ceremonies may share almost no specific content and still be legible to each other as the same kind of act.

A habit of arguing inside rather than leaving. This one is easy to miss and may be the strongest thread of all. When people in this tradition disagreed profoundly, the characteristic move was not to depart and found something new. It was to write a commentary. To take the same text the other side was using and show what it actually meant. Whole libraries exist because somebody thought somebody else had misread a sentence, and the argument was carried on in a form that kept both parties inside the same conversation.

That last thread explains something about the first three. The library is shared because the arguers kept using it. The vocabulary is shared because they needed common terms to disagree in. The tradition is not unified despite the arguing. To a real extent it is unified by the arguing.

Belonging without believing

Here is a question that exposes the shape of the thing quickly. What would it take for someone to stop being Hindu?

In a tradition organised around a creed, the answer is straightforward: stop assenting to the creed. Membership is defined by belief, so belief is what you can lose. The machinery for this exists and has names. There are formulas of assent, and there is a word for departing from them.

Look for the equivalent here and you find that the machinery is missing. There is no statement to affirm on entry and therefore nothing to withdraw. A person can hold that the divine is one and without form, or one with many forms, or many, or can be uncertain, or can hold a position so philosophically austere that it is not obvious it commits them to a divinity at all — and every one of those positions has been occupied by people the tradition regarded as unimpeachably inside it.

What organises membership instead is closer to practice, family, community and inheritance than to assent. This is sometimes summarised as a tradition of orthopraxy rather than orthodoxy: right doing rather than right thinking. That is a useful first approximation and it should be handled carefully, because it is easy to over-apply. It is not that beliefs are considered unimportant. Enormous intellectual effort has gone into getting them right, which is what the commentarial literature is. It is that belief has not been the thing membership is checked against.

Notice how much follows from that one difference.

Conversion, for one. A tradition where membership runs through belief can be entered by changing your mind, which makes conversion a well-defined act with a moment attached to it. A tradition where membership runs through practice, family and community has no equally clean equivalent, and questions about conversion into it have been answered differently in different places and periods. Some communities have well-established procedures; others regard the question as not quite well formed. This course is not going to resolve that, and you should be suspicious of any source that resolves it briskly.

Heresy, for another. The concept requires a defined orthodoxy to depart from. Where the arguments are this open, and where the characteristic response to profound disagreement is a commentary rather than an expulsion, the concept has much less to grip. Disputes here were often ferocious. They were not usually about who was inside.

And the census, for a third. Modern states need countable categories, and countable categories need boundaries. Applying an exclusive one-box-per-person religious classification to a tradition without a creed produces well-known difficulties, and a certain amount of what is written about Hindu demography is really about the shape of the form rather than the shape of the population.

The questions that misfire

Certain questions get asked about Hinduism constantly and produce strange answers. It is worth seeing why, because the strangeness is diagnostic.

Is Hinduism monotheistic or polytheistic? The tradition has held both, and several positions in between, at the same time, for a very long time. There are strands in which the divine is one and without form, strands in which the divine is one but takes many forms, strands in which the many are genuinely many, and sophisticated positions that do not sit comfortably in any of those boxes. All of them have been argued for at length, by people who read the same texts. The question presents two options and asks you to choose. The tradition's answer is that this is the wrong number of options.

What do Hindus believe? Depends which Hindus. This sounds evasive and is not. In a tradition organised around a creed, the question has an answer by construction, because the creed is what membership consists of. Here there is no creed to recite and nothing to check membership against, so the question is roughly like asking what Europeans think.

What is the Hindu holy book? Covered above. There are many, ranked differently.

When was Hinduism founded? No answer exists at the level of precision the question expects.

Notice what these have in common. Each imports a structure from elsewhere and then reports a problem when the structure is not found. The reported problem is real, but it is a fact about the question rather than about the tradition. Better questions exist, and they are mostly of the form: who holds this position, and on what basis, and who disagrees.

Every definition that has been tried

The difficulty of defining Hinduism is not something this course has noticed. It is one of the most thoroughly worked-over problems in the modern study of religion, and the pattern of the attempts is instructive.

The obvious approach is to look for an essence: a belief, a practice or a text that all Hindus share and non-Hindus do not. Try it and it comes apart in your hands. Every candidate either excludes people who are unquestionably Hindu or includes people who are not. Acceptance of the Vedas as authoritative is a serious candidate, and it has genuinely been used as a criterion, but it sits awkwardly with communities whose religious life barely touches those texts. Belief in rebirth is shared with traditions nobody would call Hinduism. Every essential definition proposed has run into counterexamples from inside.

The second approach is to give up on a single essence and define by a cluster: a list of characteristics, no one of which is required, where enough of them being present is what makes something a member. This is the family-resemblance idea from earlier in this lesson, and it is where a good deal of careful scholarship has ended up. It fits the evidence considerably better. It also gives up something people wanted, which is a sharp boundary, and it puts the definitional weight on a vague word — enough.

A third approach abandons the search and treats the category itself as the artefact: on this view "Hinduism" is a container assembled comparatively recently, largely by outsiders needing to name what they encountered, and holding together things that never regarded themselves as one thing. There is real evidence for parts of this, and it is a corrective worth taking seriously. Taken to its limit it also fails, because the shared library, the shared vocabulary and the arguments conducted between these communities across centuries are not the invention of a colonial administrator. People were arguing with each other, in shared terms, on the assumption that they were arguing within something, long before anybody supplied the English noun.

Where does that leave you? With a category that is real, has fuzzy edges, and cannot be reduced to a single criterion. That is an uncomfortable answer only if you assumed in advance that every real thing has sharp edges. Most of the interesting ones do not.

You do not need to adjudicate between these positions to take this course. You do need to know that the argument exists, so that when a source hands you a crisp one-line definition, you know you are being told less than the truth rather than more.

There is a further complication underneath the first one, and it is worth one section because it changes how the rest of the month reads.

The word "religion" carries assumptions built up largely in a European context: that religion is one distinguishable area of life, separable from law, medicine, art, kinship and cooking; that it is centrally about belief; that a person has exactly one; and that having one is a choice they made or could remake.

Very little of that maps cleanly onto what this tradition has historically been. What we would sort into "religious" and "not religious" is frequently the same activity here. Rules about eating, marriage, inheritance, when to plant, how to greet, what to do with a body, which day to begin something — these are not a religious layer applied on top of ordinary life. In much of the tradition, the distinction between the two is not one anybody was drawing.

The tradition's own preferred term for itself points at something wider than "religion" as a category, and Lesson 2 is where that term gets examined. What you should take from here is only the warning: when you read "the Hindu religion", the noun is doing more work, and less accurately, than it looks like it is doing.

This is not a reason to abandon the word. It is the word we have, this course uses it, and refusing to use it would be an affectation. It is a reason to hold it loosely.

The disagreement goes all the way in

It would be comfortable to say that Hindus agree on the fundamentals and differ in the details. It would also be false, and it is worth being exact about how false.

Take what is arguably the tradition's central question: the relation between the individual self and ultimate reality. What are you, at bottom, and what is the relation between that and whatever is finally real? On this question, the major schools of Vedānta reach conclusions that are incompatible with one another, and they reach them by reading the same verses. Advaita, Viśiṣṭādvaita and Dvaita are three of those schools. Lesson 11 reads a chapter of the Gītā alongside two of these commentaries so you can watch it happen; the names appear here only as evidence that the disagreement is internal and central, and this chapter will not attempt to explain their positions.

The same pattern repeats elsewhere. Whether ultimate reality is personal, something you can stand in relation to and address, or impersonal: the tradition divides. Whether liberation comes through ritual action or through knowledge: the tradition divides, and the division is ancient.

These are not peripheral matters. They are the middle of the subject. And everyone in the dispute remains Hindu — not by a technicality, but because that is genuinely how the tradition works. The disagreement is not a failure of the tradition to reach consensus. It is the tradition.

There is a bad way to hear everything above. It sounds like: nothing is fixed, everybody makes it up, any claim is as good as any other. That reading is wrong, and it is worth blocking directly, because it is the most common mistake people make on first contact.

The arguments in this tradition are old, careful and extremely specific. The commentarial literature is technical to a degree that surprises people who arrive expecting something soft. Positions are stated precisely, objections are raised in the strongest form the objector can manage, and answers are expected to meet them. Reading a serious commentary is closer to reading a legal argument than to reading a devotional pamphlet.

What is open is not the standard of argument. What is open is the conclusion. Several conclusions have been reached, defended rigorously, and preserved by people who found the other conclusions unpersuasive but recognisably serious.

So the plurality here is not the absence of rigour. It is a great deal of rigour that has not resolved into one answer, over a very long time, and a tradition that decided it could live with that.

Two true descriptions, and a case worth sitting with

There is a further thing to get straight early, because it will otherwise feel like inconsistency later in the course.

Ask a practising Hindu what the tradition is and you may hear that it is sanātana — enduring, without beginning, not the invention of anyone. Ask a historian and you will hear about datable texts, traceable movements, documented change and the comparatively recent history of the word itself. These sound like rival claims. Read carefully, they are usually answers to different questions.

The first is a claim about what the tradition takes itself to be and where it locates its authority: not in a founder's biography, and not in the moment a community happened to form. The second is a claim about the historical record of practices, texts and institutions. You can hold both without contradiction, and most careful writers do. Trouble starts when one is presented as debunking the other — when a historical account is offered as proof that a self-understanding is false, or when a self-understanding is offered as grounds for refusing a historical account.

This course does both, deliberately, and will try to be clear about which it is doing. When a lesson says a text was composed over a long period and reached its present form at a particular stage, that is the historian speaking. When it says that a tradition regards that text as without human authorship, that is a report of the tradition's own view, and it is reported because you cannot understand what people do with the text without it.

Neither register is the "real" one. A description of a religion that contains only what a historian can verify has left out most of what the thing is for the people who live it. A description that contains only the internal view is not a description at all; it is a restatement.

You are going to have to hold two things at once fairly often here. That is not a flaw in the presentation. It is what it takes to describe something from outside and inside at the same time.

Abstract arguments about definition become clearer against a concrete difficulty, so here is one that has no tidy answer.

Consider a family whose religious life consists almost entirely of practices: a lamp lit at a particular hour, certain days marked, certain foods prepared or avoided, a shrine in one corner of the house with a small number of images in it, a story or two told to children. Nobody in the family has read the Vedas. If pressed on the metaphysical questions this lesson has mentioned, they would not have positions on them and would find the questions faintly odd. They would say without hesitation that they are Hindu, and everyone around them would agree.

Now apply the definitions.

The essentialist criterion of accepting the Vedas as authoritative gives an awkward result. The family has never engaged with those texts. If asked whether they are authoritative they would probably say yes, in roughly the way one assents to a respected institution one has no direct dealings with. Does that count?

The family-resemblance approach handles the case much better. Enough of the threads are present — the ritual grammar, the inherited practice, the shared vocabulary in the story told to the children, the community — and the fact that no single criterion is decisive stops being a problem.

And the third approach, the one that treats the category as an outsiders' construction, has the least to say about this family, because they are not a construction. They are doing what their grandparents did.

Notice what the exercise actually demonstrates. The family is not a borderline case that a better definition would resolve. They are utterly typical, and they are the majority. Any definition that makes ordinary practising households hard to classify has told you something about the definition. This is why the tradition's centre of gravity is not where a book-based account puts it: the arguments in this course are real and consequential, and most Hindus, most of the time, are not having them. Both facts belong in an honest account.

Where all of this actually lives

The last few sections have been abstract. The tradition, in practice, is not.

Consider what any of this looks like on an ordinary evening. One household lights a lamp at dusk, at the same hour it has been lit for as long as anybody remembers, and nobody present could tell you which text prescribes it. Another sings. In another, someone sits still for twenty minutes before the day ends. In another, food is cooked and offered before it is eaten. Somewhere a family is walking to a river. Somewhere a grandmother is telling a child a story that a scholar would recognise as a version of an episode from a very old text, in a form that scholar has never seen, because it has passed through four generations of that family and been shaped by all of them.

All of these people are Hindu. None of them is doing a reduced or popular version of something that exists in a purer form in a book. This is the tradition, in the form it actually takes for most of the people who live it, and it has always taken that form. The texts and the households have been shaping each other continuously for thousands of years.

Keep that in view through the textual lessons ahead. It is easy, when reading about a religion, to end up with a picture in which the books are the reality and practice is a shadow of them. The reverse is at least as true.

A last correction before the closing, because it affects what "the tradition" means every time this course says it.

Hinduism has never been geographically uniform. What is practised in Tamil Nadu and what is practised in Assam differ in language, in text, in deity, in festival and in ritual detail, and both have been continuously Hindu for as long as the question is meaningful. Regional traditions are not local flavourings of a standard product. Some of them are older than the things that look standard.

Nor has it been static. A tradition that ran for several thousand years without a central authority was never going to be. Practices that feel timeless from inside a family often turn out to have datable histories. Others really are extremely old. Working out which is which is the historian's job and it is frequently difficult, so this course will say when something is ancient and when it is more recent, and will say when it does not know.

And it is not confined to one country. Communities carried it outward over centuries, by trade, by migration and by force, and there are places where it has been established for many generations. A reader in one of those places is not at the edge of the tradition looking in.

The practical consequence for the next twenty-nine lessons: when a lesson says "the tradition holds", you are entitled to ask which part, when, and where. Where the answer is genuinely general, the lesson will say so. Where it is not, it will name the part.

One caution, and then the next lesson takes this up properly.

"Hinduism" is an English word with an English suffix, and its history as the name of a religion is comparatively recent. That is a fact about the label. It is not a fact about the practices, texts and arguments the label points at, which are very much older. People sometimes take the recency of the word as evidence that the thing is a modern invention. The word and the thing have different ages, and confusing them produces bad history in both directions.

Lesson 2 is about the names: where they came from, who applied them to whom, and which one the tradition uses of itself.

How this course will speak

Two consequences follow from everything above, and they govern how the next twenty-nine lessons are written.

Positions get attributed. Whenever a lesson states a position, it will say whose position it is. "Hinduism teaches" is almost always the wrong sentence. "The Advaita reading holds", or "the Bhāgavata says", is usually the right one. Where one school's theology is presented as the tradition's, something has quietly gone missing, and what has gone missing is usually the people who disagreed.

Plurality does not get smoothed. Where the sources genuinely differ, you will be told that they differ. Where they can be reconciled, you will be told who does the reconciling and how. This is slower than a summary. It is also the only version that survives contact with the texts, and you will be able to tell the difference the first time you read a primary source and find it does not say what a summary told you it said.

If someone asks you what Hinduism is, you are not obliged to produce a definition. You can say what kind of thing it is.

It is what happened when a very large number of people, over a very long time, in one part of the world, asked the largest questions available and were not required to agree. What holds it together is a shared library, a shared vocabulary, a shared sense of what a religious act looks like, and a habit of staying in the argument. It has no founder, no creed and no single book, and those absences are not gaps in the account. They are the account.

The rest of this month walks into that conversation in order: what it is called, how it is passed on, what it argues about, and what it looks like in a house on an ordinary evening.