The Upaniṣads

Four things are usually said about the Upaniṣads, and three of them need correcting before the fourth can do any work.

They are said to be the last section of each Veda; to number 108; to mean "sitting near a teacher"; and to have turned away from ritual towards philosophy. The corrections are not pedantry. Each one changes what you are looking at.

What the word actually means

Take the name first, because the usual gloss is the most quoted thing about these texts and it is the least defensible.

Upaniṣad does break down as upa (near) + ni (down) + sad (sit), and from that people get "sitting down near a teacher". It is charming, and it fits the picture of pupils at a master's feet.

Now watch the word work inside its own texts.

In the Bṛhadāraṇyaka, after describing how everything issues from the self, the text says: tasyopaniṣat satyasya satyam — "its upaniṣad is 'the real of the real'." Later it gives a syllable and says tasyopaniṣad ahar iti, "its upaniṣad is 'ahar'." In the Chāndogya, a passage on chanting closes: "one who knows this upaniṣad of the chants". And in the Taittirīya Āraṇyaka — the passage this course quoted back in Lesson 03 — a teacher finishes his instruction with eṣā vedopaniṣat, "this is the upaniṣad of the Veda."

In none of these is anyone sitting anywhere. In every one, an upaniṣad is a hidden connection: a secret equivalence between one thing and another, and by extension the secret teaching in which such equivalences are given.

That is also the scholarly conclusion. Patrick Olivelle, whose annotated translation is the standard scholarly edition in English, records that the earliest usage shows upaniṣad meaning "connection" or "equivalence", that such connections are assumed to be hidden, and that the older derivation from "sitting near" a teacher is — his words — clearly untenable.

Keep the picture of the teacher and student if it helps you remember. Just do not call it the etymology.

Where they sit

Lesson 05 dismantled the four-storey diagram in which an Upaniṣad is the top floor of each Veda. The Upaniṣads are the reason that diagram feels so convincing, and the reason it is wrong.

They are not appendices bolted to the end of the Vedas. They grew inside particular Vedic schools, embedded in the ritual literature of those schools. You have already seen one proof: the Taittirīya Upaniṣad is a section of the Taittirīya Āraṇyaka, praśna seven. The Bṛhadāraṇyaka belongs to the White Yajurveda tradition and its Brāhmaṇa; the Chāndogya to a Sāmavedic one; the Aitareya to a Ṛgvedic one. Each sits in a lineage, with a Saṃhitā and a Brāhmaṇa behind it.

And here is the fact that explains why almost everyone pictures them as free-standing books. They were detached. In the first half of the second millennium CE — that is, well over a thousand years after they were composed — the early Upaniṣads were separated from the Brāhmaṇas they were part of and gathered into collections of their own.

So the "Upaniṣads" as a shelf of small books is a medieval editorial achievement. As texts they are Vedic school literature.

Why there is more than one

Because there was more than one school. Different lineages produced different teaching, and what survives is what those particular chains carried — which is the third lesson's point arriving in a new place.

Which brings us to the number.

You will be told there are 108. There is a traditional list of that length, and it has been enormously influential. But texts titled "Upaniṣad" went on being composed for a very long time, some of them into the sixteenth century, and they number in the hundreds. When collections were finally made, the count depended on where you were: 52 was a common number in the north and 108 in the south.

So 108 is a regional convention for a late collection, not the size of an ancient canon. When someone speaks of the "principal" or "major" Upaniṣads — usually a dozen or so — that is a modern and commentarial selection of the oldest ones, and it is worth asking whose selection it is.

How they teach

Dialogue, mostly, and the dialogue is doing work rather than decorating an argument. But it is not the only method, and the range is the interesting part.

A bare question. The Kena opens with one and gives no answer for some time: by whose will does the mind fly to its object? by whose command does the breath do its work? who is the god that yokes the eye and the ear? The text begins by making you feel that something ordinary is unexplained.

An everyday analogy. Uddālaka picks up clay and tells Śvetaketu that everything made of clay is known by knowing clay, the rest being names given to shapes. Then copper. Then an iron nail-parer. Three ordinary objects to make one point that no definition would have carried.

A story with something at stake. Naciketas, offered three boons by Death, spends the third on this: when a person dies there is a doubt — some say he still exists, some say he does not. Teach me this. Death offers him sons, cattle, gold, long life, anything else. He refuses them all.

Negation. The Bṛhadāraṇyaka's neti neti, "not this, not that" — an instruction about how to speak of something rather than a description of it.

Correspondence. Constantly: this part of the body answers to that part of the cosmos, this chant to that world. This is the oldest Vedic habit of all, and the Upaniṣads are full of it.

Not one voice

If you read only the famous lines you will imagine a single serene teacher. The texts are noisier than that.

Śvetaketu comes home after twelve years having mastered the whole Vedic curriculum, and the text calls him swollen-headed and stiff — and then shows that his education has missed something.

Satyakāma cannot tell his teacher his lineage, because his mother does not know it, and he repeats her answer word for word, awkward part included. The teacher accepts him and gives his reason: the boy did not depart from the truth.

Gārgī Vācaknavī questions Yājñavalkya before an assembly, is told not to press further, comes back, announces two questions, compares herself to a warrior taking up two arrows, and presses again — and what she draws out of him is one of the central passages of the corpus.

And in one striking episode the same Śvetaketu is questioned by a king and cannot answer, and his brāhmaṇa father has to go to that king as a pupil — who remarks that this knowledge had never before gone to brāhmaṇas.

On the women: two of them, Gārgī and Maitreyī, take part in these discussions, and they are introduced without any explanation of how a woman comes to be there. Olivelle reads that absence of explanation as evidence for the relatively high position of at least women of some social strata in that period, supported by a rite for obtaining a learned daughter elsewhere in the same text. That is what the evidence carries. It does not carry "women had equal status", and it is not nothing either.

Continuity, not a revolution

The textbook story says: the Vedic people did rituals, then the Upaniṣads rejected ritual and turned inward. It is too clean.

What the earlier Vedic texts do constantly is hunt for hidden connections — bandhu is their word for it — between the ritual and the cosmos. The Upaniṣads keep doing exactly that. What changes is which connection interests them: the search moves towards the human person, and towards what stands at the summit of the whole interconnected arrangement.

That is a shift of attention inside an ongoing project, not a break with it. The vocabulary of sacrifice is retained and re-read. The teacher-and-student structure is retained. The habit of correspondence is retained and intensified.

There are sharp moments. A later verse Upaniṣad, the Muṇḍaka, divides knowledge in two and puts the four Vedas and the six auxiliary sciences in the lower category, reserving the higher for that by which the imperishable is grasped. That is a real edge — and note that it comes from one of the later texts, and that it ranks the Veda rather than discarding it.

Are they all the same age?

No, and the honest answer about dates is short.

The two earliest are the Bṛhadāraṇyaka and the Chāndogya, both probably pre-Buddhist. Then the other early prose ones, Taittirīya, Aitareya and Kauṣītaki. Then the verse Upaniṣads, of which the Kena is the oldest, followed by Kaṭha, Īśā, Śvetāśvatara and Muṇḍaka. Latest are two prose texts, Praśna and Māṇḍūkya.

That relative sequence is well founded. Absolute dates are not. Olivelle puts it as sharply as a specialist can: any dating that attempts precision closer than a few centuries "is as stable as a house of cards".

And this is a different question from the traditional one. Tradition asks how a body of revelation reached us and through whom. Philology asks when particular sentences were composed. Neither answer is the other's refutation.

Why later traditions cared so much

Because of what happened after the detachment.

Once the Upaniṣads were common property rather than the possession of particular schools, they acquired a collective name: Vedānta — the end of the Veda, and by extension its essence. Around the first five centuries CE the Brahmasūtra attempted a systematic statement of their teaching, and the Upaniṣads came to be treated as the portion of the Veda carrying saving knowledge, as against the portion carrying instructions for rites.

That is the first sense of "Vedānta": a part of the Vedic corpus.

The second sense is what almost everyone means today: the families of philosophy built on the Upaniṣads together with the Bhagavadgītā and the Brahmasūtra. And here the essential thing is that they disagree. Śaṅkara, Rāmānuja and Madhva all treat these texts as authoritative and derive incompatible accounts of reality from them. Vedānta is not one philosophy, and no Upaniṣad simply teaches the doctrine of any one school.

You will meet the words brahman and ātman everywhere in this material. This lesson has deliberately not explained them. Lessons 9, 10 and 11 do that, one at a time, and they will need the room.

What to carry out of this

An upaniṣad is a hidden connection before it is a book. That is what the word does in its own texts, and it tells you what kind of reading these texts expect.

They are school literature that was later collected. Not a top floor, not an appendix, and not originally a shelf of separate volumes.

They are a conversation, not a doctrine. Many teachers across many centuries, sometimes disagreeing, often unwilling to state a conclusion at all. That is why later schools could all claim them, and why claiming them was never the same as agreeing.